<%BANNER%>

Responses of Arabidopsis to High Levels of Magnesium Sulfate and of Wheat to a Space-Flight Environment

Permanent Link: http://ufdc.ufl.edu/UFE0024227/00001

Material Information

Title: Responses of Arabidopsis to High Levels of Magnesium Sulfate and of Wheat to a Space-Flight Environment Consequences for (Extra)terrrestrial Plant Growth
Physical Description: 1 online resource (164 p.)
Language: english
Creator: Visscher, Anne
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: arabidopsis, ion, magnesium, mars, microarray, mutant, regolith, root, sulfate, tolerance, toxicity, transcriptome, transporter
Horticultural Science -- Dissertations, Academic -- UF
Genre: Horticultural Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Responses of Arabidopsis to High Levels of Magnesium Sulfate and of Wheat to a Spaceflight Environment: Consequences for (Extra)terrrestrial Plant Growth The ability to utilize in situ resources is crucial for the success of extended manned space exploration of other planetary surfaces such as the Moon or Mars. Martian regolith containing potentially phytotoxic levels of elements is a potential medium for plant growth in bioregenerative life support systems. Studies of surface materials on Mars have detected a variety of hydrated sulfate minerals, including highly soluble magnesium sulfate minerals. Localized weight percentages of magnesium sulfate can reach 10 % in the regolith. Levels of magnesium and sulfate ions toxic to crop plants have been described for soil types on Earth, such as serpentine and acid sulfate soils. We tested whether Arabidopsis knockout lines of genes encoding plasma membrane localized ion transporters in peripheral root cells can thrive in soils containing hyper elevated levels of magnesium sulfate. The selected mrs2-10 and sel1-10 mutant backgrounds do not mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype plants. Based on these findings, a microarray experiment was done to characterize the early gene expression responses of col-0 Arabidopsis roots to elevated concentrations of magnesium sulfate. The cax1-1 mutant line, which has been shown to exhibit increased tolerance for high levels of magnesium, was also included in the experiment. The results for col-0 point to a reduction in root growth and vacuolar storage of magnesium and sulfate upon high magnesium sulfate treatment. Although many transporters or their potential regulators were differentially expressed, more research is needed to discover whether any of those are involved in magnesium sulfate transport. The down-regulated expression of cax1-1 is a natural response to high magnesium sulfate in Arabidopsis, and can explain the fact that only three transcripts were differentially expressed between cax1-1 and col-0. Furthermore, we tested whether changes in gene expression patterns can be detected in wheat plants that are several generations removed from growth in space, compared to control plants with no spaceflight exposure in their lineage. We found that none of the wheat genes represented by 10,263 probes on custom-made microarrays showed a statistically significant difference in expression.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Anne Visscher.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Ferl, Robert J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024227:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024227/00001

Material Information

Title: Responses of Arabidopsis to High Levels of Magnesium Sulfate and of Wheat to a Space-Flight Environment Consequences for (Extra)terrrestrial Plant Growth
Physical Description: 1 online resource (164 p.)
Language: english
Creator: Visscher, Anne
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: arabidopsis, ion, magnesium, mars, microarray, mutant, regolith, root, sulfate, tolerance, toxicity, transcriptome, transporter
Horticultural Science -- Dissertations, Academic -- UF
Genre: Horticultural Science thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Responses of Arabidopsis to High Levels of Magnesium Sulfate and of Wheat to a Spaceflight Environment: Consequences for (Extra)terrrestrial Plant Growth The ability to utilize in situ resources is crucial for the success of extended manned space exploration of other planetary surfaces such as the Moon or Mars. Martian regolith containing potentially phytotoxic levels of elements is a potential medium for plant growth in bioregenerative life support systems. Studies of surface materials on Mars have detected a variety of hydrated sulfate minerals, including highly soluble magnesium sulfate minerals. Localized weight percentages of magnesium sulfate can reach 10 % in the regolith. Levels of magnesium and sulfate ions toxic to crop plants have been described for soil types on Earth, such as serpentine and acid sulfate soils. We tested whether Arabidopsis knockout lines of genes encoding plasma membrane localized ion transporters in peripheral root cells can thrive in soils containing hyper elevated levels of magnesium sulfate. The selected mrs2-10 and sel1-10 mutant backgrounds do not mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype plants. Based on these findings, a microarray experiment was done to characterize the early gene expression responses of col-0 Arabidopsis roots to elevated concentrations of magnesium sulfate. The cax1-1 mutant line, which has been shown to exhibit increased tolerance for high levels of magnesium, was also included in the experiment. The results for col-0 point to a reduction in root growth and vacuolar storage of magnesium and sulfate upon high magnesium sulfate treatment. Although many transporters or their potential regulators were differentially expressed, more research is needed to discover whether any of those are involved in magnesium sulfate transport. The down-regulated expression of cax1-1 is a natural response to high magnesium sulfate in Arabidopsis, and can explain the fact that only three transcripts were differentially expressed between cax1-1 and col-0. Furthermore, we tested whether changes in gene expression patterns can be detected in wheat plants that are several generations removed from growth in space, compared to control plants with no spaceflight exposure in their lineage. We found that none of the wheat genes represented by 10,263 probes on custom-made microarrays showed a statistically significant difference in expression.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Anne Visscher.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Ferl, Robert J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024227:00001


This item has the following downloads:


Full Text





RESPONSES OF ARABIDOPSIS TO HIGH LEVELS OF MAGNESIUM SULFATE AND OF
WHEAT TO A SPACEFLIGHT ENVIRONMENT; CONSEQUENCES FOR
EXTRATERRESTRIALL PLANT GROWTH



















By

ANNE MARIEKE VISSCHER


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2009






































O 2009 Anne Marieke Visscher




































To my parents









ACKNOWLEDGMENTS

I would like to thank my family and friends for supporting me during this PhD. Without

my advisors Rob Ferl and Anna-Lisa Paul, and my committee members Andrew Schuerger,

Charles Guy and Matias Kirst, this project would not have been possible. I sincerely appreciate

all the knowledge and inspiration they have offered over the years. I thank all the members of the

Ferl Lab, especially Beth Laughner, Jordan Barney, Brian Fuller, Matthew Reyes, John

Mayfield, Michael Manak and Tufan Goiklrmak for their help, insights and friendship. The

Interdisciplinary Center for Biotechnology Research at the University of Florida is recognized

for providing valuable assistance to the work presented in this dissertation. The UF College of

Agricultural and Life Sciences, the UF Department of Horticultural Sciences, the Dutch Minister

for Education, Culture and Science, and the Dutch Prins Bernard Cultuurfonds are acknowledged

for their financial support. Finally, I would like to acknowledge my previous advisors at

Wageningen University and mentors at the Institute of Ecotechnics and the Biosphere

Foundation, without whom I would not have been prepared for this proj ect.












TABLE OF CONTENTS


page

ACKNOWLEDGMENT S ............ ..... .__ ...............4....


LIST OF TABLES ............ ..... ._ ...............7...


LI ST OF FIGURE S .............. ...............8.....


LIST OF OBJECTS ............ ..... ._ ...............11...


AB S TRAC T ............._. .......... ..............._ 12...


CHAPTER


1 INTRODUCTION ................. ...............14.......... ......


Potential Phytotoxic Elements in the Regolith on Mars ................. ............................14
Analogue Soils on Earth ................ ...............16................
M agnesium Sulfate ................. ...............18.......... .....
Spaceflight Environment .............. ...............20....

2 GROWTH RESPONSES OF WILDTYPE ARABIDOPSIS AND KNOCKOUT
MUTANTS TO EXCESS LEVELS OF MAGNESIUM SULFATE; CONSEQUENCES
FOR (EXTRA)TERRESTRIAL PLANT GROWTH .............. ...............22....

Introduction............... ..............2
Re sults ................ ...............28.................
Discussion............... ...............3
Conclusions................ ..............3
M materials and M ethods .............. ...............34....


3 TRANSCRIPTOME RESPONSES OF COL-0 AND CAX1-1 ARABIDOPSIS TO
EXCESS LEVELS OF MAGNESIUM SULFATE; CONSEQUENCES FOR
(EX TRA)TERRES TRIAL PLANT GROW TH .............. ...............54....

Introduction............... ..............5
Re sults ......__................ ..........._..........5
Discussion............... ...............6
Conclusions................ ..............8
M materials and M ethods .............. ...............86....


4 EFFECTS OF A SPACEFLIGHT ENVIRONMENT ON HERITABLE CHANGES INT
WHEAT GENE EXPRESSION ................. ...............126................


Introduction............... .............12
Re sults ................ .............129.................













D iscussion............... ..............13

Conclusion .............. ... ...............136......... ......

Materials and Methods .............. ...............136....


CONCLUSIONS................ .............14


LIST OF REFERENCES ................. ...............151...............


BIOGRAPHICAL SKETCH ................. ...............164......... ......










LIST OF TABLES


Table page

2-1 Overview of wildtype and mutant Arabidopsis seed lines ................. .......................40

2-2 Overview of mutant/wildtype growth experiments .........._ ......._ .........__......40

2-3 Elemental composition of first set of six mrs2-10 plants in ppm ..........._.. ........._......41

2-4 Elemental composition of second set of six mrs2-10 plants in ppm............... .................42

2-5 ANOVA (t-test) results for specific genotype~concentration effects in the growth
experiments (Table 2-2) ................. ...............43........... ....

2-6 ANOVA (t-test) results for specific genotype~concentration effects in the ten growth
experiments (Table 2-2) ................. ...............44........... ....

3-1 Experimental conditions of one color microarray experiment .............. ....................9

3-2 Number of genes (and % of total) in GO molecular functional categories per
comparison ................. ...............90.................

3-3 Genes of Arabidopsis thaliana (col-0) with differential expression at q < 0.001 at
Tim e 1 .............. ...............9 1....

3-4 Genes of Arabidopsis thaliana (col-0) with differential expression > 3 fold at Time 1....92

3-5a Q-PCR results of col-0 gene expression at time 3 treatment versus time 1 control;
RNA sources are the same as for the microarray experiment ................. .....................93

3-5b Q-PCR non-normalized results of col-0 gene expression at time 3 treatment versus
time 1 control; RNA sources are the same as for the microarray experiment .................. .93

3-6 Q-PCR results of col-0 gene expression at time 3 treatment versus time 3 control;
repetition 1 .............. ...............93....

3-7 Q-PCR results of col-0 gene expression at time 3 treatment versus time 3 control;
repetition 2 .............. ...............93....

3-8 Genes with differential expression at q < 0.05 between Arabidopsis thaliana caxl-1
and col-0 treated for 3 hours .............. ...............94....

3-9 Genes and related primer sequences selected for Q-PCR............... ...............94.










LIST OF FIGURES


FiMr page

2-1 M~rs2-10 homozy gous knockout mutant line .............. ...............45....

2-2 RT-PCR analysis of the mrs2-10 mutant ................. ...............45........... ..

2-3 Two sets of six mrs2-10 plants growing at the Purdue lonomics proj ect facility. .............45

2-4 Element z-score values for the first set of six mrs2-10 plants .............. .....................4

2-5 Element z-score values for the second set of six mrs2-10 plants .............. ...................46

2.6 Growth experiments on agar. FW biomass of mutant lines is compared to that of
their respective wildtype backgrounds at different levels of MgSO4*7H20O in the agar
m edium .............. ...............47....

2-7 Growth experiments on soil. FW shoot biomass of mutant lines is compared to that
of their respective wildtype backgrounds at different levels of MgSO4*7H20 in the
soil medium............... ...............48.

2-8 Average fresh weight biomass of mrs2-10 and col-0 seedlings in response to
increasing concentrations of MgSO4*7H20 in agar medium.........._._... ......._._.......49

2-9 Average fresh weight shoot biomass of mrs2-10 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium ........._._.... ......_._........49

2-10 Average leaf chlorophyll content of mrs2-10 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium ........._._.... ......_._........50

2-11 Average fresh weight biomass of sell-10 and ws seedlings in response to increasing
concentrations of MgSO4*7H20 in agar medium ......_. ..........._. ........._._.....50

2-12 Average fresh weight shoot biomass of sell-10 and ws plants in response to
increasing concentrations of MgSO4*7H20 in soil medium ........._._.... ......_._........5 1

2-13 Average leaf chlorophyll content of sell-10O and ws plants in response to increasing
concentrations of MgSO4*7H20 in soil medium............... ...............51.

2-14 Average fresh weight shoot biomass of cax1-1 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium ................ ............. .......52

2-15 Average leaf chlorophyll content of caxl-1 and col-0 plants in response to increasing
concentrations of MgSO4*7H20 in soil medium............... ...............52.

2-16 Average fresh weight shoot biomass of caxl cax3 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium ................ ........_._........53











2-17 Average leaf chlorophyll content of caxl cax3 and col-0 plants in response to
increasing concentrations of MgSO4*7H 20 in soil medium ................ ............. .......53

3-1 Hydroponic Arabidopsis growth ................. ...............95................

3.2 Harvest of hydroponically grown Arabidopsis roots .........._... ......... .........._......96

3.3 Overview of microarray experiment ................. ....___ ...............96.....

3-4 Volcano plot of Time 1 .............. ...............97....

3-5 Volcano plot of Time 2 ................. ...............98........ ...

3-6 Volcano plot of Time 3 .............. ...............99....

3-7 Volcano plot of caxl-1 versus col-0 at time 3 ................ ........... ........ ...........100

3-8 Venn diagram of col-0 time series ..........._ .....___ ...............101

3-9 A hierarchical average linkage cluster analysis using uncentered correlation was
done across Time 1, 2 and 3 based on the genes with significant expression
differences at Time 1 .............. ...............102....

3-10a Differentially expressed transcripts encoding metabolic enzymes in the ethylene
biosynthesis pathway .............. ...............108....

3-10b Differentially expressed transcripts encoding metabolic enzymes in the abscisic
biosynthesis pathway .............. ...............109....

3-10c Differentially expressed transcripts encoding metabolic enzymes in the jasmonate
biosynthesis pathway .............. ...............109....

3-10d Differentially expressed transcripts encoding enzymes involved in gibberellins
m etabolism ............_._ ........ ...............110.....

3-10e Differentially expressed transcripts encoding enzymes involved in glycerolipid
metabolism ........._._. ._......_.. ...............111....

3-10f Differentially expressed transcripts encoding metabolic enzymes in the hexosamine
biosynthetic pathway ................. ...............111......... ......

3-10g Differentially expressed transcripts encoding enzymes involved in chlorophyll
breakdown ................. ...............112................

3-10h Differentially expressed transcripts encoding enzymes involved in porphyrin and
chlorophyll metabolism ................. ...............113......... ......

3-10i Differentially expressed transcripts encoding metabolic enzymes in the histidine
biosynthesis pathway ................. ...............114......... ......











3-10j Differentially expressed transcripts encoding metabolic enzymes involved in
galactose metabolism ................ ...............115................

3-10k Differentially expressed transcripts encoding enzymes involved in glycolysis ..............116

3-101 Differentially expressed transcripts encoding enzymes in the citric acid cycle ..............117

3-11 Hierarchical average linkage cluster analysis of transporter gene expression using
uncentered correlation............... ..............12

3-12 Whisker box plots representing gene expression ratio distributions for the Q-PCR
analysis of four genes showing differential expression on the microarrays in col-0 at
time 3 treatment versus time 1 control ................. ...............124........... ..

3-13 Whisker box plots representing gene expression ratio distributions for the Q-PCR
analysis of four genes showing differential expression on the microarrays in col-0 at
time 3 treatment versus time 3 control ................. ...............124........... ..

3-14 Whisker box plots representing gene expression ratio distributions for the Q-PCR
analysis of four genes showing differential expression on the microarrays in col-0 at
time 3 treatment versus time 3 control ................. ...............125........... ..

4-1 Wheat growth in the Laboratory Biosphere ................. ...............140........... ..

4-2 Flowchart of wheat experiments ................. ...............141........... ...

4-3 ANOVA (t-test) results ................. ...............142...............

4-4 Multiple testing correction: Benjamini and Hochberg method............... ..................4

4-5 Multiple testing correction: Q-value method ................. ...............144..............










LIST OF OBJECTS

ob et page

3-1 Microarray raw signal data (.xls file 35 MB) .............. ...............68....

3-2 Microarray analysis data (.xls file 1 MB) .............. ...............68....









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

RESPONSES OF ARABIDOPSIS TO HIGH LEVELS OF MAGNESIUM SULFATE AND OF
WHEAT TO A SPACEFLIGHT ENVIRONMENT; CONSEQUENCES FOR
EXTRATERRESTRIALL PLANT GROWTH

By

Anne Marieke Visscher

August 2009

Chair: Robert J. Ferl
Major: Horticultural Science

The ability to utilize in situ resources is crucial for the success of extended manned space

exploration of other planetary surfaces such as the Moon or Mars. Martian regolith containing

potentially phytotoxic levels of elements is a potential medium for plant growth in

bioregenerative life support systems. Studies of surface materials on Mars have detected a

variety of hydrated sulfate minerals, including highly soluble magnesium sulfate minerals.

Localized weight percentages of magnesium sulfate can reach 10 % in the regolith. Levels of

magnesium and sulfate ions toxic to crop plants have been described for soil types on Earth, such

as serpentine and acid sulfate soils. We tested whether Arabidopsis knockout lines of genes

encoding plasma membrane localized ion transporters in peripheral root cells can thrive in soils

containing hyper elevated levels of magnesium sulfate. The selected mrs2-10 and sell-10 mutant

backgrounds do not mitigate the constraining impacts of high magnesium sulfate concentrations

on wildtype plants. Based on these findings, a microarray experiment was done to characterize

the early gene expression responses of col-0 Arabidopsis roots to elevated concentrations of

magnesium sulfate. The caxl-1 mutant line, which has been shown to exhibit increased tolerance

for high levels of magnesium, was also included in the experiment. The results for col-0 point to









a reduction in root growth and vacuolar storage of magnesium and sulfate upon high magnesium

sulfate treatment. Although many transporters or their potential regulators were differentially

expressed, more research is needed to discover whether any of those are involved in magnesium

sulfate transport. The down-regulated expression of cax1-1 is a natural response to high

magnesium sulfate in Arabidopsis, and can explain the fact that only three transcripts were

differentially expressed between caxl-1 and col-0. Furthermore, we tested whether changes in

gene expression patterns can be detected in wheat plants that are several generations removed

from growth in space, compared to control plants with no spaceflight exposure in their lineage.

We found that none of the wheat genes represented by 10,263 probes on custom-made

microarrays showed a statistically significant difference in expression.









CHAPTER 1
INTTRODUCTION

Potential Phytotoxic Elements in the Regolith on Mars

The planet Mars differs significantly from Earth with regard to gravity, geology,

hydrology, atmosphere, radiation environment and temperature. The conditions on Mars require

habitats and suits that shield humans on manned missions from low temperature, low

atmospheric pressure and harmful radiation. Long duration missions demand the efficient use of

local planetary resources and the recycling of limited materials such as water, pressurized

atmosphere and organic matter while producing food (Barta and Henninger, 1994). Biological

processes are likely to be used for most water, air and food regeneration during long missions

(Drysdale et al., 2003). Large-scale experimentation with bioregenerative life support for human

space exploration started with the Bios-3 facility in Siberia in 1972 (Salisbury et al., 1997). In

the context of plant growth in a future bioregenerative life support system on Mars, Schuerger et

al. (2002) indicate that the use of in situ regolith may have several advantages over hydroponic

systems. These include the immediate bioavailability of plant essential ions, low-tech mechanical

support for plants, and easy access of in situ materials once on the surface. However, plant

growth may be reduced or inhibited by high levels of chemical elements in the regolith. The

most highly phytotoxic materials on Mars are thought to be present in the mobile-element

component of the regolith and composed of sulfates and chlorides (Schuerger et al., 2002). The

current study focuses on sulfate minerals and the potential toxicity of the chemical elements

these include.

Sulfate minerals have been detected on the surface of Mars both from space (remote

sensing) and on the ground (robotic landers). The formation of sulfate minerals on Mars is likely

to have been caused by oxidative, acid-sulfate weathering of basaltic surface material (Golden et









al., 2005). Possible acid-sulfate conditions proposed to have occurred include sulfuric-acid

vapors (acid fog) and waters rich in sulfuric acid. During weathering, elements such as Mg, Ca

and Fe will have leached from the basaltic parent material (e.g. olivine, feldspar, pyroxene) and

reacted with S to form sulfates and other secondary minerals such as iron oxides (Golden et al.,

2005). The Mars Express OMEGA remote sensing mission detected three principal types of

hydrated sulfate deposits: layered deposits within Valles Marineris, extended deposits exposed

from beneath younger units as in Terra Meridiani, and the dark dunes of the northern polar cap

(Bibring et al., 2006). Reflectance spectra from Valles Marineris, Margaritifer Sinus and Terra

Meridiani indicate the association of kieserite (MgSO4*H20), gypsum (CaSO4*2H20) and

polyhydrated sulfate minerals ((MgSO4*7H20), (Fe2 Fe43+(SO4)6 (OH)2*20H20),

(Fe2+Al2(SO4)4*22H20O)) with light-tone layered deposits (Gendrin et al., 2005). In the northern

circumpolar regions, calcium-rich sulfates such as gypsum have been identified in a hydrated

area correlating to the dark longitudinal dunes of Olympia Planitia (Langevin et al., 2005).

Spectra from eastern Terra Meridiani in particular suggest that kieserite is present in etched

terrain deposits (Arvidson et al., 2005).

Analyses by the Mars Exploration Rover landers at Meridiani Planum and Gusey crater

have also indicated the presence of sulfate minerals. The outcrops observed at Meridiani Planum

all contain sulfates with volume abundances of 15 to 35%. Mg- and Ca-sulfates are dominant,

and the average volume abundance of the iron hydroxide sulfate mineral j arositel in the outcrop

is around 10% (Christensen et al., 2004). At Gusey crater, Mg- and Ca-sulfates are estimated to

be present in the outcrops and rocks of the Columbia Hills up to 10 and 1 1% respectively (Ming

et al., 2006). The sulfate mineral content of the localized Paso Robles soil is estimated at 25-29%


(K, Na, H30)(Fe3-xAlx)(SO4)2(OH)6, where x <1 (Klingelhofer et al., 2004)









ferric sulfate, 10% magnesium sulfate, 3-4% calcium sulfate and 2-5% other sulfates (Ming et

al., 2006). Analyses in the 11 cm deep "The Boroughs" trench, which was dug in intercrater

terrain, suggested the presence of approximately 7-22 wt % sulfates (Mg-, Ca- and Fe-sulfate) in

the subsurface regolith (Haskin et al., 2005). Elemental analysis of several other surface and

trench soils at Meridiani Planum and Gusey crater revealed that the average weight percentage of

sulfur varied between 1.92 and 2.92 (Rieder et al., 2004) and 1.69 and 2.95 (Gellert et al., 2004)

respectively. The elemental sulfur is thought to be present in the soil in the form of sulfate

(Banin et al., 1997). The average concentrations of all elements measured in these soils were

compared with each other and with those found in soils at locations visited during the Viking 1,

Viking 2 and Pathfinder missions (Gellert et al., 2004; Rieder et al., 2004). In general, weight

percentages of maj or elements were found to be similar, with some larger differences existing for

minor elements, supporting global mixing of soil by dust storms and admixture of debris from

local rocks (Gellert et al., 2004; Rieder et al., 2004). Yen et al. (2005) also conclude that soil

compositions at five landing sites on Mars are more similar to each other than to the analyzed

rocks (Yen et al., 2005). Overall these results show that sulfate minerals are generally present in

the loose regolith in amounts related to 1.3 to 3 weight % S (Gellert et al., 2004), and that they

can be present locally in even higher quantities. Furthermore, the main cations associated with

the sulfate anion seem to be Ca2+, Mg2+ and Fe2+/3+

Analogue Soils on Earth

Soils on Earth with high levels of sulfate minerals or their individual ions (Ca2+, Fe2+,

Mg2+, SO42-) COuld be (partial) analogues for regolith high in Ca-sulfate, Fe-sulfate or Mg-sulfate

on Mars. The occurrence of high levels of CaSO4*2H20 (gypsum) minerals in soils on Earth is

confined to arid and semi-arid climates where low precipitation prevents gypsum from being

removed by leaching (Palacio et al., 2007). Among the adverse physical features of gypsum soils









are the presence of a hard soil surface crust, which can restrict seedling establishment, the

mechanical instability of the soil material due to its lack of plasticity, cohesion and aggregation,

and, in certain areas, its low porosity, which might limit the penetration of plant roots.

Chemically adverse features of gypsum soils are mainly related to the intense nutritional

impoverishment of the soil caused by the exchange of calcium for other ions retained in the soil

complex, and by the high concentration of sulfate ions (Palacio et al., 2007). Relatively soluble

ferrous or ferric sulfate minerals such as melanterite (FeSO4*7H20) and coquimbite

(Fe2(SO4)3*9H20) can form in soils when ferrous or ferric sulfate-rich solutions are desiccated.

These minerals, or the dissolved ions of them, oxidize and hydrolyze readily to form sulfuric

acid and iron (hydr)oxide minerals (Fanning and Burch, 2006). Magnesium sulfate is so soluble

that it rarely occurs in soils on Earth (Barber, 1995). Magnesium sulfate minerals such as

epsomite (MgSO4*7H20) and hexahydrite (MgSO4*6H20) are important constituents of

evaporative soil environments in cold climates and of cold desert environments, for example in

North Dakota and Saskatchewan (Chou and Seal, 2003). Survey reports regarding these soils

generally do not include information on concentrations of Mg2+ and SO42- in the soil solution

upon dissolution of magnesium sulfate minerals by precipitation, presumably because these ions

are immediately leached out from the rooting zone into the deeper soil layers.

When focusing on the chemical elements contained in sulfate minerals, we find diverse

soils on Earth that are naturally enriched in these. Soils rich in sulfur are for example saline soils,

heavy metal soils, acid sulfate soils or soils in the vicinity of volcanoes, S/CO2 vents and lignite

burns (Ernst, 1998). Iron levels toxic to plant growth generally occur in submerged soils that

allow for reduction of Fe3+, Since the reduced form Fe2+ iS directly bioavailable. Examples of soil

types that can exhibit iron toxicity when submerged are acid Ultisols and Oxisols, and acid









sulfate soils that are rich in Fe(III) oxide hydrates (Sahrawat, 2004). Interflow of Fe from

adj acent areas and salt content can also play a role in increasing soluble iron levels. When soils

that meet these criteria are used for rice production, the concentration of water-soluble Fe2+ can

increase to a few hundred milligrams per liter within weeks following submergence (Sahrawat,

2004). An example of a soil type high in bioavailable magnesium is serpentine soil. Serpentine

soils are formed from the weathered products of ultramafic rocks, whose common mineral

denominator is some form of iron magnesium silicate. In addition, serpentine soils may contain

'impurities' such as minerals with nickel, chromium and cobalt (Kruckeberg, 1999). Common

characteristics of serpentine soils are (1) high concentrations of elements such as iron,

chromium, nickel and cobalt, (2) low concentrations of plant nutrients such as nitrogen,

phosphorus and potassium, (3) a low Ca:Mg quotient compared with non-serpentine soils, and

(4) lower clay contents with lower exchange capacity than other soils (Brooks, 1987). The

infertility of serpentine soils in the context of (crop) plant growth is largely related to their

chemical composition, with the unfavorable effect of magnesium being a maj or cause of the

serpentine problem; the magnesium content of some soils can be as high as 36% MgO, and 4.32

mg/L in the soil solution (Brooks, 1987).

Magnesium Sulfate

On Mars, no direct measurements of ion concentrations in wetted regolith have so far taken

place in regions high in sulfate minerals. Factors such as parent material, weathering history,

texture, structure, organic matter content and hydrology all interact to determine bioavailability

of nutrients in a soil solution accessible to plant roots (Barber, 1995). Given the presence of

sulfur in the form of sulfate on Mars, the concentration of bioavailable SO42- in a soil solution

will depend on the relative weight percentages of the different soluble sulfate minerals in the

regolith, and the interplay between soil characteristics and hydrology mentioned above.









Regarding Ca2+, Mg2+ and Fe2+, Soluble sulfate minerals will not be the only source for these

cations in the bioavailable pool. Their overall bioavailable concentrations in the soil solution of

Martian regolith upon wetting are difficult to predict without having done direct measurements

on Mars. This is especially true in the case of Fe, which can occur in two different oxidation

states: Fe2+ and Fe3+. The relation between these forms of iron in the soil solution depends on the

redox status of the soil (Barber, 1995). Plants primarily absorb Fe2+; either directly or by first

reducing Fe3+-chelates and transporting the resulting Fe2+ (Curie and Briat, 2003). Graminaceous

plants are furthermore able to produce phytosiderophores (PS), which can form Fe3+-PS

complexes that are taken up by specific transporters (Curie and Briat, 2003).

Measurements of ion concentrations in wetted regolith in high sulfate regions on Mars

have not yet taken place. The focus of this research is therefore on documenting plant responses

to elevated levels of specific hydrated sulfate minerals in solution. Preliminary experiments

indicated that increasing concentrations of dissolved sulfate minerals that were added to 1.3%

agar medium with 0.5x Murashige and Skoog nutrient solution had different effects on growth of

wildtype Arabidopsis on Petri dishes. FeSO4*7H20 (EDTA) showed the first growth limiting

effects at lower concentrations than MgSO4*7H20, while CaSO4*2H20 did not show clear

effects up to its maximum soluble concentration. These observations indicated that among the

tested individual sulfate minerals, toxicity can be dominated by the associated cation. Similarly,

the maj or problem on sulfur-enriched saline soils is not the surplus of sulfur but rather elevated

sodium, and the selection pattern of plant species on sulfur-enriched heavy metal soils or acid

sulfate soils is generally governed more by the heavy metal cation than by the sulfur-anion

(Ernst, 1997; Ernst, 1998).









The phytotoxicity of iron upon dissolution of hydrated iron sulfates like FeSO4*7H20 is

dependent on reduced soil conditions, such as in submerged soils. The submersion of soils, as is

done for rice production, is unlikely to be implemented in a bioregenerative life support system

on Mars due to the limitation of water as a local resource. The other sulfate mineral showing

inhibition of plant growth at high bioavailable concentrations when other nutrients remain

constant is MgSO4*7H20. Magnesium sulfate is highly soluble in hydrated mineral form, which

is why it rarely occurs in soils on Earth. Hydrated forms such as kieserite (MgSO4*H20) and

epsomite (MgSO4*7H20) have been detected on Mars by the Mars Express Satellite. Magnesium

sulfate is therefore likely to become bioavailable at high concentrations compared to other

nutrients in the soil solution upon first watering of regolith in materially closed life support

systems situated in high sulfate mineral regions on Mars. This study documents the responses of

Arabidopsis to high levels of magnesium sulfate in solution and explores plant molecular

strategies to reduce accumulation of magnesium sulfate within the plant in order to enhance plant

tolerance. Since Mg2+ is a dominant toxic ion compared to SO42- when MgSO4*7H20 is

dissolved, serpentine soils that are high in Mg2+ COuld function as partial analogue soils for

regolith high in magnesium sulfate on Mars in this study.

Spaceflight Environment

Before plants or seeds arrive on Mars to be planted in a bioregenerative life support

system, they need to be transported in a spaceflight environment. Plants may be grown as part of

a life support system in the transport vehicle during a human mission to Mars. A fundamental

question in early space biology research was whether plants could complete one or more full life

cycles, from seed to seed, under spaceflight conditions. As space research progressed, plant

species such as Arabidopsis thaliana, Bra~ssica urapa, Triticum aestivum L. cultivar USU-Apogee

and Pisum sativum completed their life cycles in growth units on orbiting space stations (Merkys









et al., 1984; Levinskikh et al., 2000; Musgrave et al., 2000; Sychey et al., 2007). The primary

natural identified environmental factors that differ between a space station and a growth chamber

on Earth are gravitational forces and the radiation environment. Tropic responses in plants are

directly affected by microgravity (Brown et al., 1995), yet there is no evidence that microgravity

per se has any detrimental effect on plant metabolism (Stutte et al., 2006). However, indirect

effects of microgravity, such as reduced air circulation due to the lack of convective mixing, can

affect growth when not mediated technically (Musgrave and Kuang, 2003). Much of the early

work in orbit highlighted the necessity for tight control of the environment, including

temperature, humidity, lighting, nutrient delivery, and atmospheric composition (Stankovic,

2001). The successful life cycle experiments indicate that if the indirect effects of microgravity

and a closed environment are ameliorated, the low earth orbit spaceflight environment is not

hostile to plant growth, development and viable seed formation.

A fundamental question to follow is whether the long-term exposure (such as for one or

more life cycles) of plants to the spaceflight environment with its microgravity and radiation

parameters can cause changes in subsequent generations. Plants might be grown for multiple life

cycles as part of a life support system in a transport vehicle during a human mission to Mars.

Some of the seeds harvested from these plants could subsequently be planted in advanced life

support systems on the surface of Mars. To analyze in detail whether plants grown in spaceflight

conditions show changes in subsequent generations, gene expression in wheat plants that are

three generations removed from growth in the MIR space station was compared to gene

expression in wheat plants with no spaceflight exposure in their lineage.









CHAPTER 2
GROWTH RESPONSES OF WILDTYPE ARABIDOPSIS AND KNOCKOUT MUTANTS TO
EXCESS LEVELS OF MAGNESIUM SULFATE; CONSEQUENCES FOR
EXTRATERRESTRIALL PLANT GROWTH

Introduction

The planet Mars differs significantly from Earth with regard to gravity, geology,

hydrology, atmosphere, radiation environment and temperature. The conditions on Mars require

habitats and suits that shield humans on manned missions from low temperature, low

atmospheric pressure and harmful radiation. Long duration missions demand the efficient use of

local planetary resources and the recycling of limited materials such as water, pressurized

atmosphere and organic matter while producing food (Barta and Henninger, 1994). Biological

processes are likely to be used for most water, air and food regeneration during long missions

(Drysdale et al., 2003). Large-scale experimentation with bioregenerative life support for human

space exploration started with the Bios-3 facility in Siberia in 1972 (Salisbury et al., 1997). In

the context of plant growth in a future bioregenerative life support system on Mars, Schuerger et

al. (2002) indicate that the use of in situ regolith may have several advantages over hydroponic

systems. These include the immediate bioavailability of plant essential ions, low-tech mechanical

support for plants, and easy access of in situ materials once on the surface. However, plant

growth may be reduced or inhibited by high levels of chemical elements in the regolith. The

most highly phytotoxic materials on Mars are thought to be present in the mobile-element

component of the regolith and composed of sulfates and chlorides (Schuerger et al., 2002).

As part of the search for evidence of past water on Mars, hydrated sulfate minerals have

been detected on the surface of Mars both from space (remote sensing) and on the ground

(robotic landers). The formation of sulfate minerals on Mars is likely to have been caused by

oxidative, acid-sulfate weathering of basaltic surface material (Golden et al., 2005). Possible









acid-sulfate conditions proposed to have occurred include sulfuric-acid vapors (acid fog) and

waters rich in sulfuric acid. During weathering, elements such as Mg, Ca and Fe will have

leached from the basaltic parent material (e.g. olivine, feldspar, pyroxene) and reacted with S to

form sulfates and other secondary minerals such as iron oxides (Golden et al., 2005).

Measurements of ion concentrations in wetted regolith of high sulfate regions on Mars have not

yet taken place. The focus of this research is therefore on documenting plant responses to

elevated levels of specific hydrated sulfate minerals in solution. Preliminary experiments

indicated that increasing concentrations of dissolved MgSO4*7H20 (epsomite) that were added

to 1.3% agar medium with 0.5x Murashige and Skoog nutrient solution showed increasingly

limiting effects on growth of wildtype Arabidopsis on Petri dishes. Hydrated forms of

magnesium sulfate such as MgSO4*7H20 and MgSO4*H20 (kieserite) have been detected in

several regions by the Mars Express Satellite (Arvidson et al., 2005; Gendrin et al., 2005;

Bibring et al., 2006). Analyses by the Mars Exploration Rover landers at Meridiani Planum and

Gusey crater have also indicated the presence of high levels of magnesium sulfate minerals (up

to 10 %) in outcrops and soils (Christensen et al., 2004; Haskin et al., 2005; Ming et al., 2006).

Magnesium sulfate is highly soluble in hydrated mineral form, which is why it rarely occurs in

soils on Earth (Barber, 1995). Magnesium sulfate is therefore likely to become bioavailable at

high concentrations compared to other nutrients in the soil solution upon first watering of

regolith in materially closed life support systems situated in high sulfate mineral regions on

Mars .

Strategies to alleviate high magnesium sulfate stress in a potential bioregenerative life

support system on Mars could include (1) remediation of regolith by leaching of soluble

magnesium sulfate minerals with water before first use, (2) alleviation of nutrient deficiencies









caused by high levels of Mg2+ and SO42- ions by addition of appropriate fertilizers, and (3)

adaptation to the regolith conditions by planting crop genotypes tolerant of high magnesium

sulfate. Remediation and alleviation require additional materials (such as water) and extra time,

both of which are limited on a manned mission to Mars. The strategy of adaptation on the other

hand would allow for immediate use of the regolith for crop production in an advanced life

support system located in a region high in sulfate minerals on Mars. The focus of this study is

therefore to explore the possibility of using adaptation as a strategy on future missions by

identifying plant variants tolerant of high magnesium sulfate.

Magnesium (Mg2+) is the most abundant divalent cation in a living cell. It stabilizes

membranes and is associated with ATP in a number of enzymatic reactions (Li et al., 2001).

Magnesium is essential for the function of many enzymes, including RNA polymerases,

ATPases, protein kinases, phosphatases, glutathione synthase, and carboxylases. In higher plants,

Mg2+ is the central atom of the chlorophyll molecule and a bridging element for the aggregation

of ribosomes. Key chloroplast enzymes are strongly affected by small variations in Mg2+ leVOIS

in the cytosol and the chloroplast, exemplifying the significance of maintaining Mg2+

homeostasis in plants (Shaul, 2002). Sulfate (SO42-) iS reduced by plants to sulfide and

incorporated into cysteine. Cysteine is an integral part of proteins determining structure and

function of proteins and is involved in redox reactions. Further, cysteine is converted to the

nutritionally important amino acid methionine, as well as a wide range of sulfur-containing

metabolites, predominant among them glutathione (GSH) and S-adenosylmethionine (SAM)

(Nikiforova et al., 2006).

Movement of low-molecular weight solutes such as Mg2+ and SO42- ionS from the external

solution into the cell wall continuum (apoplasm) of roots is a passive process driven by diffusion









of ions or ion transport by mass flow of water. Carboxylic groups in the cell wall act as cation

exchangers, while anions are repelled. The main barrier to solute flux in the apoplasm of roots is

the hydrophobic wall of cells in the endodermis, which prevents passive ion movement into the

stele. Ions enter and exit the cytoplasm of cells by passive or active transport through the plasma

membrane (Marschner, 1995). Ion transport systems can be subdivided into pumps, carriers

(transporters/exchangers) and channels. Members of these groups can exhibit different levels of

specificity for the ions they transport. Plants may have evolved to cope with relatively high

levels of elements in the soil environment by limiting internal accumulation or tolerating high

internal concentrations. In the case of heavy metal tolerance for example, the following

mechanisms are described; (1) binding to the cell wall, (2) restricted influx through the plasma

membrane, (3) active efflux, (4) compartmentalization in the vacuole, (5) chelation at the cell

wall plasma membrane interface, and (6) chelation in the cytoplasm (Marschner, 1995).

In a potential bioregenerative life support system on Mars, an excess of a particular

element in the crew' s diet could affect the presence and availability of other required elements.

This study therefore focuses on the possibility of reducing accumulation of Mg2+ and SO42- ionS

within the plant as a method to enhance plant tolerance to high levels of magnesium sulfate in

the growth medium. Various efforts have illustrated that this strategy can indeed improve

tolerance to certain elements. For example, a line of transgenic wheat plants expressing an

antisense construct of the high affinity K+ transporter TaHKT2; 1 showed reduced sodium uptake

by roots and enhanced growth relative to unstressed plants compared to a control line at high

levels of NaCl in the growth medium (Laurie et al., 2002). Similarly, overexpression of the

Arabidopsis SOS1 gene, which encodes a plasma membrane Na/H' antiporter responsible for

Na+ efflux, limited Na+ accumulation and improved growth compared to control plants at high









NaCl concentrations (Shi et al., 2003). Arabidopsis is a model species in plant molecular biology

research and its genome is fully sequenced. Plasma membrane localized efflux transporters of

Mg2+ and SO42- ions have not been identified to date in the outer root cell layers of Arabidopsis

or other plant species. Several plasma membrane localized proteins in Arabidopsis roots are

known to be responsible for magnesium or sulfate ion uptake. The obj ective of this study is to

characterize the growth of selected uptake transporter gene knockout lines compared to wildtype

Arabidopsis under high levels of magnesium sulfate in both agar and soil medium to determine

whether the mutant lines show enhanced tolerance in the form of a higher fresh weight biomass.

A family of ten putative Mg2+ transport proteins (AtMRS2/AtMGT) has been identified in

Arabidopsis by several groups of researchers (Schock et al., 2000; Li et al., 2001). Reverse

transcription polymerase chain reaction (RT-PCR) analysis of ten AtMRS2 family members

showed that most members are expressed in multiple tissues, including the roots. AtMRS2-10 (=

AtMGT1) functionally complemented a bacterial mutant lacking Mg2+ transport capability and

AtMRS2-10O-GFP-expressing plants showed fluorescence in the periphery of root cells,

suggesting a plasma membrane association (Li et al., 2001). The sulfate transporter gene family

in Arabidopsis consists of 14 isoforms that show homology to one another. H -sulfate co-

transport has been determined for some of these (Hawkesford, 2003). Yoshimoto et al. (2002)

found that sulfate transporters SULTR1;1 and SULTR1;2 co-localize in the root hair, epidermal

and cortical cells that are in contact with the soil solution. In two week old roots, SULTR1;1 is

highly up-regulated under sulfur-limiting conditions, while the constitutively expressed

SULTR1;2 ensures sulfate uptake into plants under both sulfur-replete and sulfur-deficient

conditions (1500-50 CIM sulfate) (Yoshimoto et al., 2002).









For our growth experiment we selected knockout mutants of the AtMRS2-10 and

AtSULTR1;2 genes, which are indicated to encode plasma membrane localized proteins in

Arabidopsis roots that are able to import magnesium and sulfate ions respectively. A knockout

mutant of the AtMRS2-10 (AtMGT 1) gene was characterized based on the

SALK_1003 61.41.30O.x T-DNA insertion line (Alonso and Stepanova, 2003). We will refer to

this homozygous T-DNA insertion knockout line as mrs2-10. The sell-10 knockout mutant of

the AtSultrl;2 gene that we selected was previously characterized by Maruyama-Nakashita et al.

(2003). When grown for 12 days on agar with optimal levels of sulfate (1.7 mM MgSO4), the

uptake of sulfate by the sell-10 mutant was shown to be approximately 20% of wildtype in both

leaves and roots despite an observed up-regulation of SULTR1;1 gene expression (Maruyama-

Nakashita et al., 2003).

In addition to these lines we selected the caxl-1 single- and caxl/cax3 double-knockout

mutants of the CAX1 and CAX3 genes encoding vacuolar H /Ca2+ transporters characterized by

Cheng et al. (2003, 2005). Calxl-1 mutant plants grown on agar medium for 10 days were

observed to be more tolerant of Mg2+ (10 mM and 25 mM MgCl2) and Mn2+ (1.5 mM MnCl2)

stresses than wild-type plants, while being more tolerant of medium lacking Ca2+ (Cheng et al.,

2003). In a separate study, a CAX1 knockout mutant was identified through a mutant screen on

nutrient solutions reflecting low Ca:Mg ratios characteristic of serpentine soils (Bradshaw,

2005). Calxl mutants have significantly reduced levels of Mg in their leaves to the extent of 0.7

standard deviations (0) below the mean reported for wildtype (Bradshaw, 2005). Although the

Mg content of caxl mutant roots has not been analyzed so far, the reduced levels of Mg in leaves

is in line with the obj ective of this study to identify Arabidopsis variants showing improved

tolerance of high magnesium sulfate by at least partly reducing accumulation of Mg2+ Or SO42-









ions within the plant. Similar to what was observed for cax1-1 plants, double mutant caxl/cax3

plants displayed robust growth on agar medium containing 15 mM MgCl2, which caused growth

defects in the control and cax3 lines. Under regular nutrient conditions, the caxl/cax3 plants

grew more slowly and were small in comparison with the wildtype and single-mutant plants. The

double mutants showed significant differences in concentration of multiple elements, including a

reduction of Ca2+ and Mg2+ in Shoot tissue relative to wildtype (Cheng et al., 2005). We selected

the caxl-1 and caxl/cax3 lines to see if we could confirm their observed tolerance of high levels

of Mg2+ when replacing the associated anion Cl- with SO42-, and when using a different growth

medium (soil) and a longer period of growth (4 weeks).

Results

Mrs2-10 T-DNA Insertion Line Characterization

A homozygous T-DNA insertion mutant for AtMRS2-10 was identified by PCR based on

the SALK_1003 61.41.30O.x line (Figure 2-1). RT-PCR analysis showed that AtMRS2-10 mRNA

was absent from homozygous mutant leaves while it could be detected in col-0 leaves (Figure 2-

2). a-Tubulin mRNA, which was used as a constitutive control, was detected both in the

homozygous mutant and col-0 leaves (Figure 2-2). The homozygous knockout mutant was

backcrossed to col-0 three times before self-fertilization yielded homozygous backcrossed

mutants identified by PCR that produced seeds for the subsequent experiments. The knockout

mutant line was named mrs2-10. In order to analyze its ionome (elemental profile), mrs2-10

seeds were sent to the Purdue lonomics proj ect (Figure 2-3). Leaves from two sets of six mrs2-

10 plants each were analyzed for their content of elements in ppm, and results were compared to

those of col-0. Leaves from the mrs2-10 sets showed no statistically significant (p < 0.05)

differences in concentration of particular elements compared to leaves from the col-0 set, except

for boron (B) in one of the sets (Tables 2-3, 2-4, Figures 2-4, 2-5).









Biomass Comparisons of Wildtype and Mutant Lines

Transporter gene knockout mutant lines were compared to their respective wildtype

backgrounds in ten growth experiments on the basis of whole plant or shoot fresh weight

biomass levels, and of shoot chlorophyll levels (Tables 2-1, 2-2). Statistical analysis of the data

per experiment with ANOVA first of all confirmed that increasing concentrations of dissolved

MgSO4*7H20 in agar and soil significantly reduce wildtype Arabidopsis (ws, col-0) biomass

(Table 2-5). Biomass and chlorophyll level comparisons between mutant and wildtype lines at

specific concentrations per experiment were also part of the statistical analysis with ANOVA

(Table 2-6, Figures 2-8 to 2-17). Results from these comparisons indicate whether knocking out

of transporter genes encoding proteins responsible for Mg2+ and SO42- ion uptake can improve

growth at the whole plant level when plant roots are exposed to high levels of these ions in the

growth media. M~rs2-10 plants did not show a marked difference in whole plant biomass

compared to col-0 when grown on agar medium, although at 4 mM MgSO4*7H20, the slight

advantage observed of 6.8% was found to be statistically significant (Figures 2-6, 2-8, Table 2-

6). On soil, mrs2-10 FW shoot biomass and leaf chlorophyll content were indistinguishable from

that of wildtype (Figures 2-7, 2-9, 2-10, Table 2-6). Sell-10 plants exhibited a significant

reduction in whole plant biomass compared to ws on agar medium at 0, 4 and 12 mM

MgSO4*7H20 (Figures 2-6, 2-11, Table 2-6). On soil, sell-10 FW shoot biomass and leaf

chlorophyll content were indistinguishable from wildtype (Figures 2-7, 2-12, 2-13, Table 2-6).

Calx1-1 plants showed a significantly higher FW shoot biomass compared to col-0 when grown

on soil at 80 and 100 mM MgSO4*7H20 (Table 2-6). The increase in caxl-1 shoot biomass over

that of col-0 was 89% for 80 mM and 149% for 100 mM at 4 weeks (Figures 2-7, 2-14). The

absolute fresh weight of the caxl-1 shoots was still low (20%) compared to untreated col-0. Leaf

chlorophyll content was also significantly higher in caxl-1 compared to col-0 at 80 and 100 mM









MgSO4*7H20 (Figure 2-15, Table 2-6). Calxl cax3 plants showed no significantly higher FW

shoot biomass compared to col-0 when grown on soil at 80 and 100 mM MgSO4*7H20 (Table 2-

6). The average increases of 26.8% and 33.2% observed in caxl cax3 in contrast to col-0 were

not found to be statistically significant (Figures 2-7, 2-16). Leaf chlorophyll content was

significantly higher in carl cax3 compared to col-0 at 0, 80 and 100 mM MgSO4*7H20 (Figure

2-17, Table 2-6).

Discussion

The obj ective of this study was to characterize the growth of selected uptake transporter

gene knockout lines compared to wildtype Arabidopsis under high levels of magnesium sulfate

in both agar and soil medium to determine whether the mutant lines show enhanced tolerance in

the form of a higher fresh weight biomass. The results reveal that knockout mutant lines of the

known genes in Arabidopsis encoding root plasma membrane based uptake transporters of Mg2+

and SO42- ions did not show a significant increase in biomass compared to wildtype when grown

on soil treated with high levels of MgSO4*7H20 in solution. Furthermore, the tolerance of caxl-

1 mutant plants to high levels of MgCl2 in agar medium was confirmed for soil medium treated

with high levels of MgSO4*7H20 in solution, although the significant differences in biomass

after 4 weeks on MgSO4*7H20 treated soil do not appear to be as dramatic as the size

differences observed by Cheng et al. (2003) after 10 days of growth on MgCl2 treated agar.

It is not known at present whether proteins encoded by AtMRS2-10 are primarily

responsible for uptake of magnesium ions from the soil solution, or to what extent non-annotated

proteins or non-specific transport systems play a role in this process. Schock et al. (2000)

speculate that the function of the AtMRS2 gene family may be the maintenance of metal ion

homeostasis in different cellular compartments (i.e. over different cellular membrane systems).

Overexpression of AtMRS2-10 (AtMGT 1) in Nicotiana benthamniana led to increased









accumulation of magnesium (Mg), manganese (Mn), and iron (Fe) per unit dry weight and per

plant compared to wildtype plants (Deng et al., 2006). We showed that the opposite approach of

preventing AtMRS2-10 mRNA from being made did not lead to a reduction in the accumulation

of Mg or any other elements measured in the nars2-10 mutant plants compared to wildtype under

normal nutrient conditions. The observed small increase in nars2-10 biomass relative to col-0 on

agar at 4 mM MgSO4*7H20 might have been due to slight differences in average seed size and

uncontrolled environmental variation in this medium.

A large difference in results between agar and soil medium was observed for the biomass

comparisons of sell-10 and ws. This discrepancy might be attributable to differential affinity of

media components for sulfate or other nutrients. The agar growth medium we used consists of

neutral agarose and negatively charged agaropectin. Agaropectin is a complex polysaccharide

that is sulfated to some degree. Adding sulfate to the agar medium might increase the percentage

of agaropectin sulfate groups by replacing other groups such as pyruvate and methoxyl, thereby

making sulfate less bioavailable. Highly sulfated forms of agaropectin (20 30% sulfate) have

for example been found in Gelidium anzansii (Qi et al., 2008). The soil used in these experiments

contains vermiculite clay and sphagnum moss organic matter, both of which have net negative

charges unrelated to sulfate that can interact with positive cations. In the contained environment

of the soil trays, sulfate does not leach out and should generally be completely bioavailable when

in solution.

SULTR1;1 and SULTR1;2 are the two essential components of the root sulfate uptake

system; double knockout plants lack the ability to take up sulfate at low and optimal levels in the

growth medium (Yoshimoto et al., 2007). Regarding the results of the soil biomass comparisons

it could be hypothesized that the concentrations of sulfate by itself are not high enough to affect









growth of wildtype so much as to cause a reduction in biomass. The observed decline in biomass

would then be caused solely by magnesium, having a dominant effect over this concentration

range. In this scenario, a sulfate uptake mutant would be unable to show improved tolerance

based on biomass measurements. Perhaps at higher concentrations of sulfate and in the presence

of cations less phytotoxic than Mg2+, the mutant line would have a higher biomass than wildtype.

If the sulfate concentrations are indeed high enough to cause a partial effect on biomass,

alongside magnesium, it could be postulated that the reduction in SO42- uptake characteristic of

the AtSULTR1;2 mutant might be too extreme for this concentration range to lead to improved

growth compared to wildtype. In that case, mRNA down-regulation or post-translational

regulation of SULTR1;2 might be more advantageous than a complete knockout.

As mentioned in the results, carl-1 showed a small, and caxl/cax3 showed no statistically

significant improvement in growth compared to col-0 after 4 weeks on soil, while caxl-1 showed

a large improvement in size relative to col-0 after 10 days on agar medium (Cheng et al., 2005).

An explanation for the observed differences in results between soil and agar may be the reduced

availability of Mg2+ in Soil, as well as the greater environmental variation in a soil medium,

rather than the replacement of the Cl- anion by the SO42- anion. Magnesium ions are held by the

negatively charged clay and organic matter, and so may be less highly bioavailable than in agar.

The differences in biomass between col-0 and caxl-1 or caxl/cax3 on soil with high levels of

magnesium sulfate are likely to be more pronounced after an even longer period of growth.

Conclusions

Arabidopsis lines carrying knockout T-DNA insertion mutations in genes encoding known

Mg2+ and SO42- uptake transporters localized to the plasma membrane in root cells do not

mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype

Arabidopsis plants. An Arabidopsis line carrying a knockout mutation of the vacuolar CAX1









gene (caxl-1) showed a significant improvement in growth on soil treated with high levels of

MgSO4*7H20 in solution. A reduction in leaf magnesium content in caxl mutants compared to

wildtype (0.7 o below normal) was reported in a previous study (Bradshaw, 2005). Although the

Mg content of caxl mutant roots has not been analyzed so far, the reduced levels of Mg in leaves

indicate that Arabidopsis CAX1 knockout mutants are in line with our obj ective of identifying

Arabidopsis variants that show improved tolerance of high magnesium sulfate by at least partly

limiting accumulation of Mg2+ Or SO42- ions within the plant. Therefore, genes in crop species

encoding transporter proteins similar in function to the protein encoded by CAX1 in Arabidopsis

are proposed candidate genes for enhancing tolerance of crop plant species to regolith high in

soluble magnesium sulfate minerals used in an advanced life support system on Mars.

The identification by Bradshaw (2005) of a CAX1 knockout mutant tolerant of low Ca:Mg

ratios in solution, which are characteristic of serpentine soil solutions, confirms the

appropriateness of serpentine soils as partial analogue soils on Earth for regolith high in soluble

magnesium sulfate minerals on Mars. Leaf Ca:Mg molar ratios of nonserpentine plant species are

generally equal to that of the soil, while serpentine species maintain significantly higher leaf

Ca:Mg than both their nonserpentine counterparts and the soil (O'Dell et al., 2006). The authors

conclude that elevated leaf Ca:Mg in the serpentine species was achieved by selective Ca2+

transport and/or Mg2+ eXClUSion operating at the root-to-shoot translocation level, as root Ca and

Mg concentrations did not differ between serpentine and nonserpentine species. Genetic

differentiation between populations ofArabidopsis lyrata growing on granitic or serpentinic soils

was measured by using an Arabidopsis thaliana tiling array that has 2.85 million probes

throughout the genome (Turner et al., 2008). The study found significant overrepresentation of

genes involved in ion transport, and one gene in particular, calcium-exchanger 7 (CAX7), was










presented as an excellent candidate gene for adaptation to low Ca:Mg ratios in A. lyrata. It is

currently not known which transporters or their regulators could be involved in the observed

lower levels of Mg or higher Ca:Mg ratios in leaves from Arabidopsis caxl-1 and serpentine

species respectively. Results from the analysis of genetic differentiation between A. lyrata

populations include an overrepresentation of genes encoding proteins involved in ion transport.

Some of these might play a role in reducing Mg2+ accumulation within the plant.

To identify genes in Arabidopsis thaliana with potential to enhance tolerance to high

magnesium sulfate by limiting accumulation within the plant, such as genes encoding root

plasma membrane localized import systems in addition to AtMRS2-10 and AtSULTR1;2, as well

as genes encoding efflux systems or regulators of plasma membrane transporter activity, the

early transcriptome responses of col-0 and cax1-1 Arabidopsis roots to high magnesium sulfate

stress were documented (Chapter 3). The Ca:Mg ratio used in the high magnesium sulfate

treatment for the microarray transcriptome analysis experiment corresponded to ratios found in

serpentine soil solutions.

Materials and Methods

Seed Lines and Growth Experiments

An overview of the wildtype and knockout mutant Arabidopsis seed lines used in this

study is given in Table 2-1. The table lists the names of the seed lines, the genes that are knocked

out in each line, the donor institution for each line, and the original seed line where appropriate.

An overview of the mutant/wildtype growth experiments is given in Table 2-2. For each

experiment, the table lists the mutant/wildtype lines that were compared, the growth medium, the

MgSO4*7H20 concentrations, and the plant growth characteristic that was analyzed.









Mrs2-10 T-DNA Insertion Line Characterization

A homozygous knockout T-DNA insertion line for AtMRS2-10 (Atlg80900) was

identified by PCR and RT-PCR using the original SALK_100361.41.30.x line. For PCR,

oligonucleotide primers MRS2-10_LP (5 '-CAGGATCAAAGCATCGTTCTC-3 ') and MRS2-

10 RP (5'- TAGGAGC TCAGAAGAC GC AAC -3') were de si gned u si ng software avail1ab le on

the Salk Intitute websitel. In addition, the T-DNA specific primer LBb 1 (5'-

GCGTGGACCGCTTGCTGCAACT-3 ') was included as designed and recommended by the

Salk Institute Combinations of the primers were used to identify plants for which the T-DNA

insertion was present in both AtMRS2-10 alleles. Genomic DNA was extracted from T-DNA

insertion mutant leaves using the Shorty2 method and from col-0 leaves using the DNeasy Plant

Mini Kit (Qiagen). PCR was carried out using JumpStart Taq DNA polymerase (Sigma). PCR

products were separated in agarose gels and stained with SYBR Safe DNA gel stain (Invitrogen).

A confirmed homozygous T-DNA insertion mutant was backcrossed to col-0 three times before

allowing self-fertilization. Homozygous plants backcrossed 3x were identified by PCR and used

for seed generation. For RT-PCR, RNA was extracted from leaves of col-0 and a homozygous T-

DNA insertion line for AtMRS2-10 (At80900) with the RNeasy Plant Mini Kit (Qiagen). Gene

specific oligonucleotide primers MRS2-10_LP1 (5' -AGGGTTACTTTGTCGGAGA-3 ') and

MRS2-10 RP1 (5' -TACACGGGGTTTTATCTTG-3 ') were designed based on Arabidosis

genomic DNA sequence information (NCBI). Alpha-(ct)-Tubulin was used as a constitutive

control with primers according to Yoshimoto (2002). RT-PCR was carried out using the OneStep

RT-PCR kit (Qiagen). PCR products were separated in agarose gels and stained with SYBR Safe

Shttp://signal. salk.edu/tdnaprimers.html


bli w\ il ithos.ufl.edu/meteng/HansonWebpagecontents/Nceccdslto~tlAaioss2Gnmc2
DNA










DNA gel stain (Invitrogen). An ionome analysis of the backcrossed homozygous AtMRS2-10

knockout mutants (mrs2-10) was performed as part of the Purdue lonomics proj ect and results

are available online through the Purdue lonomics Information Management System3 (Baxter et

al., 2007).

Seed Production of Wildtype and Knockout Mutant Lines on Soil Medium in Black Plastic
Trays

Variation in seed characteristics between the different lines was minimized by growing

plants from each line for seed production on soil medium under similar growth conditions. Seeds

from the ws, col-0, mrs2-10, sell-10, caxl-1 or caxl cax3 lines were evenly planted (5 seeds per

tray well) on individual trays with soil medium (sphagnum peat moss, vermiculite and perlite)

that was pre-wetted with nutrient solution. The nutrient solution consisted of 2.2 g/L Murashige

and Skoog salts (Murashige and Skoog, 1962) and 0.5 g/L 2-(N-morpholino)ethanesulfonic acid

(MES) buffer. For the caxl-1 and caxl cax3 lines, the nutrient solution also contained 20-25 mM

MgSO4*7H20. The pH of the solutions was adjusted to 5.70 5.75 with KOH. The trays with

soil were covered with plastic wrap to maintain sufficient humidity and placed in a cold room at

4oC for three days. After cold treatment they were moved to a growth room at 23oC where the

seeds germinated, the plastic wrap was removed after 3-5 days, and the plants were grown until

maturity. The plants were watered twice a week from below, alternatively with tap water and the

nutrient solution described above. They were furthermore thinned to two plants per tray well

after one week to maintain sufficient spacing, and pruned to reduce growth of secondary

inflorescences. Seeds were harvested when siliques had turned yellow or light brown.






3 blip1 w il it .ionomicshub.org/home/PiiMS









Plant Growth Experiments on Agar Medium in Vertical Petri Dishes

Seeds from the ws, col-0, mrs2-10 and sell-10 lines were sterilized by soaking them for 15

minutes in a 40% (v/v) bleach solution with a drop of Tween 20

(polyoxyethylenesorbitanmonolaureate). The seeds were then washed six times with sterilized

(autoclaved) water. Each liter of agar medium was prepared by combining double distilled water,

2.2 g Murashige and Skoog salts, 1 mL 1000x Gamborg vitamins, 0.5 g MES buffer, 0.6 % (w/v)

sucrose and 1.3 % (w/v) granulated agar. This functioned as the basic and control medium to

which MgSO4*7H20 was added as 4, 8, 12, 16, 24 or 28 mM for the treatments. The pH of the

medium was adjusted to 5.70 5.75 with KOH. The medium was then autoclaved at 121oC for

21 minutes and kept in the autoclave at around 50-70oC until the medium was distributed over

the Petri dishes. A single 10 cm x 10 cm plate contained approximately 50 ml of autoclaved

medium supporting 10 sterilized seeds that were planted at regular intervals on a row 1.75 cm

from the top edge of the plate in its vertical configuration. To control for environmental variation

when comparing plant growth per concentration, 5 seeds from both a mutant line and its

associated wildtype background line were planted on a single plate, alternatively on the left and

right sides of the plates. Ten Petri dishes per concentration were used per experiment. The plates

were sealed with surgical tape (3M Micropore) and stratified for three days at 4oC before being

moved to a growth chamber at 23oC. There they were placed at a 900 angle in racks holding ten

plates each that were randomly placed on the growth benches and put perpendicular to the

fluorescent light bulbs above them. The plants were analyzed at day 13 and the experiment was

repeated three times.

Plant Growth Experiments on Soil Medium in Black Plastic Trays

Seeds from the ws, col-0, mrs2-10, sell-10, caxl-1 and caxl/cax3 lines were sterilized as

described above. To control for environmental variation when comparing plant growth per









concentration, each tray was divided in 8 sections of 6 wells each, with either a mutant line or its

associated wildtype background line planted in one section, in alternating fashion. This method

resulted in 4 sections per seed line per tray. The seeds were evenly planted (5 seeds per tray well)

on the soil medium (sphagnum peat moss, vermiculite and perlite) that was wetted with nutrient

solution. The basic and control nutrient solution consisted of 2.2 g/L Murashige and Skoog salts

and 0.5 g/L MES buffer. For the treatments, MgSO4*7H20 was added to the basic solution at 20,

60, 80 or 100 mM concentration. The pH of the solutions was adjusted to 5.70 5.75 with KOH.

The trays with soil were covered with plastic wrap to maintain sufficient humidity and placed in

a cold room at 4oC for three days. After stratification, the trays were moved to a growth room at

23oC where they were randomly placed on the growth benches. There, the seeds germinated, the

plastic wrap was removed after 3-5 days, and the surviving plants were thinned to two plants per

well (maximum of 12 plants per section) after one week to maintain sufficient spacing. The

plants were watered twice a week from below, alternatively with tap water and the appropriate

nutrient solution. The plants were analyzed at week 4 and the experiment was repeated three

times.

Fresh Weight Biomass and Chlorophyll Measurements and Statistical Analysis

Fresh weight biomass of plants grown on agar medium was measured by grouping and

weighing whole seedlings per plant line per Petri dish. The number of seeds germinated per seed

line was recorded to calculate the average whole plant FW biomass per seedling. Fresh weight

biomass of plants grown on soil was measured by grouping and weighing whole shoots of plants

per tray section. Leaf chlorophyll content of plants grown on soil medium was measured with a

Minolta SPAD-502 meter. One leaf measurement per plant per tray well was taken. Biomass or

chlorophyll levels of wildtype and knockout mutant Arabidopsis lines were evaluated per

experiment for a total of ten experiments (Table 2-2) in a mixed analysis of variance (ANOVA)









model using SAS 9.2 and JMP Genomics 7 software. Genotype and concentration were included

in the ANOVA model as fixed effects, while batch (replication set) was included as a random

effect. Unbiased estimates of biomass or chlorophyll levels (least-square means) were generated

for each effect (genotype, concentration, or genotype" concentration) per experiment. The

estimated levels of biomass or chlorophyll were then compared using a series of t-tests,

generating estimated differences and corresponding p-values. Differences with associated p-

values < 0.05 were considered significant. The results for the comparisons of interest (biomass or

chlorophyll levels between genotypes per concentration; biomass levels of wildtype between

concentrations) were represented in graphs and tables using Microsoft Excel 2007.































Analysis
whole plant FW biomass
shoot FW biomass

shoot chlorophyll

whole plant FW biomass
shoot FW biomass

shoot chlorophyll
shoot FW biomass

shoot chlorophyll
shoot FW biomass

shoot chlorophyll


Table 2-1. Overview of wildtype and mutant Arabidopsis seed lines
Seed line Knockout gene Origin Original seed line
ws

col-0

mrs2-10 AtMRS2-10; Atlg80900 Salk Institute SALK 100361.41.30.x


RIKEN

(Dr. H. Takahashi)
BCM

(Prof. K. D. Hirschi)
BCM

(Prof. K. D. Hirschi)


AtSULTR1;2; Atlg78000


AtCAX1; At2g38170


sell-10


cax1-1


caxlcax3 AtCAX1;
AtCAX3;


At2g38170 and

At3g51860


Table 2-2. Overview of mutant/wildtype growth experiments
Exp. # Mutant/wildtype set Medium MgSO4*7H20 conc.
1 mrs2-10/col-0 Agar 0, 4, 12, 20, 28 mM
2 mrs2-10/col-0 Soil 0, 20, 60, 100 mM

3 mrs2-10/col-0 Soil 0, 20, 60, 100 mM

4 sell-10/ws Agar 0, 4, 12, 20, 28 mM

5 sell-10/ws Soil 0, 20, 60, 100 mM

6 sell-10/ws Soil 0, 20, 60, 100 mM

7 caxl-1/col-0 Soil 0, 60, 80, 100 mM

8 caxl-1/col-0 Soil 0, 60, 80, 100 mM

9 caxl/cax3/col-0 Soil 0, 80, 100 mM

10 caxl/cax3/col-0 Soil 0, 80, 100 mM





Table 2-3. Elemental composition of first set of six mrs2-10 plants in ppm


Element
Li7
B11
Na23
Mg25
P31
K39
Ca43
Cr52
Mn55
Fe56
Co59
Ni60
Cu65
Zn66
Ga71
As75
Se77
Mo95
Cd111
In113
Pb208
Al27
S34
S48
Cl35
Ca44
Fe57
Se82
Rb85
Sr88
Mo98
Cdll4


Pot Avg
39.73
144.4
1285
12400
8937
38830
76030
NaN
51.04
NaN
0.2769
6.196
7.195
134.7
NaN
0.5653
NaN
NaN
NaN
NaN
NaN
NaN
12490
NaN
NaN
NaN
141.7
13.23
28.32
NaN
4.768
5.127


BG Avg
37.57
160.4
1202
12570
8830
37980
75450
NaN
53.55
NaN
0.2833
6.854
7.278
141.3
NaN
0.5997
NaN
NaN
NaN
NaN
NaN

11660
NaN
NaN
NaN
142.2
12.03
26.71

4.969
5.192


% Diff Med
0.01664
-0.06841
0.0759
-0.0058
-0.0232
-0.03327
0.05812
0
-0.09456
0
-0.04403
-0.01991
0.02622
0.05338
0
0.0155
0
0
0
0
0

0.04733
0
0
0
-0.003369
0.1389
0.00047

0.04532
0.07996


% Diff StdDev
0.09724
0.07521
0.01471
0.05367
0.09511
0.01389
0.09048

0.1339

0.1274
0.09567
0.1341
0.1084

0.1981




0.14


0.18
0.22
0.17


0.21147



0.2227


p value
0.2075
0.0434*
0.2061
0.3588
0.4173
0.3679
0.4455

0.3222

0.366
0.1277
0.4291
0.2858

0.3188








0.122




0.4832
0.2129
0.2397

0.3982
0.4674





Table 2-4. Elemental composition of second set of six mrs2-10 plants in ppm


Element
Li7
B11
Na23
Mg25
P31
K39
Ca43
Cr52
Mn55
Fe56
Co59
Ni60
Cu65
Zn66
Ga71
As75
Se77
Mo95
Cd111
In113
Pb208
Al27
S34
S48
Cl35
Ca44
Fe57
Se82
Rb85
Sr88
Mo98
Cdll4


Pot Avg
35.29
155.6
1246
12500
8893
40520
74980
NaN
58.53
NaN
0.2761
6.147
6.908
127.7
NaN
0.6195
NaN
NaN
NaN
NaN
NaN
NaN
12500
NaN
NaN
NaN
142.5
11.29
27.05
NaN
5.639
4.353


BG Avg
37.57
160.4
1202
12570
8830
37980
75450
NaN
53.55
NaN
0.2833
6.854
7.278
141.3
NaN
0.5997
NaN
NaN
NaN
NaN
NaN

11660
NaN
NaN
NaN
142.2
12.03
26.71

4.969
5.192


% Diff Med
-0.05328
-0.00365
0.07348
-0.00975
-0.02845
0.04384
0.00623
0
-0.05696
0
-0.06496
-0.0671
-0.05149
-0.07396
0
0.1045
0
0
0
0
0

0.02702
0
0
0
-0.02592
0.0371
0.03622

0.2068
-0.1131


% Diff StdDev
0.2299
0.05099
0.1367
0.05994
0.08417
0.1313
0.07332

0.2206

0.1774
0.06689
0.06632
0.03678

0.1882









0.18


0.09718

0.24



0.06383


p value
0.2963
0.2729
0.3261
0.4449
0.4487
0.1514
0.4528

0.2363

0.3803
0.0997
0.1228
0.0956

0.3905








0.0974




0.4873
0.275
0.3792

0.2169
0.1308










Table 2-5. ANOVA (t-test) results for specific genotype~concentration effects in the growth
experiments (Table 2-2). The table lists the results of differences in biomass of
wildtype Arabidopsis between concentrations of MgSO4*7H20.
Conc. Genotype _Conc. _Genotype Diff. Estimate StdErr DF tValue P-value
#4; ws whole plant FW biomass comparison on agar medium
0 ws 4 ws 1.4948 0.5912 288 2.53 0.012*
0 ws 12 ws 4.4467 0.5912 288 7.52 <.0001*
0 ws 20 ws 11.0023 0.5912 288 18.61 <.0001*
0 ws 28 ws 14.2145 0.5912 288 24.04 <.0001*
4 ws 12 ws 2.9518 0.5912 288 4.99 <.0001*
4 ws 20 ws 9.5075 0.5912 288 16.08 <.0001*
4 ws 28 ws 12.7197 0.5912 288 21.52 <.0001*
12 ws 20 ws 6.5557 0.5912 288 11.09 <.0001*
12 ws 28 ws 9.7678 0.5912 288 16.52 <.0001*
20 ws 28 ws 3.2122 0.5912 288 5.43 <.0001*
#5; ws shoot FW biomass comparison on soil medium
0 ws 20 ws 2.7275 0.3531 86 7.73 <.0001*
0 ws 60 ws 5.145 0.3531 86 14.57 <.0001*
0 ws 100 ws 7.1923 0.3531 86 20.37 <.0001*
20 ws 60 ws 2.4175 0.3531 86 6.85 <.0001*
20 ws 100 ws 4.4648 0.3531 86 12.65 <.0001*
60 ws 100 ws 2.0473 0.3531 86 5.8 <.0001*
#1; col-0 whole plant FW biomass comparison on agar medium
0 col-0 4 col-0 0.5187 0.5915 288 0.88 0.3813
0 col-0 12 col-0 10.676 0.5915 288 18.05 <.0001*
0 col-0 20 col-0 19.8547 0.5915 288 33.57 <.0001*
0 col-0 28 col-0 21.342 0.5915 288 36.08 <.0001*
4 col-0 12 col-0 10.1573 0.5915 288 17.17 <.0001*
4 col-0 20 col-0 19.336 0.5915 288 32.69 <.0001*
4 col-0 28 col-0 20.8233 0.5915 288 35.21 <.0001*
12 col-0 20 col-0 9.1787 0.5915 288 15.52 <.0001*
12 col-0 28 col-0 10.666 0.5915 288 18.03 <.0001*
20 col-0 28 col-0 1.4873 0.5915 288 2.51 0.0125*
#2 and #7; col-0 shoot FW biomass comparison on soil medium
0 col-0 20 col-0 3.4808 0.6337 86 5.49 <.0001*
0 col-0 60 col-0 7.8975 0.6337 86 12.46 <.0001*
0 col-0 80 col-0 8.302 0.4671 86 17.77 <.0001*
0 col-0 100 col-0 10.038 0.6337 86 15.84 <.0001*
20 col-0 60 col-0 4.4167 0.6337 86 6.97 <.0001*
20 col-0 100 col-0 6.5572 0.6337 86 10.35 <.0001*
60 col-0 80 col-0 3.0203 0.4671 86 6.47 <.0001*
60 col-0 100 col-0 2.1405 0.6337 86 3.38 0.0011*
80 col-0 100 col-0 0.2838 0.4671 86 0.61 0.545
































































-0.2847


1.0107


566
566
566
566


1.8972 1.0107
5.7931 1.0107
7.6306 1.0107
comparison on soil medium
-6.5781 0.3312
0.3279 0.3312
0.3345 0.3312
comparison on soil medium
8.7083 0.8883
6.2347 0.8883
8.2771 0.9219


414 9.8
414 7.02
414 8.98


t~ralue P-value


Table 2-6. ANOVA (t-test) results for specific genotype~concentration effects in the ten growth
experiments (Table 2-2). The table lists the results of differences in biomass or
chlorophyll between genotypes at the same concentration of MgSO4*7H20.


eConc.
whole plant


Conc. Genotyp~
#1: mrs2-10/col-0
0 col-0
4 col-0
12 col-0
20 col-0
28 col-0
#2: mrs2-10/col-0
0 col-0
20 col-0
60 col-0
100 col-0
#3: mrs2-10/col-0
0 col-0
20 col-0
60 col-0
100 col-0


Genotype Diff. Estimate StdErr DF
FW biomass comparison on agar medium


0 mrs2-10 0.516
4 mrs2-10 -1.4947
12 mrs2-10 -0.41
20 mrs2-10 -0.3728
28 mrs2-10 -0.2494
shoot FW biomass comparison on soil
0 mrs2-10 0.6458
20 mrs2-10 0.07167
60 mrs2-10 -0.1983
100 mrs2-10 0.07142
shoot chlorophyll comparison on soil n
0 mrs2-10 -0.4111
20 mrs2-10 -0.01528
60 mrs2-10 1.0528
100 mrs2-10 0.7072


sell-10/lws whole plant FW biomass comparison on
sell-10 0 ws -5.4527
sell-10 4 ws -3.9427
sell-10 12 ws -2.0702
sell-10 20 ws -0.717
sell-10 28 ws -0.04433


ia


0.5915
0.5915
0.5915
0.5915
0.5915
medium
0.6337
0.6337
0.6337
0.6337
medium
0.8282
0.8282
0.8282
0.846
gar medium
0.5912
0.5912
0.5912
0.5912
0.5912


0.87
-2.53
-0.69
-0.63
-0.42

1.02
0.11
-0.31
0.11

-0.5
-0.02
1.27
0.84

-9.22
-6.67
-3.5
-1.21
-0.07

-0.76
-0.55
-0.35
-0.13

0.47
0.57
1.16
-1.04

-1.69
1.01
1.99
2.44

-0.28
1.88
5.73
7.55

-19.86
0.99
1.01


0.3837
0.012*
0.4887
0.529
0.6735

0.311
0.9102
0.755
0.9105

0.6198
0.9853
0.2042
0.4035

<.0001*
<.0001*
0.0005*
0.2262
0.9403

0.4521
0.5854
0.7295
0.8985

0.6397
0.5663
0.2466
0.297

0.0941
0.3146
0.0495*
0.0165*

0.7783
0.061
<.0001*
<.0001*

<.0001*
0.3259
0.3164

<.0001*
<.0001*
<.0001*


#5:
0
20
60
100
#6:


sell-10/ws shoot FW biomass comparison on soil medium
sell-10 0 ws -0.2667 0
sell-10 20 ws -0.1933 0
sell-10 60 ws -0.1225 0
sell-10 100 ws -0.04517 0
sell-10/lws shoot chlorophyll comparison on soil medium


0


).3531
).3531
).3531
).3531

).685


0 ws
20 ws
60 ws
100 ws


sell-10
sell-10
sell-10
sepll-70


0.3208


0.3931 0.685
0.7944 0.685
-0.8181 0.7837
mparison on soil medium
-0.7908 0.4671
0.4725 0.4671
0.9305 0.4671
1.142 0.4671
prison on soil medium


#7:
0
60
80
100
#8:
0
60
80
100
#9;
0
80
100
#10
0
80
100


caxl-1/col-0 shoot FW biomass con
caxl-1 0 col-0
caxl-1 60 col-0
caxl-1 80 col-0
caxl-1 100 col-0
caxl-1/col-0 shoot chlorophyll com
caxl-1 0 col-0
caxl-1 60 col-0
caxl-1 80 col-0
caxl-1 100 col-0
caxl cax3/col-0 shoot FW biomass
caxl-1 0 col-0
caxl-1 80 col-0
caxl-1 100 col-0
: caxl cax3/col-0 shoot chlorophyll
caxl-1 0 col-0
caxl-1 80 col-0
caxl-1 100 col-0


















Figure 2-1. Mrs2-10 homozygous knockout mutant line. A homozygous T-DNA insertion mutant
for AtMRS2-10 (Atlg80900) was identified by PCR based on the
SALK 100361.41.30.x line.


col-0 mrs2-10


M R S2-10


TUB


_r


Figure 2-2. RT-PCR analysis of the mrs2-10 mutant. RNA extracted from mrs2-10 and col-0
leaves was tested for the presence of MRS2-10 mRNA with gene-specific primers. a-
Tubulin (TUB) was used as a constitutive control.


Figure 2-3. Two sets of six mrs2-10 plants growing at the Purdue lonomics proj ect facility.


SALK10061.41.30.x
MRS2-0_RPAt (RS2-10 (Atig80900)


MRS2-10_LP


cg~i











tl(SALK,100361,homo AT1C80900)







.Da h ~~..~. .~ .1 -,
;5
X .4 `C. ,
, c-
,


Elements


oiE ill-; Clrji -irji ei-


Figure 2-4. Element z-score values for the first set of six mrs2-10 plants. Each individual plant is
represented by a different color. The z score for an element indicates how far and in
what direction the element deviates from its distribution's mean, expressed in units of
its distribution's standard deviation.



11(SALK_100361.~homo AT1G80900)


a ~-.J~J`\/
'
r" Y ;S



i.i: Sii liJI': h(g:5 iii 131 13" CaJ' iAI:55 F~S;
Elements


r,_cc_.
7i'~ ~-C
L ''
`~--I._
v


V1
S
m


5rB1 PtBS )nOL19 Cdl


Figure 2-5. Element z-score values for the second set of six mrs2-10 plants. Each individual
plant is represented by a different color. The z score for an element indicates how far
and in what direction the element deviates from its distribution's mean, expressed in
units of its distribution's standard deviation.











col-O mrs2-10


m~rv2-l rol-0


we rP-ll-1 .oPll-ID we


0 mM








4 mM








12 mM








20 mM








28 mM








Figure 2.6. Growth experiments on agar. FW biomass of mutant lines is compared to that of their
respective wildtype backgrounds at different levels of MgSO4*7H20 in the agar
medium. Two Petri dishes per comparison per concentration are shown. Each dish
contains both mutant and wildtype plants, which are planted alternatively on the left
and the right side. The Petri dishes were scanned just prior to plant analysis at day 13.











Scax1-1 & col-0


sell-10 & wvs


car/ cat3~ &; col-0


0 mM










60 mM










80 mM










100 mM


20 mM










60 mM










100 mM


I


Figure 2-7. Growth experiments on soil. FW shoot biomass of mutant lines is compared to that of
their respective wildtype backgrounds at different levels of MgSO4*7H20O in the soil
medium. One flat per comparison per concentration is shown. For the comparison
between caxl cax3 and col-0, 60 mM was not tested. Yellow boxes highlight the
sections with mutant plants within each flat; the other sections contain wildtype
plants. The plants shown here are 3 weeks old and were analyzed after 4 weeks of
growth.


mrs2-10 & col-0

0 mMI ~iB

























0 1 : s s "
0 4 12 20 28

MgSO4*7H20 (mM)
Figure 2-8. Average fresh weight biomass of mrs2-10 and col-0 seedlings in response to
increasing concentrations of MgSO4*7H20 in agar medium. Bars indicate standard
error, n = 30. The asterisk indicates a statistically significant difference between
mrs2-10 and col-0 (p < 0.05) at 4 mM MgSO4*7H20 based on ANOVA.


O 20 60


MgSO4*7H20 (mM)

Figure 2-9. Average fresh weight shoot biomass of mrs2-10 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium. Bars indicate standard
error, n = 12. No statistically significant differences between mrs2-10 and col-0 at
any of the concentrations are observed based on ANOVA.


























0 20 60 100


MgSO4*7H-20 (mM)

Figure 2-10. Average leaf chlorophyll content of mrs2-10 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium. Bars indicate standard
error, n = 72. No statistically significant differences between mrs2-10 and col-0 at
any of the concentrations are observed based on ANOVA.


O 4 12 20 28


MgSO4*7H20 (mM)

Figure 2-11i. Average fresh weight biomass of sell-10O and ws seedlings in response to increasing
concentrations of MgSO4*7H20 in agar medium. Bars indicate standard error, n = 30.
Asterisks indicate statistically significant differences between sell-10 and ws (p <
0.05) at 0, 4 and 12 mM MgSO4*7H20 based on ANOVA.





























0 20 60 100


MgSO4*7H20 (mM)


Figure 2-12. Average fresh weight shoot biomass of sell-10 and ws plants in response to
increasing concentrations of MgSO4*7H20 in soil medium. Bars indicate standard
error, n = 12. No statistically significant differences between sell-10 and ws at any of
the concentrations are observed based on ANOVA.


50 -
45 -
40 -
35 -
30 -
25-

25 -
15-

20 -
5-
0-


.


20


.


100


MgSO4*7H20 (mM)

Figure 2-13. Average leaf chlorophyll content of sell-10 and ws plants in response to increasing
concentrations of MgSO4*7H20 in soil medium. Bars indicate standard error, n = 72.
No statistically significant differences between sell-10 and ws at any of the
concentrations are observed based on ANOVA.


I ws
I Ose~l-l-0











Scol-0
0 cax1-1


I


MgSO4*7H[20 (mM1)
Figure 2-14. Average fresh weight shoot biomass of cax1-1 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium. Bars indicate standard
error, n = 12. Asterisks indicate statistically significant differences between caxl-1
and col-0 (p < 0.05) at 80 and 100 mM MgSO4*7H20 based on ANOVA.


O 60 80 100


MgSO4*7H20 (mM)

Figure 2-15. Average leaf chlorophyll content of caxl-1 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium. Bars indicate standard
error, n = 72. Asterisks indicate statistically significant differences between caxl-1
and col-0 (p < 0.05) at 80 and 100 mM MgSO4*7H20 based on ANOVA.


* e





II col-0
0 caxl/cax3


MgSO4*7H20 (mM)

Figure 2-16. Average fresh weight shoot biomass of caxl/cax3 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium. Bars indicate standard
error, n = 12. The Asterisk indicates a statistically significant difference between
caxl/cax3 and col-0 (p < 0.05) at 0 mM MgSO4*7H20 based on ANOVA.


O 80 100


MgSO4*7H20 (mM)

Figure 2-17. Average leaf chlorophyll content of caxl/cax3 and col-0 plants in response to
increasing concentrations of MgSO4*7H20 in soil medium. Bars indicate standard
error, n = 72. Asterisks indicate statistically significant differences between caxl/cax3
and col-0 (p < 0.05) at 0, 80 and 100 mM MgSO4*7H20 based on ANOVA.


I









CHAPTER 3
TRANSCRIPTOME RESPONSES OF COL-0 AND CAX1-1 ARABIDOPSIS TO EXCESS
LEVEL S OF MAGNE SIUM SULFATE; CONSEQUENCES FOR (EXTRA)TERRE TRIAL
PLANT GROWTH

Introduction

Manned missions to Mars demand the efficient use of local planetary resources and the

recycling of limited materials such as water, pressurized atmosphere and organic matter while

producing food (Barta and Henninger, 1994). The use of in situ regolith for plant growth in a

future bioregenerative life support system on Mars may have several advantages over hydroponic

systems (Schuerger et al., 2002). These include the immediate bioavailability of plant essential

ions, low-tech mechanical support for plants, and easy access of in situ materials once on the

surface. However, plant growth may be reduced or inhibited by substances in the regolith, such

as high levels of hydrated magnesium sulfate minerals (Chapter 2). In a potential bioregenerative

life support system on Mars, an excess of a particular element in the crew' s diet could affect the

presence and availability of other required elements. This study therefore focuses on the

possibility of reducing accumulation of Mg2+ and SO42- ions within the plant as a method to

enhance plant tolerance to high levels of magnesium sulfate in the growth medium.

Arabidopsis is a model species in plant molecular biology research and its genome is fully

sequenced. Plasma membrane localized efflux transporters of Mg2+ and SO42- ions have not been

identified to date in the outer root cell layers of Arabidopsis or other plant species. AtMRS2-10

and AtSULTR1;2 are genes encoding a known Mg2+ and SO42- uptake transporter respectively.

The transporters are localized to the plasma membrane in root cells. Arabidopsis lines carrying

knockout T-DNA insertion mutations in AtMRS2-10 and AtSULTR1;2 did not mitigate the

constraining impacts of high magnesium sulfate concentrations on wildtype Arabidopsis plants

(Chapter 2). An Arabidopsis line carrying a knockout mutation of the vacuolar CAX1 gene










(cax1-1) showed a significant improvement in growth on soil treated with high levels of

MgSO4*7H20 in solution (Chapter 2). Although the Mg content of caxl mutant roots has not

been analyzed so far, the reduced levels of Mg in leaves (Bradshaw, 2005) indicate that

Arabidopsis CAX1 knockout mutants show improved tolerance of high magnesium sulfate by at

least partly limiting accumulation of Mg2+ Or SO42- ions within the plant.

Genome-wide analysis of Arabidopsis root transcriptome responses to elevated levels of

MgSO4*7H20 in solution could reveal what molecular processes may be involved in adaptation

to this stress, and could lead to identification of candidate genes that may play a role in

enhancing tolerance by reducing accumulation within the plant. No genome-wide transcriptome

profiling studies have been reported to date that address responses of plant species to elevated

concentrations of Mg2+ Or SO42- in the growth medium. Maruyama-Nakashita et al. (2003)

analyzed the transcriptome response of ws and sell-10 plants to sulfur deficiency stress. The

study provided a selection of -S-responsive genes related to sulfate uptake, sulfur assimilation,

remobilization of secondary sulfur metabolites, and mitigation of oxidative stress (Maruyama-

Nakashita et al., 2003). Some of the previous studies analyzing root transcriptome responses to

ion toxicities, such as responses to NaCl and heavy metals, are reviewed below. The focus for

this review is on the extent of the observed transcriptome responses, as well as the detection of

differentially expressed transporter genes, in relation to the treatment concentration and duration.

In addition, transcriptome analyses are reviewed that involve comparisons between wildtype and

transgenic plants, or between related species showing differences in tolerance to a certain ion

stress.

Maathuis et al. (2003) analyzed the Arabidopsis (col-0) root transcriptome in response to

high NaCl (80 mM) stress over a 96-h period (2, 5, 10, 24 and 96 hrs) with oligonucleotide









microarrays representing 1096 Arabidopsis transporter genes. In general, NaCl treatment did not

affect specific gene families but rather resulted in a collective transcriptional response by

affecting specific isoforms across transporter families. For example, aside from known Na+

transporters such as the vacuolar Na/H' antiporter AtNHX1, NaCl also affected the expression

of particular genes associated with transport of other nutrients such as nitrate (Maathuis et al.,

2003). Jiang et al. (2006) used microarrays with oligonucleotide probes representing 23,686

Arabidopsis genes to identify col-0 root transcripts that changed in relative abundance following

6 h, 24 h, or 48 h of hydroponic exposure to 150 mM NaCl compared to control conditions.

Among the statistically significant expression differences, 2,433 unique genes showed a NaCl-

induced increase in transcript abundance of at least 2.0 fold at one or more time points, while

2,774 unique genes showed a decrease in transcript abundance of at least 2.0 fold or more (Jiang

and Deyholos, 2006). The number of regulated genes found in this study highlights the

complexity of the response in plants exposed to 150 mM NaCl for these periods of time. Taji et

al. (2004) analyzed the gene expression profiles in Arabidopsis compared to its halophytic

relative Thellungiella halophile (salt cress) by using full-length Arabidopsis cDNA microarray

containing approximately 7000 genes. In contrast to Arabidopsis, only a few genes were induced

by treatment with 250 mM NaCl for 2 hrs in salt cress. Various genes induced by abiotic and

biotic stresses in Arabidopsis were shown to be constitutively expressed in salt cress even under

normal growth conditions (Taji et al., 2004). In another study addressing genotypic differences,

Sottosanto et al. (2004) used Affymetrix ATH1 arrays with probes representing 22,746 genes to

compare transcription profiles between a T-DNA insertion mutant of AtNHX1 (nhx1) and a

'rescued' line (NHX1::nhx1), which were both exposed to 100 mM NaCl for 12 and 48 hrs, as

well as 1 and 2 wks. 147 transcripts showed both salt responsiveness and a significant influence









of AtNHX1, including signaling proteins, DNA binding proteins, components of the protein

processing and trafficking machinery, cell wall related proteins and those involved in sulfur

metabolism (Sottosanto et al., 2004).

Several studies analyzing plant transcriptome responses to toxic levels of metal ions have

been reported. Herbette et al. (2006) investigated the transcriptional regulation in response to

cadmium treatment (5 and 50 CLM for 2, 6 and 30 hrs) in both roots and leaves of hydroponically

grown Arabidopsis (ecotype Columbia), using a whole genome microarray containing 24,576

independent probe sets. The number of differentially expressed genes was lower in leaves than in

roots, and regulation of gene expression in response to Cd seems time-regulated rather than dose-

regulated. One of the main responses observed in roots was the induction of genes involved in

sulfur assimilation-reduction and glutathione (GSH) metabolism (Herbette et al., 2006). In

another study, van de Mortel et al. (2006) examined the root transcript profiles of Arabidopsis

and its heavy metal accumulating relative Thlaspi caerulescens grown for 7 days under deficient

(0 CIM), sufficient (2 or 100 CIM), and excess (25 or 1000 CIM) supply of zinc (ZnSO4), with the

main aim of establishing which genes are most likely to be relevant for adaptation to high zinc

exposure in 7: caerulescens. Results emphasized the role of previously implicated zinc

homeostasis genes in adaptation to high zinc exposure, but also suggest a similar role for many

more, as yet uncharacterized genes, often without any known function (van de Mortel et al.,

2006).

This study is the first to document genome-wide plant root transcriptome responses to

elevated levels of magnesium sulfate using microarrays. The obj ective is to analyze which genes

are differentially expressed as part of the primary stress response in roots of a non-tolerant

species such as wildtype Arabidopsis (col-0) compared to unexposed col-0 roots. This could lead









to identification of candidate genes in Arabidopsis with potential to enhance tolerance to high

magnesium sulfate by limiting accumulation within the plant. Potential candidate genes are those

encoding root plasma membrane localized import or efflux systems, or regulators of plasma

membrane transporter activity. In order to capture part of the primary stress response, three early

time periods (45 min. 90 min. and 3 hrs) of col-0 exposure to a non-lethal concentration of high

MgSO4*7H20 in a hydroponic growth medium were chosen. In addition, a set of col-0 plants

was exposed to a control solution for 45 minutes. The treatment concentration of MgSO4*7H20

was based on the high Mg:Ca ratio that can occur in serpentine soils on Earth. Serpentine soils

can be a partial analogue for regolith high in magnesium sulfate on Mars because of their high

amount of bioavailable magnesium (Chapter 2). The caxl-1 mutant was also exposed to elevated

MgSO4*7H20 for 3 hours to determine which genes are differentially expressed in the CAX1

knockout mutant background compared to exposed col-0 at this time. Genes that are

differentially expressed between the genotypes could point to some of the downstream molecular

processes eventually leading to enhanced tolerance for caxl-1 at the whole plant level, including

reduced leaf Mg content and increased fresh weight biomass, after days or weeks of exposure in

agar or soil medium (Chapter 2). Some of the genes involved in downstream processes may

themselves be candidate genes for enhanced tolerance, such as those encoding (regulators of)

plasma membrane based channels that transport Mg2+

Results

A microarray experiment was conducted with conditions described in Table 3-1, Figures 3-

1, 3-2, 3-3, and the Materials and Methods section. Sample sets were compared using a series of

t-tests; the col-0 time 1, col-0 time 2 and col-0 time 3 treatment sets were compared to the col-0

time 1 control sample set, and these comparisons are referred to in the text, tables and figures as

Time 1, Time 2 and Time 3 respectively. The caxl-1 time 3 treatment set was compared to the









col-0 time 3 treatment set; this comparison is not referred to in abbreviated form. The resulting

gene expression data for each comparison are presented as volcano plots (Figures 3-4, 3-5, 3-6,

3-7). The volcano plots show the log2 values of the fold changes in gene expression between two

compared sample sets on the x-axes. The y-axes show the -logl0 p-values corresponding to the

log2 fold change values. Expression differences (fold change) corresponding to p-values for

which q < 0.05 are considered significant and are indicated in red. The number of genes with a

significant difference in expression between treatment and control sets increases from 325 for

Time 1, to 1516 for Time 2, and 3265 for Time 3. The number of genes with a significant

difference in expression over 2 fold increases accordingly, from 100 for Time 1, to 248 for Time

2, and 445 for Time 3. Between the caxl-1 time 3 treatment and col-0 time 3 treatment sample

sets, only 4 transcripts show a significant difference in expression, all over 2 fold. Genes with

significant expression differences at Time 1, 2 and 3 were annotated according to Gene Ontology

(GO) terminology. The genes were grouped per GO functional category for each comparison,

and the resulting numbers for the three comparisons are listed in Table 3-2. The results show that

the number of genes per functional category is increasing with time for each category. The

percentage of genes in each functional category changes slightly across the Time 1, 2 and 3

comparisons; increasing in some categories and decreasing in others (Table 3-2).

Transcripts col-0 Time Series

Since the experimental design for the microarray experiment involving the col-0 time

series does not control for diurnally regulated genes at Time 2 and 3, we will focus on genes with

significant expression differences at Time 1, and genes with significant expression differences

that are shared between Time 1 and one or more of the other comparisons (Time 2 and 3). As

mentioned above, 325 genes show a significant difference in expression at Time 1. Of these 325

genes, 34 are differentially expressed at q < 0.001 (Table 3-3), while 36 are differentially









expressed over 3 fold (Table 3 -4). A Venn diagram was made based on a comparison of genes

with significant expression differences at Time 1, 2 and 3. The results show that of the 325 genes

expressed differentially at Time 1, 74 genes are unique to Time 1, 48 are shared between 1 and 2,

155 genes are shared between all three comparisons, and 48 are shared between 1 and 3 (Figure

3-8). The 325 differentially expressed genes at Time I were also compared with a cluster of 197

genes identified by Ma and Bohnert (2007) that are differentially expressed in response to broad

range of diverse stress conditions: cold, osmotic, salinity, wounding, and biotic stresses

(including treatments with elicitors). The comparison showed that at least 18 of the 325 genes

differentially expressed at Time 1 are universally responsive to stress conditions. A hierarchical

average linkage cluster analysis using uncentered correlation was done across Time 1, 2 and 3

based on the genes with significant expression differences at Time 1 (Figure 3-9). The results

show the exact expression patterns of the genes that are shared with Time 2, Time 3, or both.

Groups of functionally related genes

The 325 genes that showed a statistically significant difference in expression at Time 1 can

be grouped into functional categories according to GO annotations (Table 3-2). The overview

shows that the molecular function of a large portion of the genes is currently unknown. A closer

look at the genes whose molecular function is known or putatively known reveals the following

(Figure 3-9a-d, Object 3-2). Within the group genes encoding transcription factors, multiple

AP2/EREBF, bHLH, heat shock, MYB (KAN4, MYB50), NAC (NAM) TCP (PCF l, PCF2) and

zinc finger (C2H2 and C3HC4 type) transcription factors, as well as a MADS box (AGL45),

WRKY (WRKY70), and scarecrow transcription factor can be recognized. Among the genes

encoding proteins with kinase activity, several protein kinases, lectin protein kinases, and

leucine-rich repeat transmembrane protein kinases can be distinguished. Three genes encoding

cyclins, which regulate the activity of cyclin-dependent kinases, are differentially expressed at









Time 1 and 2, but not at Time 3. Regarding phosphatase activity, genes encoding protein

phosphatase 2C (PP2C) family proteins are well represented. Within the group of genes encoding

membrane based transporters, those encoding cyclic nucleotide-gated channels CNGC 19 and

CNGC 1 are differentially expressed at the early time points. The expression of two genes

encoding potassium transporters is up-regulated, with the expression of HAK5 being up-

regulated across the time points. Transcripts encoding calcium-transporting ATP-ase ACAl2 and

calcium exchanger CAX1 are differentially expressed at Time 1 and beyond. Other genes

encoding transporters that show differential expression are amino-acid transporters, an organic

cation transporter, and inorganic phosphate transporter PHT1i. Among transcripts encoding

proteins with ion binding activity, those encoding calcium-binding proteins, and in particular

calmodulins, are well represented at Time 1 and further on. The expression of genes encoding

calcium-binding copines and copper-binding plastocyanins is down-regulated. A few transcripts

encoding proteins responsive to inorganic ions such as phosphate and nitrate can also be

distinguished. Genes encoding hormone-responsive proteins are dominated by those encoding

auxin-responsive proteins. The expression of these genes is up-regulated at Time 1, and for many

of the genes this is the same at Time 2 and 3. A large group of disease resistance proteins (TIR-

NBS-LRR or variations thereof) shows that the expression of the genes encoding these is

generally down-regulated across the time points. Genes encoding cell wall related proteins such

as (fasciclin-like) arabinogalactans, xyloglucan endotransglycosylases, exotosins, putative lipid

transfer proteins, beta-expansins, a pectin acetylesterase and a hydroxyproline-rich glycoprotein

all show down-regulated expression, while multiple genes encoding alpha-expansins and

polygalacturonases/pectinases all show up-regulated expression. The expression of two

transcripts encoding anionic peroxidases and of one that encodes a glutathione S-transferase is










up-regulated, while that of two transcripts encoding SEC 14/phosphoglyceride transfer family

proteins is down-regulated. Other differentially expressed genes encoding metabolic enzymes

include those that encode glycoside and glycosyl hydrolases, glycosyl transferases. UDP-

glucosyl transferases, beta-amylases and nudix hydrolases.

Remaining genes that encode metabolic enzymes that do not directly group together were

analyzed by overlaying the significant gene expression ratios per time point on a metabolic map

by using the Aracyc Omics viewer of the Arabidosis Information Center (TAIR). Genes

encoding 1 -aminocyclopropane-1l-carboxylate synthase in the ethylene biosynthesis pathway

show up-regulated expression across the three time points (Figure 3-10a). A transcript encoding

a 9-cis-epoxycarotenoid dioxygenase/neoxanthin cleavage enzyme in the abscisic biosynthesis

pathway shows up-regulated expression in all comparisons as well (Figure 3-10b). In the

jasmonate biosynthesis pathway, the expression of a lipoxygenase encoding gene is up-regulated

at Time 1, and several genes encoding 12-oxophytodienoate reductases show down-regulated

expression at Time 3 (Figure 3-10c). Transcripts encoding gibberellin 2-oxidases 7 and 8 show

up-regulated expression at the time points, while a transcript encoding gibberellin 2-oxidase 1

shows down-regulated expression at Time 3 (Figure 3-10d). A gene encoding

monogalactosyldiacylglycerol synthase 2 shows down-regulated expression across the time

points as part of glycerolipid metabolism (Figure 3-10e). A transcript encoding the rate-limiting

enzyme glucosamine-fructose-6-phosphate aminotransferase shows up-regulated expression at

Time 1 and 3 in the hexosamine biosynthetic pathway (Figure 3-10f). A gene encoding a Rieske

[2Fe-2S] domain-containing protein similar to Pheophorbide A oxygenase shows up-regulated

expression at Time 1 and Time 3. Pheophorbide A oxygenase is part of the chlorophyll

breakdown pathway (Figure 3-10g). A Glutamyl-tRNA reductase 2 encoding gene shows up-










regulated expression at Time 1 and 2, and a sirohydrochlorin ferrochelatase encoding gene shows

up-regulated expression at Time 2 and 3, as part of porphyrin and chlorophyll metabolism

(Figure 3-10h). A transcript encoding phosphoribosyl-ATP pyrophosphohydrolase shows up-

regulated expression across the time points as part of the histidine biosynthesis pathway (Figure

3-10i). A gene encoding UDP-glucose 4-epimerase shows down-regulated expression at Time 1

and 2 as part of galactose metabolism (Figure 3-10j), and phosphofructokinase encoding genes

show up-regulated expression at Time 1 and 3 as part of the glycolysis pathway (Figure 3-10k).

A transcript encoding cytosolic malate dehydrogenase shows up-regulated expression at Time 1,

while a transcript encoding mitochondrial malate dehydrogenase shows down-regulated

expression at Time 3, and a transcript encoding malate synthase shows up-regulated expression

at Time 2 and 3 (Figure 3-101). Finally, genes encoding proteins grouped by their structural

motifs, such as Armadillo, F-box and U-box family proteins, as well as DC1 domain containing

proteins, are represented by multiple members at Time 1, and some of these are shared with

Time 2 and 3. The role of functional groups as a whole and of individual gene members within

the various functional groups in the response of Arabidopsis roots to high levels of magnesium

sulfate will where possible be addressed in the discussion.

Transcripts Encoding Membrane Based Transporters in the Col-0 Time Series

Because of our interest in identifying plant variants with enhanced tolerance to elevated

levels of magnesium sulfate and the possibility of differential expression of transporter genes as

possible reasons for such enhanced tolerance, the responses of genes encoding membrane based

transporters were documented not only at Time 1, but also across the other time points. Since

gene expression differences are not fully controlled for diurnal effects for the Time 2 and 3

comparisons in the col-0 time series, these effects will have to be ruled out in follow-up studies.

A hierarchical average linkage cluster analysis was done based on transporter genes with









significant expression differences at Time 1, 2 or 3 using uncentered correlation. The results

show that distinct clusters of expression patterns can be distinguished within the group of

transporter genes across the three time points. For example, some transporter genes are

differentially expressed from Time 1 or 2 onwards, while most are not differentially expressed

until Time 3. A few are differentially expressed at Time 1 and 2 but not 3, Time 2 only, or at

Time 1 and 3 but not 2 (Figure 3-11).

Groups of functionally related transporters

When grouping the known transporter genes according to the type of molecules they

transport, the following can be observed (Figure 3-1 1, Obj ect 3-2). Genes encoding ion

transporters such as magnesium transporter MRS2-10 (Atlg80900) and Mg2+ H+ exchanger

MHX1 show up-regulated expression at Time 2 and 3, while the expression of MRS2-7

(At~g09690) is up-regulated at Time 3. Sulfate transporter SULTR3;4 (At3gl5990) shows

down-regulated expression from Time 2 onwards, while SULTR3;1 (At3g51895) and SULTR4;1

(At~gl 3550) show down-regulated expression at Time 3. The expression of the gene encoding

Ca2+ H+ antiporter CAX1 is down-regulated almost two-fold starting at Time 1, while CAX2 and

CAX3 begin to show down-regulated expression at Time 2. The gene encoding cation exchanger

CHX20 is on the other hand up-regulated in expression from Time 2 onwards.

Genes encoding the high affinity nitrate transporter ACH1, the ammonium transporter

AMT1.1, and the inorganic phosphate transporters PHT1 and PHT2 all show up-regulated

expression, while those encoding a putative inorganic phosphate transporter (At2g29650),

phosphate transporter PT2, and the putative mitochondrial phosphate transporter (At3g48850)

show down-regulated expression. Transcripts encoding metal transporters NRAMP1 and ZIP5,

as well as a ZIP Zinc transporter domain containing protein (Atlg68100), an iron transporter-

related protein (At2g3 8460), a ferroportin-related protein (At~g26820), and a metal transporter









family protein (At3g08650), all show up-regulated expression. The expression of four transcripts

encoding potassium transporters is up-regulated at different time points, with that of HAK5

consistently and increasingly across the three comparisons. The expression of potassium channel

encoding genes AKT 1 and AKT4 is up-regulated, while that of AKT2 is greatly down-regulated.

Transcripts encoding chlorine channel proteins CLC-a and CLC-b show down-regulated

expression, while a transcript encoding a putative chloride channel-like protein (At5g33280)

shows up-regulated expression. The expression of three genes encoding putative cation-chloride

cotransporters is up-regulated, while that of a gene encoding K-Cl cotransporter type 1 protein-

related (At3g58370) is down-regulated. Two transcripts encoding cation efflux family proteins

are differentially expressed in contrasting ways, and the expression of a transcript encoding an

anion exchange family protein (At3g62270) is down-regulated.

Genes encoding ion-transporting ATP-ases, such as calcium-transporting ATP-ases, are

well represented among the differentially expressed transporter genes. Genes encoding ACA1,

ACA9, ACA10 and ACAl2, as well as ECA2, all show down-regulated expression, while those

encoding ACA2, ACA4 and ACA8 show up-regulated expression. The expression of two

transcripts encoding E1-E2 type ATPase family proteins is up-regulated, while two transcripts

encoding copper exporting ATP-ases show contrasting expression patterns. The expression of

genes encoding phospholipid-transporting ATPase 1/magnesium-ATPase 1 (ALAl) and an

anion-transporting ATPase family protein (At5g60730) is down-regulated from Time 2 onwards.

Three transcripts encoding putative proton pump ATP-ases show up-regulated expression at

Time 3.

Genes encoding cyclic nucleotide gated ion channels CNGC1 and CNGC19 show

differential expression at the early time points only, while those encoding CNGC10, CNGC13









and CNGC15 are differentially expressed at Time 2 or later. Two transcripts encoding

mechanosensitive ion channel domain-containing proteins show up-regulated expression. A

group of genes encoding maj or intrinsic family proteins (MIP) show down-regulated expression

at Time 3, with the exception of At4g0 1470, which shows up-regulated expression. The

expression of multiple genes encoding plasma membrane intrinsic proteins (PIP lB, IC, 2A, 2C),

and two tonoplast intrinsic proteins (TIP3.1) is down-regulated. Several transcript encoding

mitochondrial import (Timl7/Tim22/Tim23) family proteins show up-regulated expression at

Time 3, except for At3gl01 10, which shows down-regulated expression. A number of genes

encoding mitochondrial substrate carrier family proteins are also differentially expressed, with

At4g24570 showing up-regulated expression across all comparisons. At4g24570 is shown to be

universally responsive to stress (Ma and Bohnert, 2007). The expression of a group of transcripts

encoding nodulin MtN21 family and nodulin-related proteins is up-regulated at Time 3, again

with an exception in the form of the expression of At4g3 0420.

With respect to transporters of specific metabolic products, two genes encoding auxin

efflux carrier proteins show down-regulated expression, while a gene encoding auxin transport

protein PIN3 shows up-regulated expression. A transcript encoding a C4-dicarboxylate

transporter/malic acid transport family protein (At4g27970) shows down-regulated expression at

Time 2. The expression of a gene encoding an organic cation transporter-related protein

(Atlgl6390) is greatly down-regulated across all time points. Transcripts encoding glutathione

S-conjugate ABC transporters MRP1 and MRP2, and a putative glycerol-3-phosphate

transporter/permease (At3g47420) all show down-regulated expression at Time 3. The

expression of a gene encoding a putative lysine and histidine specific transporter (At3g0 1760) is

down-regulated across the three time points, while that of Atlg25530 is up-regulated from Time









2 onwards. A transcript encoding an UDP-galactose/UDP-glucose transporter shows up-

regulated expression at Time 3, while transcripts encoding the sucrose transporters/sucrose-

proton symporters SUC1, SUC3 and SUC5 all show down-regulated expression. Finally,

transporter gene families represented by multiple members with little annotation, and with mixed

differential expression patterns amongst the members, are the amino acid transporter family, the

proton-dependent oligopeptide transport family, the sugar transporter family, the MATE efflux

family, the ABC transporter family and the integral membrane family.

Q-PCR confirmation

The expression of four transporter genes that showed a significant difference in expression

at Time 3 was analyzed by quantitative PCR. Time 3 refers to the microarray comparison of the

col-0 time 3 treatment sample set with the col-0 time 1 control sample set (Table 3-1). Because

the control sample set is harvested at time 1, the Time 2 and Time 3 comparisons are not

controlled for diurnal effects. In order to confirm that the expression differences observed at

Time 2 or 3 for several transporter genes of relevance to this study are not due to diurnal effects,

The treatment at time 3 was repeated with a different set of hydroponically grown col-0 plants

and a control set was harvested at time 3 as well (Figure 3-2 D). In addition to confirming their

expression using the same RNA sources as those used for the microarray experiment, RNA was

used from these sets of independently grown plants. The genes analyzed were CAX1, MRS2-10,

SULTR3;4 and NRAMPl. CAX1 and MRS2-10 were chosen to be able to relate the microarray

results for these genes to the knockout mutant results of Chapter 2. SULTR3;4 was selected

because it represents a sulfate transporter gene not previously reported to be regulated by sulfur

levels. NRAMP1 was included because it encodes a transporter of ions other than Mg2+, SO42- Of

Ca2+. The differences in gene expression between treatment and control as measured by Q-PCR

are most similar to the results achieved with the microarray experiment when using the same









RNA sources (Tables 3-5a, 3-5b, 3-6, 3-7). When using RNA from the independently grown and

diurnally controlled plants for the Q-PCR analysis, the expression differences are similar to the

microarray results for CAX1 and NRAMP 1, less pronounced for SULTR3;4, while MRS2-10

shows no difference at all (Tables 3-6, 3-7, Figures 3-12, 3-13, 3-14). The discrepancy between

the Q-PCR and the microarray results for SULTR3;4 and MRS2-10 may be due to diurnal

regulation of the expression of these transcripts between time 1 and time 3 in the microarray

analysis. Despite this, the gene expression difference of SULTR3;4 between treatment and

control at time 3 is still significant during one of the Q-PCR analysis repetitions. The small

difference in expression of MRS2-10 between treatment and control does not qualify as

significant in any of the Q-PCR analyses, even when using the same RNA sources as those used

for the microarray experiment, which may indicate a limitation of the SYBR green dye based Q-

PCR analysis.

Transcripts Caxl-1/Col-0 Comparison

Only 4 transcripts showed a significant difference in expression between the caxl-1 time 3

treatment and col-0 time 3 treatment sample sets (Table 3-8). Among these, the transcript with

the smallest q-values and largest differences in expression is CAX1, the gene that is knocked out

in the mutant caxl-1, and which is represented on the microarray by two different probes. The

other three transcripts are At3g01345, an expressed protein with similarity to beta-galactosidases

in Arabidopsis, At4g07526, an unknown protein with similarity to other unknown proteins in

Arabidopsis, and chromosomal region CHR2:011819877-0118 19818, with no significant

similarity to other nucleotide sequences.

Supplemental Files

Obj ect 3-1. Microarray raw signal data (.xls Eile 3 5 MB)

Object 3-2. Microarray analysis data (.xls Eile 1 MB)









Discussion

The obj ective of the current study was to quantify the early transcriptome responses of

wildtype Arabidopsis (col-0) roots to elevated MgSO4*7H20 and to possibly identify candidate

genes that play a role in enhancing tolerance to MgSO4*7H20 by limiting accumulation within

the plant. Our results showed 325 differentially expressing genes in col-0 after 45 minutes of

exposure compared to col-0 unexposed for 45 minutes (Time 1) that could be organized into

functionally related groups and many of which were shared with either Time 2 or 3, or with both.

In addition we described the transporter genes differentially expressed at all three time points.

Finally, we showed that only three transcripts are differentially expressed between caxl-1 and

col-0. In this discussion we will attempt to clarify how the results might relate to the treatment

given.

Transcripts Col-0 Time Series

Phytohormones like abscisic acid (ABA), ethylene, salicylic acid and j asmonates, and

other signalling molecules such as nutrients and intracellular second messengers phospholipidss,

Ca2+ and reactive oxygen species), have been implicated in mediating cellular events associated

with the response of the plant to abiotic stress (Magnan et al., 2008). Genes encoding enzymes

involved in the ABA, ethylene and jasmonic acid biosynthesis pathways show differential

expression at Time 1 and beyond. Transcripts encoding enzymes 9-cis-epoxycarotenoid

dioxygenase and aldehyde oxidase 3 in the ABA biosynthesis pathway show up-regulated

expression, pointing to increased ABA synthesis. Genes encoding 1 -aminocyclopropane- 1-

carboxylate synthases in the ethylene biosynthesis pathway show up-regulated expression across

the time points, while several genes encoding enzymes with 1 -aminocyclopropane-1 -carboxylate

oxidase activity are differentially expressed at Time 3, indicating possible changes in ethylene

synthesis. In the j asmonate biosynthesis pathway, a lipoxygenase encoding gene shows up-










regulated expression at Time 1, while a gene encoding a 12-oxophytodienoate reductase shows

down-regulated expression at Time 2 and 3, pointing to possible changes in jasmonate synthesis.

The expression of a gene encoding a flavin monooxygenase was up-regulated at Time 1. This

enzyme shows similarity to YUCCA, a flavin monooxygenase-like enzyme in Arabidopsis that

catalyzes hydroxylation of the amino group of tryptamine, a rate-limiting step in tryptophan-

dependent auxin biosynthesis and which causes elevated levels of free auxin in dominant mutant

plants (Zhao et al., 2001). Two Arabidopsis cytochrome P450s, CYP79B2 and CYP79B3, which

convert tryptophan (Trp) to indole-3-acetaldoxime (IAOx) in vitro, were also shown to be critical

enzymes in auxin biosynthesis in vivo (Zhao et al., 2002). The cytochrome P450 that is encoded

by a gene with up-regulated expression at Time 1 in this study is CYP94A1, which has not been

associated with auxin biosynthesis, but with fatty acid hydroxylation. Products of the

Auxin/Indole-3 -Acetic Acid (Aux/IAA) and auxin response factor (ARF) gene families can

function in a negative (auto)feedback loop in which particular combinations of ARF and

Aux/IAA proteins control specific auxin-mediated responses via regulation of gene expression,

such as for example the combination of IAAl9 and ARF7, which may constitute a negative

feedback loop to regulate differential growth responses of hypocotyls and lateral roots in

Arabidopsis (Tatematsu et al., 2004). Genes encoding IAAl and IAAl9 both show up-regulated

expression at Time 1 in this study, while the other genes encoding auxin-responsive proteins

(GH3, SAUR) at Time 1 also show up-regulated expression. These results indicate the

occurrence of auxin-mediated responses such as negative regulation of lateral root growth in the

presence of elevated levels of magnesium sulfate. The expression of several genes encoding

gibberellin 2 oxidases, which are involved in gibberrellin (GA) inactivation, is up-regulated at

the different time points, indicating a reduction in GA in this phase. One of these genes









(Atlg02400) is differentially expressed in response to a broad range of stresses (Ma and Bohnert,

2007). The involvement of auxin in the regulation of GA metabolism genes points at the

existence of a complex regulatory network (Frigerio et al., 2006).

A large number of genes in the Arabidopsis genome encode EF hand-containing Ca2+-

binding proteins, such as calmodulin(-like) and calcineurin B-like proteins, which can act as Ca2+

sensors that relay Ca2+ Signals through Ca2+-induced conformational changes that can affect

interactions with an abundance of target proteins and modulate target protein activity

(McCormack et al., 2005). Calmodulin-like 9 (CML9) is for example involved in salt stress

tolerance through its effect on the ABA-mediated pathways in Arabidopsis (Magnan et al.,

2008). The genes encoding Ca2+-binding proteins that are differentially expressed at Time 1 in

this study are not functionally associated to specific signaling pathways at this point. Three of the

differentially expressed genes (At2g41410, At2g46600 and At3gl0300) are universally

responsive to stress (Ma and Bohnert, 2007). Calcineurin B-like (CBL) proteins specifically

target CBL-interacting protein kinases (CIPKs) such as CIPK9, which is required for low-

potassium tolerance in Arabidopsis (Pandey et al., 2007). The expression of the gene encoding

CIPK9 is up-regulated at Time 1. Another gene with differential expression at Time 1 is

At2g30040, which encodes a protein kinase family protein and is known to show a response to a

broad range of stresses (Ma and Bohnert, 2007). The maj ority of transcripts encoding kinases

that are differentially expressed at Time 1 encode functionally unspecified protein kinases as

well as receptor-like protein kinases (RLKs) with extracytoplasmic leucine-rich repeats (LRRs)

or with an extracellular lectin-like domain. RLKs are membrane-spanning proteins with a

predicted signal sequence and a cytoplasmic kinase domain that have been implicated in a wide

range of signal transduction pathways (Fontes et al., 2004). Several genes encoding protein









phosphatase 2C family proteins, which are potentially involved in phosphorylation-mediated

signaling, are differentially expressed at Time 1. Protein degradation can also play a role in

signal transduction and transcriptional regulation. Genes related to protein degradation are also

differentially expressed at Time 1, such as those encoding S-phase kinase-associated protein 1

(SKPl) interacting F-box proteins (up-regulated), U-box proteins, Arm/U-box proteins, and an

ubiquitin family protein (down-regulated).

The exact function of many of the differentially expressed genes at Time 1 encoding

transcription factors is currently unknown, although some are associated with tolerance to other

abiotic stresses. Members belonging to the NAC (NAM) and bHLH transcription factor families

are coordinated in their up-regulated expression at Time 1. Several genes encoding C2H2 type

zinc finger family proteins were detected, which play a crucial role in many metabolic pathways

as well as in stress response and defense activation in plants. C2H2 type zinc finger proteins can

have a putative repression activity in the defense and stress response of plants, which is thought

to occur via their ethylene-responsive element-binding factor (ERF)-associated amphiphilic

repression (EAR) domain (Ciftci-Yilmaz and Mittler, 2008). An example of this is ZAT7, which

renders plants more tolerant to salinity stress when constitutively expressed, and which is

therefore thought to function in Arabidopsis as a suppressor of a repressor of defense responses

(Ciftci-Yilmaz et al., 2007). This study shows that the expression of the gene encoding ZAT7 is

up-regulated after 45 minutes of exposure to elevated levels of magnesium sulfate and that it is

increasingly up-regulated with time up to at least 3 hours after treatment initiation. The functions

of the other two differentially expressed genes at Time 1 encoding C2H2 type transcription

factors have not been determined yet. AP2/ERF superfamily transcription factors play significant

roles in regulating plant biotic and abiotic stress-responsive gene expression and can be divided









into several families, one of which is the ERF family with a single AP2/ERF domain (Zhang et

al., 2008). An ERF transcription factor of the DREB l/CBF subfamily was shown to bind DRE-

like motifs in the promoter of gibberellin 2 oxidase 7 (GA2ox7), which is an enzyme encoded by

a gene whose expression is up-regulated in Arabidopsis under high-salinity stress, and which

reduces endogenous GA levels with the result that growth is repressed for stress adaptation

(Magome et al., 2008). Some of the AP2/ERF transcription factors encoded by differentially

expressed genes at Time 1 may therefore be involved in the up-regulation of the expression of

genes encoding gibberellin 2 oxidases, such as GA2ox7, in this study. Another transcription

factor, WRKY70, functions as a negative regulator of developmental senescence and is involved

in plant defense signaling pathways (Ulker et al., 2007). In the current study, the expression of

the gene encoding WRKY70 is stably down-regulated across the three time points.

Several differentially expressed genes point at senescence-related responses in Arabidopsis

roots exposed to high levels of magnesium sulfate. Two genes encoding senescence-associated

proteins (At2g25690, At4g35985) show up-regulated expression at Time 1. Two genes encoding

hairpin induced (HIN1) proteins (At4g01410, At2g27080) which previously showed up-

regulated expression during leaf senescence in tobacco (Takahashi et al., 2004), also show up-

regulated expression at Time 1. At4g3 5985 and At2g27080 are shown to be responsive to a

broad ranges stress (Ma and Bohnert, 2007). A gene encoding a Rieske [2Fe-2S] domain-

containing protein similar to pheophorbide A oxygenase, which is part of the chlorophyll

breakdown pathway, shows up-regulated expression at Time 1 and 3. It is possible that it is

transported to the shoot in the form of RNA or protein, or that the gene product has a different

function in roots in the absence of chlorophyll.










A general pattern of down-regulated expression of genes encoding disease resistance

proteins was observed at Time 1. Consistent with the need for a rapid response to pathogen

attack, many NBS-LRR-encoding genes are constitutively expressed at low levels in healthy,

unchallenged tissue, and little is known about the regulation of these genes (McHale et al.,

2006). Copines such as BON1 and BONZAll, which are calcium-dependent membrane-binding

proteins, have been implicated in the repression of several disease resistance genes (Yang et al.,

2006). In this study, the expression of BON1 and BONZAll-related genes is down-regulated,

while the expression of the maj ority of disease resistance genes is not seen to be up-regulated at

Time 1. At later time points the expression of the maj ority of disease resistance genes is still

down-regulated, although the number of genes showing up-regulated expression increases.

Many of the differentially expressed genes detected at Time 1 encode cell wall related

proteins. Arabinogalactan proteins for example are plant glycoproteins that consist of a core-

protein backbone O-glycosylated by one or more complex carbohydrates consisting of galactan

and arabinose as main components. Fasciclin-like arabinogalactan proteins contain one or two

fasciclin domains thought to be involved in protein-protein interactions that are variably

modified with a glycosylphosphatidylinositol lipid anchor (Seifert and Roberts, 2007). Cell

surface arabinogalactan proteins in the roots of Arabidopsis were shown to influence the

organization of cortical microtubules, which in turn affect the orientation of cellulose

microfibrils that constitute an ordered, fibrous phase of the cell wall (Nguema-Ona et al., 2007).

In roots, arabinogalactan proteins are implied in root elongation, the orientation of root

patterning and root hair growth (Seifert and Roberts, 2007). Genes encoding arabinogalactan

proteins show down-regulated expression at Time 1, indicating a possible disorganization of

cortical microtubules and consequently of cellulose microfibrils, and a reduction in root










elongation and root hair growth. The gene encoding UGE4 also shows down-regulated

expression at Time 1 in this study. A loss-of-function allele of UDP-glucose 4-epimerase 4

(UGE4), which freely interconverts UDP-glucose and UDP-galactose resulted in a 20% decrease

in Arabidopsis root cell wall galactose content that correlated with a reduction in root length

(Rosti et al., 2007). The loss of UGE4 specifically affected xyloglucan galactosylation by

causing a 40% reduction in the relative total abundance of galactose-containing xyloglucan

oligosaccharides in roots. Xyloglucans are polysaccharides binding noncovalently to cellulose,

thereby coating and cross-linking adj acent cellulose microfibrils (Rose et al., 2002). Xyloglucan

galactosyltransferases, which show sequence similarities to the glucuronosyltransferase domain

of exostosins, are involved in the biosynthesis of xyloglucans by attaching D-galactose to

specific xylose residues (Madson et al., 2003). The expression of two genes encoding xyloglucan

galactosyltransferases is down-regulated at Time 1, thereby affecting the structure of

xyloglucans. A class of enzymes known as xyloglucan endotransglucosylases/hydrolases (XTHs)

catalyzes the endo-cleavage of xyloglucan polymers and the subsequent transfer of the newly

generated reducing ends to other polymeric or oligomeric xyloglucan molecules (Liu et al.,

2007). Xyloglucan endotransglucosylases/hydrolases have been associated with primary root

growth and root hair initiation in Arabidopsis (Vissenberg et al., 2001; Liu et al., 2007). The

down-regulated expression at Time 1 of genes encoding these enzymes indicates a possible

reduction in the associated processes. Expansins are cell wall loosening proteins with a proposed

nonenzymatic action mechanism in which noncovalent bonds are disrupted that connect matrix

polysaccharides to the surface of cellulose (Sampedro and Cosgrove, 2005). All the a-expansin

proteins that have been characterized so far have a pH optimum for cell-wall extension of about

4, which permits the cell to regulate a-expansin activity rapidly by modulating wall pH through









the activity of the plasma membrane H+ ATPase, which pumps protons to the cell wall

(Sampedro and Cosgrove, 2005). The expression of genes encoding a-expansin proteins is up-

regulated at Time 1, while that of three transcripts encoding putative proton pump ATP-ases is

up-regulated at Time 3, possibly indicating an increase in a-expansin activity with time. The

expression of two genes encoding P-expansin proteins on the other hand is down-regulated; a P-

expansin gene in barley is tightly related to root hair initiation (Kwasniewski and Szarejko,

2006). Transcripts encoding lipid transfer proteins, which in this study show down-regulated

expression, can facilitate the ongoing extension process of cell walls in tobacco and wheat,

where they were associated to the presence of a P-expansin (Nieuwland et al., 2005). A gene

encoding a pectin acetylesterase also shows down-regulated expression at Time 1. Pectin

acetylesterases catalyze the deacetylation of esterified pectin, which diminishes the pectin

backbone hydrophobicity and increases its solubility in water, while it makes the polysaccharide

more accessible to pectin-degrading enzymes (Vercauteren et al., 2002). The expression of

transcripts encoding polygalacturonases on the contrary is up-regulated. These are enzymes that

hydrolyze the homogalacturonan (pectin backbone) of the plant cell wall and are therefore

involved in cell wall loosening. In general, the genes encoding proteins involved in cell-wall

loosening show down-regulated expression in this study where they are involved in root (hair)

growth, while those showing up-regulated expression might be the ones involved in expansion of

existing cells.

The up-regulated expression of giberrellin 2 oxidases mentioned earlier in the discussion

also indicates a possible decrease in root growth. Gibberellins (GAs) promote growth by

targeting the degradation of DELLA repressor proteins in the endodermis, where cell expansion

is rate-limiting for elongation of other tissues and therefore of the root as a whole (Ubeda-Tomas









et al., 2008). Other genes potentially associated with root growth processes are for example the

transcription factor scarecrow-like 11 and the putative replication protein Al, both of which

show down-regulated expression at Time 1. Scarecrow-like 11 is related to SCARECROW,

which is involved in root cortex cell proliferation, stem cell renewal, and the gravitropic response

(Cui and Benfey, 2009). Finally, the expression of two anionic peroxidases, one of which is a

putative lignin forming anionic peroxidase, is strongly up-regulated at Time 1. Anionic

peroxidases belong to class III peroxidases, which are generally secreted into the cell wall or the

surrounding medium and the vacuole where they catalyse the reduction of H202 by taking

electrons from various donor molecules such as phenolic compounds, lignin precursors, auxin, or

secondary metabolites, or where they are involved in the formation of various reactive oxygen

species (ROS) (Cosio and Dunand, 2009). Lignin may form a barrier against metal entrance

(Kovacik and Klejdus, 2008).

Many genes encoding proteins of known and unknown function are now implicated in the

early response of Arabidopsis roots to high levels of magnesium sulfate. Several of the genes

described above have been shown to be involved in an abiotic stress response in previous studies,

but the maj ority can only be functionally interpreted on a family or superfamily level.

Transcripts Encoding Transport Proteins Col-0 Time Series

Vacuolar transporters

Storage vacuoles contain mainly inorganic salts and water, enabling the plant to reach a

large size and surface area using a minimum of energy for the synthesis of organic metabolites.

The vacuole also serves as an internal reservoir of metabolites and nutrients and takes part in

cytosolic ion homeostasis (Martinoia et al., 2000). The vacuolar MHX transporter functions as an

exchanger of H+ with Mg2+ and Zn2+ ions and is expressed in the vascular cylinder in close

association with the xylem tracheary elements and in the root epidermis (Shaul et al., 1999). The









gene encoding the MHX transporter shows up-regulated expression at Time 2 and 3, indicating

the possible storage of excess Mg2+ in the vacuole. The tonoplast-localized H /SO42-

cotransporter SULTR4; 1 mediates the efflux of sulfate from vacuoles in the pericycle and xylem

parenchyma cells of the vascular tissues along the entire root (Kataoka et al., 2004). The

expression of the gene encoding SULTR4; 1 is up-regulated at Time 3, pointing at the possible

retention of excess SO42- in the vacuole. Several members of the CAX (cation exchanger) family

show differential expression in response to elevated magnesium sulfate. The Ca2+ H+ antiporter

CAX1, which shows down-regulated expression across the time points, is one of the few

transporter genes that are already differentially expressed at Time 1. Its down-regulated

expression was confirmed by Q-PCR at Time 3 when controlled for diurnal effects. CAX1 is

localized to the tonoplast and responsible for 50% of the Ca2+ H+ antiport activity there (Cheng

et al., 2003). The expression of the CAX2 antiporter is down-regulated at Time 2 and 3. CAX2

has been shown to transport Ca2+ and Mn2+ into the vacuole in Arabidopsis and other plant

species (Edmond et al., 2009). CAX3, a weak Ca2+ VACUOlar transporter, also shows down-

regulated expression from Time 2 onwards. CAX1 and CAX3 can be combined as "hetero-

CAX" complexes to form functional transporters with distinct transport properties (Zhao et al.,

2009). The down-regulated expression of the genes encoding these three vacuolar Ca2+ H+

antiporters indicates a possible shortage of calcium in the cytosol. The gene encoding CHX20

shows up-regulated expression at Time 2 and 3. CHX20 is localized to endomembranes, where it

is thought to play a critical role in osmoregulation and possibly pH modulation by exchanging

K+ for H+ (Padmanaban et al., 2007). The expression of two tonoplast intrinsic proteins (TIP),

which are vacuolar aquaporins, is also down-regulated. Aquaporins are water channel proteins,

but some have also been shown to transport CO2, H202, boron or silicon in addition to H20










(Maurel, 2007). An organic cation transporter (OCT3), which is localized to the vacuolar

membrane, shows down-regulated expression across the time points. OCT transporters are

differentially expressed in response to various abiotic stresses (Kufner and Koch, 2008). ATP-

binding cassette (ABC) type transporters such as MRP1 and MRP2, which show down-regulated

expression at Time 3, are directly energized by Mg-ATP and do not depend on the

electrochemical force. Their substrates are organic anions formed by conjugation, e.g. to

glutathione (Martinoia et al., 2000).

Plasma membrane transporters

The slightly up-regulated expression of the gene encoding magnesium transporter MRS2-

10 at Time 2 and 3 corresponds with the observation described in Chapter 2 that a knockout

mutant of MRS2-10 does not show improved growth in the form of higher FW biomass

compared to col-0 at elevated levels of magnesium sulfate. Q-PCR results did not show a

statistically significant difference in expression for MRS2-10 at Time 3, with or without control

for diurnal effects, although the difference in expression was closer to the microarray results

without the control at Time 3. The expression of the gene encoding MRS2-7, a low-affinity

magnesium transporter of the same family, is also up-regulated. Both MRS2-10 and MRS2-7

have been shown to transport other ions in addition to Mg2+ (Deng et al., 2006; Mao et al., 2008).

Schock et al. (2002) speculate that, in line with the large number of AtMRS2 family members,

the function of the AtMRS2 gene family may be the maintenance of metal ion homeostasis in

different cellular compartments (i.e. over different cellular membrane systems). The expression

of the gene encoding SULTR1;2, a high-affinity sulfate transporter, showed no significant

differences at the three time points. The fact that expression of the gene is not down-regulated in

the early adaptation response of Arabidopsis roots to high levels of magnesium sulfate supports

the outcome of the FW biomass comparison between the SULTR1;2 knockout mutant sell-10









and col-0 described in Chapter 2, which showed no advantage for sell-10. The gene encoding

EIL3, one of the known transcriptional regulators of SULTR1;2 (Maruyama-Nakashita et al.,

2006), did not show a significant difference in expression at any of the time points either.

Interestingly, two sulfate transporter genes of the same family as SULTR1;2 show down-

regulated expression; the gene encoding SULTR3;4 at Time 2 and 3, and the gene encoding

SULTR3;1 at Time 3. The spatial and subcellular localization of these transporters is not known,

and so far no influence on the expression of Group 3 sulfate transporters by the sulfur status of

plants has been reported (Buchner et al., 2004). This study shows for the first time that genes

encoding Group 3 sulfate transporters SULTR3;1 and SULTR3;4 are differentially expressed in

roots upon exposure to high levels of sulfate. The down-regulated expression of the gene

encoding SULTR3;4 was confirmed by Q-PCR at Time 3, although its down-regulated

expression was less pronounced when controlled for diurnal effects.

Several genes encoding transporters of other inorganic nutrients were also differentially

expressed. The expression of the gene encoding potassium channel HAK5 is strongly up-

regulated at Time 1 and continues to increase at Time 2 and 3. The expression of the HAK5

encoding gene is up-regulated by growth conditions that result in a hyperpolarized root plasma

membrane potential (Nieves-Cordones et al., 2008). The inorganic phosphate transporter PHT 1

shows up-regulated expression at Time 1 and 3, which may point to an increased demand for

phosphate. Several genes encoding root ion transporters that are differentially expressed in this

study, such as ACH1 (AtNRT2.1) AMT1.1 and PT2, are diurnally regulated (Lejay et al., 2003).

Whether they are also regulated by high magnesium sulfate will have to be determined in future

studies. The down-regulated expression of the gene encoding CNGC 1 at Time 1 and 2 could

indicate a temporary reduction in Ca2+ uptake. Plant cyclic nucleotide gated channels (CNGCs)









are a large family of proteins that are thought to be ligand-gated, nonselective inwardly

conducting cell membrane cation channels. CNGC 1 protein is primarily expressed in the roots of

Arabidopsis seedlings where it contributes (along with other channels) to Ca2+ uptake into plants

(Ma et al., 2006).

Transcripts caxl-1/col-0 Comparison

The differential expression of only four transcripts between caxl-1 and col-0 plants at time

3 indicates that the vast maj ority of the root transcriptome responses to high magnesium sulfate

are the same for the two genotypes at this time after initiation of treatment. As described above,

CAX1 is one of the few transporter genes that are already differentially expressed at Time 1 in

col-0, which means that down-regulated expression of CAX1 is part of the natural response of

Arabidopsis to elevated levels of magnesium sulfate. The difference between the down-regulated

level of CAX1 in col-0 and the absence of CAX1 in the knockout mutant might not be large

enough, and the time passed between initiation of treatment and root harvest may not be long

enough, to reveal many of the downstream molecular processes eventually leading to enhanced

tolerance for caxl-1 at the whole plant level, including reduced leaf Mg content and increased

fresh weight biomass, after days or weeks of exposure in agar or soil medium (Chapter 2).

The three transcripts showing a difference in abundance in caxl-1 compared to col-0 other

than CAX1 itself currently do not have a known function. Follow-up experiments need to be

done to discover the function of these three transcripts and whether they can help explain the

observed higher tolerance of caxl-1 plants to elevated magnesium sulfate in the growth medium

compared to col-0. The root transcriptome responses of caxl-1 and col-0 could be compared at

later time points after initial exposure to high magnesium sulfate. Doing so might reveal

additional differentially expressed genes that play a role in the whole plant level tolerance

difference exhibited by these genotypes. Some of the genes involved in downstream processes









may themselves be candidate genes for enhanced tolerance, such as those encoding (regulators

of) plasma membrane based channels that transport Mg2+

Candidate Genes for enhanced tolerance

One of the main obj ectives of this study was to identify potential transporter genes

encoding plasma membrane based transporters of Mg2+ and SO42, Or their transcriptional or post-

translational regulators, that are differentially expressed in the early phases of Arabidopsis root

adaptation to elevated levels of magnesium sulfate. Based on the up-regulated expression of the

gene encoding the known vacuolar transporter MHX, which transports Mg2+ into the vacuole,

and the down-regulated expression of the gene encoding SULTR4;1i, which exports SO42- fTOm

the vacuole, it is concluded that both Mg2+ and SO42- are available to the plant in excessive

amounts. Plasma membrane localized transporters of Mg2+ and SO42- may therefore also be

differentially expressed. Genes encoding known plasma membrane based importers of Mg2+ and

SO42-, Such as MRS2-10, SULTR1;1 or SULTR1;2, did not show down-regulated expression.

One possible way of Mg2+ entry in roots may be through putative homologs of the wheat rca

channel, which is defined as a calcium channel, but which is permeable to a wide variety of

monovalent and divalent cations including Ca2+, Mg2+, Mn2+, Cd2+, CO2+, Ni2+ K and Na

(Shaul, 2002). Among the genes encoding transporters of unknown function in the col-0 time

series that show a significant difference in expression, candidate genes for improving plant

tolerance of high magnesium sulfate by reducing accumulation within the plant might be found.

Additional research is needed to determine whether for example any of the maj or intrinsic

proteins, cyclic nucleotide gated channels, integral membrane proteins, cation efflux family

proteins, ATPase E1-E2 type family proteins, MATE efflux family proteins, transporter-related

proteins and putative transporters, are plasma membrane localized and able to transport Mg2+

into or out of peripheral root cells, or export SO42- fTOm the same cells. Follow-up studies are









also needed to determine the localization of Group 3 sulfate transporters SULTR3;1 and

SULTR3;4 in the root. For transporters differentially expressed at Time 2 and 3, diurnal effects

will have to be ruled out if this has not been done already. The differential expression of post-

translational regulators of plasma membrane transporters could be an important way to affect ion

uptake and efflux; candidate regulators could for example be kinases and phosphates that are

differentially expressed at Time 1. Based on these first results it is not possible to identify

specific candidate genes other than SULTR3;1 and SULTR3;4 whose (natural) over- or under-

expression could improve tolerance of Arabidopsis to elevated levels of magnesium sulfate.

Even for a stress as widely studied as high NaC1, no plasma membrane localized

transporters responsible for uptake of Na' from the soil solution have been identified yet in

Arabidopsis roots. In other plant species, Na+ has been shown to enter roots passively, via

voltage independent (or weakly voltage-dependent) nonselective cation channels and possibly

via other Na+ transporters, such as some members of the high-affinity K+ transporter (HKT)

family (Munns and Tester, 2008). For example, a line of transgenic wheat plants expressing an

antisense construct of the high affinity K+ transporter TaHKT2; 1 showed reduced sodium uptake

by roots and enhanced growth relative to unstressed plants compared to a control line at high

levels of NaCl in the growth medium (Laurie et al., 2002). With respect to efflux, several plasma

membrane ion transporters have been identified in Arabidopsis. Overexpression of the

Arabidopsis SOS1 gene, which encodes a plasma membrane Na /H+ antiporter responsible for

Na' efflux, limited Na' accumulation and improved growth compared to control plants at high

NaCl concentrations (Shi et al., 2003). Overexpression of BOR4, an Arabidopsis borate exporter

detected in the plasma membranes of the distal sides of root epidermal cells, generated plants

that are tolerant of high boron levels (Miwa et al., 2007). The Arabidopsis NO3~ Offlux









transporter NAXT1 is localized in the plasma membrane and mainly expressed in cortex cells of

mature roots. It was found to be up-regulated at the post-transcriptional level in wildtype upon

acid treatment (Segonzac et al., 2007). This example shows that regulation of import or efflux

transporters may not occur at the transcriptional level and can therefore be missed by a

microarray experiment.

Conclusions

Arabidopsis col-0 root transcriptome analysis reveals over 300 differentially expressed

genes at Time 1. Genes of known function include those encoding calcium-binding proteins,

kinases, transcription factors, enzymes involved in hormone metabolism, disease resistance

proteins and many cell wall related proteins. The responses of the genes encoding cell wall

related proteins indicate a possible reduction in root growth when col-0 is exposed to high

concentrations of magnesium sulfate. Some of the genes of known or unknown function were

previously associated with specific or broad ranges of abiotic stresses, but not necessarily in

roots or at these time points. Over 200 genes encoding membrane based transporters were

differentially expressed across the col-0 time series. The expression of genes encoding known

plasma membrane based importers of Mg2+ and SO42- ions, such as MRS2-10, SULTR1;1 or

SULTR1;2, was not down-regulated. This corresponds with the observation reported in Chapter

2 that Arabidopsis lines carrying knockout T-DNA insertion mutations in AtMRS2-10 and

AtSULTR1;2 did not mitigate the constraining impacts of high magnesium sulfate

concentrations on wildtype Arabidopsis plants. The differential expression of genes encoding

known tonoplast localized transporters of Mg2+ and SO42- ions indicate a possible storage of

excess Mg2+ and SO42- ions in the vacuole. Future research can reveal whether any of the

differentially expressed transporter genes of unknown protein localization, protein function, or

both, are candidates to enhance tolerance to high levels of soluble magnesium sulfate minerals in









Martian regolith by reducing accumulation of Mg2+ and SO42- ions within the plant. For

example, the localization of sulfate transporters SULTR3;1 and SULTR3;4 within Arabidopsis

root tissue and cells can be analyzed in follow-up studies to see if the genes encoding these

transporters are candidates for enhanced tolerance. Potential regulators of membrane based

transporter activity, such as kinases, which are encoded by differentially expressed genes across

the col-0 time series, could also be analyzed. Since gene expression differences are not fully

controlled for diurnal effects for the Time 2 and 3 comparisons in the col-0 time series, these

effects will have to be ruled out in follow-up studies.

The down-regulation of cax1-1 gene expression is a natural response to high magnesium

sulfate in col-0 that is already seen at Time 1. Together with the down-regulation of CAX2 and

CAX3 gene expression at later time points it indicates a possible shortage of calcium in the

cytosol experienced by col-0 when exposed to high concentrations of magnesium sulfate. Only

three transcripts were differentially expressed between caxl-1 and col-0 at 3 hours after initiation

of treatment. Follow-up experiments could be done to discover the function of these three

transcripts. The root transcriptome of caxl-1 and col-0 could furthermore be compared at later

time points after initial exposure to high magnesium sulfate to reveal additional differentially

expressed transcripts that could indicate the molecular processes eventually leading to the

tolerance difference exhibited by these genotypes after days or weeks of growth. Some of the

transcripts involved in downstream processes may themselves be candidates for enhanced

tolerance, such as those encoding (regulators of) plasma membrane based channels that transport

Mg2+












Hydroponic Root Growth

Few studies have focused on developing methods to grow Arabidopsis thaliana

hydroponically, and only one system is commercially available at present (Tocquin et al., 2003).

Detailed analyses of root enzyme activity and gene expression indicated that it is highly

important to continuously provide oxygen to the root zone through aeration (Smeets et al., 2008).

We designed a hydroponic set-up for Arabidopsis that is generally based on the systems

described by Tocquin et al. (2003) and Smeets et al. (2008). Glass containers (2.6 L) were

covered along the glass with opaque black plastic sheeting and across the top with opaque plastic

lids. 18 Holes (d = 1.43 cm) were made in each opaque plastic lid with a hole puncher at regular

intervals. A 2-cm-long rockwool plug (d = 1.59 cm) was placed in each hole. Containers were

filled with a nutrient solution adjusted to pH 5.7 by KOH that was replaced once a week. It

consisted of distilled water, 0.25 g/L MES buffer, and 1/32x MS salts during week 1 and 2, while

the concentration of MS salts increased to 1/16x during week 3. About ten sterilized Arabidopsis

seeds were planted per rockwool plug. 18 containers were planted with col-0, while 6 were

planted with cax1-1. Containers were covered with transparent plastic wrap during the first 3-5

days of germination. Holes were made in the plastic wrap after 1-3 days depending on the

general atmospheric humidity. The containers were randomly placed over 3 benches (8

containers/bench). Air was supplied to the roots by one 3W air pump per 8 containers; each

container had an airstone made from glass beads. After one week, seedlings were thinned to one

seedling per rockwool plug, so that every container supported 18 plants (Figure 3-1). At day 21,

col-0 and cax1-1 roots were exposed to the basic nutrient solution (0.25 g/L MES, 1/16x MS, pH

5.7) with an additional 2.08 mM magnesium sulfate (total Ca:Mg ratio = 1:15). Col-0 was

exposed for 45 min. (time 1), 90 min. (time 2), or 3 hrs (time 3), while carl-1 was exposed for 3


Materials and Methods









hrs only. The col-0 control received no extra magnesium sulfate and was harvested at 45 min.

(time 1) together with the first col-0 treatment (Table 3-1). Four replicate containers were

harvested for the control and the treatments (Figure 3-2). Roots were cut below the rockwool

plug and pooled per container before being flash frozen in liquid nitrogen and stored at -80oC.

Microarray Procedures, Statistical Analysis and Data display

RNA was extracted from root samples with the RNeasy Plant Mini Kit (Qiagen). Samples

were weighed while still frozen and reagents were adjusted to the total measured weight for the

grinding and lysing step. Two replicate extractions were completed per sample with lysate

volumes corresponding to 100 mg frozen wet weight each. The remainder of a sample was stored

at -80oC as cleared lysate according to the Qiagen protocol. The quality and quantity of the

extracted RNA was checked with denaturing agarose gels stained with ethidium bromide, and the

NanoDrop 1000 Spectrophotometer (Thermo Scientific). RNA from each sample was amplified

and labeled with cy3 dye by using Agilent Quick Amp labeling kit, one color. A total of 20

samples were hybridized to five 4x44k Arabidopsis microarray slides (Agilent), which were then

washed and scanned before data was extracted (Figure 3-3). Microarray handling and data

extraction was done at the Interdisciplinary Center for Biotechnology Research (ICBR) at the

University of Florida according to the One-Color Microarray-Based Gene Expression Analysis

(Quick Amp Labeling) Protocol (version 5.7). Median signal intensities were quantile

normalized using R software. Log2 transformed normalized data were evaluated in a mixed

analysis of variance (ANOVA) model using SAS 9 and JMP Genomics 7 software. Microarray

slide was included in the ANOVA model as a random effect to account for the covariance of

samples hybridized to the same microarray (i.e. control for spot effect) (Jin et al., 2001;

Wolfinger et al., 2001). Unbiased estimates of transcript abundance (least-square means) were

generated for each sample set. The transcript level estimates of different sample sets were then










compared using a series of t-tests; the col-0 time 1, col-0 time 2 and col-0 time 3 treatment sets

were compared to the col-0 time 1 control sample set, while the caxl-1 time 3 set was compared

to the col-0 time 3 set (Table 3-1, Figure 3-3). The p-values generated in these comparisons were

corrected for multiple testing by controlling the False Discovery Rate (FDR) with the Q-value

procedure in the Q-value 1.0 package (default settings) of the R software (Storey and Tibshirani,

2003). Lists of genes with statistically significant changes in expression at p < 0.05 between

sample sets were further organized, analyzed and displayed in tables and figures using JMP

Genomics 7, R software, GeneVenn, Cluster 2. 11, TreeView 1.60, the National Center for

Biotechnology Information (NCBI) BLAST tool, and the Arabidopsis Information Resource

(TAIR) Aracyc Omics viewer and GO annotations tools.

Quantitative Real-time PCR

Wildtype Arabidopsis plants (col-0) were grown hydroponically for 21 days as described

above. At day 21, four containers received a replacement of the basic nutrient solution (0.25 g/L

MES, 1/16x MS, pH5.7), while another four containers received the basic solution with an

additional 2.08 mM magnesium sulfate (total Ca:Mg ratio = 1:15). Roots were exposed for three

hours before being harvested per container. RNA was extracted and checked for quality and

quantity as described above. RNA from these independently grown plants exposed or unexposed

for 3 hrs, as well as RNA from plants exposed for 3 hrs and unexposed for 45 min. in the

microarray experiment, were used for the quantitative PCR (Q-PCR) analyses. Q-PCR was

performed using TaqMan reverse transcription reagents and Power SYBR Green PCR Master

Mix (Applied Biosystems). 1 ug of total RNA was reverse transcribed per sample for a total of

three control and three treatment samples. Primers were designed for 5 genes using Primer

Express (Applied Biosystems) (Table 3-9). At2g32170 was chosen as a stably expressing

reference gene appropriate for abiotic stress treatment (Czechowski et al., 2005). Each










template/primer pair combination was run in triplicate. The relative increase or decrease of

mRNA abundance between the two sample sets was calculated by using the Pfaffl method, and

statistical analysis of the results was done with the REST 2008 2.0.7 software.









Table 3-1. Experimental conditions of one color microarray experiment
Biol.
Treatment Time Seedline .. Arrays
replicates
Control solution 45 min. = time 1 col-0 4 4
Magnesium sulfate 45 min. = time 1 col-0 4 4
Magnesium sulfate 90 min. = time 2 col-0 4 4
Magnesium sulfate 3 hrs = time 3 col-0 4 4
Magnesium sulfate 3 hrs = time 3 caxl-1 4 4
Gene expression was compared between sample sets. The comparison of col-0 time 1 treatment
with col-0 time 1 control is referred to in the text, tables and figures as Time 1. Col-0 time 2
treatment versus col-0 time 1 control is referred to as Time 2. Col-0 time 3 treatment versus col-0
time 1 control is referred to as Time 3. The comparison of carl-1 time 3 treatment with col-0
time 3 treatment is not referred to in abbreviated form.

Table 3-2. Number of genes (and % of total) in GO molecular functional categories per
comparison


Functional Category
DNA or RNA binding

hydrolase activity
kinase activity
nucleic acid binding
nucleotide binding
other binding
other enzyme activity
other molecular functions

protein binding
receptor binding or activity
structural molecule activity
transcription factor activity
transferase activity
transporter activity
unknown molecular functions


Time 1

22 (6.43%)
27 (7.89%)
13 (3.80%)
5 (1.46%)
12 (3.51%)
42 (12.28%)
23 (6.73%)
6 (1.75%)
34 (9.94%)
8 (2.34%)
1 (0.29%)
33 (9.65%)
16 (4.68%)
14 (4.09%)
86 (25.15%)


Time 2

103 (6.58%)
105 (6.71%)
83 (5.30%)
25 (1.60%)
64 (4.09%)
155 (9.90%)
127 (8.12%)
54 (3.45%)
138 (8.82%)
22 (1.41%)
18 (1.15%)
117 (7.48%)
122 (7.80%)
80 (5.11%)
352 (22.49%)


Time 3

233 (6.93%)
253 (7.53%)
165 (4.91%)
52 (1.55%)
152 (4.52%)
331 (9.85%)
347 (10.32%)
128 (3.81%)
262 (7.80%)
32 (0.95%)
30 (0.89%)
230 (6.84%)
265 (7.88%)
202 (6.01%)
679 (20.20%)


The table shows the number (and percentage of total) of significantly differentially expressed
genes per GO functional category in col-0 roots exposed to magnesium sulfate at Time 1, Time 2
and Time 3










Table 3-3. Genes of Arabidopsis thaliana (col-0) with differential expression at q < 0.001 at
Time 1


q-value
6.71E-07
6.71E-07
1.99E-06
2.44E-06

6.25E-06
8.93E-06
1.20E-05
1.20E-05

1.20E-05
1.20E-05
1.43E-05
1.65E-05
1.80E-05
3.48E-05
6.97E-05
9.42E-05
0.000216711
0.000283167
0.000295879
0.000299824
0.00032124
0.000405444
0.000410629
0.000416943
0.000427249
0.000432439
0.00044696
0.000465244

0.000465244
0.000468475
0.000510184

0.000528206
0.000565577
0.00090829


Log2(FC)
-1.2264
-2.0693
1.4961
1.5416

-1.3399
0.7094
0.7914
-0.907

-1.6379
2.6714
-1.2778
1.1068
1.673
-1.149
1.1562
-1.1368
-0.9759
1.3473
2.0824
-0.7518
2.2018
2.3327
-1.0069
-0.5479
-1.4143
1.1365
2.023
-1.0837

-1.585
-1.4471
0.9195

-1.0809
-0.7772
1.9088


Gene and DNA region description
phosphate-responsive protein, putative (EXO) [At4g08950.1]
rhomboid family protein [At4g23070.1]
expressed protein [At4g29780.1]
AP2/EREBP-like transcription factor LEAFY PETIOLE, putative
[At5gl3910.1]
expressed protein [At5g25240.1]
kinase interacting family protein [At2g30500.1]
senescence-associated protein-related [At2g25690.1]
glycoside hydrolase family 28 protein/polygalacturonase
(pectinase) family protein [At3g06770.1]
expressed protein [At4g3 723 5. 1]
expressed protein [At5g38700.1]
calmodulin, putative [At3gl0190.1]
expressed protein [Atlg74450.1]
nodulin-related [At2g30300.1]
arabinogalactan-protein (AGPl17) [At2g23130. 1]
calcium-binding EF hand family protein [At3gl0300.1]
expressed protein [At4g01140. 1]
expressed protein [At2gl7300.1]
lipoxygenase, putative [Atlg72520.1]
immediate-early fungal elicitor family protein [At3g02840.1]
expressed protein [Atlg09812.1]
Unknown [CHR2:009686022-009686081]
anionic peroxidase, putative [Atlgl4550.1]
arabinogalactan-protein (AGP21) [Atlg55330.1]
expressed protein [Atlg22882.1]
BON1-associated protein (BAPl)-related [At2g45760. 1]
exocyst subunit EXO70 family protein [At2g28650.1]
anionic peroxidase, putative [Atlgl4540.1]
expressed protein predicted protein, Arabidopsis thaliana
[At4g35320.1]
Unknown [CHR1:021477192-02 1477133]
heat shock transcription factor family protein [At3g63350.1]
auxin-responsive protein/indoleacetic acid-induced protein 19
(IAAl9) [At3gl5540.1]
RNA recognition motif (RRM)-containing protein [Atlg78260. 1]
arabinogalactan-protein (AGP7) [At5g65390.1]
basic helix-loop-helix (bHLH) family protein [At2g22760.1]





Table 3-4. Genes ofArabidopzsis thaliana (col-0) with differential expression > 3 fold at Time 1


Log2(FC)
3.2099
3.1098
2.6714
2.6151
2.556
2.554
2.4955
2.3327
2.2018
2.0836
2.0824
2.023
1.9782
1.9294
1.9088
1.8946
1.8802


1.8732
1.8605
1.8518
1.8469
1.8359
1.7794
1.7716
1.7374
1.6878
1.673
1.6424
1.6067
-1.585
-1.6379
-1.7618

-1.7916
-2.0347
-2.0693
-2.3484


q-value
0.001474678
0.009201495
1.20E-05
0.01425427
0.003229586
0.001396907
0.00128496
0.000405444
0.00032124
0.001823929
0.000295879
0.00044696
0.003274863
0.04072004
0.00090829
0.01474273
0.04969909


0.01911007
0.004903664
0.03187784
0.04757759
0.02514971
0.01188793
0.0074956
0.003857111
0.005354
1.80E-05
0.003447976
0.02599538
0.000465244
1.20E-05
0.005505618

0.04439508
0.02980178
6.71E-07
0.02514971


Gene and DNA region descriptions
expansion, putative (EXPl2) [At3gl5370.1]
F-box family protein/SKP1 interacting partner 3-related [At2g02320. 1]
expressed protein [At5g38700.1]
expressed protein [Atlg20310. 1]
expressed protein [Atlg06135.1]
9-cis-epoxycarotenoid dioxygenase, putative [Atlg01140.1]
expansion, putative (EXPl17) [At4g0 163 0.1]
anionic peroxidase, putative [Atl1g4550.1]
Unknown [CHR2:009686022-009686081]
nitrate-responsive NOI protein, putative [At2gl7660.1]
immediate-early fungal elicitor family protein [At3g02840.1]
anionic peroxidase, putative [Atl1g 4540.1]
expressed protein [At5g50335.1]
Unknown [CHR1:027987259-027987200]
basic helix-loop-helix (bHLH) family protein [At2g22760. 1]
polygalacturonase, putative/pectinase, putative [At2g43890.1]
S-adenosyl-L-methionine carboxyll methyltransferase family protein
[At3g44870.1]
auxin-responsive family protein [At4gl2410.1]
1 -aminocyclopropane-1l-carboxylate synthase, putative [At5g65 800. 1]
no apical meristem (NAM) family protein [At2g46770.1]
AP2 domain-containing transcription factor, putative [At4g34410.1]
ovate family protein [At4gl4860. 1]
CBL-interacting protein kinase 9 (CIPK9) [Atlg78390.1]
calmodulin-related protein, putative [Atlg76640.1]
glutamate decarboxylase, putative [At2g02010. 1]
ethylene-responsive element-binding protein, putative [At5g25190. 1]
nodulin-related [At2g30300. 1]
leucine-rich repeat family protein [Atlg78230.1]
lectin protein kinase, putative [Atlg70130.1]
Unknown [CHR1:021477192-02 1477133]
expressed protein [At4g37235.1]
disease resistance protein (TIR-NBS-LRR class), putative
[At4gl6860.1]
Unknown [CHR1: 021477409-02147735 0]
cDNA clone RAFLO9-31-N12 3', mRNA sequence [AV802416]
rhomboid family protein [At4g23070. 1]
disease resistance protein (TIR-NBS class), putative [At3g04210.1]










Table 3-5a. Q-PCR results of col-0 gene expression at time 3 treatment versus time 1 control;
RNA sources are the same as for the microarray experiment
Gene Type Reaction Expression Std. Error 95% C.I. P-value Result
Efficiency (array)
At2g32170 REF 0.7657 1
CAX 1 TRG 0.8076 0.57 (0.602) 0.436 -0.766 0.366 0.925 0 DOWN
MRS2-10 TRG 0.8548 1.035 (1.183) 0.758 1.377 0.628 1.466 0.714
SULTR3;4 TRG 0.7749 0.557 (0.541) 0.462 0.691 0.362 0.773 0 DOWN
NRAMP 1 TRG 0.8599 1.196 (1.317) 0.886 -1.515 0.787 -2.537 0.107

Table 3-5b. Q-PCR non-normalized results of col-0 gene expression at time 3 treatment versus
time 1 control; RNA sources are the same as for the microarray experiment
Gene Type Reaction Expression Std. Error 95% C.I. P-value Result
Efficiency (array)
At2g32170 REF 0.7657 1.117 0.989 1.264 0.800 -1.518 0.051
CAX 1 TRG 0.8076 0.637 (0.602) 0.535 -0.775 0.460 0.860 0 DOWN
MRS2-10 TRG 0.8548 1.156 (1.183) 0.915 -1.435 0.829 1.662 0.06
SULTR3;4 TRG 0.7749 0.622 (0.541) 0.506 -0.766 0.391 -0.924 0 DOWN
NRAMP 1 TRG 0.8599 1.336 (1.317) 1.037 1.627 0.898 -2.862 0.004 UP

Table 3-6. Q-PCR results of col-0 gene expression at time 3 treatment versus time 3 control;
repetition 1
Gene Type Reaction Expression Std. Error 95% C.I. P-value Result
Efficiency (array)
At2g32170 REF 0.76571
CAX 1 TRG 0.8076 0.641 (0.602) 0.429 0.966 0.286 1.337 0.004 DOWN
MRS2-10 TRG 0.8548 1.017 (1.183) 0.876 1.195 0.720 1.335 0.763
SULTR3;4 TRG 0.7749 0.861 (0.541) 0.695 1.059 0.553 1.162 0.049 DOWN
NRAMP 1 TRG 0.8599 1.128 (1.317) 0.970 1.381 0.638 1.524 0.156

Table 3-7. Q-PCR results of col-0 gene expression at time 3 treatment versus time 3 control;
repetition 2
Gene Type Reaction Expression Std. Error 95% C.I. P-value Result
Efficiency (array)
At2g32170 REF 0.76571
CAX 1 TRG 0.8076 0.565 (0.602) 0.491 -0.673 0.418 -0.734 0 DOWN
MRS2-10 TRG 0.8548 0.986 (1.183) 0.786 -1.206 0.657 1.603 0.864
SULTR3;4 TRG 0.7749 0.907 (0.541) 0.749 1.075 0.589 1.572 0.26
NRAMP 1 TRG 0.8599 1.241 (1.317) 1.033 1.522 0.890 1.984 0.008 UP









Table 3-8. Genes with differential expression at q < 0.05 between Arabidopsis thaliana caxl-1
and col-0 treated for 3 hours
q-value Log2(FC) Gene and DNA region description
1.66E-09 -3.0461 calcium exchanger (CAX 1) [At2g3 8170. 1]

3.28E-07 -3.8209 calcium exchanger (CAX 1) [At2g3 8170.3]i

0.001201795 -2.7293 Expressed protein [At3g01345.1]

0.004291231 -1.1998 hypothetical protein [At4g07526.1]

0.005396832 -1.0903 Unknown [CHR2: 011819877-01181981 8]

Table 3-9. Genes and related primer sequences selected for Q-PCR
Gene ID Primer sequence

At2g32170 (reference gene) FW: 5'- GTTAAAT CAT GAC CATGGCAGT GT- 3'
RV: 5' -CTACATCAACCAGAGGAACATGTGT-3 '


At2g38170 (CAX1)


Atlg80900 (MRS2-10)


At3gl5990 (SULTR3;4)


Atlg80830 (NRAMPl)


FW: 5' -GCGACTCAGATTGGCTTATTCG-3 '

RV: 5' -GATCCATATTAATTCCCAAAATCCA-3 '

FW: 5' -TTCTCTGTCTGCGCCAGTTTC-3 '

RV: 5 '-GGC TC C TTACAATGC TC AAGC T-3 '

FW : 5 '- GGT GAAGC TGT GGC TGATC TC -3 '

RV: 5' -GCTCCATCTTCAGAAACAGTCTCTCT-3 '

FW: 5' -ACAGGATCTGGACGGTCTCAA-3 '
RV: 5 '-GATGAGT GGAGAAT TGGAGAAGCT -3 '








































D

C


Figure 3-1. Hydroponic Arabidopsis growth. Panels A, B and C show the set-up for the
microarray experiment. Panel D shows the set-up for the real-time Q-PCR
experiment.

























Figure 3.2. Harvest of hydroponically grown Arabidopsis roots. A) Example of Arabidopsis
roots after 21 days of growth. B) Example of root harvest per container for the
microarray and real-time Q-PCR experiments.


Time 3


vs col-0


Figure 3.3 Overview of microarray experiment. A) Scan of a single microarray with a single cy3-
labeled cRNA sample hybridized to it. The experiment included 20 arrays in total,
each with a single cy3-labeled sample hybridized to it. B) Diagram of the sample sets.
Each set consisted of 4 biological replicates. Gene expression was compared among
the sets as indicated by the arrows.
































*







r*, *:
*r ** i* *
** d

.*Olq~ I


-16x -8x -4x -2x


2x 4x 8x


(U
3
CO-
IFJ
0-
C1S
m cc> -
o



ef

q = 0.05


Log2 (Fo d Change)


Figure 3-4. Volcano plot of Time 1. Graph with the x-axis showing log2 values of the fold
changes in gene expression between col-0 exposed to magnesium sulfate for 45
minutes (n = 4) and col-0 exposed to the control nutrient solution for 45 minutes (n =
4). Each dot represents one of 37478 transcripts. Vertical lines indicate absolute fold
change values as indicated on top of the graph. The y-axis shows the -logl0 p-values
corresponding to the log2 fold change values. The horizontal line indicates the -logl0
p-value where the q-value is 0.05. Transcripts whose expression difference (fold
change) corresponds to a p-value for which q < 0.05 are indicated in red.





























(I):



i
?j *
tl

4i

dr
r, +
+id ir, i i
:...-,.L~1..........:---------:---------
I i i
i"~"j 'il

i+ i *r~~ ~ i +i


-32x -16x -8x -4x -2x


2x 4x 8x 16x 32x


(U
3
CO-
IFJ
0-
C1S
m cc> -
o



ef


q = 0.05
hl-


Log2 (Fo d Change)


Figure 3-5. Volcano plot of Time 2. Graph with the x-axis showing log2 values of the fold
changes in gene expression between col-0 exposed to magnesium sulfate for 90
minutes (n = 4) and col-0 exposed to the control nutrient solution for 45 minutes (n =
4). Each dot represents one of 37478 transcripts. Vertical lines indicate absolute fold
change values as indicated on top of the graph. The y-axis shows the -logl0 p-values
corresponding to the log2 fold change values. The horizontal line indicates the -logl0
p-value where the q-value is 0.05. Transcripts whose expression difference (fold
change) corresponds to a p-value for which q < 0.05 are indicated in red.
















1 t


I I I


-32x -16x -8x -4x -2x


2x 4x 8x 16x 32x


*! i


*+~+













i* ** *
*:


~9 i N
ir i ~I i ~



Q .'


~i
f


i i i
i i i i


(U
3
CO-
IFJ
0-
C1S
m cc> -
o



ef



q = 0.05
hl-


Log2 (Fo d Change)


Figure 3-6. Volcano plot of Time 3. Graph with the x-axis showing log2 values of the fold
changes in gene expression between col-0 exposed to magnesium sulfate for 3 hours
(n = 4) and col-0 exposed to the control nutrient solution for 45 minutes (n = 4). Each
dot represents one of 37478 transcripts. Vertical lines indicate absolute fold change
values as indicated on top of the graph. The y-axis shows the -logl0 p-values
corresponding to the log2 fold change values. The horizontal line indicates the -logl0
p-value where the q-value is 0.05. Transcripts whose expression difference (fold
change) corresponds to a p-value for which q < 0.05 are indicated in red.







































0:* .
I *: .


-32x -16x -8x -4x


2x 4x 8x 16x 32x


(U
3
CO-
IFJ
0-
C1S
m cc> -
o
9=


Log2 (Fo d Change)


Figure 3-7. Volcano plot of caxl-1 versus col-0 at time 3. Graph with the x-axis showing log2
values of the fold changes in gene expression between caxl-1 exposed to magnesium
sulfate for 3 hours (n = 4) and col-0 exposed to magnesium sulfate for 3 hours (n = 4).
Each dot represents one of 37478 transcripts. Vertical lines indicate absolute fold
change values as indicated on top of the graph. The y-axis shows the -logl0 p-values
corresponding to the log2 fold change values. The horizontal line indicates the -logl0
p-value where the q-value is 0.05. Transcripts whose expression difference (fold
change) corresponds to a p-value for which q < 0.05 are indicated in red.










Time 2


Time 3


Figure 3-8. Venn diagram of col-0 time series. The diagram shows the results of comparing sets
of genes with significant expression differences at Time 1, Time 2 and Time 3. 74
genes are unique to Time 1, 48 are shared with Time 2, 48 are shared with Time 3,
and 155 genes are shared between Time 1, 2 and 3.


Tim~e 1















myb family transcription factor (MYB50) [Atlg57560.1]
Dot-type zine finger domain-containing protein [At3q52440.1]
protein phosphatase 2C, putative / PP2C, putative [At2g30020.1)
protein phosphatase 2C, putative / PP2C, putative [At2g20630.2)
Unknown
flavin-containing monooxygenase family protein [At~gd3890.1]
COP1-interacting protein-related similar to CIP7 [Atl1l7360.1]
heat shock transcription factor family protein [At3g22830.1]
calcium-binding protein, putative [At3g59440.1]
glucosamine--fructose-6-phosphate aminotransferase, putative [At3g24090]
protein kinase family protein contains protein kinase domain [At2g30040.1]
expressed protein [Atlg52905.1]
Rieske [2Pe-2S] domain-containing protein similar to lethal leaf-spot 1 [At3944880.1]
expressed protein [Atl1g9180.1]
expressed protein [At5g45630.1]
expressed protein [At2g20010.1]
calmodulin-related protein, putative [Atlg76640.1]
phosphofructokinase family protein [At~g56630.1]
expressed protein [At5g36925.1]
basic helix-loop-helix (bHLH) family protein [At2g46510.1]
polygalacturonase, putative / pectinase, putative [Atlg05650.1)
basic helix-loop-helix (bHLM) family protein [Atlg59640.1]
harpin-induced family protein / HIN1 family protein [At4g01410.1]
myb family transcription factor [At5gd5580.1]
expressed protein [Atlg04490.1]
disease resistance protein (TIR-NBS-LRR class), putative [At5g45200.1]
polygalacturonase, putative / pectinase, putative [Atlg05660.1)
calcium-transporting ATPase 2, plasma membrane-type (ACA2) [At4g37640.1]
glycosyl transferase family 1 protein [Atlg73160.1]
dibydroflavono1 4-reductase family/dihydrokaempterol 4-reductase family [At2g45400.1]
endonuclease/exonuclease/phosphatase family protein [Atlg02270.1]
calmodulin-like protein (MSS3) [At2g43290.1]
oxidoreductase, 20G-Fe(II) oxygenase family protein [At4g21200.1]
inorganic phosphate transporter (PHT1) (PT1) [At5g43350.1]
nodulin-related weak similarity to nodule-specific protein H1570 [At2g30300.1]
AP2/EREBP-like transcription factor LEAFY PETIOLE, putative [At5gl3910.1]


Figure 3-9. A hierarchical average linkage cluster analysis using uncentered correlation was done
across Time 1, 2 and 3 based on the genes with significant expression differences at
Time 1. Yellow denotes a higher, and blue a lower expression of a gene in the treated
plants versus the control. The figure shows the expression patterns of genes
differentially expressed at Time 1 that are shared with Time 2, Time 3, or both.











NC domain-containing protein [At~gl6i360.1]
expressed protein [At2g34910.1]
kinase interacting family protein [At2g30500.1]
IAV521081 Arabidopsis thaliana cDNA clone APZ43b09F 3', mrRNA sequence [AV521081]
1-aminocyclopropane-1-carboxylate synthase/ACC synthese, putative [At5g65800.1]
'I mitochondrial substrate carrier family protein [At4g24570.1]
mexpteressedpotsien [Atprteng55720.1]t9166.
nitat-glcresonsive NOI protein utrativrae [Atmgl7660.1] tt~5~10r
9-eiMUD-glcpoxrotnosy/DP-gluosyl trasnsferaseh famly prot einzye [At atvg55710.1]
9-eais-epoxycrotenoi diox fmygenase/noxn Ithi cevae nympuatv [t1739
li Hin picale meritemn (AM) amlyprten Atg238.1
expressed protein [Atlg38700.1]
oaexpressed protein [At~gl4870.1]
oae familyd protein [Atlgl4860.1]
euxpnressdproteine [Aotei/ng06135.1]idue roen 9(AA9 [tg55
auxinressdpronsive potei/ndlaetcacdidue roen 9(AA9 [tg5501
exresslo/ed protein [At atfaig29780.1]tg083.1
amat/ndilobea-ca rtenin repat fail potin Atg083.
MutT/nuedixfm protein [At3g73540.1]
expressebidpoting [F Atd ag03280.1]Atg130.
s~calcium-bindraing F and iamily protein-ead [At4g10300.1]
sexnescnce/dhydationasoiae proeinrelte [At~g35985.1]
expanseinl puateive (EXP11 [At20.190
hypotheatihcal poterina [Ati-cntg35200.1] Atg2S1.1
DnAJni heat shock N-tminal doaincotanig roei [t.21101
anionic peroxidase, putative [Atlg14540.1]
nioneic p-earoxdae putgal tive [Atlgl4550.1] tg084.1
immedicate-earlyt uNgal lcio family protein [At~g02840.1)
glno apical eristemn (NAM) amlyprten Atg477.
glyc inger(C2h2 prtyein [At iy g07135.1] Atg409.
zinc faingepr C2H2 ty1/pe)/Y family protein (ZA7 At2g46090.1]
EXSc family protein/ERD1/XPR1/SYG1 faily protein [tg3295.1
zin figelctr (C2H2 ptyefamily protiaein [Atatg46080.1]80.
epolygalactrotense putatvepetiaepuatve[t.4390
epressed ptrotaein [Atatvg50335.1]uatve[t~5950
auxin-responsive protein/indoleacetic acid-induced protein 1 (IAA1) [At4g14560.1]
expolygalactrotense putatvepetiaepuatve[t.59501
Expressed protein [Atg411030.1]
exprmaessed prbxlotei [At atvg34340.1]0.
glutx amate ecrboxyla/seputtv A10001
M F-bo family prnciteion/SKP1 interacting partner 3-related [At2g02230.1]
mybfamly rancritio fator(KAN4) [At5g42630.1]
expansion, putative (EXPl2) [At39l5370.1]
Hypothetical protein [Atlg20530.1]
auxin-responsive protein, putative/small auxin up RNA (SAURD) [At2g24400.1]
SKP1 interacting partner 4 (SKIP4) [At3g61350.1]
disease resistance family protein [At2g34930.1]
Whistidine biosynthesis bifunctional protein (HISIE) [Atlg31860.1]
expansion, putative (EXP9) [At5g02260.1]
expansion, putative (EXP17) [At4q01630.1]
ethylene-responsive element-binding protein, putative [At5g25190.1]
potassium transporter (HAK5) [At4gl3420.1]
Unknown [CHR2:009686022-009686081]
expressed protein [At1g74450.1]
auxin-responsive GH3 Eamily protein [At2g14960.1]


Figure 3-9. Continued












CBL-interacting protein kinase 9 (CIPK9) [AtlgO1140]
oxidoreductase, 20G-Fe(II) oxygenase family protein [At3g11180.1]
F-box family protein / SKP1 interacting partner 3-related [At2902320.1]
expressed protein [At2g34170.1]
auxin-responsive family protein [At4gl2410.1]
expressed protein [At5g02010.1]
Unknown
hypothetical protein [At~gl9200.1]
basic helix-loop-helix (bHLH) family protein [At2g22770.1]
MutT/nudix family protein [Atlgl8300.1]
basic helix-loop-helix (bHLH) family protein [At2g22760.1]
leucine-rich repeat family protein [Atlg78230.1]
polygalacturonase, putative / pectinase, putative [At2g43870.1]
glutamyl-tRNA reductase 2 / GluTR (HEMA2) [Atlg09940.1]
senescence-associated protein-related [At2g25690.1]
exocyst subunit EXO70 family protein [At2g28650.1]
cyclic nucleotide-binding transporter 2 / CHBT2 (CNGC19) [At3q17690.1]
beta-amylase, putative / 1,4-alpha-D-glucan maltohydrolase, putative [At3g23920]
basic helix-loop-helix (bHLH) family protein [At4g05170.1]
auxin-resistance protein, putative [At2g332410.1]
armadillo/beta-catenin repeat family protein [Atlg60190.1]
AP2 domain-containing transcription factor, putative [At5g65130.1]
AP2 domain-containing transcription factor, putative [At4g34410.1]
AP2 domain-containing transcription factor, putative [Atlg22810.1]
Unknown
U-box domain-containing protein similar to CMPG1 [At3g52450.1]
touch-responsive protein / calmodulin-related protein 2 [At5g37770.1]
tobamovirus multiplication protein 3, putative [Atigl4530.1]
syntaxin 21 (SYP21) / PEP12 homolog [At5gl6830.1]
S-adenosyl-L-methionine carboxyll methyltransferase family protein [At~g44870.1]
protein phosphatase 2C family protein [At3g09400.1)
potassium transporter family protein [At4gl9960.1]
pentatricopeptide (PPR) repeat-containing protein [Atlg09680.1]
nodulin MtN21 family protein [At5g40210.1]
nodulin family protein [At2gl6660.1]
no apical meristem (NAM) family protein similar to NAC1 [Atdg01540.1]
malate dehydrogenase, cytosolic, putative [Atig04410]
lipoxygenase, putative [Atlg72520.1]
lectin protein kinase, putative similar to receptor lectin kinase 3 [At1g70130.1]
hypothetical protein [At3g56670.1]
hypothetical protein [Atlq7536i0.1]
hypothetical protein ; expression supported by MPSS [At3g09950.1]
harpin-induced protein-related / HIN1-related [At2g27080.1]
glutathione S-transferase (103-1A) [At2g29450.1]
gibberellin 2-oxidase, putative similar to GA2ox2 [Atlg02400.1]
expressed protein similar to axi 1 [At5g01100.1]
expressed protein hin1 protein [At3g54200.1]
expressed protein [At3gl5810.1]
expressed protein [At5g20190.1]
expressed protein [At5gl3220.1]
expressed protein [At5gl2010.1]
expressed protein [At~g11070.1]
expressed protein [At4g27654.1]
expressed protein [At4g27652.1]
expressed protein [At2g21500.2]
expressed protein [Atlg74950.1]
expressed protein [Atig56660.1]
expressed protein [Atig20310.1]
expressed protein ; expression supported by MPSS [At4g21500.1]
expressed protein [At5g37550.1]
expressed protein [At5g03210.1]
expressed protein [At4g01360.1]
expressed protein [At2g20240.1]
expressed protein [At2gl8970.1]
ethylene-responsive element-binding family protein [Atig28160.1]
esterase/1ipase/thioesterase family protein [At2g39420.1]
cytochrome P450 family protein similar to Cytochrome P450 94A1 [At3g01900.1]
cyclin, putative similar to mritotic eyelin a2-type [Atlg77390.1]


Figure 3-9. Continued












expressed protein [At2q43340.1]
AP2 domain-containing protein [At5g52020.1]
expressed protein [Atlg26210.1]
eukaryotic translation initiation factor 5, putative [Atlg77840]
expressed protein [At5g01970.1]
hydrolase, alpha/beta fold family protein [Atig52750.1]
expressed protein [Atlg22882.1]
calmodulin, putative similar to calmodulin NtCaM13 [At3g10190.1]
expressed protein [At4g00770.1]
cyclic nucleotide-gated channel (CNGC1) [At5g53130.1]
myosin heavy chain-related similar to myosin heavy chain [Atlg64330.1]
F-box family protein-related similar to A3 protein [At2g03530]
expressed protein [At3g08600.1]
disease resistance protein (TIR class), putative [Atlg61100.1]
expressed protein ; expression supported by MPSS [Atlg09812.1]
cyclin family protein similar to cyclin 2 [Trypanosoma brucei] [At3g63120.1]
microtubules associated protein (MAP65/ASE1) family protein [At2g01910.1]
lectin protein kinase family protein [At5g60280.1]
protein phosphatase 2C family protein [At5g66080]
plastocyanin-like domain-containing protein [At1g64640.1]
expressed protein weak similarity to enterophilin-2L [At3g52920.11
expressed protein [At2g20142.1]
Unknown
expressed protein [At1g72240.1]
DNA cross-link repair protein-related [At3g26680.1]
expressed protein predicted protein, Arabidopsis thaliana [At4g35320.1]
xyloglucan:xyloglucosyl transferase, putative [Atlg11545.1]
UDP-glucose 4-opimerase, putative [Atlg64440]
xyloglucan :xyloglucosy1 transferase, putative [At4g03210]
SEC14 cytosolic factor/phosphoglyceride transfer family protein [Atlg30690]
glycoside hydrolase family 28 protein/polygalacturonase (pectinase) [At4g33440.1]
beta-expansin, putative (EXPB1) identical to beta-expansin [At2g20750.1]
hypothetical protein [At5g35870.1]


Figure 3-9. Continued












protease inhibitor/seed storage/1ipid transfer protein (LTP) family protein [At4gl2510.1]
protease inhibitor/seed storage/1ipid transfer protein (LTP) family protein [Atlg62510.1]
arabinogalactan-protein, putative (AGP22) [At5g153250.1)
organic cation transporter-related [Atl1l6390.1]
calcium-transportingJ ATPase, plasma membrane-type, putative (ACA12) [At3g63380.1]
FAD-binding domain-containing protein [At4g20840.1]
expressed protein [Atl1l3540.1]
phosphate-responsive protein, putative (EXO) [At4g08950.1]
amino acid transporter family protein [At3g56200.1]
expansion family protein (EXPL2) [At4g38400.1]
expressed protein [At4gl7220.1]
disease resistance protein (TIR-NBS-LRR class), putative [At4gl6950.1]
glycosyl hydrolase family 9 protein endo-1,4-beta-glucanase precursor [At4g39000.1]
disease resistance protein (TIR-NBS class) [At1g72890.1]
speckle-type POZ protein-related [At3g56230.1]
monogalactosyldiacylglycerol synthase, putative [At5g20410.1]
transferase family protein [At~g02890.1]
fasciclin-like arabinogalactan-protein, putative [At5g44130.1]
arabinogalaetan-protein (AGP4) [At5g10430.1)
protein kinase family protein [At4g00330.1]
expressed protein [Atlg53625.1]
tasciclin-like arabinogalactan-protein (FLA9) [Atlg03870.11
expressed protein [At5g24105.1]
calcium-binding protein, putative [Atlg21550.1]
calcium exchanger (CAX1) [At2g38170.1]
DC1 domain-containing protein [At2gl7590.1]
SEC14 cytosolic factor/phosphoglyceride transfer family protein [Atiq22530.1]
leucine-rich repeat transmembrane protein kinase, putative [At5g24100.1]
expressed protein [At4gl4380.1]
expressed protein [At3g07470.1]
exostosin family protein [At5g41250.1]
protein kinase family protein [At5g61570.1]
calmodulin, putative [At2g41410.1)
WRKY family transcription factor DHA-binding protein 4 [At3g56400.1]
protease inhibitor/seed storage/1ipid transfer protein (LTP) family protein [At2gd5180.1]
calmodulin-binding protein [At5g26920.1]
RHA recognition motif (RRM)-containing protein [Atlg22330.1]
BOH1-associated protein (BAPl)-related [At2g45760.1]
AV802416 Arabidopsis thaliana cDNA clone RAFL09-31-N12 3', mRNA sequence (AV802416]
plastocyanin-like domain-containing protein [IAtlg08500.1]
MutT/nudix family protein [At4gl2720.1]
expressed protein [Atlg05880.1]
expressed protein [At5g54240.1]
UDP-glucoronosyl/UDP-glucosyl transferase family protein [At2g26480.1]
disease resistance protein (TIR-NBS-LRR class), putative [At5g41750.1]


Figure 3-9. Continued












DC1 domain-containing protein [At5g54030.1]
U-box domain-containing protein [Atl901680.1]
RHA recognition motif (RRM)-containing protein [Atlg78260.1]
expressed protein [At3gl5115.1]
expressed protein [Atlg06930.1]
disease resistance protein (TIR-NBS-LRR class), putative [Atdgl6920.1]
tumor susceptibility protein-related [At2g38830.1]
expressed protein [At4g37240.1]
ankyrin repeat family protein [Atlg11740.1]
Unknown [CHR1:021477409-021477350]
expressed protein [Atlg22335.1]
disease resistance protein (MBS-LRR lasss, putative [At5g38350.1]
arabinogalactan-protein (AGP7) [At5g65390.1)
MADS-box protein (AGL45) [At3g05860.1]
leucine-rich repeat family protein [At4gl8760.1]
expressed protein [At2gl7300.1]
lysine and histidine specific transporter, putative [At3g01760.1]
arabinogalactan-protein (AGP21) [At1g55330.1]
Unknown [CHR1:007886848-007886789)
expressed protein [At~ig65340.1]
disease resistance protein (TIR-NBS-LRR class), putative [Atdgl6860.1]
disease resistance protein (TIR-NBS-LRR class), putative [At5g46470.1]
expressed protein [At3g07460.1]
neurofilament protein-related [At3g05900.1]
expressed protein [Atlg47740.1]
expressed protein [At3g50120.1]
expressed protein [At4g37235.1]
armadillo/beta-catenin repeat famnily/Ul-box domain-containing protein [AtLg23030.1]
protein kinase family protein protein kinase family [At3g24720.1]
fasciclin-like arabinogalactan-protein (FLA2) [At4gl2730.11
rhomboid family protein [Atdg23070.1]
glycoside hydrolase family 28 protein/polygalacturonase family protein [At3g06770.1]
arabinogalactan-protein (AGIP17) [At2g23130.1]
expressed protein [At5g25240.1]
TCP family transcription factor, putative [At5g23280.1)
exostosin family protein [At5g22940.1]
arabinogalactan-protein (AGP17) [At2g23130.2]
TCP family transcription factor, putative [At5g08330.1]

beta-amylase (BMY1) / 1,4-alpha-D-glucan maltohydrolase [At4gl5210.1]
leucine-rich repeat family protein [At5g66330.1]
expressed protein [Atlg22110.1]
DNAJ heat shock Ni-terminal domain-containing protein [At2g47440.1]
BP611361 RAFL16 Arabidopsis thaliana cDNA clone [BP611361]
hypotetical protein [Atl1g7620.1]
expressed protein [At59g1070.1]
scarecrow transcription factor family protein scarecrow-like 11 [At3g46600.1]
heat shock transcription factor family protein [At3g63350.1]
DC1 domain-containing protein [Atlg55440.1]
tribelix DNA-binding protein / GT-2 factor (GT2) identical to GT2 actor [At1g76890.1]
proton-dependent oligopeptide transport (POT) family protein [At5g46040.1]
ubiquitin family protein [At1g22050.1]
Toll-Interleukin-Resistance (TIR) domain-containing protein [At2g20145.1]
protein kinase, putative [At3g57730.1]
protein kinase family protein [At3g09780.1]
OTU-like cysteine protease family protein [At2g27350.1]
O-methyltransferase, putative [Atlg21100.1]
leucine-rich repeat transmembrane protein kinase, putative [Atl1l2460.1]
hydroxyproline-rich glycoprotein family protein [Atlg23050.1]
expressed protein predicted proteins, Arabidopsis thaliana [At4g23870.1]
expressed protein predicted protein, Arabidopsis thaliana [At5gl9190.1]
expressed protein [Atl1l9835.1)
expressed protein ; expression supported by MPSS [At3g30580.1]
expressed protein [At5g62280.1]
expressed protein [At4gl6400.1]
expressed protein [At2g29525.1]
expressed protein [Atlg67230.1]
expressed protein [Atlg51430.1]
eukaryotic translation initiation factor 5, putative [Atlg36730]
cyclin family protein similar to cyclin 2 [At2g45080.1]
AP2 domain-containing transcription factor, putative [Atlg22985.1]
zinc finger (C3HC4-type RING finger) family protein [At~gl6720.1]
U-box domain-containing protein similar to CMPG1 [Atlg49780.1]


Figure 3-9. Continued












B L-methionine C


L-methionine
SAM-1 AT4G01850
AT3G17390 AT2G36880
S-adenosyl- L-methionine
ACS10 ACS1 ACS12
ACS2 ACS4 ACS51.86


L-methionine
SAM-1 AT4G01850
AT3G17390 AT2G36880
enosyl-L-methionine
ACS10 ACS1 ACS12
ACS2 ACS4 ACS51t22
AT5G28360 ACS8
ACS7 ACS8 ACS11
ACS9 3.58 ATSiG36160
clopro wane-1 -carboxylate
AT3G47190 .'_ `
AT5G43440 0.79 AlT3G61400
AT2G30840 AT2G30830


AT2G19590 AT1G12010 0.621
AT1GO3400 EFE ACO2
ethylene


SAM-1 AT4G01850
AT3G17390 AT2G36880
.-methionine
ACS10 ACS1
ACS12 ACS2 ACS4
ACS5 2.09 AT5G28360
ACS8 ACS7 0.852
ACS6 ACS11 1.24
ACS9 3.38 AT5G36160
mne-1 -carboxylate
AT3G47190
AT5G43450 AT5G43440
AT3G61400 AT2G30840
AT2G30830 AT2G25450
AT2G19590 AT1Gl2010
AT1G03400 EFE ACO2


S-adenosyl-









1-aminocycloprop


S-ade


AT5G28360 ACS8
ACS7 ACSG ACS11
ACS9 AT5G36160
me-1-carboxylate
AT3G47190
AT5G43450 AT5G43440
AT3G61400 AT2G30840
AT2G30830 AT2G25450
AT2G19590 ATI1012010
AT1GO3400 EFE ACO2
ene


1-aminocycloprop


1-aminocy~


ethyl


ethylene


Figure 3-10a. Differentially expressed transcripts encoding metabolic enzymes in the ethylene
biosynthesis pathway. Up-regulated expression of genes encoding 1-
aminocyclopropane-1 -carboxylate synthase (ACS) at Time 1 (A), Time 2 (B) and
Time 3 (C), and differential expression of genes encoding several enzymes with 1-
aminocyclopropane-1 -carboxylate oxidase activity at Time 3 (C). The Log2 value of
the difference in expression is indicated next to the enzyme or gene identifier in color.










A violaxanthin B violaxanthin C violaxanihin
ra~s-neoxanthin tranrs-neoxanthin tras-neoxanthin
9'-cis-neoxanthin 9'-cis-neoxanthin 9'-cis-neoxanthin

INCED925 NCED2iG310 NCED9 3.31 NCED2 NCED6 NCED9 2.4 NCED2 NCED6
NCED NCE3 A1 G3100NCED3 1.71 AT1 G30100 NCED3 AT1 G30100
xanthoxin xanthoxin xanthoxin
IABA2 APB2 /ABA2
abscisic aldehyde abscisic aldehyde abscisic aldehyde
lAA3 IAAO3 AAO3 0.462
(+)-abscisate (+)-abacisate (+)-abscisate

Figure 3-10b. Differentially expressed transcripts encoding metabolic enzymes in the abscisic
biosynthesis pathway. Up-regulated expression of genes encoding 9-cis-
epoxycarotenoid dioxygenase/neoxanthin cleavage enzyme (NCED) at Time 1 (A),
Time 2 (B) and Time 3 (C), and up-regulated expression of a gene encoding aldehyde
oxidase 3 at Time 3 (C). The Log2 value of the difference in expression is indicated
next to the enzyme or gene identifier in color.

A a phosphatidylcholine~ B a phosphatidylcholine C a phosphatidylcholine
dad1 AT1 G31480 dad1 AT1 G31480 dad1 AT1 G31480
(9,1 2,15}-linolenate (9,1 2,15)-linolenate (9,12,15)-linolenate
LOX1 LOX2 AT3G22400 LOX1 LOX2 LOX1 LOX2
AT1 G72520 1.35 AT3 G22400 AT1 G72520 AT3G22400 AT1 G72520
AT1 G67560 AT1 017420 AT1 G67560 AT1 017420 AT1 G67560 AT1 017420
13(S yhydroperoxylinolenate 13(S )hydroperoxylinolenate 13(8)-hydroperoxylinolenate
AOS AOS AOS
12E,1 3(S)-epoxylinolenate 12,13(S)-epoxylinolenate 12,13(85epoxylinolenate
A`T3G25770 AT3G25780 AT3G25770 AT3G25780 AT3G25770 AT3G25780
AT3G2576i0 ATI1G3280 AT3G25760 AT1G13280 AT3G25760 AT1G13280
12-oxo-cis-1 0,1 5-phytodienoate 12-oxo-cis-1 0,1 5-phytodienoate 12-oxo-cis-1 0,1 5-phytodienoate
AT1 G76680 AT1 G76680 AT1 G76i680 -21.41
AT2GO6050 AT1G76690 AT2006050 AT1G76690
3-oxo-2-(cls-2--pentenslrl- 3-oxo-2-(cis-2'-pentenyl
cyclopentane-1 -octanoate cyclopentane-1 -octanoate
I.,,,,.,I.,,,,.,3-oxo-2-(cis-2'-pentenyl)-
-,*+th--l. -e~othe. Icyclopeniane-1 -octanoate
(+}7-isojasmonate (+}7-isojasmonate Ihypothetical

(-)-asmonate (-)-asmonate ()7ioamnt

IAT1 619640 AT1 19640 (-)-jasmonate
(-ljasmonic acid methyl ester (-}jasmonic acid methyl ester lAT10964
(-)-asmonic acid methyl ester
Figure 3-10c. Differentially expressed transcripts encoding metabolic enzymes in the j asmonate
biosynthesis pathway. Up-regulated expression of a gene encoding lipoxygenase at
Time 1 (A) and down-regulated expression of several genes encoding 12-
oxophytodienoate reductase at Time 3 (C). The Log2 value of the difference in
expression is indicated next to the enzyme or gene identifier in color.












A GA20
IGA~ox3
GA29
IG~ox3
GA~veatabolite



B GA2o
CA~lox3
GA29
IGA~x3
GA~g-catabolite


GAs
GA2ox2 GA2ox1 GA~ox3 GA2ox2
AT10 02400 0.6i56
GA2oxI GA2ox2 AT1G47990 GA2ox1
GAs


GP~og GA40x GA~i2
GAsjox3o GA~ox GA2ox1 GA~ox1 GACox2 G~Alo GA12o3Go1GAo2
GAsi GAs* GA20x8 1.56 GA20x7
SGAi~ox3 GA~ox1 GA~ox2 GAE~ox GA~ox2 G~a~o3 G~ile
GAsi-catabolite GAz4-catabolite


GAss GA44
GA2ox3 GA~ox1 GA~ox2 GAP~ox3 GA2oxt GA~ox2a
GA2ox8 1.56 GA2ox7 G~ge
GAgy


IGA~ox1 GA2ox2 GAb~ox3
GAg-catabolite
GAI GAs GA4 GAi2
GA2ox2 GA2ox1 GA~ox3 GA2ox2 AT1G02400 GA2ox3 GA2ox2 GA2ox1 GA2ox1 GA2ox2 GA2ox3 GA2ox3 GA2ox1 GA2ox2
AT1G47990 GA2ox1 GAst GAs+ GA~ox8 GA2ox71.82
GA2ox1 GA2ox2 GA, GAP2ox3 GA~ox1 GA2ox2 IGA~ox1 GA2ox2 GA~ox3 GAl1o
IGA2ox1 GA2ox2 GA2ox3 GAsi-catabolite GAz4-catabolite
GArcatabolite


GA~ox3 GA~ox1 GA~ox2 GA~ox3 GA~ox1 GA~ox2
GA~ox8 GA2ox71.82 GAss


GA



GAs

ss-cat


Figure 3-10d. Differentially expressed transcripts encoding enzymes involved in gibberellins metabolism. Up-regulated expression of
genes encoding gibberellin 2-oxidase (Ga2ox8, Ga2ox7) at Time 1 (A), Time 2 (B) and Time 3 (C), and down-regulated
expression of a gene encoding gibberellin 2-oxidase (Ga2oxl) at Time 3 (C). The Log2 value of the difference in
expression is indicated next to the enzyme or gene identifier in color.


C GA20 GA,

IGA2ox3 GA2ox2 GA~ox3 GA2ax2
GA2oxi -1 72 AT1GO2400 AT1G47990
GA2s GAb2ox1 -1.72
SGA~ox3 GA2ox1 -1 72 GA~ox2 GA,
GA2rcatabolite GAlox1 -1 72 GA!
GA~ox2 GA2ox3
GArcatabolite


g GA4 GA12 GASS GA44
GAr2ox3 GA~ox2 GACi~ox1-1 72 GA~u2ox3 GA~oox17 -172GALox3 GA201-1.72 GA-2ox3 G~oi'A2ox1- 7 A~x
GA2ox1 -1.72 IGA2ox2 GAOX3 GA C2ox2 G~ox8 1 68 1 GA2ox2 GA2ox8 1.68 GAge
iGAs+ GA2ox7 1.54 GA C2ox7 1.54
GA2ox3 GA~ox1-1 72 GA2ox2 GA~o~x1-1 72 GAllo GA97
abolit! 1GA2ox2 GA2ox3
GAz4-catabolite









a 1,2-ciacylglycerol
IMGD3 MGD2-0.478 MGD1
a monogalac~tosyldiacylgycerol

a p,p digalactosyldi~acyvlglcerol a trigalactosylcliac:ylglycerol
a digalactosyldiacylglycerol


a 1,2-diacylglycerol
SMGD3


MGD1


a monogalactosyldiacylgIycerol

a p,p digalac:tosyldiacylglycerol a trigalactosyldiac:ylglycerol
a digalac:tosyldiacylglycerol


a 1,2-diacylglyceral
MILGD3


MGD1


a monogalactosyldiacylgycerol

a P,B digalactosyldiacylglycerol a trigalactosylcliacylglycerol
a digalac:tosyldiacylglycerol

Figure 3-10e. Differentially expressed transcripts encoding enzymes involved in glycerolipid
metabolism. Down-regulated expression of a gene encoding
monogalactosyldiacylglycerol synthase 2 (MGD2) at Time 1 (A), Time 2 (B) and
Time 3 (C). The Log2 value of the difference in expression is indicated next to the
enzyme or gene identifier in color.


A D-fructose-6-phosphate

D-glucosamine-6-phosphate
IGNA1
N-ac:etyl-D-glucosamine-6i-phosphate
IAT5G18070
N-acetyl-glucosamine-1I-phosphate
UDP-N-acetyl- D-glucosamine


B D-fructose-6-phosphate
InAT3G2090
D-glucosamine-6-phosphate
IGNA1
N-acetyl- D-gluco~samine-6-phosphate
AT5G18070
N-acetyl-glucosamine-1 -phosphate
UD P-N-acetyl- D-glucosamine


C D-fru ctose-6 -pho sphate
SAT3G24090 0.308
D-glucosamine-6-phosphate
IGNA1I
N-ac etyl- D- glucosamine-6-phosphate
IAT5G18070
N-acetyl-glucosamine-1l-phosphate
UD P-N-acetyl- D-glucosamine


Figure 3-10f. Differentially expressed transcripts encoding metabolic enzymes in the hexosamine
biosynthetic pathway. Up-regulated expression of a gene encoding glucosamine-
fructose-6-phosphate aminotransferase at Time 1 (A) and Time 3 (C). The Log2 value
of the difference in expression is indicated next to the enzyme or gene identifier in
color.










B C


chlorophyll la
CL2 CLH1
phytol chlorophyllide a
pheophorbide a

RCC pyropheaphorbide a
RCCR

pFCC


2LH CLH1
phytol .hlorophyllidt a
pheophorbide a

RCC pyropheophorbide a
IRCCR
pFCC


2LH CLH1
phytol chlorophyllide a
pheophorbide a
0 ,
RC pyropheophorbide a
IRCCR
pFCC


Figure 3-10g. Differentially expressed transcripts encoding enzymes involved in chlorophyll
breakdown. Up-regulated expression of a gene encoding pheophorbide A oxygenase
(PAO) at Time 1 (A) and Time 3 (C). The Log2 value of the difference in expression
is indicated next to the enzyme or gene identifier in color.










L-glutomate!
/AT5G0ie50 AT15G26710
L-glutamyl-tRNAel
HEtMA2 HEMA1 At2g31250i
glutamate-1 -semialdehyde
GSA1 AT3G48730
5-amino-levulinate
HnEMB1 ATl1 G4318
porphobilinogen
IHEMC
hydroxymethylbilane
AT2G26540


B L-glutamate
AT5G64050 AT5G26710
L-glutamyl-tRNAlu
HEMA2 0.403
H-EMA1 At2g31250
glutamate-1 -semialdehyde
GSA1 AT3G48730
5-amino-levulinate
IHEMB1 AT1 G44318
porphobilinogen
IHEMC
hydroxymethylbilane
AnT2G2540
uroporphyrinogen-Ill
14930 AT2G40490~ AT5G40850
coproporphyrinogen III precorrin-1
IHEMF1 AT4GO3205 IAT5G40850
protoporphyrinogen IX precorrin-2
IPPOX AT5G14220O sirohydrochlorin

AT5G26030 siroheme
AT2G30390 AT10192100
protoheme 18
home a


uroporphyrinogen-llI
1~4930 AT2G40490 ~ AT5G40850
coproporphyrinogen III precorrin-1
HnEMF1 AT4GO3205 JAT5G40850
protoporphyrinogen IX precorrin-2
IPPON AT5G14220 sirohydrochlorinI


protoporphyrin IN
AT5G26030
AT2G30390 AT1G19200
protoheme IX
heme o


SAtSirB 0.142
siroheme


Figure 3-10h. Differentially expressed transcripts encoding enzymes involved in porphyrin and chlorophyll metabolism. Up-regulated
expression of a gene encoding glutamyl-tRNA reductase 2 (HEMA2) at Time 1 (A) and Time 2 (B), and up-regulated
expression of a gene encoding sirohydrochlorin ferrochelatase (AtSir B) at Time 2 (B) and Time 3 (C). The Log2 value of
the difference in expression is indicated next to the enzyme or gene identifier in color.


A L-glutamate
AT5G64050 AT5G26710
L-glutamyl-tRNAl
HEMA2 0.765
H1EMA1 At2g31250
gllutamate-1 -semialdehyde
GSA1 AT3G48730
5-amino-levulinate
/HEMB1 AT1 G44318
porphobilinogen
JHEMC
hydroxymethylbilane
AT2G240
uroporphyrinogen-lli
1~49530 AT2G40490~ AT5jG40850
coproporphyrinogen III precorrin-1
IHEMF1 AT4GO33205 IAT5G40850
protoporphyrinogen 18 precorrin-2
PPOXAT5G4220 sirohyd ochlorin
protoporphyrin IX AtirB
AT5G26030 siroheme
SAT2G30390 AT1G19200
protohemeIX
home a










A 5-phosphoribosyl 1-pyrophosphate
A~tATP-PRT 1 AtATP-PRT2
phosphoribosyl-ATP

phosphoribosyl-AMP

pho sp or Ibosylf'or mimno AICA: R-phof ph4 te
AT2G36230
phosphoribulosylformimino-AI CAR- P
IAT4G26900
AIC ~D-erythro-imridazole-gllycerol-phosphae
(AT3G22425 AT4G14910
imidarzole acetol-phosphate
IAT5G10330 AT1G71920
L-histidinol-phosphate
histidinoI
AT5G3890
histidinal
AT56390
L-histidine

C 5-phosphoribosyl 1-pyrophosphate
A~tATP-PRTI 1 AtTP-PRT
phosphoribosyl-ATP
IAT1 G1860 027
:Ihosphorl~bosybAMP
SAT1 G31860 0.27
phosphoribosylformiminoAI CAR-phosphate
AT2G36230
phosphoribulosylformimino-AI CAR P
IAT4G26900
Al OAR D-erythro-imnidazole-glycerol-phosphat
JAT3G2425 AT4G14910
imidazole acetol-phosphate
ATS5G10330 AT1lG71920J
L-histidinol-phosphate
histidinol
/AT5G83890
histidinal
IATSG6389
L-histidine


B 5-phosphoribosyl 1-pyrophosphate
AtAP~TP-PRT1 AtATP-PRT2~
phosphoribosyl-ATP
AT G31860ss 0 241
phosphoribosyl-AM P
SAT1 G31860 0 241
phosphoriborsylformiminoAI CAR-phosphate
AT2G36230
phosphoribulosylformimino-AI CAR- P
IAT4G28900
AI C. RD-erythro-imidazole-glycerol-phosphat
AT3G22425 AT4G14910
imidazole acetol-phosphate
(AT5G10330 AT1G71920
L-histidinol-phosphate
histidinal
(AT5G63890
histidinal
IAT5iG63890
L-histidine


Figure 3-10i. Differentially expressed transcripts encoding metabolic enzymes in the histidine
biosynthesis pathway. Up-regulated expression of a gene encoding phosphoribosyl-
ATP pyrophosphohydrolase at Time 1 (A), Time 2 (B) and Time 3 (C) The Log2
value of the difference in expression is indicated next to the enzyme or gene identifier
in color.












A D-galactose
ARA1 AGK1 AT5G14470
AT3G01640 AT3G10700
AT3G42850 AT1001220
c- D-galactose1I-phosphate
AT5G52560

SAT5G18200 AT5G17310
UDP-galactose
UGE4 -0 577 UGE3
UGE2 UGE5 UGE1
LIGE4 -0 577 AT3G28530
SUS5 SUSI SU UDP-D-glucose 171 AJ296 A1267
SUS5SUS1SUS3AT5G52560 AT5G39320 AT5G15490
SUS4 SUS6 SUS2ATG30ATG71AT230A1260
sucrose m- D-glucose l-phosphate UDP-D-glucuronate

B D-galactose
ARA1 AGK1
ATSG14470 AT3G01640
AT3G10700 0.265
AT3G42850 AT1G01220
pc-D-galactose l -phosphate
AT5G52560

SAT5G18200 AT5G17310
UDP-galactose


UGE2 i..il-. 1 UGE1
ATl-C I:I:P3G28530


SUS5 SUS1 SUS3 I AI5G52560 AT5G39320 AT5G15490
SUS4 SUS6 SUS2 I :;Ir-I ::rr AT5G17310 Ii~':i: IIi AT1G26570
SUCrose eL-D-glucose l -phosphate UDP-D-glucuronate

C D-galactose
ARA1 AGK1
;s~l 1 AT3G01640
AT3G10700 0.29 AT3G42850
AT1 G01220 0.248
o-D-galactose 1-phosphate
rT1;5,5d50 AT5G18200 0.326


UDP-galactose
UGE4 UGE3 UGE2 UGES
UGE1 UGE4 AT3G28530
UD-D-glucose
SUS5 SUS1 SUS3 TE550 AT5G39320 AT5G15490
SUS4 SUS 6SUS2 ,i H 11; T3G293i0 -0.225 AT1 G26570
sucrose el-D-glucose 1-phosphate UDP-D-glucuronate


Figure 3-10j. Differentially expressed transcripts encoding metabolic enzymes involved in
galactose metabolism. Down-regulated expression of a gene encoding UDP-glucose
4-epimerase (UGE4) at Time 1 (A) and Time 2 (B). The Log2 value of the difference
in expression is indicated next to the enzyme or gene identifier in color.











































PGK
1,3-diphospi


3-phosph



2-phosphi



phosphoe


pyruvate


Figure 3-10k. Differentially expressed transcripts encoding enzymes involved in glycolysis. Up-
regulated expression of genes encoding phosphofructokinase at Time 1 (A) and Time
3 (C). The Log2 value of the difference in expression is indicated next to the enzyme
or gene identifier in color.


e a--- mvCCXaroxucrne onesanate


A 0-D-glucose-6-phosphate
(PCll AT5G42740
D-fructose-6-phosphate


B P-D-glucose-

D-fractose-l


6-phosphate
(PGll ATSG42740
6-phosphate
MS5G56630
AT5G6~1580 AT5G47810
AT4G29220 AT4G32840
ATG26270 AT2G22480
bisphosphate
AT5GO3690
AT438970 AT4G26530
AT4G26520 AT3G52930
AT2G336460 AT2G01140
AT2G21330 AT1G18270
5 @4L TG1170


P-D-glucose-6-phosphate
SPGIl AT5G412740
D-fructose-6-phosphate
AT5G566300.17 AT5G61580
ATSG47810 0.391
AT4G29220 AT4G32840
AT4G26270 AT2G22480
fructose-1,6-bisphosphate
AT5GO3690
AT4G38970 AT4G26530
A4250AT3G~52930
AT2G36460 AT2G01140
phosphate AT2G21330 AT1G18270
AT35540..RQ2170.


AT5G61580 AT5G47810
AT4G29220 AT4G32840
AT4G26270 AT2G22480
fructose-1,6-bispholsphate
AT5GO3690
AT4G38970 AT4G26530 ute16
AT4G26520 AT3G52930
AT2G36460 AT2G01140
AT2G;21330 AT1018270
AT31 T2G2170
D-~cradhye*91"i-spat ihydroxyaceton phosphate
CP12-2 PRK GapA-1 T3
GapA-1 GapC-1 GapA-1
CP12-2 PR
GapB GapCp-1 GapA-2
GapCp-2 GapC-2 Gp- a
GapB GapCI
1,3-ciphosphateglycerate
AT43G12780 AIT1GS6190 PGK GapCp-2
3-phsphalyceate1,3-diphosp ateglycerate
rj-phophoglcertAT3Gn12780 i
A8T4GO9520
AT3G08590 AT1G09780 3phshgyrte
AT4GO!
2-pholsph glycerate
AT2G6530AT3G08590 r
AT2G9560AT1G4030 2-phosphoglycerate
AT2G31
phosphoenopyruvate
PKp3 PKpl PKp2 A2250
PKpl AT5GS3680 poponprvt
PKp3 PKp
AT5G56350 AT5G08570
AT4G26390 AT3GS2990 PlA
AT3G55650 AT3G55810 A5530
AT3G49180 AT3GD4050 A4230
AT3G2596i0 AT2G36580 A3560
pyruateAT3G49160 ,
AT3G25960 r
pyruvate


3-phosphat


e-


dihydrox~yaceton
IK GapA-1
D-glyceraldehydl
C-1 GapA-1
p-1 GapA-2


CPt2-2 PRK( GapA-1


GapA-1 GapB GapCp-1
GapA-2 GapCp-2 GapC-2
lateglycerate
IAT3G12780 ATIG56190 PGK
~glycerate
AT4GO9520
AT3Ga8590 ATIGO9780
aglycerate
A12G36530 AT2029560n-0 222


agyruvate
Pl~p3 PKpl PKp2
PKpl AT5G63680
AT5G56350 AT5G08570
AT4G26390 AT3G52990
AT3G55650 AT3G55810
AT3G49160 AT3G04050
AT3G25960 AT2G36580


GapC-2

AT1GS6190


9950
AT1009780


6530
AT1G74030


l1 PKp2
;G63680
AT5GO8570
AT3GS2990
AT3G55810
AT3GO40SO
AT2G36580











A afattr acid
AT3G05970 AT5G27600
AT1 G77590 AT1 G49430
acetyl-CoA AT2G47240 AT2GO4350
acyl-C

AT5G65110 AT3G 4
AT5G65110 AT4G16
AT2 4T5 KAT1
AT3G51840 AT1GO6300
AT1G06290 AT3G06690
AT2G35690 AT1G06;310
a 3-ketoacyl-CoA a trans-2- .noyl-CoA

AT3G6860 AT4G29010 AT3GO6860
AT5G43280 AT4G160





a( L3%tShydroxyacyl-CoA

AT2G e780 AT3Gi5020)


glyoxylate oxaloacetate
succinate AT3G58750 AT2G42790
SDH8 SDH7 SDH71 AT3G21720 AT1G21440
AT3G58740 AT3G60100
SDH6 SDH5 SDH4 ioirt i t
AT2GO5710
SDH3 SDH2 SDH1
AT2GO5710
fumarate
nrscossonrrcnlo cisaconAT4eG26970 AT4G35830
mate
AT1G53240 AT2G22780
AT5GO9660 AT3G53910
r~l-l1iI3;1AT3G15020
AT3G47520 AT5G56720
AT5G58330 AT5G43330
oxaloacetate
PPC4 AT3G42628
PPC1 PPC2 PPC3
phosphoenopyruvate


Figure 3-101. Differentially expressed transcripts encoding enzymes in the citric acid cycle. Up-
regulated expression of a gene encoding malate dehydrogenase (cytosolic) at Time 1
(A), and down-regulated expression of a gene encoding malate dehydrogenase
(mitochondrial) at Time 3 (C). The Log2 value of the difference in expression is
indicated next to the enzyme or gene identifier in color.











B a fatty acid
AT3G05970 AT5G27600
AT1 G77590 AT1 G49430
acetyl-CoA AT2G47240 AT2GO4350
acyl-C

AT5G65110 AT3 4
AT5G65110 AT4G16
AT2 t~T5 KAT1
AT3G51840 AT1G06300
AT1G06290 AT3G06690
AT2G35690 AT1G06310
a 3-ketoacyl-CoA a traw2- noyl-CoA

AT3G6860 AT4G29010 AT3G06860
1 ~AT5G43280 AT4G160







A 270AT3G1
A1G04410 AT3G53910
glyoxylate oxaloacetate
succnateAT3G58750 AT2G42790
SDH8 SDH7 SDH7sucnt AT3G21720 AT1 G21440
AT3G58740 AT3G60100
SDH6 SDH5 SDH4 isc aecitrate
SDH3 SDH2 SDH1ATG71
fumarae 00510 AT4G26970

AT5G50950 AT2G47510 cis-aconitaterc '1G5:0i1
malate
AT1G53240 AT2G22780
AT5GO9660 AT3G53910
AT1G04410 AT3G15020
AT3G47520 AT5G56720
AT5G58330 AT5G43330
oxaloacetate
PPC4 AT3G42628
PPC1 PPC2 PPC3
phosphoenoityruvate


Figure 3-101. Continued



























a 3















suct
SDH8 SDH7 SDH7
SDH6 SDHS SDH4
SDH3 SDH2 SDH1
fum
ATSG50950 AT2G47510


a fatty acid
AT3G05970 AT5G27600
AT1 G77590 AT1 G49430
A AT2G47240 0.548 AT2GO4350
5an acyl- CoT AT4 G 16760
AT5G65110 AT3G51840
AT5G65110 AT4G16760
5 KAT1AT3G51840 ATI G06300
AT1G06290 AT3G06690
;111'1!1-- -: -AT1 G06310


AT4G29010
AT5G43280


270AT3G1
T1 04410 AT3G5391
ylate! oxaloacetate
AT3G58750 AT2G42790
AT3G21720 AT1G21440
AT3G58740 AT3G60100

ATG05710
A005710 AT4G26970

cis-aconitate TG58001


Ai;T1 0.53240 -0 291 AT2G22780
AT5GO9660 AT3G53910
AT1 GD4410 AT3G15020
AT3G47520 AT5G56720
AT5G58330 AT5G43330
oxaloacetate
PPC4 AT3G42628
PPC1 PPC2 PPC3
phosphoenti~yruvate


Figure 3-101. Continued















sugar transporter family protein [At4g04750.1]
sugar transporter family protein [At3g05150.1]
sucrose transporter/sucrose-proton symporter (SUC3) [At2g02860.1]
proton-dependent oligopeptide transport (POT) family protein [At3gl6180.1]
phosphate transporter (PT2) [At2g38940.1]
mitochondrial substrate carrier family protein [At5g61810.1]
chloride channel protein (CLC-b) [At3g27170.1]
calcium-transporting ATPase 2, endoplasmic reticulum-type (ECA2) [At4g00900.1]
C4-dicarboxylate transporter/malic acid transport family protein [At4g27970.1]
ABC transporter family protein [At3g60160.1]
transporter-related [At4gl7550.1]

uproton-dependent oligopeptide transport (POT) family protein [At5g46040.1]
cyclic nucleotide-regulated ion channel (CNGC1) [At5g53130.1]
anion exchange family protein [At3g62270.1]
proton-dependent oligopoptide transport (POT) family protein [At3g47960.1]
ABC transporter family protein NBD-like protein POP [At5g02270.1]
ABC transporter family protein [At5g44110.1]
tonoplast intrinsic protein, putative [At2g25810.1]
cyclic nucleotide-regulated ion channel, putative (CHGC13) [At4g01010.1]
calcium-transporting ATPase 1, plasma membrane-type [Atlg27770.2)
plasma membrane intrinsic protein, putative [At4g00430.1]
ABC transporter family protein transport protein ABC-C [At3g47780.1]
calcium-transporting ATPase, plasma membrane-type, putative (ACA10) [At4g29900.1]
MAE efflux family protein [At2g04070.1]
proton-dependent oligopeptide transport (POT) family protein [At~g38100.1]
delta tonoplast integral protein (delta-TIP) [At3gl6240.1]
copper-exporting ATPase, putative [Atlg63440.1]
anion-transporting ATPase family protein [At5g60730.1]
AFG1-like ATease family protein [At4g30490.1]
sulfate transporter, putative [At3gl5990.1]
phospholipid-transporting ATPase 1/magnesium-ATPase 1 (ALA1) [At5g04930.1]
calcium exchanger (CAX2) [At3gl3320.1]
calci um- transport ing ATPasea, plasma membrane-type, putat ive (ACA9 ) [At3g21 180 .1]i
calcium-transporting ATPase, plasma membrane-type, putative (ACA10) [At4g29900.1]
secretory carrier membrane protein (SCAMP) family protein [Atlg03550.1]
cation exchanger, putative (CAX3) [At3g51860.1]
amino acid transporter family protein [At3g11900.1]
lysine and histidine specific transporter, putative [At3g01760.1]
calcium exchanger (CAX1) [At2g38170.1]
organic cation transporter-related [Atl1l6390.1]
calcium-transporting ATPase, plasma membrane-type, putative (ACA12) [At3g63380.1]
amino acid transporter family protein [At3g56200.1]


Figure 3-11i. Hierarchical average linkage cluster analysis of transporter gene expression using
uncentered correlation. The cluster analysis is based on transporter genes with
significant expression at Time 1, 2 or 3. Yellow denotes a higher, and blue a lower
expression of a gene in the treated plants versus the control. The figure shows that
distinct clusters of expression patterns can be distinguished within the group of
transporter genes across the three comparisons.












plasma membrane intrinsic protein 18 (PIP1B)/PIPL.2 (PIP1.2) [At2g45960.1]
plasma membrane intrinsic protein (SIMIP) [At4g35100.1]
organic cation transporter family protein [At3g20660.1]
nodulin MtN21 family protein [At4g30420.1]
mitochondrial substrate carrier family protein [At2g35800.1]
mitochondrial phosphate transporter, putative [At3g48850.1]
mitochondrial import Timl7/Tim22/Tim23 family protein [At3g10110.1]
MATE efflux protein-related [At4g29140.1]
MATE efflux protein-related [At2g04066.1)
MATE efflux protein-related [Atig58340.11
MATE efflux family protein [At3q23550.1]
MATE efflux family protein [At3g21690.1]
MATE efflux family protein [Atlg73700.1]
MATE efflux family protein [Atlg66760.2]
MATE efflux family protein [Atlg61890.1]
MATE efflux family protein [Atlg51340.1]
MATE efflux family protein [Atigl5170.1]
major intrinsic protein-related/MIP-related [At~gl8290.1]
major intrinsic family protein/MIP family protein [At5g60660.1]
major intrinsic family protein/MIP family protein [At4g23400.1]
major intrinsic family protein/MIP family protein [At3g04090.1]
major intrinsic family protein MIP family protein [At4gl7340.1]
K-C1 Co-transporter type 1 protein-related/KCC1 protein-related [At3g58370.1]
integral membrane transporter family protein [Atlg79710.1]
integral membrane protein, putative; MSF protein [Atlg75220.1]
integral membrane family protein [At4g25830.1]
integral membrane family protein [At2q27370.1]
inorganic phosphate transporter, putative [At2g29650.1]
glycerol-3-phosphate transporter/permease, putative, [At3g47420.1]
glutathione S-conjugate ABC transporter (MRP2) [At2g34660.1]
glutathione S-conjugate ABC transporter (MRPI) [Ati~g30400.1]
chloride channel protein (CLC-a) [At5g40890.1]
cation efflux family protein [At2g39450.1]
calcium exchanger (CAX1) [At2g38170.3]
auxin efflux carrier family protein [At2g17500.1]
auxin efflux carrier family protein [Atlg76520.1]
amino acid transporter family protein [At3g30390.1]
amino acid transporter family protein [At3g09340.1]
amino acid transporter family protein [At2g41190.1]
amino acid permease family protein [At5g05630.1]
amino acid permease family protein [At5g04770.1]
amino acid carrier, putative/amino acid permease, putative [Atlg77380.1]
ABC transporter family protein putative multi resistance protein mrp [At3959140.1]
ABC transporter family protein ATP-binding cassette-sub-family [At3g55090.1]
ABC transporter family protein [At5g64840.1]
ABC transporter family protein [At3g28345.1]
ABC transporter family protein [Atlg53270.1]
ABC transporter family protein [Atig31770,1]
transporter-related [At2g38060.1]
transporter-related [Atig79410.1]
transporter-related [Atlg20840.1)
transporter-related [At5gl37150,1]
transporter, putative [At5g53550.1]
transporter, putative [At~g24380.1]
transporter, putative [At4g25220.1]
tonoplast intrinsic protein, alpha (TIP3.1) [Atlg73190.1]
sulfate transporter family protein [At5gl3550.1]
sulfate transporter (ST1) [At3g51895.1]
sugar transporter, putative [At3q20460.1]
sugar transporter, putative [At3g05165.1)
sugar transporter family protein [At2g43330.1]
sugar transporter family protein [Atlg30220.1]
sucrose transporter/sucrose-proton symporter (SUC5) [Atlg71890.1]
sucrose transporter/sucrose-proton symporter (SUC1) [Atlg71880.1]
sodium/dicarboxylate cotransporter, putative [At~g47560.1]
proton-dependent oligopeptide transport (POT) family protein [At4g21680.1]
proton-dependent oligopeptide transport (POT) family protein [At3q53960.1]
proton-dependent oligopeptide transport (POT) family protein [At2g37900.1]
potassium channel protein 2 (AKT2) (AKT3) [At~g22200.1]
plasma membrane intrinsic protein 20 (PIP2C)/PIP2.3 (PIP2.3) [At2g37180.1]
plasma membrane intrinsic protein 2A (PIP2A)/PIP2.1 (PIP2.1) [At3g53420.1]
plasma membrane intrinsic protein 10 (PIPlC)/PIP1,3 (PIP1.3) [Atlg01620.1]
Figure 3-11. Continued











potassium transporter family protein [At4gl9960.1]
cyclic nucleotide-binding transporter 2/CNBT2 (CNGC19) [At3gl7690.1]


nodulin-related [At5g40230.1]
metal transporter, putative (ZIP5) [Atlg05300.1]
MATE efflux family protein [At2g38330.1]
amino acid transporter family protein [At5g02170.1]
transporter-related [At5g57100.1]
sugar transporter, putative [At5g23270.1]
sugar transporter, putative [At4g02050.1]
sugar transporter family protein [At5gl3740.1]
sugar transporter family protein [At4gl6480.1]
sugar transport protein (STP4) [At3ql9930.1]
proton-dependent oligopeptide transport (POT) family protein
proton-dependent oligopeptide transport (POTP) family protein
potassium transporter, putative (KT2) [At2g40540.1]
oligopeptide transporter OPT family protein [At5g64410.1]


[At5g46050.1]
[Atlg22570.1]


chloride channel-like (CLC) protein, putative [At5g33280.1]
cation/hydrogen exchanger (CHX20)/ antiporter family 2 (CPA2) [At3g53720.1]
lysine and histidine specific transporter, putative [Atlg25530.1]
metal transporter family protein [At3q08650.1]
proton-dependent oligopeptide transport (POT) family protein [At3g45650.1]
integral membrane protein, putative [At5g62820.1]
magnesium/proton exchanger (MHX1) [At2g47600.1]
MATE efflux family protein [At3g26590.1]
magnesium transporter CorA-like family protein (MGT1) (MRS2) [At1g80900.1]
ABC transporter family protein [At2g41700.1]
transporter-related [At5g20380.1]
integral membrane family protein [At4g27860.1]
integral membrane TerC family protein [At5gl2130.1]
ferroportin-related [At5g26820.1]
mechanosensitive ion channel domain-containing protein [At5gl9520.1]
cation-chloride cotransporter, putative [Atlg30450.3]
integral membrane family protein [At4g27870.1]
potassium transporter (MAK5) [At4gl3420.1]
mitochondrial substrate carrier family protein [At4g24570.1]


Figure 3-11. Continued











inorganic phosphate transporter (PHT1) (PT1) [At5g43350.1]
calcium-transporting ATPase 2, plasma membrane-type (ACA2) [At4g37640.1]
iron transporter-related [At2g38460.1]
inward rectifying potassium channel, putative (KAT3) (AKT4) (KC1) [At4g32650.1]
integral membrane transporter family protein [Atig64890.1]
integral membrane family protein [At5gl9930.1]
inorganic phosphate transporter (PHT2) [At5g43370.1]
importin beta-2 subunit family protein [At3g59020.1]
high-affinity nitrate transporter (ACH1) [Atlg08090.1]
hexose transporter, putative [Atlg79820.1]
cyclic nucleotide-regulated ion channel, putative (CNGC15) [At2928260.1]
cyclic nucleotide-regulated ion channel (CNGC10) (ACBKL) [Atlg01340.1]
copper-exporting ATPase/responsive-to-antagonist 1 [At5g44790.1]
contains ZIP Zinc transporter domain [Atlg68100.1]
choline transporter-related [At3gl5380.1]
cation-chloride cotransporter, putative [Atlg30450.2]
cation-chloride cotransporter, putative [Atlg30450.1]
cation efflux family protein [Atl1l6310.1]
calcium-transporting ATPase 8, plasma membrane-type (ACAS) [At5g57110.1]
calcium-transporting ATPase 4, plasma membrane-type (ACA4) [At2g41560.1]
auxin transport protein, putative (PIN3) [Atlg70940.1]
ATPase, plasma membrane-type, putative/proton pump, putative [At3q60330.1]
ATPase, plasma membrane-type, putative/proton pump, putative [At3g42640.1]
ATPase E1-E2 type family protein [At4g30120.1]
ATPase E1-E2 type family protein [At4g30110.1]
ATPase 1, plasma membrane-type, putative/proton pump 1, putative [At2gl8960.1]
ammonium transporter 1, member 1 (AMT1.1) [At4gl3510.1]
amino acid transporter family protein [At5g38820.1]
amino acid transporter family protein [Atlg47670.1]
amino acid permease, putative [At5g01240.1]
amino acid permease family protein [Atl1l7120.1]
ABC transporter family protein AbcA [At3g47770.1]
ABC transporter family protein [At4gl5236.1]
ABC transporter family protein [At3q16340.1]
ABC transporter family protein [Atlg71960.1]
ABC transporter family protein [Atlg66i950.1]
ABC transporter family protein [Atlg65410.1]
ABC transporter family protein [Atlg51500.1]
UDP-galactose/UDP-glucose transporter [At2g02810.1]
transporter-related [At4g35300.1]
transporter-related [At4g09810.1]
transporter-related [At3g43790.1]
transport protein, putative [At2g21630.1]
proton-dependent oligopeptide transport (POT) family protein [At3g45660.1]
potassium transporter family protein [At5g09400.1]
potassium channel protein 1 (AKT1) [At2g26650.1]
porin, putative [At3g49920.1]
oligopeptide transporter OPT family protein [At4g27730.1]
oligopeptide transporter OPT family protein [At1g48370.1]
NRAMP metal ion transporter 1 (NRAMP1) [Atlg80830.1]
nodulin-related/integral membrane family protein [At5g45370.1]
nodulin MtN21 family protein; Integral membrane protein [Atlg75500.1]
nodulin MtN21 family protein [At5g40240.1]
nodulin MtN21 family protein [At3g53210.1]
nodulin MtN21 family protein [Atlg44800.1]
multidrug resistance P-glycoprotein, putative [At3g28860.1]
mitochondrial substrate carrier family protein [At5g64970.1]
mitochondrial import Timl7/Tim22/Tim23 family protein [At4g26670.1]
mitochondrial import Timl7/Tim22/Tim23 family protein [At3g25120.1]
mitochondrial import Timl7/Tim22/Tim23 family protein [Atl1l7530.1]
mitochondrial import inner membrane translocase (TIM17) [At2g37410.1]
mechanosensitive ion channel domain-containing protein [At4g00290.1]
MATE efflux family protein [At4g22790.1]
MATE efflux family protein [Atlg66780.1]
MATE efflux family protein [Atlg33080.1]
mannitol transporter, putative [At3gl8830.1]
major intrinsic family protein/MIP family protein [At4g01470.1]
magnesium transporter CorA-like family protein (MRS2-7) [At5g09690.1]


Figure 3-11. Continued



















































ULJ------------------------------------


1.


0.5 +


SULTR3;4


NIRAMP1


CAXi


MRS2-10


Gene
Figure 3-12. Whisker box plots representing gene expression ratio distributions for the Q-PCR
analysis of four genes showing differential expression on the microarrays in col-0 at
time 3 treatment versus time 1 control. RNA sources were the same as for the
microarray experiment. Results show permutated expression data that are calculated
by the REST 2008 statistical analysis software, which uses randomization techniques.
The graph gives an impression of the expression ratio distribution per gene related to
the results presented in Tables 3-5a and 3-5b.


3


MRS2-10


SULTF;4


NRAMP1


Figure 3-13. Whisker box plots representing gene expression ratio distributions for the Q-PCR
analysis of four genes showing differential expression on the microarrays in col-0 at
time 3 treatment versus time 3 control. RNA came from plants treated and controlled
for diurnal effects at time 3. Results represent the first repetition of the analysis and
show permutated expression data that are calculated by the REST 2008 statistical
analysis software, which uses randomization techniques. The graph gives an
impression of the expression ratio distribution per gene related to the results presented
in Table 3-6.


Relative Expression


Relative Expression
















LIJ



CAX1 MR~S2-10 SULTfQ34 NRAMP1
Giene
Figure 3-14. Whisker box plots representing gene expression ratio distributions for the Q-PCR
analysis of four genes showing differential expression on the microarrays in col-0 at
time 3 treatment versus time 3 control. RNA came from plants treated and controlled
for diurnal effects at time 3. Results represent the second repetition of the analysis
and show permutated expression data that are calculated by the REST 2008 statistical
analysis software, which uses randomization techniques. The graph gives an
impression of the expression ratio distribution per gene related to the results presented
in Table 3-7.


Relative Expression









CHAPTER 4
EFFECTS OF A SPACEFLIGHT ENVIRONMENT ON HERITABLE CHANGES IN WHEAT
GENE EXPRE SSION

Introduction

A fundamental question in space biology is whether the long-term exposure (such as for

one or more life cycles) of plants to the spaceflight environment with its microgravity and

radiation parameters, can cause changes in subsequent generations. Ways to address this question

include making genomic, proteomic, metabolic and morphologic comparisons. In this study we

used a genomic approach by testing for heritable differences in wheat gene expression patterns in

plants grown from seed whose progenitors were set in a spaceflight environment the space

station MIR. The molecular response of Arabidopsis and wheat to the spaceflight environment

has been evaluated in terms of genome-wide patterns of differential gene expression between

orbital and ground control plants (Paul et al., 2005; Stutte et al., 2006). These studies examined

the direct reaction to a novel environment; they do not reflect spaceflight induced changes to the

genome. The present study represents the first time that gene expression analysis is used to assay

for potential spaceflight-induced changes in later generations. An overview of experiments to

date addressing these two fundamental questions whether plant species can complete one or

more of their life cycles in space, and whether changes can be detected in subsequent generations

after such long-term growth in orbit is given below, with experiments involving wheat being

reviewed last.

In 1982, Arabidopsis thaliana (Arabidopsis) plants completed for the first time a full life

cycle in space while in a growth unit on the Salyut 7 space station. In this first example of seed

to seed development, the siliques and seeds that had formed under spaceflight conditions were

shorter and smaller respectively compared to those of the ground controls. When the space seeds

were subsequently grown on Earth, 42% developed as normal seed-bearing plants as compared









to 67.6% of the control seeds. The percentages of seeds that did not germinate or that were lethal

at an early growth stage were both higher for the space seeds than for the control seeds (Merkys

et al., 1984).

During the greenhouse 3 experiment in 1997 on space station MIR, Bra~ssica urapa L.

(Brassica) completed two full life cycles and a third partial cycle (Musgrave et al., 2000). Seeds

harvested from the first set of plants grown on MIR were planted alongside original seeds for the

second experiment. Second-generation space plants were significantly smaller than those grown

from original seeds, while on the ground there was no distinction between the corresponding two

control sets (Musgrave et al., 2000). With respect to late stages of seed development and

maturation of Brassica in microgravity, the study found that Einal seed size was diminished,

reserves were also stored in the form of starch rather than the expected protein and lipid, and

ripening by the silique occurred in a basipetal manner, rather than simultaneously along the

silique. The authors propose that these results might be explained by indirect effects of

microgravity on the silique microenvironment (Musgrave et al., 2000). Altered gas movement in

microgravity would be expected to change the dynamics of oxygen and carbon dioxide exchange

by tissues enclosed within these unique gaseous microenvironments (Kuang et al., 2005).

Between March 2003 and April 2005, Hyve successive generations of Pisum sativum (dwarf

pea) were grown in Lada on the International Space Station (ISS). Characteristics such as plant

height, number of pods per plant, number of peas per plant, dry plant biomass, and dry pea mass

per plant did not differ significantly between the plants in space and on the ground for the 1st

generation (Sychey et al., 2007). Quantitative results for the subsequent four generations were

not given. Random Amplified Polymorphic DNA (RAPD) analysis, as well as an examination of









chromosome aberrations, was performed on plants grown from several first generation space and

ground control seeds. Results did not reveal genetic polymorphisms (Sychey et al., 2007).

The life cycle experiments with wheat that establish the foundation of the experiments

presented here were initiated in 1991. Triticunt aestivunt L. cy. Super Dwarf plants nearly

completed their full life cycle in the Svetoblock M unit on the MIR station by forming three

seed-heads. The plants were returned to Earth after 167 days to finish the maturation process,

with two of the heads yielding a total of 28 mature seeds, most of which subsequently produced

healthy plants and viable seeds (Mashinsky et al., 1994). In 1996, another life cycle experiment

(greenhouse 2) was carried out on MIR. An improved growth unit called the SVET-2 SG

resulted in synchronous growth stages of wheat during its life cycle in space and on the ground

(Ivanova et al., 1998). The plants in space formed sterile heads due to the concentration of

ethylene in the MIR atmosphere (Ivanova et al., 1998; Levinskikh et al., 2000; Campbell et al.,

2001; Bubenheim et al., 2003; Salisbury et al., 2003). As a result, the Triticunt aestivunt L.

cultivar USU-Apogee, which is more resistant to ethylene, was used during the greenhouse 4

experiment in 1998 (Levinskikh et al., 2000; Sychey et al., 2001). Of the 52 Apogee seeds

planted on orbit, 12 developed into mature plants that produced a total of 508 seeds. The number

of seeds per plant was 3 8% lower than that of the ground control plants. 45 of the seeds collected

in space were subsequently planted on Earth. Growth after germination did not deviate

significantly from plants grown from ground control seeds (Levinskikh et al., 2000).

During the subsequent greenhouse 5 experiment in 1999, first generation space seeds were

planted as well as seeds without a history in space (Levinskikh et al., 2001). Only one plant

grown from the first generation matured and produced the second generation of space seeds

(Levinskikh et al., 2001; Sychey et al., 2001). The plant of the second space generation was not










morphologically different from those of the first generation or the ground controls (Levinskikh et

al., 2001; Sychey et al., 2001). Furthermore, plants grown on Earth from first- and second-

generation space seeds, as well as from ground control seeds, were also morphologically similar

(Levinskikh et al., 2001).

For the present study we return to the experiment in 1991 during which Super Dwarf wheat

plants set seed on space station MIR and were then returned to Earth to Einish maturation

(Mashinsky et al., 1994). The seeds formed by the plants flown on MIR and their offspring will

be referred to as 'MIR flight' seeds and plants, and those formed by the related ground control

plants and their offspring as 'MIR ground control' seeds and plants. The first generation MIR

flight and ground control seeds were grown for another two generations on the ground during a

study that screened for possible aftereffects of spaceflight factors on plant growth and

development (Mashinsky and Nechitailo, 1996; Mashinsky et al., 1997). Some of the retrieved

third generation MIR flight and ground control seeds were recently grown alongside an

additional control set consisting of Apogee wheat in the soil-based closed ecological facility

"Laboratory Biosphere" (Dempster et al., 2004; Nelson et al., 2008). The Apogee wheat will be

referred to as 'Biosphere control' plants. The goal of the presented work is to test whether

changes in gene expression can be detected in plants that are three generations removed from

long-term growth in space, compared to control plants that have not had a history in space. For

this purpose, leaf tissue was harvested for gene expression analysis using specifically designed

wheat microarrays.

Results

The MIR flight and Biosphere control plants were grown at the same time within the

Laboratory Biosphere facility to produce leaf tissue for gene expression analysis (Figure 4-1). A

comparison of gene expression between the plant sets was done via pooled reference RNA










(Figure 4-2). The transcript levels of each gene were compared by running a series of t-tests.

Results from the t-tests are presented in Figure 4-3 in the form of p-values (y-axis) plotted

against fold change (x-axis). Differences in transcript abundance larger than 2-fold change were

detected for 363 genes, and larger than 4-fold change for 25 genes. For 593 genes, differences in

expression corresponding to p-values < 0.05 were observed between the sets. The p-value

indicates the probability that the observed difference in expression between the sets is falsely

positive (type I error). For a single test, the hypothesis that a difference is significant is

commonly accepted when the type I error is 5% or lower. However, when analyzing a large

number of genes at the same time, such as with microarrays, the p-values of the individual tests

need to be adjusted so that the overall incidence of false-positives is reduced (Bretz et al., 2005).

Several methods are commonly used. The Bonferroni method for example controls the

familywise error rate (FWER), which is the probability of committing at least one type I error,

while the Benj amini and Hochberg and Q-value methods control the false discovery rate (FDR),

which is related to the expected proportion of false-positives among all significant results (Bretz

et al., 2005). The Benj amini and Hochberg and the Q-value methods are therefore less stringent

with respect to type I error control than Bonferroni, and allow the false negative rate to be kept

relatively low. To ensure that few real differences are falsely marked as insignificant, we chose

to correct with the Benjamini and Hochberg or Q-value method.

After correcting for multiple testing with Benj amini and Hochberg or Q-value, none of the

differences between the fourth generation MIR flight and the Biosphere control plants were

significant at the p < 0.05 or q < 0.05 level (Figures 4-4, 4-5). These results show that a

comparison of two biological replicates of both the experimental set and control set did not

identify any statistically significant differences in their patterns of gene expression. This









conclusion applies to the specific transcripts matching the 10,263 oligonucleotide probes on our

arrays; with respect to the remaining transcripts in the genome we cannot make any statements.

The plants comprising the experimental and control set are very closely related cultivars,

although not identical. Sequence polymorphisms within a species can result in differential

hybridization of transcripts to probes and thereby increase the incidence of false positives (Kirst

et al., 2006). Since we did not Eind any positives after applying the Benj amini and Hochberg and

Q-value methods, we have not increased the stringency of our multiple testing correction in the

direction of controlling false positives, such as with the Bonferroni method, which means that we

were able to maintain our false negative rate relatively low. We are aware of approaches used to

'recapture' potential false negatives that may exist despite a low false negative rate. For example

Wu et al. (2008) first used a traditional statistical analysis (ANOVA) and a multiple test

correction to define a set of 93 genes that showed highly significant differences in expression

levels between two strains of mice. By clustering analysis they then defined another 39 genes

where the differences between the expression levels of the two strains did not reach the criteria

of statistical significance (Wu et al., 2008). Since our analysis did not yield any significant

changes in gene expression, we need not apply this strategy to our results.

In summarizing the results of this analysis, we did not find differences in gene expression

that were statistically significant between the MIR plants three generations removed from

spaceflight exposure, and the Laboratory Biosphere ground control plants. Documents with

detailed tables listing all the 10,263 probes, their related wheat target sequence, raw signal values

as well as statistical results are available as supplemental data.

Discussion

This study has tested for the first time whether there are any heritable changes in gene

expression patterns in leaf tissue from plants three generations removed from spaceflight.









Several groups have analyzed gene expression differences between spaceflight-exposed plants

themselves, rather than their offspring, and their related ground controls. When 24-day-old

Apogee wheat leaves harvested during a first generation of spaceflight exposure were compared

to leaves from control plants, Stutte et al. (2006) did not Eind differences in gene expression

greater than 2-fold. In that case, leaf material harvested from two plants in space and on the

ground was pooled per set and mRNA extracted from the two pools was subj ected to a dye-swap

microarray experiment (Stutte et al., 2006). Paul et al. (2005) exposed 7-day-old Arabidopsis

seedlings to spaceflight for 5 days before harvesting. Shoot material from approximately 20

Arabidopsis seedlings from two different Plant Growth Chambers was pooled for both the

spaceflight and the control set. The resulting two samples were differentially labeled and

hybridized to a single microarray. The expression of 182 genes differed more than 4-fold

between the spaceflight and ground control samples. Of these genes, 50 were expressed at

moderate to high levels where additional confidence in fold-change values can be derived (Paul

et al., 2005). It should be stressed that these studies examined patterns of gene expression as a

function of a metabolic reaction to spaceflight and do not reflect spaceflight induced changes to

the genome. In contrast, the current study represents the first time that gene expression is used to

screen for heritable changes in plants induced by long-term spaceflight. Wheat plants were

exposed to the spaceflight environment during one generation and compared to control plants

three generations later. Although differences in gene expression larger than 2-fold or 4-fold

change were detected, none of the differences were shown to be statistically significant after

correction for multiple testing.

Heritable changes in gene expression can be caused by changes to the base sequence of the

genetic material (genetic mutations) or by chromatin and DNA modifications that do not involve









base sequence alteration (epigenetics). The main potential cause of heritable mutations to which

an organism is exposed in the spaceflight environment is ionizing radiation. Ionizing radiation in

low earth orbit consists of galactic cosmic rays (85% protons, 12% helium ions, 1% heavy ions

of charge 3, 2% electrons and positrons), energetic electrons and protons trapped in Earth's

geomagnetic field, solar energetic particles (electrons, protons and heavier charged particles up

to iron) and albedo neutrons and protons (Benton and Benton, 2001). Interactions between these

particles and the atomic nuclei of the spacecraft materials produce a variety of secondary

particles, including protons, helium ions, neutrons, and recoil heavy nuclei, that can be of higher

linear energy transfer (charged particles) or radiation weighting factors (neutrons) than the

primary radiation flux (Benton et al., 2002). Mutation induction per charged particle increases

with linear energy transfer up to about 200 keV/Cpm (Kiefer, 2002). On MIR, secondary particles

from proton-induced target fragmentation interactions were found to be the largest contributor to

the linear energy transfer (LET) spectrum above 100 keV/Clm (Benton et al., 2002). In low earth

orbit, mean dose rates of ionizing radiation have tended to be below 500 CIGy/day, regardless of

orbital inclination, solar cycle phase, and spacecraft orientation and shielding (Benton and

Benton, 2001).

When passing through cells, ionizing radiation may interact with DNA. This can result in

molecular damage such as single strand breaks, double strand breaks, base damage, and DNA-

protein cross links (Nikj oo et al., 1999). Double strand breaks in plant cells are mainly repaired

via non-homologous end j oining, which is associated with deletions and insertions of DNA

sequence (Gorbunova and Levy, 1997). Early space biology studies used various indirect

methods to indicate the possibility of mutations caused by ionizing radiation in satellites, space

vehicles and space stations. These methods include the association of high charge and energy










(HZE) particle tracts in plant seeds with chromosomal aberrations or visible phenotypes in

seedlings grown from these seeds (Peterson et al., 1977; Bucker and Facius, 1981; Nevzgodina et

al., 1984; Nevzgodina et al., 1989), and, more recently, random amplified polymorphic DNA-

polymerase chain reaction (Sychey et al., 2007). DNA-sequencing is a direct and detailed

method used to detect mutations related to ionizing radiation. For example, Arabidopsis mutants

induced by irradiating pollen with gamma rays (150-600 Gy) or carbon ions (40-150 Gy) carried

1- or 4-bp deletions, which were transmitted normally, or extremely large deletions of up to >6

Mbp, most of which were not transmitted to progeny (Naito et al., 2005). Ionizing radiation

studies involving plant or mammalian cells often analyze the molecular damage resulting from

relatively high irradiation doses (Schmidt and Kiefer, 1998; Nikj oo et al., 1999; Naito et al.,

2005). After a 40-day exposure to 0.25-0.51 mGy/day on MIR, the averaged mutation frequency

of the rpsL gene in yeast did not differ from that of the same gene in the ground control samples.

However, the greater part of the Mir mutant samples were found to have a total or large deletion

in the rpsL sequence, suggesting that space radiation containing high-LET might have caused

deletion-type mutations (Fukuda et al., 2000).

In the event of transmissible small intragenic deletions or insertions in plants growing in a

space station, the likelihood of finding differences in gene expression between these plants, or

their offspring, and the ground controls appears to be small. First of all, the small deletion or

insertion would have to affect the binding of proteins involved in transcription of the particular

gene. If this is the case, then the ploidy of the plant species will determine the extent to which

overall transcription levels of the gene or genes are changed. I: aestivum is a hexaploid species,

which means that with potentially three functional copies of a gene, a change in expression of

one allele may have only subtle effects overall (Dubcovsky and Dvorak, 2007). For a self-










compatible species like T. aestivum, the relative presence of the mutant allele in subsequent

generations will be the same for the total population, but may differ in smaller sample

populations. With a random distribution of affected alleles, this could result in a higher or lower

incidence of mutant alleles in a small sample compared to the overall population. The same

hypothetical scenario applies to potentially larger deletions of the size of promoters or coding

regions, in case they occur in low earth orbit and are transmissible.

Heritable changes in gene expression can also be associated with changes in the

methylation state of cytosine bases in DNA and the modification state of histone proteins

(Henderson and Jacobsen, 2007). DNA and histone modification in plants are interrelated and

mediated in part by enzymes involved in addition or removal of methyl or acetyl groups and by

small interfering RNAs (Chen and Tian, 2007; Henderson and Jacobsen, 2007). In tobacco

plants, a close correlation was found between demethylation and expression of a tobacco gene

encoding a glycerophosphodiesterase-like protein upon exposure to abiotic stresses such as

aluminum, salt and low temperature (Choi and Sano, 2007). For an environmentally triggered

DNA and histone modification to be heritable, it would have to be transmitted over many rounds

of mitotic DNA replication in sporophytic tissues, through the differentiation of gametophyte

precursor cells, meiosis, and postmeiotic mitoses of haploid gametophytes (Takeda and

Paszkowski, 2006). Identified environmental factors relevant to spaceflight, such as microgravity

or ionizing radiation, have so far not been shown to cause heritable histone or DNA

modifications changing gene expression in the offspring of exposed plants. A study of

Arabidopsis plants growing near Chernobyl in 1989 found that the genomic DNA of two

subsequent generations grown under laboratory conditions was hypermethylated compared to

control plants. Seeds collected on site several years later gave rise to offspring with less










hypermethylated DNA (Kovalchuk et al., 2004). The authors suggested that DNA

hypermethylation could be an immediate plant response to ionizing radiation. Aside from this

observation, DNA and histone modifications initiated by environmental factors have to date not

been reported to be inherited epigenetically in plants.

Conclusion

This study is a first attempt to answer the question whether long-term exposure to the

spaceflight environment in low earth orbit space stations can cause significant, heritable changes

in gene expression patterns in plants. Some of the factors to which plants are exposed in a

spaceflight environment that are potentially able to cause such differences are described in the

discussion. Our analysis of plants three generations removed from spaceflight exposure

compared to plants with no exposure in their lineage indicate that exposure to the spaceflight

environment for one generation does not result in changes in gene expression that are heritable.

Future space biology experiments addressing heritable changes in gene expression could use a

larger number of biological replicates to ensure that the false negative rate is indeed as low as

possible. Furthermore, whole genome arrays could be used that contain probes targeting all

known transcripts. However, we do not imply to be able to foresee whether or in what way this

may change the current outcome. The effects of still longer exposure to spaceflight, such as for

several contiguous generations, on gene expression in exposed plants and their offspring could

also be tested.

Materials and Methods

Plant Species and Cultivars

Two closely related cultivars of wheat were used in this study: T. aestivum L. cy. Super

Dwarf and T. aestivum L. cy. USU-Apogee. USU-Apogee originated from a cross between

'Parula' and 'Super Dwarf, both of which were obtained from the CIMMYT germplasm









collection in 1984. Parula was selected for its small leaf size, and Super Dwarf was selected for

its short stature (25 cm tall). USU-Apogee is a dwarf (45-50 cm tall), hard red spring wheat with

a rapid development and resistance to Ca-induced leaf tip necrosis, leading to high yields in

controlled environments such as bioregenerative life support systems in space (Bugbee and

Koerner, 1997; Bugbee et al., 1997).

MIR and Ground Experiments

Triticum aestivum L. cy. Super Dwarf was grown in the Svetoblock M unit on MIR for 167

days in 1991. Ionite (ion exchange substrate) was used as a soil substitute. Light intensity ranged

between 135 and 175 Cpmol m-2 S-1. Air temperature was 22 + 20C and atmospheric CO2 WAS

6000 ppm. Three heads appeared in the boot (i.e. each surrounded by a leaf) in two plants while

the plants were still on MIR. After return to Earth on October 10, 1991, three spikes developed

from these, two of which contained a total of 28 seeds. The MIR ground control plants were

grown in the Svetoblock M unit inside the Svet greenhouse at a laboratory in Moscow

(Mashinsky et al., 1994). First generation MIR flight and ground control seeds were grown for an

additional two generations on the ground during subsequent experiments on Earth. The

experiments took place at the Institute of General Genetics of the Russian Academy of Science.

The plants were grown under the same light, temperature and CO2 COnditions as the plants on

MIR. Third generation seeds were harvested in 1997 (Nechitailo, data not shown).

Laboratory Biosphere Experiment

Third generation MIR flight and ground control seeds were planted in the Laboratory

Biosphere facility in New Mexico alongside T. aestivum L. cy. USU-Apogee in February 2005.

The lighting regime was 13 h light/1 1 h dark at a light intensity of 960 Cpmol m-2 S-1, 45 mol m-2

dayl supplied by high-pressure sodium lamps. Atmospheric CO2 ranged between 300 and 3000

ppm daily during the maj ority of the growing season. Temperatures ranged from 21 to 29 OC









during light hours and from 20 to 24 OC at night. Two of the third generation MIR flight seeds

germinated, grew and matured, versus one of the third generation MIR ground control seeds.

Two of the Apogee wheat plants were selected as additional controls that will be referred to as

'Biosphere controls'.

Sample Preparation

Leaves were harvested from the two fourth-generation MIR flight plants, the fourth-

generation MIR ground control plant, and the two Biosphere control plants. Tissues were

immediately stored in RNAlater. Two leaves per plant were pooled and total RNA was extracted

with the Qiagen RNeasy Plant Mini Kit. Aliquots of the MIR flight, Biosphere control, and MIR

ground control RNA samples were mixed to obtain a reference RNA sample.

Microarray Experimental Design and Data Analysis

Custom-made long-oligonucleotide (60-mer) wheat microarrays were developed by the

Interdisciplinary Center for Biotechnology Research (Gainesville, FL) in collaboration with

Agilent Technologies (Foster City, CA). A set of 10,263 60-mer wheat specific probes was

designed based on Triticum aestivum L. cy. Apogee EST sequences and in situ synthesized on

microarrays (2x22K format). The samples were amplified and labeled by converting target

mRNA to cRNA according to the manufacturer's protocol (Low RNA Input Fluorescent Linear

Amplifieation Kit Protocol [v. 4], Agilent Technologies). The two biological replicates from the

MIR flight and Biosphere control sample sets, and the one MIR ground control sample, were

Cy5-labeled and hybridized to the oligonucleotide microarrays with Cy3-labeled reference

cRNA. After hybridization, slides were washed, scanned and the data extracted according to

standard procedures (Two-Color Microarray-Based Gene Expression Analysis [v.5.5], Agilent

Technologies). Because there was only one biological replicate available for the MIR ground









control sample, it was excluded from the subsequent statistical analysis and will not be discussed

further.

Median signal intensities, detected by scanning the microarrays with the MIR flight,

Biosphere control and reference samples hybridized to the probes, were analyzed using a two-

step strategy (Chu et al., 2002; Hsieh et al., 2003). Initially, probe measurements were centered

relative to the microarray mean and log2 transformed. The signals of the MIR flight, Biosphere

control and reference samples were then evaluated in a mixed analysis of variance (ANOVA)

model, and unbiased estimates of transcript abundance (least-square means) were generated for

each sample set. Microarray slide was included in the ANOVA model as a random effect to

account for the covariance of samples hybridized to the same microarray (i.e. control for spot

effect) (Jin et al., 2001; Wolfinger et al., 2001). The transcript level estimates of the MIR flight

and the Biosphere control samples were then compared using a series of t-tests. The p-values

generated in this comparison were corrected for multiple testing using the Benj amini and

Hochberg procedure in the Multtest package of the R software (Benj amini and Hochberg, 1995),

and the Q-value procedure in the Q-value 1.0 package (default settings) of the R software (Storey

and Tibshirani, 2003).



























A B


A: MIR flight #1

B: MIR flight #2

C: MIR ground control

D: Biosphere control #1

E: Biosphere control #2



D E


Figure 4-1. Wheat growth in the Laboratory Biosphere. Photographs of individual wheat plants
growing in the Laboratory Biosphere from which leaves were collected for gene
expression analysis. Top row: fourth generation MIR flight plants A and B, and
fourth generation MIR ground control plant C (T. aestivum cy, Super Dwarf). Bottom
row: Biosphere control plants D and E (7: aestivum cy. USU-Apogee). Gene
expression patterns were compared between the MIR flight and the Biosphere control
sets by using microarrays and statistical analysis.




































Biosphere control #1
vs Reference

MIR flight #1
vs Reference


MIR ground control
vs Reference

MIR flight #2
vs Reference


Biosphere control #2
vs Reference

Microarray slides with
Sample and Reference
cRNA


Seeds planted on MIR


Seeds planted on the ground
as controls


Seed maturation and harvest of first
generation on the ground



First generation MIR flight and ground
control seeds are grown for another two
generations on the ground; third genera-
tion seeds are harvested


~Third generation MIR flight and ground
control seeds are planted in the Laboratory
Biosphere; fourth generation leaf tissue is
collected for gene expression analysis.


Seeds are planted in the Laboratory
Biosphere as controls; leaf tissue is
collected for gene expression analysis


Figure 4-2. Flowchart of wheat experiments. Flowchart showing an overview of the history of
the plant samples comprising the MIR flight and Biosphere control sets, as well as a
diagram of how the experimental and reference cRNA samples were hybridized to the
microarrays.










-2x 2x 4x


-3 -2- 012

Fod hage(og) iophr coto/MI lih






forr fas pstv es.S











-4x -2x 2x 4x
1.4
0.05
1.2




o0.8




0.2


0.



-3 -2 -1 0 1 2 3 4

Fold change (log2) Biosphere control/MIlR flight


Figure 4-4. Multiple testing correction: Benj amini and Hochberg method. ANOVA (t-test) p-
values corrected for multiple testing with Benj amini and Hochberg are plotted against
the corresponding log2 fold changes in transcripts between the Biosphere control and
MIR flight sets; corrected p-value = 0.05 is indicated by the horizontal line with the
label on the right. The scatter plot shows that none of the corrected p-values
associated with the 10.263 comparisons in gene expression are smaller than 0.05.











-4x -2x 2x 4x
1.4

1.3 0.05


o1.1








$ 0.8



0.7
-3 -2 -1 0 1 2 3 4

Fold change (log2) Biosphere control/MIR flight



Figure 4-5. Multiple testing correction: Q-value method. ANOVA (t-test) p-values corrected for
multiple testing with Q-value are plotted against the corresponding log2 fold changes
in transcripts between the Biosphere control and MIR flight sets; q-value = 0.05 is
indicated by the horizontal line with the label on the right. The scatter plot shows that
none of the corrected p-values (= q-values) associated with the 10.263 comparisons in
gene expression are smaller than 0.05.









CHAPTER 5
CONCLUSIONS

Manned missions to Mars demand the efficient use of local planetary resources and the

recycling of limited materials such as water, pressurized atmosphere and organic matter while

producing food (Barta and Henninger, 1994). The use of in situ regolith for plant growth in a

future bioregenerative life support system on Mars may have several advantages over hydroponic

systems (Schuerger et al., 2002). These include the immediate bioavailability of plant essential

ions, low-tech mechanical support for plants, and easy access of in situ materials once on the

surface. However, plant growth may be reduced or inhibited by substances in the regolith, such

as high levels of hydrated magnesium sulfate minerals (Chapter 1 and 2). In a potential

bioregenerative life support system on Mars, an excess of a particular element in the crew' s diet

could affect the presence and availability of other required elements. This study therefore focuses

on the possibility of reducing accumulation of Mg2+ and SO42- ions within the plant as a method

to enhance plant tolerance to high levels of magnesium sulfate in the growth medium.

Arabidopsis is a model species in plant molecular biology research and its genome is fully

sequenced. Plasma membrane localized efflux transporters of Mg2+ and SO42- ions have not been

identified to date in the outer root cell layers of Arabidopsis or other plant species. AtMRS2-10

and AtSULTR1;2 are genes encoding a known Mg2+ and SO42- uptake transporter respectively.

Arabidopsis lines carrying knockout T-DNA insertion mutations in AtMRS2-10 and

AtSULTR1;2 did not mitigate the constraining impacts of high magnesium sulfate

concentrations on wildtype Arabidopsis plants (Chapter 2). An Arabidopsis line carrying a

knockout mutation of the vacuolar CAX1 gene (caxl-1) showed a significant improvement in

growth on soil treated with high levels of MgSO4*7H20 in solution (Chapter 2). A reduction in

leaf magnesium content in caxl mutants compared to wildtype (0.7 o below normal) was









reported in a previous study (Bradshaw, 2005). Although the Mg content of caxl mutant roots

has not been analyzed so far, the reduced levels of Mg in leaves indicate that Arabidopsis CAX1

knockout mutants are in line with our obj ective of identifying Arabidopsis variants that show

improved tolerance of high magnesium sulfate by at least partly limiting accumulation of Mg2+

or SO42- ions within the plant. Therefore, genes in crop species encoding transporter proteins

similar in function to the protein encoded by CAX1 in Arabidopsis are proposed candidate genes

for enhancing tolerance of crop plant species to regolith high in soluble magnesium sulfate

minerals used in an advanced life support system on Mars.

The identification by Bradshaw (2005) of a CAX1 knockout mutant tolerant of low Ca:Mg

ratios in solution, which are characteristic of serpentine soil solutions, confirms the

appropriateness of serpentine soils as partial analogue soils on Earth for regolith high in soluble

magnesium sulfate minerals on Mars. Leaf Ca:Mg molar ratios of nonserpentine plant species are

generally equal to that of the soil, while serpentine species maintain significantly higher leaf

Ca:Mg than both their nonserpentine counterparts and the soil (O'Dell et al., 2006). The authors

conclude that elevated leaf Ca:Mg in the serpentine species was achieved by selective Ca2+

transport and/or Mg2+ eXClUSion operating at the root-to-shoot translocation level, as root Ca and

Mg concentrations did not differ between serpentine and nonserpentine species. Genetic

differentiation between populations ofArabidopsis lyrata growing on granitic or serpentinic soils

was measured by using an Arabidopsis thaliana tiling array that has 2.85 million probes

throughout the genome (Turner et al., 2008). The study found significant overrepresentation of

genes involved in ion transport, and one gene in particular, calcium-exchanger 7 (CAX7), was

presented as an excellent candidate gene for adaptation to low Ca:Mg ratios in A. lyrata. It is

currently not known which transporters or their regulators could be involved in the observed









lower levels of Mg or higher Ca:Mg ratios in leaves from Arabidopsis caxl-1 and serpentine

species respectively. Results from the analysis of genetic differentiation between A. lyrata

populations include an overrepresentation of genes encoding proteins involved in ion transport.

Some of these transporters might play a role in reducing Mg2+ accumulation within the plant.

This study is the first to document genome-wide plant root transcriptome responses to

elevated levels of magnesium sulfate based on the high Mg:Ca ratio that can occur in serpentine

soils using microarrays (Chapter 3). The obj ective was to analyze which genes are differentially

expressed as part of the primary stress response in roots of a non-tolerant species such as

wildtype Arabidopsis thaliana (col-0) compared to unexposed col-0 roots. This could lead to

identification of candidate genes in Arabidopsis with potential to enhance tolerance to high

magnesium sulfate by limiting accumulation within the plant. The caxl-1 mutant was also

exposed to elevated MgSO4*7H20 to determine which genes are differentially expressed in the

CAX1 knockout mutant background compared to exposed col-0. Genes that are differentially

expressed between the genotypes could point to some of the downstream molecular processes

eventually leading to enhanced tolerance for caxl-1 at the whole plant level, including reduced

leaf Mg content and increased fresh weight biomass, after days or weeks of exposure in agar or

soil medium. Some of the transcripts involved in downstream processes may themselves be

candidates for enhanced tolerance, such as those encoding (regulators of) plasma membrane

based channels that transport Mg2+

Transcriptome responses of Arabidopsis col-0 roots exposed to high magnesium sulfate for

45 min. compared to col-0 exposed to a control solution for 45 min. (Time 1) reveal over 300

differentially expressed genes. Genes of known function include those encoding calcium-binding

proteins, kinases, transcription factors, enzymes involved in hormone metabolism, disease









resistance proteins and many cell wall related proteins. The responses of the genes encoding cell

wall related proteins indicate a possible reduction in root growth when col-0 is exposed to high

concentrations of magnesium sulfate. Some of the genes of known or unknown function were

previously associated with specific or broad ranges of abiotic stresses, but not necessarily in

roots or at these time points. Over 200 genes encoding membrane based transporters were

differentially expressed across the col-0 time series. The expression of genes encoding known

plasma membrane based importers of Mg2+ and SO42- ions, such as MRS2-10, SULTR1;1 or

SULTR1;2, was not down-regulated. This corresponds with the observation reported in Chapter

2 that Arabidopsis lines carrying knockout T-DNA insertion mutations in AtMRS2-10 and

AtSULTR1;2 did not mitigate the constraining impacts of high magnesium sulfate

concentrations on wildtype Arabidopsis plants. The differential expression of genes encoding

known tonoplast localized transporters of Mg2+ and SO42- ions indicate a possible storage of

excess Mg2+ and SO42- ions in the vacuole. Future research can reveal whether any of the

differentially expressed transporter genes of unknown protein localization, protein function, or

both, are candidates to enhance tolerance to high levels of soluble magnesium sulfate minerals in

Martian regolith by reducing accumulation of Mg2+ and SO42- ions within the plant. For

example, the localization of sulfate transporters SULTR3;1 and SULTR3;4 within Arabidopsis

root tissue and cells can be analyzed in follow-up studies to see if the genes encoding these

transporters are candidates for enhanced tolerance. Potential regulators of membrane based

transporter activity, such as kinases, which are encoded by differentially expressed genes across

the col-0 time series, could also be analyzed. Since gene expression differences are not fully

controlled for diurnal effects for the Time 2 and 3 comparisons in the col-0 time series, these

effects will have to be ruled out in follow-up studies.









The down-regulation of cax1-1 gene expression is a natural response to high magnesium

sulfate in col-0 that is already seen at Time 1. Together with the down-regulation of CAX2 and

CAX3 gene expression at later time points it indicates a possible shortage of calcium in the

cytosol experienced by col-0 when exposed to high concentrations of magnesium sulfate. Only

three transcripts were differentially expressed between caxl-1 and col-0 at 3 hours after initiation

of treatment. Follow-up experiments could be done to discover the function of these three

transcripts. The root transcriptome of caxl-1 and col-0 could furthermore be compared at later

time points after initial exposure to high magnesium sulfate to reveal additional differentially

expressed transcripts that could indicate the molecular processes eventually leading to the

tolerance difference exhibited by these genotypes after days or weeks of growth. Some of the

transcripts involved in downstream processes may themselves be candidates for enhanced

tolerance, such as those encoding (regulators of) plasma membrane based channels that transport

Mg2+

Crop plants might be grown for multiple life cycles as part of a life support system in a

transport vehicle during a human mission to Mars. Within the transport vehicle, crop plants will

be exposed to lower gravitational forces and higher radiation fluxes compared to a growth

chamber on Earth. Some of the seeds harvested from these plants could subsequently be planted

in advanced life support systems on the surface of Mars. To analyze in detail whether plants

grown in spaceflight conditions can show changes in subsequent generations compared to control

plants, gene expression in wheat plants that are three generations removed from growth in the

MIR space station was compared to gene expression in wheat plants with no spaceflight

exposure in their lineage (Chapter 4). Our gene expression analysis results indicate that exposure

to the spaceflight environment for one generation does not result in changes in gene expression









that are heritable. Future space biology experiments addressing heritable changes in gene

expression could use a larger number of biological replicates to ensure that the false negative rate

is indeed as low as possible. Furthermore, whole genome arrays could be used that contain

probes targeting all known transcripts. However, we do not imply to be able to foresee whether

or in what way this may change the current outcome. The effects of still longer exposure to

spaceflight, such as for several contiguous generations, on gene expression in exposed plants and

their offspring could also be tested.









LIST OF REFERENCES


Alonso JM, Stepanova AN (2003) T-DNA mutagenesis in Arabidopsis. Methods Mol Biol 236:
177-188

Arvidson RE, Poulet F, Bibring JP, Wolff M, Gendrin A, Morris RV, Freeman JJ,
Langevin Y, Mangold N, Bellucci G (2005) Spectral reflectance and morphologic
correlations in eastern Terra Meridiani, Mars. Science 307: 1591-1594

Banin A, Han FX, Kan I, Cicelsky A (1997) Acidic volatiles and the Mars soil. J. Geophys.
Res. 102: 13,341-313,356

Barber SA (1995) Soil nutrient bioavailability: a mechanistic approach, Ed 2nd. John Wiley and
Sons, Inc

Barta DJ, Henninger DL (1994) Regenerative life support systems: why do we need them?
Adv Space Res 14: 403-410

Baxter I, Ouzzani M, Orcun S, Kennedy B, Jandhyala SS, Salt DE (2007) Purdue ionomics
information management system: an integrated functional genomics platform. Plant
Physiol 143: 600-611

Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful
approach to multiple testing. Journal of the royal statistical society 57: 289-300

Benton ER, Benton EV (2001) Space radiation dosimetry in low Earth orbit and beyond. Nucl
Instrum Methods Phys Res B 184: 255-294

Benton ER, Benton EV, Frank AL, Leonov A, Gaskin J (2002) Dosimetric results from the
Mir orbital station. Radiat Prot Dosimetry 100: 489-494

Bibring JP, Langevin Y, Mustard JF, Poulet F, Arvidson R, Gendrin A, Gondet B,
Mangold N, Pinet P, Forget F, Berthe M, Bibring JP, Gendrin A, Gomez C, Gondet
B, Jouglet D, Poulet F, Soufflot A, Vincendon M, Combes M, Drossart P, Encrenaz T,
Fouchet T, Merchiorri R, Belluci G, Altieri F, Formisano V, Capaccioni F, Cerroni P,
Coradini A, Fonti S, Korabley O, Kottsov V, Ignatiev N, Moroz V, Titov D, Zasova L,
Loiseau D, Mangold N, Pinet P, Doute S, Schmitt B, Sotin C, Hauber E, Hoffmann H,
Jaumann R, Keller U, Arvidson R, Mustard JF, Duxbury T, Forget F, Neukum G
(2006) Global mineralogical and aqueous mars history derived from OMEGA/Mars
Express data. Science 312: 400-404

Bradshaw HD, Jr. (2005) Mutations in CAX1 produce phenotypes characteristic of plants
tolerant to serpentine soils. New Phytol 167: 81-88

Bretz F, Landgrebe J, Brunner E (2005) Multiplicity issues in microarray experiments.
Methods Inf Med 44: 43 1-437

Brooks RR (1987) Serpentine and its vegetation. Dioscorides Press









Brown AH, Chapman DK, Johnsson A, Heathcote D (1995) Gravitropic responses of the
Avena coleoptile in space and on clinostats. I. Gravitropic response thresholds. Physiol
Plant 95: 27-33

Bubenheim DL, Stieber J, Campbell WF, Salisbury FB, Levinski M, Sytchev V, Podolsky I,
Chernova L, Pdolsky I (2003) Induced abnormality in Mir- and Earth-grown Super
Dwarf wheat. Adv Space Res 31: 229-234

Buchner P, Takahashi H, Hawkesford MJ (2004) Plant sulphate transporters: co-ordination of
uptake, intracellular and long-distance transport. J Exp Bot 55: 1765-1773

Bucker H, Facius R (1981) The role of HZE particles in spaceflight: results from spaceflight
and ground-based experiments. Acta Astronaut 8: 1099-1107

Bugbee B, Koerner G (1997) Yield comparisons and unique characteristics of the dwarf wheat
cultivar 'USU-Apogee'. Adv Space Res 20: 1891-1894

Bugbee B, Koerner G, Albrechtsen R, Dewey W, Clawson S (1997) Registration of cultivars:
registration of 'USU-Apogee' wheat. Crop Sci 37: 626

Campbell WF, Salisbury FB, Bugbee B, Klassen S, Naegle E, Strickland DT, Bingham GE,
Levinskikh M, Iljina GM, Veselova TD, Sytchev VN, Podolsky I, McManus WR,
Bubenheim DL, Stieber J, Jahns G (2001) Comparative floral development of Mir-
grown and ethylene-treated, Earth-grown Super Dwarf wheat. J Plant Physiol 158: 105 1-
1060

Chen ZJ, Tian L (2007) Roles of dynamic and reversible histone acetylation in plant
development and polyploidy. Biochim Biophys Acta 1769: 295-307

Cheng NH, Pittman JK, Barkla BJ, Shigaki T, Hirschi KD (2003) The Arabidopsis caxl
mutant exhibits impaired ion homeostasis, development, and hormonal responses and
reveals interplay among vacuolar transporters. Plant Cell 15: 347-364

Cheng NH, Pittman JK, Shigaki T, Lachmansingh J, LeClere S, Lahner B, Salt DE,
Hirschi KD (2005) Functional association of Arabidopsis CAX1 and CAX3 is required for
normal growth and ion homeostasis. Plant Physiol 138: 2048-2060

Choi CS, Sano H (2007) Abiotic-stress induces demethylation and transcriptional activation of a
gene encoding a glycerophosphodiesterase-like protein in tobacco plants. Mol Genet
Genomics 277: 589-600

Chou IM, Seal RR, 2nd (2003) Determination of epsomite-hexahydrite equilibria by the
humidity-buffer technique at 0.1 MPa with implications for phase equilibria in the system
MgSO4-H20. Astrobiology 3: 619-630









Christensen PR, Wyatt MB, Glotch TD, Rogers AD, Anwar S, Arvidson RE, Bandfield JL,
Blaney DL, Budney C, Calvin WM, Fallacaro A, Fergason RL, Gorelick N, Graff TG,
Hamilton VE, Hayes AG, Johnson JR, Knudson AT, McSween HY, Jr., Mehall GL,
Mehall LK, Moersch JE, Morris RV, Smith MD, Squyres SW, Ruff SW, Wolff MJ
(2004) Mineralogy at Meridiani Planum from the Mini-TES experiment on the
Opportunity Rover. Science 306: 1733-1739

Chu TM, Weir B, Wolfinger R (2002) A systematic statistical linear modeling approach to
oligonucleotide array experiments. Math Biosci 176: 35-51

Ciftci-Yilmaz S, Mittler R (2008) The zinc Einger network of plants. Cell Mol Life Sci 65:
1150-1160

Ciftci-Yilmaz S, Morsy MR, Song L, Coutu A, Krizek BA, Lewis MW, Warren D,
Cushman J, Connolly EL, Mittler R (2007) The EAR-motif of the Cys2/His2-type zinc
Einger protein Zat7 plays a key role in the defense response of Arabidopsis to salinity
stress. J Biol Chem 282: 9260-9268

Cosio C, Dunand C (2009) Specific functions of individual class III peroxidase genes. J Exp
Bot 60: 391-408

Cui H, Benfey PN (2009) Interplay between SCARECROW, GA and LIKE
HETEROCHROMATIN PROTEIN 1 in ground tissue patterning in the Arabidopsis root.
Plant J

Curie C, Briat JF (2003) Iron transport and signaling in plants. Annu Rev Plant Biol 54: 183-
206

Czechowski T, Stitt M, Altmann T, Udvardi MK, Scheible WR (2005) Genome-wide
identification and testing of superior reference genes for transcript normalization in
Arabidopsis. Plant Physiol 139: 5-17

Dempster WF, Van Thillo M, Alling A, Allen JP, Silverstone S, Nelson M (2004) Technical
review of the Laboratory Biosphere closed ecological system facility. Adv Space Res 34:
1477-1482

Deng W, Luo K, Li D, Zheng X, Wei X, Smith W, Thammina C, Lu L, Li Y, Pei Y (2006)
Overexpression of an Arabidopsis magnesium transport gene, AtMGT1, in Nicotiana
benthamniana confers Al tolerance. J Exp Bot 57: 423 5-4243

Drysdale AE, Ewert MK, Hanford AJ (2003) Life support approaches for Mars missions. Adv
Space Res 31: 51-61

Dubcovsky J, Dvorak J (2007) Genome plasticity a key factor in the success of polyploid wheat
under domestication. Science 316: 1862-1866









Edmond C, Shigaki T, Ewert S, Nelson MD, Connorton JM, Chalova V, Noordally Z,
Pittman JK (2009) Comparative analysis of CAX2-like cation transporters indicates
functional and regulatory diversity. Biochem J 418: 145-154

Ernst WH (1998) Sulfur metabolism in higher plants: potential for phytoremediation.
Biodegradation 9: 311-318

Ernst WHO (1997) Life-history syndromes and the ecology of plants from high sulphur-
habitats. In WJ Cram, LJ De Kok, I Stulen, C Brunold, H Rennenberg, eds, Sulphur
metabolism in higher plants: molecular, ecophysiological and nutritional aspects.
Backhuys Publishers, pp 131-146

Fanning DS, Burch SN (2006) Sulfate and sulfide minerals. Encyclopedia of Soil Science 1:
1714-1716

Fontes EP, Santos AA, Luz DF, Waclawovsky AJ, Chory J (2004) The geminivirus nuclear
shuttle protein is a virulence factor that suppresses transmembrane receptor kinase activity.
Genes Dev 18: 2545-2556

Frigerio M, Alabadi D, Perez-Gomez J, Garcia-Carcel L, Phillips AL, Hedden P, Blazquez
MA (2006) Transcriptional regulation of gibberellin metabolism genes by auxin signaling
in Arabidopsis. Plant Physiol 142: 553-563

Fukuda T, Fukuda K, Takahashi A, Ohnishi T, Nakano T, Sato M, Gunge N (2000)
Analysis of deletion mutations of the rpsL gene in the yeast Saccharomyces cerevisiae
detected after long-term flight on the Russian space station Mir. Mutat Res 470: 125-132

Gellert R, Rieder R, Anderson RC, Bruckner J, Clark BC, Dreibus G, Economou T,
Klingelhofer G, Lugmair GW, Ming DW, Squyres SW, D'Uston C, Wanke H, Yen A,
Zipfel J (2004) Chemistry of rocks and soils in Gusey Crater from the Alpha Particle X-
ray Spectrometer. Science 305: 829-832

Gendrin A, Mangold N, Bibring JP, Langevin Y, Gondet B, Poulet F, Bonello G, Quantin
C, Mustard J, Arvidson R, LeMouelic S (2005) Sulfates in Martian layered terrains: the
OMEGA/Mars Express view. Science 307: 1587-1591

Golden DC, Ming DW, Morris RV, Mertzman SA (2005) Laboratory-simulated acid-sulfate
weathering of basaltic materials: implications for formation of sulfates at Meridiani
Planum and Gusey crater, Mars. J. Geophys. Res. 110

Gorbunova V, Levy AA (1997) Non-homologous DNA end joining in plant cells is associated
with deletions and filler DNA insertions. Nucleic Acids Res 25: 4650-4657










Haskin LA, Wang A, Jolliff BL, McSween HY, Clark BC, Des Marais DJ, McLennan SM,
Tosca NJ, Hurowitz JA, Farmer JD, Yen A, Squyres SW, Arvidson RE, Klingelhofer
G, Schroder C, de Souza PA, Jr., Ming DW, Gellert R, Zipfel J, Bruckner J, Bell JF,
3rd, Herkenhoff K, Christensen PR, Ruff S, Blaney D, Gorevan S, Cabrol NA,
Crumpler L, Grant J, Soderblom L (2005) Water alteration of rocks and soils on Mars at
the Spirit rover site in Gusey crater. Nature 436: 66-69

Hawkesford MJ (2003) Transporter gene families in plants: the sulphate transporter gene
family; redundancy or specialization? Physiol Plant 117: 155-163

Henderson IR, Jacobsen SE (2007) Epigenetic inheritance in plants. Nature 447: 418-424

Herbette S, Taconnat L, Hugouvieux V, Piette L, Magniette ML, Cuine S, Auroy P,
Richaud P, Forestier C, Bourguignon J, Renou JP, Vavasseur A, Leonhardt N (2006)
Genome-wide transcriptome profiling of the early cadmium response of Arabidopsis roots
and shoots. Biochimie 88: 1751-1765

Hsieh WP, Chu TM, Wolfinger RD, Gibson G (2003) Mixed-model reanalysis of primate data
suggests tissue and species biases in oligonucleotide-based gene expression profiles.
Genetics 165: 747-757

Ivanova TN, Kostov PT, Sapunova SM, Dandolov IW, Salisbury FB, Bingham GE, Sytchov
VN, Levinskikh MA, Podolski IG, Bubenheim DB, Jahns G (1998) Six-month space
greenhouse experiments: a step to creation of future biological life support systems. Acta
Astronaut 42: 11-23

Jiang Y, Deyholos MK (2006) Comprehensive transcriptional profiling ofNaCl-stressed
Arabidopsis roots reveals novel classes of responsive genes. BMC Plant Biol 6: 25

Jin W, Riley RM, Wolfinger RD, White KP, Passador-Gurgel G, Gibson G (2001) The
contributions of sex, genotype and age to transcriptional variance in Drosophila
melan2oga~ster. Nat Genet 29: 389-395

Katsoka T, Watanabe-Takahashi A, Hayashi N, Ohnishi M, Mimura T, Buchner P,
Hawkesford MJ, Yamaya T, Takahashi H (2004) Vacuolar sulfate transporters are
essential determinants controlling internal distribution of sulfate in Arabidopsis. Plant Cell
16: 2693-2704

Kiefer J (2002) Mutagenic effects of heavy charged particles. J Radiat Res (Tokyo) 43 Suppl:
S21-25

Kirst M, Caldo R, Casati P, Tanimoto G, Walbot V, Wise RP, Buckler ES (2006) Genetic
diversity contribution to errors in short oligonucleotide microarray analysis. Plant
Biotechnol J 4: 489-498










Klingelhofer G, Morris RV, Bernhardt B, Schroder C, Rodionov DS, de Souza PA, Jr., Yen
A, Gellert R, Evlanov EN, Zubkov B, Foh J, Bonnes U, Kankeleit E, Gutlich P, Ming
DW, Renz F, Wdowiak T, Squyres SW, Arvidson RE (2004) Jarosite and hematite at
Meridiani Planum from Opportunity's Mossbauer Spectrometer. Science 306: 1740-1745

Kovacik J, Klejdus B (2008) Dynamics of phenolic acids and lignin accumulation in metal-
treated Matricaria chamnomilla roots. Plant Cell Rep 27: 605-615

Kovalchuk I, Abramov V, Pogribny I, Kovalchuk O (2004) Molecular aspects of plant
adaptation to life in the Chernobyl zone. Plant Physiol 135: 357-363

Kruckeberg AR (1999) Serpentine barrens of western North America. In RC Anderson, JS
Fralish, JM Baskin, eds, Savannas, barrens, and rock outcrop plant communities of North
America. Cambridge University Press, pp 309-321

Kuang A, Popova A, McClure G, Musgrave ME (2005) Dynamics of storage reserve
deposition during Bra~ssica rapa L. pollen and seed development in microgravity. Int J
Plant Sci 166: 85-96

Kufner I, Koch W (2008) Stress regulated members of the plant organic cation transporter
family are localized to the vacuolar membrane. BMC Res Notes 1: 43

Kwasniewski M, Szarejko I (2006) Molecular cloning and characterization of beta-expansin
gene related to root hair formation in barley. Plant Physiol 141: 1149-1158

Langevin Y, Poulet F, Bibring JP, Gondet B (2005) Sulfates in the north polar region of Mars
detected by OMEGA/Mars Express. Science 307: 1584-1586

Laurie S, Feeney KA, Maathuis FJ, Heard PJ, Brown SJ, Leigh RA (2002) A role for HKT 1
in sodium uptake by wheat roots. Plant J 32: 139-149

Lejay L, Gansel X, Cerezo M, Tillard P, Muller C, Krapp A, von Wiren N, Daniel-Vedele
F, Gojon A (2003) Regulation of root ion transporters by photosynthesis: functional
importance and relation with hexokinase. Plant Cell 15: 2218-2232

Levinskikh MA, Sychev VN, Derendiaeva TA, Signalova OB, Podol'skii IG, Avdeev SV,
Bingheim GE (2001) [Growth and development of plants in a row of generations under
the conditions of spaceflight (experiment Greenhouse-5)]. Aviakosm Ekolog Med 35: 45-
49

Levinskikh MA, Sychev VN, Derendiaeva TA, Signalova OB, Podol'skii IG, Padalka GI,
Avdeev SV, Bingham GE (2000) [Growth of wheat from seed-to-seed in spaceflight].
Aviakosm Ekolog Med 34: 44-49

Levinskikh MA, Sychev VN, Derendyneva TA, Signalova OB, Salisbury FB, Campbell WF,
Bingham GE, Bubenheim DL, Jahns G (2000) Analysis of the spaceflight effects on
growth and development of Super Dwarf wheat grown on the Space Station Mir. J Plant
Physiol 156: 522-529










Li L, Tutone AF, Drummond RS, Gardner RC, Luan S (2001) A novel family of magnesium
transport genes in Arabidopsis. Plant Cell 13: 2761-2775

Liu YB, Lu SM, Zhang JF, Liu S, Lu YT (2007) A xyloglucan endotransglucosylase/hydrolase
involves in growth of primary root and alters the deposition of cellulose in Arabidopsis.
Planta 226: 1547-1560

Ma S, Bohnert HJ (2007) Integration ofArabidopsis thaliana stress-related transcript profiles,
promoter structures, and cell-specific expression. Genome Biol 8: R49

Ma W, Ali R, Berkowitz GA (2006) Characterization of plant phenotypes associated with loss-
of-function of AtCNGC 1, a plant cyclic nucleotide gated cation channel. Plant Physiol
Biochem 44: 494-505

Maathuis FJ, Filatov V, Herzyk P, Krijger GC, Axelsen KB, Chen S, Green BJ, Li Y,
Madagan KL, Sanchez-Fernandez R, Forde BG, Palmgren MG, Rea PA, Williams
LE, Sanders D, Amtmann A (2003) Transcriptome analysis of root transporters reveals
participation of multiple gene families in the response to cation stress. Plant J 35: 675-692

Madson M, Dunand C, Li X, Verma R, Vanzin GF, Caplan J, Shoue DA, Carpita NC,
Reiter WD (2003) The MUR3 gene of Arabidopsis encodes a xyloglucan
galactosyltransferase that is evolutionarily related to animal exostosins. Plant Cell 15:
1662-1670

Magnan F, Ranty B, Charpenteau M, Sotta B, Galaud JP, Aldon D (2008) Mutations in
AtCML9, a calmodulin-like protein from Arabidopsis thaliana, alter plant responses to
abiotic stress and abscisic acid. Plant J 56: 575-589

Magome H, Yamaguchi S, Hanada A, Kamiya Y, Oda K (2008) The DDF 1 transcriptional
activator upregulates expression of a gibberellin-deactivating gene, GA2ox7, under high-
salinity stress in Arabidopsis. Plant J 56: 613-626

Mao DD, Tian LF, Li LG, Chen J, Deng PY, Li DP, Luan S (2008) AtMGT7: An Arabidopsis
Gene Encoding a Low-Affinity Magnesium Transporter. J Integr Plant Biol 50: 1530-1538

Marschner H (1995) Mineral nutrition of higher plants, Ed second. Academic Press, London

Martinoia E, Massonneau A, Frangne N (2000) Transport processes of solutes across the
vacuolar membrane of higher plants. Plant Cell Physiol 41: 1 175-1 186

Maruyama-Nakashita A, Inoue E, Watanabe-Takahashi A, Yamaya T, Takahashi H (2003)
Transcriptome profiling of sulfur-responsive genes in Arabidopsis reveals global effects of
sulfur nutrition on multiple metabolic pathways. Plant Physiol 132: 597-605

Maruyama-Nakashita A, Nakamura Y, Tohge T, Saito K, Takahashi H (2006) Arabidopsis
SLIM1 is a central transcriptional regulator of plant sulfur response and metabolism. Plant
Cell 18: 323 5-325 1









Mashinsky AL, Ivanova I, Derendyneva T, Nechitailo GS, Salisbury F (1994) "From seed-to-
seed" experiment with wheat plants under spaceflight conditions. Adv Space Res 14: 13-19

Mashinsky AL, Nechitailo GS (1996) Ground-based experiments with plants on modelling the
action of gravity on the organism. Dr 47th International Astronautical Congress, Beijing,
China

Mashinsky AL, Nechitailo GS, Archipov MV (1997) Aftereffect of spaceflight factors in
plants. hz 48th International Astronautical Congress, Turin, Italy

Maurel C (2007) Plant aquaporins: novel functions and regulation properties. FEBS Lett 581:
2227-2236

McCormack E, Tsai YC, Braam J (2005) Handling calcium signaling: Arabidopsis CaMs and
CMLs. Trends Plant Sci 10: 383-389

McHale L, Tan X, Koehl P, Michelmore RW (2006) Plant NB S-LRR proteins: adaptable
guards. Genome Biol 7: 212

Merkys AJ, Laurinavicius RS, Svegzdiene DV (1984) Plant growth, development and
embryogenesis during Salyut-7 flight. Adv Space Res 4: 55-63

Ming DW, Mittlefehldt DW, Morris RV, Golden DC, Gellert R, Yen A, Clark BC, Squyres
SW, Farrand WH, Ruff SW, Arvidson RE, Klingelhofer G, McSween HY, Rodionov
DS, Schroder C, de Souza Jr. PA, Wang A (2006) Geochemical and mineralogical
indicators for aqueous processes in the Columbia Hills of Gusey crater, Mars. J. Geophys.
Res. 111

Miwa K, Takano J, Omori H, Seki M, Shinozaki K, Fujiwara T (2007) Plants tolerant of
high boron levels. Science 318: 1417

Munns R, Tester M (2008) Mechanisms of salinity tolerance. Annu Rev Plant Biol 59: 651-681

Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco
tissue cultures. Physiol Plant 15: 473-497

Musgrave ME, Kuang A (2003) Plant reproductive development during spaceflight. Adv Space
Biol Med 9: 1-23

Musgrave ME, Kuang A, Xiao Y, Stout SC, Bingham GE, Briarty LG, Levenskikh MA,
Sychev VN, Podolski IG (2000) Gravity independence of seed-to-seed cycling in Bra~ssica
urapa. Planta 210: 400-406

Naito K, Kusaba M, Shikazono N, Takano T, Tanaka A, Tanisaka T, Nishimura M (2005)
Transmissible and nontransmissible mutations induced by irradiating Arabidopsis thaliana
pollen with gamma-rays and carbon ions. Genetics 169: 881-889









Nelson M, Dempster WF, Allen JP, Silverstone S, Alling A (2008) Cowpeas and pinto beans:
performance and yields of candidate space crops in the Laboratory Biosphere closed
ecological system. Adv Space Res 41: 748-753

Nevzgodina LV, Maksimova EN, Kaminskaya EV (1989) Effects of prolonged exposure of
lettuce seeds to HZE particles on orbital stations. Adv Space Res 9: 53-58

Nevzgodina V, Gaubin Y, Kovaley EE, Planel H, Clegg J, Gasset G, Maximova EN,
Pianezzi B (1984) Changes in developmental capacity of artemia cyst and chromosomal
aberrations in lettuce seeds flown aboard Salyut-7 (Biobloc III experiment). Adv Space
Res 4: 71-76

Nguema-Ona E, Bannigan A, Chevalier L, Baskin TI, Driouich A (2007) Disruption of
arabinogalactan proteins disorganizes cortical microtubules in the root ofArabidopsis
thaliana. Plant J 52: 240-251

Nieuwland J, Feron R, Huisman BA, Fasolino A, Hilbers CW, Derksen J, Mariani C (2005)
Lipid transfer proteins enhance cell wall extension in tobacco. Plant Cell 17: 2009-2019

Nieves-Cordones M, Miller AJ, Aleman F, Martinez V, Rubio F (2008) A putative role for
the plasma membrane potential in the control of the expression of the gene encoding the
tomato high-affinity potassium transporter HAK5. Plant Mol Biol 68: 521-532

Nikiforova VJ, Bielecka M, Gakiere B, Krueger S, Rinder J, Kempa S, Morcuende R,
Scheible WR, Hesse H, Hoefgen R (2006) Effect of sulfur availability on the integrity of
amino acid biosynthesis in plants. Amino Acids 30: 173-183

Nikjoo H, Munson RJ, Bridges BA (1999) RBE-LET relationships in mutagenesis by ionizing
radiation. J Radiat Res (Tokyo) 40 Suppl: 85-105

O'Dell RE, James JJ, Richards JH (2006) Congeneric serpentine and nonserpentine shrubs
differ more in leaf Ca:Mg than in tolerance of low N, low P, or heavy metals. Plant and
Soil 280: 49-64

Padmanaban S, Chanroj S, Kwak JM, Li X, Ward JM, Sze H (2007) Participation of
endomembrane cation/H' exchanger AtCHX20 in osmoregulation of guard cells. Plant
Physiol 144: 82-93

Palacio S, Escudero A, Montserrat-Marti G, Maestro M, Milla R, Albert MJ (2007) Plants
living on gypsum: beyond the specialist model. Ann Bot (Lond) 99: 333-343

Pandey GK, Cheong YH, Kim BG, Grant JJ, Li L, Luan S (2007) CIPK9: a calcium sensor-
interacting protein kinase required for low-potassium tolerance in Arabidopsis. Cell Res
17: 411-421

Paul AL, Popp MP, Gurley WB, Guy C, Norwood KL, Ferl RJ (2005) Arabidopsis gene
expression patterns are altered during spaceflight. Adv Space Res 37: 1175-1181









Peterson DD, Benton EV, Tran M, Yang T, Freeling M, Craise L, Tobias CA (1977)
Biological effects of high-LET particles on corn-seed embryos in the Apollo-Soyuz Test
Project Biostack III experiment. Life Sci Space Res 15: 151-155

Qi H, Li D, Zhang J, Liu L, Zhang Q (2008) Study on extraction of agaropectin from Gelidium
ama~nsii and its anticoagulant activity. Chinese Journal of Oceanology and Limnology 26:
186-189

Rieder R, Gellert R, Anderson RC, Bruckner J, Clark BC, Dreibus G, Economou T,
Klingelhofer G, Lugmair GW, Ming DW, Squyres SW, d'Uston C, Wanke H, Yen A,
Zipfel J (2004) Chemistry of rocks and soils at Meridiani Planum from the Alpha Particle
X-ray Spectrometer. Science 306: 1746-1749

Rose JK, Braam J, Fry SC, Nishitani K (2002) The XTH family of enzymes involved in
xyloglucan endotransglucosylation and endohydrolysis: current perspectives and a new
unifying nomenclature. Plant Cell Physiol 43: 1421-1435

Rosti J, Barton CJ, Albrecht S, Dupree P, Pauly M, Findlay K, Roberts K, Seifert GJ
(2007) UDP-glucose 4-epimerase isoforms UGE2 and UGE4 cooperate in providing UDP-
galactose for cell wall biosynthesis and growth ofArabidopsis thaliana. Plant Cell 19:
1565-1579

Sahrawat KL, (2004) Iron toxicity in wetland rice and the role of other nutrients. J. Plant
Nutrition 27: 1471-1504

Salisbury FB, Campbell WF, Carman JG, Bingham GE, Bubenheim DL, Yendler B,
Sytchev V, Levinskikh MA, Ivanova I, Chernova L, Podolsky I (2003) Plant growth
during the Greenhouse II experiment on the Mir orbital station. Adv Space Res 31: 221-
227

Salisbury FB, Gitelson JI, Lisovsky GM (1997) Bios-3: Siberian experiments in
bioregenerative life support. Bioscience 47: 575-585

Sampedro J, Cosgrove DJ (2005) The expansion superfamily. Genome Biol 6: 242

Schmidt P, Kiefer J (1998) Deletion-pattern analysis of Alpha Particle and X-ray induced
mutations at the HPRT locus of V79 Chinese hamster cells. Mutat Res 421: 149-161

Schock I, Gregan J, Steinhauser S, Schweyen R, Brennicke A, Knoop V (2000) A member of
a novel Arabidopsis thaliana gene family of candidate Mg2+ ion transporters complements
a yeast mitochondrial group II intron-splicing mutant. Plant J 24: 489-501

Schuerger AC, Ming DW, Newsom HE, Ferl RJ, McKay CP (2002) Near-term lander
experiments for growing plants on Mars: requirements for information on chemical and
physical properties of Mars regolith. Life Support Biosph Sci 8: 137-147










Segonzac C, Boyer JC, Ipotesi E, Szponarski W, Tillard P, Touraine B, Sommerer N,
Rossignol M, Gibrat R (2007) Nitrate efflux at the root plasma membrane: identification
of an Arabidopsis excretion transporter. Plant Cell 19: 3760-3777

Seifert GJ, Roberts K (2007) The biology of arabinogalactan proteins. Annu Rev Plant Biol 58:
137-161

Shaul O (2002) Magnesium transport and function in plants: the tip of the iceberg. Biometals
15: 309-323

Shaul O, Hilgemann DW, de-Almeida-Engler J, Van Montagu M, Inz D, Galili G (1999)
Cloning and characterization of a novel Mg2+ H+ exchanger. EMBO J 18: 3973-3980

Shi H, Lee BH, Wu SJ, Zhu JK (2003) Overexpression of a plasma membrane Na /H
antiporter gene improves salt tolerance in Arabidopsis thaliana. Nat Biotechnol 21: 81-85

Smeets K, Ruytinx J, Van Belleghem F, Semane B, Lin D, Vangronsveld J, Cuypers A
(2008) Critical evaluation and statistical validation of a hydroponic culture system for
Arabidopsis thaliana. Plant Physiol Biochem 46: 212-218

Sottosanto JB, Gelli A, Blumwald E (2004) DNA array analyses of Arabidopsis thaliana
lacking a vacuolar Na /H+ antiporter: impact of AtNHX1 on gene expression. Plant J 40:
752-771

Stankovic B (2001) 2001: a plant space odyssey. Trends Plant Sci 6: 591-593

Storey JD, Tibshirani R (2003) Statistical significance for genome-wide studies. Proc Natl
Acad Sci U SA 100: 9440-9445

Stutte GW, Monje O, Hatfield RD, Paul AL, Ferl RJ, Simone CG (2006) Microgravity
effects on leaf morphology, cell structure, carbon metabolism and mRNA expression of
dwarf wheat. Planta 224: 1038-1049

Sychev VN, Levenskikh MA, Gostimsky SA, Bingham GE, Podolsky IG (2007) Spaceflight
effects on consecutive generations of peas grown onboard the Russian segment of the
International Space Station. Acta Astronaut 60: 426-432

Sychev VN, Shepeley EY, Meleshko GI, Gurieva TS, Levinskikh MA, Podolsky IG,
Dadasheva OA, Popov VV (2001) Main characteristics of biological components of
developing life support system observed during the experiments aboard orbital complex
Mir. Adv Space Res 27: 1529-1534

Taji T, Seki M, Saton M, Sakurai T, Kobayashi M, Ishiyama K, Narusaka Y, Narusaka M,
Zhu JK, Shinozaki K (2004) Comparative genomics in salt tolerance between
Arabidopsis and Arabidopsis-related halophyte salt cress using Arabidopsis microarray.
Plant Physiol 135: 1697-1709









Takahashi Y, Berberich T, Yamashita K, Uehara Y, Miyazaki A, Kusano T (2004)
Identification of tobacco HIN1l and two closely related genes as spermine-responsive genes
and their differential expression during the Tobacco mosaic virus-induced hypersensitive
response and during leaf- and flower-senescence. Plant Mol Biol 54: 613-622

Takeda S, Paszkowski J (2006) DNA methylation and epigenetic inheritance during plant
gametogenesis. Chromosoma 115: 27-35

Tatematsu K, Kumagai S, Muto H, Sato A, Watahiki MK, Harper RM, Liscum E,
Yamamoto KT (2004) MASSUGU2 encodes Aux/IAAl9, an auxin-regulated protein that
functions together with the transcriptional activator NPH4/ARF7 to regulate differential
growth responses of hypocotyl and formation of lateral roots in Arabidopsis thaliana. Plant
Cell 16: 379-393

Tocquin P, Corbesier L, Havelange A, Pieltain A, Kurtem E, Bernier G, Perilleux C (2003)
A novel high efficiency, low maintenance, hydroponic system for synchronous growth and
flowering ofArabidopsis thaliana. BMC Plant Biol 3: 2

Turner TL, von Wettberg EJ, Nuzhdin SV (2008) Genomic analysis of differentiation
between soil types reveals candidate genes for local adaptation in Arabidopsis lyrata. PLoS
ONE 3: e3183

Ubeda-Tomas S, Swarup R, Coates J, Swarup K, Laplaze L, Beemster GT, Hedden P,
Bhalerao R, Bennett MJ (2008) Root growth in Arabidopsis requires gibberellin/DELLA
signalling in the endodermis. Nat Cell Biol 10: 625-628

Ulker B, Shahid Mukhtar M, Somssich IE (2007) The WRKY70 transcription factor of
Arabidopsis influences both the plant senescence and defense signaling pathways. Planta
226: 125-137

van de Mortel JE, Almar Villanueva L, Schat H, Kwekkeboom J, Coughlan S, Moerland
PD, Ver Loren van Themant E, Koornneef M, Aarts MG (2006) Large expression
differences in genes for iron and zinc homeostasis, stress response, and lignin biosynthesis
distinguish roots of Arabidopsis thaliana and the related metal hyperaccumulator Thlaspi
caerulescens. Plant Physiol 142: 1127-1147

Vercauteren I, Niebel A, Van Montagu M, Gheysen G (2002) Arabidopsis thaliana mutants
with an altered susceptibility to the root-knot nematode M~eloidogyne incognita. Meded
Rijksuniv Gent Fak Landbouwkd Toegep Biol Wet 67: 51-61

Vissenberg K, Fry SC, Verbelen JP (2001) Root hair initiation is coupled to a highly localized
increase of xyloglucan endotransglycosylase action in Arabidopsis roots. Plant Physiol
127: 1125-1135

Wolfinger RD, Gibson G, Wolfinger ED, Bennett L, Hamadeh H, Bushel P, Afshari C,
Paules RS (2001) Assessing gene significance from cDNA microarray expression data via
mixed models. J Comput Biol 8: 625-637









Wu J, Lenchik NI, Gerling IC (2008) Approaches to reduce false positives and false negatives
in the analysis of microarray data: applications in type 1 diabetes research. BMC
Genomics 9 Suppl 2: S12

Yang S, Yang H, Grisafi P, Sanchatjate S, Fink GR, Sun Q, Hua J (2006) The BON/CPN
gene family represses cell death and promotes cell growth in Arabidopsis. Plant J 45: 166-
179

Yen AS, Gellert R, Schroder C, Morris RV, Bell JF, 3rd, Knudson AT, Clark BC, Ming
DW, Crisp JA, Arvidson RE, Blaney D, Bruckner J, Christensen PR, DesMarais DJ,
de Souza PA, Jr., Economou TE, Ghosh A, Hahn BC, Herkenhoff KE, Haskin LA,
Hurowitz JA, Joliff BL, Johnson JR, Klingelhofer G, Madsen MB, McLennan SM,
McSween HY, Richter L, Rieder R, Rodionov D, Soderblom L, Squyres SW, Tosca
NJ, Wang A, Wyatt M, Zipfel J (2005) An integrated view of the chemistry and
mineralogy of Martian soils. Nature 436: 49-54

Yoshimoto N, Inoue E, Watanabe-Takahashi A, Saito K, Takahashi H (2007)
Posttranscriptional regulation of high-affinity sulfate transporters in Arabidopsis by sulfur
nutrition. Plant Physiol 145: 378-388

Yoshimoto N, Takahashi H, Smith FW, Yamaya T, Saito K (2002) Two distinct high-affinity
sulfate transporters with different inducibilities mediate uptake of sulfate in Arabidopsis
roots. Plant J 29: 465-473

Zhang G, Chen M, Chen X, Xu Z, Guan S, Li LC, Li A, Guo J, Mao L, Ma Y (2008)
Phylogeny, gene structures, and expression patterns of the ERF gene family in soybean
(Glycine max L.). J Exp Bot 59: 4095-4107

Zhao J, Shigaki T, Mei H, Guo YQ, Cheng NH, Hirschi KD (2009) Interaction between
Arabidopsis Ca2+ H+ exchangers CAX1 and CAX3. J Biol Chem 284: 4605-4615

Zhao Y, Christensen SK, Fankhauser C, Cashman JR, Cohen JD, Weigel D, Chory J
(2001) A role for flavin monooxygenase-like enzymes in auxin biosynthesis. Science 291:
306-309

Zhao Y, Hull AK, Gupta NR, Goss KA, Alonso J, Ecker JR, Normanly J, Chory J, Celenza
JL (2002) Trp-dependent auxin biosynthesis in Arabidopsis: involvement of cytochrome
P450s CYP79B2 and CYP79B3. Genes Dev 16: 3100-3112









BIOGRAPHICAL SKETCH

Anne Visscher was born in the Netherlands and completed her master' s degree in soil,

water and atmosphere at the Wageningen University in 2003. She has had a long-standing

interest in advanced life support systems and further developed this interest through an internship

with the Biosphere Foundation during her master' s program. This provided a connection with the

lab of Rob Ferl at the University of Florida, where she subsequently commenced her PhD proj ect

in January 2005. She received her PhD from the University of Florida in the Summer of 2009.





PAGE 1

RESPONSES OF ARAB IDOPSIS TO HIGH LEV ELS OF MAGNESIUM SULFATE AND OF WHEAT TO A SPACEFLIGHT ENVIRONMENT; CONSEQUENCES FOR (EXTRA)TERRESTRIAL PLANT GROWTH By ANNE MARIEKE VISSCHER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Anne Marieke Visscher 2

PAGE 3

To my parents 3

PAGE 4

ACKNOWLEDGMENTS I would like to thank my family and friends for supporting me during this PhD. Without my advisors Rob Ferl and Anna-Lisa Paul, and my committee members Andrew Schuerger, Charles Guy and Matias Kirst, th is project would not have been possible. I sincerely appreciate all the knowledge and inspiration th ey have offered over the years. I thank all the members of the Ferl Lab, especially Beth Laughner, Jordan Barney, Brian Fuller, Matthew Reyes, John Mayfield, Michael Manak and Tufan Gk rmak for their help, insi ghts and friendship. The Interdisciplinary Center for Biot echnology Research at the Univer sity of Florida is recognized for providing valuable assistance to the work pr esented in this disserta tion. The UF College of Agricultural and Life Sciences, the UF Department of Horticultu ral Sciences, the Dutch Minister for Education, Culture and Science, and the Du tch Prins Bernard Cultuurfonds are acknowledged for their financial support. Finally, I would lik e to acknowledge my previous advisors at Wageningen University and mentors at the Institute of Ecotechnics and the Biosphere Foundation, without whom I would not have been prepared for this project. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .........................................................................................................................8LIST OF OBJECTS .......................................................................................................................11ABSTRACT ...................................................................................................................................12CHAPTER 1 INTRODUCTION................................................................................................................. .14Potential Phytotoxic Elements in the Regolith on Mars .........................................................14Analogue Soils on Earth .........................................................................................................16Magnesium Sulfate .................................................................................................................18Spaceflight Environment ........................................................................................................202 GROWTH RESPONSES OF WILDTY PE ARABIDOPSIS AND KNOCKOUT MUTANTS TO EXCESS LEVELS OF MAGNESIUM SULFATE; CONSEQUENCES FOR (EXTRA)TERRESTRI AL PLANT GROWTH............................................................22Introduction .............................................................................................................................22Results .....................................................................................................................................28Discussion ...............................................................................................................................30Conclusions .............................................................................................................................32Materials and Methods ...........................................................................................................343 TRANSCRIPTOME RESPONSES OF COL-0 AND CAX1-1 ARABIDOPSIS TO EXCESS LEVELS OF MAGNESIUM SULFATE; CONSEQUENCES FOR (EXTRA)TERRESTRIAL PLANT GROWTH.....................................................................54Introduction .............................................................................................................................54Results .....................................................................................................................................58Discussion ...............................................................................................................................69Conclusions .............................................................................................................................84Materials and Methods ...........................................................................................................864 EFFECTS OF A SPACEFLIGHT ENVIRO NMENT ON HERITABLE CHANGES IN WHEAT GENE EXPRESSION...........................................................................................126Introduction ...........................................................................................................................126Results ...................................................................................................................................129 5

PAGE 6

Discussion .............................................................................................................................131Conclusion ............................................................................................................................136Materials and Methods .........................................................................................................136CONCLUSIONS ..........................................................................................................................145LIST OF REFERENCES .............................................................................................................151BIOGRAPHICAL SKETCH .......................................................................................................164 6

PAGE 7

LIST OF TABLES Table page 2-1 Overview of wildtype and mutant Arabidopsis seed lines .................................................402-2 Overview of mutant/wild type growth experiments ...........................................................402-3 Elemental composition of first set of six mrs2-10 plants in ppm ......................................412-4 Elemental composition of second set of six mrs2-10 plants in ppm ..................................422-5 ANOVA (t-test) results for specific genot ype*concentration eff ects in the growth experiments (Table 2-2) .....................................................................................................432-6 ANOVA (t-test) results for specific genotype*concentration effects in the ten growth experiments (Table 2-2) .....................................................................................................443-1 Experimental conditions of one color microarray experiment ..........................................903-2 Number of genes (and % of total) in GO molecular functional categories per comparison .........................................................................................................................903-3 Genes of Arabidopsis thaliana ( col-0 ) with differential expression at q < 0.001 at Time 1 ................................................................................................................................913-4 Genes of Arabidopsis thaliana ( col-0 ) with differential expression > 3 fold at Time 1 ....923-5a Q-PCR results of col-0 gene expression at time 3 trea tment versus time 1 control; RNA sources are the same as for the microarray experiment ............................................933-5b Q-PCR non-normalized results of col-0 gene expression at time 3 treatment versus time 1 control; RNA sources are the same as for the microarray experiment ...................933-6 Q-PCR results of col-0 gene expression at time 3 trea tment versus time 3 control; repetition 1 .........................................................................................................................933-7 Q-PCR results of col-0 gene expression at time 3 trea tment versus time 3 control; repetition 2 .........................................................................................................................933-8 Genes with differential expression at q < 0.05 between Arabidopsis thaliana cax1-1 and col-0 treated for 3 hours ..............................................................................................943-9 Genes and related primer sequences selected for Q-PCR ..................................................94 7

PAGE 8

LIST OF FIGURES Figure page 2-1 Mrs2-10 homozygous knockout mutant line .....................................................................452-2 RT-PCR analysis of the mrs2-10 mutant ...........................................................................452-3 Two sets of six mrs2-10 plants growing at the Pur due Ionomics project facility ..............452-4 Element z-score values for the first set of six mrs2-10 plants ...........................................462-5 Element z-score values for the second set of six mrs2-10 plants ......................................462.6 Growth experiments on agar. FW biomass of mutant lines is compared to that of their respective wildt ype backgrounds at different levels of MgSO H O in the agar medi um4 2..............................................................................................................................472-7 Growth experiments on soil. FW shoot biomass of mutant lines is compared to that of their respective wildtype backgr ounds at different levels of MgSO H O in the soil medium4 2........................................................................................................................482-8 Average fresh weight biomass of mrs2-10 and col-0 seedlings in response to increasing concentrations of MgSO H O in agar medium4 2............................................492-9 Average fresh weight shoot biomass of mrs2-10 and col-0 plants in response to increasing concentrations of MgSO H O in soil medium4 2.............................................492-10 Average leaf chlorophyll content of mrs2-10 and col-0 plants in response to increasing concentrations of MgSO H O in soil medium4 2.............................................502-11 Average fresh weight biomass of sel1-10 and ws seedlings in response to increasing concentrations of MgSO H O in agar medium4 2.............................................................502-12 Average fresh weight shoot biomass of sel1-10 and ws plants in response to increasing concentrations of MgSO H O in soil medium4 2.............................................512-13 Average leaf chlorophyll content of sel1-10 and ws plants in response to increasing concentrations of MgSO H O in soil medium4 2...............................................................512-14 Average fresh weight shoot biomass of cax1-1 and col-0 plants in response to increasing concentrations of MgSO H O in soil medium4 2.............................................522-15 Average leaf chlorophyll content of cax1-1 and col-0 plants in response to increasing concentrations of MgSO H O in soil medium4 2...............................................................522-16 Average fresh weight shoot biomass of cax1/cax3 and col-0 plants in response to increasing concentrations of MgSO H O in soil medium4 2.............................................53 8

PAGE 9

2-17 Average leaf chlorophyll content of cax1/cax3 and col-0 plants in response to increasing concentrations of MgSO H O in soil medium4 2.............................................533-1 Hydroponic Arabidopsis growth ........................................................................................953.2 Harvest of hydroponically grown Arabidopsis roots .........................................................963.3 Overview of microarray experiment ..................................................................................963-4 Volcano plot of Time 1 ......................................................................................................973-5 Volcano plot of Time 2 ......................................................................................................983-6 Volcano plot of Time 3 ......................................................................................................993-7 Volcano plot of cax1-1 versus col-0 at time 3 .................................................................1003-8 Venn diagram of col-0 time series ...................................................................................1013-9 A hierarchical average linkage cluster analysis using uncente red correlation was done across Time 1, 2 and 3 based on th e genes with significant expression differences at Time 1 .......................................................................................................1023-10a Differentially expressed transcripts encoding metabolic enzymes in the ethylene biosynthesis pathway .......................................................................................................1083-10b Differentially expressed transcripts encoding metabolic enzymes in the abscisic biosynthesis pathway .......................................................................................................1093-10c Differentially expressed transcripts encoding metabolic enzymes in the jasmonate biosynthesis pathway .......................................................................................................1093-10d Differentially expressed transcripts encoding enzymes involved in gibberellins metabolism .......................................................................................................................1103-10e Differentially expressed transcripts encoding enzymes involved in glycerolipid metabolism .......................................................................................................................1113-10f Differentially expressed transcripts encoding metabolic enzymes in the hexosamine biosynthetic pathway .......................................................................................................1113-10g Differentially expressed transcripts encoding enzymes involved in chlorophyll breakdown ........................................................................................................................1123-10h Differentially expressed transcripts encoding enzymes involved in porphyrin and chlorophyll metabolism ...................................................................................................1133-10i Differentially expressed transcripts encoding metabolic enzymes in the histidine biosynthesis pathway .......................................................................................................114 9

PAGE 10

3-10j Differentially expressed transcript s encoding metabolic enzymes involved in galactose metabolism .......................................................................................................1153-10k Differentially expressed transcripts encoding enzymes involved in glycolysis ..............1163-10l Differentially expressed transcripts encoding enzymes in the citric acid cycle ..............1173-11 Hierarchical average linkage cluster anal ysis of transporter gene expression using uncentered correlation ......................................................................................................1203-12 Whisker box plots representing gene expr ession ratio distributions for the Q-PCR analysis of four genes showing differe ntial expression on the microarrays in col-0 at time 3 treatment versus time 1 control .............................................................................1243-13 Whisker box plots representing gene expr ession ratio distributions for the Q-PCR analysis of four genes showing differe ntial expression on the microarrays in col-0 at time 3 treatment versus time 3 control .............................................................................1243-14 Whisker box plots representing gene expr ession ratio distributions for the Q-PCR analysis of four genes showing differe ntial expression on the microarrays in col-0 at time 3 treatment versus time 3 control .............................................................................1254-1 Wheat growth in the Laboratory Biosphere .....................................................................1404-2 Flowchart of wheat experiments ......................................................................................1414-3 ANOVA (t-test) results ....................................................................................................1424-4 Multiple testing correction: Benjamini and Hochberg method ........................................1434-5 Multiple testing correction: Q-value method ...................................................................144 10

PAGE 11

LIST OF OBJECTS Object page 3-1 Microarray raw signal data (.xls file 35 MB) ....................................................................683-2 Microarray analysis data (.xls file 1 MB) ..........................................................................68 11

PAGE 12

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Par tial Fulfillment of the Requirements for the Degree of Doctor of Philosophy RESPONSES OF ARABIDOPSIS TO HIGH LEV ELS OF MAGNESIUM SULFATE AND OF WHEAT TO A SPACEFLIGHT ENVIRONMENT; CONSEQUENCES FOR (EXTRA)TERRESTRIAL PLANT GROWTH By Anne Marieke Visscher August 2009 Chair: Robert J. Ferl Major: Horticultural Science The ability to utilize in situ resources is crucial for the success of extended manned space exploration of other planetary surfaces such as the Moon or Mars. Martian regolith containing potentially phytotoxic levels of elements is a potential medium for plant growth in bioregenerative life support systems. Studies of surface materials on Mars have detected a variety of hydrated sulfate minerals, including highly soluble magnesium sulfate minerals. Localized weight percentages of magnesium sulf ate can reach 10 % in th e regolith. Levels of magnesium and sulfate ions toxic to crop plants have been descri bed for soil types on Earth, such as serpentine and acid sulfate so ils. We tested whether Arab idopsis knockout lines of genes encoding plasma membrane localized ion transporters in peripheral root cells can thrive in soils containing hyper elevated levels of magnesium sulfate. The selected mrs2-10 and sel1-10 mutant backgrounds do not mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype plants. Based on these findings, a mi croarray experiment was done to characterize the early gene expression responses of col-0 Arabidopsis roots to elev ated concentrations of magnesium sulfate. The cax1-1 mutant line, which has been show n to exhibit increased tolerance for high levels of magnesium, was also in cluded in the experiment. The results for col-0 point to 12

PAGE 13

13 a reduction in root growth and vacuolar storag e of magnesium and sulfate upon high magnesium sulfate treatment. Although many tr ansporters or their potential re gulators were differentially expressed, more research is need ed to discover whether any of those are involved in magnesium sulfate transport. The down-regulated expression of cax1-1 is a natural response to high magnesium sulfate in Arabidopsis, and can explai n the fact that only th ree transcripts were differentially expressed between cax1-1 and col-0 Furthermore, we tested whether changes in gene expression patterns can be detected in wh eat plants that are seve ral generations removed from growth in space, compared to control plants with no spaceflight expo sure in their lineage. We found that none of the wheat genes represented by 10,263 probes on custom-made microarrays showed a statistically significant difference in expression.

PAGE 14

CHAPTER 1 INTRODUCTION Potential Phytotoxic El ements in the Regolith on Mars The planet Mars differs significantly from Earth with regard to gravity, geology, hydrology, atmosphere, radiation environment and temperature. The conditions on Mars require habitats and suits that shield humans on manned missions from low temperature, low atmospheric pressure and harmful radiation. L ong duration missions demand the efficient use of local planetary resources and the recycling of limited materials such as water, pressurized atmosphere and organic matter while produci ng food (Barta and Henni nger, 1994). Biological processes are likely to be used for most wate r, air and food regenera tion during long missions (Drysdale et al., 2003). Large-scale experimenta tion with bioregenerative life support for human space exploration started with the Bios-3 facility in Siberia in 1972 (Salisbury et al., 1997). In the context of plant growth in a future bioregenerative life s upport system on Mars, Schuerger et al. (2002) indicate that the use of in situ re golith may have several advantages over hydroponic systems. These include the immediate bioavailability of plant essential ions low-tech mechanical support for plants, and easy access of in situ materials once on the surface. However, plant growth may be reduced or inhibited by high leve ls of chemical elements in the regolith. The most highly phytotoxic materials on Mars are thought to be pr esent in the mobile-element component of the regolith and composed of sulfat es and chlorides (Schuerger et al., 2002). The current study focuses on sulfate minerals and the potential toxicity of the chemical elements these include. Sulfate minerals have been detected on the surface of Mars both from space (remote sensing) and on the ground (robotic landers). The formation of sulfate minerals on Mars is likely to have been caused by oxidative, acid-sulfate we athering of basaltic surface material (Golden et 14

PAGE 15

al., 2005). Possible acid -sulfate conditions propose d to have occurred include sulfuric-acid vapors (acid fog) and waters rich in sulfuric acid. During weathe ring, elements such as Mg, Ca and Fe will have leached from the basaltic parent material (e.g. olivine, feldspar, pyroxene) and reacted with S to form sulfates and other seconda ry minerals such as iron oxides (Golden et al., 2005). The Mars Express OMEGA remote sensing mission detected three principal types of hydrated sulfate deposits: layered deposits with in Valles Marineris, extended deposits exposed from beneath younger units as in Terra Meridiani, and the dark dunes of the northern polar cap (Bibring et al., 2006). Reflectance sp ectra from Valles Marineris, Margaritifer Sinus and Terra Meridiani indicate the associ ation of kieserite (MgSO4H2O), gypsum (CaSO4H2O) and polyhydrated sulfate minerals ((MgSO 4H2O), (Fe2+Fe4 3+(SO4)6 (OH)220H2O) (Fe2+Al2(SO4)422H2O)) with light-tone layered deposits (Gendrin et al., 2005). In the northern circumpolar regions, calcium-rich sulfates such as gypsum have been identified in a hydrated area correlating to the dark l ongitudinal dunes of Olympia Plan itia (Langevin et al., 2005). Spectra from eastern Terra Meridiani in particular suggest that kieserite is present in etched terrain deposits (Arv idson et al., 2005). Analyses by the Mars Exploration Rover land ers at Meridiani Planum and Gusev crater have also indicated the presence of sulfate mi nerals. The outcrops observed at Meridiani Planum all contain sulfates with volume abundances of 15 to 35%. Mgand Ca-sulfates are dominant, and the average volume abundance of the iron hydroxide sulfate mineral jarosite1 in the outcrop is around 10% (Christensen et al., 2004). At Guse v crater, Mgand Ca-sulfates are estimated to be present in the outcrops and rocks of the Colu mbia Hills up to 10 and 11% respectively (Ming et al., 2006). The sulfate mineral c ontent of the localized Paso Robles soil is estimated at 25-29% ) 1 (K, Na, H3O)(Fe3xAlx)(SO4)2(OH)6, where x <1 (Klingelhofer et al., 2004 15

PAGE 16

ferric sulfate, 10% magnesium sulfate, 3-4% calc ium sulfate and 2-5% other sulfates (Ming et al., 2006). Analyses in the 11 cm deep The Boroughs trench, which was dug in intercrater terrain, suggested the presence of approximately 7 wt % sulfates (Mg-, Caand Fe-sulfate) in the subsurface regolith (Haskin et al., 2005). El emental analysis of several other surface and trench soils at Meridiani Planum and Gusev crater revealed that the averag e weight percentage of sulfur varied between 1.92 and 2.92 (Rieder et al., 2004) and 1.69 and 2.95 (Gellert et al., 2004) respectively. The elemental sulfur is thought to be present in th e soil in the form of sulfate (Banin et al., 1997). The average concentrations of all elements measured in these soils were compared with each other and w ith those found in soils at loca tions visited during the Viking 1, Viking 2 and Pathfinder missions (G ellert et al., 2004; Rieder et al., 2004). In general, weight percentages of major elements were found to be si milar, with some larger differences existing for minor elements, supporting global mixing of soil by dust storms and admixture of debris from local rocks (Gellert et al., 2004; Rieder et al., 2004). Yen et al. (2005) also conclude that soil compositions at five landing sites on Mars are more similar to each other than to the analyzed rocks (Yen et al., 2005). Overall th ese results show that sulfate mi nerals are generally present in the loose regolith in amounts related to 1.3 to 3 we ight % S (Gellert et al., 2004), and that they can be present locally in even higher quantities. Furthermore, th e main cations associated with the sulfate anion seem to be Ca2+, Mg2+ and Fe2+/3+. Analogue Soils on Earth Soils on Earth with high levels of sulfate minerals or their individual ions (Ca2+, Fe2+, Mg2+, SO4 2-) could be (partial) analogues for regolith high in Ca-sulfate, Fe-sulfate or Mg-sulfate on Mars. The occurrence of high levels of CaSO4H2O (gypsum) minerals in soils on Earth is confined to arid and semi-arid climates wher e low precipitation prevents gypsum from being removed by leaching (Palacio et al., 2007). Among the adverse physic al features of gypsum soils 16

PAGE 17

are the presence of a hard soil su rface crust, which can restrict seedling establishment, the mechanical instability of the soil material due to its lack of plasticit y, cohesion and aggregation, and, in certain areas, its low porosity, which might limit the penetra tion of plant roots. Chemically adverse features of gypsum soils ar e mainly related to the intense nutritional impoverishment of the soil caused by the exchange of calcium for other ions retained in the soil complex, and by the high concentration of sulfate ions (Palacio et al., 2007). Relatively soluble ferrous or ferric sulfate minera ls such as melanterite (FeSO4H2O) and coquimbite (Fe2(SO4)3H2O) can form in soils when ferrous or fe rric sulfate-rich solutions are desiccated. These minerals, or the dissolved ions of them oxidize and hydrolyze read ily to form sulfuric acid and iron (hydr)oxide minerals (Fanning an d Burch, 2006). Magnesium sulfate is so soluble that it rarely occurs in soils on Earth (Barber, 1995). Magnesium sulfate minerals such as epsomite (MgSO 4H2O) and hexahydrite (MgSO4H2O) are important constituents of evaporative soil environments in cold climates a nd of cold desert environments, for example in North Dakota and Saskatchewan (Chou and Seal, 2003). Survey reports re garding these soils generally do not include inform ation on concentrations of Mg2+ and SO4 2in the soil solution upon dissolution of magnesium sulfate minerals by precipitation, presumably because these ions are immediately leached out from the roo ting zone into the deeper soil layers. When focusing on the chemical elements contained in sulfate minerals, we find diverse soils on Earth that are naturally enriched in these. Soils rich in sulfur are for example saline soils, heavy metal soils, acid sulfate soils or soils in the vicinity of volcanoes, S/CO2 vents and lignite burns (Ernst, 1998). Iron levels toxic to plant gr owth generally occur in submerged soils that allow for reduction of Fe3+, since the reduced form Fe2+ is directly bioavailable. Examples of soil types that can exhibit iron toxicity when subm erged are acid Ultisols and Oxisols, and acid 17

PAGE 18

sulfate soils that are rich in Fe(III) oxide hydr ates (Sahrawat, 2 004). Interflow of Fe from adjacent areas and salt content can also play a ro le in increasing soluble iron levels. When soils that meet these criteria are used for rice pr oduction, the concentra tion of water-soluble Fe2+ can increase to a few hundred milligrams per liter within weeks following submergence (Sahrawat, 2004). An example of a soil type high in bioavail able magnesium is serpentine soil. Serpentine soils are formed from the weathered products of ultramafic rocks, whose common mineral denominator is some form of iron magnesium si licate. In addition, serp entine soils may contain impurities such as minerals with nickel, chromium and cobalt (Kruckeberg, 1999). Common characteristics of serpentine soils are (1) hi gh concentrations of elements such as iron, chromium, nickel and cobalt, (2) low concentr ations of plant nutrien ts such as nitrogen, phosphorus and potassium, (3) a low Ca:Mg quotient compared with non-serpentine soils, and (4) lower clay contents with lower exchange capacity than other soils (Brooks, 1987). The infertility of serpentine soils in the context of (crop) plant growth is largely related to their chemical composition, with the unfavorable eff ect of magnesium being a major cause of the serpentine problem; the magnesium content of so me soils can be as high as 36% MgO, and 4.32 mg/L in the soil solu tion (Brooks, 1987). Magnesium Sulfate On Mars, no direct measurements of ion concentr ations in wetted regoli th have so far taken place in regions high in sulfate minerals. Factor s such as parent mate rial, weathering history, texture, structure, organic matter content and hyd rology all interact to determine bioavailability of nutrients in a soil solution accessible to pl ant roots (Barber, 1995). Given the presence of sulfur in the form of su lfate on Mars, the concentration of bioavailable SO4 2in a soil solution will depend on the relative weight percentages of the different soluble sulfate minerals in the regolith, and the interplay between soil char acteristics and hydrology mentioned above. 18

PAGE 19

Regarding Ca2+, Mg2+ and Fe2+, soluble sulfate minerals will not be the only source for these cations in the bioavailable pool. Their overall bioava ilable concentrations in the soil solution of Martian regolith upon wetting are difficult to pred ict without having done direct measurements on Mars. This is especially true in the case of Fe, which can occur in two different oxidation states: Fe2+ and Fe3+. The relation between these forms of iron in the soil solution depends on the redox status of the soil (Barber, 1995). Plants primarily absorb Fe2+; either directly or by first reducing Fe3+-chelates and transporting the resulting Fe2+ (Curie and Briat, 2003). Graminaceous plants are furthermore able to produce phytosiderophores (PS), which can form Fe3+-PS complexes that are taken up by specific tr ansporters (Curie and Briat, 2003). Measurements of ion concentrations in wetted regolith in high sulfate regions on Mars have not yet taken place. The focus of this research is therefore on documenting plant responses to elevated levels of specific hydrated sulfat e minerals in solution. Preliminary experiments indicated that increasing concentrations of disso lved sulfate minerals that were added to 1.3% agar medium with 0.5x Murashige and Skoog nutrient solution had different effects on growth of wildtype Arabidopsis on Petri dishes. FeSO4H2O (EDTA) showed the first growth limiting effects at lower concentrations than MgSO4H2O, while CaSO4H2O did not show clear effects up to its maximum soluble concentrati on. These observations indicated that among the tested individual sulfate minerals toxicity can be dominated by th e associated cation. Similarl the major problem on sulfur-enriche d saline soils is not the surplus of sulfur but rather elevat sodium, and the selection pattern of plant species on sulfur-enric hed heavy metal soils or acid sulfate soils is generally governed more by th e heavy metal cation than by the sulfur-anion (Ernst, 1997; Ernst, 1998). y, ed 19

PAGE 20

The phytotoxicity of iron upon dissolution of hydrated iron su lfates like FeSO4H2O is dependent on reduced soil conditions, such as in submerged soils. The submersion of soils, as is done for rice production, is unlikely to be implem ented in a bioregenerat ive life support system on Mars due to the limitation of water as a loca l resource. The other su lfate mineral showing inhibition of plant growth at high bioavailable concentrations when other nutrients remain constant is MgSO4H2O. Magnesium sulfate is highly solu ble in hydrated mineral form, which is why it rarely occurs in soils on Earth. Hydrated forms such as kieserite (MgSO4H2O) and epsomite (MgSO4H2O) have been detected on Mars by th e Mars Express Satellite. Magnesium sulfate is therefore likely to become bioavail able at high concentrations compared to other nutrients in the soil solution upon first watering of regolith in materially closed life support systems situated in high sulfate mineral regions on Mars. This study documents the responses of Arabidopsis to high levels of magnesium sulf ate in solution and explores plant molecular strategies to reduce accumulation of magnesium sulfate w ithin the plant in or der to enhance plant tolerance. Since Mg2+ is a dominant toxic ion compared to SO4 2when MgSO4H2O is dissolved, serpentine soils that are high in Mg2+ could function as part ial analogue soils for regolith high in magnesium sulfate on Mars in this study. Spaceflight Environment Before plants or seeds arrive on Mars to be planted in a bior egenerative life support system, they need to be transported in a spaceflight environment. Plants may be grown as part of a life support system in the transport vehicle during a human mission to Mars. A fundamental question in early space biology research was whethe r plants could complete one or more full life cycles, from seed to seed, under spaceflight conditions. As space research progressed, plant species such as Arabidopsis thaliana Brassica rapa, Triticum aestivum L. cultivar USU-Apogee and Pisum sativum completed their life cycles in growth units on orbiting space stations (Merkys 20

PAGE 21

21 et al., 1984; Levinskikh et al., 2000; Musgrave et al., 2000; Sych ev et al., 2007). The primary natural identified environmental factors that di ffer between a space station and a growth chamber on Earth are gravitational forces and the radiation environment. Tropic responses in plants are directly affected by microgravity (Brown et al., 1995), yet there is no evid ence that microgravity per se has any detrimental eff ect on plant metabolism (Stutte et al., 2006). However, indirect effects of microgravity, such as reduced air circulation due to th e lack of convective mixing, can affect growth when not mediated technically (Musgrave and Kuang, 2003). Much of the early work in orbit highlighted the necessity for tight control of the environment, including temperature, humidity, lighting, nutrient deli very, and atmospheric composition (Stankovic, 2001). The successful life cycle expe riments indicate that if the i ndirect effects of microgravity and a closed environment are ameliorated, the low earth orbit spaceflight environment is not hostile to plant growth, development and viable seed formation. A fundamental question to follow is whether the long-term exposure (such as for one or more life cycles) of plants to the spaceflight environment with its microgravity and radiation parameters can cause changes in subsequent generations. Plants might be grown for multiple life cycles as part of a life support system in a transport vehicle during a human mission to Mars. Some of the seeds harvested from these plants could subsequently be planted in advanced life support systems on the surface of Mars. To analyze in detail whether plants grown in spaceflight conditions show changes in subsequent generati ons, gene expression in wheat plants that are three generations removed from growth in the MIR space station was compared to gene expression in wheat plants with no spaceflight exposure in their lineage.

PAGE 22

CHAPTER 2 GROWTH RESPONSES OF WILDTYPE ARABIDOPSIS AND KNOCKOUT MUTANTS TO EXCESS LEVELS OF MAGNESIUM SULFATE; CONSEQUENCES FOR (EXTRA)TERRESTRIAL PLANT GROWTH Introduction The planet Mars differs significantly from Earth with regard to gravity, geology, hydrology, atmosphere, radiation environment and temperature. The conditions on Mars require habitats and suits that shield humans on manned missions from low temperature, low atmospheric pressure and harmful radiation. L ong duration missions demand the efficient use of local planetary resources and the recycling of limited materials such as water, pressurized atmosphere and organic matter while produci ng food (Barta and Henni nger, 1994). Biological processes are likely to be used for most wate r, air and food regenera tion during long missions (Drysdale et al., 2003). Large-scale experimenta tion with bioregenerative life support for human space exploration started with the Bios-3 facility in Siberia in 1972 (Salisbury et al., 1997). In the context of plant growth in a future bioregenerative life s upport system on Mars, Schuerger et al. (2002) indicate that the use of in situ re golith may have several advantages over hydroponic systems. These include the immediate bioavailability of plant essential ions low-tech mechanical support for plants, and easy access of in situ materials once on the surface. However, plant growth may be reduced or inhibited by high leve ls of chemical elements in the regolith. The most highly phytotoxic materials on Mars are thought to be pr esent in the mobile-element component of the regolith and composed of sulf ates and chlorides (Sc huerger et al., 2002). As part of the search for evidence of past water on Mars, hydrated sulfate minerals have been detected on the surface of Mars both fr om space (remote sensing) and on the ground (robotic landers). The formation of sulfate minera ls on Mars is likely to have been caused by oxidative, acid-sulfate weatheri ng of basaltic surface material (Golden et al., 2005). Possible 22

PAGE 23

acid-sulfate conditions proposed to have occurred include sulfur ic-acid vapors (acid fog) and waters rich in sulfuric acid. During weathering, elements such as Mg, Ca and Fe will ha ve leached from the basaltic parent material (e.g. olivine, feldspar, pyroxene) and reacted with S to form sulfates and other secondary minerals such as iron oxides (Golden et al., 2005). Measurements of ion concentrations in wetted regolith of high sulfate regions on Mars have not yet taken place. The focus of this research is therefore on documenting plant responses to elevated levels of specific hydrated sulfate mi nerals in solution. Preliminary experiments indicated that increasing concentrations of dissolved MgSO4H2O (epsomite) that were added to 1.3% agar medium with 0.5x Murashige and Skoog nutrient solution showed increasingl limiting effects on growth of wildtype Arabi dopsis on Petri dishes. Hydrated forms of magnesium sulfate such as MgSO y 4H2O and MgSO4H2O (kieserite) have been detected in several regions by the Mars Express Satellite (Arvidson et al., 2005; Gendrin et al., 2005; Bibring et al., 2006). Analyses by the Mars Explor ation Rover landers at Meridiani Planum and Gusev crater have also indicated the presence of high levels of magnesium sulfate minerals (up to 10 %) in outcrops and soils (Christensen et al., 2004; Haskin et al., 2005; Ming et al., 2006). Magnesium sulfate is highly soluble in hydrated mineral form, which is why it rarely occurs in soils on Earth (Barber, 1995). Magne sium sulfate is therefore likely to become bioavailable at high concentrations compared to other nutrients in the soil solution upon first watering of regolith in materially closed life support syst ems situated in high sulfate mineral regions on Mars. Strategies to alleviate high ma gnesium sulfate stress in a po tential bioregenerative life support system on Mars could include (1) remediation of regolith by leaching of soluble magnesium sulfate minerals with water before firs t use, (2) alleviation of nutrient deficiencies 23

PAGE 24

caused by high levels of Mg2+ and SO4 2ions by addition of appropr iate fertilizers, and (3) adaptation to the regolith conditions by planti ng crop genotypes tolera nt of high magnesium sulfate. Remediation and alleviat ion require additional materials (such as water) and extra time, both of which are limited on a manned mission to Mars. The strategy of adaptation on the other hand would allow for immediate use of the re golith for crop production in an advanced life support system located in a region high in sulfate minerals on Mars. The focus of this study is therefore to explore the possibi lity of using adaptation as a strategy on future missions by identifying plant variants tolera nt of high magnesium sulfate. Magnesium (Mg2+) is the most abundant divalent cati on in a living cell. It stabilizes membranes and is associated with ATP in a num ber of enzymatic reactions (Li et al., 2001). Magnesium is essential for the function of many enzymes, including RNA polymerases, ATPases, protein kinases, phosphatases, glutathione synthase, and carboxyl ases. In higher plants, Mg2+ is the central atom of the chlorophyll molecule and a bridgi ng element for the aggregation of ribosomes. Key chloroplast enzymes are st rongly affected by small variations in Mg2+ levels in the cytosol and the chloroplast, exem plifying the significanc e of maintaining Mg2+ homeostasis in plants (Shaul, 2002). Sulfate (SO4 2-) is reduced by plants to sulfide and incorporated into cysteine. Cysteine is an inte gral part of proteins determining structure and function of proteins and is invol ved in redox reactions. Further, cysteine is converted to the nutritionally important amino acid methionine, as well as a wide range of sulfur-containing metabolites, predominant among them glutathi one (GSH) and S-adenosylmethionine (SAM) (Nikiforova et al., 2006). Movement of low-molecular weight solutes such as Mg2+ and SO4 2ions from the external solution into the cell wall conti nuum (apoplasm) of roots is a passi ve process driven by diffusion 24

PAGE 25

of ions or ion transport by mass flow of water. Carboxylic groups in the cell wall act as cation exchangers, while anions are repelled. The m ain ba rrier to solute flux in the apoplasm of roots is the hydrophobic wall of cells in th e endodermis, which prevents passive ion movement into the stele. Ions enter and exit the cytoplasm of cells by passive or active transport through the plasma membrane (Marschner, 1995). Ion transport syst ems can be subdivided into pumps, carriers (transporters/exchangers) and channels. Members of these groups can exhibit different levels of specificity for the ions they transport. Plants may have evolved to cope with relatively high levels of elements in the soil environment by limiting internal accumulation or tolerating high internal concentrations. In the case of hea vy metal tolerance for example, the following mechanisms are described; (1) binding to the ce ll wall, (2) restricted in flux through the plasma membrane, (3) active efflux, (4) compartmentalizati on in the vacuole, (5) chelation at the cell wall plasma membrane interface, and (6) chelation in the cytoplasm (Marschner, 1995). In a potential bioregenerativ e life support system on Mars, an excess of a particular element in the crews diet could affect the pres ence and availability of other required elements. This study therefore focuses on the possi bility of reducing accumulation of Mg2+ and SO4 2ions within the plant as a method to enhance plant tolerance to high le vels of magnesium sulfate in the growth medium. Various efforts have illust rated that this strate gy can indeed improve tolerance to certain elements. For example, a line of transgenic wheat plants expressing an antisense construct of the high affinity K+ transporter TaHKT2;1 showed reduced sodium uptake by roots and enhanced growth relative to unstres sed plants compared to a control line at high levels of NaCl in the growth medium (Lau rie et al., 2002). Similarl y, overexpression of the Arabidopsis SOS1 gene, which encodes a plasma membrane Na+/H+ antiporter responsible for Na+ efflux, limited Na+ accumulation and improved growth compared to control plants at high 25

PAGE 26

NaCl concentrations (Shi et al ., 2003). Arabidopsis is a model species in plant mo lecular biology research and its genome is fully sequenced. Plas ma membrane localized efflux transporters of Mg2+ and SO4 2ions have not been identified to date in the outer root cell layers of Arabidopsis or other plant species. Several plasma membrane localized proteins in Arabidopsis roots are known to be responsible for magnesium or sulfat e ion uptake. The objective of this study is to characterize the growth of select ed uptake transporter gene knoc kout lines compared to wildtype Arabidopsis under high levels of magnesium sulfate in both agar and soil medium to determine whether the mutant lines show enhanced tolerance in the form of a higher fresh weight biomass. A family of ten putative Mg2+ transport proteins (AtMRS2/At MGT) has been identified in Arabidopsis by several groups of researchers (Schock et al., 2000; Li et al., 2001). Reverse transcription polymerase chain reaction (RT-PCR) analysis of ten AtMRS2 family members showed that most members are expressed in multiple tissues, including the roots. AtMRS2-10 (= AtMGT1) functionally complemented a bacterial mutant lacking Mg2+ transport capability and AtMRS2-10-GFPexpressing plan ts showed fluorescence in th e periphery of root cells, suggesting a plasma membrane asso ciation (Li et al., 2001). The su lfate transporter gene family in Arabidopsis consists of 14 isoforms that show homology to one another. H+-sulfate cotransport has been determined for some of these (Hawkesford, 2003). Yoshimoto et al. (2002) found that sulfate transporters SULTR1;1 and SULTR 1;2 co-localize in the root hair, epidermal and cortical cells that are in c ontact with the soil solution. In two week old roots, SULTR1;1 is highly up-regulated under sulf ur-limiting conditions, while the constitutively expressed SULTR1;2 ensures sulfate uptake into plants un der both sulfur-replete and sulfur-deficient conditions (1500-50 M sulfate) (Yoshimoto et al., 2002). 26

PAGE 27

For our growth experiment we selected knockout mutants of the AtMRS2-10 and AtSULTR1;2 genes, which are indicated to enco de plasma membrane localized proteins in Arabidopsis roots that are able to import ma gnesium and sulfate ions respectively. A knockout mutant of the AtMRS2-10 (AtMGT1) gene was characterized based on the SALK_100361.41.30.x T-DNA insertion line (Alonso a nd Stepanova, 2003). We will refer to this homozygous T-DNA insertion knockout line as mrs2-10 The sel1-10 knockout mutant of the AtSultr1;2 gene that we selected was previo usly characterized by Mar uyama-Nakashita et al. (2003). When grown for 12 days on agar with optimal levels of su lfate (1.7 mM MgSO4), the uptake of sulfate by the sel1-10 mutant was shown to be approximately 20% of wildtype in both leaves and roots despite an observed up-regulat ion of SULTR1;1 gene expression (MaruyamaNakashita et al., 2003). In addition to these lines we selected the cax1-1 singleand cax1/cax3 double-knockout mutants of the CAX1 and CAX3 genes encoding vacuolar H+/Ca2+ transporters characterized by Cheng et al. (2003, 2005). Cax1-1 mutant plants grown on agar medium for 10 days were observed to be more tolerant of Mg2+ (10 mM and 25 mM MgCl2) and Mn2+ (1.5 mM MnCl2) stresses than wild-type plants, while be ing more tolerant of medium lacking Ca2+ (Cheng et al., 2003). In a separate study, a CAX1 knockout mutant was identified through a mutant screen on nutrient solutions reflecting low Ca:Mg ratios ch aracteristic of serpentine soils (Bradshaw, 2005). Cax1 mutants have significantly reduced levels of Mg in their leaves to the extent of 0.7 standard deviations ( ) below the mean reported for wild type (Bradshaw, 2005). Although the Mg content of cax1 mutant roots has not been analyzed so fa r, the reduced levels of Mg in leaves is in line with the objective of this study to identify Arabidopsis variants showing improved tolerance of high magnesium sulfate by at least partly reducing accumulation of Mg2+ or SO4 227

PAGE 28

ions within the plant. Sim ilar to what was observed for cax1-1 plants, double mutant cax1/cax3 plants displayed robust growth on ag ar medium containing 15 mM MgCl2, which caused growth defects in the control and cax3 lines. Under regular nutrient conditions, the cax1/cax3 plants grew more slowly and were small in comparison with the wildtype and single-mutant plants. The double mutants showed significant differences in concentration of multiple elements, including a reduction of Ca2+ and Mg2+ in shoot tissue relative to wildtype (Cheng et al., 2005). We selected the cax1-1 and cax1/cax3 lines to see if we could confirm th eir observed tolerance of high levels of Mg2+ when replacing the associated anion Clwith SO4 2-, and when using a different growth medium (soil) and a longer period of growth (4 weeks). Results Mrs2-10 T-DNA Insertion Line Characterization A homozygous T-DNA insertion mutant for At MRS2-10 was identified by PCR based on the SALK_100361.41.30.x line (Figure 2-1). RT-PCR an alysis showed that AtMRS2-10 mRNA was absent from homozygous mutant l eaves while it could be detected in col-0 leaves (Figure 22). -Tubulin mRNA, which was used as a constit utive control, was de tected both in the homozygous mutant and col-0 leaves (Figure 2-2). The homozygous knockout mutant was backcrossed to col-0 three times before self-fertilization yielded homozygous backcrossed mutants identified by PCR that produced seeds for the subsequent experiments. The knockout mutant line was named mrs2-10 In order to analyze its i onome (elemental profile), mrs2-10 seeds were sent to the Purdue Ionomics projec t (Figure 2-3). Leaves from two sets of six mrs210 plants each were analyzed for their content of elements in ppm, and results were compared to those of col-0 Leaves from the mrs2-10 sets showed no statistica lly significant (p < 0.05) differences in concentration of particular elements compared to leaves from the col-0 set, except for boron (B) in one of the sets (Tables 2-3, 2-4, Figures 2-4, 2-5). 28

PAGE 29

Biomass Comparisons of Wildtype and Mutant Lines Transporter gene knockout mutant lines were compared to their respective wildtype backgrounds in ten growth experim ents on the basis of whole plant or shoot fresh weight biomass levels, and of shoot chloro phyll levels (Tables 2-1, 2-2). St atistical analysis of the data per experiment with ANOVA first of all confirmed that increasing concentrations of dissolved MgSO4H2O in agar and soil significantly reduce wildtype Arabidopsis ( ws col-0 ) biomass (Table 2-5). Biomass and chlorophyll level comp arisons between mutant and wildtype lines at specific concentrations per experiment were also part of the statisti cal analysis with ANOVA (Table 2-6, Figures 2-8 to 2-17) Results from these comparisons indicate whether knocking out of transporter genes encoding proteins responsible for Mg2+ and SO4 2ion uptake can improve growth at the whole plan t level when plant roots are exposed to high levels of these ions in the growth media. Mrs2-10 plants did not show a marked difference in whole plant biomass compared to col-0 when grown on agar medi um, although at 4 mM MgSO4H2O, the slight advantage observed of 6.8% was found to be stat istically significant (F igures 2-6, 2-8, Table 26). On soil, mrs2-10 FW shoot biomass and leaf chlorophyll content were indistinguishable from that of wildtype (Figures 2-7, 2-9, 2-10, Table 2-6). Sel1-10 plants exhibited a significant reduction in whole plant biomass compared to ws on agar medium at 0, 4 and 12 mM MgSO4H2O (Figures 2-6, 2-11, Table 2-6). On soil, sel1-10 FW shoot biomass and leaf chlorophyll content were indisti nguishable from wildtype (Fi gures 2-7, 2-12, 2-13, Table 2-6). Cax1-1 plants showed a signi ficantly higher FW shoot biomass compared to col-0 when grown on soil at 80 and 100 mM MgSO4H2O (Table 2-6). The increase in cax1-1 shoot biomass over that of col-0 was 89% for 80 mM and 149% for 100 mM at 4 weeks (Figures 2-7, 2-14). The absolute fresh weight of the cax1-1 shoots was still low (20% ) compared to untreated col-0. Leaf chlorophyll content was also significantly higher in cax1-1 compared to col-0 at 80 and 100 mM 29

PAGE 30

MgSO4H2O (Figure 2-15, Table 2-6). Cax1/cax3 plants showed no significantly higher FW shoot biomass compared to col-0 when grown on soil at 80 and 100 mM MgSO4H2O (Table 26). The average increases of 26.8% and 33.2% observed in cax1/cax3 in contrast to col-0 were not found to be statistically significant (Figures 2-7, 2-16) Leaf chlorophyll content was significantly higher in cax1/cax3 compared to col-0 at 0, 80 and 100 mM MgSO4H2O (Figure 2-17, Table 2-6). Discussion The objective of this study was to characteri ze the growth of select ed uptake transporter gene knockout lines compared to wildtype Arab idopsis under high levels of magnesium sulfate in both agar and soil medium to determine whether the mutant lines show enhanced tolerance in the form of a higher fresh weight biomass. The results reveal that knockout mutant lines of the known genes in Arabidopsis encoding root plas ma membrane based uptake transporters of Mg2+ and SO4 2ions did not show a significan t increase in biomass compar ed to wildtype when grown on soil treated with high levels of MgSO4H2O in solution. Furthermor e, the tolerance of cax11 mutant plants to high levels of MgCl2 in agar medium was confirmed for soil medium treated with high levels of MgSO eated agar. 4H2O in solution, although the signifi cant differences in biomass after 4 weeks on MgSO4H2O treated soil do not appear to be as dramatic as the size differences observed by Cheng et al. ( 2003) after 10 days of growth on MgCl2 tr It is not known at present whether prot eins encoded by AtMRS2-10 are primarily responsible for uptake of magnesium ions from the soil solution, or to what extent non-annotated proteins or non-specific transport systems play a role in this process. Schock et al. (2000) speculate that the function of the AtMRS2 gene family may be the maintenance of metal ion homeostasis in different cellular compartments (i.e. over different cellular membrane systems). Overexpression of AtMRS2-10 (AtMGT1) in Nicotiana benthamiana led to increased 30

PAGE 31

accumulatio n of magnesium (Mg), manganese (Mn), and iron (Fe) per unit dry weight and per plant compared to wildtype plants (Deng et al ., 2006). We showed that the opposite approach of preventing AtMRS2-10 mRNA from being made di d not lead to a reduction in the accumulation of Mg or any other elements measured in the mrs2-10 mutant plants compared to wildtype under normal nutrient conditions. The observed small increase in mrs2-10 biomass relative to col-0 on agar at 4 mM MgSO4H2O might have been due to slight di fferences in average seed size and uncontrolled environmental va riation in this medium. A large difference in results between agar a nd soil medium was observed for the biomass comparisons of sel1-10 and ws This discrepancy might be attribut able to differential affinity of media components for sulfate or ot her nutrients. The agar growth medium we used consists of neutral agarose and negatively charged agarope ctin. Agaropectin is a complex polysaccharide that is sulfated to some degree. Adding sulfate to the agar medium might increase the percentage of agaropectin sulfate groups by replacing other groups such as pyruvate and methoxyl, thereby making sulfate less bioavailable. Highly sulfated forms of agaropectin (20 30% sulfate) have for example been found in Gelidium amansii (Qi et al., 2008). The soil used in these experiments contains vermiculite clay and sphagnum moss organic matter, both of which have net negative charges unrelated to sulfate that can interact with positive cations. In the contained environment of the soil trays, sulfate does not leach out and should generally be completely bioavailable when in solution. SULTR1;1 and SULTR1;2 are the two essentia l components of the root sulfate uptake system; double knockout plants lack th e ability to take up sulfate at low and optimal levels in the growth medium (Yoshimoto et al., 2007). Regard ing the results of the soil biomass comparisons it could be hypothesized that the concentrations of sulfate by itsel f are not high enough to affect 31

PAGE 32

growth of wildtype so much as to cause a reduction in biomass. The observed decline in biomass would then be caused solely by m agnesium, havi ng a dominant effect ove r this concentration range. In this scenario, a sulfate uptake mutant would be unable to show improved tolerance based on biomass measurements. Perhaps at higher concentrations of sulfat e and in the presence of cations less phytotoxic than Mg2+, the mutant line would have a higher biomass than wildtype. If the sulfate concentrations are indeed high enough to cause a partia l effect on biomass, alongside magnesium, it could be post ulated that the reduction in SO4 2uptake characteristic of the AtSULTR1;2 mutant might be too extreme for this concentra tion range to lead to improved growth compared to wildtype. In that case, mRNA down-regulation or post-translational regulation of SULTR1;2 might be more advantageous than a complete knockout. As mentioned in the results, cax1-1 showed a small, and cax1/cax3 showed no statistically significant improvement in growth compared to col-0 after 4 weeks on soil, while cax1-1 showed a large improvement in size relative to col-0 after 10 days on agar medium (Cheng et al., 2005). An explanation for the observed differences in results between soil and ag ar may be the reduced availability of Mg2+ in soil, as well as the greater envi ronmental variation in a soil medium, rather than the replacement of the Clanion by the SO4 2anion. Magnesium ions are held by the negatively charged clay and organi c matter, and so may be less highl y bioavailable than in agar. The differences in biomass between col-0 and cax1-1 or cax1/cax3 on soil with high levels of magnesium sulfate are likely to be more pronoun ced after an even l onger period of growth. Conclusions Arabidopsis lines carrying knockout T-DNA inse rtion mutations in genes encoding known Mg2+ and SO4 2uptake transporters localized to the plasma membrane in root cells do not mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype Arabidopsis plants. An Arabidopsis line carry ing a knockout mutation of the vacuolar CAX1 32

PAGE 33

gene ( cax1-1) showed a significant improveme nt in grow th on soil treated with high levels of MgSO4H2O in solution. A reduction in leaf magnesium content in cax1 mutants compared to wildtype (0.7 below normal) was reported in a prev ious study (Bradshaw, 2005). Although the Mg content of cax1 mutant roots has not been analyzed so fa r, the reduced levels of Mg in leaves indicate that Arabidopsis CAX1 knockout mutant s are in line with our objective of identifying Arabidopsis variants that show improved tolerance of high magnesi um sulfate by at least partly limiting accumulation of Mg2+ or SO4 2ions within the plant. Therefore, genes in crop species encoding transporter proteins similar in function to the protein encoded by CAX1 in Arabidopsis are proposed candidate genes for en hancing tolerance of crop plan t species to regolith high in soluble magnesium sulfate minerals used in an advanced life support system on Mars. The identification by Bradshaw (2005) of a CAX1 knockout mutant tolerant of low Ca:Mg ratios in solution, which are characteristic of serpentine soil so lutions, confirms the appropriateness of serpentine soils as partial anal ogue soils on Earth for regolith high in soluble magnesium sulfate minerals on Mars. Leaf Ca:Mg molar ratios of nonserpentine plant species are generally equal to that of the soil, while serp entine species maintain significantly higher leaf Ca:Mg than both their nonserpentine counterpart s and the soil (O'Dell et al., 2006). The authors conclude that elevated leaf Ca:Mg in the serpentine species was achieved by selective Ca2+ transport and/or Mg2+ exclusion operating at the root-to-shoot translocati on level, as root Ca and Mg concentrations did not differ between se rpentine and nonserpentine species. Genetic differentiation between populations of Arabidopsis lyrata growing on granitic or serpentinic soils was measured by using an Arabidopsis thaliana tiling array that has 2.85 million probes throughout the genome (Turner et al., 2008). Th e study found significant overrepresentation of genes involved in ion transport, and one gene in particular, calcium-exchanger 7 (CAX7), was 33

PAGE 34

presented as an excellent candidate gene for adap tation to low Ca:Mg ratios in A. lyrata It is currently not known which transpor ters or their regulators could be involved in the observed lower levels of Mg or higher Ca:Mg ratios in leaves from Arabidopsis cax1-1 and serpentine species respectively. Results from the anal ysis of genetic differentiation between A. lyrata populations include an overrepresen tation of genes encoding proteins involved in ion transport. Some of these might play a role in reducing Mg2+ accumulation within the plant. To identify genes in Arabidopsis thaliana with potential to enhance tolerance to high magnesium sulfate by limiting accumulation within the plant, such as genes encoding root plasma membrane localized import systems in a ddition to AtMRS2-10 and AtSULTR1;2, as well as genes encoding efflux systems or regulators of plasma membrane transporter activity, the early transcriptome responses of col-0 and cax1-1 Arabidopsis roots to high magnesium sulfate stress were documented (Chapt er 3). The Ca:Mg ratio used in the high magnesium sulfate treatment for the microarray transcriptome analys is experiment corresponded to ratios found in serpentine soil solutions. Materials and Methods Seed Lines and Growth Experiments An overview of the wildtype and knockout mutant Arabidopsis seed lines used in this study is given in Table 2-1. The table lists the na mes of the seed lines, th e genes that are knocked out in each line, the donor institution for each line, and the original seed line where appropriate. An overview of the mutant/wildtype growth ex periments is given in Table 2-2. For each experiment, the table lists the mutant/wildtype lin es that were compared, the growth medium, the MgSO4H2O concentrations, and the plant growth characteristic that was analyzed. 34

PAGE 35

Mrs2-10 T-DNA Insertion Line Characterization A hom ozygous knockout T-DNA insertion line for AtMRS2-10 (At1g80900) was identified by PCR and RT-PCR using the original SALK_100361.41.30.x line. For PCR, oligonucleotide primers MRS2-10_LP (5 -CAGGATCAAAGCATCGTTCTC-3) and MRS210_RP (5-TAGGAGCTCAGAAGACGCAAC-3) were designed using software available on the Salk Intitute website1. In addition, the T-DNA specific primer LBb1 (5GCGTGGACCGCTTGCTGCAACT-3) was include d as designed and recommended by the Salk Institute1. Combinations of the primers were used to identify plants for which the T-DNA insertion was present in both AtMRS2-10 alle les. Genomic DNA was extracted from T-DNA insertion mutant leaves using the Shorty2 method and from col-0 leaves using the DNeasy Plant Mini Kit (Qiagen). PCR was car ried out using JumpStart Ta q DNA polymerase (Sigma). PCR products were separated in agarose gels and st ained with SYBR Safe DN A gel stain (Invitrogen). A confirmed homozygous T-DNA insertion mutant was backcrossed to col-0 three times before allowing self-fertilization. Homozygous plants backcrossed 3x were identified by PCR and used for seed generation. For RT-PCR, RNA was extracted from leaves of col-0 and a homozygous TDNA insertion line for AtMRS2-10 (At80900) with the RNeasy Plant Mini Kit (Qiagen). Gene specific oligonucleotide pr imers MRS2-10_LP1 (5-AGGGTTACTTTGTCGGAGA-3) and MRS2-10_RP1 (5-TACACGGGGTTTTATCTTG-3) were designed based on Arabidosis genomic DNA sequence information (NCBI). Alpha-( )-Tubulin was used as a constitutive control with primers according to Yoshimoto (2002). RT-PCR was carried out using the OneStep RT-PCR kit (Qiagen). PCR products were separated in agarose gels and stained with SYBR Safe 1 http://signal.salk.edu/tdnaprimers.html 2 http://www.hos.ufl.edu/meteng/HansonW ebp a gecontents/NucleicAcidIsolation.html#Arabidopsis%20Genomic%20 DNA 35

PAGE 36

DNA gel stain (Invitrogen). An ionome analysis of the backcrossed homo zygous AtMRS2-10 knockout mutants (mrs2-10) was performed as part of the Purdue Ionomics project and results are available online through the Purdue Ionomics Information Management System3 (Baxter et al., 2007). Seed Production of Wildtype and Knockout Mutant Lines on Soil Med ium in Black Plastic Trays Variation in seed characteristics between th e different lines was minimized by growing plants from each line for seed production on soil medium under similar growth conditions. Seeds from the ws col-0 mrs2-10, sel1-10 cax1-1 or cax1/cax3 lines were evenly planted (5 seeds per tray well) on individual trays with soil medium (sphagnum peat moss, vermiculite and perlite) that was pre-wetted with nutrient solution. The nutrient solution consisted of 2.2 g/L Murashige and Skoog salts (Murashige and Skoog, 1962) and 0.5 g/L 2-(N-morpholino)ethanesulfonic acid (MES) buffer. For the cax1-1 and cax1/cax3 lines, the nutrient soluti on also contained 20-25 mM MgSO4H2O. The pH of the solutions was adjusted to 5.70 5.75 with KOH. The trays with soil were covered with plastic wrap to maintain sufficient humidity and placed in a cold room at 4oC for three days. After cold treatment th ey were moved to a growth room at 23oC where the seeds germinated, the plastic wrap was removed after 3-5 days, and the plants were grown maturity. The plants were watered twice a week fr om below, alternatively with tap water and the nutrient solution described above. They were furt hermore thinned to two plants per tray well after one week to maintain sufficient spaci ng, and pruned to reduce growth of secondary inflorescences. Seeds were harvested when siliques had turned yellow or light brown. until 3 http://www.ionomicshub.org/home/PiiMS 36

PAGE 37

Plant Growth Experiments on Agar Medium in Vertical P etri Dishes Seeds from the ws col-0 mrs2-10 and sel1-10 lines were sterilized by soaking them for 15 minutes in a 40% (v/v) bleach solution with a drop of Tween 20 (polyoxyethylenesorbitanmonolaureate) The seeds were then washed six times with sterilized (autoclaved) water. Each liter of agar medium was prepared by combining double distilled water, 2.2 g Murashige and Skoog salts, 1 mL 1000x Gamborg vitamins, 0.5 g MES buffer, 0.6 % (w/v) sucrose and 1.3 % (w/v) granulated agar. This f unctioned as the basic and control medium to which MgSO4H2O was added as 4, 8, 12, 16, 24 or 28 mM for the treatments. The pH of the medium was adjusted to 5.70 5.75 with KOH. The medium was then autoclaved at 121oC for 21 minutes and kept in th e autoclave at around 50-70oC until the medium was distributed over the Petri dishes. A single 10 cm x 10 cm plate contained approximately 50 ml of autoclaved medium supporting 10 sterilized seeds that were planted at regular intervals on a row 1.75 cm from the top edge of the plate in its vertical configuration. To c ontrol for environmental variation when comparing plant growth per concentration, 5 seeds from both a mutant line and its associated wildtype background lin e were planted on a single plate, alternatively on the left and right sides of the plates. Ten Petr i dishes per concentration were used per experiment. The plates were sealed with surgical tape (3M Micropore) and stratified for three days at 4oC before being moved to a growth chamber at 23oC. There they were placed at a 90o angle in racks holding ten plates each that were randomly placed on the growth benches and put perpendicular to the fluorescent light bulbs above them. The plants we re analyzed at day 13 and the experiment was repeated three times. Plant Growth Experiments on Soil Medium in Black Plastic Trays Seeds from the ws col-0 mrs2-10 sel1-10, cax1-1 and cax1/cax3 lines were sterilized as described above. To control for environmental variation when comparing plant growth per 37

PAGE 38

concentration, each tray was divided in 8 sections of 6 wells each, with either a mutant line or its associated wildtype background line planted in one section, in alternating fashion. This method resulted in 4 sections per seed line per tray. The seeds were evenly planted (5 seeds per tray well) on the soil medium (sphagnum peat moss, vermic ulite and perlite) that was wetted with nutrient solution. The basic and control nutrient solution consisted of 2.2 g/L Murashige and Skoog salts and 0.5 g/L MES buffer. For the treatments, MgSO4H2O was added to the basic solution at 20, 60, 80 or 100 mM concentration. The pH of the solutions was adjusted to 5.70 5.75 with KOH. The trays with soil were covered with plastic wrap to maintain sufficient humidity and placed in a cold room at 4oC for three days. After stratification, the trays were moved to a growth room at 23oC where they were randomly placed on the growth benches. There, the seeds germinated, the plastic wrap was removed after 3-5 days, and the su rviving plants were thinned to two plants per well (maximum of 12 plants per section) after one week to maintain sufficient spacing. The plants were watered twice a week from below, a lternatively with tap water and the appropriate nutrient solution. The plants were analyzed at week 4 and the experiment was repeated three times. Fresh Weight Biomass and Chlorophyll Mea surements and Statistical Analysis Fresh weight biomass of plants grown on agar medium was measured by grouping and weighing whole seedlings per plant line per Petri dish. The number of seeds germinated per seed line was recorded to calculate the average whol e plant FW biomass per seedling. Fresh weight biomass of plants grown on soil was measured by grouping and weighing whole shoots of plants per tray section. Leaf chlorophyll content of pl ants grown on soil medium was measured with a Minolta SPAD-502 meter. One leaf measurement per plant per tr ay well was taken. Biomass or chlorophyll levels of wildtype and knockout mutant Arabidop sis lines were evaluated per experiment for a total of ten e xperiments (Table 2-2) in a mixe d analysis of variance (ANOVA) 38

PAGE 39

39 model using SAS 9.2 and JMP Genomics 7 software Genotype and concentration were included in the ANOVA model as fixed effects, while batc h (replication set) was included as a random effect. Unbiased estimates of biomass or chlorop hyll levels (least-square means) were generated for each effect (genotype, concentration, or genotype*concentration) per experiment. The estimated levels of biomass or chlorophyll were then compared using a series of t-tests, generating estimated differences and correspondi ng p-values. Differences with associated pvalues < 0.05 were considered significant. The re sults for the comparisons of interest (biomass or chlorophyll levels between genotype s per concentration; biomass levels of wildtype between concentrations) were represented in gra phs and tables using Microsoft Excel 2007.

PAGE 40

Table 2-1. Overview of wildtype an d m utant Arabidopsis seed lines Seed line Knockout gene Origin Original seed line ws col-0 mrs2-10 AtMRS2-10; At1g80900 Salk Institute SALK_100361.41.30.x RIKEN sel1-10 AtSULTR1;2; At1g78000 (Dr. H. Takahashi) BCM cax1-1 AtCAX1; At2g38170 (Prof. K. D. Hirschi) AtCAX1; At2g38170 and BCM cax1/cax3 AtCAX3; At3g51860 (Prof. K. D. Hirschi) Table 2-2. Overview of mutant/wildtype growth experiments Exp. # Mutant/wildtype set Medium MgSO4H2O conc. Analysis 1 mrs2-10/col-0 Agar 0, 4, 12, 20, 28 mM whole plant FW biomass 2 mrs2-10/col-0 Soil 0, 20, 60, 100 mM shoot FW biomass 3 mrs2-10/col-0 Soil 0, 20, 60, 100 mM shoot chlorophyll 4 sel1-10/ws Agar 0, 4, 12, 20, 28 mM whole plant FW biomass 5 sel1-10/ws Soil 0, 20, 60, 100 mM shoot FW biomass 6 sel1-10/ws Soil 0, 20, 60, 100 mM shoot chlorophyll 7 cax1-1/col-0 Soil 0, 60, 80, 100 mM shoot FW biomass 8 cax1-1/col-0 Soil 0, 60, 80, 100 mM shoot chlorophyll 9 cax1/cax3/col-0 Soil 0, 80, 100 mM shoot FW biomass 10 cax1/cax3/col-0 Soil 0, 80, 100 mM shoot chlorophyll 40

PAGE 41

Table 2-3. Elem ental compos ition of first set of six mrs2-10 plants in ppm Element Pot Avg BG Avg % Diff Med % Diff StdDev p value Li7 39.73 37.57 0.01664 0.09724 0.2075 B11 144.4 160.4 -0.06841 0.07521 0.0434* Na23 1285 1202 0.0759 0.01471 0.2061 Mg25 12400 12570 -0.0058 0.05367 0.3588 P31 8937 8830 -0.0232 0.09511 0.4173 K39 38830 37980 -0.03327 0.01389 0.3679 Ca43 76030 75450 0.05812 0.09048 0.4455 Cr52 NaN NaN 0 0 Mn55 51.04 53.55 -0.09456 0.1339 0.3222 Fe56 NaN NaN 0 0 Co59 0.2769 0.2833 -0.04403 0.1274 0.366 Ni60 6.196 6.854 -0.01991 0.09567 0.1277 Cu65 7.195 7.278 0.02622 0.1341 0.4291 Zn66 134.7 141.3 0.05338 0.1084 0.2858 Ga71 NaN NaN 0 0 As75 0.5653 0.5997 0.0155 0.1981 0.3188 Se77 NaN NaN 0 0 Mo95 NaN NaN 0 0 Cd111 NaN NaN 0 0 In113 NaN NaN 0 0 Pb208 NaN NaN 0 0 Al27 NaN S34 12490 11660 0.04733 0.1147 0.122 S48 NaN NaN 0 0 Cl35 NaN NaN 0 0 Ca44 NaN NaN 0 0 Fe57 141.7 142.2 -0.003369 0.1382 0.4832 Se82 13.23 12.03 0.1389 0.2227 0.2129 Rb85 28.32 26.71 0.00047 0.1878 0.2397 Sr88 NaN Mo98 4.768 4.969 0.04532 0.2113 0.3982 Cd114 5.127 5.192 0.07996 0.2648 0.4674 41

PAGE 42

Table 2-4. Elem ental compos ition of second set of six mrs2-10 plants in ppm Element Pot Avg BG Avg % Diff Med % Diff StdDev p value Li7 35.29 37.57 -0.05328 0.2299 0.2963 B11 155.6 160.4 -0.00365 0.05099 0.2729 Na23 1246 1202 0.07348 0.1367 0.3261 Mg25 12500 12570 -0.00975 0.05994 0.4449 P31 8893 8830 -0.02845 0.08417 0.4487 K39 40520 37980 0.04384 0.1313 0.1514 Ca43 74980 75450 0.00623 0.07332 0.4528 Cr52 NaN NaN 0 0 Mn55 58.53 53.55 -0.05696 0.2206 0.2363 Fe56 NaN NaN 0 0 Co59 0.2761 0.2833 -0.06496 0.1774 0.3803 Ni60 6.147 6.854 -0.0671 0.06689 0.0997 Cu65 6.908 7.278 -0.05149 0.06632 0.1228 Zn66 127.7 141.3 -0.07396 0.03678 0.0956 Ga71 NaN NaN 0 0 As75 0.6195 0.5997 0.1045 0.1882 0.3905 Se77 NaN NaN 0 0 Mo95 NaN NaN 0 0 Cd111 NaN NaN 0 0 In113 NaN NaN 0 0 Pb208 NaN NaN 0 0 Al27 NaN S34 12500 11660 0.02702 0.09718 0.0974 S48 NaN NaN 0 0 Cl35 NaN NaN 0 0 Ca44 NaN NaN 0 0 Fe57 142.5 142.2 -0.02592 0.08584 0.4873 Se82 11.29 12.03 0.0371 0.1281 0.275 Rb85 27.05 26.71 0.03622 0.06383 0.3792 Sr88 NaN Mo98 5.639 4.969 0.2068 0.2646 0.2169 Cd114 4.353 5.192 -0.1131 0.2038 0.1308 42

PAGE 43

Table 2-5. ANOVA (t-test) results for specific genotype*concentration e ff ects in the growth experiments (Table 2-2). The table lists th e results of differences in biomass of wildtype Arabidopsis between concentrations of MgSO4H2O. Conc. Genotype _Conc. _Genotype Diff. Estimate StdErr DF tValue P-value #4; ws whole plant FW biomass comparison on agar medium 0 ws 4 ws 1.4948 0.5912 288 2.53 0.012* 0 ws 12 ws 4.4467 0.5912 288 7.52 <.0001* 0 ws 20 ws 11.0023 0.5912 288 18.61 <.0001* 0 ws 28 ws 14.2145 0.5912 288 24.04 <.0001* 4 ws 12 ws 2.9518 0.5912 288 4.99 <.0001* 4 ws 20 ws 9.5075 0.5912 288 16.08 <.0001* 4 ws 28 ws 12.7197 0.5912 288 21.52 <.0001* 12 ws 20 ws 6.5557 0.5912 288 11.09 <.0001* 12 ws 28 ws 9.7678 0.5912 288 16.52 <.0001* 20 ws 28 ws 3.2122 0.5912 288 5.43 <.0001* #5; ws shoot FW biomass comparison on soil medium 0 ws 20 ws 2.7275 0.3531 86 7.73 <.0001* 0 ws 60 ws 5.145 0.3531 86 14.57 <.0001* 0 ws 100 ws 7.1923 0.3531 86 20.37 <.0001* 20 ws 60 ws 2.4175 0.3531 86 6.85 <.0001* 20 ws 100 ws 4.4648 0.3531 86 12.65 <.0001* 60 ws 100 ws 2.04 73 0.3531 86 5.8 <.0001* #1; col-0 whole plant FW biomass comparison on agar medium 0 col-0 4 col-0 0.51 87 0.59 15 288 0.88 0.3813 0 col-0 12 col-0 10.676 0.5915 288 18.05 <.0001* 0 col-0 20 col-0 19.8547 0.5915 288 33.57 <.0001* 0 col-0 28 col-0 21.342 0.5915 288 36.08 <.0001* 4 col-0 12 col-0 10.1573 0.5915 288 17.17 <.0001* 4 col-0 20 col-0 19.336 0.5915 288 32.69 <.0001* 4 col-0 28 col-0 20.8233 0.5915 288 35.21 <.0001* 12 col-0 20 col-0 9.1787 0.5915 288 15.52 <.0001* 12 col-0 28 col-0 10.666 0.5915 288 18.03 <.0001* 20 col-0 28 col-0 1.4873 0.5915 288 2.51 0.0125* #2 and #7; col-0 shoot FW biomass comparison on soil medium 0 col-0 20 col-0 3.4808 0.6337 86 5.49 <.0001* 0 col-0 60 col-0 7.8975 0.6337 86 12.46 <.0001* 0 col-0 80 col-0 8.302 0.4671 86 17.77 <.0001* 0 col-0 100 col-0 10.0 38 0.63 37 86 15.84 <.0001* 20 col-0 60 col-0 4.4167 0.6337 86 6.97 <.0001* 20 col-0 100 col-0 6.5572 0.6337 86 10.35 <.0001* 60 col-0 80 col-0 3.02 03 0.46 71 86 6.47 <.0001* 60 col-0 100 col-0 2.1405 0.6337 86 3.38 0.0011* 80 col-0 100 col-0 0.2838 0.4671 86 0.61 0.545 43

PAGE 44

Table 2-6. ANOVA (t-test) results for specific ge notype*concentration effect s in the ten growth experim ents (Table 2-2). The table lists th e results of differences in biomass or chlorophyll between genotypes at the same concentration of MgSO4H2O. Conc. Genotype _Conc. _Genotype Diff. Estimate StdErr DF tValue P-value #1; mrs2-10 / col-0 whole plant FW biomass comparison on agar medium 0 col-0 0 mrs2-10 0.516 0.5915 288 0.87 0.3837 4 col-0 4 mrs2-10 -1.4947 0.5915 288 -2.53 0.012* 12 col-0 12 mrs2-10 -0.41 0.5915 288 -0.69 0.4887 20 col-0 20 mrs2-10 -0.3728 0.5915 288 -0.63 0.529 28 col-0 28 mrs2-10 -0.2494 0.5915 288 -0.42 0.6735 #2; mrs2-10 / col-0 shoot FW biomass comparison on soil medium 0 col-0 0 mrs2-10 0.6458 0.6337 86 1.02 0.311 20 col-0 20 mrs2-10 0.07167 0.6337 86 0.11 0.9102 60 col-0 60 mrs2-10 -0.1983 0.6337 86 -0.31 0.755 100 col-0 100 mrs2-10 0.07142 0.6337 86 0.11 0.9105 #3; mrs2-10 / col-0 shoot chlorophyll comparison on soil medium 0 col-0 0 mrs2-10 -0.4111 0.8282 560 -0.5 0.6198 20 col-0 20 mrs2-10 -0.01528 0.8282 560 -0.02 0.9853 60 col-0 60 mrs2-10 1.0528 0.8282 560 1.27 0.2042 100 col-0 100 mrs2-10 0.7072 0.846 560 0.84 0.4035 #4; sel1-10 / ws whole plant FW biomass comparison on agar medium 0 sel1-10 0 ws -5.4527 0.5912 288 -9.22 <.0001* 4 sel1-10 4 ws -3.9427 0.59 12 288 -6.67 <.0001* 12 sel1-10 12 ws -2.0702 0.5912 288 -3.5 0.0005* 20 sel1-10 20 ws -0 .7 1 7 0.5912 288 -1.21 0.2262 28 sel1-10 28 ws -0.04433 0.5912 288 -0.07 0.9403 #5; sel1-10 / ws shoot FW biomass comparison on soil medium 0 sel1-10 0 ws -0.2667 0.3531 86 -0.76 0.4521 20 sel1-10 20 ws -0.1933 0.3531 86 -0.55 0.5854 60 sel1-10 60 ws -0.1225 0.3531 86 -0.35 0.7295 100 sel1-10 100 ws -0.04517 0.3531 86 -0.13 0.8985 #6; sel1-10 / ws shoot chlorophyll comparison on soil medium 0 sel1-10 0 ws 0.3208 0.685 532 0.47 0.6397 20 sel1-10 20 ws 0.3931 0.685 532 0.57 0.5663 60 sel1-10 60 ws 0.7944 0.685 532 1.16 0.2466 100 sel1-10 100 ws -0.8181 0.7837 532 -1.04 0.297 #7; cax1-1 / col-0 shoot FW biomass comparison on soil medium 0 cax1-1 0 col-0 -0.7908 0.4671 86 -1.69 0.0941 60 cax1-1 60 col-0 0.4725 0.4671 86 1.01 0.3146 80 cax1-1 80 col-0 0.9305 0.4671 86 1.99 0.0495* 100 cax1-1 100 col-0 1. 14 2 0.4671 86 2.44 0.0165* #8; cax1-1 / col-0 shoot chlorophyll comparison on soil medium 0 cax1-1 0 col-0 -0.2847 1.0107 566 -0.28 0.7783 60 cax1-1 60 col-0 1.8972 1.0107 566 1.88 0.061 80 cax1-1 80 col-0 5. 79 3 1 1.0107 566 5.73 <.0001* 100 cax1-1 100 col-0 7.6306 1.0107 566 7.55 <.0001* #9; cax1/cax3 / col-0 shoot FW biomass comparison on soil medium 0 cax1-1 0 col-0 -6.5781 0.3312 64 -19.86 <.0001* 80 cax1-1 80 col-0 0.3279 0.3312 64 0.99 0.3259 100 cax1-1 100 col-0 0.3345 0.3312 64 1.01 0.3164 #10; cax1/cax3 / col-0 shoot chlorophyll comparison on soil medium 0 cax1-1 0 col-0 8.7083 0.8883 414 9.8 <.0001* 80 cax1-1 80 col-0 6.2347 0.8883 414 7.02 <.0001* 100 cax1-1 100 col-0 8.2771 0.9219 414 8.98 <.0001* 44

PAGE 45

Figure 2-1. Mrs2-10 homozygous knockout mutant line. A homozygous T-DNA insertion mutant for AtMRS2-10 (At1g80900) was identified by PCR based on the SALK_100361.41.30.x line. Figure 2-2. RT-PCR analysis of the mrs2-10 mutant. RNA extracted from mrs2-10 and col-0 leaves was tested for the presence of MRS2-10 mRNA with gene-specific primers. Tubulin (TUB) was used as a constitutive control. Figure 2-3. Two sets of six mrs2-10 plants growing at the Pur due Ionomics project facility. 45

PAGE 46

Figure 2-4. Element z-score values for the first set of six mrs2-10 plants. Each individual plant is represented by a different color. The z scor e for an element indicates how far and in what direction the element deviates from its distribution's mean, expressed in units of its distribution's standard deviation. Figure 2-5. Element z-score values for the second set of six mrs2-10 plants. Each individual plant is represented by a different color. Th e z score for an element indicates how far and in what direction the element deviates from its distribution's mean, expressed in units of its distribution's standard deviation. 46

PAGE 47

Figure 2.6. Growth experiments on agar. FW biomass of mutant lines is compar ed to that of their respective wildtype backgrounds at different levels of MgSO4H2O in the agar medium. Two Petri dishes per comparison per concentration ar e shown. Each dish contains both mutant and wildtype plants, which are planted alte rnatively on the left and the right side. The Petri dishes were sca nned just prior to plan t analysis at day 13. 47

PAGE 48

Figure 2-7. Growth experiments on soil. FW shoot biom ass of mutant lines is compared to that of their respective wildt ype backgrounds at different levels of MgSO4H2O in the soil medium. One flat per comparison per con centration is shown. For the comparison between cax1/cax3 and col-0 60 mM was not tested. Yellow boxes highlight the sections with mutant plants within each flat; the othe r sections contain wildtype plants. The plants shown here are 3 weeks old and were analyzed after 4 weeks of growth. 48

PAGE 49

Figure 2-8. Average fresh weight biomass of mrs2-10 and col-0 seedlings in response to increasing concentrations of MgSO4H2O in agar medium. Bars indicate standard error, n = 30. The asterisk indicates a statistically significant difference between mrs2-10 and col-0 (p < 0.05) at 4 mM MgSO4H2O based on ANOVA. Figure 2-9. Average fresh we ight shoot biomass of mrs2-10 and col-0 plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indicate standard error, n = 12. No statistically significant differences between mrs2-10 and col-0 at any of the concentrations are observed based on ANOVA. 49

PAGE 50

Figure 2-10. Average leaf chlorophyll content of mrs2-10 and col-0 plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indicate standard error, n = 72. No statistically significant differences between mrs2-10 and col-0 at any of the concentrations are observed based on ANOVA. Figure 2-11. Average fresh weight biomass of sel1-10 and ws seedlings in response to increasing concentrations of MgSO4H2O in agar medium. Bars indicate standard error, n = 30. Asterisks indicate statistically significant differences between sel1-10 and ws (p < 0.05) at 0, 4 and 12 mM MgSO4H2O based on ANOVA. 50

PAGE 51

Figure 2-12. Average fresh weight shoot biomass of sel1-10 and ws plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indicate standard error, n = 12. No statistically significant differences between sel1-10 and ws at any of the concentrations are observed based on ANOVA. Figure 2-13. Average leaf chlorophyll content of sel1-10 and ws plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indi cate standard error, n = 72. No statistically significant differences between sel1-10 and ws at any of the concentrations are observed based on ANOVA. 51

PAGE 52

Figure 2-14. Average fresh weight shoot biomass of cax1-1 and col-0 plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indicate standard error, n = 12. Asterisks indicate statistically signi ficant differences between cax1-1 and col-0 (p < 0.05) at 80 and 100 mM MgSO4H2O based on ANOVA. Figure 2-15. Average leaf chlorophyll content of cax1-1 and col-0 plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indicate standard error, n = 72. Asterisks indicate statistically signi ficant differences between cax1-1 and col-0 (p < 0.05) at 80 and 100 mM MgSO4H2O based on ANOVA. 52

PAGE 53

53 Figure 2-16. Average fresh weight shoot biomass of cax1/cax3 and col-0 plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indicate standard error, n = 12. The Asterisk indicates a statistically significant difference between cax1/cax3 and col-0 (p < 0.05) at 0 mM MgSO4H2O based on ANOVA. Figure 2-17. Average leaf chlorophyll content of cax1/cax3 and col-0 plants in response to increasing concentrations of MgSO4H2O in soil medium. Bars indicate standard error, n = 72. Asterisks indicate statistically signi ficant differences between cax1/cax3 and col-0 (p < 0.05) at 0, 80 and 100 mM MgSO4H2O based on ANOVA.

PAGE 54

CHAPTER 3 TRANSCRIPTOME RE SPONSES OF COL-0 AND CAX1-1 ARABIDOPSIS TO EXCESS LEVELS OF MAGNESIUM SULFATE; CONSEQUENCES FOR (EXTRA)TERRESTRIAL PLANT GROWTH Introduction Manned missions to Mars demand the efficient use of local planetary resources and the recycling of limited materials such as water, pressurized atmosphere and organic matter while producing food (Barta and Henninger, 1994). The use of in situ regolith for plant growth in a future bioregenerative life support system on Ma rs may have several advantages over hydroponic systems (Schuerger et al., 2002). These include th e immediate bioavailabili ty of plant essential ions, low-tech mechanical support for plants, an d easy access of in situ materials once on the surface. However, plant growth may be reduced or inhibited by substances in the regolith, such as high levels of hydrated magnesium sulfate minera ls (Chapter 2). In a po tential bioregenerative life support system on Mars, an exce ss of a particular element in th e crews diet co uld affect the presence and availability of other required elements. This study therefore focuses on the possibility of reducing accumulation of Mg2+ and SO4 2ions within the plant as a method to enhance plant tolerance to high levels of magnesium sulfate in the growth medium. Arabidopsis is a model species in plant molecular biology research and its genome is fully sequenced. Plasma membrane loca lized efflux transporters of Mg2+ and SO4 2ions have not been identified to date in the outer root cell layers of Arabidopsis or other plant species. AtMRS2-10 and AtSULTR1;2 are genes encoding a known Mg2+ and SO4 2uptake transporter respectively. The transporters are localized to the plasma membrane in root cells. Ar abidopsis lines carrying knockout T-DNA insertion mutations in AtMRS2 -10 and AtSULTR1;2 did not mitigate the constraining impacts of high magnesium sulfate concentrations on wildty pe Arabidopsis plants (Chapter 2). An Arabidopsis line carrying a knockout mutation of the vacuolar CAX1 gene 54

PAGE 55

( cax1-1 ) showed a significant improvement in grow th on soil treated with high levels of MgSO4H2O in solution (Chapter 2). Although the Mg content of cax1 mutant roots has not been analyzed so far, the reduced levels of Mg in leaves (Brads haw, 2005) indicate that Arabidopsis CAX1 knockout mutants show improve d tolerance of high magnesium sulfate by at least partly limiting accumulation of Mg2+ or SO4 2ions within the plant. Genome-wide analysis of Arabidopsis root tran scriptome responses to elevated levels of MgSO4H2O in solution could reveal what molecula r processes may be involved in adaptation to this stress, and could lead to identification of candidate genes that may play a role in enhancing tolerance by reducing accumulation with in the plant. No genome-wide transcriptome profiling studies have been reported to date that address responses of plant species to elevated concentrations of Mg2+ or SO4 2in the growth medium. Maruyama-Nakashita et al. (2003) analyzed the transcriptome response of ws and sel1-10 plants to sulfur deficiency stress. The study provided a selection of -S-responsive genes related to sulfate uptake, sulfur assimilation, remobilization of secondary sulfur metabolites, and mitigation of oxidative stress (MaruyamaNakashita et al., 2003). Some of the previous studies analyzing root transcriptome responses to ion toxicities, such as responses to NaCl and heavy metals, are reviewed below. The focus for this review is on the extent of the observed transcriptome responses, as we ll as the detection of differentially expressed transporter genes, in re lation to the treatment c oncentration and duration. In addition, transcriptome analyses are reviewed that involve co mparisons between wildtype and transgenic plants, or between related species showing differences in tole rance to a certain ion stress. Maathuis et al. (2003) analyzed the Arabidopsis ( col-0 ) root transcriptome in response to high NaCl (80 mM) stress over a 96-h period (2, 5, 10, 24 and 96 hrs) with oligonucleotide 55

PAGE 56

microarrays representing 1096 Arabi dopsis transporter genes. In ge neral, NaCl treatment did not affect specific gene families but rather resulted in a collective transcriptional response by affecting specific isoforms across transporte r families. For example, aside from known Na+ transporters such as the vacuolar Na+/H+ antiporter AtNHX1, NaCl also affected the expression of particular genes associated with transport of other nutrients such as nitrate (Maathuis et al., 2003). Jiang et al. (2006) used microarrays with oligonucleotide pr obes representing 23,686 Arabidopsis genes to identify col-0 root transcripts that change d in relative abundance following 6 h, 24 h, or 48 h of hydroponic exposure to 150 mM NaCl compared to control conditions. Among the statistically significant expression di fferences, 2,433 unique genes showed a NaClinduced increase in transcript a bundance of at least 2.0 fold at one or more time points, while 2,774 unique genes showed a decrease in transcript abundance of at least 2.0 fold or more (Jiang and Deyholos, 2006). The number of regulated genes found in this study highlights the complexity of the response in plants exposed to 150 mM NaCl for these periods of time. Taji et al. (2004) analyzed the gene e xpression profiles in Arabidopsis compared to its halophytic relative Thellungiella halophila (salt cress) by using full-leng th Arabidopsis cDNA microarray containing approximately 7000 genes. In contrast to Arabidopsis, only a few genes were induced by treatment with 250 mM NaCl for 2 hrs in sa lt cress. Various genes induced by abiotic and biotic stresses in Arabidopsis were shown to be constitutively expressed in salt cress even under normal growth conditions (Taji et al., 2004). In another study addressi ng genotypic differences, Sottosanto et al. (2004) used A ffymetrix ATH1 arrays with pr obes representing 22,746 genes to compare transcription profiles between a T-DNA insertion mutant of AtNHX1 ( nhx1 ) and a 'rescued' line (NHX1:: nhx1 ), which were both exposed to 10 0 mM NaCl for 12 and 48 hrs, as well as 1 and 2 wks. 147 transcripts showed bot h salt responsiveness and a significant influence 56

PAGE 57

of AtNHX1, including signaling proteins, DNA bi nding proteins, compon ents of the protein processing and trafficking machinery, cell wall re lated proteins and thos e involved in sulfur metabolism (Sottosanto et al., 2004). Several studies analyzing plant transcriptome re sponses to toxic levels of metal ions have been reported. Herbette et al. (2006) investigated the transcrip tional regulation in response to cadmium treatment (5 and 50 M for 2, 6 and 30 hrs) in both roots and leaves of hydroponically grown Arabidopsis (ecotype Columbia), us ing a whole genome microarray containing 24,576 independent probe sets. The number of differentially expressed genes was lower in leaves than in roots, and regulation of gene e xpression in response to Cd seem s time-regulated rather than doseregulated. One of the main responses observed in roots was the induction of genes involved in sulfur assimilationreduction and glutathione (GSH) metabolism (Herbe tte et al., 2006). In another study, van de Mortel et al. (2006) examined the root transcript profiles of Arabidopsis and its heavy metal accumulating relative Thlaspi caerulescens grown for 7 days under deficient (0 M), sufficient (2 or 100 M), and excess (25 or 1000 M) supply of zinc (ZnSO4), with the main aim of establishing which genes are most lik ely to be relevant for adaptation to high zinc exposure in T. caerulescens Results emphasized the role of previously implicated zinc homeostasis genes in adaptation to high zinc exposure, but also suggest a similar role for many more, as yet uncharacterized genes, often wit hout any known function (van de Mortel et al., 2006). This study is the first to document genome-w ide plant root transcriptome responses to elevated levels of magnesium sulfate using microarrays. The objective is to analyze which genes are differentially expressed as part of the prim ary stress response in roots of a non-tolerant species such as wildtype Arabidopsis ( col-0 ) compared to unexposed col-0 roots. This could lead 57

PAGE 58

to identification of candidate ge nes in Arabidopsis with potentia l to enhance tolerance to high magnesium sulfate by lim iting accumulation within th e plant. Potential candidate genes are those encoding root plasma membrane localized import or efflux systems, or regulators of plasma membrane transporter activity. In order to capture part of the primary stress response, three early time periods (45 min. 90 min. and 3 hrs) of col-0 exposure to a non-lethal concentration of high MgSO4H2O in a hydroponic growth medium we re chosen. In addition, a set of col-0 plants was exposed to a control solution for 45 mi nutes. The treatment c oncentration of MgSO4H2O was based on the high Mg:Ca ratio that can occu r in serpentine soils on Earth. Serpentine soils can be a partial analogue for regolith high in ma gnesium sulfate on Mars because of their high amount of bioavailable magnesium (Chapter 2). The cax1-1 mutant was also exposed to el MgSO evated 4H2O for 3 hours to determine which genes ar e differentially expressed in the CAX1 knockout mutant background compared to exposed col-0 at this time. Genes that are differentially expressed between the genotypes could point to some of the downstream molecular processes eventually leading to enhanced tolerance for cax1-1 at the whole plant level, including reduced leaf Mg content and increased fresh weight biomass, after days or weeks of exposure in agar or soil medium (Chapter 2). Some of the genes involved in downstream processes may themselves be candidate genes for enhanced tole rance, such as those encoding (regulators of) plasma membrane based channels that transport Mg2+. Results A microarray experiment was conducted with conditions described in Table 3-1, Figures 31, 3-2, 3-3, and the Materials and Methods section. Sample sets were compared using a series of t-tests; the col-0 time 1, col-0 time 2 and col-0 time 3 treatment sets were compared to the col-0 time 1 control sample set, and these comparisons ar e referred to in the text, tables and figures as Time 1, Time 2 and Time 3 respectively. The cax1-1 time 3 treatment set was compared to the 58

PAGE 59

col-0 time 3 treatm ent set; this comparison is not re ferred to in abbreviated form. The resulting gene expression data for each comparison are pres ented as volcano plots (Figures 3-4, 3-5, 3-6, 3-7). The volcano plots show the log2 values of the fold changes in gene expression between two compared sample sets on the x-axes. The y-axes show the log10 p-valu es corresponding to the log2 fold change values. Expre ssion differences (fold change) corresponding to p-values for which q < 0.05 are considered significant and are indicated in red. The number of genes with a significant difference in expressi on between treatment and control sets increases from 325 for Time 1, to 1516 for Time 2, and 3265 for Time 3. The number of genes with a significant difference in expression over 2 fold increases acc ordingly, from 100 for Time 1, to 248 for Time 2, and 445 for Time 3. Between the cax1-1 time 3 treatment and col-0 time 3 treatment sample sets, only 4 transcripts show a significant difference in expressi on, all over 2 fold. Genes with significant expression differences at Time 1, 2 an d 3 were annotated according to Gene Ontology (GO) terminology. The genes were grouped per GO functional category for each comparison, and the resulting numbers for the three comparisons are listed in Table 3-2. The results show that the number of genes per functional category is increasing with time for each category. The percentage of genes in each functional category changes sligh tly across the Time 1, 2 and 3 comparisons; increasing in some categories and decreasing in others (Table 3-2). Transcripts col-0 Time Series Since the experimental design for the microarray experiment involving the col-0 time series does not control for diurnally regulated ge nes at Time 2 and 3, we will focus on genes with significant expression differences at Time 1, and genes with significant expression differences that are shared between Time 1 and one or more of the other comparisons (Time 2 and 3). As mentioned above, 325 genes show a significant diffe rence in expression at Time 1. Of these 325 genes, 34 are differentially expressed at q < 0.001 (Table 3-3), while 36 are differentially 59

PAGE 60

expressed over 3 fold (Table 3-4). A Venn diagram was m ade based on a comparison of genes with significant expression differences at Time 1, 2 and 3. The results show that of the 325 genes expressed differentially at Time 1, 74 genes are unique to Time 1, 48 are shared between 1 and 2, 155 genes are shared between all three comparis ons, and 48 are shared between 1 and 3 (Figure 3-8). The 325 differentially expressed genes at Ti me 1 were also compared with a cluster of 197 genes identified by Ma and Bohnert (2007) that ar e differentially expressed in response to broad range of diverse stress conditions: cold, os motic, salinity, wounding, and biotic stresses (including treatments with elicito rs). The comparison showed that at least 18 of the 325 genes differentially expressed at Time 1 are universally responsive to stress conditions. A hierarchical average linkage cluster analysis using uncente red correlation was done across Time 1, 2 and 3 based on the genes with significan t expression differences at Time 1 (Figure 3-9). The results show the exact expression pattern s of the genes that are shared with Time 2, Time 3, or both. Groups of functionally related genes The 325 genes that showed a statistically significant difference in expression at Time 1 can be grouped into functional categories according to GO annotations (Table 3-2). The overview shows that the molecular function of a large por tion of the genes is currently unknown. A closer look at the genes whose molecu lar function is known or putat ively known reveals the following (Figure 3-9a-d, Object 3-2). Within the group ge nes encoding transcription factors, multiple AP2/EREBF, bHLH, heat shock, MYB (KAN4, MYB50), NAC (NAM) TCP (PCF1, PCF2) and zinc finger (C2H2 and C3HC4 type) transcrip tion factors, as well as a MADS box (AGL45), WRKY (WRKY70), and scarecrow transcription factor can be recognized. Among the genes encoding proteins with kinase activity, several protein kinases, lectin protein kinases, and leucine-rich repeat transmembr ane protein kinases can be dis tinguished. Three genes encoding cyclins, which regulate the activity of cyclin-dep endent kinases, are differentially expressed at 60

PAGE 61

Time 1 and 2, but not at Tim e 3. Regardi ng phosphatase activity, ge nes encoding protein phosphatase 2C (PP2C) family proteins are well represented. Within the group of genes encoding membrane based transporters, those encoding cyclic nucleotide-gate d channels CNGC19 and CNGC1 are differentially expre ssed at the early time points. The expression of two genes encoding potassium transporters is up-regul ated, with the expression of HAK5 being upregulated across the time points. Transcripts encoding calcium-transporting ATP-ase ACA12 and calcium exchanger CAX1 are differentially expressed at Time 1 and beyond. Other genes encoding transporters that show differential e xpression are amino-acid transporters, an organic cation transporter, and inorga nic phosphate transporter PHT1. Among transcripts encoding proteins with ion binding activ ity, those encoding calcium-bindi ng proteins, and in particular calmodulins, are well represented at Time 1 a nd further on. The expression of genes encoding calcium-binding copines and coppe r-binding plastocyanins is down -regulated. A few transcripts encoding proteins responsive to inorganic ions such as phospha te and nitrate can also be distinguished. Genes encoding hormone-responsive proteins are dominated by those encoding auxin-responsive proteins. The expression of these genes is up-regulated at Time 1, and for many of the genes this is the same at Time 2 and 3. A large group of disease resistance proteins (TIRNBS-LRR or variations thereof) shows that the expression of the genes encoding these is generally down-regulated across the time points. Genes encoding cell wall re lated proteins such as (fasciclin-like) arab inogalactans, xyloglucan endotransglycos ylases, exotosins, putative lipid transfer proteins, beta-expansins, a pectin acet ylesterase and a hydroxyproline-rich glycoprotein all show down-regulated expression, while multiple genes encoding alpha-expansins and polygalacturonases/pectinases all show up-re gulated expression. The expression of two transcripts encoding anionic peroxidases and of one that encodes a glutat hione S-transferase is 61

PAGE 62

up-regulated, while that of two transcripts encoding SEC14/phosphoglyceride transfer family proteins is down-regulated. Other dif ferentially expressed genes encoding metabolic enzymes include those that encode gl ycoside and glycosyl hydrolases glycosyl transferases. UDPglucosyl transferases, beta-a mylases and nudix hydrolases. Remaining genes that encode metabolic enzyme s that do not directly group together were analyzed by overlaying the significant gene expre ssion ratios per time point on a metabolic map by using the Aracyc Omics viewer of the Arabidosis Information Center (TAIR). Genes encoding 1-aminocyclopropane-1-carboxylate synt hase in the ethylene biosynthesis pathway show up-regulated expression acro ss the three time points (Figure 3-10a). A transcript encoding a 9-cis-epoxycarotenoid dioxygena se/neoxanthin cleavage enzyme in the abscisic biosynthesis pathway shows up-regulated expression in all comparisons as well (Figure 3-10b). In the jasmonate biosynthesis pathway, the expression of a lipoxygenase encoding gene is up-regulated at Time 1, and several genes encoding 12-oxophyt odienoate reductases show down-regulated expression at Time 3 (Figure 3-10c). Transcripts encoding gibberellin 2-oxidases 7 and 8 show up-regulated expression at the time points, while a transcript encoding gi bberellin 2-oxidase 1 shows down-regulated expression at Ti me 3 (Figure 3-10d). A gene encoding monogalactosyldiacylglycerol synthase 2 s hows down-regulated expr ession across the time points as part of glycerolipid metabolism (Figur e 3-10e). A transcript encoding the rate-limiting enzyme glucosamine-fructose-6-phosphate aminot ransferase shows up-regulated expression at Time 1 and 3 in the hexosamine biosynthetic pa thway (Figure 3-10f). A gene encoding a Rieske [2Fe-2S] domain-containing protein similar to Pheophorbide A oxygenase shows up-regulated expression at Time 1 and Time 3. Pheophorbide A oxygenase is part of the chlorophyll breakdown pathway (Figure 3-10g). A Glutamyl -tRNA reductase 2 encoding gene shows up62

PAGE 63

regulated expression at Time 1 and 2, and a sirohydrochlorin ferro chelatase encoding gene shows up-regulated expression at Tim e 2 and 3, as part of porphyrin and chlorophyll metabolism (Figure 3-10h). A transcript encoding phosphoribosyl-ATP pyrophos phohydrolase shows upregulated expression across the time points as part of the histidine biosynthesis pathway (Figure 3-10i). A gene encoding UDP-glucose 4-epimeras e shows down-regulated expression at Time 1 and 2 as part of galactose metabolism (Figur e 3-10j), and phosphofructokinase encoding genes show up-regulated expression at Time 1 and 3 as part of the glycolysis pathway (Figure 3-10k). A transcript encoding cytosolic malate dehydrogenase shows up-regulated expression at Time 1, while a transcript encoding mitochondrial malate dehydrogenase shows down-regulated expression at Time 3, and a tran script encoding malate synthase shows up-regulated expression at Time 2 and 3 (Figure 3-10l). Finally, genes encoding protei ns grouped by their structural motifs, such as Armadillo, F-box and U-box family proteins, as well as DC1 domain containing proteins, are represented by multiple members at Time 1, and some of these are shared with Time 2 and 3. The role of functional groups as a whole and of individual gene members within the various functional groups in the response of Ar abidopsis roots to high levels of magnesium sulfate will where possible be addressed in the discussion. Transcripts Encoding Membrane Based Transporters in the Col-0 Time Series Because of our interest in identifying plant va riants with enhanced tolerance to elevated levels of magnesium sulfate and the possibility of differential expression of transporter genes as possible reasons for such enhanced tolerance, the responses of genes encoding membrane based transporters were documented not only at Time 1, but also across the other time points. Since gene expression differences are not fully contro lled for diurnal effects for the Time 2 and 3 comparisons in the col-0 time series, these effects will have to be ruled out in follow-up studies. A hierarchical average linkage cluster analys is was done based on transporter genes with 63

PAGE 64

significant expression differences at Time 1, 2 or 3 using uncentered correlation. The results show that distinct clusters of expression patterns can be distinguis hed within the group of transporter genes across the three time points. For example, some transporter genes are differentially expressed from Time 1 or 2 onwar ds, while most are not differentially expressed until Time 3. A few are differentially expressed at Time 1 and 2 but not 3, Time 2 only, or at Time 1 and 3 but not 2 (Figure 3-11). Groups of functionally related transporters When grouping the known transporter genes acc ording to the type of molecules they transport, the following can be observed (F igure 3-11, Object 3-2) Genes encoding ion transporters such as magnesium tr ansporter MRS2-10 (At1g80900) and Mg2+/H+ exchanger MHX1 show up-regulated expression at Time 2 and 3, while the expression of MRS2-7 (At5g09690) is up-regulated at Time 3. Sulfate transporter SULTR3;4 (At3g15990) shows down-regulated expression from Time 2 onw ards, while SULTR3;1 (At3g51895) and SULTR4;1 (At5g13550) show down-regulated expression at Ti me 3. The expression of the gene encoding Ca2+/H+ antiporter CAX1 is down-regulated almost tw o-fold starting at Time 1, while CAX2 and CAX3 begin to show down-regulated expression at Time 2. The gene encoding cation exchanger CHX20 is on the other hand up-regulate d in expression from Time 2 onwards. Genes encoding the high affinity nitrate tr ansporter ACH1, the ammonium transporter AMT1.1, and the inorganic phosphate transporters PHT1 and PHT2 all show up-regulated expression, while those encodi ng a putative inorganic phosph ate transporter (At2g29650), phosphate transporter PT2, and the putative m itochondrial phosphate transporter (At3g48850) show down-regulated expression. Transcripts en coding metal transporters NRAMP1 and ZIP5, as well as a ZIP Zinc trans porter domain containing protei n (At1g68100), an iron transporterrelated protein (At2g38460), a ferroportin-related protein (At5g26820), and a metal transporter 64

PAGE 65

family protein (At3g08650), all show up-regulated expression. The expression of four transcripts encoding potassium transporters is up-regulated at different time point s, with that of HAK5 consistently and increasingly across the three co mparisons. The expression of potassium channel encoding genes AKT1 and AKT4 is up-regulated, while that of AKT2 is greatly down-regulated. Transcripts encoding chlorine channel prot eins CLC-a and CLC-b show down-regulated expression, while a transcript encoding a putat ive chloride channellike protein (At5g33280) shows up-regulated expression. The expression of three genes enc oding putative cation-chloride cotransporters is up-regulated, wh ile that of a gene encoding K-Cl cotransporter type 1 proteinrelated (At3g58370) is down-regulat ed. Two transcripts encoding cat ion efflux family proteins are differentially expressed in c ontrasting ways, and the expression of a transcript encoding an anion exchange family protein (At3g62270) is down-regulated. Genes encoding ion-transporting ATP-ases, such as calcium-transporting ATP-ases, are well represented among the differe ntially expressed transporter genes. Genes encoding ACA1, ACA9, ACA10 and ACA12, as well as ECA2, a ll show down-regulated expression, while those encoding ACA2, ACA4 and ACA8 show up-re gulated expression. The expression of two transcripts encoding E1-E2 type ATPase family proteins is up-regulated, while two transcripts encoding copper exporting ATP-ases show contra sting expression patter ns. The expression of genes encoding phospholipid-transporting ATPase 1/magnesium-ATPase 1 (ALA1) and an anion-transporting ATPase family protein (At5g60730) is down-re gulated from Time 2 onwards. Three transcripts encoding putat ive proton pump ATP-ases show up-regulated expression at Time 3. Genes encoding cyclic nucleotide gated ion channels CNGC1 and CNGC19 show differential expression at the early time poi nts only, while those encoding CNGC10, CNGC13 65

PAGE 66

and CNGC15 are differentially expressed at Tim e 2 or later. Two transcripts encoding mechanosensitive ion channel domain-containing proteins show up-regulated expression. A group of genes encoding major intrinsic family proteins (MIP) show dow n-regulated expression at Time 3, with the exception of At4g01470, which shows up-regulated expression. The expression of multiple genes encoding plasma membrane intrinsic proteins (PIP1B, 1C, 2A, 2C), and two tonoplast intrinsic prot eins (TIP3.1) is down-regulate d. Several transcript encoding mitochondrial import (Tim17/Tim22/Tim23) family proteins show up-re gulated expression at Time 3, except for At3g10110, which shows down-re gulated expression. A number of genes encoding mitochondrial substrate carrier family pr oteins are also differe ntially expressed, with At4g24570 showing up-regulated expression across all comparisons. At4g24570 is shown to be universally responsive to stress (Ma and Bohnert, 2007). The expre ssion of a group of transcripts encoding nodulin MtN21 family and nodulin-related proteins is up-regulated at Time 3, again with an exception in the form of the expression of At4g30420. With respect to transporters of specific metabolic products, two genes encoding auxin efflux carrier proteins show dow n-regulated expression, while a gene encoding auxin transport protein PIN3 shows up-regulated expressi on. A transcript enc oding a C4-dicarboxylate transporter/malic acid transport family protei n (At4g27970) shows down-re gulated expression at Time 2. The expression of a gene encoding an organic cation transp orter-related protein (At1g16390) is greatly down-regulated across all time points. Transcript s encoding glutathione S-conjugate ABC transporters MRP1 and MRP2, and a putative glycerol-3-phosphate transporter/permease (At3g47420) all show dow n-regulated expressi on at Time 3. The expression of a gene encoding a putative lysine and histidine specific tr ansporter (At3g01760) is down-regulated across the three time points, while that of At1g25530 is up-regulated from Time 66

PAGE 67

2 onwards. A transcript enc oding an UDP-galactose/UDP-glu cose transporter shows upregulated expression at Time 3, whi le transcripts encoding the sucrose transporters/sucroseproton symporters SUC1, SUC3 and SUC5 a ll show down-regulated expression. Finally, transporter gene families represented by multiple members with little annotation, and with mixed differential expression patterns amongst the member s, are the amino acid transporter family, the proton-dependent oligopeptide tr ansport family, the sugar tran sporter family, the MATE efflux family, the ABC transporter family and the integral membrane family. Q-PCR confirmation The expression of four transporter genes that showed a significant difference in expression at Time 3 was analyzed by quantit ative PCR. Time 3 refers to the microarray comparison of the col-0 time 3 treatment sample set with the col-0 time 1 control sample set (Table 3-1). Because the control sample set is harvested at time 1, the Time 2 and Time 3 comparisons are not controlled for diurnal effects. In order to conf irm that the expression differences observed at Time 2 or 3 for several transporte r genes of relevance to this study are not due to diurnal effects, The treatment at time 3 was repeated w ith a different set of hydroponically grown col-0 plants and a control set was harvested at time 3 as we ll (Figure 3-2 D). In addi tion to confirming their expression using the same RNA sources as thos e used for the microarray experiment, RNA was used from these sets of independently grown plants. The genes analyzed were CAX1, MRS2-10, SULTR3;4 and NRAMP1. CAX1 and MRS2-10 were chos en to be able to relate the microarray results for these genes to the knockout mutant results of Chapter 2. SULTR3;4 was selected because it represents a sulfate transporter gene not previously reported to be regulated by sulfur levels. NRAMP1 was included because it enco des a transporter of ions other than Mg2+, SO4 2or Ca2+. The differences in gene expression between treatment and control as measured by Q-PCR are most similar to the results achieved with the microarray experiment when using the same 67

PAGE 68

RNA sources (Tables 3-5a, 3-5b, 3-6, 3-7). When using RNA from the independently grown and diurnally controlled plants for the Q-PCR analysis, the expres sion differences are similar to the microarray results for CAX1 and NRAMP1, less pronounced for SULTR3;4, while MRS2-10 shows no difference at all (Tables 3-6, 3-7, Fi gures 3-12, 3-13, 3-14). Th e discrepancy between the Q-PCR and the microarray results for SULT R3;4 and MRS2-10 may be due to diurnal regulation of the expression of these transcript s between time 1 and time 3 in the microarray analysis. Despite this, the gene expression difference of SU LTR3;4 between treatment and control at time 3 is still significant during one of the Q-PCR analysis repetitions. The small difference in expression of MRS2-10 between treatment and control does not qualify as significant in any of the Q-PCR analyses, even when using the same RNA sources as those used for the microarray experiment, which may indica te a limitation of the SYBR green dye based QPCR analysis. Transcripts Cax1-1/Col-0 Comparison Only 4 transcripts showed a significant difference in expression between the cax1-1 time 3 treatment and col-0 time 3 treatment sample sets (Table 3-8). Among these, the transcript with the smallest q-values and largest differences in expression is CAX1, the gene that is knocked out in the mutant cax1-1 and which is represented on the mi croarray by two different probes. The other three transcripts are At3g01345, an expressed protein with sim ilarity to beta-galactosidases in Arabidopsis, At4g07526, an unknow n protein with similarity to other unknown proteins in Arabidopsis, and chromosomal region CHR2:011819877-011819818, with no significant similarity to other nucleotide sequences. Supplemental Files Object 3-1. Microarray raw si gnal data (.xls file 35 MB) Object 3-2. Microarray analys is data (.xls file 1 MB) 68

PAGE 69

Discussion The objective of the current st udy was to quantify the early transcriptome responses of wildtype Arabidopsis ( col0 ) roots to elevated MgSO4H2O and to possibly identify candidate genes that play a role in enhancing tolerance to MgSO4H2O by limiting accumulation within the plant. Our results showed 325 differentially expressing genes in col-0 after 45 minutes of exposure compared to col-0 unexposed for 45 minutes (Time 1) that could be organized into functionally related groups and many of which were shared with either Time 2 or 3, or with both. In addition we described the tran sporter genes differentially expre ssed at all three time points. Finally, we showed that only three transc ripts are differentiall y expressed between cax1-1 and col-0 In this discussion we will attempt to clarify how the results might relate to the treatment given. Transcripts Col-0 Time Series Phytohormones like abscisic aci d (ABA), ethylene, salicyli c acid and jasmonates, and other signalling molecules such as nutrients a nd intracellular second messengers (phospholipids, Ca2+ and reactive oxygen species), have been implic ated in mediating cellular events associated with the response of the plant to abiotic stress (Magnan et al ., 2008). Genes encoding enzymes involved in the ABA, ethylene and jasmonic aci d biosynthesis pathways show differential expression at Time 1 and beyond. Transcript s encoding enzymes 9-cis-epoxycarotenoid dioxygenase and aldehyde oxidase 3 in the AB A biosynthesis pathway show up-regulated expression, pointing to increas ed ABA synthesis. Genes encoding 1-aminocyclopropane-1carboxylate synthases in the ethy lene biosynthesis pathway show up-regulated expression across the time points, while several genes encoding enzymes with 1-aminoc yclopropane-1-carboxylate oxidase activity are differentially expressed at Time 3, indicating possi ble changes in ethylene synthesis. In the jasmonate biosynthesis pa thway, a lipoxygenase encoding gene shows up69

PAGE 70

regulated expression at Time 1, whi le a gene encoding a 12-oxophytodienoate reductase shows down-regulated expression at Time 2 and 3, pointing to possible changes in jasmonate synthesis. The expression of a gene encoding a flavin m onooxygenase was up-regula ted at Time 1. This enzyme shows similarity to YUCCA, a flavin monooxygenase-like enzyme in Arabidopsis that catalyzes hydroxylation of the amino group of tr yptamine, a rate-limiting step in tryptophandependent auxin biosynthesis and which causes elev ated levels of free auxin in dominant mutant plants (Zhao et al., 2001). Two Arabidopsis cy tochrome P450s, CYP79B2 and CYP79B3, which convert tryptophan (Trp) to indole3-acetaldoxime (IAOx) in vitro, were also shown to be critical enzymes in auxin biosynthesis in vivo (Zhao et al., 2002). The cytochrome P450 that is encoded by a gene with up-regulated expression at Time 1 in this study is CYP94A1, which has not been associated with auxin biosynthesis, but w ith fatty acid hydroxylat ion. Products of the Auxin/Indole-3-Acetic Acid (A ux/IAA) and auxin response factor (ARF) gene families can function in a negative (auto)feedback loop in which particular combinations of ARF and Aux/IAA proteins control specif ic auxin-mediated responses via regulation of gene expression, such as for example the combination of I AA19 and ARF7, which may constitute a negative feedback loop to regulate diffe rential growth responses of hypocotyls and lateral roots in Arabidopsis (Tatematsu et al., 2004). Genes encoding IAA1 and IAA19 both show up-regulated expression at Time 1 in this study, while the other genes encoding auxin-responsive proteins (GH3, SAUR) at Time 1 also show up-regulat ed expression. These results indicate the occurrence of auxin-mediated responses such as negative regulation of latera l root growth in the presence of elevated levels of magnesium su lfate. The expression of several genes encoding gibberellin 2 oxidases, which are involved in gi bberrellin (GA) inactivation, is up-regulated at the different time points, indicating a reducti on in GA in this phase. One of these genes 70

PAGE 71

(At1g02400) is differentially expressed in response to a broad range of stresses (Ma and Bohnert, 2007). The involvement of auxin in the regul ation of GA metabolism genes points at the existence of a complex regulatory network (Frigerio et al., 2006). A large number of genes in the Arabidops is genome encode EF hand-containing Ca2+binding proteins, such as calmodulin(-like) and cal cineurin B-like proteins, which can act as Ca2+ sensors that relay Ca2+ signals through Ca2+-induced conformational ch anges that can affect interactions with an abundan ce of target proteins and m odulate target protein activity (McCormack et al., 2005). Calmodu lin-like 9 (CML9) is for exam ple involved in salt stress tolerance through its effect on the ABA-mediated pathways in Arabi dopsis (Magnan et al., 2008). The genes encoding Ca2+-binding proteins that are differentially expressed at Time 1 in this study are not functionally associ ated to specific signaling pathwa ys at this point. Three of the differentially expressed genes (At2 g41410, At2g46600 and At3g10300) are universally responsive to stress (Ma and Bohnert, 2007). Ca lcineurin B-like (CBL) proteins specifically target CBL-interacting protein kinases (CIPKs) such as CIPK9, which is required for lowpotassium tolerance in Arabidopsis (Pandey et al., 2007). The expression of the gene encoding CIPK9 is up-regulated at Time 1. Another gene with differen tial expression at Time 1 is At2g30040, which encodes a protein kinase family pr otein and is known to show a response to a broad range of stresses (Ma and Bohnert, 2007). The majority of transcripts encoding kinases that are differentially expressed at Time 1 enc ode functionally unspecifi ed protein kinases as well as receptor-like protein ki nases (RLKs) with extracytoplas mic leucine-rich repeats (LRRs) or with an extracellular lectin-like domain. RLKs are membrane-spanning proteins with a predicted signal sequence and a cyt oplasmic kinase domain that have been implicated in a wide range of signal transduction pathways (Fontes et al., 2004). Several genes encoding protein 71

PAGE 72

phosphatase 2C family proteins, which are pote ntially involved in phos phorylation-m ediated signaling, are differentially expressed at Time 1. Protein degradation can also play a role in signal transduction and transcriptional regulation. Genes related to protein degradation are also differentially expressed at Time 1, such as t hose encoding S-phase kina se-associated protein 1 (SKP1) interacting F-box proteins (up-regulated), U-box proteins, Arm/ U-box proteins, and an ubiquitin family protein (down-regulated). The exact function of many of the differen tially expressed genes at Time 1 encoding transcription factors is currently unknown, although some are associated with tolerance to other abiotic stresses. Members belonging to the NAC (NAM) and bHLH transcription factor families are coordinated in their up-regul ated expression at Time 1. Seve ral genes encoding C2H2 type zinc finger family proteins were detected, which play a crucial role in many metabolic pathways as well as in stress response and defense activatio n in plants. C2H2 type zinc finger proteins can have a putative repression activity in the defens e and stress response of plants, which is thought to occur via their ethylene-responsive elemen t-binding factor (ERF)-associated amphiphilic repression (EAR) domain (Ciftci-Yilmaz and Mittle r, 2008). An example of this is ZAT7, which renders plants more tolerant to salinity stress when constitutively expressed, and which is therefore thought to function in Ar abidopsis as a suppressor of a repressor of defense responses (Ciftci-Yilmaz et al., 2007). This study shows that the expression of the gene encoding ZAT7 is up-regulated after 45 minutes of e xposure to elevated levels of ma gnesium sulfate and that it is increasingly up-regulated with tim e up to at least 3 hours after tr eatment initiation. The functions of the other two differentially expressed genes at Time 1 encoding C2H2 type transcription factors have not been determined yet. AP2/ERF superfamily transcription factors play significant roles in regulating plant biotic and abiotic stress-responsive gene expression and can be divided 72

PAGE 73

into several f amilies, one of which is the ERF family with a single AP2/ERF domain (Zhang et al., 2008). An ERF transcription factor of th e DREB1/CBF subfamily was shown to bind DRElike motifs in the promoter of gibberellin 2 oxidase 7 (GA2ox7), which is an enzyme encoded by a gene whose expression is up-regulated in Arabidopsis under high-salinity stress, and which reduces endogenous GA levels with the result that growth is repressed for stress adaptation (Magome et al., 2008). Some of the AP2/ERF tr anscription factors encoded by differentially expressed genes at Time 1 may therefore be i nvolved in the up-regulation of the expression of genes encoding gibberellin 2 oxidases, such as GA2ox7, in this study. Another transcription factor, WRKY70, functions as a ne gative regulator of developmen tal senescence an d is involved in plant defense signaling pathways (Ulker et al., 2007). In the current study, the expression of the gene encoding WRKY70 is stably down -regulated across the three time points. Several differentially expressed genes point at senescence-relat ed responses in Arabidopsis roots exposed to high levels of magnesium sulf ate. Two genes encoding senescence-associated proteins (At2g25690, At4g35985) show up-regulated expression at Time 1. Two genes encoding hairpin induced (HIN1) pr oteins (At4g01410, At2g27080) which previously showed upregulated expression during leaf senescence in tobacco (Takahashi et al., 2004), also show upregulated expression at Time 1. At4g35985 and At2g27080 are shown to be responsive to a broad ranges stress (Ma and Bohnert, 2007). A gene encoding a Rieske [2Fe-2S] domaincontaining protein similar to pheophorbide A oxygenase, which is part of the chlorophyll breakdown pathway, shows up-regulate d expression at Time 1 and 3. It is possible that it is transported to the shoot in the form of RNA or protein, or that the gene product has a different function in roots in the absence of chlorophyll. 73

PAGE 74

A general pattern of down-regulated expres sion of genes encoding disease resistance proteins was observed at Time 1. Consisten t with the need for a rapid resp onse to pathogen attack, many NBS-LRR-encoding genes are constitu tively expressed at low levels in healthy, unchallenged tissue, and little is known about the regulation of these genes (McHale et al., 2006). Copines such as BON1 and BONZAI1, which are calcium -dependent membrane-binding proteins, have been implicated in the repression of several disease resist ance genes (Yang et al., 2006). In this study, the expression of BON1 an d BONZAI1-related genes is down-regulated, while the expression of the majority of disease resistance genes is not seen to be up-regulated at Time 1. At later time points the expression of the majority of disease resistance genes is still down-regulated, although the number of genes showing up-regulated e xpression increases. Many of the differentially expressed genes detected at Time 1 encode cell wall related proteins. Arabinogalactan proteins for example ar e plant glycoproteins th at consist of a coreprotein backbone O-glycosylated by one or more complex carbohydrates consisting of galactan and arabinose as main component s. Fasciclin-like arabinogalactan proteins contain one or two fasciclin domains thought to be involved in pr otein-protein interactio ns that are variably modified with a glycosylphosphatidylinositol lip id anchor (Seifert and Roberts, 2007). Cell surface arabinogalactan proteins in the roots of Arabidopsis were shown to influence the organization of cortical microt ubules, which in turn affect the orientation of cellulose microfibrils that constitute an ordered, fibrous phase of the cell wall (Nguema-Ona et al., 2007). In roots, arabinogalactan prot eins are implied in root elongation, the orientation of root patterning and root hair growth (Seifert and Roberts, 2007). Genes encoding arabinogalactan proteins show down-regulated e xpression at Time 1, indicating a possible disorganization of cortical microtubules and consequently of cellu lose microfibrils, and a reduction in root 74

PAGE 75

elongation and root hair growth. The gene encoding UGE4 also shows down-regulated expression at Time 1 in this study. A loss-of-f unction allele of UDP-glucose 4-epim erase 4 (UGE4), which freely interconvert s UDP-glucose and UDP-galactose resulted in a 20% decrease in Arabidopsis root cell wall galactose content th at correlated with a re duction in root length (Rosti et al., 2007). The loss of UGE4 specifica lly affected xyloglucan galactosylation by causing a 40% reduction in the relative total a bundance of galactose-containing xyloglucan oligosaccharides in roots. Xyloglucans are polysaccharides binding noncovalently to cellulose, thereby coating and cross-linking adjacent cellulose microfibrils (Rose et al., 2002). Xyloglucan galactosyltransferases, which show sequence sim ilarities to the glucur onosyltransferase domain of exostosins, are involved in the biosynthe sis of xyloglucans by attaching D-galactose to specific xylose residues (Madson et al., 2003). The expression of two genes encoding xyloglucan galactosyltransferases is dow n-regulated at Time 1, there by affecting the structure of xyloglucans. A class of enzymes known as xylogl ucan endotransglucosylases/hydrolases (XTHs) catalyzes the endo-cleavage of xyloglucan polymer s and the subsequent tr ansfer of the newly generated reducing ends to other polymeric or oligomeric xyloglucan molecules (Liu et al., 2007). Xyloglucan endotransglucosylases/hydrolases have been associated with primary root growth and root hair initiation in Arabidopsis (Vissenberg et al., 2001; Liu et al., 2007). The down-regulated expression at Time 1 of gene s encoding these enzymes indicates a possible reduction in the associated proce sses. Expansins are cell wall loos ening proteins with a proposed nonenzymatic action mechanism in which noncovalent bonds are disrupted that connect matrix polysaccharides to the surface of cellulose (Sampedro and Cosgrove, 2005). All the -expansin proteins that have been characterized so far ha ve a pH optimum for cellwall extension of about 4, which permits the cell to regulate -expansin activity rapidly by modulating wall pH through 75

PAGE 76

the activity of the plasma membrane H+ ATPase, which pumps protons to the cell wall (Sampedro and Cosgrove, 2005). The expression of genes encoding -expansin proteins is upregulated at Time 1, while that of three transcri pts encoding putative prot on pump ATP-ases is up-regulated at Time 3, possibl y indicating an increase in -expansin activity with time. The expression of two genes encoding -expansin proteins on the othe r hand is down-regulated; a expansin gene in barley is tightly related to root hair initiation ( Kwasniewski and Szarejko, 2006). Transcripts encoding lipid transfer protei ns, which in this study show down-regulated expression, can facilitate the ongoing extension process of cell walls in tobacco and wheat, where they were associated to the presence of a -expansin (Nieuwland et al., 2005). A gene encoding a pectin acetylesterase also shows down-regulated expression at Time 1. Pectin acetylesterases catalyze the deacety lation of esterified pectin, which diminishes the pectin backbone hydrophobicity and in creases its solubility in water, while it makes the polysaccharide more accessible to pectin-degrading enzymes (V ercauteren et al., 2002). The expression of transcripts encoding polygalacturon ases on the contrary is up-regul ated. These are enzymes that hydrolyze the homogalacturonan (p ectin backbone) of the plan t cell wall and are therefore involved in cell wall loosening. In general, the genes encoding proteins involved in cell-wall loosening show down-regulated e xpression in this study where they are involved in root (hair) growth, while those showing up-regulated expression might be the ones involved in expansion of existing cells. The up-regulated expression of giberrellin 2 oxidases mentioned earlier in the discussion also indicates a possible decrea se in root growth. Gibberel lins (GAs) promote growth by targeting the degradation of DE LLA repressor proteins in the endodermis, where cell expansion is rate-limiting for elongation of ot her tissues and therefore of the root as a whole (Ubeda-Tomas 76

PAGE 77

et al., 2008). Other genes potential ly associated with root growth processes are for example the transc ription factor scarecrow-like 11 and the putative replication prot ein A1, both of which show down-regulated expression at Time 1. Scarecrow-like 11 is related to SCARECROW, which is involved in root cortex cell proliferation, stem cell rene wal, and the gravitropic response (Cui and Benfey, 2009). Finally, th e expression of two anionic per oxidases, one of which is a putative lignin forming anionic peroxidase, is strongly up-regulated at Time 1. Anionic peroxidases belong to class III peroxidases, which are generally secreted in to the cell wall or the surrounding medium and the vacuole wher e they catalyse the reduction of H2O2 by taking electrons from various donor molecules such as phenolic compounds, lignin precursors, auxin, or secondary metabolites, or where they are involved in the form ation of various reactive oxygen species (ROS) (Cosio and Dunand, 2009). Lignin ma y form a barrier against metal entrance (Kovacik and Klejdus, 2008). Many genes encoding proteins of known and unknown function are now implicated in the early response of Arabidopsis roots to high leve ls of magnesium sulfate. Several of the genes described above have been shown to be involved in an abiotic stress response in previous studies, but the majority can only be functionally inte rpreted on a family or superfamily level. Transcripts Encoding Transport Proteins Col-0 Time Series Vacuolar transporters Storage vacuoles contain mainly inorganic sa lts and water, enabling the plant to reach a large size and surface area using a minimum of en ergy for the synthesis of organic metabolites. The vacuole also serves as an internal reservoi r of metabolites and nutrients and takes part in cytosolic ion homeostasis (Martinoi a et al., 2000). The vacuolar M HX transporter functions as an exchanger of H+ with Mg2+ and Zn2+ ions and is expressed in th e vascular cylinder in close association with the xylem trachear y elements and in the root ep idermis (Shaul et al., 1999). The 77

PAGE 78

gene encoding the MHX transporter shows up-regul ated expression at Ti me 2 and 3, indicating the possib le storage of excess Mg2+ in the vacuole. The tonoplast-localized H+/SO4 2cotransporter SULTR4;1 mediates the efflux of sulfate from vacuoles in the pericycle and xylem parenchyma cells of the vascular tissues along the entire root (K ataoka et al., 2004). The expression of the gene encoding SULTR4;1 is up -regulated at Time 3, poi nting at the possible retention of excess SO4 2in the vacuole. Several members of the CAX (cation exchanger) family show differential expression in response to elevated magnesium sulfate. The Ca2+/H+ antiporter CAX1, which shows down-regulated expression across the time points, is one of the few transporter genes that are already differentially expressed at Time 1. Its down-regulated expression was confirmed by Q-PCR at Time 3 when controlled fo r diurnal effects. CAX1 is localized to the tonoplast and responsible for 50% of the Ca2+/H+ antiport activity there (Cheng et al., 2003). The expression of the CAX2 antipor ter is down-regulated at Time 2 and 3. CAX2 has been shown to transport Ca2+ and Mn2+ into the vacuole in Ar abidopsis and other plant species (Edmond et al., 2009). CAX3, a weak Ca2+ vacuolar transporter, also shows downregulated expression from Time 2 onwards. C AX1 and CAX3 can be combined as heteroCAX complexes to form functiona l transporters with distinct tr ansport properties (Zhao et al., 2009). The down-regulated expression of the genes encoding these three vacuolar Ca2+/H+ antiporters indicates a possible shortage of cal cium in the cytosol. The gene encoding CHX20 shows up-regulated expression at Time 2 and 3. C HX20 is localized to endomembranes, where it is thought to play a critical role in osmoregulation and possibly pH modulation by exchanging K+ for H+ (Padmanaban et al., 2007). The expression of two tonoplast intrinsic proteins (TIP), which are vacuolar aquaporins, is also down-re gulated. Aquaporins are water channel proteins, but some have also been shown to transport CO2, H2O2, boron or silicon in addition to H2O 78

PAGE 79

(Maurel, 2007). An organic cati on transporter (OCT3), which is localized to the vacuol ar membrane, shows down-regulated expression acro ss the time points. OCT transporters are differentially expressed in response to various abiotic stresses (Kufner and Koch, 2008). ATPbinding cassette (ABC) type tran sporters such as MRP1 and MRP2, which show down-regulated expression at Time 3, are directly energized by Mg-ATP and do not depend on the electrochemical force. Their substrates are organic anions formed by conjugation, e.g. to glutathione (Marti noia et al., 2000). Plasma membrane transporters The slightly up-regulated expr ession of the gene encoding magnesium transporter MRS210 at Time 2 and 3 corresponds with the observa tion described in Chapter 2 that a knockout mutant of MRS2-10 does not show improved growth in the form of higher FW biomass compared to col-0 at elevated levels of magnesium sulfate. Q-PCR results did not show a statistically significant difference in expression fo r MRS2-10 at Time 3, with or without control for diurnal effects, although the difference in expression was clos er to the microarray results without the control at Time 3. The expression of the gene encoding MRS2-7, a low-affinity magnesium transporter of the same family, is also up-regulated. Both MRS2-10 and MRS2-7 have been shown to transport other ions in addition to Mg2+ (Deng et al., 2006; Mao et al., 2008). Schock et al. (2002) speculate that, in line with the large number of AtMRS2 family members, the function of the AtMRS2 gene family may be the maintenance of metal ion homeostasis in different cellular compartments (i.e. over differe nt cellular membrane systems). The expression of the gene encoding SULTR1;2, a high-affinity sulfate transporter, showed no significant differences at the three time points. The fact that expression of the gene is not down-regulated in the early adaptation response of Arabidopsis root s to high levels of magnesium sulfate supports the outcome of the FW biomass comparison between the SULTR1;2 knockout mutant sel1-10 79

PAGE 80

and col-0 described in Chapter 2, wh ich showed no advantage for sel1-10 The gene encoding EIL3, one of the known transcriptional regulators of SULTR1;2 (Maruyam a-Nakashita et al., 2006), did not show a significant difference in e xpression at any of the time points either. Interestingly, two sulfate transporter genes of the same family as SULTR1;2 show downregulated expression; the gene encoding SULTR 3;4 at Time 2 and 3, and the gene encoding SULTR3;1 at Time 3. The spatial and subcellular localiz ation of these trans porters is not known, and so far no influence on the expression of Group 3 sulfate transporters by the sulfur status of plants has been reported (Buc hner et al., 2004). This study shows for the first time that genes encoding Group 3 sulfate transporters SULTR3;1 and SULTR3;4 are differe ntially expressed in roots upon exposure to high levels of sulfate. The down-regulated expression of the gene encoding SULTR3;4 was confirmed by Q-PCR at Time 3, although its down-regulated expression was less pronounced when controlled for diurnal effects. Several genes encoding transporte rs of other inorganic nutrien ts were also differentially expressed. The expression of the gene en coding potassium channel HAK5 is strongly upregulated at Time 1 and conti nues to increase at Time 2 a nd 3. The expression of the HAK5 encoding gene is up-regulated by growth conditions that result in a hyper polarized root plasma membrane potential (Nieves-Co rdones et al., 2008). The inorga nic phosphate transporter PHT1 shows up-regulated expression at Time 1 and 3, which may point to an increased demand for phosphate. Several genes encoding root ion transporters that are differentially expressed in this study, such as ACH1 (AtNRT2.1) AMT1.1 and PT2, are diurnally regulated (Lejay et al., 2003). Whether they are also regulated by high magnesium sulfate will have to be determined in future studies. The down-regulated expr ession of the gene encoding CNGC1 at Time 1 and 2 could indicate a temporary reduction in Ca2+ uptake. Plant cyclic nucleo tide gated channels (CNGCs) 80

PAGE 81

are a large fam ily of proteins that are t hought to be ligand-gated, nonselective inwardly conducting cell membrane cation channels. CNGC1 prot ein is primarily expressed in the roots of Arabidopsis seedlings where it contribut es (along with ot her channels) to Ca2+ uptake into plants (Ma et al., 2006). Transcripts cax1-1/col-0 Comparison The differential expression of only four transcripts between cax1-1 and col-0 plants at time 3 indicates that the vast major ity of the root transcriptome re sponses to high magnesium sulfate are the same for the two genotypes at this time af ter initiation of treatmen t. As described above, CAX1 is one of the few transporter genes that ar e already differentially expressed at Time 1 in col-0 which means that down-regulated expression of CAX1 is part of the natural response of Arabidopsis to elevated levels of magnesium su lfate. The difference between the down-regulated level of CAX1 in col-0 and the absence of CAX1 in the knockout mutant might not be large enough, and the time passed between initiation of treatment and root harvest may not be long enough, to reveal many of the downstream molecula r processes eventually leading to enhanced tolerance for cax1-1 at the whole plant level, including reduced leaf Mg content and increased fresh weight biomass, after days or weeks of exposure in agar or so il medium (Chapter 2). The three transcripts showing a difference in abundance in cax1-1 compared to col-0 other than CAX1 itself currently do not have a know n function. Follow-up experiments need to be done to discover the function of these three tran scripts and whether they can help explain the observed higher tolerance of cax1-1 plants to elevated magnesium sulfate in the growth medium compared to col-0 The root transcriptome responses of cax1-1 and col-0 could be compared at later time points after initial exposure to high magnesium sulfate. Doing so might reveal additional differentially expresse d genes that play a role in the whole plant level tolerance difference exhibited by these genotypes. Some of the genes involved in downstream processes 81

PAGE 82

may the mselves be candidate genes for enhanced tolerance, such as those encoding (regulators of) plasma membrane based channels that transport Mg2+. Candidate Genes for enhanced tolerance One of the main objectives of this study was to identify potential transporter genes encoding plasma membrane based transporters of Mg2+ and SO4 2, or their transcriptional or posttranslational regulators, that are differentially expressed in the ear ly phases of Arabidopsis root adaptation to elevated levels of magnesium sulf ate. Based on the up-regulated expression of the gene encoding the known vacuolar transporter MHX, which transports Mg2+ into the vacuole, and the down-regulated expression of the gene encoding SULTR4;1, which exports SO4 2from the vacuole, it is concluded that both Mg2+ and SO4 2are available to th e plant in excessive amounts. Plasma membrane lo calized transporters of Mg2+ and SO4 2may therefore also be differentially expressed. Genes encoding know n plasma membrane based importers of Mg2+ and SO4 2-, such as MRS2-10, SULTR1;1 or SULTR1;2, did not show down-regulated expression. One possible way of Mg2+ entry in roots may be through putative homologs of the wheat rca channel, which is defined as a calcium channel, but which is permeable to a wide variety of monovalent and divalent cations including Ca2+, Mg2+, Mn2+, Cd2+, Co2+, Ni2+ K+, and Na+ (Shaul, 2002). Among the genes encoding tr ansporters of unknown function in the col-0 time series that show a significant difference in expression, candidate genes for improving plant tolerance of high magnesium sulfate by reduci ng accumulation within the plant might be found. Additional research is needed to determine whether for example any of the major intrinsic proteins, cyclic nucleotide gate d channels, integral membrane proteins, cation efflux family proteins, ATPase E1-E2 type family proteins, MATE efflux family proteins, transporter-related proteins and putative transporters, are plasma membrane localized and able to transport Mg2+ into or out of periphera l root cells, or export SO4 2from the same cells. Follow-up studies are 82

PAGE 83

also needed to determine th e localization of Group 3 sulfate transporters S ULTR3;1 and SULTR3;4 in the root. For transporters differentia lly expressed at Time 2 and 3, diurnal effects will have to be ruled out if th is has not been done already. Th e differential expression of posttranslational regulators of plasma membrane transporters could be an important way to affect ion uptake and efflux; candidate regulators could for example be kinases and phosphates that are differentially expressed at Time 1. Based on thes e first results it is not possible to identify specific candidate genes other than SULTR3;1 and SULTR3;4 whose (natural) overor underexpression could improve tolerance of Arabidopsis to elevated levels of magnesium sulfate. Even for a stress as widely studied as high NaCl, no plasma membrane localized transporters responsible for uptake of Na+ from the soil solution have been identified yet in Arabidopsis roots. In other plant species, Na+ has been shown to enter roots passively, via voltage independent (or weakly voltage-depen dent) nonselective cation channels and possibly via other Na+ transporters, such as some members of the high-affinity K+ transporter (HKT) family (Munns and Tester, 2008). For example, a line of transgenic wheat plants expressing an antisense construct of the high affinity K+ transporter TaHKT2;1 showed reduced sodium uptake by roots and enhanced growth relative to unstres sed plants compared to a control line at high levels of NaCl in the growth medium (Laurie et al., 2002). With respect to efflux, several plasma membrane ion transporters have been identi fied in Arabidopsis. Overexpression of the Arabidopsis SOS1 gene, which encodes a plasma membrane Na+/H+ antiporter responsible for Na+ efflux, limited Na+ accumulation and improved growth compared to control plants at high NaCl concentrations (Shi et al ., 2003). Overexpression of BOR4, an Arabidopsis borate exporter detected in the plasma membranes of the distal sides of root epidermal cells, generated plants that are tolerant of high boron levels (Miwa et al., 2007). The Arabidopsis NO3 efflux 83

PAGE 84

transporter NAXT1 is localized in the plasma mem b rane and mainly expressed in cortex cells of mature roots. It was found to be up-regulated at the post-transcriptiona l level in wildtype upon acid treatment (Segonzac et al., 2007). This exam ple shows that regulati on of import or efflux transporters may not occur at the transcrip tional level and can therefore be missed by a microarray experiment. Conclusions Arabidopsis col-0 root transcriptome analysis rev eals over 300 differentially expressed genes at Time 1. Genes of known function incl ude those encoding calc ium-binding proteins, kinases, transcription factors, enzymes involved in hormone metabolism, disease resistance proteins and many cell wall rela ted proteins. The responses of the genes encoding cell wall related proteins indicate a possibl e reduction in root growth when col-0 is exposed to high concentrations of magnesium sulfate. Some of the genes of known or unknown function were previously associated with specific or broad ra nges of abiotic stresses, but not necessarily in roots or at these time points. Over 200 genes encoding membrane based transporters were differentially expressed across the col-0 time series. The expression of genes encoding known plasma membrane based importers of Mg2+ and SO4 2ions, such as MRS2-10, SULTR1;1 or SULTR1;2, was not down-regulated. This corresponds with the observation reported in Chapter 2 that Arabidopsis lines ca rrying knockout T-DNA insertion mutations in AtMRS2-10 and AtSULTR1;2 did not mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype Arab idopsis plants. The differential expression of genes encoding known tonoplast localized transporters of Mg2+ and SO4 2ions indicate a possible storage of excess Mg2+ and SO4 2ions in the vacuole. Future rese arch can reveal whether any of the differentially expressed transporter genes of unknown protein localization, protein function, or both, are candidates to enha nce tolerance to high levels of so luble magnesium sulfate minerals in 84

PAGE 85

Martian regolith by redu cing accumulation of Mg2+ and SO4 2ions within the plant. For example, the localization of sulfate transpor ters SULTR3;1 and SULTR3;4 within Arabidopsis root tissue and cells can be analyzed in follow-up studies to see if the genes encoding these transporters are candidates for enhanced tolera nce. Potential regulators of membrane based transporter activity, such as kinases, which ar e encoded by differentially expressed genes across the col-0 time series, could also be analyzed. Sinc e gene expression differences are not fully controlled for diurnal effects for the Time 2 and 3 comparisons in the col-0 time series, these effects will have to be ruled out in follow-up studies. The down-regulation of cax1-1 gene expression is a natura l response to high magnesium sulfate in col-0 that is already seen at Time 1. Togeth er with the down-regulation of CAX2 and CAX3 gene expression at later time points it in dicates a possible shortage of calcium in the cytosol experienced by col-0 when exposed to high concentrations of magnesium sulfate. Only three transcripts were differentially expressed between cax1-1 and col-0 at 3 hours after initiation of treatment. Follow-up experiments could be do ne to discover the function of these three transcripts. The root transcriptome of cax1-1 and col-0 could furthermore be compared at later time points after initial exposure to high magnesium sulfate to reveal additional differentially expressed transcripts that could indicate the molecular processes eventually leading to the tolerance difference exhibited by these genotypes after days or we eks of growth. Some of the transcripts involved in downstream processes may themselves be candidates for enhanced tolerance, such as those encodi ng (regulators of) plasma membrane based channels that transport Mg2+. 85

PAGE 86

Materials and Methods Hydroponic Root Growth Few studies have focused on developing methods to grow Arabidopsis thaliana hydroponically, and only one system is commercially available at presen t (Tocquin et al., 2003). Detailed analyses of root enzyme activity and gene expression indica ted that it is highly important to continuously provide oxygen to the root zone through aeratio n (Smeets et al., 2008). We designed a hydroponic set-up for Arabidopsis that is generally based on the systems described by Tocquin et al. ( 2003) and Smeets et al. (2008). Gl ass containers (2.6 L) were covered along the glass with opaque black plastic sheeting and acro ss the top with opaque plastic lids. 18 Holes (d = 1.43 cm) were made in each opaque plastic lid with a hole puncher at regular intervals. A 2-cm-long rockwool plug (d = 1.59 cm ) was placed in each hole. Containers were filled with a nutrient so lution adjusted to pH 5.7 by KOH that was replaced once a week. It consisted of distilled water, 0.25 g/L MES buffer, and 1/32x MS sa lts during week 1 and 2, while the concentration of MS salts increased to 1/16x during week 3. A bout ten sterilized Arabidopsis seeds were planted per rockwool plug 18 containers were planted with col-0 while 6 were planted with cax1-1 Containers were covered with transp arent plastic wrap during the first 3-5 days of germination. Holes were made in the plastic wrap after 1-3 days depending on the general atmospheric humidity. The containers were randomly placed over 3 benches (8 containers/bench). Air was supplied to the roots by one 3W air pump per 8 containers; each container had an airstone made from glass beads. After one wee k, seedlings were thinned to one seedling per rockwool plug, so that every contai ner supported 18 plants (F igure 3-1). At day 21, col-0 and cax1-1 roots were exposed to the basic nutri ent solution (0.25 g/L MES, 1/16x MS, pH 5.7) with an additional 2.08 mM magnesiu m sulfate (total Ca:Mg ratio = 1:15). Col-0 was exposed for 45 min. (time 1), 90 min. (time 2), or 3 hrs (time 3), while cax1-1 was exposed for 3 86

PAGE 87

hrs only. The col-0 control received no extra magnesium sulfate and was harvested at 45 min. (time 1) together with the first col-0 treatment (Table 3-1). Four replicate containers were harvested for the control and the treatments (F igure 3-2). Roots were cut below the rockwool plug and pooled per container befo re being flash frozen in liqu id nitrogen and stored at -80oC. Microarray Procedures, Statistica l Analysis and Data display RNA was extracted from root samples with the RNeasy Plant Mini Kit (Qiagen). Samples were weighed while still frozen and reagents were adjusted to the total measured weight for the grinding and lysing step. Two replicate extracti ons were completed per sample with lysate volumes corresponding to 100 mg frozen wet weight each. The remainder of a sample was stored at -80oC as cleared lysate according to the Qiagen protocol. The quality and quantity of the extracted RNA was checked with denaturing agarose gels stained with ethidium bromide, and the NanoDrop 1000 Spectrophotometer (Thermo Scientif ic). RNA from each sample was amplified and labeled with cy3 dye by using Agilent Quic k Amp labeling kit, one color. A total of 20 samples were hybridized to five 4x44k Arabidopsis microarray slides (Agilent), which were then washed and scanned before data was extracte d (Figure 3-3). Microarr ay handling and data extraction was done at the Inte rdisciplinary Center for Biotechnology Research (ICBR) at the University of Florida according to the One-Colo r Microarray-Based Gene Expression Analysis (Quick Amp Labeling) Protocol (version 5.7). Median signal intensities were quantile normalized using R software. Log2 transformed normalized data were evaluated in a mixed analysis of variance (ANOVA) model using SAS 9 and JMP Genomics 7 software. Microarray slide was included in the ANOVA m odel as a random effect to account for the covariance of samples hybridized to the same microarray (i.e. control for spot effect) (Jin et al., 2001; Wolfinger et al., 2001). Unbiased estimates of transcript abunda nce (least-square means) were generated for each sample set. The transcript leve l estimates of different sample sets were then 87

PAGE 88

compared using a series of t-tes ts; the col-0 time 1, col-0 time 2 and col-0 time 3 treatment sets were compared to the col-0 time 1 control sample set, while the cax1-1 time 3 set was compared to the col-0 time 3 set (Table 3-1, Figure 3-3). The pvalues generated in these comparisons were corrected for multiple testing by controlling the False Discovery Rate (FDR) with the Q-value procedure in the Q-value 1.0 package (default setti ngs) of the R software (Storey and Tibshirani, 2003). Lists of genes with stat istically significant changes in expression at p < 0.05 between sample sets were further orga nized, analyzed and displayed in tables and figures using JMP Genomics 7, R software, GeneVenn, Cluster 2.11, TreeView 1.60, the National Center for Biotechnology Information (NCBI) BLAST tool, and the Arabidopsis Information Resource (TAIR) Aracyc Omics viewer and GO annotations tools. Quantitative Real-time PCR Wildtype Arabidopsis plants ( col-0 ) were grown hydroponically for 21 days as described above. At day 21, four containers received a replacement of the basic nutrient solution (0.25 g/L MES, 1/16x MS, pH5.7), while another four contai ners received the basic solution with an additional 2.08 mM magnesium sulf ate (total Ca:Mg ratio = 1:15). Roots were exposed for three hours before being harvested per container. R NA was extracted and checked for quality and quantity as described above. R NA from these independently grow n plants exposed or unexposed for 3 hrs, as well as RNA from plants exposed for 3 hrs and unexposed for 45 min. in the microarray experiment, were used for the quantitative PCR (Q-PCR) analyses. Q-PCR was performed using TaqMan reverse transcripti on reagents and Power SYBR Green PCR Master Mix (Applied Biosystems). 1 ug of total RNA was reverse transcribed per sample for a total of three control and three treatment samples. Prim ers were designed for 5 genes using Primer Express (Applied Biosystems) (Table 3-9). At2g32170 was chosen as a stably expressing reference gene appropriate fo r abiotic stress treatment (Czechowski et al., 2005). Each 88

PAGE 89

89 template/primer pair combination was run in triplicate. The relative increase or decrease of mRNA abundance between the two sample sets was calculated by using the Pfaffl method, and statistical analysis of the results was done with the REST 2008 2.0.7 software.

PAGE 90

Table 3-1. Experime ntal conditions of one color microarray experiment Treatment Time Seedline Biol. replicates Arrays Control solution 45 min. = time 1 col-0 4 4 Magnesium sulfate 45 min. = time 1 col-0 4 4 Magnesium sulfate 90 min. = time 2 col-0 4 4 Magnesium sulfate 3 hrs = time 3 col-0 4 4 Magnesium sulfate 3 hrs = time 3 cax1-1 4 4 Gene expression was compared between sample sets. The comparison of col-0 time 1 treatment with col-0 time 1 control is referred to in th e text, tables and figures as Time 1. Col-0 time 2 treatment versus col-0 time 1 control is referred to as Time 2. Col-0 time 3 treatment versus col-0 time 1 control is referred to as Time 3. The comparison of cax1-1 time 3 treatment with col-0 time 3 treatment is not referred to in abbreviated form. Table 3-2. Number of genes (and % of total) in GO molecular functional categories per comparison Functional Category Time 1 Time 2 Time 3 DNA or RNA binding 22 (6.43%) 103 (6.58%) 233 (6.93%) hydrolase activity 27 (7.89%) 105 (6.71%) 253 (7.53%) kinase activity 13 (3.80%) 83 (5.30%) 165 (4.91%) nucleic acid binding 5 (1.46%) 25 (1.60%) 52 (1.55%) nucleotide binding 12 (3.51%) 64 (4.09%) 152 (4.52%) other binding 42 ( 12.28%) 155 (9.90%) 331 (9.85%) other enzyme activity 23 (6.73%) 127 (8.12%) 347 (10.32%) other molecular functions 6 (1.75%) 54 (3.45%) 128 (3.81%) protein binding 34 (9.94%) 138 (8.82%) 262 (7.80%) receptor binding or activity 8 (2.34%) 22 (1.41%) 32 (0.95%) structural molecule activity 1 (0.29%) 18 (1.15%) 30 (0.89%) transcription fact or activity 33 (9.65%) 117 (7.48%) 230 (6.84%) transferase activity 16 (4.68%) 122 (7.80%) 265 (7.88%) transporter activity 14 (4.09%) 80 (5.11%) 202 (6.01%) unknown molecular functions 86 (25.15%) 352 (22.49%) 679 (20.20%) The table shows the number (and per centage of total) of significa ntly differentially expressed genes per GO functional category in col-0 roots exposed to magnesium sulfate at Time 1, Time 2 and Time 3 90

PAGE 91

Table 3-3. Genes of Arabidopsis thaliana ( col-0 ) with differential expr ession at q < 0.001 at Tim e 1 q-value Log2(FC) Gene and DNA region description 6.71E-07 -1.2264 phosphate-responsive protein, putative (EXO) [At4g08950.1] 6.71E-07 -2.0693 rhomboid fam ily protein [At4g23070.1] 1.99E-06 1.4961 expressed protein [At4g29780.1] 2.44E-06 1.5416 AP2/EREBP-like transcription factor LEAFY PETIOLE, putative [At5g13910.1] 6.25E-06 -1.3399 expressed protein [At5g25240.1] 8.93E-06 0.7094 kinase interacti ng family protein [At2g30500.1] 1.20E-05 0.7914 senescence-associated protein-related [At2g25690.1] 1.20E-05 -0.907 glycoside hydrolase fa mily 28 protein/polygalacturonase (pectinase) family protein [At3g06770.1] 1.20E-05 -1.6379 expressed protein [At4g37235.1] 1.20E-05 2.6714 expressed protein [At5g38700.1] 1.43E-05 -1.2778 calmodulin, putative [At3g10190.1] 1.65E-05 1.1068 expressed protein [At1g74450.1] 1.80E-05 1.673 nodulin-related [At2g30300.1] 3.48E-05 -1.149 arabinogalactan-protein (AGP17) [At2g23130.1] 6.97E-05 1.1562 calcium-binding EF hand family protein [At3g10300.1] 9.42E-05 -1.1368 expressed protein [At4g01140.1] 0.000216711 -0.9759 expressed protein [At2g17300.1] 0.000283167 1.3473 lipoxygenase, putative [At1g72520.1] 0.000295879 2.0824 immediate-early fungal elicitor family protein [At3g02840.1] 0.000299824 -0.7518 expressed protein [At1g09812.1] 0.00032124 2.2018 Unknown [CHR2:009686022-009686081] 0.000405444 2.3327 anionic peroxidase, putative [At1g14550.1] 0.000410629 -1.0069 arabinogalactan-pr otein (AGP21) [At1g55330.1] 0.000416943 -0.5479 expressed protein [At1g22882.1] 0.000427249 -1.4143 BON1-associated protein (BAP1)-related [At2g45760.1] 0.000432439 1.1365 exocyst subunit EXO70 family protein [At2g28650.1] 0.00044696 2.023 anionic peroxidase, putative [At1g14540.1] 0.000465244 -1.0837 expressed protei n predicted protein, Arabidopsis thaliana [At4g35320.1] 0.000465244 -1.585 Unknown [CHR1:021477192-021477133] 0.000468475 -1.4471 heat shock transcription factor family protein [At3g63350.1] 0.000510184 0.9195 auxin-responsive protein/i ndoleacetic acid-i nduced protein 19 (IAA19) [At3g15540.1] 0.000528206 -1.0809 RNA recognition motif (RRM )-containing protein [At1g78260.1] 0.000565577 -0.7772 arabinogalactan-pr otein (AGP7) [At5g65390.1] 0.00090829 1.9088 basic helix-loop-helix (bHLH) family protein [At2g22760.1] 91

PAGE 92

Table 3-4. Genes of Arabidopsis thaliana ( col-0 ) with dif ferential expres sion > 3 fold at Time 1 Log2(FC) q-value Gene and DNA region descriptions 3.2099 0.001474678 expansin, putative (EXP12) [At3g15370.1] 3.1098 0.009201495 F-box family protein/SKP1 interacting partner 3-related [At2g02320.1] 2.6714 1.20E-05 expressed protein [At5g38700.1] 2.6151 0.01425427 expressed protein [At1g20310.1] 2.556 0.003229586 expressed protein [At1g06135.1] 2.554 0.001396907 9-cis-epoxycarotenoi d dioxygenase, putative [At1g01140.1] 2.4955 0.00128496 expansin, putative (EXP17) [At4g01630.1] 2.3327 0.000405444 anionic peroxidase, putative [At1g14550.1] 2.2018 0.00032124 Unknown [CHR2:009686022-009686081] 2.0836 0.001823929 nitrate-responsive NOI protein, putative [At2g17660.1] 2.0824 0.000295879 immediate-early fungal elicitor family protein [At3g02840.1] 2.023 0.00044696 anionic peroxidase, putative [At1g14540.1] 1.9782 0.003274863 expressed protein [At5g50335.1] 1.9294 0.04072004 Unknown [CHR1:027987259-027987200] 1.9088 0.00090829 basic helix-loop-helix (bHLH) family protein [At2g22760.1] 1.8946 0.01474273 polygalacturonase, puta tive/pectinase, putative [At2g43890.1] 1.8802 0.04969909 S-adenosyl-L-methionine:car boxyl methyltransferase family protein [At3g44870.1] 1.8732 0.01911007 auxin-responsive family protein [At4g12410.1] 1.8605 0.004903664 1-aminocyc lopropane-1-carboxylate synthase, putative [At5g65800.1] 1.8518 0.03187784 no apical meristem (NAM) family protein [At2g46770.1] 1.8469 0.04757759 AP2 domain-containing transcription factor, putative [At4g34410.1] 1.8359 0.02514971 ovate family protein [At4g14860.1] 1.7794 0.01188793 CBL-interacting prot ein kinase 9 (CIPK9) [At1g78390.1] 1.7716 0.0074956 calmodulin-related protein, putative [At1g76640.1] 1.7374 0.003857111 glutamate deca rboxylase, putative [At2g02010.1] 1.6878 0.005354 ethylene-responsive element-binding protein, putative [At5g25190.1] 1.673 1.80E-05 nodulin-related [At2g30300.1] 1.6424 0.003447976 leucine-rich repeat family protein [At1g78230.1] 1.6067 0.02599538 lectin protein kinase, putative [At1g70130.1] -1.585 0.000465244 Unknown [CHR1:021477192-021477133] -1.6379 1.20E-05 expressed protein [At4g37235.1] -1.7618 0.005505618 disease resistance protein (TIR-NBS-LRR class), putative [At4g16860.1] -1.7916 0.04439508 Unknown [CHR1:021477409-021477350] -2.0347 0.02980178 cDNA clone RAFL0931-N12 3', mRNA sequence [AV802416] -2.0693 6.71E-07 rhomboid family protein [At4g23070.1] -2.3484 0.02514971 disease resistance protein (TIR-NBS class), putative [At3g04210.1] 92

PAGE 93

Table 3-5a. Q-PCR results of col-0 gene expression at time 3 trea tm ent versus time 1 control; RNA sources are the same as for the microarray experiment Gene Type Reaction Efficiency Expression (array) Std. Error 95% C.I. P-value Result At2g32170 REF 0.7657 1 CAX1 TRG 0.8076 0.57 (0.602) 0.436 0.766 0.366 0.925 0 DOWN MRS2-10 TRG 0.8548 1.035 (1.183) 0.758 1.377 0.628 1.466 0.714 SULTR3;4 TRG 0.7749 0.557 (0.541) 0.462 0.691 0.362 0.773 0 DOWN NRAMP1 TRG 0.8599 1.196 (1.317) 0.886 1.515 0.787 2.537 0.107 Table 3-5b. Q-PCR non-normalized results of col-0 gene expression at time 3 treatment versus time 1 control; RNA sources are the same as for the microarray experiment Gene Type Reaction Efficiency Expression (array) Std. Error 95% C.I. P-value Result At2g32170 REF 0.7657 1.117 0.989 1.264 0.800 1.518 0.051 CAX1 TRG 0.8076 0.637 (0.602) 0.535 0.775 0.460 0.860 0 DOWN MRS2-10 TRG 0.8548 1.156 (1.183) 0.915 1.435 0.829 1.662 0.06 SULTR3;4 TRG 0.7749 0.622 (0.541) 0.506 0.766 0.391 0.924 0 DOWN NRAMP1 TRG 0.8599 1.336 (1.317) 1.037 1.627 0.898 2.862 0.004 UP Table 3-6. Q-PCR results of col-0 gene expression at time 3 treatment versus time 3 control; repetition 1 Gene Type Reaction Efficiency Expression (array) Std. Error 95% C.I. P-value Result At2g32170 REF 0.7657 1 CAX1 TRG 0.8076 0.641 (0.602) 0.429 0.966 0.286 1.337 0.004 DOWN MRS2-10 TRG 0.8548 1.017 (1.183) 0.876 1.195 0.720 1.335 0.763 SULTR3;4 TRG 0.7749 0.861 (0.541) 0.695 1.059 0.553 1.162 0.049 DOWN NRAMP1 TRG 0.8599 1.128 (1.317) 0.970 1.381 0.638 1.524 0.156 Table 3-7. Q-PCR results of col-0 gene expression at time 3 treatment versus time 3 control; repetition 2 Gene Type Reaction Efficiency Expression (array) Std. Error 95% C.I. P-value Result At2g32170 REF 0.7657 1 CAX1 TRG 0.8076 0.565 (0.602) 0.491 0.673 0.418 0.734 0 DOWN MRS2-10 TRG 0.8548 0.986 (1.183) 0.786 1.206 0.657 1.603 0.864 SULTR3;4 TRG 0.7749 0.907 (0.541) 0.749 1.075 0.589 1.572 0.26 NRAMP1 TRG 0.8599 1.241 (1.317) 1.033 1.522 0.890 1.984 0.008 UP 93

PAGE 94

Table 3-8. Genes with differentia l expression at q < 0.05 between Arabidopsis thaliana cax1-1 and col-0 treated for 3 hours q-value Log2(FC) Gene and DNA region description 1.66E-09 -3.0461 calcium exchanger (CAX1) [At2g38170.1] 3.28E-07 -3.8209 calcium exchanger (CAX1) [At2g38170.3] 0.001201795 -2.7293 Expressed protein [At3g01345.1] 0.004291231 -1.1998 hypothetical protein [At4g07526.1] 0.005396832 -1.0903 Unknown [CHR2:011819877-011819818] Table 3-9. Genes and related primer sequences selected for Q-PCR Gene ID Primer sequence At2g32170 (reference gene) FW: 5 -GTTAAATCATGACCATGGCAGTGT-3 RV: 5-CTACATCAACCAGAGGAACATGTGT-3 At2g38170 (CAX1) FW: 5-GCGACTCAGATTGGCTTATTCG-3 RV: 5-GATCCATATTAATTCCCAAAATCCA-3 At1g80900 (MRS2-10) FW: 5-TTCTCTGTCTGCGCCAGTTTC-3 RV: 5-GGCTCCTTACAATGCTCAAGCT-3 At3g15990 (SULTR3;4) FW: 5-GGTGAAGCTGTGGCTGATCTC-3 RV: 5-GCTCCATCTTCAGAAACAGTCTCTCT-3 At1g80830 (NRAMP1) FW: 5-AC AGGATCTGGACGGT CTCAA-3 RV: 5-GATGAGTGG AGAATTGGAGAAGCT-3 94

PAGE 95

Figure 3-1. Hydroponic Arabidopsis growth. Panels A, B and C show the set-up for the microarray experiment. Panel D shows the set-up for the real-time Q-PCR experiment. 95

PAGE 96

Figure 3.2. Harvest of hydroponically grown Arabi dopsis roots. A) Example of Arabidopsis roots after 21 days of growth. B) Exampl e of root harvest per container for the microarray and real-time Q-PCR experiments. Figure 3.3 Overview of microarray experiment. A) Scan of a single microarray with a single cy3labeled cRNA sample hybridized to it. The experiment included 20 arrays in total, each with a single cy3-labeled sample hybridized to it. B) Diagram of the sample sets. Each set consisted of 4 biological replic ates. Gene expression was compared among the sets as indicated by the arrows. 96

PAGE 97

Figure 3-4. Volcano plot of Time 1. Graph with the x-axis showing log2 values of the fold changes in gene expression between col-0 exposed to magnesium sulfate for 45 minutes (n = 4) and col-0 exposed to the control nutrient solution for 45 minutes (n = 4). Each dot represents one of 37478 transcript s. Vertical lines indicate absolute fold change values as indicated on top of the graph. The y-axis shows the log10 p-values corresponding to the log2 fold change valu es. The horizontal line indicates the log10 p-value where the q-value is 0.05. Transc ripts whose expression difference (fold change) corresponds to a p-value for which q < 0.05 are indicated in red. 97

PAGE 98

Figure 3-5. Volcano plot of Time 2. Graph with the x-axis showing log2 values of the fold changes in gene expression between col-0 exposed to magnesium sulfate for 90 minutes (n = 4) and col-0 exposed to the control nutrient solution for 45 minutes (n = 4). Each dot represents one of 37478 transcript s. Vertical lines indicate absolute fold change values as indicated on top of the graph. The y-axis shows the log10 p-values corresponding to the log2 fold change valu es. The horizontal line indicates the log10 p-value where the q-value is 0.05. Transc ripts whose expression difference (fold change) corresponds to a p-value for which q < 0.05 are indicated in red. 98

PAGE 99

Figure 3-6. Volcano plot of Time 3. Graph with the x-axis showing log2 values of the fold changes in gene expression between col-0 exposed to magnesium sulfate for 3 hours (n = 4) and col-0 exposed to the control nutrient solu tion for 45 minutes (n = 4). Each dot represents one of 37478 tran scripts. Vertical lines indi cate absolute fold change values as indicated on top of the gra ph. The y-axis shows the log10 p-values corresponding to the log2 fold change valu es. The horizontal line indicates the log10 p-value where the q-value is 0.05. Transc ripts whose expression difference (fold change) corresponds to a p-value for which q < 0.05 are indicated in red. 99

PAGE 100

Figure 3-7. Volcano plot of cax1-1 versus col-0 at time 3. Graph with the x-axis showing log2 values of the fold changes in gene expression between cax1-1 exposed to magnesium sulfate for 3 hours (n = 4) and col-0 exposed to magnesium sulfate for 3 hours (n = 4). Each dot represents one of 37478 transcripts. Vertical lines indicate absolute fold change values as indicated on top of the graph. The y-axis shows the log10 p-values corresponding to the log2 fold change valu es. The horizontal line indicates the log10 p-value where the q-value is 0.05. Transc ripts whose expression difference (fold change) corresponds to a p-value for which q < 0.05 are indicated in red. 100

PAGE 101

Figure 3-8. Venn diagram of col-0 time series. The diagram shows the results of comparing sets of genes with significant e xpression differences at Time 1, Time 2 and Time 3. 74 genes are unique to Time 1, 48 are shared with Time 2, 48 are shared with Time 3, and 155 genes are shared between Time 1, 2 and 3. 101

PAGE 102

Figure 3-9. A hierarchical average linkage cluster analysis usi ng uncentered correlation was done across Time 1, 2 and 3 based on the genes w ith significant expression differences at Time 1. Yellow denotes a higher, and blue a lower expression of a gene in the treated plants versus the control. The figure shows the expression patterns of genes differentially expressed at Time 1 that are shared with Time 2, Time 3, or both. 102

PAGE 103

Figure 3-9. Continued 103

PAGE 104

Figure 3-9. Continued 104

PAGE 105

Figure 3-9. Continued 105

PAGE 106

Figure 3-9. Continued 106

PAGE 107

Figure 3-9. Continued 107

PAGE 108

Figure 3-10a. Differentially expr essed transcripts encoding metabolic enzymes in the ethylene biosynthesis pathway. Up-regulated expression of genes encoding 1aminocyclopropane-1-carboxylate synthase (A CS) at Time 1 (A), Time 2 (B) and Time 3 (C), and differential expression of genes encoding several enzymes with 1aminocyclopropane-1-carboxylate oxidase activ ity at Time 3 (C). The Log2 value of the difference in expression is indicated next to the enzyme or gene identifier in color. 108

PAGE 109

109 Figure 3-10b. Differentially expres sed transcripts encoding metabolic enzymes in the abscisic biosynthesis pathway. Up-regulated expression of genes encoding 9cis epoxycarotenoid dioxygenase/neoxanthin cleav age enzyme (NCED) at Time 1 (A), Time 2 (B) and Time 3 (C), and up-regulat ed expression of a gene encoding aldehyde oxidase 3 at Time 3 (C). The Log2 value of the difference in expression is indicated next to the enzyme or gene identifier in color. Figure 3-10c. Differentially expr essed transcripts encoding metabolic enzymes in the jasmonate biosynthesis pathway. Up-regul ated expression of a gene encoding lipoxygenase at Time 1 (A) and down-regulated expre ssion of several genes encoding 12oxophytodienoate reductase at Time 3 (C). The Log2 value of the difference in expression is indicated next to the en zyme or gene identifier in color.

PAGE 110

Figure 3-10d. Differentially expres sed transcripts encoding enzymes involved in gi bberellins metabolism. Up-regulated expressio n of genes encoding gibberellin 2-oxidase (Ga 2ox8, Ga2ox7) at Time 1 (A), Time 2 (B) and Time 3 (C), and down-regulated expression of a gene encoding gibberell in 2-oxidase (Ga2ox1) at Time 3 (C). The Log2 value of the difference in expression is indicated next to the en zyme or gene identifier in color. 110

PAGE 111

Figure 3-10e. Differentially expr essed transcripts encoding enzymes involved in glycerolipid metabolism. Down-regulated expression of a gene encoding monogalactosyldiacylglycerol synthase 2 (MGD2) at Time 1 (A), Time 2 (B) and Time 3 (C). The Log2 value of the differen ce in expression is i ndicated next to the enzyme or gene identifier in color. Figure 3-10f. Differentially expressed transcript s encoding metabolic enzymes in the hexosamine biosynthetic pathway. Up-regulated expres sion of a gene encoding glucosaminefructose-6-phosphate aminotransferase at Ti me 1 (A) and Time 3 (C). The Log2 value of the difference in expression is indicated next to the enzyme or gene identifier in color. 111

PAGE 112

112 Figure 3-10g. Differentially expr essed transcripts encoding enzymes involved in chlorophyll breakdown. Up-regulated expression of a gene encoding pheophorbide A oxygenase (PAO) at Time 1 (A) and Time 3 (C). The Log2 value of the difference in expression is indicated next to the enzyme or gene identifier in color.

PAGE 113

Figure 3-10h. Differentially expressed tran scripts encoding enzymes involved in porp hyrin and chlorophyll metabolism. Up-regula ted expression of a gene encoding glutamyl -tRNA reductase 2 (HEMA2) at Time 1 (A) and Time 2 (B), and up-regulated expression of a gene encoding sirohydrochlorin ferrochelatase (AtSir B) at Time 2 (B) and Time 3 (C). The Log2 value of the difference in expression is indicated next to the enzyme or gene identifier in color. 113

PAGE 114

Figure 3-10i. Differentially expressed transcript s encoding metabolic enzymes in the histidine biosynthesis pathway. Up-regulated expres sion of a gene encoding phosphoribosylATP pyrophosphohydrolase at Time 1 (A), Ti me 2 (B) and Time 3 (C) The Log2 value of the difference in expr ession is indicated next to th e enzyme or gene identifier in color. 114

PAGE 115

Figure 3-10j. Differentially expressed transcripts encoding metabolic enzymes involved in galactose metabolism. Down-regulated expr ession of a gene encoding UDP-glucose 4-epimerase (UGE4) at Time 1 (A) and Time 2 (B). The Log2 value of the difference in expression is indicated next to the enzyme or gene identifier in color. 115

PAGE 116

Figure 3-10k. Differentially expres sed transcripts encoding enzyme s involved in glycolysis. Upregulated expression of genes encoding phospho fructokinase at Time 1 (A) and Time 3 (C). The Log2 value of the difference in ex pression is indicated next to the enzyme or gene identifier in color. 116

PAGE 117

Figure 3-10l. Differentially expre ssed transcripts encoding enzymes in the citric acid cycle. Upregulated expression of a gene encoding ma late dehydrogenase (cyt osolic) at Time 1 (A), and down-regulated expression of a gene encoding malate dehydrogenase (mitochondrial) at Time 3 (C). The Log2 value of the difference in expression is indicated next to the enzyme or gene identifier in color. 117

PAGE 118

Figure 3-10l. Continued 118

PAGE 119

Figure 3-10l. Continued 119

PAGE 120

Figure 3-11. Hierarchical average linkage cluster analysis of transporter gene expression using uncentered correlation. The cluster analysis is based on transporter genes with significant expression at Time 1, 2 or 3. Yellow denotes a higher, and blue a lower expression of a gene in the treated plants versus the control. The figure shows that distinct clusters of expres sion patterns can be distingui shed within the group of transporter genes across the three comparisons. 120

PAGE 121

Figure 3-11. Continued 121

PAGE 122

Figure 3-11. Continued 122

PAGE 123

Figure 3-11. Continued 123

PAGE 124

Figure 3-12. Whisker box plots representing gene expression ratio distributions for the Q-PCR analysis of four genes showing differe ntial expression on the microarrays in col-0 at time 3 treatment versus time 1 control. RNA sources were the same as for the microarray experiment. Results show permutated expression data that are calculated by the REST 2008 statistical analysis softwa re, which uses randomization techniques. The graph gives an impression of the expres sion ratio distribution per gene related to the results presented in Tables 3-5a and 3-5b. Figure 3-13. Whisker box plots representing gene expression ratio distributions for the Q-PCR analysis of four genes showing differe ntial expression on the microarrays in col-0 at time 3 treatment versus time 3 control. R NA came from plants tr eated and controlled for diurnal effects at time 3. Results represent the first re petition of the analysis and show permutated expression data that are calculated by the REST 2008 statistical analysis software, which uses randomi zation techniques. The graph gives an impression of the expression ratio distribution per gene related to the results presented in Table 3-6. 124

PAGE 125

125 Figure 3-14. Whisker box plots representing gene expression ratio distributions for the Q-PCR analysis of four genes showing differe ntial expression on the microarrays in col-0 at time 3 treatment versus time 3 control. R NA came from plants tr eated and controlled for diurnal effects at time 3. Results repres ent the second repetition of the analysis and show permutated expression data that are calculated by the REST 2008 statistical analysis software, which uses randomi zation techniques. The graph gives an impression of the expression ratio distribution per gene related to the results presented in Table 3-7.

PAGE 126

CHAPTER 4 EFFECTS OF A S PACEFLIGHT ENVIRONMEN T ON HERITABLE CHANGES IN WHEAT GENE EXPRESSION Introduction A fundamental question in space biology is wh ether the long-term exposure (such as for one or more life cycles) of pl ants to the spaceflight environm ent with its microgravity and radiation parameters, can cause changes in subseq uent generations. Ways to address this question include making genomic, proteomic, metabolic and morphologic comparisons. In this study we used a genomic approach by testing for heritable di fferences in wheat gene expression patterns in plants grown from seed whose progenitors were set in a spaceflight environment the space station MIR. The molecular res ponse of Arabidopsis and wheat to the spaceflight environment has been evaluated in terms of genome-wide pa tterns of differential gene expression between orbital and ground control plants (Paul et al., 200 5; Stutte et al., 2006). These studies examined the direct reaction to a novel e nvironment; they do not reflect sp aceflight induced changes to the genome. The present study represents the first time that gene expression analysis is used to assay for potential spaceflight-induced changes in late r generations. An overview of experiments to date addressing these two fundamental questions whether plant species can complete one or more of their life cycles in space, and whether ch anges can be detected in subsequent generations after such long-term growth in orbit is given below, with experiments involving wheat being reviewed last. In 1982, Arabidopsis thaliana (Arabidopsis) plants completed for the first time a full life cycle in space while in a growth unit on the Salyut 7 space station. In this first example of seed to seed development, the siliques and seeds that had formed under spaceflight conditions were shorter and smaller respectively compared to thos e of the ground controls When the space seeds were subsequently grown on Earth, 42% develope d as normal seed-bearing plants as compared 126

PAGE 127

to 67.6% of the control seeds. The percentages of s eeds that did not germinate or that we re lethal at an early growth stage were both higher for th e space seeds than for the control seeds (Merkys et al., 1984). During the greenhouse 3 experime nt in 1997 on space station MIR, Brassica rapa L. (Brassica) completed two full life cycles and a th ird partial cycle (Musgrave et al., 2000). Seeds harvested from the first set of pl ants grown on MIR were planted al ongside original seeds for the second experiment. Second-generation space plants were significantly smaller than those grown from original seeds, while on the ground there was no distinction betw een the corresponding two control sets (Musgrave et al., 2000). With respect to late stages of seed development and maturation of Brassica in micr ogravity, the study found that fina l seed size was diminished, reserves were also stored in the form of starch rather than the expect ed protein and lipid, and ripening by the silique occurred in a basipetal manner, rather than simultaneously along the silique. The authors propose that these results might be explained by indirect effects of microgravity on the silique microenvironment (Mus grave et al., 2000). Altered gas movement in microgravity would be expected to change th e dynamics of oxygen and carbon dioxide exchange by tissues enclosed within these unique ga seous microenvironments (Kuang et al., 2005). Between March 2003 and April 2005, fi ve successive generations of Pisum sativum (dwarf pea) were grown in Lada on the International Sp ace Station (ISS). Characteristics such as plant height, number of pods per plant, number of peas per plant, dry plant biomass, and dry pea mass per plant did not differ significantly between th e plants in space and on the ground for the 1st generation (Sychev et al., 2007). Quantitative resu lts for the subsequent four generations were not given. Random Amplified Polymorphic DNA (RAPD) analysis, as well as an examination of 127

PAGE 128

chromosome aberrations, was performe d on plants grown from several first generation space and ground control seeds. Results did not reveal genetic polymorphisms (Sychev et al., 2007). The life cycle experiments with wheat that establish the foundation of the experiments presented here were initiated in 1991. Triticum aestivum L. cv. Super Dwarf plants nearly completed their full life cycle in the Svetobl ock M unit on the MIR station by forming three seed-heads. The plants were returned to Earth after 167 days to finish the maturation process, with two of the heads yielding a total of 28 mature seeds, most of which subsequently produced healthy plants and viable seeds (Mashinsky et al., 1994). In 1996, anothe r life cycle experiment (greenhouse 2) was carried out on MIR. An improved growth unit called the SVET-2 SG resulted in synchronous growth stages of wheat during its life cycle in space and on the ground (Ivanova et al., 1998). The plants in space formed sterile heads due to the concentration of ethylene in the MIR atmosphere (Ivanova et al ., 1998; Levinskikh et al ., 2000; Campbell et al., 2001; Bubenheim et al., 2003; Salisbur y et al., 2003). As a result, the Triticum aestivum L. cultivar USU-Apogee, which is more resistan t to ethylene, was used during the greenhouse 4 experiment in 1998 (Levinskikh et al., 2000; Sychev et al., 2001). Of the 52 Apogee seeds planted on orbit, 12 developed into mature plants that produced a total of 508 seeds. The number of seeds per plant was 38% lower than that of the ground control pl ants. 45 of the seeds collected in space were subsequently planted on Eart h. Growth after germination did not deviate significantly from plants grown from gr ound control seeds (Levinskikh et al., 2000). During the subsequent greenhouse 5 experiment in 1999, first generation space seeds were planted as well as seeds without a history in space (Levinskikh et al ., 2001). Only one plant grown from the first generation matured and pr oduced the second generation of space seeds (Levinskikh et al., 2001; Sychev et al., 2001). The plant of the second space generation was not 128

PAGE 129

morphologically different from those of the first generation or the ground c ontrols (Levinskikh et al., 2001; Sychev et al., 2001). Furthermore, plants grown on Earth from firstand secondgeneration space seeds, as well as from ground cont rol seeds, were also morphologically similar (Levinskikh et al., 2001). For the present study we return to the experiment in 1991 during which Super Dwarf wheat plants set seed on space station MIR and were then returned to Earth to finish maturation (Mashinsky et al., 1994). The seeds formed by the plants flown on MIR and their offspring will be referred to as MIR flight seeds and plan ts, and those formed by the related ground control plants and their offspring as MIR ground control seeds and plants. The first generation MIR flight and ground control seeds were grown for another two generations on the ground during a study that screened for possible aftereffects of spaceflight factors on plant growth and development (Mashinsky and Nechitailo, 1996; Ma shinsky et al., 1997). Some of the retrieved third generation MIR flight and ground control seeds were recently grown alongside an additional control set consisting of Apogee wheat in the soil-based closed ecological facility Laboratory Biosphere (Dempster et al., 2004; Nelson et al., 2008). The Apogee wheat will be referred to as Biosphere control plants. The goal of the presented work is to test whether changes in gene expression can be detected in plants that are three generations removed from long-term growth in space, compared to control plants that have not had a history in space. For this purpose, leaf tissue was harvested for gene expression analysis using specifically designed wheat microarrays. Results The MIR flight and Biosphere control plants were grown at the same time within the Laboratory Biosphere facility to produce leaf tissue for gene e xpression analysis (Figure 4-1). A comparison of gene expression between the pl ant sets was done via pooled reference RNA 129

PAGE 130

(Figure 4-2). The transcript levels of each gene were compared by running a series of t-tests. Results from the t-tests are presented in Figure 4-3 in the form of p-values (y-axis) plotted against fold change (x-axis). Diffe rences in transcript abundance la rger than 2-fold change were detected for 363 genes, and larger than 4-fold ch ange for 25 genes. For 593 genes, differences in expression corresponding to p-values < 0.05 were observed between the sets. The p-value indicates the probability that the observed differe nce in expression between the sets is falsely positive (type I error). For a single test, the hypothesis that a difference is significant is commonly accepted when the type I error is 5% or lower. However, when analyzing a large number of genes at the same time, such as with microarrays, the p-values of the individual tests need to be adjusted so that the overall inciden ce of false-positives is reduced (Bretz et al., 2005). Several methods are commonly used. The Bonf erroni method for example controls the familywise error rate (FWER), which is the prob ability of committing at least one type I error, while the Benjamini and Hochberg and Q-value me thods control the false discovery rate (FDR), which is related to the expected proportion of false-positives am ong all significant results (Bretz et al., 2005). The Benjamini and Hochberg and the Q-value methods are therefore less stringent with respect to type I error control than Bonferr oni, and allow the false ne gative rate to be kept relatively low. To ensure that few real differences are falsely marked as insignificant, we chose to correct with the Benjamini and Hochberg or Q-value method. After correcting for multiple test ing with Benjamini and Hochbe rg or Q-value, none of the differences between the fourth generation MIR flight and the Bi osphere control plants were significant at the p < 0.05 or q < 0.05 level (Figures 4-4, 4-5). These results show that a comparison of two biological replicates of both the experimental set and control set did not identify any statistically signi ficant differences in their patt erns of gene expression. This 130

PAGE 131

conclusion applies to the specifi c transcripts matching the 10,263 oligonucleotide probes on our arrays; with respect to the rem aining transcripts in the ge nome we cannot make any statements. The plants comprising the experi mental and control set are very closely related cultivars, although not identical. Sequence polymorphisms within a species can result in differential hybridization of transcripts to pr obes and thereby increase the in cidence of false positives (Kirst et al., 2006). Since we did not find any positives after applying the Benjamini and Hochberg and Q-value methods, we have not increased the stri ngency of our multiple testing correction in the direction of controlling fa lse positives, such as with the Bonferroni method, which means that we were able to maintain our false negative rate rela tively low. We are aware of approaches used to recapture potential fals e negatives that may exist despite a low false negative rate. For example Wu et al. (2008) first used a traditional statistica l analysis (ANOVA) and a multiple test correction to define a se t of 93 genes that showed highly si gnificant differences in expression levels between two strains of mi ce. By clustering analysis they then defined another 39 genes where the differences between the expression levels of the two strains did not reach the criteria of statistical significan ce (Wu et al., 2008). Since our analys is did not yield any significant changes in gene expression, we need not apply this strategy to our results. In summarizing the results of this analysis, we did not find differenc es in gene expression that were statistically significant between the MIR plants three generations removed from spaceflight exposure, and the Laboratory Bios phere ground control plants. Documents with detailed tables listing all the 10,263 probes, their related wheat target sequence, raw signal values as well as statistical results are available as supplemental data. Discussion This study has tested for the first time whethe r there are any herita ble changes in gene expression patterns in leaf tissue from plants three generations removed from spaceflight. 131

PAGE 132

Several groups have analyzed gene expression differences between spaceflight-exposed plants themselves, rather than their offspring, and their related ground controls. When 24-day-old Apogee wheat leaves harvested during a first ge n eration of spaceflight exposure were compared to leaves from control plants, Stutte et al. (2006) did not find differen ces in gene expression greater than 2-fold. In that case, leaf material harvested from two plants in space and on the ground was pooled per set and mRNA extracted fr om the two pools was subjected to a dye-swap microarray experiment (Stutte et al., 2006). Paul et al. (2005) exposed 7day-old Arabidopsis seedlings to spaceflight for 5 days before ha rvesting. Shoot material from approximately 20 Arabidopsis seedlings from two different Pl ant Growth Chambers was pooled for both the spaceflight and the control set. The resulting two samples were diffe rentially labeled and hybridized to a single microarray. The expressi on of 182 genes differed more than 4-fold between the spaceflight and ground control samples. Of these genes, 50 were expressed at moderate to high levels where a dditional confidence in fold-change values can be derived (Paul et al., 2005). It should be stressed that these st udies examined patterns of gene expression as a function of a metabolic reaction to spaceflight and do not reflect spaceflight induced changes to the genome. In contrast, the current study represents the first time that gene expression is used to screen for heritable changes in plants induced by long-term spaceflight. Wheat plants were exposed to the spaceflight environment during on e generation and compared to control plants three generations later. Although differences in gene expression larger than 2-fold or 4-fold change were detected, none of the differences were shown to be statistically significant after correction for multiple testing. Heritable changes in gene expression can be caused by changes to the base sequence of the genetic material (genetic mutations) or by chro matin and DNA modifications that do not involve 132

PAGE 133

base sequence alteration (epigenetics). The main potential cause of heritable m utations to which an organism is exposed in the spaceflight enviro nment is ionizing radiation. Ionizing radiation in low earth orbit consists of galactic cosmic rays (85% protons, 12% helium ions, 1% heavy ions of charge 3, 2% electrons and positrons), energetic electrons and protons trapped in Earths geomagnetic field, solar energetic particles (el ectrons, protons and heavier charged particles up to iron) and albedo neutrons and protons (Ben ton and Benton, 2001). Interactions between these particles and the atomic nuclei of the spacecra ft materials produce a variety of secondary particles, including protons, helium ions, neutrons and recoil heavy nuclei, that can be of higher linear energy transfer (charged particles) or radiation weight ing factors (neutrons) than the primary radiation flux (Benton et al., 2002). Mutation i nduction per charged particle increases with linear energy transfer up to about 200 keV/ m (Kiefer, 2002). On MIR, secondary particles from proton-induced target fragmentation interactions were found to be the largest contributor to the linear energy transfer (LET) spectrum above 100 keV/ m (Benton et al., 2002). In low earth orbit, mean dose rates of ionizing radiation have tended to be below 500 Gy/day, regardless of orbital inclination, solar cycle phase, and spacec raft orientation and shielding (Benton and Benton, 2001). When passing through cells, ionizi ng radiation may interact with DNA. This can result in molecular damage such as single strand break s, double strand breaks, base damage, and DNAprotein cross links (Nikjoo et al ., 1999). Double strand breaks in plant cells are mainly repaired via non-homologous end joining, which is associat ed with deletions and insertions of DNA sequence (Gorbunova and Levy, 1997). Early space biology studies used various indirect methods to indicate the possibility of mutations caused by ionizing radiation in satellites, space vehicles and space stations. These methods includ e the association of high charge and energy 133

PAGE 134

(HZE) particle tracts in plant seeds with ch rom osomal aberrations or visible phenotypes in seedlings grown from these seeds (Peterson et al., 1977; Bucker and Facius, 1981; Nevzgodina et al., 1984; Nevzgodina et al., 1989), and, more recently, random amplified polymorphic DNApolymerase chain reaction (Sychev et al., 2007 ). DNA-sequencing is a direct and detailed method used to detect mutations related to ioni zing radiation. For example, Arabidopsis mutants induced by irradiating pollen with gamma rays (150 Gy) or carbon ions (40 Gy) carried 1or 4-bp deletions, which were transmitted norma lly, or extremely large deletions of up to >6 Mbp, most of which were not transmitted to pr ogeny (Naito et al., 2005 ). Ionizing radiation studies involving plant or mammalian cells ofte n analyze the molecular damage resulting from relatively high irradiation doses (Schmidt and Kiefer, 1998; Nikj oo et al., 1999; Naito et al., 2005). After a 40-day exposure to 0.25-0.51 mGy/day on MIR, the averaged mutation frequency of the rpsL gene in yeast did not differ from that of the same gene in the ground control samples. However, the greater part of the Mir mutant samp les were found to have a total or large deletion in the rpsL sequence, suggesting that space ra diation containing high-LET might have caused deletion-type mutations (Fukuda et al., 2000). In the event of transmissible small intragenic deletions or insertions in plants growing in a space station, the likelihood of finding differences in gene expression between these plants, or their offspring, and the ground controls appears to be small. First of all, the small deletion or insertion would have to affect the binding of proteins involved in transcription of the particular gene. If this is the case, then the ploidy of th e plant species will determine the extent to which overall transcription levels of the gene or genes are changed. T. aestivum is a hexaploid species, which means that with potentially three functiona l copies of a gene, a change in expression of one allele may have only subtle effects overall (Dubcovsky and Dvor ak, 2007). For a self134

PAGE 135

compatible s pecies like T. aestivum the relative presence of the mutant allele in subsequent generations will be the same for the total population, but may differ in smaller sample populations. With a random distribu tion of affected alleles, this could result in a higher or lower incidence of mutant alleles in a small sample compared to the overa ll population. The same hypothetical scenario applies to potentially larger deletions of th e size of promoters or coding regions, in case they occur in low earth orbit and are transmissible. Heritable changes in gene expression can also be associated with changes in the methylation state of cytosine bases in DNA and the modificati on state of histone proteins (Henderson and Jacobsen, 2007). DNA and histone m odification in plants are interrelated and mediated in part by enzymes involved in addition or removal of methyl or acetyl groups and by small interfering RNAs (Chen and Tian, 2007; Henderson and Jacobsen, 2007). In tobacco plants, a close correlation was found between demethylation and expression of a tobacco gene encoding a glycerophosphodiesterase-like protei n upon exposure to abiotic stresses such as aluminum, salt and low temperature (Choi and Sano, 2007). For an environmentally triggered DNA and histone modification to be heritable, it would have to be transmitted over many rounds of mitotic DNA replication in sporophytic ti ssues, through the differentiation of gametophyte precursor cells, meiosis, and postmeiotic m itoses of haploid gametophytes (Takeda and Paszkowski, 2006). Identified environmental factors relevant to spaceflight, such as microgravity or ionizing radiation, have so far not been shown to cause heritable histone or DNA modifications changing gene expression in th e offspring of exposed plants. A study of Arabidopsis plants growing near Chernobyl in 1989 found that the genomic DNA of two subsequent generations grown under laboratory conditions was hypermethylated compared to control plants. Seeds collected on site several years later gave rise to offspring with less 135

PAGE 136

hypermethylated DNA (Kovalchuk et al., 2004 ). The authors suggested that DNA hypermethylation could be an imm e diate plant response to ionizi ng radiation. Aside from this observation, DNA and histone modifi cations initiated by environmental factors have to date not been reported to be inherite d epigenetically in plants. Conclusion This study is a first attempt to answer the question whether long-term exposure to the spaceflight environment in low earth orbit space stations can cause significant, heritable changes in gene expression patterns in plants. Some of the factors to which plants are exposed in a spaceflight environment that are potentially able to cause such differences are described in the discussion. Our analysis of plants three ge nerations removed from spaceflight exposure compared to plants with no exposure in their li neage indicate that expos ure to the spaceflight environment for one generation does not result in changes in gene expression that are heritable. Future space biology experiments addressing her itable changes in gene expression could use a larger number of biological replicates to ensure th at the false negative rate is indeed as low as possible. Furthermore, whole genome arrays coul d be used that contai n probes targeting all known transcripts. However, we do not imply to be able to foresee whether or in what way this may change the current outcome. The effects of still longer exposure to spaceflight, such as for several contiguous generations, on gene expression in exposed plants and their offspring could also be tested. Materials and Methods Plant Species and Cultivars Two closely related cultivars of wheat were used in this study: T. aestivum L. cv. Super Dwarf and T. aestivum L. cv. USU-Apogee. USU-Apogee originated from a cross between 'Parula' and 'Super Dwarf', both of which we re obtained from the CIMMYT germplasm 136

PAGE 137

collection in 1984. Parula was select ed for its small leaf size, and Super Dwarf was selected for its short stature (25 cm tall). USU-Apogee is a dwarf (45-50 cm tall), hard red spring wheat with a rapid development and resistance to Ca-induced leaf tip necrosis, lead ing to high yields in controlled environments such as bioregener ative life support system s in space (Bugbee and Koerner, 1997; B ugbee et al., 1997). MIR and Ground Experiments Triticum aestivum L. cv. Super Dwarf was grown in the Svetoblock M unit on MIR for 167 days in 1991. Ionite (ion exchange substrate) was used as a soil substitute. Light intensity ranged between 135 and 175 mol m-2 s-1. Air temperature was 22 2C and atmospheric CO2 was 6000 ppm. Three heads appeared in the boot (i.e. each surrounded by a leaf) in two plants while the plants were still on MIR. After return to Earth on Octobe r 10, 1991, three spikes developed from these, two of which contained a total of 28 seeds. The MIR ground control plants were grown in the Svetoblock M unit inside the Svet greenhouse at a laboratory in Moscow (Mashinsky et al., 1994). First ge neration MIR flight and ground c ontrol seeds were grown for an additional two generations on the ground durin g subsequent experiments on Earth. The experiments took place at the Institute of Genera l Genetics of the Russian Academy of Science. The plants were grown under the same light, temperature and CO2 conditions as the plants on MIR. Third generation seeds were harveste d in 1997 (Nechitailo, data not shown). Laboratory Biosphere Experiment Third generation MIR flight and ground cont rol seeds were planted in the Laboratory Biosphere facility in New Mexico alongside T. aestivum L. cv. USU-Apogee in February 2005. The lighting regime was 13 h light/11 h dark at a light intensity of 960 mol m-2 s-1, 45 mol m-2 day-1 supplied by high-pressure sodium lamps. Atmospheric CO2 ranged between 300 and 3000 ppm daily during the majority of the growing season. Temperat ures ranged from 21 to 29 C 137

PAGE 138

during light hours and from 20 to 24 C at night. Two of the thir d generation MIR flight seeds germinated, grew and matured, versus one of the third generation MIR ground control seeds. Two of the Apogee wheat plants were selected as additional controls that will be referred to as Biosphere controls. Sample Preparation Leaves were harvested from the two fourth-g eneration MIR flight plants, the fourthgeneration MIR ground control plant, and the two Biosphere c ontrol plants. Tissues were immediately stored in RNAlater. Two leaves per plant were pooled and total RNA was extracted with the Qiagen RNeasy Plant Mini Kit. Aliquots of the MIR flight, Biosphere control, and MIR ground control RNA samples were mixed to obtain a reference RNA sample. Microarray Experimental Design and Data Analysis Custom-made long-oligonucleotide (60-mer) wh eat microarrays were developed by the Interdisciplinary Center for Bi otechnology Research (Gainesville, FL) in collaboration with Agilent Technologies (Foster City, CA). A set of 10,263 60-mer wheat specific probes was designed based on Triticum aestivum L. cv. Apogee EST sequences and in situ synthesized on microarrays (2x22K format). The samples were amplified and labeled by converting target mRNA to cRNA according to the manufacturers protocol (Low RNA Input Fluorescent Linear Amplification Kit Protocol [v. 4] Agilent Technologies). The two biological replicates from the MIR flight and Biosphere control sample sets and the one MIR ground control sample, were Cy5-labeled and hybridized to the oligonucleo tide microarrays with Cy3-labeled reference cRNA. After hybridization, slides were washed, scanned and the data extracted according to standard procedures (Two-Color Microarray-Based Gene Expression Analysis [v.5.5], Agilent Technologies). Because there was only one biol ogical replicate available for the MIR ground 138

PAGE 139

139 control sample, it was excluded from the subsequent statistical analysis and will not be discussed further. Median signal intensities, detected by scanning the micr oarrays with the MIR flight, Biosphere control and reference samples hybridized to the probes, were analyzed using a twostep strategy (Chu et al., 2002; Hsieh et al., 2003). Initially, pr obe measurements were centered relative to the microarray mean and log2 transformed. The signals of the MIR flight, Biosphere control and reference samples were then evaluated in a mixed analysis of variance (ANOVA) model, and unbiased estimates of transcript abundance (least-square means) were generated for each sample set. Microarray slide was includ ed in the ANOVA model as a random effect to account for the covariance of samples hybridized to the same microarray (i.e. control for spot effect) (Jin et al., 2001; Wolfinge r et al., 2001). The transcript leve l estimates of the MIR flight and the Biosphere control samples were then comp ared using a series of t-tests. The p-values generated in this comparison were corrected for multiple testing using the Benjamini and Hochberg procedure in the Multtest package of the R software (Benjamini and Hochberg, 1995), and the Q-value procedure in the Q-value 1.0 package (default setti ngs) of the R software (Storey and Tibshirani, 2003).

PAGE 140

Figure 4-1. Wheat growth in the Laboratory Bios phere. Photographs of individual wheat plants growing in the Laboratory Biosphere from which leaves were collected for gene expression analysis. Top row: fourth generation MIR fli ght plants A and B, and fourth generation MIR ground control plant C ( T. aestivum cv, Super Dwarf). Bottom row: Biosphere control plants D and E ( T. aestivum cv. USU-Apogee). Gene expression patterns were compared between the MIR flight and the Biosphere control sets by using microarrays and statistical analysis. 140

PAGE 141

Figure 4-2. Flowchart of wheat e xperiments. Flowchart showing an overview of the history of the plant samples comprising the MIR flight and Biosphere control sets, as well as a diagram of how the experimental and refe rence cRNA samples were hybridized to the microarrays. 141

PAGE 142

Figure 4-3. ANOVA (t-test) results. ANOVA (t-test) p-values not corrected for multiple testing are plotted against the corre sponding log2 fold changes in transcripts between the Biosphere control and MIR f light sets; p-value = 0.05 is indicated by the horizontal line with the label on the right. The scatter plot shows the distri bution of genes whose log2 differential expression levels have pvalues smaller than 0.05 before correcting for false positives. 142

PAGE 143

Figure 4-4. Multiple testing correction: Benj amini and Hochberg method. ANOVA (t-test) pvalues corrected for multiple testing with Benjamini and Hochberg are plotted against the corresponding log2 fold changes in tran scripts between the Biosphere control and MIR flight sets; corrected p-value = 0.05 is indicated by the horizon tal line with the label on the right. The scatte r plot shows that none of the corrected p-values associated with the 10.263 comparisons in gene expression are smaller than 0.05. 143

PAGE 144

144 Figure 4-5. Multiple testing corr ection: Q-value method. ANOVA (t-test) p-values corrected for multiple testing with Q-value are plotted against the corresponding log2 fold changes in transcripts between the Biosphere cont rol and MIR flight se ts; q-value = 0.05 is indicated by the horizontal line with the label on the right. The scatter plot shows that none of the corrected p-values (= q-values ) associated with the 10.263 comparisons in gene expression are smaller than 0.05.

PAGE 145

CHAPTER 5 CONCLUSIONS Manned m issions to Mars demand the efficient use of local planetary resources and the recycling of limited materials such as water, pressurized atmosphere and organic matter while producing food (Barta and Henninger, 1994). The use of in situ regolith for plant growth in a future bioregenerative life support system on Ma rs may have several advantages over hydroponic systems (Schuerger et al., 2002). These include th e immediate bioavailabili ty of plant essential ions, low-tech mechanical support for plants, an d easy access of in situ materials once on the surface. However, plant growth may be reduced or inhibited by substances in the regolith, such as high levels of hydrated magnesium sulfate minerals (Chapter 1 and 2). In a potential bioregenerative life support system on Mars, an excess of a particular element in the crews diet could affect the presence and av ailability of other required elemen ts. This study therefore focuses on the possibility of reducing accumulation of Mg2+ and SO4 2ions within the plant as a method to enhance plant tolerance to high levels of magnesium sulfate in the growth medium. Arabidopsis is a model species in plant mol ecular biology research and its genome is fully sequenced. Plasma membrane loca lized efflux transporters of Mg2+ and SO4 2ions have not been identified to date in the outer root cell layers of Arabidopsis or other plant species. AtMRS2-10 and AtSULTR1;2 are genes encoding a known Mg2+ and SO4 2uptake transporter respectively. Arabidopsis lines carrying knockout T-DNA in sertion mutations in AtMRS2-10 and AtSULTR1;2 did not mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype Arab idopsis plants (Chapter 2). An Arabidopsis line carrying a knockout mutation of the vacuolar CAX1 gene ( cax1-1 ) showed a significant improvement in growth on soil treated w ith high levels of MgSO4H2O in solution (Chapter 2). A reduction in leaf magnesium content in cax1 mutants compared to wildtype (0.7 below normal) was 145

PAGE 146

reported in a previous study (Brads haw, 2005). Although th e Mg content of cax1 mutant roots has not been analyzed so far, th e reduced levels of Mg in leaves indicate that Arabidopsis CAX1 knockout mutants are in line with our objective of identifying Arabidopsis variants that show improved tolerance of high magne sium sulfate by at least partly limiting accumulation of Mg2+ or SO4 2ions within the plant. Therefore, genes in crop species encoding transporter proteins similar in function to the protein encoded by CA X1 in Arabidopsis are proposed candidate genes for enhancing tolerance of crop plant species to regolith high in soluble magnesium sulfate minerals used in an advanced life support system on Mars. The identification by Bradshaw (2005) of a CAX1 knockout mutant tolerant of low Ca:Mg ratios in solution, which are characteristic of serpentine soil so lutions, confirms the appropriateness of serpentine soils as partial anal ogue soils on Earth for regolith high in soluble magnesium sulfate minerals on Mars. Leaf Ca:Mg molar ratios of nonserpentine plant species are generally equal to that of the soil, while serp entine species maintain significantly higher leaf Ca:Mg than both their nonserpentine counterpart s and the soil (O'Dell et al., 2006). The authors conclude that elevated leaf Ca:Mg in the serpentine species was achieved by selective Ca2+ transport and/or Mg2+ exclusion operating at the root-to-shoot translocati on level, as root Ca and Mg concentrations did not differ between se rpentine and nonserpentine species. Genetic differentiation between populations of Arabidopsis lyrata growing on granitic or serpentinic soils was measured by using an Arabidopsis thaliana tiling array that has 2.85 million probes throughout the genome (Turner et al., 2008). Th e study found significant overrepresentation of genes involved in ion transport, and one gene in particular, calcium-exchanger 7 (CAX7), was presented as an excellent candidate gene for adaptation to low Ca:Mg ratios in A. lyrata It is currently not known which transpor ters or their regulators could be involved in the observed 146

PAGE 147

lower levels of Mg or higher Ca:Mg ratios in leaves from Arabidopsis cax1-1 and serp entine species respectively. Results from the anal ysis of genetic differentiation between A. lyrata populations include an overrepresen tation of genes encoding proteins involved in ion transport. Some of these transporters might play a role in reducing Mg2+ accumulation within the plant. This study is the first to document genome-w ide plant root transcriptome responses to elevated levels of magnesium sulfate based on th e high Mg:Ca ratio that can occur in serpentine soils using microarrays (Chapter 3). The objective was to analyz e which genes are differentially expressed as part of the primary stress respons e in roots of a non-tole rant species such as wildtype Arabidopsis thaliana ( col-0 ) compared to unexposed col-0 roots. This could lead to identification of candidate gene s in Arabidopsis with potential to enhance tolerance to high magnesium sulfate by limiting accumulation within the plant. The cax1-1 mutant was also exposed to elevated MgSO4H2O to determine which genes are differentially expressed in the CAX1 knockout mutant background compared to exposed col-0 Genes that are differentially expressed between the genotypes could point to some of the downstream molecular processes eventually leading to enhanced tolerance for cax1-1 at the whole plant level, including reduced leaf Mg content and increased fres h weight biomass, after days or weeks of exposure in agar or soil medium. Some of the tran scripts involved in downstream processes may themselves be candidates for enhanced tolerance, such as t hose encoding (regulators of) plasma membrane based channels that transport Mg2+. Transcriptome responses of Arabidopsis col-0 roots exposed to high magnesium sulfate for 45 min. compared to col-0 exposed to a control solution for 45 min. (Time 1) reveal over 300 differentially expressed genes. Genes of known function include those encoding calcium-binding proteins, kinases, transcripti on factors, enzymes involved in hormone metabolism, disease 147

PAGE 148

resistance proteins and many cell wall related pr oteins. The responses of the genes encoding cell wall related proteins indicate a possibl e reduction in root growth when col-0 is exposed to high concentrations of magne sium sulfate. Some of the genes of known or unknown function were previously associated with specific or broad ra nges of abiotic stresses, but not necessarily in roots or at these time points. Over 200 genes encoding membrane based transporters were differentially expressed across the col-0 time series. The expression of genes encoding known plasma membrane based importers of Mg2+ and SO4 2ions, such as MRS2-10, SULTR1;1 or SULTR1;2, was not down-regulated. This corresponds with the observation reported in Chapter 2 that Arabidopsis lines ca rrying knockout T-DNA insertion mutations in AtMRS2-10 and AtSULTR1;2 did not mitigate the constraining impacts of high magnesium sulfate concentrations on wildtype Arab idopsis plants. The differential expression of genes encoding known tonoplast localized transporters of Mg2+ and SO4 2ions indicate a possible storage of excess Mg2+ and SO4 2ions in the vacuole. Future rese arch can reveal whether any of the differentially expressed transporter genes of unknown protein localization, protein function, or both, are candidates to enha nce tolerance to high levels of so luble magnesium sulfate minerals in Martian regolith by reducing accumulation of Mg2+ and SO4 2ions within the plant. For example, the localization of sulfate transpor ters SULTR3;1 and SULTR3;4 within Arabidopsis root tissue and cells can be analyzed in follow-up studies to see if the genes encoding these transporters are candidates for enhanced tolera nce. Potential regulators of membrane based transporter activity, such as kinases, which ar e encoded by differentially expressed genes across the col-0 time series, could also be analyzed. Sinc e gene expression differences are not fully controlled for diurnal effects for the Time 2 and 3 comparisons in the col-0 time series, these effects will have to be ruled out in follow-up studies. 148

PAGE 149

The down-regulation of cax1-1 gene expression is a natura l response to high magnesium sulfate in co l-0 that is already seen at Time 1. Togeth er with the down-regulation of CAX2 and CAX3 gene expression at later time points it in dicates a possible shortage of calcium in the cytosol experienced by col-0 when exposed to high concentrations of magnesium sulfate. Only three transcripts were differentially expressed between cax1-1 and col-0 at 3 hours after initiation of treatment. Follow-up experiments could be do ne to discover the function of these three transcripts. The root transcriptome of cax1-1 and col-0 could furthermore be compared at later time points after initial exposure to high magnesium sulfate to reveal additional differentially expressed transcripts that could indicate the molecular processes eventually leading to the tolerance difference exhibited by these genotypes after days or we eks of growth. Some of the transcripts involved in downstream processes may themselves be candidates for enhanced tolerance, such as those encodi ng (regulators of) plasma membrane based channels that transport Mg2+. Crop plants might be grown for multiple life cy cles as part of a life support system in a transport vehicle during a human mi ssion to Mars. Within the tran sport vehicle, crop plants will be exposed to lower gravitational forces and hi gher radiation fluxes co mpared to a growth chamber on Earth. Some of the seeds harvested fro m these plants could subsequently be planted in advanced life support systems on the surface of Mars. To analyze in detail whether plants grown in spaceflight conditions can show changes in subsequent generations compared to control plants, gene expression in wheat plants that are three generations removed from growth in the MIR space station was compared to gene expr ession in wheat plants with no spaceflight exposure in their lineage (Chapter 4). Our gene e xpression analysis results indicate that exposure to the spaceflight environment for one generation does not result in changes in gene expression 149

PAGE 150

150 that are heritable. Future space biology experi ments addressing heritable changes in gene expression could use a larger number of biological replicates to ensu re that the false negative rate is indeed as low as possible. Furthermore, w hole genome arrays could be used that contain probes targeting all known transcripts. However, we do not imply to be able to foresee whether or in what way this may change the current ou tcome. The effects of still longer exposure to spaceflight, such as for several contiguous generati ons, on gene expression in exposed plants and their offspring could also be tested.

PAGE 151

LIST OF REFERENCES Alonso JM, Stepanova AN (2003) T-DNA mutagene sis in Arabidopsis. Methods Mol Biol 236: 177-188 Arvidson RE, Poulet F, Bibring JP, Wol ff M, Gendrin A, Morris RV, Freeman JJ, Langevin Y, Mangold N, Bellucci G (2005) Spectral reflectance and morphologic correlations in eastern Terra Meridiani, Mars. Science 307: 1591-1594 Banin A, Han FX, Kan I, Cicelsky A (1997) Acidic volatiles and the Mars soil. J. Geophys. Res. 102: 13,341-313,356 Barber SA (1995) Soil nutrient bioavaila bility: a mechanistic approach, Ed 2nd. John Wiley and Sons, Inc Barta DJ, Henninger DL (1994) Regenerative life support systems: why do we need them? Adv Space Res 14: 403-410 Baxter I, Ouzzani M, Orcun S, Kennedy B, Jandhyala SS, Salt DE (2007) Purdue ionomics information management system: an integr ated functional genomics platform. Plant Physiol 143: 600-611 Benjamini Y, Hochberg Y (1995) Controlling the false discove ry rate: a practical and powerful approach to multiple testing. Journa l of the royal statistical society 57: 289-300 Benton ER, Benton EV (2001) Space radiation dosimetry in low Earth orbit and beyond. Nucl Instrum Methods Phys Res B 184: 255-294 Benton ER, Benton EV, Frank AL, Leonov A, Gaskin J (2002) Dosimetric results from the Mir orbital station. Ra diat Prot Dosimetry 100: 489-494 Bibring JP, Langevin Y, Mustard JF, Poulet F, Arvidson R, Gendrin A, Gondet B, Mangold N, Pinet P, Forget F, Berthe M, Bibring JP, Gendrin A, Gomez C, Gondet B, Jouglet D, Poulet F, Soufflot A, Vincendon M, Combes M, Drossart P, Encrenaz T, Fouchet T, Merchiorri R, Belluci G, Altieri F, Formisano V, Capaccioni F, Cerroni P, Coradini A, Fonti S, Korablev O, Kottsov V, Ignatiev N, Moroz V, Titov D, Zasova L, Loiseau D, Mangold N, Pinet P, Doute S, Schmitt B, Sotin C, Hauber E, Hoffmann H, Jaumann R, Keller U, Arvidson R, Mustar d JF, Duxbury T, Forget F, Neukum G (2006) Global mineralogical and aqueous mars history derived from OMEGA/Mars Express data. Science 312: 400-404 Bradshaw HD, Jr. (2005) Mutations in CAX1 produce phenotypes char acteristic of plants tolerant to serpentine soils. New Phytol 167: 81-88 Bretz F, Landgrebe J, Brunner E (2005) Multiplicity issues in microarray experiments. Methods Inf Med 44: 431-437 Brooks RR (1987) Serpentine and its ve getation. Dioscorides Press 151

PAGE 152

Brown AH, Chapman DK, Johnsson A, Heathcote D (1995) Gravitropi c responses of the Avena coleoptile in space and on clinostats. I. Gravitropic response thresholds. Physiol Plant 95: 27-33 Bubenheim DL, Stieber J, Campbell WF, Salisbury FB, Levinski M, Sytchev V, Podolsky I, Chernova L, Pdolsky I (2003) Ind uced abnormality in Mirand Earth-grown Super Dwarf wheat. Adv Space Res 31: 229-234 Buchner P, Takahashi H, Hawkesford MJ (2004) Plant sulphate tran sporters: co-ordination of uptake, intracellular and long-di stance transport. J Exp Bot 55: 1765-1773 Bucker H, Facius R (1981) The role of HZE particles in spaceflight: results from spaceflight and ground-based experiments. Acta Astronaut 8: 1099-1107 Bugbee B, Koerner G (1997) Yield comparisons and unique characteristics of the dwarf wheat cultivar 'USU-Apogee'. Adv Space Res 20: 1891-1894 Bugbee B, Koerner G, Albrechts en R, Dewey W, Clawson S (1997) Registration of cultivars: registration of 'USU-Apogee' wheat. Crop Sci 37: 626 Campbell WF, Salisbury FB, Bugbee B, Klassen S, Naegle E, Strickland DT, Bingham GE, Levinskikh M, Iljina GM, Veselova TD Sytchev VN, Podolsky I, McManus WR, Bubenheim DL, Stieber J, Jahns G (2001) Comparative floral development of Mirgrown and ethylene-treated, Earth-grown Super Dwarf wheat. J Plant Physiol 158: 10511060 Chen ZJ, Tian L (2007) Roles of dynamic and reversib le histone acetylation in plant development and polyploidy. Biochim Biophys Acta 1769: 295-307 Cheng NH, Pittman JK, Barkla BJ, Shigaki T, Hirschi KD (2003) The Arabidopsis cax1 mutant exhibits impaired ion homeostasis, development, and hormonal responses and reveals interplay among vacuolar transporters. Plant Cell 15: 347-364 Cheng NH, Pittman JK, Shigaki T, Lachmans ingh J, LeClere S, Lahner B, Salt DE, Hirschi KD (2005) Functional association of Arabi dopsis CAX1 and CAX3 is required for normal growth and ion homeostasis. Plant Physiol 138: 2048-2060 Choi CS, Sano H (2007) Abiotic-stress induces demethylat ion and transcriptional activation of a gene encoding a glycerophosphodiesterase-lik e protein in tobacco plants. Mol Genet Genomics 277: 589-600 Chou IM, Seal RR, 2nd (2003) Determination of epsom ite-hexahydrite equilibria by the humidity-buffer technique at 0.1 MPa with imp lications for phase equilibria in the system MgSO4-H2O. Astrobiology 3: 619-630 152

PAGE 153

Christensen PR, Wyatt MB, Glotch TD, Roger s AD, Anwar S, Arvidson RE, Bandfield JL, Blaney DL, Budney C, Calvin WM, Fallacaro A, Fergason RL, Gorelick N, Graff TG, Hamilton VE, Hayes AG, Johnson JR, K nudson AT, McSween HY, Jr., Mehall GL, Mehall LK, Moersch JE, Morris RV, Smith MD, Squyres SW, Ruff SW, Wolff MJ (2004) Mineralogy at Meridiani Planum from the Mini-TES experiment on the Opportunity Rover. Science 306: 1733-1739 Chu TM, Weir B, Wolfinger R (2002) A systematic statistical linear modeling approach to oligonucleotide array experiments. Math Biosci 176: 35-51 Ciftci-Yilmaz S, Mittler R (2008) The zinc finger networ k of plants. Cell Mol Life Sci 65: 1150-1160 Ciftci-Yilmaz S, Morsy MR, Song L, Co utu A, Krizek BA, Lewis MW, Warren D, Cushman J, Connolly EL, Mittler R (2007) The EAR-motif of the Cys2/His2-type zinc finger protein Zat7 plays a key role in the defense response of Arabidopsis to salinity stress. J Biol Chem 282: 9260-9268 Cosio C, Dunand C (2009) Specific functions of individual class III peroxidase genes. J Exp Bot 60: 391-408 Cui H, Benfey PN (2009) Interplay between SCARECROW, GA and LIKE HETEROCHROMATIN PROTEIN 1 in ground tissu e patterning in the Arabidopsis root. Plant J Curie C, Briat JF (2003) Iron transport a nd signaling in plants. Annu Rev Plant Biol 54: 183206 Czechowski T, Stitt M, Altmann T, Udvardi MK, Scheible WR (2005) Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol 139: 5-17 Dempster WF, Van Thillo M, Alling A, Allen JP, Silverstone S, Nelson M (2004) Technical review of the Laboratory Biosphere closed ecological system facility. Adv Space Res 34: 1477-1482 Deng W, Luo K, Li D, Zheng X, Wei X, Sm ith W, Thammina C, Lu L, Li Y, Pei Y (2006) Overexpression of an Arabidopsis magnesium transport gene, AtMGT1, in Nicotiana benthamiana confers Al tolerance. J Exp Bot 57: 4235-4243 Drysdale AE, Ewert MK, Hanford AJ (2003) Life support approach es for Mars missions. Adv Space Res 31: 51-61 Dubcovsky J, Dvorak J (2007) Genome plasticity a key factor in the success of polyploid wheat under domestication. Science 316: 1862-1866 153

PAGE 154

Edmond C, Shigaki T, Ewert S, Nelson MD Connorton JM, Chalova V, Noordally Z, Pittman J K (2009) Comparative analysis of CAX2-like cation transporters indicates functional and regulator y diversity. Biochem J 418: 145-154 Ernst WH (1998) Sulfur metabolism in higher plants: potential for phytoremediation. Biodegradation 9: 311-318 Ernst WHO (1997) Life-history syndromes and the ecology of plants from high sulphurhabitats. In WJ Cram, LJ De Kok, I Stulen, C Brunold, H Rennenberg, eds, Sulphur metabolism in higher plants: molecular, ecophysiological and nutritional aspects. Backhuys Publishers, pp 131-146 Fanning DS, Burch SN (2006) Sulfate and sulfide minera ls. Encyclopedia of Soil Science 1: 1714-1716 Fontes EP, Santos AA, Luz DF, Waclawovsky AJ, Chory J (2004) The geminivirus nuclear shuttle protein is a virulence factor that s uppresses transmembrane receptor kinase activity. Genes Dev 18: 2545-2556 Frigerio M, Alabadi D, Perez-Gomez J, Garci a-Carcel L, Phillips AL, Hedden P, Blazquez MA (2006) Transcriptional regula tion of gibberellin metabolis m genes by auxin signaling in Arabidopsis. Plant Physiol 142: 553-563 Fukuda T, Fukuda K, Takahashi A, Ohnishi T, Nakano T, Sato M, Gunge N (2000) Analysis of deletion mutations of the rpsL gene in the yeast Saccharomyces cerevisiae detected after long-term flight on the Russian space station Mir. Mutat Res 470: 125-132 Gellert R, Rieder R, Anderson RC, Bruckn er J, Clark BC, Dreibus G, Economou T, Klingelhofer G, Lugmair GW, Ming DW, Squyres SW, D'Uston C, Wanke H, Yen A, Zipfel J (2004) Chemistry of rocks and soils in Gusev Crater from th e Alpha Particle Xray Spectrometer. Science 305: 829-832 Gendrin A, Mangold N, Bibring JP, Langevin Y, Gondet B, Poulet F, Bonello G, Quantin C, Mustard J, Arvidson R, LeMouelic S (2005) Sulfates in Martian layered terrains: the OMEGA/Mars Express view. Science 307: 1587-1591 Golden DC, Ming DW, Morris RV, Mertzman SA (2005) Laboratory-simulated acid-sulfate weathering of basaltic materials: implications for formation of sulfates at Meridiani Planum and Gusev crater, Mars. J. Geophys. Res. 110 Gorbunova V, Levy AA (1997) Non-homologous DNA end joini ng in plant cells is associated with deletions and filler DNA insertions. Nucleic Acids Res 25: 4650-4657 154

PAGE 155

Haskin LA, Wang A, Jolliff BL, McSween HY, Clark BC, Des Marais DJ, McLennan SM, Tosca NJ, Hurowi tz JA, Farmer JD, Yen A, Squyres SW, Arvidson RE, Klingelhofer G, Schroder C, de Souza PA, Jr., Ming D W, Gellert R, Zipfel J, Bruckner J, Bell JF, 3rd, Herkenhoff K, Christensen PR, Ruff S, Blaney D, Gorevan S, Cabrol NA, Crumpler L, Grant J, Soderblom L (2005) Water alteration of rocks and soils on Mars at the Spirit rover site in Gusev crater. Nature 436: 66-69 Hawkesford MJ (2003) Transporter gene families in pl ants: the sulphate transporter gene family; redundancy or specialization? Physiol Plant 117: 155-163 Henderson IR, Jacobsen SE (2007) Epigenetic inheritance in plants. Nature 447: 418-424 Herbette S, Taconnat L, Hugouvieux V, Pie tte L, Magniette ML, Cuine S, Auroy P, Richaud P, Forestier C, Bourguignon J, Renou JP, Vavasseur A, Leonhardt N (2006) Genome-wide transcriptome profiling of the ear ly cadmium response of Arabidopsis roots and shoots. Biochimie 88: 1751-1765 Hsieh WP, Chu TM, Wolfinger RD, Gibson G (2003) Mixed-model reanalysis of primate data suggests tissue and species biases in oligonuc leotide-based gene expression profiles. Genetics 165: 747-757 Ivanova TN, Kostov PT, Sapunova SM, Dandolov IW, Salisbury FB, Bingham GE, Sytchov VN, Levinskikh MA, Podolski IG, Bubenheim DB, Jahns G (1998) Six-month space greenhouse experiments: a step to creation of future biological life support systems. Acta Astronaut 42: 11-23 Jiang Y, Deyholos MK (2006) Comprehensive transcriptional profiling of NaCl-stressed Arabidopsis roots reveals novel classes of responsive genes. BMC Plant Biol 6: 25 Jin W, Riley RM, Wolfinger RD, White KP, Passador-Gurgel G, Gibson G (2001) The contributions of sex, genotype and ag e to transcriptional variance in Drosophila melanogaster Nat Genet 29: 389-395 Kataoka T, Watanabe-Takahashi A, Hayashi N, Ohnishi M, Mimura T, Buchner P, Hawkesford MJ, Yamaya T, Takahashi H (2004) Vacuolar sulf ate transporters are essential determinants control ling internal distribution of su lfate in Arabidopsis. Plant Cell 16: 2693-2704 Kiefer J (2002) Mutagenic effects of heavy charged particles. J Radiat Res (Tokyo) 43 Suppl: S21-25 Kirst M, Caldo R, Casati P, Tanimoto G, Walbot V, Wise RP, Buckler ES (2006) Genetic diversity contribution to errors in short oligonucleot ide microarray analysis. Plant Biotechnol J 4: 489-498 155

PAGE 156

Klingelhofer G, Morris RV, Bernhardt B, Schroder C, Rodionov DS, de Souza PA, Jr., Yen A, Gellert R, Evlanov EN, Zubkov B, Foh J, Bonnes U, Kankeleit E, Gutlich P, Ming DW, Renz F, Wdowiak T, Squyres SW, Arvidson RE (2004) Jarosite and hematite at Meridiani Planum from Opportunity's Mossbauer Spectrometer. Science 306: 1740-1745 Kovacik J, Klejdus B (2008) Dynamics of phenolic acid s and lignin accumulation in metaltreated Matricaria chamomilla roots. Plant Cell Rep 27: 605-615 Kovalchuk I, Abramov V, Pogribny I, Kovalchuk O (2004) Molecular aspects of plant adaptation to life in the Ch ernobyl zone. Plant Physiol 135: 357-363 Kruckeberg AR (1999) Serpentine barrens of western North America. In RC Anderson, JS Fralish, JM Baskin, eds, Savannas, barrens, a nd rock outcrop plant communities of North America. Cambridge University Press, pp 309-321 Kuang A, Popova A, McClure G, Musgrave ME (2005) Dynamics of storage reserve deposition during Brassica rapa L. pollen and seed development in microgravity. Int J Plant Sci 166: 85-96 Kufner I, Koch W (2008) Stress regulated members of the plant organic cation transporter family are localized to the v acuolar membrane. BMC Res Notes 1: 43 Kwasniewski M, Szarejko I (2006) Molecular cloning and char acterization of beta-expansin gene related to root hair formation in barley. Plant Physiol 141: 1149-1158 Langevin Y, Poulet F, Bibring JP, Gondet B (2005) Sulfates in the north polar region of Mars detected by OMEGA/Mars Express. Science 307: 1584-1586 Laurie S, Feeney KA, Maathuis FJ, Heard PJ, Brown SJ, Leigh RA (2002) A role for HKT1 in sodium uptake by wheat roots. Plant J 32: 139-149 Lejay L, Gansel X, Cerezo M, Tillard P, Mu ller C, Krapp A, von Wiren N, Daniel-Vedele F, Gojon A (2003) Regulation of root ion tran sporters by photosynthesis: functional importance and relation with hexokinase. Plant Cell 15: 2218-2232 Levinskikh MA, Sychev VN, Derendiaeva TA Signalova OB, Podol'skii IG, Avdeev SV, Bingheim GE (2001) [Growth and development of pl ants in a row of generations under the conditions of spaceflight (experime nt Greenhouse-5)]. Aviakosm Ekolog Med 35: 4549 Levinskikh MA, Sychev VN, Derendiaeva TA Signalova OB, Podol'skii IG, Padalka GI, Avdeev SV, Bingham GE (2000) [Growth of wheat from seed-to-seed in spaceflight]. Aviakosm Ekolog Med 34: 44-49 Levinskikh MA, Sychev VN, Derendyaeva TA, Signalova OB, Salisbury FB, Campbell WF, Bingham GE, Bubenheim DL, Jahns G (2000) Analysis of the spaceflight effects on growth and development of Super Dwarf wh eat grown on the Space Station Mir. J Plant Physiol 156: 522-529 156

PAGE 157

Li L, Tutone AF, Drummond RS, Gardner RC, Luan S (2001) A novel family of m agnesium transport genes in Arabidopsis. Plant Cell 13: 2761-2775 Liu YB, Lu SM, Zhang JF, Liu S, Lu YT (2007) A xyloglucan endot ransglucosylase/hydrolase involves in growth of primary root and alters the deposition of cellulose in Arabidopsis. Planta 226: 1547-1560 Ma S, Bohnert HJ (2007) Integration of Arabidopsis thaliana stress-related transcript profiles, promoter structures, and cell-sp ecific expression. Genome Biol 8: R49 Ma W, Ali R, Berkowitz GA (2006) Characterization of plant phenotypes associated with lossof-function of AtCNGC1, a plant cyclic nucleotide gated ca tion channel. Plant Physiol Biochem 44: 494-505 Maathuis FJ, Filatov V, Herzyk P, Krijger GC, Axelsen KB, Chen S, Green BJ, Li Y, Madagan KL, Sanchez-Fernandez R, Forde BG, Palmgren MG, Rea PA, Williams LE, Sanders D, Amtmann A (2003) Transcriptome analysis of root transporters reveals participation of multiple gene families in the response to cation stress. Plant J 35: 675-692 Madson M, Dunand C, Li X, Verma R, Va nzin GF, Caplan J, Shoue DA, Carpita NC, Reiter WD (2003) The MUR3 gene of Ar abidopsis encodes a xyloglucan galactosyltransferase that is evolutionarily related to animal exostosins. Plant Cell 15: 1662-1670 Magnan F, Ranty B, Charpenteau M, Sotta B, Galaud JP, Aldon D (2008) Mutations in AtCML9, a calmodulin-like protein from Arabidopsis thaliana alter plant responses to abiotic stress and abscisic acid. Plant J 56: 575-589 Magome H, Yamaguchi S, Hanada A, Kamiya Y, Oda K (2008) The DDF1 transcriptional activator upregulates expression of a gi bberellin-deactivating gene, GA2ox7, under highsalinity stress in Arabidopsis. Plant J 56: 613-626 Mao DD, Tian LF, Li LG, Chen J, Deng PY, Li DP, Luan S (2008) AtMGT7: An Arabidopsis Gene Encoding a Low-Affinity Magnesiu m Transporter. J Integr Plant Biol 50: 1530-1538 Marschner H (1995) Mineral nutrition of higher plan ts, Ed second. Academic Press, London Martinoia E, Massonneau A, Frangne N (2000) Transport processe s of solutes across the vacuolar membrane of higher plants. Plant Cell Physiol 41: 1175-1186 Maruyama-Nakashita A, Inoue E, Watanab e-Takahashi A, Yamaya T, Takahashi H (2003) Transcriptome profiling of sulf ur-responsive genes in Arabidopsis reveals global effects of sulfur nutrition on multiple metabolic pathways. Plant Physiol 132: 597-605 Maruyama-Nakashita A, Nakamura Y, Tohge T, Saito K, Takahashi H (2006) Arabidopsis SLIM1 is a central transcripti onal regulator of plant sulfur response and metabolism. Plant Cell 18: 3235-3251 157

PAGE 158

Mashinsky AL, Ivanova I, Derendyae va T, Nechitailo GS, Salisbury F (1994) "From seed-toseed" experim ent with wheat plants under spaceflight conditions. Adv Space Res 14: 13-19 Mashinsky AL, Nechitailo GS (1996) Ground-based experiments with plants on modelling the action of gravity on the organism. In 47th International Astronautical Congress, Beijing, China Mashinsky AL, Nechitailo GS, Archipov MV (1997) Aftereffect of spaceflight factors in plants. In 48th International Astronau tical Congress, Turin, Italy Maurel C (2007) Plant aquaporins: novel functions and regulation properties. FEBS Lett 581: 2227-2236 McCormack E, Tsai YC, Braam J (2005) Handling calcium signaling: Arabidopsis CaMs and CMLs. Trends Plant Sci 10: 383-389 McHale L, Tan X, Koehl P, Michelmore RW (2006) Plant NBS-LRR proteins: adaptable guards. Genome Biol 7: 212 Merkys AJ, Laurinavicius RS, Svegzdiene DV (1984) Plant growth, development and embryogenesis during Salyut-7 flight. Adv Space Res 4: 55-63 Ming DW, Mittlefehldt DW, Morris RV, Golden DC, Gellert R, Yen A, Clark BC, Squyres SW, Farrand WH, Ruff SW, Arvidson RE, Klingelhofer G, McSween HY, Rodionov DS, Schroder C, de Souza Jr. PA, Wang A (2006) Geochemical and mineralogical indicators for aqueous processes in the Columbia Hills of Gusev crater, Mars. J. Geophys. Res. 111 Miwa K, Takano J, Omori H, Seki M, Shinozaki K, Fujiwara T (2007) Plants tolerant of high boron levels. Science 318: 1417 Munns R, Tester M (2008) Mechanisms of salinity tolerance. Annu Rev Plant Biol 59: 651-681 Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 15: 473-497 Musgrave ME, Kuang A (2003) Plant reproductive developm ent during spaceflight. Adv Space Biol Med 9: 1-23 Musgrave ME, Kuang A, Xiao Y, Stout SC, Bingham GE, Briarty LG, Levenskikh MA, Sychev VN, Podolski IG (2000) Gravity independence of seed-to-seed cycling in Brassica rapa. Planta 210: 400-406 Naito K, Kusaba M, Shikazono N, Takano T, Tanaka A, Tanisaka T, Nishimura M (2005) Transmissible and nontransmissible mutations induced by irradiating Arabidopsis thaliana pollen with gamma-rays and carbon ions. Genetics 169: 881-889 158

PAGE 159

Nelson M, Dempster WF, Allen JP, Silverstone S, Alling A (2008) Cowpeas and pinto beans: perform ance and yields of candidate space crops in the Laboratory Biosphere closed ecological system. Adv Space Res 41: 748-753 Nevzgodina LV, Maksimova EN, Kaminskaya EV (1989) Effects of prolonged exposure of lettuce seeds to HZE particles on orbital stations. Adv Space Res 9: 53-58 Nevzgodina V, Gaubin Y, Kovalev EE, Plan el H, Clegg J, Gasset G, Maximova EN, Pianezzi B (1984) Changes in developmental capac ity of artemia cyst and chromosomal aberrations in lettuce seeds flown aboard Sa lyut-7 (Biobloc III experiment). Adv Space Res 4: 71-76 Nguema-Ona E, Bannigan A, Chevalier L, Baskin TI, Driouich A (2007) Disruption of arabinogalactan proteins di sorganizes cortical micr otubules in the root of Arabidopsis thaliana. Plant J 52: 240-251 Nieuwland J, Feron R, Huisman BA, Fasolino A, Hilbers CW, Derksen J, Mariani C (2005) Lipid transfer proteins enhance cell wall extension in tobacco. Plant Cell 17: 2009-2019 Nieves-Cordones M, Miller AJ, Al eman F, Martinez V, Rubio F (2008) A putative role for the plasma membrane potential in the contro l of the expression of the gene encoding the tomato high-affinity potassium transporter HAK5. Plant Mol Biol 68: 521-532 Nikiforova VJ, Bielecka M, Gakiere B, Kr ueger S, Rinder J, Kempa S, Morcuende R, Scheible WR, Hesse H, Hoefgen R (2006) Effect of sulfur ava ilability on the integrity of amino acid biosynthesis in plants. Amino Acids 30: 173-183 Nikjoo H, Munson RJ, Bridges BA (1999) RBE-LET relationships in mutagenesis by ionizing radiation. J Radiat Res (Tokyo) 40 Suppl: 85-105 O'Dell RE, James JJ, Richards JH (2006) Congeneric serpentine and nonserpentine shrubs differ more in leaf Ca:Mg than in tolerance of low N, low P, or heavy metals. Plant and Soil 280: 49-64 Padmanaban S, Chanroj S, Kwak JM, Li X, Ward JM, Sze H (2007) Participation of endomembrane cation/H+ exchanger AtCHX20 in osmore gulation of guard cells. Plant Physiol 144: 82-93 Palacio S, Escudero A, Montserrat-Mart i G, Maestro M, Milla R, Albert MJ (2007) Plants living on gypsum: beyond the specialist model. Ann Bot (Lond) 99: 333-343 Pandey GK, Cheong YH, Kim BG, Grant JJ, Li L, Luan S (2007) CIPK9: a calcium sensorinteracting protein kinase required for low-pot assium tolerance in Arabidopsis. Cell Res 17: 411-421 Paul AL, Popp MP, Gurley WB, Guy C, Norwood KL, Ferl RJ (2005) Arabidopsis gene expression patterns are altered during spaceflight. Adv Space Res 37: 1175-1181 159

PAGE 160

Peterson DD, Benton EV, Tran M, Yang T, Freeling M, Craise L, Tobias CA (1977) Biological effects of high-LET particles on co rn-seed embryos in the Apollo-Soyuz Test Project Biostack III exp eriment. Life Sci Space Res 15: 151-155 Qi H, Li D, Zhang J, Liu L, Zhang Q (2008) Study on extraction of agaropectin from Gelidium amansii and its anticoagulant activity. Chin ese Journal of Oceanology and Limnology 26: 186-189 Rieder R, Gellert R, Anderson RC, Bruckn er J, Clark BC, Dreibus G, Economou T, Klingelhofer G, Lugmair GW, Ming DW, Squyres SW, d'Uston C, Wanke H, Yen A, Zipfel J (2004) Chemistry of rocks and soils at Me ridiani Planum from the Alpha Particle X-ray Spectrometer. Science 306: 1746-1749 Rose JK, Braam J, Fry SC, Nishitani K (2002) The XTH family of enzymes involved in xyloglucan endotransglucosylation and endohydr olysis: current perspectives and a new unifying nomenclature. Plant Cell Physiol 43: 1421-1435 Rosti J, Barton CJ, Albrecht S, Dupree P, Pauly M, Findlay K, Roberts K, Seifert GJ (2007) UDP-glucose 4-epimerase isoforms UGE2 and UGE4 cooperate in providing UDPgalactose for cell wall biosynthesis and growth of Arabidopsis thaliana Plant Cell 19: 1565-1579 Sahrawat KL (2004) Iron toxicity in wetland rice and the role of other nutrients. J. Plant Nutrition 27: 1471-1504 Salisbury FB, Campbell WF, Carman JG, Bingham GE, Bubenheim DL, Yendler B, Sytchev V, Levinskikh MA, Ivanova I, Chernova L, Podolsky I (2003) Plant growth during the Greenhouse II experiment on th e Mir orbital station. Adv Space Res 31: 221227 Salisbury FB, Gitelson JI, Lisovsky GM (1997) Bios-3: Siberian experiments in bioregenerative life support. Bioscience 47: 575-585 Sampedro J, Cosgrove DJ (2005) The expansin superfamily. Genome Biol 6: 242 Schmidt P, Kiefer J (1998) Deletion-pattern analysis of Alpha Part icle and X-ray induced mutations at the HPRT locus of V79 Chinese hamster cells. Mutat Res 421: 149-161 Schock I, Gregan J, Steinhauser S, Schweyen R, Brennicke A, Knoop V (2000) A member of a novel Arabidopsis thaliana gene family of candidate Mg2+ ion transporters complements a yeast mitochondrial group II intr on-splicing mutant. Plant J 24: 489-501 Schuerger AC, Ming DW, Newsom HE, Ferl RJ, McKay CP (2002) Near-term lander experiments for growing plants on Mars: re quirements for information on chemical and physical properties of Mars re golith. Life Support Biosph Sci 8: 137-147 160

PAGE 161

Segonzac C, Boyer JC, Ipotesi E, Sz ponarski W, Tillard P, Touraine B, Sommerer N, Rossignol M, Gibrat R (2007) Nitrate efflux at the root plasma membrane: identification of an Arabidopsis excretion transporter. Plant Cell 19: 3760-3777 Seifert GJ, Roberts K (2007) The biology of arabinogalactan proteins. Annu Rev Plant Biol 58: 137-161 Shaul O (2002) Magnesium transport and function in plants: the tip of the iceberg. Biometals 15: 309-323 Shaul O, Hilgemann DW, de-Almeida-Eng ler J, Van Montagu M, Inz D, Galili G (1999) Cloning and characterization of a novel Mg2+/H+ exchanger. EMBO J 18: 3973-3980 Shi H, Lee BH, Wu SJ, Zhu JK (2003) Overexpression of a plasma membrane Na+/H+ antiporter gene improves salt tolerance in Arabidopsis thaliana Nat Biotechnol 21: 81-85 Smeets K, Ruytinx J, Van Belleghem F, Sema ne B, Lin D, Vangronsveld J, Cuypers A (2008) Critical evaluation a nd statistical valida tion of a hydroponic culture system for Arabidopsis thaliana Plant Physiol Biochem 46: 212-218 Sottosanto JB, Gelli A, Blumwald E (2004) DNA array analyses of Arabidopsis thaliana lacking a vacuolar Na+/H+ antiporter: impact of AtNHX1 on gene expression. Plant J 40: 752-771 Stankovic B (2001) 2001: a plant space odyssey. Trends Plant Sci 6: 591-593 Storey JD, Tibshirani R (2003) Statistical significance fo r genome-wide studies. Proc Natl Acad Sci U S A 100: 9440-9445 Stutte GW, Monje O, Hatfield RD, Paul AL, Ferl RJ, Simone CG (2006) Microgravity effects on leaf morphology, cell structure, carbon metabolism and mRNA expression of dwarf wheat. Planta 224: 1038-1049 Sychev VN, Levenskikh MA, Gostimsky SA, Bingham GE, Podolsky IG (2007) Spaceflight effects on consecutive generations of peas grown onboard the Russian segment of the International Space Station. Acta Astronaut 60: 426-432 Sychev VN, Shepelev EY, Meleshko GI, Guri eva TS, Levinskikh MA, Podolsky IG, Dadasheva OA, Popov VV (2001) Main characteristics of biological components of developing life support system observed duri ng the experiments aboard orbital complex Mir. Adv Space Res 27: 1529-1534 Taji T, Seki M, Satou M, Sakurai T, Kobayas hi M, Ishiyama K, Narusaka Y, Narusaka M, Zhu JK, Shinozaki K (2004) Comparative genomics in salt tolerance between Arabidopsis and Arabidopsis-related halophyt e salt cress using Ar abidopsis microarray. Plant Physiol 135: 1697-1709 161

PAGE 162

Takahashi Y, Berberich T, Yamashita K, Uehara Y, Miyazaki A, Kusano T (2004) Identificatio n of tobacco HIN1 and two closel y related genes as spermine-responsive genes and their differential expression during the Tobacco mosaic virus-induced hypersensitive response and during leafand flow er-senescence. Plant Mol Biol 54: 613-622 Takeda S, Paszkowski J (2006) DNA methylation and epigen etic inheritance during plant gametogenesis. Chromosoma 115: 27-35 Tatematsu K, Kumagai S, Muto H, Sato A, Watahiki MK, Harper RM, Liscum E, Yamamoto KT (2004) MASSUGU2 encodes Aux/IAA19, an auxin-regulated protein that functions together with the transcriptional activator NPH4/ARF7 to regulate differential growth responses of hypocotyl and formation of lateral roots in Arabidopsis thaliana Plant Cell 16: 379-393 Tocquin P, Corbesier L, Havelange A, Pielta in A, Kurtem E, Bernier G, Perilleux C (2003) A novel high efficiency, low maintenance, hydroponic system for synchronous growth and flowering of Arabidopsis thaliana. BMC Plant Biol 3: 2 Turner TL, von Wettberg EJ, Nuzhdin SV (2008) Genomic analysis of differentiation between soil types reveals candidate genes for local adaptation in Arabidopsis lyrata PLoS ONE 3: e3183 Ubeda-Tomas S, Swarup R, Coates J, Swarup K, Laplaze L, Beemster GT, Hedden P, Bhalerao R, Bennett MJ (2008) Root growth in Arabi dopsis requires gibberellin/DELLA signalling in the endodermis. Nat Cell Biol 10: 625-628 Ulker B, Shahid Mukhtar M, Somssich IE (2007) The WRKY70 transcription factor of Arabidopsis influences both the plant sene scence and defense signa ling pathways. Planta 226: 125-137 van de Mortel JE, Almar Villanueva L, Schat H, Kwekkeboom J, Coughlan S, Moerland PD, Ver Loren van Themaat E, Koornneef M, Aarts MG (2006) Large expression differences in genes for iron and zinc homeostasis, stress response, and lignin biosynthesis distinguish roots of Arabidopsis thaliana and the related metal hyperaccumulator Thlaspi caerulescens Plant Physiol 142: 1127-1147 Vercauteren I, Niebel A, Van Montagu M, Gheysen G (2002) Arabidopsis thaliana mutants with an altered susceptibility to the root-knot nematode Meloidogyne incognita Meded Rijksuniv Gent Fak Landbouwkd Toegep Biol Wet 67: 51-61 Vissenberg K, Fry SC, Verbelen JP (2001) Root hair initiation is coupled to a highly localized increase of xyloglucan endotransglycosylase action in Arabidopsis roots. Plant Physiol 127: 1125-1135 Wolfinger RD, Gibson G, Wolfinger ED, Benne tt L, Hamadeh H, Bushel P, Afshari C, Paules RS (2001) Assessing gene significance from cDNA mi croarray expression data via mixed models. J Comput Biol 8: 625-637 162

PAGE 163

163 Wu J, Lenchik NI, Gerling IC (2008) Approaches to reduce false positives and false negatives in the analysis of microarray data: applic ations in type 1 di abetes research. BMC Genomics 9 Suppl 2: S12 Yang S, Yang H, Grisafi P, Sanchatjate S, Fink GR, Sun Q, Hua J (2006) The BON/CPN gene family represses cell death and prom otes cell growth in Arabidopsis. Plant J 45: 166179 Yen AS, Gellert R, Schroder C, Morris RV, Bell JF, 3rd, Knudson AT, Clark BC, Ming DW, Crisp JA, Arvidson RE, Blaney D, Bruckner J, Christensen PR, DesMarais DJ, de Souza PA, Jr., Economou TE, Ghosh A, Hahn BC, Herkenhoff KE, Haskin LA, Hurowitz JA, Joliff BL, Johnson JR, Klin gelhofer G, Madsen MB, McLennan SM, McSween HY, Richter L, Rieder R, Rodion ov D, Soderblom L, Squyres SW, Tosca NJ, Wang A, Wyatt M, Zipfel J (2005) An integrated view of the chemistry and mineralogy of Martian soils. Nature 436: 49-54 Yoshimoto N, Inoue E, Watanabe-Ta kahashi A, Saito K, Takahashi H (2007) Posttranscriptional regulation of high-affinity sulfate transpor ters in Arabidopsis by sulfur nutrition. Plant Physiol 145: 378-388 Yoshimoto N, Takahashi H, Smith FW, Yamaya T, Saito K (2002) Two distinct high-affinity sulfate transporters with different inducibilit ies mediate uptake of su lfate in Arabidopsis roots. Plant J 29: 465-473 Zhang G, Chen M, Chen X, Xu Z, Guan S, Li LC, Li A, Guo J, Mao L, Ma Y (2008) Phylogeny, gene structures, and expression patterns of the ERF gene family in soybean ( Glycine max L.). J Exp Bot 59: 4095-4107 Zhao J, Shigaki T, Mei H, Guo YQ, Cheng NH, Hirschi KD (2009) Interaction between Arabidopsis Ca2+/H+ exchangers CAX1 and CAX3. J Biol Chem 284: 4605-4615 Zhao Y, Christensen SK, Fankhauser C, Ca shman JR, Cohen JD, Weigel D, Chory J (2001) A role for flavin monooxygenase-like enzymes in auxin biosynthesis. Science 291: 306-309 Zhao Y, Hull AK, Gupta NR, Goss KA, Alonso J, Ecker JR, Normanly J, Chory J, Celenza JL (2002) Trp-dependent auxin biosynthesis in Arabidopsis: involvement of cytochrome P450s CYP79B2 and CYP79B3. Genes Dev 16: 3100-3112

PAGE 164

BIOGRAPHICAL SKETC H Anne Visscher was born in the Netherlands and completed her masters degree in soil, water and atmosphere at the Wageningen Un iversity in 2003. She has had a long-standing interest in advanced life support systems and furt her developed this intere st through an internship with the Biosphere Foundation during her master s program. This provided a connection with the lab of Rob Ferl at the Univers ity of Florida, where she subseq uently commenced her PhD project in January 2005. She received her PhD from th e University of Florida in the Summer of 2009. 164