Citation
Effects of Soil and Plant on Arsenic Accumulation by Arsenic Hyperaccumulator Pteris vittata L

Material Information

Title:
Effects of Soil and Plant on Arsenic Accumulation by Arsenic Hyperaccumulator Pteris vittata L
Creator:
GONZAGA, MARIA ISIDORIA SILVA ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Arsenic ( jstor )
Biomass ( jstor )
Ferns ( jstor )
Phytoremediation ( jstor )
Plant roots ( jstor )
Plants ( jstor )
Rhizosphere ( jstor )
Soil science ( jstor )
Soils ( jstor )
Species ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Maria Isidoria Silva Gonzaga. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
7/24/2006
Resource Identifier:
496614577 ( OCLC )

Downloads

This item has the following downloads:

silva_m ( .pdf )

silva_m_Page_001 ( .txt )

silva_m_Page_002 ( .txt )

silva_m_Page_003 ( .txt )

silva_m_Page_004 ( .txt )

silva_m_Page_005 ( .txt )

silva_m_Page_006 ( .txt )

silva_m_Page_007 ( .txt )

silva_m_Page_008 ( .txt )

silva_m_Page_009 ( .txt )

silva_m_Page_010 ( .txt )

silva_m_Page_011 ( .txt )

silva_m_Page_012 ( .txt )

silva_m_Page_013 ( .txt )

silva_m_Page_014 ( .txt )

silva_m_Page_015 ( .txt )

silva_m_Page_016 ( .txt )

silva_m_Page_017 ( .txt )

silva_m_Page_018 ( .txt )

silva_m_Page_019 ( .txt )

silva_m_Page_020 ( .txt )

silva_m_Page_021 ( .txt )

silva_m_Page_022 ( .txt )

silva_m_Page_023 ( .txt )

silva_m_Page_024 ( .txt )

silva_m_Page_025 ( .txt )

silva_m_Page_026 ( .txt )

silva_m_Page_027 ( .txt )

silva_m_Page_028 ( .txt )

silva_m_Page_029 ( .txt )

silva_m_Page_030 ( .txt )

silva_m_Page_031 ( .txt )

silva_m_Page_032 ( .txt )

silva_m_Page_033 ( .txt )

silva_m_Page_034 ( .txt )

silva_m_Page_035 ( .txt )

silva_m_Page_036 ( .txt )

silva_m_Page_037 ( .txt )

silva_m_Page_038 ( .txt )

silva_m_Page_039 ( .txt )

silva_m_Page_040 ( .txt )

silva_m_Page_041 ( .txt )

silva_m_Page_042 ( .txt )

silva_m_Page_043 ( .txt )

silva_m_Page_044 ( .txt )

silva_m_Page_045 ( .txt )

silva_m_Page_046 ( .txt )

silva_m_Page_047 ( .txt )

silva_m_Page_048 ( .txt )

silva_m_Page_049 ( .txt )

silva_m_Page_050 ( .txt )

silva_m_Page_051 ( .txt )

silva_m_Page_052 ( .txt )

silva_m_Page_053 ( .txt )

silva_m_Page_054 ( .txt )

silva_m_Page_055 ( .txt )

silva_m_Page_056 ( .txt )

silva_m_Page_057 ( .txt )

silva_m_Page_058 ( .txt )

silva_m_Page_059 ( .txt )

silva_m_Page_060 ( .txt )

silva_m_Page_061 ( .txt )

silva_m_Page_062 ( .txt )

silva_m_Page_063 ( .txt )

silva_m_Page_064 ( .txt )

silva_m_Page_065 ( .txt )

silva_m_Page_066 ( .txt )

silva_m_Page_067 ( .txt )

silva_m_Page_068 ( .txt )

silva_m_Page_069 ( .txt )

silva_m_Page_070 ( .txt )

silva_m_Page_071 ( .txt )

silva_m_Page_072 ( .txt )

silva_m_Page_073 ( .txt )

silva_m_Page_074 ( .txt )

silva_m_Page_075 ( .txt )

silva_m_Page_076 ( .txt )

silva_m_Page_077 ( .txt )

silva_m_Page_078 ( .txt )

silva_m_Page_079 ( .txt )

silva_m_Page_080 ( .txt )

silva_m_Page_081 ( .txt )

silva_m_Page_082 ( .txt )

silva_m_Page_083 ( .txt )

silva_m_Page_084 ( .txt )

silva_m_Page_085 ( .txt )

silva_m_Page_086 ( .txt )

silva_m_Page_087 ( .txt )

silva_m_Page_088 ( .txt )

silva_m_Page_089 ( .txt )

silva_m_Page_090 ( .txt )

silva_m_Page_091 ( .txt )

silva_m_Page_092 ( .txt )

silva_m_Page_093 ( .txt )

silva_m_Page_094 ( .txt )

silva_m_Page_095 ( .txt )

silva_m_Page_096 ( .txt )

silva_m_Page_097 ( .txt )

silva_m_Page_098 ( .txt )

silva_m_Page_099 ( .txt )

silva_m_Page_100 ( .txt )

silva_m_Page_101 ( .txt )

silva_m_Page_102 ( .txt )

silva_m_Page_103 ( .txt )

silva_m_Page_104 ( .txt )

silva_m_Page_105 ( .txt )

silva_m_Page_106 ( .txt )

silva_m_Page_107 ( .txt )

silva_m_Page_108 ( .txt )

silva_m_Page_109 ( .txt )

silva_m_Page_110 ( .txt )

silva_m_Page_111 ( .txt )

silva_m_Page_112 ( .txt )

silva_m_Page_113 ( .txt )

silva_m_Page_114 ( .txt )

silva_m_Page_115 ( .txt )

silva_m_Page_116 ( .txt )

silva_m_Page_117 ( .txt )

silva_m_Page_118 ( .txt )

silva_m_Page_119 ( .txt )

silva_m_Page_120 ( .txt )

silva_m_Page_121 ( .txt )

silva_m_Page_122 ( .txt )

silva_m_Page_123 ( .txt )

silva_m_Page_124 ( .txt )

silva_m_Page_125 ( .txt )

silva_m_Page_126 ( .txt )

silva_m_Page_127 ( .txt )

silva_m_Page_128 ( .txt )

silva_m_Page_129 ( .txt )

silva_m_Page_130 ( .txt )

silva_m_Page_131 ( .txt )

silva_m_Page_132 ( .txt )

silva_m_Page_133 ( .txt )

silva_m_Page_134 ( .txt )

silva_m_Page_135 ( .txt )

silva_m_Page_136 ( .txt )

silva_m_Page_137 ( .txt )

silva_m_Page_138 ( .txt )

silva_m_Page_139 ( .txt )

silva_m_Page_140 ( .txt )

silva_m_Page_141 ( .txt )

silva_m_Page_142 ( .txt )

silva_m_Page_143 ( .txt )

silva_m_Page_144 ( .txt )

silva_m_Page_145 ( .txt )

silva_m_Page_146 ( .txt )


Full Text













EFFECTS OF SOIL AND PLANT ON ARSENIC ACCUMULATION BY ARSENIC
HYPERACCUMULATOR Pteris vittata L












By

MARIA ISIDORIA SILVA GONZAGA


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


2006

































Copyright 2006

By

Maria Isidoria Silva Gonzaga

































To the memory of my father Henrique, whose love and inspiration have accompanied me
through the years; to my mother Ines who, together with my father, gave me all the
values that guide my life; and to my kids Jamile, Thomas and Genna, the best fruits of
my existence.















ACKNOWLEDGMENTS

I wish to thank my supervisory committee chair (Dr. Lena Q. Ma) for providing

valuable academic advices, critical support and guidance during my graduate study. I

thank my supervisory committee members (Jean-Claude Bonzongo, Nicholas B.

Comerford, Robert Stamps, and Roy Dean Rhue) for their assistance and advice. I also

wish to especially thank Dr. Nick Comerford for helping me apply to the University of

Florida and also for helping me and my family during our first few days in Gainesville.

Special thanks also go to Dr. Jean-Claude Bonzongo for his advisement and valuable

friendship.

I am grateful for the assistance and support of my husband, Dr. Jorge Santos, who

provided experimental and statistical advice and also helped in conducting my

experiments. I am grateful to Mr. Tom Luongo for his assistance with chemical analysis

and proofreading manuscripts; to Dr. Mristunjai Srivastava for his assistance in providing

fern plants and proofreading manuscripts; to Dr. John Thomas for his friendship and for

proofreading manuscripts; and Mr. Bill Reeve, for his friendship and help with chemical

analyses. I also wish to thank the members of the Biogeochemistry of Trace Metals

Laboratory (Gina Kertulius-Tartar, Abioye Fayiga, Joonki Yoon, Rocky Cao, Mike Tu,

and Autumm Wan), for what they have taught me. I thank Mrs. Marlin Pam for her help

and patience.

I would like to acknowledge the Brazilian institution (CAPES) for providing

financial support during my Ph.D. program. I am grateful to Dr. Paulo Gabriel Nacif (and









members of the Soil Chemistry Department of the Agronomy School, at the Federal

University of Bahia) for granting me permission to pursue my degree. I also thank Dr.

Joelito Rezende for all the encouragement and help, keeping me in the scientific life.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ....................................................... ............ ....... ....... ix

LIST OF FIGURES ......... ......................... ...... ........ ............ xi

A B S T R A C T .............................................. ..........................................x iii

CHAPTER

1 IN TRODU CTION ................................................. ...... .................

2 LITERATURE REVIEW ........................................................................4

A rsenic ........................ ............ ...... .................... ...................... ....... 4
O ccurrence, A availability and Toxicity ........................................ .....................4
Remediation of Arsenic Contaminated Soils .................................................6
P hytorem edition ........................ ......................................................... ............... .. 7
Phytoextraction................................ ........... ...... .. ........ .. .. .. ........ .. 7
Phytostabilization ...................................... ........................... .8
Phytoim obilization .................................................................... .8
Phytovolatilization....................................... .. .... ... ............. ....
Phytoextraction of A rsenic .................................................................... ............... 9
H yperaccum ulator Plants ........... .. ....... .. ........... .................... ............... 13
Arsenic Hyperaccumulation by Pteris vittata ............................................. 16
Arsenic Species in Pteris vittata............................... ....................18
Arsenic Tolerance and Detoxification Mechanisms................ ..................19
Effects of Heavy Metals on Arsenic Hyperaccumulation..............................21
Effect of Phosphate on Arsenic Hyperaccumulation.......... ............... 22
Mycorrhizae Association and its Role in Arsenic Hyperaccumulation...............24

3 ARSENIC ACCUMULATION BY Pteris vittata L FROM ARSENIC
CON TAM IN A TED SOILS ............................................... ............................. 27

In tro du ctio n ...................................... ................................................ 2 7
M materials and M ethods ....................................................................... ..................29
Soil Collection and Characterization................... ...... ....................... 29
Experimental Design and Statistical Analyses............... ...............29









E xperim mental Set up .................................................. .............................. 29
Sam pling, D igestion and A analysis ........................................... .....................31
A rsenic Fractionation ....................... ...................... ........... ...............3 1
Single E xtractions........ ................................................................. .... .... .... 32
R results and D iscu ssion ................................. .............. ................ ..... .......... 33
Plant G row th........................ ....................................... 33
Arsenic Concentrations in the Frond Biomass ............................................ 36
Plant A rsenic R em oval from Soils ........................................... .....................38
Soil A rsenic D distribution ............... ....... ..... .............................................39
Comparison of Extractable Arsenic Using Single Extraction Methods .............44
Plant uptake and Arsenic Availability in Soils as Measured by Single
E x tractan ts............................. .................................................. ............... 4 4

4 EFFECTS OF PLANT AGE ON THE ARSENIC ACCUMULATION BY THE
HYPERACCUMULATION Pteris vittata L......................................................49

In tro du ctio n ...................................... ................................................ 4 9
M materials and M methods ....................................................................... ..................50
Plant Propagation .................................................. .. ....... .. ........ .. 50
Soil C haracterization ...................... .. .. ..................... .... .. ........... 50
E xperim ental D esign ........................................ ........................... .......... 1
C hem ical A naly sis......... .......................................................... .... ........52
D ata A n aly sis................................................ ................ 5 3
R results and D iscu ssion ................................. .............. ................ ..... .......... 53
A rsenic U ptak e ..............................................................53
Plant Biomass ................................................ 56
Phosphorus Distribution in the Plants ...................................... ............... 56
Concentrations of Other Nutrients in Plants..................................................59

5 ARSENIC ACCUMULATION AND ROOT CHARACTERISTICS OF Pteris
vittata L. AND Nephrolepis exaltata L ............................ ...... ...................... 63

Introdu action .......................... ............. .........................................................63
M materials and M methods ....................................................................... ..................66
Soil C haracterization ...................... .. .. ..................... .... .. ........... 66
E xperim ental Set up ................................................. ............................... 66
Root M easurem ent ............... .. .... .......... ...... ............ ... .... .. ............ 68
C hem ical A nalysis........... ........................................................ ...... .... .....69
A rsenic Fractionation .......................................... .. .. ...... ...............69
D ata A n a ly sis ................................................................................................. 7 0
R results and D discussion ............... ... ......... ................................ ...... .... .......... 70
Root Biomass, Root Length, Root Area, and Root Diameter ...........................70
Frond Biomass, Arsenic Uptake, Bioconcentration and Translocation .............72
Root Uptake Efficiency of Arsenic and Phosphorus................ ..................76
P lant N utrient U ptake ................... ...................................................................78
Arsenic Reduction and Distribution in the Soil ................................................79
Influence of Plants on Water-Soluble Arsenic, Soil pH and DOC................. 84









6 ARSENIC ACCUMULATION BY TWO HYPERACCUMULATOR Pteris
SPECIES FROM TWO ARSENIC CONTAMINATED SOILS..............................90

In tro du ctio n ...................................... ................................................ 9 0
M materials and M ethods ....................................................................... ..................9 1
Soil Collection and Characterization.................... ...... ...................... 91
C hem ical A naly sis........... ........................................................ .. .... ...... .94
D ata A n a ly sis ................................................................................................. 9 4
R results and D discussion ....................................................................................... .....94
Arsenic Accumulation and Distribution in the Ferns............... .... ........... 94
Plant Biomass of P. biaurita and P. vittata...................... .......................98
R hizosphere C hem istry ............................................. ............................ 100
E lem ental C om position ............................................. .............. ............105

7 CON CLU SION S ........... .. .... .... ... .... ................................ 111

LIST O F R EFEREN CE S ....... ............................................................. ............... 115

BIOGRAPHICAL SKETCH ......... ....... ............. ............. 132















LIST OF TABLES


Table pge

3-1 Selected chemical and physical characteristics of six arsenic contaminated soils...30

3-2 Arsenic sequential extraction procedure ...................................... ............... 32

3-3 The frond arsenic concentrations ofPteris vittata after the first, second and,
third harvest in six soils............. .... .................................................... .. .... ..... .. 37

3-4 Arsenic reduction (%) in each fraction in six soils as a result of plant arsenic
uptake by P. vittata............ ....... .................................. .................43

3-5 Concentrations of arsenic extracted by different extraction methods in six soils
before plant grow th. ......................................... ........................... 45

3-6 Percent of the arsenic taken up by P. vittata in each extracting method.................46

4-1 Plant biomass, bioconcentration factor and translocation factor ofP. vittata of
different ages after 8 weeks of growth in an arsenic-contaminated soil ................54

4-2 Macronutrients (K, Ca, and Mg) (g kg-1) and micronutrients (Fe, Zn, and Mn)
content (mg kg-1) in the fronds and roots of P. vittata of different ages.................61

5- 1 Selected properties of the soils used in this study ................ ............... 66

5-2 The root biomass (dw), root length, and root area of P. vittata and N. exaltata
after 8 weeks of growth in a As-contaminated soil (As-soil) and control soil.........71

5-3 The frond biomass (dw), arsenic accumulation, and bioconcentration and
translocation factors of P. vittata and N. exaltata .............................. ............... 76

5-4 The root uptake efficiency (RUE) as measured by As and P uptake (mg plant-)
by the fronds and roots of P. vittata and N. exaltata..............................................78

5-5 Nutrients concentrations (g kg-1) in the fronds and roots of P. vittata and N.
exaltata after 8 weeks of growth in an As-contaminated soil and a control soil. ....80

5-6 Arsenic distribution in different fractions in the bulk soil after 8 weeks of plant
g ro w th ...................................... ................................................... 8 1









5-7 Arsenic distribution in different fractions in the rhizosphere of P. vittata (PV)
and N. exaltata (NE) after 8 weeks of plant growth................. ......................82

6-1 Selected properties of the soils used in this study ....................................................92

6-2 Comparison of the arsenic concentration (mg kg-1), translocation factor (TF) and
bioconcentration factor (BF) of P. vittata and P. biaurita ................... ..............96

6-3 The macronutrients (P, K, Ca, Mg) (g kg-1) and micronutrients (Fe, Zn, Mn)
content (mg kg-1) in the fronds and the roots of P. vittata and P. biaurita ..........06
















LIST OF FIGURES


Figure pge

2-1 Factors that influence the efficiency of phytoextraction......................................... 11

2-2 Conceptual response of hyperaccumulator, indicator and excluder plants for
metal concentrations in the aboveground tissues. ............. ..................................... 15

3-1 Fond biomass ofPteris vittata after first, second and third harvests in six soils.....34

3-2 Arsenic accumulation by P. vittata after first, second and third harvests in six
s o ils .............................................................................. 3 8

3-3 Arsenic removed from the soil by Pteris vittata after three harvests.....................40

3-4 Arsenic concentrations (mg kg-1) in different fractions in soils before and after
plant growth............. ......... ...... ............................. .........41

4-1 Pteris vittata of 1.5, 4, 10, and 16 old months before transplant...........................51

4-2 Arsenic accumulation in the fronds and roots of P. vittata of different ages after
8 weeks of growth in an arsenic-contaminated soil. ..............................................54

4-3 The frond biomass of P. vittata of different ages before and after 8 weeks of
growth in an arsenic-contaminated soil......................................... ............... 57

4-5 Effect of plant age on phosphorus and arsenic distribution in the fronds of P.
vittata after 8 weeks of growth in an arsenic-contaminated soil..............................59

4-6 Effect of plant age on the phosphorus and arsenic distribution in the roots of P.
vittata after 8 weeks of growth in an arsenic-contaminated soil.............................60

5.1 Cross section of the rhizopot used in the study ...................................................67

5-2 The root length density (Lv) of P. vittata and N. exaltata as a function of root
diameter in a control soil (a) and an As-contaminated soil (b) .............................74

5-3 The root surface area density (Av) of P. vittata and N. exaltata as a function of
root diameter in a control soil (a) and an As-contaminated soil (b).........................75

5-4 Comparison of water-soluble arsenic in the bulk and rhizosphere soil of P.
vittata and N. exaltata after 8 weeks of growth in an arsenic-contaminated soil.....86









5-5 Comparison of the soil pH in the bulk and rhizosphere soil of P. vittata and N.
exaltata after 8 weeks of growth in an arsenic-contaminated soil. ..........................88

5-6 Comparison of dissolved organic carbon (DOC) in the bulk and rhizosphere soil
of P. vittata and N. exaltata ......... ....... ......... .................. ............... 89

6-1 Pteris biaurita and Pteris vittata growing in an arsenic-contaminated soil.............95

6-2 The frond and root biomass (mg kg-1 DW) of P. vittata and P. biaurita after 8
weeks of growth in two arsenic contaminated soils (Soil 1: 153; Soil2: 266)........99

6-3 The bulk soil and rhizosphere pH of P. vittata and P. biaurita after 8 weeks of
growth in two arsenic contaminated soils. .................................. .................101

6-4 The bulk soil and rhizosphere dissolved organic carbon (DOC) of P. vittata and
P. biaurita after 8 weeks of growth in two arsenic contaminated soils ...............102

6-5 The water-soluble P concentrations in the bulk soil and rhizosphere of P. vittata
and P. biaurita.......................... ..... .. ....... .......... 103

6-6 The water-soluble As concentrations in the bulk soil and rhizosphere of P.
vittata and P biaurita .......... ........................................ .............. .. .... ......... 104















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

EFFECTS OF SOIL AND PLANT ON ARSENIC ACCUMULATION BY ARSENIC
HYPERACCUMULATOR Pteris vittata L

By

Maria Isidoria Silva Gonzaga

May 2006

Chair: Lena Q. Ma
Major Department: Soil and Water Science

Phytoextraction using Pteris vittata L., an arsenic hyperaccumulator plant, could

be potentially applied to remediate arsenic-contaminated sites worldwide. Pteris vittata

grew well and took up large amounts of arsenic in six arsenic-contaminated soils with

different properties and different sources of arsenic contamination. However, efficiency

of the ferns in continually taking up arsenic from the soils decreased with time, and

varied with growing season. The arsenic concentrations in different chemical fractions

before and after plant uptake showed that P. vittata took up arsenic from the most

available and also from the less available pools in all soils. The change in arsenic

availability in the rhizosphere was evaluated by comparing P. vittata with a non-arsenic-

hyperaccumulator Nephrolepis exaltata. As expected, P. vittata removed more arsenic

from the soil than the non-arsenic-hyperaccumulator fern. Besides the efficient

detoxification mechanisms ofP. vittata, its more extensive and finer root systems, and its

greater capacity to change the soil pH and produce root exudates in the rhizosphere likely









contributed to the difference. Furthermore, the experiments with P. vittata of different

physiological ages revealed that younger plants (because of their higher metabolic and

growth rate) were more efficient than older plants in taking up arsenic. Therefore,

younger plants are recommended for phytoremediation. Comparing two arsenic

hyperaccumulators in the genus Pteris showed that arsenic hyperaccumulator ferns

diffred in their ability to take up arsenic when growing under the same soils and

environmental conditions. The concentrations of arsenic in both plants increased with the

soil arsenic concentration, however, P. vittata performed better, regardless of the arsenic

level in soils. After 4 weeks of growth, P. biaurita showed signs of stress and probably

would not have survived longer under those conditions, while P. vittata showed no

toxicity symptom. Our results showed that, in implementing an arsenic phytoextraction

project, many aspects related to the plant must be considered.














CHAPTER 1
INTRODUCTION

Arsenic contamination in the environment from both anthropogenic and natural

sources occurs in many parts of the world and is a global problem. Arsenic-contaminated

soils, sediments, and sludge are the major sources of arsenic contamination in food chain,

surface water, groundwater, and drinking water (Frankenberger and Arshad, 2002). Other

potential sources of arsenic contamination are the chemicals used extensively in

agriculture such as pesticides, insecticides, defoliants, wood preservatives, and soil

sterilants (Azcue and Nriagu, 1994).

In many areas, arsenic levels in the environment have exceeded the safe threshold

for human health. Epidemiological studies have documented various adverse effects of

arsenic on humans and animals. Conventional remediation technologies have been used

to clean up metal-contaminated sites because they are relatively insensitive to the

heterogeneity in contaminated matrix, and can function over a wide range of oxygen, pH,

pressure, temperature, and osmotic potentials (Cunningham et al., 1997). However, they

are expensive and time-consuming, often hazardous to workers, and produce secondary

wastes (Lombi et al., 2000). Of the disadvantages of conventional remediation methods,

cost is the primary driving force behind the search for alternative remediation

technologies, such as phytoremediation.

Phytoremediation is the use of plants and their associated microbes to reduce,

remove, degrade, or immobilize environmental contaminants in soil and water systems

and can be applied for both organic and inorganic pollutants (Salt et al., 1998).









Phytoextraction, one of the strategies of phytoremediation, has attracted much attention

as an environmentally-friendly low-input remediation technique that uses plants that

extract heavy metals from the soil and accumulate them in the harvestable aboveground

biomass (McGrath et al., 2002). However, the effectiveness of a phytoremediation plan is

plant-dependent. Phytoextraction can be accomplished by using either tolerant high

biomass plant species or hyperaccumulator plant species. Plants native to the target area

should be considered, since they are adapted to the local climate, insects, and diseases.

Hyperaccumulators are plants that can take up and concentrate more than 0.1% of

a given element in their tissue (Brooks, 1998). Metal hyperaccumulation is a rare

phenomenon in terrestrial higher plants. The first arsenic hyperaccumulator, Pteris vittata

L., was identified by Ma et al. (2001), followed by Pityrogramma calomelanos L.

(Francesconi et al., 2002) and many other species of the Pteris genus such as P. cretica

L., P. longifolia L., P. umbrosa L., and P. argyraea L. (Zhao et al., 2002), and P.

quadriaurita L., P. ryiunkensis L. and P. biaurita (Srivastava et al., 2006).

Pteris vittata is a fast-growing fern and has the potential to produce relatively

high biomass while accumulating a large amount of arsenic in its aboveground tissue.

These characteristics and its ability to survive in different environments make this plant

an exceptional candidate for use in phytoremediation of arsenic-contaminated soils.

To increase the efficiency of phytoextraction of arsenic contaminated soils using

P. vittata, besides understanding its arsenic tolerance and detoxification mechanisms, it is

important to learn more about the biological processes involved, such as plant arsenic

uptake, plant nutritional requirements, rhizosphere processes and mobilization and






3


bioavailability of arsenic in the soil. Experiments included in our study had the following

objectives:

1. Evaluate the efficiency of P. vittata in extracting arsenic from different arsenic-
contaminated soils

2. Evaluate the importance of the physiological age of P. vittata on its arsenic uptake
and plant growth

3. Determine the characteristics of the root systems and changes in the rhizosphere of
P. vittata in comparison with a non-hyperaccumulator

4. Compare the efficiency ofP. vittata in arsenic phytoextraction in comparison with
Pteris biaurita, another arsenic hyperaccumulator.














CHAPTER 2
LITERATURE REVIEW

Arsenic

Occurrence, Availability and Toxicity

Arsenic concentrations range from below 10 mg kg1 in non-contaminated soils

(Adriano, 1986) to as high as 30,000 mg kg-1 in contaminated soils (Vaughan, 1993). In

rocks, arsenic is concentrated in magmatic sulphide minerals and iron ore. The most

important arsenic ores include arsenical pyrite or arsenopyrite (FeAsS), realgar (AsS),

and orpiment (As2S3).

Arsenic has been identified as a major toxic contaminant in many countries.

Various human activities have elevated the arsenic levels in soils such as the production

and the use of arsenical pesticides (fungicides, herbicides, and insecticides). In addition,

manufacture of arsenic-based compounds, smelting of arsenic-containing ores, and

combustion of fossil fuels have also contributed to arsenic contamination in soils, water,

and atmosphere (Azcue and Nriagu, 1994). An increase in industrialization has also lead

to an increase in the amount of arsenic present in biosolids. Arsenic deposits from the

atmosphere, runoff, and effluents of industries often increase the concentration of arsenic

in biosolids. Woolson (1983) reported a range of 0 to 188 mg kg-1 As dry weight of

biosolids. Biosolids are often disposed on land and may then increase arsenic

concentrations in the top 20 cm of soil by up to 0.15% (O'Neill, 1990).

Sorption onto soil particles is an important process for arsenic immobilization in

soils. Many studies have been devoted to arsenic sorption on well-characterized minerals









or soil particles (clay, oxides of Al, Fe, and Mn, calcium carbonates, and/or organic

matter) (Sadiq, 1997; Dobran and Zagury, 2005). The high affinity of arsenic for oxide

surfaces is well known and is affected by several biogeochemical factors such as pH,

redox potential, and competing ions (Adriano, 2001). The activity of arsenic in soil

solution is mostly controlled by surface complexation reactions on oxides/hydroxides of

Al, Mn, and especially Fe (Inskeep et al., 2002). The mobility of arsenic in soil is pH

dependent as arsenate is preferentially sorbed onto hydrous oxides in the pH range of 4 to

7; whereas, arsenite is preferentially sorbed for in the pH range of 7 to 10 (Pierce and

Moore, 1982). Frost and Griffin (1977) have shown that arsenic sorption onto kaolinite

and montmorillonite is pH dependent, while Lin and Puls (2000) confirmed that arsenate

is more strongly sorbed onto these minerals than arsenite. Furthermore, arsenate

adsorption onto humic substances reaches a maximum at pH 5.5, while arsenite

adsorption peaks at much higher pH of 8.5 (Thanabalasingam and Pickering, 1986).

The bioavailability, toxicity, and chemical behavior of arsenic compounds are

largely influenced by the form and speciation of arsenic. In natural systems, arsenic can

exist in four oxidation states: (-3), (0), (+3), and (+5). Arsenate [As(+5)] and arsenite

[As(+3)] are the main forms present in soils (Harper and Haswell, 1988). Arsenate

prevails under aerobic conditions and is somewhat less toxic and also less mobile than

arsenite (dominant form under anaerobic conditions) because arsenate sorbs more

strongly than arsenite onto minerals (Pierce and Moore, 1982). Generally, inorganic

arsenicals are more toxic than organic arsenicals while the trivalent oxidation state is

more toxic than the pentavalent oxidation state (Adriano, 2001).









Remediation of Arsenic Contaminated Soils

Many remediation techniques are available to address the problems in

contaminated sites. However, the relatively high capital expenditure, unsuitability for

large areas, and environmental disruption are some of the shortcomings of those

techniques. No single soil-remediation technique is suitable for all situations. Careful

investigation of the contaminated site characteristics, contaminant problem, treatment

options and treatment timeframe must be considered.

Current remediation methods for arsenic-contaminated soils include soil removal

and washing, physical stabilization, and/or the use of chemical amendments, all of which

are expensive and disruptive, with an average cost of $ 404,700 per ha (Raskin et al.,

1997). Following are some selected current remediation technologies for arsenic-

contaminated soils (USEPA, 2002a).

a. Capping: a hard cover is placed on the surface of a contaminated soil. It

is a simple method to reduce contaminant exposure. However, it does not

remove contaminants from the soil.


b. Solidification and stabilization: the contaminated soil is mixed with

stabilizers to reduce the arsenic mobility in a soil. It can be relatively

costly.


c. Vitrification: arsenic is chemically bonded inside a glass matrix forming,

silicoarsenates.


d. Soil flushing: uses water, chemicals, or organic to mobilize arsenic in a

soil and then flush it from the soil.









e. Phytoremediation/phytoextraction: uses plants to take up arsenic from

soil.


Ex situ remediation methods

f Excavation: physical removal and disposal of a contaminated soil in a

designated landfill. Even though it produces rapid results, excavation is

often expensive because of the operation, transport, and special landfill

requirements.


g. Soil washing/Acid extraction: based on suspension or dissolution of

arsenic in a water-based solution to concentrate the contaminant.


Phytoremediation

Phytoremediation includes several methods that use plants to either remove

contaminants or render them harmless in soil and water systems. It can be applied to

both organic and inorganic contaminants in soil and water (Salt et al., 1998). This

practice has been growing in popularity, because of its overall cost-effectiveness

(Watanabe, 1997; Salt et al., 1998; Kabata-Pendias and Pendias, 2001). The term

phytoremediation includes the following strategies.

Phytoextraction

Phytoextraction is the use of contaminant-accumulating plants, which are able to

extract and translocate contaminants to the harvestable parts. Phytoextraction is the most

effective strategy of phytoremediation, although technically the most difficult one. It

uses tolerant plants that concentrate soil contaminants in their aboveground biomass and

the contaminant-enriched biomass can then be properly disposed (Kramer, 2005).









Phytostabilization

Phytostabilization refers to the use of contaminant-tolerant plants for mechanical

stabilization of contaminated soils to prevent soil erosion and to reduce air-borne

transport and leaching of contaminants. It is used to provide a vegetation cover for a

contaminated site, thus preventing wind and water erosion (Kramer, 2005). Plants that

are suitable for phytostabilization have an extensive root system, provide a good soil

cover, are tolerant to the contaminants, and ideally immobilize the contaminants in the

rhizosphere. Arsenic-tolerant plants that may be potentially used for phytostabilization

purposes have been known for a long time (Rocovich and West, 1975; Porter and

Peterson, 1977; Benson et al., 1981).

Phytoimobilization

Phytoimmoblization is the use of plants to decrease the mobility and

bioavailability of contaminants by altering soil factors that lower contaminant mobility

by formation of precipitates and insoluble compounds and by sorption onto the roots.

Based on the chemical similarities between arsenic and phosphorus, there may be

formation of arsenic/lead compounds as shown for phosphorus/lead precipitates in the

rhizosphere ofAgrostis capillaris (Cotter-Howells et al., 1999). Other plant-mediated

processes of arsenic immobilization at the soil-root interface involve pH reduction and

oxidation of the root environment by 02 release from roots. Doyle and Otte (1997) found

accumulation of arsenic onto the iron plaque in the oxidized rhizosphere of salt marsh

plants, which may provide an effective immobilization and detoxification mechanism for

the plants.









Phytovolatilization

Phytovolatilization is the use of plants to volatilize contaminants and has been

demonstrated for Hg and Se. In the case of Hg, this was achieved by genetic

manipulation of plants (Rugh et al., 1996); whereas phytovolatilization of Se occurs

naturally in plants (Terry and Zayed, 1994). De Souza et al. (1999) demonstrated that

rhizosphere bacteria can enhance the Se volatilization and accumulation in plants.

Volatilization of arsenic is also known to occur in natural environments (Frankenberger

and Arshad, 2002), but rhizosphere studies have not been reported for the formation of

gaseous arsenicals enhancement at the soil-root interface. Available information on

arsenic volatilization for soil suggests that in the absence of plant roots, volatile

compounds account only for small proportions of total arsenic (Turpeinen et al., 1999).

Phytoextraction of Arsenic

Arsenic is a nonessential element for plants, and inorganic arsenic species are

generally phytotoxic. Under normal conditions, arsenic concentrations in terrestrial

plants are usually less than 10 mg kg-1 (Matschullat, 2000). Different plants contain

different levels of arsenic. The following plants contain arsenic in an increasing order:

cabbage (0.020 to0.050 mg kg-1) < carrots (0.040 to 0.080) < grass (0.020 to 0.160) <

potatoes (0.020 to 0.200) < lettuce (0.020 to 0.250) < mosses and lichens (0.26) < ferns

(1.3) (Matschullat, 2000).

An average toxicity threshold of 40 mg kg-1 was established for crop plants

(Sheppard, 1992). The chemical behavior of arsenic is largely similar to that of

phosphorus in soils and plants. In all plant species tested so far, arsenate is taken up via

the phosphate transport systems (Asher and Reay, 1979; Lee, 1992; Meharg and Macnair,

1992). As a phosphate analog, arsenate can replace phosphate in many biochemical









processes, thus disrupt phosphate metabolism in plants. For example, arsenate can

disrupt mitochondrial oxidative phosphorylation, thus the production of nucleotide

adenosine triphosphate, the main energy source for cells. This process is known as

arsenolysis, or the hydrolytic process whose first step is the replacement of arsenate for

phosphate (Oremland and Stolz, 2003). Unlike arsenate, arsenite reacts with sulfhydryl

groups of enzymes and tissue proteins, leading to inhibition of cellular function and death

(Meharg and Hartley-Whitaker, 2002).

The transfer of arsenic from soils to plants is low for most plant species. This is

due to several factors including: i) the restricted uptake by plant roots, ii) the limited

translocation of arsenic from roots to shoots, iii) arsenic phytotoxicity even at low

concentrations in plant tissues, and iv) the low bioavailability of arsenic in soil (Wang et

al., 2002).

Among phytoremediation methods, phytoextraction is the most suitable for

arsenic contaminated soils. It represents one of the largest economic opportunities for

phytoremediation. This is due to the size and scope of environmental problems

associated with arsenic contaminated soils, minimal environmental disturbance, public

acceptance and the competitive advantage offered by the plant-based remediation

technology (Raskin et al., 1997; Susarla, 2002).

After harvesting arsenic-enriched plants, the weight and volume of contaminated

material can be further reduced, transported, and disposed off site as hazardous material

(Salt et al., 1998; Ma et al., 2001). Successful application of phytoextraction to arsenic

contaminated soils depends on many factors, among which the concentration and









bioavailability of arsenic, the nutrient and water status of the soils and the capacity of the

plant to access the arsenic (Figure 2-1).

The arsenic bioavailability in soils is regulated by various physical, chemical and

biological processes and their interactions (Ernst, 1996). The physical characteristics of

soils have a great influence on the bioavailability of arsenic. For instance, soils with very

fine texture offer a physical resistance that may demand a lot of energy from the plant in

order to penetrate deeper layers. In this situation, arsenic may be inaccessible for plant

uptake.






Plant biomass / Plant specie



Contaminant bioavailability Contaminant concentration
Phytoreinediation

Uptake Rhizosphere processes



Soil characteristics Nutrient bioavailability

Figure 2-1. Factors that influence the efficiency of phytoextraction

Arsenic speciation and chemical conditions such as soil acidity and redox

potential can largely determine arsenic bioavailability (Wenzel et al., 2002). The soil

biota can strongly modify the chemical and physical conditions and processes, which

determine arsenic bioavailability. Rhizosphere interactions plays a key role in controlling

nutrient bioavailability to crop plants (Hinsinger, 2001) as well as in the understanding of

the processes occurring during phytoremediation (Lombi et al., 2001). The availability of









arsenic in a soil around the roots is strongly affected by root exudates and root

depositions mucilagee and border cells) but also by microbial activities (Lombi et al.,

2000). The processes occurring in the rhizosphere such as plant uptake, changes of pH

and redox potential, root exudation and etc. alter the chemical compositions of the soil-

root interface. It may also influence the bioavailability of arsenic in the soil and

consequently the efficiency of phytoextraction. For instance, under oxic soil conditions,

an increase in the rhizosphere pH could favor mobilization of labile and exchangeable

As(V), enhancing plant uptake (Wenzel et al., 2002). Also, nitrogen fertilization as

nitrate (NO3s) can potentially increase the rhizosphere pH, and thus possibly enhance

arsenic accumulation in plant tissues.

The symbiotic association of plant roots with mycorrhizal fungi in arsenic

contaminated soils can influence its bioavailability by exploiting a greater soil volume

and by solubilizing the arsenic. Therefore, the interactions of arsenic with soil matrix and

the ability of plants to continually accumulate and detoxify arsenic in their shoot system

are essential to the phytoremediation concept.

Phytoextraction can be accomplished by using either tolerant high biomass plant

species or hyperaccumulator plant species. The growth and remediation potential has

been assessed for various plants including cottonwood (Populus deltoides Bartr.), cypress

(Taxodium distichum L.), eucalyptus (Eucalyptus amplifolia Naudin, E. camaldulensis

Dehnh., and E. grandis Hill), and leucaena (Leucaena leucocephala L.), which are all

potential high biomass species. However, the use of hyperaccumulator plants has the

advantage of producing a more concentrated residue, reducing the final disposal of the

contaminant-rich biomass.









Unfortunately, most metal hyperaccumulator plants grow slowly with low

biomass, while plants that produce a high biomass are usually sensitive to high metal

concentrations. The energy costs of metal tolerance mechanisms are partially responsible

for this phenomenon (trade-off hypothesis). There are, however, exceptions (e.g. the

nickel hyperaccumulator Berkheya coddii and arsenic hyperaccumulator ferns),

indicating the capacity of a plant to accumulate and tolerate high metal concentrations in

shoots and to produce high amounts of dry matter is not always mutually exclusive

(Robinson et al., 1997; Ma et al., 2001).

Plant species used in the phytoextraction of arsenic should be able to thrive in a

contaminated site and at the same time accumulating a substantial amount of arsenic in

the aboveground tissues. They should also be responsive to agricultural practices

designed to enhance arsenic accumulation and to allow repeated planting and harvesting

of arsenic-rich biomass (Tu and Ma, 2002). However, continuous phytoextraction

depends on the natural ability of a plant to accumulate, translocate and tolerate high

concentrations of arsenic over the entire growth cycle (Garbisu and Alkorta, 2001) as

well as the arsenic bioavailability in the soil.

Hyperaccumulator Plants

Plants show several response patterns when growing in the presence of toxic

metals, such as tolerance, indicator and hyperaccumulation (Figure 2-2). Most plants are

sensitive even to low concentrations, others have developed resistance and a reduced

number behave as hyperaccumulators of a metal (Baker and Brooks, 1989; Schat et al.,

1999). Tolerant species are those that can grow in a soil with metal concentrations toxic

to most other plants. However, they are not necessarily indicators or hyperaccumulators,









as tolerant non-accumulators can exclude metals from entering the root tissue. Both

indicator species and hyperaccumulators are also tolerant (Assuncgo et al. 2001; Bert et

al. 2003). On the other hand, hyperaccumulators take up particularly high amounts of a

toxic element in their shoots during normal growth and reproduction cycle (Baker and

Whiting 2002). Metal resistance in species with exclusion strategy is frequently based on

reduced metal uptake into the roots, preferential storage of metals in the root vacuoles

and restricted translocation into the shoots. Hyperaccumulators, in contrast, take up more

metals, store a lower proportion of them in the root vacuoles, and export higher amounts

to the shoots.

Hyperaccumulators are plants that can take up and concentrate greater than 0.1%

of a given element in their tissue (Brooks, 1998). Metal hyperaccumulation is a rare

phenomenon in terrestrial higher plants. To date, over 400 plant species have been

identified as metal hyperaccumulators, representing <0.2% of all angiosperms (Brooks,

1998; Baker et al., 2000; Ma et al.' 2001). Approximately two-thirds of the known

hyperaccumulators are Ni accumulators. This is because of the widespread occurrence of

Ni-rich ultramafic (serpentine) soils and the long history of geobotanical studies of

ultramafic floras. Plant species that are able to hyperaccumulate Cd, Co, Cu, Pb, Zn, and

As are much less numerous.

Hyperaccumulator plants have evolved internal mechanisms that allow them to

take up and tolerate high concentrations of metals that would be toxic to other organisms

(Lasat, 2002). They are adapted to the particular environmental conditions of their habitat

and high metal accumulation may have contributed to their defense against herbivores

and fungal infections (Martens and Boyd, 2002).









The definition of a hyperaccumulator has to take into consideration not only the

metal concentration in the aboveground biomass, but also the metal concentration in the

soil. Both bioaccumulation factor (BF) and translocation factor (TF) have to be

considered while evaluating whether a particular plant is a hyperaccumulator (Ma et al.,

2001). The term BF, defined as the ratio of metal concentration in plant biomass to that in

the soil, has been used to determine the effectiveness of plants in removing metals from

soil. The term TF, defined as the ratio of metal concentrations in the shoots to those in

the roots of a plant, has been used to determine the effectiveness of plants in translocating

metals from the roots to the shoots (Tu and Ma, 2002). Therefore, an arsenic

hyperaccumulator plant should have BF > 1 and TF > 1 as well as total accumulation >

1,000 mg kg-1 arsenic in plant biomass.



Hyperaccumulator




q Excluder


Indicator





Metal Cone in soil
Figure 2-2. Conceptual response of hyperaccumulator, indicator and excluder plants for
metal concentrations in the aboveground tissues in relation to increasing metal
concentrations in a soil. (Adapted from Ghosh and Singh (2005).

While some plants can survive in an environment containing high concentrations

of metals, they may not show an ability of hyperaccumulating the metals. For example,

Agrostis tenuis concentrated 3, 470 mg kg-1 of arsenic when growing in a soil containing









as high as 26, 500 mg kg -1 of arsenic. Even though the plant arsenic concentration was

very high, it can not be characterized as an arsenic hyperaccumulator. This is because the

BF and TF were both lower than one.

Arsenic Hyperaccumulation by Pteris vittata

Arsenic hyperaccumulators were discovered only recently and the majority of

them are fern species in the Pteris genus. Ma et al. (2001) reported the first known

arsenic hyperaccumulator plant, Pteris vittata L, commonly known as Chinese Brake

fern. The plant was found on a site contaminated with the wood preservative chromated

copper arsenate in Central Florida. Afterwards, other arsenic hyperaccumulator ferns

have been identified in Florida and elsewhere (Francesconi et al., 2002; Zhao et al., 2002;

Srivastava et al., 2005).

Other non fern plants have also been reported as "arsenic hyperaccumulators"

(>1,000 |tg g-1 As) growing on mine wastes from various sites in the United Kingdom

(Porter and Peterson, 1977) and on smelter wastes in northeast Portugal (De Koe, 1994).

However, though accumulating large amounts of arsenic, these plants do not concentrate

arsenic, i.e. arsenic concentrations in the plant are lower than those in the soil. Hence,

they should not be classified as arsenic hyperaccumulators.

Although there are a number of ferns known to accumulate arsenic, P. vittata is by

far the most studied arsenic hyperaccumulator plant. It is native from China (Nelson,

2000) but is widespread in the old world, occurring from Europe to Asia. In the U.S., this

fern grows in the southeast and southern California (Ma et al., 2001). In Florida, P.

vittata is one of the only three naturalized exotic ferns (Nelson, 2000). Pteris vittata ferns

are very diverse and have survived in great numbers and adapted to both a vastly









changing environment and competition from seed plants. They are widespread thriving in

both temperate and tropical climates (Matschullat, 2000). Also, the distribution of P.

vittata is controlled by its requirement of a well-drained alkaline substrate exposed to

abundant sunshine (Ma et al., 2001).

Pteris vittata has an extraordinary capacity not only to tolerate and take up arsenic

but also translocate large amounts of arsenic to the fronds, with frond concentrations

reaching levels up to 100-fold greater than the soil concentrations. In addition, this plant,

with its high biomass, is an easy-to-grow, vigorous perennial that is resistant to disease

and pests, and exhibits a high arsenic accumulation rate. While natural arsenic

concentrations in most plants seldom exceed 1 mg kg-1 (Porter and Peterson, 1977), after

6 weeks of growth P. vittata accumulated 438-755 mg kg-1 As in a uncontaminated soil

and 3,525 to 6,805 mg kg-1 As in a contaminated soil (Ma et al., 2001). It also removed

26% of the added arsenate after 18 weeks of growth (Tu and Ma, 2002). Liao et al.

(2004) showed that, within seven months, P. vittata could extract up to 7.8 % of the

arsenic from a soil containing 64 mg kg-1 As.

The ability of P. vittata to take up high concentrations of arsenic and sequester it

into aboveground portions when grown on either uncontaminated or arsenic-enriched

soils implies that the fern has highly effective arsenic scavenging mechanisms. Another

interesting characteristic of the ferns is their pereniality, which allows for successive

cuttings of the aboveground biomass while growing in contaminated soils.

An intriguing observation was made by Tu et al. (2002). In a pot experiment, they

observed that addition of 50 mg kg-1 arsenate to a sandy soil increased the fern biomass

by 107%. After 12 weeks of growth, P. vittata produced more aboveground biomass in









soils containing 50 and 100 mg kg-1 arsenic compared to the control. The increase in

plant biomass may be due to the arsenic-enhanced phosphate concentrations in the soils.

However, when the soil arsenic concentration reached 200 mg kg-1, there was a slight

decrease in fern biomass. Another interesting fact is that arsenic has been shown to leach

from P. vittata fronds as they senesce (Tu et al., 2003). This may pose a potential

drawback to the use of P. vittata in phytoremediation of arsenic contaminated soils, as the

arsenic may be returned to the soil.

To date, the only non-Pteris fern to exhibit the ability of arsenic hyperaccumulation

is Pityrogramma calomelanos (Francesconi et al., 2002). Its fronds accumulated 2,760 to

8,350 mg kg-1 arsenic when growing in soil containing 135 to 510 mg kg-1 arsenic.

Interestingly, the fronds with the greatest arsenic concentration were collected from the

ferns growing in the lowest arsenic concentration in the soil (135 mg kg-1 As)

(Francesconi et al. (2002). In the case of P. vittata, studies have shown that the arsenic

concentration in the fronds increase with increasing arsenic concentration in the soil.

Arsenic Species in Pteris vittata

The form of arsenic accumulated by plants is important in determining its

suitability to remediate arsenic contaminated sites, because the arsenic-rich fronds will

need to be disposed properly to avoid further contamination (Ma et al., 2001). Pteris

vittata has been shown to be capable of taking up both inorganic and organic arsenic

species including arsenate, arsenite and monomethylarsonic acid (MMA), concentrating

up to 93% of the arsenic in the fronds (Ma et al., 2001; Kertulis et al., 2005). Research on

arsenic hyperaccumulation by P. vittata showed that arsenic exists in the plant mostly as

inorganic species, and up to 94 % of the arsenic in the fronds is present as arsenite.

Similar results were observed when arsenic was supplied to the ferns in several different









forms (Zhang et al., 2002; Tu et al., 2003; Kertulis et al., 2005). Regardless of the arsenic

species supplied to the fern, >90% of the total arsenic in the roots is present as arsenate,

versus approximately 94% arsenite in the fronds. In both studies, low concentrations of

organic arsenic were found in the fern. Similarly, in a study involving P. calomelanos,

most of the arsenic in its fronds was present as arsenite. Only trace amounts of

monomethylarsonate (MMA) and dimethylarsinate (DMA) were found in a few samples

(Francesconi et al., 2002).

The uptake rates of arsenite and arsenate by P. vittata roots may not be equal,

because the species are taken up through different mechanisms. For instance, Wang et al.

(2002) found that arsenate was taken up more quickly by P. vittata than was arsenite,

especially in the absence of phosphate. The authors suggest that this is due to arsenate

being assimilated via phosphorus-suppressible uptake system in the roots.

Arsenic Tolerance and Detoxification Mechanisms

Plants often contain trace concentrations of contaminants of concern. At low levels,

plants can usually metabolize or dispose of these compounds without significant injury.

Generally, at high concentrations in soil or water, plants are not able to metabolize

contaminants. However, some plants can survive and even grow well when they

accumulate high concentrations of toxic elements as is the case of the hyperaccumulator

plants. In ferns like P. vittata, arsenic is taken up at high rates and at concentrations

proportional to the arsenic concentrations in the growth media at least up to a certain

point where arsenic availability becomes a limiting factor and its arsenic concentration

exceeds its detoxification ability (Ma et al., 2001; Zhang et al., 2002; Wang et al., 2002;

Kertulis et al., 2005).









The fact that P. vittata could survive in a soil spiked with 1,500 mg kg-1 arsenic and

concentrate 2.3 % of arsenic in its biomass indicates that it is equipped with efficient

mechanisms for detoxifying accumulated arsenic. Mechanisms employed by plants to

detoxify metals include chelation, compartmentalization, biotransformation and cellular

repair (Salt et al., 1998). For example, heavy metals are generally transported and

deposited in a vacuole as metal chelates. Baker et al. (2000) reported that the solution

concentration of free metal ions taken up by plants into their tissues is reduced greatly

when they are chelated by specific high-affinity ligands (like oxygen-donor, sulfur-donor,

or nitrogen-donor ligands). Sulfur-donor ligands (like metallothioneins and

phytochelatins) form highly stable complexes with heavy metals, because sulfur is a

better electron donor than oxygen. In fact, the reduction of arsenate to arsenite is

catalyzed by glutathione (GSH) in microbes (Rosen, 2002). However, this has not been

demonstrated in P. vittata.

The formation of arsenite-tris-thiolate complexes has been demonstrated both in

vivo and in vitro by electrospray ionization mass spectroscopy (Schmoger et al., 2000)

and x-ray absorption spectroscopy (Pickering et al., 2000) with the thiolate derived from

either GSH or phytochelatins (PCs). The majority of the reports on arsenic toxicity in

plants show a clear role for PCs in the detoxification of arsenic (Schmoger et al., 2000;

Hartley-Whitaker et al., 2001, 2002; Schat et al., 2002). Reina et al. (2005) demonstrated

that PCs were the most abundant thiols in white lupin under higher arsenic exposure

levels than other plants could tolerate. Together, GSH and PCs were able to complex the

majority of arsenic in the shoots. However, the role of PCs seems to be minor in arsenic

hyperaccumulator ferns (Zhao et al., 2003; Raab et al., 2004; Zhang et al., 2004).









As for arsenic hyperaccumulators P. vittata and P. calomelanos the reduction of

arsenate to arsenite inside plant cells occurs as well (Ma et al., 2001; Francesconi et al.,

2002). This reduction of arsenate inside the plant cells is intriguing because arsenite is

more toxic than arsenate. Additionally, P. vittata was shown to have only 4.5% of its

arsenic completed with phytochelatins, as a glutathione-arsenite- phytochelatin complex

(Zhao et al., 2003). In a study by Raab et al. (2004), the arsenic hyperaccumulator, P.

cretica, had only 1% of its arsenic completed with phytochelatins. The conclusion

reached in both studies was that the phytochelatins may act as shuttles for transporting

arsenic in a non-toxic form through the cytoplasm and into the vacuoles. However,

arsenic complexation with phytochelatins by itself does not account for the high efficient

detoxification mechanism in arsenic hyperaccumulator ferns, suggesting a novel

mechanism of arsenic tolerance in P. vittata.

Arsenic detoxification in microorganisms includes methylation and

biotransformation. Some bacteria enzymatically reduce arsenate to arsenite by Ars C, and

the arsenite is then pumped out by the membrane protein Ars B (Cai and Ma, 2003).

None of these mechanisms were identified in the ferns. Therefore, an important gap in the

arsenic hyperaccumulation mechanism is how the ferns rapidly translocate arsenic from

the roots to the fronds, and still able to survive the exceedingly high concentrations of

arsenite in the fronds, which may perturb the cellular function by disrupting the

sulfhydryl groups of proteins.

Effects of Heavy Metals on Arsenic Hyperaccumulation

Many contaminated sites contain a variety of contaminants. So another important

point to be considered in phytoremediation is the ability of arsenic hyperaccumulator

plants to grow in soils contaminated with other heavy metals. The ability of P. vittata to









survive in soil contaminated with Cu and Cr is confirmed by the fact that this fern was

first discovered in a soil contaminated with CCA (chromated-copper-arsenate). The total

arsenic in that soil was 131 mg kg-1, Cr was 40.6 mg kg-1, and Cu was 8.30 mg kg1.

Fayiga et al. (2004) studied the effects of Cd, Ni, Pb, and Zn on the arsenic

accumulation by P. vittata. They observed the concentrations of heavy metals (50 and

200 mg kg-1) negatively affect the growth of the fern. Still total plant biomass increased

by 12-fold after 8 weeks of growth.

However, the efficiency of arsenic accumulation by the fern depended on the

concentration of the metals in the soil (Fayiga et al., 2004; An et al., 2006). In the study

of Fayiga et al. (2004), arsenic uptake decreased with an increase in the metal

concentration except in Pb-treated soils. Though effective in taking up arsenic, P. vittata

had a limited capability to take up other metals. Arsenic transfer factors ranging from

15.8 to 46.1, indicating that P. vittata was able to effectively translocate arsenic in the

presence of other metals. Caille et al. (2004) reported that P. vittata grew poorly in a soil

heavily contaminated with As, Cu, Pb and Zn, probably as a result of Zn and Cu toxicity.

The concentrations of Zn and Cu in the plants were well above the toxicity thresholds of

100-500 mg kg-1 Zn and 20 mg kg-1 Cu, respectively (Kabata-Pendias and Pendias, 2001).

Phytotoxicity of Zn and Cu not only decreased plant growth, but also arsenic uptake by

P. vittata, resulting in a negligible phytoextraction of arsenic from the soil.

Contamination of multiple metals or metalloids, particularly at high concentrations, thus

presents a challenge for phytoremediation.

Effect of Phosphate on Arsenic Hyperaccumulation

The competitive effect of arsenate and phosphate in soils (Livesey and Huang,

1981; Manning and Goldberg, 1997; Smith et al., 2002; Tu and Ma, 2003) and their









interactions in plants (Asher and Reay, 1979; Meharg and Macnair, 1992; Fourquerean

and Cai, 2001) have been demonstrated. Basically, the uptake of arsenate and phosphate

by plants has been reported to be competitive (Tu and Ma, 2003). Moreover, arsenate

interferes with phosphate metabolism causing toxicity in plants. On the other hand,

phosphate may be able to alleviate arsenate toxicity by improving phosphate nutrition

(Sneller et al., 1999). Thus, the presence of phosphorus in contaminated soils plays a role

in the phytoextraction process.

Tu and Ma (2003) studied the influence of phosphate on the arsenic

phytoextraction by P. vittata. They observed that, at low and medium arsenate levels (50

to 200 mg kg-1, respectively), phosphate had slight effect on arsenate accumulation by

and growth of P. vittata. However, phosphate substantially increased plant biomass and

arsenate accumulation by alleviating arsenate phytotoxicity at high arsenate levels (400

mg kg-1). The authors suggested a minimum P/As molar ratio of 1:2 in soil solution for

effective removal of arsenic by the plant.

Fayiga et al. (2006) reported that addition of phosphate rock in a soil spiked with

arsenic and other metals increased plant arsenic uptake, from 608 to 1,046 mg kg-1 in the

fronds. This shows the ability of phosphate rock to aid in arsenic accumulation by P.

vittata in a multi-metal system by reducing the toxic effects of the metals on the fern;

thereby, enhancing plant arsenic uptake. Boisson et al. (1999) also reported an increase in

arsenic uptake by plants (Zea mays cv. Volga and Phaseolus vulgaris cv. Limburgse

vroege) after applying hydroxyapatite to a soil contaminated with Zn, Pb, Cu, Cd and As.

They suggested that the increased phosphate concentration in the soil solution might be









responsible for increased plant uptake, since phosphate can displace adsorbed arsenate

from soils and increase the availability of arsenate (Smith and Naidu, 1998).

In a chromated-copper-arsenate contaminated sandy soil, addition of a large

amount of rock phosphate (15 g kg-1) also increased arsenate uptake by P. vittata (Cao et

al., 2003). Therefore, phosphorus additions, in the first place, enhanced plant growth, and

secondly, mobilized exchangeable arsenic resulting in an increased total arsenic uptake.

Despite the appeal of increased arsenic availability from application of soil

amendments such as phosphorus fertilizers for the purpose of phytoremediation, it is

environmentally paradoxical in that the benefit of enhanced arsenic removal from soils

may be offset by the risk of increased arsenic leaching into ground water. Therefore, the

rate of arsenic uptake should be higher than the rate of arsenic release in the soil. In this

regard, a very extensive root system is a valuable characteristic in a hyperaccumulator

plant.

Mycorrhizae Association and its Role in Arsenic Hyperaccumulation

Mycorrhizae have a well-documented role in increasing the plant uptake of

phosphorus (Smith and Read, 1997) and other poorly mobile elements, and are

recognized as an important component of bioremediation strategies for soils

contaminated with heavy metals (Khan et al., 2000). In fact, mycorrhizae are integral,

functional components of plant root systems and the fungi involved can play an important

role either in alleviating metal toxicity in plants or enhancing metal uptake. For instance,

Weissenhorn and Leyval (1995) reported higher concentrations of heavy metals in plants,

even resulting in toxic levels, due to arbuscular mycorrhizae (AM) colonization, while

Heggo et al. (1990) observed reduced metal concentrations in the shoots due to









mycorrhizal colonization. On the other hand, Galli et al. (1995) found no effects exerted

by AM fungi on metal concentrations in plant shoots and roots.

Since arsenic and phosphorus are chemical analogues, it is expected that

mycorrhizal symbiosis may be involved in arsenic uptake as well as P uptake by arsenic

hyperaccumulator plants. However, few studies have been done in this regard. Ying et al.

(2004) investigated the colonization and diversity of AM fungi associated with common

pteridophytes and found no defined relationship between mycorrhizae and plants under

field condition. Out of all arsenic hyperaccumulator ferns, only association of P. vittata

with mycorrhizae has been investigated (Al agely et al., 2005; Leung et al., 2006).

In a study on the effect of increasing levels of arsenic and phosphorus on P.

vittata infected with mycorrhizae, Al Agely et al. (2005) observed that the AM fungi not

only tolerated arsenic amendment, but its presence increased the frond biomass at the

highest arsenic application rate (100 mg kg-1). The AM fungi also increased arsenic

uptake across a range of phosphorus levels, while phosphorus uptake was generally

increased only when there was no arsenic amendment, suggesting an important role of

AM fungi in arsenic accumulation by P. vittata.

The percentage of mycorrhizal infection in P. vittata increased as soil arsenic

concentrations increased (Leung et al., 2006). The authors also reported an increase in

arsenic accumulation by P. vittata infected with mycorrhizae and improvement of the

nutrient status of the plants. The presence of mycorrhizae likely increased the amount of

P transporters at hyphae level and, consequently, the amount of arsenic taken up by the

plant. The presence of mycorrhizae also seems to be related to the formation of thiol like

glutathione (Schutzendubel and Andrea, 2002), a phenolic defense system. It is






26


hypothesized that the production of thiols by P. vittata, induced by the increased arsenic

concentration, alleviated the toxicity of arsenic (Cai et al., 2004) and the plant can readily

retain arsenic with the aid of mycorrhizae (Diaz et al., 1996). Therefore, it is possible that

P. vittata may derive some benefit from the symbiosis by forming low molecular weight

thiols.














CHAPTER 3
ARSENIC ACCUMULATION BY Pteris vittata L FROM ARSENIC
CONTAMINATED SOILS

Introduction

Arsenic contamination in soils, groundwater, and drinking water is a serious

concern as it can affect both human and animal health. The extensive use of arsenic

compounds such as pesticides, insecticides, defoliants, wood preservatives, and soil

sterilants in the past (Azcue and Nriagu, 1994) has left a legacy in the history of arsenic

contamination in the environment and a burden for the future generations.

For instance, pesticides containing arsenic trioxide were largely used in cattle dip

vats in the southeastern USA to eradicate cattle fever tick resulting in many soils

contaminated with arsenic (Thomas et al., 2000). The production of the wood

preservative containing chromated-copper-arsenate (CCA) represented 67 % of the

arsenic used in 1992. Besides the contamination of soils in the sites of operation, CCA-

treated wood can potentially contaminate soils through the leaching of arsenic from the

wood. Smelting and mining activities represent significant sources of arsenic

contamination because pyrometallurgical production processes lead to large emissions of

Pb, Zn, Cu, Cd and As (Boisson et al., 1999). Moreover, herbicides containing

monosodium methane arsonate (MSMA) are still used for weed control in Florida (Cai et

al., 2002).

The use of plants to remove arsenic from soils is a fairly new technology. It is an

efficient and less costly way to treat contaminated soils. Hyperaccumulator plants possess









highly efficient mechanisms to acquire and concentrate arsenic in the plants. However,

not all contaminated sites are suitable to be treated by means of phytoextraction. This is

because the arsenic concentration in a soil, its availability and the depth of arsenic

contamination can be limiting factors. Moreover, the plant to be used for this purpose

must be tolerant to arsenic, adapted to the local soil and climate characteristics and take

up a large amount of arsenic.

During phytoextraction process, several crops of hyperaccumulator plants may be

needed to sequentially reduce soil arsenic concentration (Raskin et al., 1994). However,

continuous phytoextraction depends on the natural ability of a plant to accumulate,

translocate and tolerate high concentrations of metals over the whole growth cycle

(Garbisu and Alkorta, 2001).

Pteris vittata is a fern species identified by Ma et al. (2001) as an efficient

arsenic-hyperaccumulator plant. Because it hyperaccumulates an extremely high level of

arsenic (up to 2.3%) in its aboveground biomass and it is easy to grow in a variety of soil

environments, this plant has the potential to clean up arsenic contaminated sites

nationwide, and potentially worldwide. Furthermore, the perennial nature of P. vittata

make the phytoextraction process even more cost-effective since no replanting after

harvest is needed. However, practical issues such as the time required to achieve a given

target level, the long term efficiency of the process, and the arsenic pools depleted by the

plant still need to be addressed.

The aim of this study was thus to 1) assess the efficiency of repeated harvests of

the fronds of P. vittata growing in soils with different sources of arsenic contamination,

and 2) investigate the effects of plant arsenic uptake on arsenic redistribution in soils.









Materials and Methods

Soil Collection and Characterization

Five arsenic contaminated soils plus a soil with naturally high arsenic (Marl soil)

were used for this study. The soils were: contaminated with arsenical insecticide (Avon

soil), arsenical wood preservative (CCA soil), arsenical pesticide (CDV soil), arsenical

herbicide (EDS soil), and mining activities (Mining soil). Selected properties of the soils

are shown in Table 3-1. The soils were collected, air-dried, and passed through 2 mm

sieve. They were analyzed for: CEC by the ammonium acetate method (Thomas, 1982),

organic matter content by the Walkley Black method (Nelson and Sommers (1982), soil

texture by the pipette method (Day, 1965) and soil pH in 1:2 soil: water. Concentrations

of P, Ca, Mg, and trace elements were determined using the EPA Method 3050.

Experimental Design and Statistical Analyses

Six soils with different sources of arsenic contamination were used to grow P. vittata.

Each treatment was replicated four times and pots without plants were included as

controls. The pots were arranged in a completely randomized design.

SAS software was used for all statistical analyses (SAS Institute, 1987). Means and

standard deviations of arsenic concentrations were calculated for different treatments.

Analysis of variance was used to assess significant differences among treatments.

Experimental Set up

Pteris vittata used for the experiments were of similar age and size. One-year-old

ferns were transferred (1 per pot) to 2- gallon-size plastic pots filled with 4 kg of arsenic

contaminated soil. The soils were thoroughly mixed with Osmocote, extended time-

release base fertilizer (N-P-K = 18-6-12) at a rate of 2 g kg-1 soil. The fertilization was









applied annually. After transplanting, the ferns were watered to 60 % of the field

capacity.

Table 3-1. Selected chemical and physical characteristics of six arsenic contaminated
soils


Soil characteristic Marl Avon CCA CDV EDS Mining
pH 7.85 6.70 7.00 6.76 6.70 6.75
CECa 26.6 4.50 4.40 16.8 22.8 12.0
OM (g kg-1) 1.80 19.6 11.0 26.5 28.0 4.20
Total As (mg kg-1) 22.2 26.5 110 211 640 214
Extractable Asc (mg kg-1) 22.7 14.8 58.8 45.9 499 22.5
Extractable P (mg kg-1) 0.36 31.9 24.8 19.2 96.0 8.07
Extractable Ca (mg kg-1) 21000 1892 2960 2040 1602 4920
Extractable Mg (mg kg-1) 120 28.8 130 64.8 336 63.6
Fe oxides (mg kg-l)b 3.96 509 1322 5577 1117 1626
Al oxides (mg kg-l)b 5.23 632 884 1163 1056 179
Sand (g kg-1) 401 866 882 840 851 807
Silt (g kg-1) 410 89 91 140 117 152
Clay (g kg-1) 189 45 27 20 32 41
a:CEC: cation exchange capacity (Cmolc kg-1)
b: Amorphous Fe and Al extracted by 0.2 M oxalic acid + ammonium oxalate solution
c: Extractable elements were obtained using Mehlich III solution.

The first harvest was performed in the month of October, 2003, four months after

the transplant. The second (April, 2004) and third (October, 2004) harvests were

performed in a six months interval from the first harvest and from each other. The plants

aboveground biomass was analyzed for dry biomass weight and total arsenic

concentration.









Sampling, Digestion and Analysis

The plant samples were dried in a 65C oven for approximately 48 hours, weighed

and then ground into powder through a 1-mm mesh screen using a Wiley Mill for

digestion. Also, soil samples collected together with the plant samples were air-dried and

analyzed.

The soil samples were digested according to the EPA Method 305 la and the plant

samples a modified EPA Method 3051 (Chen and Ma, 1998; Komar et al., 1999). One

blank, one Standard Reference Material from National Institute of Standards and

Technology (NIST), one duplicate and one spiked sample were included for every 20

samples. Soil and plant samples were digested with nitric acid using a Hot Block

digestion system (Environmental Express, Mt. Pleasant, SC; EPA Method 3051).

Approximately 0.5-1.0 g of air-dried soil or 0.1-0.5 g of dry plant sample were

mixed with 1:1 HNO3 and allowed to set for approximately 24 hours. They were heated

at 105C for 2 hours and then cooled for 3 minutes. The samples were mixed with 1 ml

of 30% H202 and placed on the block digester for 15 additional minutes. After the

second heating, the samples were cooled completely and diluted to a 50 mL volume with

distilled water. A filter cartridge was placed in the bottom of the digestion tube. The

digested samples were analyzed for As concentration with a SIMMA 6000 graphite

furnace atomic absorption spectrophotometer (GFAAS, Perkin-Elmer, Norwalk, CT)

using the EPA SW 846 method 7060 A.

Arsenic Fractionation

Since it is assumed that plant arsenic uptake will substantially reduce soil arsenic

availability, it is important to determine how plant arsenic uptake affects arsenic









redistribution in a soil. A sequential extraction procedure was used to account for the

changes in different soil arsenic-pools as a function of plant uptake. According to Wenzel

et al. (2001), soil samples were fractionated into five arsenic fractions with decreasing

availability: (N) non-specifically bound, (S) specifically bound, (A) amorphous hydrous

oxide-bound, (C) crystalline hydrous oxide-bound and (R) residual (Table 3-2). Samples

from each fraction, with the exception of the residual fraction, were centrifuged at 3,500

rpm for 15 min and 20C after each extraction and/or wash. The supernatants were

collected and filtered through Whatman 42 filter paper and analyzed for arsenic

concentration using GFAAS.


Table 3-2. Arsenic sequential extraction procedure
Fractions Extracting solution
N- Non-specifically- bound (NH4)2SO4 0.05M
S- Specifically-bound (NH4)H2P04 0.05M
A-Amorphous hydrous NH4-oxalate buffer
oxide-bound (0.2 M) pH 3.25
C-Crystalline-hydrous- NH4-oxalate buffer
oxide-bound (0.2 M) + ascorbic
acid (0.1M) pH 3.25


Extraction condition
4 h shaking, 20C
16 h shaking, 20C
4 h shaking, 20C


30 min in a water
basin at 96 +3 C in
the light


R-Residual HN03/H202 Hot block digestion 1:50
*SSR=Soil solution ratio

Single Extractions

Water-extractable arsenic was obtained using a 1:10 soil to deionized water ratio,

shaking for two hours and centrifuging at 3500 rpm for 15 min (Olsen and Sommers,

1982). Mehlich III method (Mehlich, 1984) was commonly used as a standard procedure

in soil testing to assess the amount of P available in the soil for growing plants. As a

phosphorus analogue, this method was chosen to correlate plant arsenic availability in


SSR*
1:25
1:25
1:25


1:25









soil as well. Mehlich III extractable arsenic was obtained using Mehlich III extracting

solution (0.2 M CH3COOH, 0.25 MNH4NO3, 0.015 MNH4F, 0.013 MHNO3 and 0.001

MEDTA) after shaking for 5 min. Arsenic was also extracted with 1.6 mM organic acids

(phytic acid -1.4mM + oxalic acid 0.2 mM), in a 1:20 soil to solution ratio, shaking for

24 hours and centrifuging at 3500 rpm for 15 min. The concentrations of organic acids

used in this study were similar to those reported by Tu et al. (2004) in the root exudates

ofP. vittata. All the extracts were filtered through Whatman 42 filter paper and analyzed

for arsenic concentration using GFAAS.

Results and Discussion

Plant Growth

The aboveground biomass production and the re growth capacity are important

factors in the phytoextraction of arsenic using perennial plants since multiple harvests

will be needed in order to remove the arsenic from the soil (Fayiga and Ma, 2005).

Despite the different soil properties and arsenic concentration among the six soils,

the ferns grew well in all six soils during the first four months of the experiment (June-

October 2003), without showing any toxicity symptom (Figure 3-1). The plants biomass

was uniform in all soils ranging from 24.8 to 33.5 g plant-', greater than the biomass data

reported for P. vittata growing in pot experiments (Tu and Ma, 2002; Fayiga et al., 2004).

It should be pointed out that the 12-month-old plants used in this study were small, with

5-6 fronds.

There was a significant reduction in the frond biomass production in the

subsequent harvests, except for the ferns growing in the Marl and Mining soils in the

third harvest. For all soils, there was a significant decrease in the frond biomass in the









second harvest (40, 81, 82, 84, 79, and 71 % for the Marl, Avon, CCA, CDV, EDS and

Mining soils, respectively), likely due to the seasonal effect (cooler climate from October

to April). For all soils, October harvests yielded larger plant biomass than the April

harvest, except for the Marl and EDS soils.


Marl -22 Avon -27 CCA 110 CDV 211 Mining- 214 EDS 640
Soil


Figure 3-1. Fond biomass ofPteris vittata after first, second and third harvests in six
soils. Means followed by the same letter within the same soil are not different
by the Duncan test at p < 0.05.

Due to the alkaline nature of the Marl soil, condition that favors the fern growth,

dry matter yield was comparable in all three harvests. On the other hand, plants

drastically failed to regrow in the EDS soil after the first harvest, probably because of the

extremely high solubility of arsenic in this soil (Table 3-1).

Even though P. vittata has shown to be very tolerant to arsenic exposition, Tu and

Ma (2002) reported that addition of 500 mg kg-1 As to a sandy soil reduced the fern









biomass by 64 %, a common symptom of arsenic toxicity (Kabata Pendias and Pendias,

2001). However, in their study, addition of 200 mg kg-1 As did not affect biomass

production, suggesting a much higher tolerance of P. vittata as compared to normal

plants. The difference in arsenic concentrations in the Avon, CCA and CDV soils (Table

3-1) did not influence the dry matter yield since all had the same trends among different

harvests (Figure 3-1). However, the better performance of the CCA soil was likely due to

the higher pH and Ca concentration, and also due to the fact that this was the soil where

P. vittata was found naturally growing (Ma et al., 2001).

Overall, the regrowth capacity of the plants was related to the temperature effect.

That is, the difference in plant performance between harvests is due to colder temperature

prevailing during December to April. According to Jones (1998), P. vittata prefers

warmer climate. In a field experiment, Kertulius (2005) also reported greater biomass

production when P. vittata plants were harvested in the months of October and

December. Another factor that, associated with the temperature, could also explain the

poor regrowth capacity of the plants after the first harvest was the clipping procedure. All

the fronds, including the coiled young fronds (fiddleheads), were removed in the first

harvest. This did not happen in the second harvest.

The effect of successive fern cultivation on plant biomass and arsenic removal by

P. vittata is relatively scarce and inconsistent. For instance, pot experiments using P.

vittata have shown that the amount of biomass harvested decreased after two or three

successive cuttings as well as the amount of arsenic extracted from a soil (McGrath et al.,

2002). However, Kertulius (2005) found positive results when growing P. vittata under

field conditions. Likely, the exploitation of a greater soil volume and the amount of









residue produced at field condition may account for some of the differences between the

studies in pot (McGrath et al., 2002) and field condition (Ketulius, 2005).

Arsenic Concentrations in the Frond Biomass

The total arsenic concentration in the soils used in this study varied from 22 mg

kg-1 in the Marl soil to 640 mg kg-1 in the EDS soil (Table 3-1). The arsenic concentration

in the EDS soil was greater than the arsenic level of 500 mg kg-1 used by Tu and Ma

(2002), which was spiked as arsenate. Despite the high arsenic level, the plants grew well

and took up arsenic from all soils.

The frond arsenic concentrations ranged from 166 to 6,151 mg kg-1 in the first

harvest, from 110 to 3,056 mg kg-1 in the second harvest, and from 162 to 2,139 mg kg-1

in the third harvest (Table 3-3). The highest arsenic concentrations in plants were

observed in the EDS (from 2139 to 6151 mg kg-1), CDV (477 to 1872 mg kg-1) and

Mining (423 to 1079 mg kg-1) soils (higher soil arsenic concentrations) and the lowest in

the Marl (110 to 166 mg kg-1) soil (the lowest soil arsenic concentration).

Even though the Marl and Avon soils had similar arsenic concentrations, the

plants took up more arsenic from the Avon soil.

There was no significant difference in the frond arsenic concentrations of P.

vittata in the first and third harvest for all soils, except for the CDV and EDS soils.

However, cooler climate conditions not only affected plant biomass production but also

plant arsenic accumulation. Fern frond arsenic concentrations were 34, 46, 51, 75, 61 and

50% lower in the second harvest compared to those in the first harvest for the Marl,

Avon, CCA, CDV, Mining and EDS soils, respectively (Table 3-3).









Table 3-3. The frond arsenic concentrations ofPteris vittata after the first, second and,
third harvest in six soils.
Soil As levels Frond arsenic (mg kg -1)

mg kg 1 Harvest 1 Harvest2 Harvest 3

Marl 22 166 a 110b 162 a

Avon 27 336 a 181 b 280 a

CCA 110 659 a 325 b 747 a

CDV 211 1872 a 477 b 715 b

Mining 215 1079 a 423.4 b 962 a

EDS 638 6151 a 3056 b 2139 b

Means followed by the same letter in a row and within the same soil are not different by
the Duncan test at P < 0:05 within the same soil.
The frond arsenic concentrations increased with soil arsenic concentrations, in all

three harvests with a correlation coefficient of 0.97-0.98. However, the ability of the

plants to remove arsenic from the soils reduced with time. The highest arsenic

concentration (6,151 mg kg-1) was observed in the plants growing in the EDS soil, which

also had the highest soil arsenic concentration, both total and extractable (638 and 498

mg Kg-1 As, respectively; Table 3-1).

The high organic matter content of the EDS soil likely plays a role in the

availability of arsenic. However, little information is known on arsenic reactions with

organic matter. In fact, it is suggested that soil organic matter contributes very little to the

arsenic sorption in soils due to their anionic nature (Livesey and Huang, 1981; Wenzel et

al., 2001a). Therefore, considering the sandy nature and arsenic concentration of the EDS

soil, phytoextraction might not be a good alternative for this soil, especially if there is any










risk of arsenic exposure. Besides the long time required to cleaning up, there is the risk of

potential arsenic leaching.

Plant Arsenic Removal from Soils

Pteris vittata had significantly higher arsenic accumulation (arsenic concentration

X plant biomass) in the first harvest as compared with the second and third harvest from

the Avon, CDV and EDS soils (Figure 3-2).Therefore, the first harvest accounted for

most of the arsenic removed from these soils. In the Marl, CCA and Mining soils, the

accumulation of arsenic in the first harvest was similar to that of the third harvest.


200
E Harvest 1 (October 2003)
[] Harvest 2 (April 2004)
160 O Harvest 3 (October 2004)


0 120


80


40


0
Marl 22 Avon 27 CCA- 110 CDV- 211 Mining 214 EDS 640

Total Arsenic (mg kg-1)
Figure 3-2. Arsenic accumulation by P. vittata after first, second and third harvests in six
soils. Values represent mean + standard deviation (n=4).

If the Mining soil is not considered, arsenic accumulation followed the trend

arsenic concentrations in the soils. The total arsenic removed by the plants varied from

1.71 to 4.54 mg pot- in the Marl soil and from 17.0 to 175 mg pot- in the EDS soil.

More espicificaly, the plants removed 10, 13.5, 42.7, 70.9, 53.3, and 210 mg arsenic per









pot from the Marl, Avon, CCA, CDV, Mining and EDS soils, respectively (Figure 3-2)

after the three harvests. Even though arsenic hyperaccumulation did occur in the ferns

growing in the EDS soil, the rate of arsenic removal was low relative to the amount of

soluble and extractable arsenic in this soil.

The percentage of arsenic removed from each soil by the fern uptake was used as

a parameter of relative comparison of the soils with a wide range of arsenic

concentration. For instance, the plants removed 8.2 % of arsenic from the EDS soil

(higher arsenic content) after the three harvests, while the plants growing in the Marl and

Avon soils (soils with the lowest arsenic content) removed approximately 12 % of the

soil arsenic. The lower percent of arsenic removal from the EDS soil as compared to the

Marl and Avon soils was likely due to the failure of the plant in regrowing after the first

harvest. Similarly, the percent of arsenic removed from the CCA soil (9.5 %) was greater

than from the Mining soil (5.8 %). In this case, despite the similar growth of the plants in

both soils, most of the arsenic in the Mining soil is unavailable to the plants.

In general, the accumulation of arsenic by P. vittata translates to arsenic reduction

of 6.4 to 13 % (Figure 3-3). In spite of the greater amounts of arsenic extracted from the

first and third harvest as compared to the second harvest, it was still difficult to project

the time required for soil arsenic reduction to meet the required limits because the plants

failed to regrow after the third harvest.

Soil Arsenic Distribution

Despite all the criticisms to the fractionation procedures, it is important in

environmental research because it may provide useful information on the reactivity,

mobility and availability of the elements (Wenzel et al., 2001).










16

14




10 -
8-

2 6

4

2

0
Marl 22 Avon 27 CCA- 110 CDV 211 Mining 215 EDS 638
Soil
Figure 3-3. Arsenic removed from the soil by Pteris vittata after three harvests. Bars
represent mean standard deviation (n=4).

Figure 3-4 shows arsenic distribution in the five fractions based on sequential

extraction before and after plant growth. Among the five fractions, the N and S fractions

are considered to be the most plant-available, whereas R the least plant-available. The

sum ofN and S fractions constituted 17.5, 22.0, 32.1, 21.7, 4.33 and 46.3 % of the

arsenic in the Marl, Avon, CCA, CDV, Mining and EDS soils, respectively (Figure 3-4).

As expected, arsenic in these soils was primarily associated with the A fraction, ranging

from 40.0 to 58.7 %, except for the Marl soil which had a substantial amount of the

arsenic (59.1%) associated with the C fraction and only 9.28 % associated with the N+S

fraction. In the Mining soil, the A and C fractions constituted 84.3 % of the total arsenic.

In the EDS soil, the majority of the arsenic was distributed among the first three fractions

in the order A > S > N, further explaining the high arsenic availability in this soil (499

out of 640 mg kg-1 total As was extractable As).






































ArAenic fraction


N S A C R
Arsenic fractiorm


Figure 3-4. Arsenic concentrations (mg kg-1) in different fractions in soils before and
after plant growth. Values represent mean + standard deviation (n=4). N=
non-specifically bound, S= specifically bound, A= amorphous hydrous oxide-
bound, C= crystalline hydrous oxide-bound and R= residual fraction.

Wenzel et al. (2001) fractionated arsenic in 20 arsenic contaminated soils and also

showed that most of the arsenic was associated with the amorphous fraction. The

different soils with different sources of arsenic contamination used in this study

confirmed that, regardless of the source or form of arsenic that enter the soil, arsenic

reacts with soil components and becomes less available. It seems that the predominant

reactions occur with hydrous Fe and Al oxides which mostly coat soil particles (Pierce

and Moore, 1980; Smith et al., 1998).









Plant growth influenced arsenic mobilization from all five arsenic fractions in the

soils, except the residual fraction, which is the most stable one. Among the fractions, the

A fraction contributed the most for the arsenic reduction (45.3, 48.0, 72.0 and 59.0 %) for

the Avon, CCA, CDV and mining soils, respectively (Table 3-4). This was not observed

for the Marl and EDS soils (Figure 3-4; Table 3-4). For instance, in the Marl soil, the

highest arsenic mobilization occurred in the C fraction, which was also the fraction with

the highest arsenic concentration. Even though the greatest arsenic mobilization in the

EDS soil occurred in the S fraction (46.4 %), both the N and A fractions also contributed

to a great extent. The treatments without plants showed that the arsenic mobilization was

primarily a result of plant growth, although some arsenic mobilization was also observed

in the absence of plants. This was probably due to wetting and drying cycles during the

experimental period.

Although arsenic has a high affinity to soil, when the environmental condition

changes, the mobility and even speciation of arsenic can change. For instance, the

presence of plants leads to the exudation of some organic anions into a soil, which

directly impacts the mobilization of arsenic in the soil.

Total arsenic concentration is not a good predictor of its bioavailability because

only arsenic dissolved in water can be transported to the roots and taken up by plants

(Ritchie and Sposito, 1995). Plants tend to first take up the most available fraction of

arsenic from the soils and as this pool becomes smaller, some of the arsenic from other

fractions will be slowly transformed to water-soluble fraction to reestablish their

equilibrium (McGrath et al., 2000).









Table 3-4. Arsenic reduction (%) in each fraction in six soils as a result of plant arsenic
uptake by P. vittata.
As removed
Soil (mg pot-) N1 S A C R
----------------------- % As ------------------------

Marl 9.93 13.9 9.65 3.47 62.32 10.6
Avon 13.5 20.9 27.8 45.3 4.71 1.28
CCA 41.9 15.0 16.3 47.9 16.2 4.60
CDV 72.1 5.60 18.9 71.6 1.85 2.03
Mining 50.0 1.17 7.18 58.7 26.9 5.98
EDS 210 28.0 46.4 21.2 3.26 1.56
1 N= non-specifically bound, S= specifically bound, A= amorphous hydrous oxide-
bound, C= crystalline hydrous oxide-bound and R= residual fraction.
2 the fraction with the highest arsenic reduction for a given soil was in bold font.

As plants take up more arsenic, there is less available arsenic present in the soils,

resulting in reduced available arsenic in soils and reduced plant arsenic uptake.

Therefore, it is reasonable to assume that, only arsenic in N fraction can be readily taken

up by the plant and arsenic in other fractions has to convert to N fraction before being

taken up by the plant. In other words, N fraction acts as both a source of arsenic for the

plant and a sink for arsenic in the soil, thus linking arsenic in the soil with the arsenic

taken up by the plant.

High proportion of the total arsenic in historically polluted soils is mostly present

in the residual fraction (Voigt et al., 1996; Kavanagh et al., 1997). Thus, most of the

schemes commonly used tend to underestimate arsenic availability (Gleyzes et al., 2001).

However, the present results show that the presence of arsenic in the residual fraction was

low, even in the soils where the presence of arsenic is of natural origin, such as Marl and

Mining soils. This could indicate that the scheme we used seems to be more suitable for

the study of arsenic fractionation in these polluted soils.









Comparison of Extractable Arsenic Using Single Extraction Methods

In this study, three single extraction methods were used in order to relate arsenic

availability in soils and plant uptake. The arsenic concentrations extracted with these

methods are shown in Table 3-5. Generally, the extractable arsenic followed the

descending order of Mehlich III > organic acids > water-soluble in all soils. When

calculated in percent of the total arsenic determined using the EPA 3050 method, the

water soluble arsenic, Mehlich III and organic acid extractable arsenic represented 1.62,

102 and, 28.2 % (Marl), 5.81, 54.8 and, 49.3 % (Avon), 5.32, 53.4, 12.1% (CCA), 0.06,

21.7 and, 18.9 % (CDV), 1.46, 10.5 and, 3.96 % (Mining) and 21.3, 78 and 38 % (EDS),

respectively. The Mehlich III was expected to extract more arsenic than the water and

the organic acids, irrespective the soil because of its chemical composition, a dilute acid-

fluoride-EDTA solution at pH 2.5. The result showed that the presence of low molecular

weight organic acids in soils can be potentially dangerous as it increase the mobility and

availability of arsenic in the soil. For instance, in the Avon and CDV soils, the organic

acid extractable arsenic was comparable to the Mehlich III extractable arsenic. Therefore,

the presence of typical plants that do not hyperaccumulate arsenic in arsenic

contaminated soils has the potential to mobilize arsenic from the soil since the exudation

of organic acids is a common phenomenon in the rhizosphere of the plants.

Plant uptake and Arsenic Availability in Soils as Measured by Single Extractants

Arsenic phytoextraction from contaminated soils depends, among other factors,

on arsenic bioavailability and the capacity of a plant to access arsenic. Arsenic

bioavailability is regulated by physical, chemical and biological processes and their

interactions, which makes the simulation of element availability to plants with chemical

extractants a difficult task (Carter, 1993). On the other hand, the total concentration of an









element is only important for evaluating the toxicity potential of soils, but they say little

about elemental bioavailability.

Table 3-5. Concentrations of arsenic extracted by different extraction methods in six soils
before plant growth.
Soil Arsenic extracting solutions
Water Mehlich III Organic acids Total
------------------------------- As (mg kg-1)----------------------
Marl 0.36 0.15 (1.6)a 22.7 1.70 (102) 6.26 0.89 (28) 22.2 4.11
Avon 1.57 + 0.05 (5.9) 14.8 + 0.88 (55) 13.3 0.05 (49) 27.0 + 1.00
CCA 5.85 0.29 (5.3) 58.8 + 5.02 (54) 13.3 1.03 (12) 110+ 1.73.
CDV 0.12 0.03 (0.1) 45.9 20.5 (22) 39.9 1.00 (19) 211 2.65
Mining 3.13 0.15 (1.5) 22.5 1.00 (11) 8.49 0.97 (4.0) 214 16.5
EDS 136 4.04 (21) 498 + 57.5 (78) 242 30.1 (38) 640 5.52
Results are mean standard deviations (n=3). a: Results in the brackets are percentage of
the total.
In this section, the relationship between the soil arsenic extracted by three

different methods and the arsenic taken up by the plants was evaluated.

Water-soluble arsenic grossly underestimated the plant arsenic uptake by 242%

(based on the average of five soils, excluding the CDV soil), whereas the Mehlich III

over-estimated by 72%. Therefore, neither water nor Mehlich III solution was good

indicators of the plant available arsenic. On the other hand, arsenic taken up by P. vittata

constituted 74% of the arsenic extracted using organic acids, only 26% underestimation.

Based on arsenic extracted with water, the arsenic taken up by P. vittata implies

that a greater amount of arsenic was mobilized from other pools over time. This was

particularly true for the CDV soil in which the amount of arsenic taken up by the plants

was over 15,500 times greater than the amount of water-extractable arsenic (Table 3-6).









Because low molecular weight organic acids used in this study are products of

root exudates, microbial secretions, and plant and animal residue decomposition in soils

(Stevenson, 1986; Zhang et al., 1997), and they are found in the rhizosphere ofP. vittata

(Tu et al., 2004), we hypothesized that the organic acids extractable arsenic would be a

good predictor of plant available arsenic in soils. Phytic acid and oxalic acid were the two

main low molecular weight organic acids identified in the root exudates ofP. vittata (Tu

et al., 2004). Furthermore, the role of oxalic acid in mobilizing mineral elements has been

widely reported (Jones, 1998).

In fact, in this study, the organic acid extractable arsenic was the best method to

predict plant arsenic uptake by P. vittata (Table 3-6).

Arsenic bioavailability in soils has been found to be enhanced by plant root

exudates (Dinkelaker et al., 1989; Ohwaki and Hirata, 1992), suggesting that root

exudates could be important in the mobilization of soil arsenic and arsenic accumulation

by a plant.

Table 3-6. Percent of the arsenic taken up by P. vittata in each extracting method.

Soil Water-Soluble Mehlich III Organic acids
Marl 836* 11 40
Avon 216 23 95
CCA 183 18 81
CDV 16475 45 46
Mining 435 60 159
EDS 39 11 22
Average 342# 28 74
*Obtained by dividing the amount of arsenic taken up by P. vittata by the amount of
water-soluble arsenic. A value of 836% indicates that P. vittata took up 736% more
arsenic than the amount of water soluble arsenic.
# excluding the CDV soil.









The percent of arsenic taken up by P. vittata according to the organic acids

extraction method varied with soils. The best correlation was found for Avon (95 %),

CCA (81 %) and Mining (159 %). In the Mining soil, P. vittata took up 59 % more

arsenic than predicted by the organic acid method. However, the plants took up 60, 54

and 78 % less arsenic than predicted for the Marl, CDV and EDS soils, respectively.

Tu et al. (2004) reported that P. vittata exuded 40-106% more phytic acid and

300-500% times more oxalic acid under arsenic stress than a non arsenic

hyperaccumulator, N. exaltata fern. It also mobilized significantly greater amounts of

arsenic from both arsenic minerals and a CCA soil (data not shown). Under the same

concentrations, phytic acid mobilized more arsenic from the soils than oxalic acid, likely

due to its stronger complexation capability and greater acidity (Tu et al., 2004).

Oxalic acid has been widely used as an extractant for plant available nutrients

from soil, including phosphate (Fransson, 2001; Sagoe et al., 1998) due to its typical

presence in plant root exudates (Pinton et al., 2001) and its capability of both proton

donation and ion complexation.

Arsenic extraction with water underestimated the availability of soil arsenic to P.

vittata in all soils except the EDS soil. This was because 21% of the arsenic was water

soluble in the EDS soil (Table 3-5) whereas the plant took up 39% water-soluble arsenic,

i.e., the amount of arsenic taken up by the plant was less than the amount of water-

soluble arsenic in the EDS soil. On the other hand, the concentrations of water-soluble

arsenic were low in all other soils. The amount of arsenic taken up by P. vittata after

three harvests was consistent with the fact that more arsenic was solubilized over time

from other pools (Figure 3-3).









Because Mehlich III extracts a large proportion of total soil arsenic, it

overestimated plant available arsenic. The amount of arsenic taken up by P. vittata was

lower than the amount of arsenic extracted using Mehlich III extractant. Only Mehlich III

extractable arsenic in the Mining soil was close to the amount of arsenic taken up by the

plant (60 %). Results from this study support the conclusion of McLaughlin (2002) that

indices of metal phytoavailability in soils that are based on the determination of the most

available metal pools (water-soluble and organic acids), while not perfect, appear to

better predict risk from metal contamination than do total metal concentrations or metals

removed from soils by strong extractants (Mehlich III). Therefore, organic acids might be

more suitable than water and Mehlich III to assess arsenic availability to P. vittata in the

soils used in this study.














CHAPTER 4
EFFECTS OF PLANT AGE ON THE ARSENIC ACCUMULATION BY THE
HYPERACCUMULATION Pteris vittata L.

Introduction

Arsenic is a ubiquitous trace constituent in soil and plants. However, natural and

anthropogenic activities have resulted in many areas polluted with arsenic. The extensive

contamination of soil and water by arsenic worldwide has renewed the interest in more

cost-effective remediation technologies such as phytoremediation/phytoextraction.

The potential use of hyperaccumulator plants such as Pteris vittata L. for the

phytoremediation of arsenic-contaminated soils has been reported (Komar et al., 1998;

Ma et al., 2001; Tu and Ma, 2002). Several desirable characteristics, such as the ability to

concentrate arsenic in the fronds, large biomass, fast growth, easy reproduction,

resistance to adverse soil characteristics and its perennial nature, make P. vittata suitable

for phytoremediation (Ma et al., 2001).

It is known that during the sigmoidal phase of a plant growth cycle, dry matter

and nutrient accumulation occur at a maximal rate, although the rate of nutrient uptake

varies with developmental conditions of a plant and is different for different nutrients

(Garcia et al., 2003). There exists a set of plant physiological characteristics that change

with plant age, which either isolated or combined with other factors influence biomass

production, and nutrient and contaminant accumulation. For instance, young roots are

generally considered to have higher nutrient uptake activity than old roots (Eshel and

Waisel, 1972; Mengel and Barber, 1974: Vegh, 1991; Yanai, 1994) and the root systems









at different growth stages may not have the same specific nutrient uptake activity (Chen

and Barber, 1990; Smethurst and Comerford, 1993). This is important because the

accumulation ability of a plant is highly influenced by the effectiveness of its root

system.

Little information is available about arsenic hyperaccumulation in plants at different

ages. Furthermore, no study has been reported for P. vittata of different ages growing in

soil conditions. Therefore, identifying the plant age that allows maximal arsenic

accumulation may play an important role in successful phytoextraction. The aim of this

study was to assess the influence of the physiological ages of P. vittata on its arsenic

removal from a contaminated soil.

Materials and Methods

Plant Propagation

Pteris vittata plants used in this study were propagated in our laboratory. An

assembly of cool-white fluorescent lamps supplied an 8-h photoperiod to the plants with

an average photon flux of 825 [[mol m-2 s-1

Soil Characterization

The soil used in this study (sandy, siliceous, hyperthermic grossarenic paleudult)

was collected from an abandoned chromated- copper-arsenate (CCA) wood preservation

site in north central Florida. The soil had pH (1:2 soil:H20) 7.30; cation exchange

capacity (Thomas, 1982) 4.4 Cmol(-) kg-1; soil organic matter (Nelson and Sommers,

1982) 11 g kg-'; total As 153 mg kg-1; 880 g kg-1 sand; 90 g kg-1 silt, and 30 g kg-1 clay.









Experimental Design

This greenhouse study, set up as completely randomized design, assessed the

influence of four physiological ages of P. vittata (1.5, 4, 10 and 16 months after

transplant from the sporophytic phase) (Figure 4-1) on the arsenic removal of P. vittata

from a contaminated soil. The plants of different ages were chosen based on their

availability. Each treatment had four replications. Three control pots, soil without

plants, were also included.

Air-dried soil (2.5 kg) was weighed into each pot and thoroughly mixed with 3.0 g

of Osmocote, extended time-release fertilizer (18-6-12) (Scotts-Sierra Horticultural

Products Co., Marysville, OH). After one week of equilibrium under field capacity, one

plant was transplanted into each pot. The pre-experimental frond and root biomass was

taken before transplantation. The plants were allowed to grow for 8 weeks. All plants

were watered throughout the study to keep the soil at approximately 70% of field

capacity.


Figure 4-1. Pteris vittata of 1.5, 4, 10, and 16 old months before transplant.









After 8 weeks of growth, plants were harvested and separated into roots and

fronds. Plant tissues were washed thoroughly with tap water, and then rinsed with

deionized water. The fronds and roots were oven-dried for 3 d at 650 C, weighed and

ground with a Wiley mill to pass through a 1mm mesh screen for chemical analysis. Soil

samples from each pot were collected from the rhizosphere and bulk soil. The

rhizosphere was defined as the soil attached to the roots. The rhizosphere soil was gently

shaken from the roots and sieved through 2 mm mesh screen to separate it from the roots.

Chemical Analysis

Plants were digested using a modified version of the EPA Method 3050A for the Hot

Block Digestion System (Environmental Express, Mt. Pleasant, SC). Analysis was

performed with a transversely heated, Zeeman background correction equipped graphite

furnace atomic absorption spectrophotometer (Perkin Elmer SIMMA 6000, Norwalk,

CT). The analysis of K, Ca, Mg, Zn, Cu, Fe and Mn was performed with a flame atomic

absorption spectrophotometer (Varian model FS 220, Australia) with the same digestate.

Quality control of arsenic analysis was assured by including the Standard Reference

Material 1547 (Peach Leaves). Total P analysis was carried out using a modified

molybdenum blue method (Carvalho et al., 1998). This involved reduction of the As in

digestates from As (V) to As (III) with L-cysteine to minimize its interference with

phosphate analysis. Tightly capped test tubes are incubated at 800C for 5 min to allow

complete reduction of arsenate into arsenite. P was determined, after cooling, by a

double-beam spectrophotometer (Shimadzu UVI60U, Shimadzu Corp., Columbia, MD).









Data Analysis

All results were expressed as an average of four replicates. Treatment effects

were determined by analysis of variance according to the General Linear Model

procedure of the Statistical Analysis System (SAS Institute Inc., 1987). Duncan test at a

5% probability was used for post-hoc comparisons in order to separate treatment

differences.

Results and Discussion

Arsenic Uptake

Arsenic uptake in P. vittata ranged from 3.98 to 6.18 mg plant1 in the fronds and

from 0.94 to 2.84 mg plant- in the roots (Figure 4-2). The As uptake in the fronds of 1.5-

month old P. vittata was 36% greater than those in the 4- and 16-month old plants and

similar to that of the 10-month old plants. On the other hand, the arsenic accumulation in

the roots of the 1.5-month old ferns was lower than the rest. Interesting to note that, even

though the 10-month old plants accumulated as much arsenic in the fronds as those in the

1.5-month old ferns, they accumulated more arsenic in the roots as compared to the plants

of other ages.

Transfer factor (TF) (As in fronds/As in roots) was calculated to evaluate the relative

effectiveness of plants in moving arsenic from the roots to the fronds. This characteristic

of P. vitatta in accumulating high arsenic concentrations in its fronds makes this plant

suitable for phytoextraction. The TF values for 1.5, 4, 10 and 16 months' old plants

were 3.20, 2.05, 1.63 and, 1.60, respectively (data not shown). The TF reduced with plant

age. The 1.5-month old plants were the more efficient in translocating arsenic from the

roots to the fronds.






54










~3--- S





r r
.5 4 10 16
Plsant I.r tage

Figure 4-2. Arsenic accumulation in the fronds and roots of P. vittata of different ages
after 8 weeks of growth in an arsenic-contaminated soil. Means with the same
letter for the fronds or the roots are not significantly different (P < 0.05) based
on Duncan's multiple range test.

Table 4-1. Plant biomass, bioconcentration factor and translocation factor ofP. vittata of
different ages after 8 weeks of growth in an arsenic-contaminated soil.
Plant growth Biomass DW Bioconcentration Translocation

stage (g pot-) Factor Factor

(months) Frond Root Frond Root

1.5 12.7 b 6.8b 2.9 a 0.9b 3.0 a

4 12.0b 8.8b 1.9b 1.1 ab 1.8b

10 20.7 a 16.0 a 1.9 b 1.3 a 1.6b

16 13.1 b 8.9b 2.0b 0.5 c 3.8 a

1Means (3 reps.) followed by the same letter within the column are not significantly
different (P < 0.05) based on Duncan's multiple range test.

Tu et al. (2004) also reported that young fern plants were more efficient in removing

arsenic from a hydroponic system than older ferns. Arsenic levels in the growth medium

seem to determine the extent of arsenic accumulation by plants of different age. For a









given plant, arsenic tends to be preferentially stored in young fronds at low arsenic level

and in old fronds at high arsenic level in the media (Tu and Ma, 2002; Tu and Ma, 2005).

The ability of young P. vittata to accumulate more arsenic may be related to several

factors. The concentration of glutathione, a sulfur-containing tripeptide thiol and a

precursor of phytochelatins, a very important antioxidant involved in plant detoxification

(Scott et al., 1993), seems to decrease with the plant age (Hatton et al. 1996).

In P. vittata, the glutathione concentration tends to increase with the arsenic

concentration in the plant (Cao et al., 2004). Elsewhere, it has also been reported that

plant contaminant absorption decreases as plant is more mature, hence tolerance increase

with age (Wilcut et al., 1989; Leah et al., 1995). Knuteson et al. (2001) reported that

four-week-old parrot feather and canna were more tolerant of simazine than two-week-

old plants. Additionally, the root system at different ages has different nutrient uptake

activities. The average uptake rate per unit of root decreases as a plant matures (Barber,

1984).

The fact that P. vittata ferns are very diverse, easy to adapt to different environments

and are widespread thriving in both temperate and tropical climates (Matschullat, 2000)

raises the concerns on using these ferns for phytoremediation of contaminated sites as

they are classified as type II invasive species. One aspect that contributes to this is their

life cycle and breeding system. The large quantities of spores produced and released by

the ferns are easily dispersed to long distances favoring colonization of wide ecological

niches (Bondada and Ma, 2002). Therefore, the use of younger plants besides being more

efficient in arsenic uptake could also minimize the risk of uncontrolled ferns propagation

since younger plants could grow longer up to the phase of spore maturation in a









contaminated site. The arsenic rich biomass could be harvested before the spores

maturation phase as compared with older plants, which produce spores earlier after being

transplanted in a field.

Plant Biomass

Arsenic hyperaccumulation efficiency depends on both plant biomass production

and arsenic concentration in the plant tissue (Baker et al., 1991; Salt et al., 1998; Ma et

al., 2001). Therefore, the biomass of P. vittata was determined for all plant ages. After 8

weeks of growth, the frond and root biomass of the 10-month old plants was greater than

the rest (Figures 4-3 and 4-4).

Because the initial differences in the frond and root biomass of the treatments, the

similarity of final root and frond biomass among plants of different ages indicated that

plant biomass output occurred at different rates. Therefore, the net biomass (final minus

initial biomass) for each plant age and for each plant part (frond and root) was calculated.

Even though the final biomass of the 10-month old plants was greater than the plants of

other ages, the rate of increase in biomass was greater in the 1.5-month old plants,

suggesting a higher metabolic activity.

Phosphorus Distribution in the Plants

As a P chemical analogue, arsenic is taken up by plants via the phosphate

transport system (Meharg and Hartkey-Whitaker, 2002). Phosphate concentrations in P.

vittata of different ages were determined to examine its relationship with arsenic

concentrations.

During vegetative growth phase, P contents in plants usually ranges from 97 to

161 mM of plant dry matter (Marschner, 2003).










25


o20
--
a
-15



-C
510
.5

a 5


SInitial find
Final frond
a


1.5 4 10 16
Plant growth stage

Figure 4-3. The frond biomass of P. vittata of different ages before and after 8 weeks of
growth in an arsenic-contaminated soil. Means with the same letter are not
significantly different (P < 0.05) based on Duncan's multiple range test.


1 1.


E


Plant growth stage


Figure 4-4. The root biomass of P. vittata of different ages before and after 8 weeks of
growth in an arsenic-contaminated soil. Means with the same letter are not
significantly different (P < 0.05) based on Duncan's multiple range test.


18

16

1 -
~14
-4
S12


8-





2 -

0-
6-


SInltial root
- Final root




--









In this study, P concentrations ranged from 140 to 183 mM in the fronds and from

97 to 124 mM in the roots, which were typical of most plants. Phosphate concentrations

in the fronds and roots increased with plant age (Figures 4-5 and 4-6). As for arsenic, its

concentrations in the roots increased with plant age whereas those in the fronds

decreased. Therefore, accumulation of both arsenic and P in the roots was the highest in

the 16-month old plants. On the other hand, the tissues of older P. vittata had lower

arsenic concentrations in the fronds and higher arsenic concentrations in the roots, as

compared with the younger plants. It is interesting to note that the molar ratios of P/As

stayed approximately one in the roots for all plant age (Figure 4-6) whereas those in the

fronds increased with plant age (Figure 4-5).

The effectiveness in arsenic hyperaccumulation by P. vittata is related to a series

of detoxification mechanisms which are important to mitigate arsenic toxicity in the roots

and fronds (Lombi et al., 2002). One of them is the ability to maintain lower

concentrations of arsenic in the roots and higher in the fronds, thereby minimizing the

potential damage to the roots. The lower arsenic concentrations in the roots of the

younger plants suggest a more effective arsenic removal process by P. vittata.

Another important factor is the ability ofP. vittata to manipulate phosphate in the

plant biomass. Tu and Ma (2003c) reported that P. vittata maintained greater amount of

phosphate in the roots than Nephrolepis exaltata, a non arsenic-hyperaccumulator fern. It

is hypotesized that the ability of P. vittata in maintaining higher concentrations of

phosphorus in the roots may constitute one of its mechanisms of arsenic tolerance (Tu

and Ma, 2003c).







59



200 7
-uP
4As



I ^-------- p
160

5


120





o

2

40




0 0
1.5 4 10 16
Plant growth stage (Month)

Figure 4-5. Effect of plant age on phosphorus and arsenic distribution in the fronds of P.
vittata after 8 weeks of growth in an arsenic-contaminated soil. Bars represent
standard deviations of four replicates.

Concentrations of Other Nutrients in Plants

The K concentrations in the fronds and in the roots of P. vittata were within the

normal range for most plants (Havlin et al., 2005). However, the K concentration in the

fronds of the 1.5-month old plants was higher than those in other ages (Table 4-2).The K

distribution in the fronds and the roots in plants of different ages was similar to that of

arsenic suggesting synergetic relationship of K with arsenic in the plants. Tu and Ma

(2005) reported that arsenic and K had similar distribution in the fronds of P. vittata and

speculated that K may function as a counterbalancing ion in these plants.












160 3









100
140




05
120

2



t 80 ----------------1 5


10 0
80 15





05
2 0 -----------------------------------


20


0 0
1.5 4 10 16
Plant growth stage (Month)


Figure 4-6. Effect of plant age on the phosphorus and arsenic distribution in the roots of
P. vittata after 8 weeks of growth in an arsenic-contaminated soil. Bars
represent standard deviations of four replicates.

The Ca and Mg distribution in the fronds and roots of P. vittata were similar to


that of P, and increased with plant age (Table 4-1). Also, the plants that accumulated


more arsenic (1.5- and 4-month old plants) in the fronds had lower Ca. Tu and Ma (2005)


reported that Ca concentrations in the fronds were inversely related to those of arsenic


and were higher in older fronds growing in a soil spiked with up to 30 mg kg-1 As. Due to


the very low mobility of calcium in the phloem, the growth of the young parts of the plant


is dependent upon the concurrent uptake of calcium.


Among other roles of calcium in plants, the ionic form of calcium is required in


the vacuole of plant cells as a couter-cation for inorganic and organic anions (Marschner,


2001). As arsenic forms compounds with calcium in soil and arsenic is hypotesized to be









stored in the cell vacuoles, it would be reasonable to assume that the calcium

concentration in the plant would increase with increasing arsenic concentrations in the

plant tissues. However, based on the the results of this study and the studies of Tu and

Ma (2005) and Fayiga and Ma (2005), that does not seem to be the case. In fact, the fact

that calcium concentration in the older plants increased while arsenic concentrations

decreased suggests a limited role of calcium in arsenic detoxification by the plants.

Table 4-2. Macronutrients (K, Ca, and Mg) (g kg-1) and micronutrients (Fe, Zn, and Mn)
content (mg kg-1) in the fronds and roots of P. vittata of different ages after 8
weeks of growth in an arsenic contaminated soil.
Elemental Frond growth stage (Month) Root growth stage (Month)

content
1.5 4 10 16 1.5 4 10 16

Potassium 18.8 a 14.2 b 15.6 b 15.3 b 8.70 ab 10.9 a 10.0 a 5.50 b

Calcium 13.5 b 15.5 b 20.4 a 20.7 a 21.5 a 20.0 a 23.4 a 23.8 a

Magnesium 2.80 b 3.40 b 4.10 a 4.20 a 2.10b 1.70 b 2.90 b 16.6 a

Iron 69.2 ab 75.0 a 58.0 bc 45.0 c 1235 b 1397 b 1669 b 5627 a

Zinc 43.6 a 37.0a 39.5 a 44.3 a 114 a 131 a 123 a 122 a

Manganese 15.1 bc 28.6 a 20.3 b 13.9 c 41.3b 40.5b 39.5b 70.4a

Means followed by the same letter in a row and within each plant part are not different
by the Duncan test at P < 0.05

The concentrations of Fe, Mn and Zn were within the range of most plants

(Havlin et al., 2005). While the Fe concentrations in the fronds of the 1.5-and 4-months

old plants were higher, the Fe concentrations in the roots of the 16-month old plants were

higher than those in other plant ages. Tu and Ma (2005) observed an inverse correlation

between arsenic and Fe in P. vittata, but they offered no explanation. Chen et al. (2003)

reported an unusual high concentration of Fe in the epidermis ofPteris nervosa roots,

indicating the formation of Fe plaques. Other studies with non hyperaccumulator plants






62


also showed the formation of Fe plaques and the inhibition of arsenic uptake in roots

(Otte et al., 1991; Colleen et al., 2002).














CHAPTER 5
ARSENIC ACCUMULATION AND ROOT CHARACTERISTICS OF Pteris vittata L.
AND Nephrolepis exaltata L

Introduction

For successful application of phytoextraction, the hyperaccumulator plants must

be metal-tolerant, adapt to soil and climate environment of their intended use, and

effectively absorb the metal from soils (Keller et al., 2003; Schwartz et al., 1999). Since

heavy metals are present in soils predominantly as insoluble forms, their bioavailability is

generally low. Therefore, the efficacy of phytoextraction largely depends on the

development of extensive plant root systems in a contaminated soil (Schwartz et al.,

1999).

Plant root systems perform many important functions including water and nutrient

uptake, anchorage to the soil, and the establishment of biotic interactions in a

rhizosphere. The characteristics of a root system are important in plant nutrient

acquisition, particularly for elements of low mobility such as phosphorus and arsenic.

Plant root systems can be characterized based on length, surface area, biomass and

their relationship to plant shoots (Van Noordwijk and Brouwer, 1991; Jungk, 2002). The

efficiency of roots to transfer metals from a soil to the shoots can be characterized by net

influx per unit root length, and root length per unit of shoot biomass. Structural root

properties, such as diameter of the roots and the formation of root hairs, may also exert an

influence on nutrient uptake from the immediate vicinity of a root cylinder (Waisel and

Eshel, 2002).









According to Ernst (1996), most hyperaccumulators have limited root systems.

However, in a number of studies, unexplained growth enhancements were observed in

plants subjected to a mild stress from toxic metals. For example, the root growth of

Betulapendula seedlings was induced by low concentrations of cadmium (Gussarsson,

1994). Also, enhanced root elongation, root biomass production, and root hair formation

were found in plants of a tolerant population of Silene vulgaris treated with lead chloride

(Wierzbicka and Panufnik, 1998). Jiang and Liu (1999) reported stimulated root growth

of Brassicajuncea plants with low concentrations of lead nitrate. Schwartz et al (1999)

and Whiting et al (2000) reported a preferential root development in the Zn

hyperaccumulator Thiaspi cearulences in the presence of Zn.

Little information is available on the root system ofP. vittata. The process by

which P. vittata mobilizes and takes up arsenic is not well known. Hyperaccumulator

species may release root exudates containing chelators to enhance heavy metal uptake,

translocation and resistance (Wenzel et al., 2003). The ability of a plant to exude large

quantities of dissolved organic carbon (DOC) and to change the rhizosphere pH may

enhance the arsenic bioavailability in soils, thereby increasing its arsenic uptake (Tu et

al., 2004). The fate and bioavailability of arsenic in the rhizosphere can be different from

that of the bulk soil (Fitz et al., 2003). Depending on plant and soil factors, rhizosphere

pH can be up to two units different from the bulk soil (Marschner, 2003). Factors

affecting the rhizosphere, such as pH, plant nutritional status, organic acids excretion, and

CO2 production by roots and rhizosphere microorganisms, are accountable for the

differences (Marschner, 2003).









Several attempts have been made to measure the bioavailability of metals in soils.

Tessier at al. (1979) used a sequential extraction technique for determining the labile

metal in soils. This approach is based on the fact that metal associated with various

geochemical phases varies in its chemical reactivity and bioavailability. The drawback of

the method is its non-selectivity and metal redistribution among geochemical phases

(Howard and Brink, 1999). Despite its limitation, fractionation provides an understanding

of the relative mobility and bioavailability of metals in soils (Fitz and Wenzel, 2002;

Krishnamurti and Naidu, 2002), because plant metal uptake or metal toxicity is related to

those fractions (Chlopecka and Adriano, 1996; Guo and Yost, 1998). Water-extractable

and exchangeable forms of metals are usually considered to be the most available to

plants (Petruzzelli, 1989).

This study was conducted aiming to evaluate and compare two fern species, an

As-hyperaccumulator (P. vittata) and a non As-hyperaccumulator (Nephrolepis exaltata),

in terms of: 1) root system characteristics (biomass, length, surface area and diameter); 2)

plant biomass and arsenic uptake; and 3) root uptake efficiency of arsenic; and 4) effects

of plant arsenic uptake on arsenic distribution and bioavailability in the rhizosphere and

bulk soils. The information obtained from this study should be useful for determining the

plant density and effective root depth ofP. vittata for phytoextraction purpose as well as

for providing the critical linkage between the ability of P. vittata in solubilizing soil

arsenic and arsenic hyperaccumulation.









Materials and Methods

Soil Characterization

The soils used in this study were collected (0-20 cm) from an abandoned

chromated-copper-arsenate (CCA) wood preservation site (As-contaminated soil) and

from a non-contaminated site (control soil) in Florida. The site was contaminated with

arsenic from using CCA wood preservative between 1951-1962 (Komar, 1999). The soil

analyses were performed as described in the Chapter 4 and summarized in Table 5-1.

Table 5- 1. Selected properties of the soils used in this study

Property As-contaminated soil Control soil
pH (1:2 soil/water ratio) 7.30 +0.01* 6.81 +0.02
Organic matter content (g kg-1) 11.0 +0.30 13.2 + 0.42
CEC** (cmol(+) kg-1) 4.40+0.02 5.60+0.04
Total As (mg kg-1) 101 +5.40 0.41+0.01
Water-soluble As (mg kg-1) 0.2 +0.02 0.003+0.001
Mechlic III extractable P (mg kg-1) 21.8 + 0.30 87.2 +0.77
Sand (g kg-) 882 + 4.32 892 +4.76
Silt (g kg-1) 91+0.40 75 +0.30
Clay (g kg-1) 27 +0.01 33 +0.02
Bulk density (g cm-3) 1.29 1.39
Values represent mean standard deviation;
*Cation exchange capacity

Experimental Set up

The study employed a completely randomized design in a 2 x 2 split plot scheme.

Two fern species (P. vittata and N. exaltata) were used as the main plot and the chemistry

of the rhizosphere and bulk soil as the sub-plot. Each treatment was replicated four times.









A known weight (2.5 kg) of air dried and sieved (2 mm) soil was poured into

plastic bags and thoroughly mixed with 3.0 g of Osmocote extended time-release base

fertilizer (18-6-12) (Scotts-Sierra Horticultural Products Co., Marysville, OH), and then

poured into plastic pots (2.5-L rhizopot).


The rhizopot (16 cm in height and 15 cm in diameter; Figure 5-1) was used to

grow the plants. Plastic frames (13 cm in height and 7 cm in diameter) covered with

nylon mesh cloth (mesh size 45 [im) were used to separate the rhizosphere from the bulk

soil in the rhizopot.


Figure 5-1. Cross section of the rhizopot used in the study.

Pteris vittata was propagated in our laboratory whereas N. exaltata was procured

from a nearby nursery (Milestone Agriculture, Inc., Apopka, FL). Efforts were made to

ensure visual uniformity across all plants. The average biomass per plant before the

transplant was 0.70 g for the fronds and 0.45 g for the roots.


45wm mesh
cloth +
plastic net









Root growth was limited within the nylon cloth in the central compartment (500 g

soil). One week before the study, the soil was set to equilibrate at water holding capacity.

One healthy fern of similar age with five to six fronds was transplanted into each pot.

The plants grew for 8 weeks in a greenhouse with an average night/day

temperature of 14/30C and an average photosynthetically active radiation flux of 825

[tmol m-2 s-1. The plants were watered throughout the study to keep the soil at

approximately 70% of its field capacity. At the end of the experiment, the ferns were

harvested and each plant was separated into roots and fronds. Rhizosphere and bulk soil

were collected and used for analysis of arsenic, soil pH and dissolved organic carbon

(DOC). Rhizosphere soil was defined as the soil attached to the roots, which was

removed from the roots by shaking gently. The rhizosphere soil was sieved to remove the

roots while keeping the roots intact as much as possible. The bulk soil was defined as the

soil outside the central compartment.

The plants were washed thoroughly with tap water, and then rinsed with distilled

water. The fronds were oven-dried for 3 days at 650C, weighed and ground using a

Willey mill to 60-mesh fineness for chemical analysis. The roots were immediately

separated and scanned to prevent dehydration. After scanning, the roots were also oven-

dried, weighed and ground for chemical analysis.

Root Measurement

The roots were washed with water, spread out on a transparent tray and then the

image was digitalized using a scanner (Envisions88005) at 400 dpi resolutions with

Vistascan software (UMAX Data system, Inc). When necessary, root samples were

divided into sub-samples for more accurate measurement. The digitalized images were









processed by the program GSROOTS (Guddanti and Chambers, 1993), which was

programmed to yield root length data of pre-defined diameter classes (<0.10; 0.10 < d <

0.25; 0.25 2.0 mm). Root

length density (Lv, root length per unit volume of soil) and root area density (Av, root

surface area per unit volume of soil) were calculated. Unit root length and unit root area

were used in all discussions unless otherwise specified. The root biomass was also

determined.

Chemical Analysis

Plants and soils digestion and analyses were performed as described in the

Chapter 4.

Rhizosphere and bulk soil samples were evaluated for water-soluble arsenic and

DOC in 1:4 soil to water ratio (Olsen and Sommers, 1982), obtained after shaking (1 h),

centrifuging (15 min at 3500 g) and filtering (0.45 |tm syringe filter). Arsenic in solution

was determined by the same method described above. The concentration of DOC was

measured using a TOC-5050A TOC analyzer (Shimadzu, Japan). Quality control of

arsenic analysis was included using Standard Reference Materials 1547 (Peach Leaves)

and 2710 (soil) (US NIST, MD).

Arsenic Fractionation

The improved sequential extraction procedure (Wenzel et al. 2001) was used to

fractionate arsenic into five operationally-defined fractions, including non-specifically

bound (N), specifically bound (S), amorphous hydrous oxide-bound (A), crystalline

hydrous-oxide-bound (C), and residual (R), as described in the Chapter 3.









Data Analysis

All results were expressed as an average of four replications. Treatment effects

were determined by analysis of variance according to General Linear Model procedure of

the Statistical Analysis System (SAS Institute Inc., 1987, Cary, NC). The Duncan test at

5% of probability was used for post-hoc comparisons to separate treatment differences.

Pearson correlation coefficients were calculated between the root characteristics

and different plant parameters.

Results and Discussion

Root Biomass, Root Length, Root Area, and Root Diameter

Plant roots are in direct contact with, and are affected by soil constituents in

addition to regulating many plant processes. Measurement of below ground plant

productivity may provide information on root growth, distribution, water and nutrient

supply and plant potential of removing contaminant (Van Noordwijk and Brouwer,

1991). Root-soil contact is an especially important factor for the uptake of less mobile

elements such as P and arsenic from a soil. However, the available information on the

root characteristics of P. vittata has been limited to the measurement of root biomass (Tu

at al., 2002; Tu and Ma, 2003; Liao et al., 2004).

Root characteristics including root biomass, length and surface areas were

determined for both plants. They are important for estimating plant water and nutrient

uptake (Zhuang et al. 2001). In both soils, P. vittata had a more extensive root system

than N. exaltata (Table 5-2). The root biomass of P. vittata was 3.0 and 2.4 times greater

than that ofN. exaltata in the As-contaminated soil and in the control soil, respectively.









The root biomass accounted for approximately 25% of plant biomass for both plant

species (Tables 5-2 and 5-3).

Several studies have reported that the presence of contaminants stimulated the

growth of plant roots (Gussarsson, 1994; Wierzbicka and Panufnik, 1998; Jiang and Liu,

1999). Other studies indicate a severe inhibition of root growth. For instance, Schwartz et

al. (1999) reported that Thiaspi caerulences growing in a soil spiked with Zn had reduced

root growth while it explored a large soil volume with enhanced Zn acquisition in an

unpolluted soil. However, the root growth of T. caerulescens, a Zn hyperaccumulator,

was higher than that of Thlaspi arvense, a non-hyperaccumulator, when grown in a Zn-

amended soil (Schwartz et al., 1999; Whiting et al., 2000).

Table 5-2. The root biomass (dw), root length, and root area of P. vittata and N. exaltata
after 8 weeks of growth in a As-contaminated soil (As-soil) and control soil.

Root biomass Root length Root area
Fern species (g dw pot-f) (cm cm-3) (cm cm-2
As-contaminated soil


P. vittata 1.8a 4.3a 1.3 a

N. exaltata 0.6b 1.1b 0.4 b
Control soil

P. vittata 2.2a 6.2 a 1.8 a

N. exaltata 0.9b 2.1 b 0.7 b
Means followed by the same letter in a column within the same soil are not significantly
different at p < 0.05 (n=4).

The root length and root area of P. vittata were 3.9 and 3.2 times greater than that

of N. exaltata in the As-contaminated soil, and were 2.9 and 2.6 times greater in the

control soil, respectively (Table 5-2). In our study, the root length and surface area were

lower in the As-contaminated soil than in the control soil for both fern species. Although









the overall chemical and physical characteristics of the two soils (Table 5-1) were similar,

the higher P content and low As content of the control soil must have accounted for the

better development of the plant roots in that soil.

Root diameter determines the extent of the soil volume that potentially supplies

nutrients and contaminants to plants. For both soils, P. vittata had a greater root length

for all classes of diameter than N. exaltata (Figure 5-2). The roots of 0.25 to 1.00 mm of

diameter comprised a major proportion of P. vittata roots as measured by root length. For

both soils and plant species, the roots < 0.50 mm diameter comprised over 40% of the

total root length, while those of 0.50 to 1.00 mm diameter contributed another 25%. The

fine roots (0.25 to 1.00 mm) ofP. vittata contributed more to the total root length than

did N. exaltata. This is important because fine roots allow the root system to exploit a soil

volume more effectively while coarser roots have greater transport capacity (Forde and

Lorenzo, 2001).

For all classes of root diameter, P. vittata had a greater root area than N. exaltata.

The roots with diameter >1 mm represented 67 to 71 % of the total root area for N.

exaltata, whereas they only accounted for 59 to 60 % for P. vittata (Figure 5-3). Based

on the root characteristics (root biomass, length and area) of the two fern species, P.

vittata presented a much greater potential to remove As and nutrients from soils than N.

exaltata regardless of the soil characteristics.

Frond Biomass, Arsenic Uptake, Bioconcentration and Translocation

Good biomass production and high arsenic accumulation ability are two of many

desirable treats of hyperaccumulator plants for a successful application of

phytoextraction. Although the two ferns species had similar frond biomass









(approximately 0.7 g dw plant-) before the experiment, P. vittata had a greater biomass

(5.5 -6.6 g) than N. exaltata (1.8 2.8 g) after 8-week of growth (Table 5-3). The frond

biomass of P. vittata was 3.1 and 2.4 times greater than that of N. exaltata in the As-

contaminated soil and in the control soil, respectively.

This observation confirms the faster growing characteristic ofP. vittata when

compared with other fern species, which is desirable for a plant to be used for

phytoremediation purpose. Other studies have shown that P. vittata can grow normally or

even have the growth stimulated in the presence of arsenic (Tu and Ma, 2002).


The amount of arsenic taken up by P. vittata (2.51 mg per plant) was 29 times

greater than that of N. exaltata (0.09 mg per plant) in the arsenic-contaminated soil

(Table 5-3). Tu and Ma (2004) found a 5.8 fold increase in arsenic accumulation of P.

vittata over N. exaltata in a hydroponic study. Huang et al. (2004) reported that P. vittata

and Pteris cretica, another arsenic hyperaccumulator, removed more arsenic from water

than N. exaltata. While 92% of the plant arsenic was accumulated in the fronds of P.

vittata, only 37% was in the fronds ofN. exaltata (Table 5-3).

The ratio between the contaminant concentrations in the tissue relative to that in

the soil (bioconcentration factor, BF) as well as the partitioning of the contaminant

between the fronds and roots (transfer factor, TF) were determined aiming to gain insight

of the arsenic accumulating capability of the plants.












> 2.00 iN. exalata
S1.00-2.00 9I BP. vitati

0.75-1.00

S0.50-0.75

S 0.25-0.50

o 0.10-0.25

<0.10

0 0.5 1 1.5 2 2.5 3 3.5 4
Lv (cm cm-3)




> 2.00 N. elta
di e i aN.aedsil(tt a
S1.00-2.00 ersn P. vittata

0.75-1.00

S0.50 0.753

0.25-0.50

S0. 10-0.25

<0.10

0 0.5 1 1.5 2 2.5 3 3.5 4

Lv (cm cln-3)




Figure 5-2. The root length density (Lv) of P. vittata and N. exaltata as a function of root
diameter in a control soil (a) and an As-contaminated soil (b) after 8 weeks of
growth. Bars represent standard deviations of four replicates.











>2JO0

1.00-2100

0.75-1.00

mg
: 0.150-0.75

z 0.25-050

S0.10-025

< 0.10


0 0.1 0.2 0.3 OA 0.5 06 0.7 O
Av (cin2 cm-3)


Si"l-"-"- 111111111111111111 I I IIIII--
,RRNMeIIeIIe~


ON. exaltata
[0 P. vittata


0.10-025

<0.10

0 0.1 02 0.3 0.4 0.5


Av (cmn2 cm-3)





Figure 5-3. The root surface area density (Av) of P. vittata and N. exaltata as a function
of root diameter in a control soil (a) and an As-contaminated soil (b) after 8
weeks of growth. Bars represent standard deviations of four replicates


1111 111111111111111111111111111111111 11111111111 I


ON. exaltata
[ P. vittata


> 2.00

1.00-2100

0.75-1JO0

050-0.75


I II III III II III 11 1 1 II I1 11111 1111111111111111111 1-


.- Ill...-..-l l


D


TMTM-

RIIIIIIIIIIIIIIflH









Table 5-3. The frond biomass (dw), arsenic accumulation, and bioconcentration and
translocation factors of P. vittata and N. exaltata after 8 weeks of growth in a
arsenic-contaminated soil and control soil.
Frond As Bioconcentration Transfer
Fern biomass (mg plant-) factor (BF) factor
species (g plant-) (TF)
Frond Root Frond Root
As-contaminated soil
P. vittata 5.5 a 2.30 a 0.21 a 1.7 a 0.5 a 3.6 a
N. exaltata 1.8b 0.03 b 0.06 b 0.1 b 0.4 a 0.2 b
Control soil
P. vittata 6.6 a 0.003 a 0.00 0.55 a 2.5 a 0.2 a
N. exaltata 2.8 b 0.000 b 0.00 0.03 b 0.3 b 0.1 b
Means followed by the same letter in a column within the same soil are not significantly
different at p < 0.05 (n=4).

We assumed that plant root characteristics play a more important role in

determining BF than TF. The fronds (1.7) and roots (0.5) BF of P. vittata was 17 and 1.3

time greater than that ofN. exaltata in the arsenic-contaminated soil (Table 5-3).The fact

that P. vittata had a more extensive and more efficient root system may have contributed

to its greater BF as compared with N. exaltata. The transfer factor (TF=3.6) of P. vittata

was 18 folds greater than that of N. exaltata (Table 5-3). Huang et al. (2004) also

reported that N. exaltata was unable to translocate the absorbed arsenic from the roots to

the fronds, which was expected for a non-hyperaccumulator plant specie. The ability of

P. vittata to translocate arsenic from the roots to the fronds is considered to be one of the

tolerance and detoxification mechanisms involved in its arsenic hyperaccumulation.

Root Uptake Efficiency of Arsenic and Phosphorus

Root uptake efficiency (RUE), which is defined as elemental accumulation in

plant tissues (fronds or roots) per unit root biomass, length or area, can be used to









evaluate how efficient a plant is in taking up an element (Table 5-4). The arsenic RUE

for the fronds of P. vittata measured by root biomass, root length and root area was 15-23

times greater than that ofN. exaltata in both soils. This means that not only did P. vittata

have a greater root biomass but also a more efficient root uptake system for arsenic.

Because N. exaltata tended to accumulate arsenic in the roots, the differences in the

arsenic RUE for the roots between the two plants were smaller, 8-10 times in the control

soil and similar in the arsenic contaminated soil (Table 5-4).

Because arsenic is an analogue to phosphorus, we also determined the P RUE for

the fronds and roots of both plants. The P RUE in the control soil and for roots in the

arsenic contaminated soil was similar between the two plant species (Table 5-4).

However, the P RUE for the fronds was greater in N. exaltata than P. vittata in the

arsenic-contaminated soil.

These RUE results are rather interesting because they indicate that despite the

similarity between P and arsenic, both P. vittata and N. exaltata root systems tended to

remove P from the soil preferentially and more efficiently than arsenic.

Recently, it has been found that some natural hyperaccumulators proliferate their

roots positively in patches of high metal availability.In contrast, non-accumulators

actively avoid these areas, and this is one of the mechanisms by which

hyperaccumulators absorb more metals when grown in the same soil (Schwartz et al.,

1999; Whiting et al., 2000).

In the greenhouse study it was observed that most of the roots of P. vittata tended

to concentrate in the top 7 cm of the pot (data not shown). The observation gathered in

this study points to the need to evaluate the behavior ofP. vittata root system under field









conditions in an arsenic-contaminated soil as well as the effective root depth of the plant.

The answer to this question will help to determine the effective depth that the plant will

be able to clean up arsenic -contaminated soil.

Table 5-4. The root uptake efficiency (RUE) as measured by As and P uptake (mg
plant-) by the fronds and roots of P. vittata and N. exaltata per unit of root
parameters (root dry weight-RB, length- Lv, and surface area-Av) in a
As-contaminated soil and control soil.
Root As-contaminated soil Control soil
characteristics P. vittata N. exaltata P. vittata N. exaltata
Frond
As-RB* 1277 a 61 b 1.7 a 0.1 b
As-Lv** 552 a 31 b 0.6 a 0.04 b
As-Av* 1913 a 85 b 2.1 a 0.1 b
P-RB* 5400 b 10500 a 11600 a 9000 a
P-Lv** 2300 b 5600 a 3800 a 3600 a
P-Av*** 8200 b 14500 a 12500 a 10800 a
Root
As-RB 116 a 92 b 2.6 a 0.3 b
As-Lv 50 a 47 a 0.9 a 0.1 b
As-Av 176 a 124 a 3.2 a 0.3 b
P-RB 2400 a 1700 a 2900 a 2500 a
P-Lv 1100 a 900 a 1000 a 1100 a
P-Av 3700 a 2400 a 3200 a 3200 a
*: ug plant- (As or P)/g root; **: ug plant-1 (As or P)/cm cm-3 root; ***: ug plant- (As or
P)/cm2 cm-3 root.Means followed by the same letter in a row within the same soil are not
different at P < 0.05 (n=4).

Plant Nutrient Uptake

Balanced supply of essential nutrients is one of the most important factors in

increasing crop yields (Fageria, 2001). The frond P (1.8 -3.5 g kg-1), K (1.28 2.51g kg-

1), Ca (1.26-1.58 g kg-1), and Mg (2.8-4.8 g kg-1) concentrations of both fern species









(Table 5-5) were within the normal concentration range for most plant species

(Marschner, 2003). They were also similar to those reported by Tu and Ma (2005).

For the As-contaminated soil, the P and Mg concentrations in the fronds of N.

exaltata were greater than that of P. vittata (Table 5-5). Positive interaction between P

and Mg are expected since Mg is an activator of kinase enzymes and activates most

reactions involving phosphatase transfer (Faglia, 2001). In most plants, P is generally

concentrated in the upper parts or reproductive organs (Marschner, 2003). However, in

this study, the frond P concentration of P. vittata (1.8 g kg-1) was lower than that in the

roots (2.4 g kg-1) in the arsenic-contaminated soil. This result agrees with Tu and Ma

(2004) who found similar trend for P concentration in fronds and roots of P. vittata and

N. exaltata growing in a hydroponic solution. Arsenic hyperaccumulation by P. vittata

may be facilitated by its high arsenic influx rate and its high molar P/As ratio in the roots

resulting from both high arsenic TF and low P TF (Tu and Ma, 2004).

In the control soil, N. exaltata had a greater K and Mg concentration in the fronds

(1.7-1.9 fold) as well as Mg in the roots than P. vittata. Plant potassium content

influences the plant biomass and arsenic content (Komar, 1999). In the present study

neither plant biomass nor arsenic uptake of N.exaltata was related with the status of K in

the plant. It is unclear why N. exaltata took up more K in both soils, especially in the

fronds.

Arsenic Reduction and Distribution in the Soil

Whereas they caused no significant change in total arsenic concentration in the

bulk soil (Table 5-6), N. exaltata and P. vittata reduced arsenic concentrations in the

rhizosphere soil by 28.6 and 40.7% (Table 5-7).









Table 5-5. Nutrients concentrations (g kg-1) in the fronds and roots of P. vittata and N.
exaltata after 8 weeks of growth in an As-contaminated soil and a control soil.
Fern fronds roots

species P K Ca Mg P K Ca Mg

As-contaminated soil

P. vittata 1.8b 13.3a 15.8a 3.2b 2.4a 8.9a 17.3a 3.0a

N. exaltata 3.4a 25.1a 14.9a 4.8a 1.7a 7.8a 17.2a 2.8b

Control soil

P. vittata 3.5a 12.8b 15.0a 2.8b 2.8a 7.6b 16.0a 2.5a

N. exaltata 3.0a 21.7a 12.6a 4.4a 2.5a 10.3a 13.3a 2.6a

Means followed by the same letter in a column within the same soil are not significantly
different at p < 0.05 (n=4).

To achieve the arsenic reduction from 105 to 75.0 mg kg-1 for N. exaltata and 105

to 62.3 mg kg-1 for P. vittata in the rhizosphere soil, the soil mass required to contribute

the 0.09 and 2.51 mg arsenic (Table 5-3) constituted only 3 and 61g out of 2.5 kg,

respectively.

Similar to total arsenic, no change was observed for different fractions of arsenic

in the bulk soil with or without a plant (Table 5-6). Most of the arsenic was present in the

A fraction (61.5%) and C fraction (22.2%), contributing a total of 83.7% (Table 5-6). The

N and R fractions accounted for only 4.63 and 2.93%.

Unlike the bulk soil, the decrease in arsenic concentrations was observed across

all five arsenic fractions in the rhizosphere soil (Table 5-7). Nephrolepis exaltata and P.

vittata reduced arsenic concentrations in the A fraction (As-A) from 66.7 to 42.0 and 66.7

to 30.9 mg kg-', respectively, accounting for 76 and 67% arsenic decrease in the

rhizosphere.









Table 5-6. Arsenic distribution in different fractions in the bulk soil after 8 weeks of plant
growth.
As fractions No plant P. vittata N. exaltata As distribution

---------- mg kg ------------- ----%---

N** 5.00 a* 5.30 a 5.40 a 4.63

S 9.80 a 9.50 a 8.60 a 9.07

A 66.4 a 63.8 a 63.1 a 61.5

C 24.0 a 24.4 a 22.4 a 22.2

R 3.20 a 3.20 a 3.70 a 2.96

Sum 108 106 103 100

Total As 105 99 97

Recovery (%) 103 107 106

Means with the same letter in each row are not significantly different according to
Tukey HSD test at 5 % level.
Fractions N, S, A, C and R stand for non-specifically bound, specifically bound,
amorphous Al and Fe hydrous-oxide bound, crystalline hydrous-oxide bound, and
residual arsenic fraction;
*Based on the treatment with no plant.












Table 5-7. Arsenic distribution in different fractions in the rhizosphere of P. vittata (PV) and N. exaltata (NE) after 8 weeks of plant
growth.
As fractions Soil As concentrations % As in each fraction* As reductions*** % of total As
(mg kg-1) (mg kg-1) reductions***
No plant PV NE No plant PV NE PV NE PV NE

N**"" 4.30 a 2.0 b 4.10 a*1 4.63 4.18 6.03 2.30 0.20 4.38 0.62

S 10.7 a 3.9 c 6.50 b 9.07 8.16 9.56 6.80 4.18 13.0 12.9

A 66.7 a 30.9 c 42.0 b 61.5 64.6 61.8 35.2 24.6 67.0 76.2

C 16.0 a 9.7 b 13.7 ab 22.2 20.3 20.1 6.40 2.30 12.2 7.12

R 2.80 a 1.3 b 1.70 b 2.96 2.09 2.50 1.50 1.10 2.86 3.41
00
Sum 101 a 47.8 c 68.0 b 100 100 100 52.5 32.3 100 100

Total As 105 a 62.3 c 75.0 b 40.6 28.5

Recovery (%) 95.7 76.7 90.7


SMeans with the same letter in each row are not significantly different according to Tukey HSD test at 5 % level.
* Based on total arsenic of 101, 47.8 and 68.0 mg kg-1.
SBased on no-plant treatment for each fraction.
SBased on total reduction of 52.5 and 32.3 mg kg-1.
Fractions N, S, A, C and R stand for non-specifically bound, specifically bound, amorphous Al and Fe hydrous-oxide bound,
crystalline hydrous-oxide bound, and residual arsenic fraction;









Compared to N. exaltata, P. vittata removed more arsenic from each fraction. For

example, reductions in arsenic concentration in the A, S and C fractions were 24.6, 4.18

and 2.30 mg kg-1 for N. exaltata and 35.2, 6.80, and 6.40 mg kg-1 for P. vittata (Table 5-

7). However, plant uptake had little effect on the relative arsenic distribution among the

five fractions in the rhizosphere of two plant species, with 61.8-64.6% in the A fraction,

20.1-20.3% in the C fraction and 8.16-9.54% in the S fraction. This was because arsenic

in all fractions was reduced, with an average reduction of 52.7% and 27.0% for P. vittata

and N. exaltata, respectively (Table 5-7)

The results for arsenic distribution in different fractions obtained in our study

were consistent with the data reported by Wenzel et al. (2001) where they determined

arsenic fractionations in 20 Austria soils with differing levels of arsenic contamination

(96-2,183 mg kg 1). Arsenic in those soils was also mostly present in the A fraction

(42.3%) followed by C fraction (29.2%), accounting for a total of 71.5% or 12.2% less

than that obtained in our study. However, they reported concentration of 0.24 and 17.5%

in the N (most available) and R (least available) fractions. Comparatively, the soil used in

our study had 4.39% more arsenic in the most available fraction and 14.5% less arsenic in

the least available fraction, indicating that arsenic in the soil used in our study had greater

availability than in the soils from their study.

The decrease in arsenic concentration in the five arsenic fractions of the

rhizosphere soil did not follow arsenic availability according to the sequential extraction,

rather it was positively correlated to the arsenic concentrations in each fraction (r = 0.995

for both plants, n=4). This means that the fraction having the highest arsenic

concentration, i.e. the A fraction, had the greatest reduction in its concentration. The fact









that arsenic concentration in the N fraction, the most available, was lower with P. vittata

(2.0 mg kg-1) than N. exaltata (4.1 mg kg-1) was consistent with the greater ability of P.

vittata to take up arsenic. The results also imply that arsenic transfer from less-available

fractions to the N fraction was slower than the arsenic depletion by P. vittata. Fitz et al.

(2003) reported a much lower arsenic transformation in P. vittata rhizosphere in a soil

with a higher buffer capacity. The high rate of arsenic mobilization (reduction in the A

fraction) observed in the rhizosphere of P. vittata can also be attributed to the sandy

nature of the soil (88% sand) (Table 5-1), i.e. low ability to retain arsenic, which was

supported by low oxalate extractable Fe and Al (267 and 260 mg kg-1) (Table 5-1).

The mass balance of the total arsenic in the soil, as determined by the sum of the

fractions, was evaluated (Table 5-6 and 5-7). Good recoveries of total soil arsenic (103-

107% in the bulk soil and 77-96% in the rhizosphere) indicated that the difference

between the two methods ranged from 3 to 24%. Despite all the criticism attributed to

sequential extraction, the procedure was able to provide some insight into the

mobilization and availability of arsenic in contaminated soils. Additionally, the sum of

the fractions was in good agreement with the total arsenic values.

Influence of Plants on Water-Soluble Arsenic, Soil pH and DOC

The water soluble arsenic in the rhizosphere of P. vittata (0.7 mg kg-1) was two

times lower than that in the N. exaltata (1.4 mg kg-1), which was similar to the no plant

treatment (1.7 mg kg-1). Compared to the bulk soil, reduction in water-soluble arsenic in

the rhizosphere soil by P. vittata and N. exaltata were 58.9 and 17.6% (Fig 5-4). The

high water-soluble arsenic in the N. exaltata rhizosphere was consistent with the results

found for arsenic in the N fraction, the most available arsenic.









Concentration of water-soluble metal is a good indicator of bioavailability for

plant uptake in soils (McBride, 1994) since plants preferentially take up nutrients from

soil solution (Linehan et al., 1985). The depletion or the enrichment of an element in the

rhizosphere is determined by the capacity of a soil to replenish the soluble or

exchangeable forms of the element (Hinsinger, 1998).

Concentration of water-soluble metal is a good indicator of bioavailability for

plant uptake in soils (McBride, 1994) since plants preferentially take up nutrients from

soil solution (Linehan et al., 1985). The depletion or the enrichment of an element in the

rhizosphere is determined by the capacity of a soil to replenish the soluble or

exchangeable forms of the element (Hinsinger, 1998).

The supply of elements such as P and As to the plant rhizosphere is limited by

diffusion. When plant outpaces the soil supply capacity, the depletion occurs in the

rhizosphere, as it was the case of P. vittata (Table 5-7).

When the ability of the plant removes arsenic from the soil is lower than the

arsenic diffusion rate, arsenic accumulates in the rhizosphere, as it was the case of N.

exaltata (Table 5-7). Thus, lower water-soluble arsenic in the rhizosphere of P. vittata

was due to the higher ability of the plant to deplete the arsenic from the rhizosphere as

compared with N. exaltata. These results are consistent with the more extensive root

system of P. vittata, as well as with the arsenic reduction in its rhizosphere.

The changes in rhizosphere pH and DOC may have a great effect on soil arsenic

bioavailability. The influence of plant growth on soil pH was reflected by the pH

difference in the bulk and rhizosphere soil. Plants had no effect on the bulk soil pH for all

treatments (Figure 5-5).







86



2.5
Bulk soil
B Rhizosphere




1.5






0.5



0
P. vittata N. exaltata Control

Treatments

Figure 5-4. Comparison of water-soluble arsenic in the bulk and rhizosphere soil of P.
vittata and N. exaltata after 8 weeks of growth in an arsenic-contaminated
soil. Bars represent standard deviations of four replicates.

However, the rhizosphere pH of P. vittata (7.66) and N. exaltata (7.18) were 0.4

units higher and -0.13 lower, respectively, than that in the no plant treatment.


Youssef and Chino (1989) and Luo et al. (2000) also reported an increase of pH in

the rhizosphere of Triticum aestivum and Thlaspi cearulences. Yet Fitz et al. (2003)

found no change in the rhizosphere pH of P. vittata. Since the soil used in their

experiment had high carbonate content and high buffer capacity, their results were not

unexpected. The increase in soil pH observed in the rhizosphere of P. vittata in our study

was more likely due to the low buffer capacity of the sandy soil as well as the

cation/anion balance caused by high excretion of hydroxyl groups in response to high

absorption of arsenic. The pH buffering capacity of the soil and the initial soil pH are the

main factors determining the extent to which plant roots can change rhizosphere pH




Full Text

PAGE 1

EFFECTS OF SOIL AND PLANT ON AR SENIC ACCUMULATION BY ARSENIC HYPERACCUMULATOR Pteris vittata L By MARIA ISIDORIA SILVA GONZAGA 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 2006

PAGE 2

Copyright 2006 By Maria Isidoria Silva Gonzaga

PAGE 3

To the memory of my father Henrique, whos e love and inspiration have accompanied me through the years; to my moth er Ines who, together with my father, gave me all the values that guide my life; and to my kids Jamile, Thomas and Genna, the best fruits of my existence.

PAGE 4

iv ACKNOWLEDGMENTS I wish to thank my supervisory committee chair (Dr. Lena Q. Ma) for providing valuable academic advices, critical suppor t and guidance during my graduate study. I thank my supervisory committee member s (Jean-Claude Bonzongo, Nicholas B. Comerford, Robert Stamps, and Roy Dean Rhue ) for their assistance and advice. I also wish to especially thank Dr. Nick Comerford for helping me apply to the University of Florida and also for helping me and my family during our first few days in Gainesville. Special thanks also go to Dr. Jean-Claude Bonzongo for his advisement and valuable friendship. I am grateful for the assistance and s upport of my husband, Dr. Jorge Santos, who provided experimental and statistical a dvice and also helped in conducting my experiments. I am grateful to Mr. Tom Luongo for his assistance with chemical analysis and proofreading manuscripts; to Dr. Mristunjai Srivastava for his assistance in providing fern plants and proofreading manuscripts; to Dr. John Thomas for his friendship and for proofreading manuscripts; and Mr. Bill Reeve, fo r his friendship and help with chemical analyses. I also wish to thank the member s of the Biogeochemistry of Trace Metals Laboratory (Gina Kertulius-Tartar, Abioye Fayiga, Joonki Yoon, Rocky Cao, Mike Tu, and Autumm Wan), for what they have taught me. I thank Mrs. Marlin Pam for her help and patience. I would like to acknowledge the Brazili an institution (CAPES) for providing financial support during my Ph.D. program. I am grateful to Dr. Paulo Gabriel Nacif (and

PAGE 5

v members of the Soil Chemistry Department of the Agronomy School, at the Federal University of Bahia) for granting me permi ssion to pursue my degree. I also thank Dr. Joelito Rezende for all the encouragement and help, keeping me in the scientific life.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES...........................................................................................................xi ABSTRACT.....................................................................................................................xi ii CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................4 Arsenic........................................................................................................................ ..4 Occurrence, Availability and Toxicity..................................................................4 Remediation of Arsenic Contaminated Soils........................................................6 Phytoremediation..........................................................................................................7 Phytoextraction......................................................................................................7 Phytostabilization..................................................................................................8 Phytoimobilization................................................................................................8 Phytovolatilization.................................................................................................9 Phytoextraction of Arsenic...........................................................................................9 Hyperaccumulator Plants............................................................................................13 Arsenic Hyperaccumulation by Pteris vittata .....................................................16 Arsenic Species in Pteris vittata ..........................................................................18 Arsenic Tolerance and De toxification Mechanisms............................................19 Effects of Heavy Metals on Arsenic Hyperaccumulation...................................21 Effect of Phosphate on Ar senic Hyperaccumulation...........................................22 Mycorrhizae Association and its Ro le in Arsenic Hyperaccumulation...............24 3 ARSENIC ACCUMULATION BY Pteris vittata L FROM ARSENIC CONTAMINATED SOILS........................................................................................27 Introduction.................................................................................................................27 Materials and Methods...............................................................................................29 Soil Collection and Characterization...................................................................29 Experimental Design and Statistical Analyses....................................................29

PAGE 7

vii Experimental Set up............................................................................................29 Sampling, Digestion and Analysis......................................................................31 Arsenic Fractionation..........................................................................................31 Single Extractions................................................................................................32 Results and Discussion...............................................................................................33 Plant Growth........................................................................................................33 Arsenic Concentrations in the Frond Biomass....................................................36 Plant Arsenic Removal from Soils......................................................................38 Soil Arsenic Distribution.....................................................................................39 Comparison of Extractable Arseni c Using Single Extraction Methods..............44 Plant uptake and Arsenic Availability in Soils as Measured by Single Extractants........................................................................................................44 4 EFFECTS OF PLANT AGE ON THE ARSENIC ACCUMULATION BY THE HYPERACCUMULATION Pteris vittata L..............................................................49 Introduction.................................................................................................................49 Materials and Methods...............................................................................................50 Plant Propagation.................................................................................................50 Soil Characterization...........................................................................................50 Experimental Design...........................................................................................51 Chemical Analysis...............................................................................................52 Data Analysis.......................................................................................................53 Results and Discussion...............................................................................................53 Arsenic Uptake....................................................................................................53 Plant Biomass......................................................................................................56 Phosphorus Distribution in the Plants.................................................................56 Concentrations of Other Nutrients in Plants........................................................59 5 ARSENIC ACCUMULATION AND ROOT CHARACTERISTICS OF Pteris vittata L. AND Nephrolepis exaltata L......................................................................63 Introduction.................................................................................................................63 Materials and Methods...............................................................................................66 Soil Characterization...........................................................................................66 Experimental Set up............................................................................................66 Root Measurement...............................................................................................68 Chemical Analysis...............................................................................................69 Arsenic Fractionation..........................................................................................69 Data Analysis.......................................................................................................70 Results and Discussion...............................................................................................70 Root Biomass, Root Length, Root Area, and Root Diameter.............................70 Frond Biomass, Arsenic Uptake, Bioc oncentration and Translocation..............72 Root Uptake Efficiency of Arsenic and Phosphorus...........................................76 Plant Nutrient Uptake..........................................................................................78 Arsenic Reduction and Di stribution in the Soil...................................................79 Influence of Plants on Water-Soluble Arsenic, Soil pH and DOC......................84

PAGE 8

viii 6 ARSENIC ACCUMULATION BY TWO HYPERACCUMULATOR Pteris SPECIES FROM TWO ARSENI C CONTAMINATED SOILS...............................90 Introduction.................................................................................................................90 Materials and Methods...............................................................................................91 Soil Collection and Characterization...................................................................91 Chemical Analysis...............................................................................................94 Data Analysis.......................................................................................................94 Results and Discussion...............................................................................................94 Arsenic Accumulation and Di stribution in the Ferns..........................................94 Plant Biomass of P. biaurita and P. vittata .........................................................98 Rhizosphere Chemistry.....................................................................................100 Elemental Composition.....................................................................................105 7 CONCLUSIONS......................................................................................................111 LIST OF REFERENCES.................................................................................................115 BIOGRAPHICAL SKETCH...........................................................................................132

PAGE 9

ix LIST OF TABLES Table page 3-1 Selected chemical and physical character istics of six arsenic contaminated soils...30 3-2 Arsenic sequential extraction procedure..................................................................32 3-3 The frond arsenic concentrations of Pteris vittata after the first, second and, third harvest in six soils............................................................................................37 3-4 Arsenic reduction (%) in each fraction in six soils as a result of plant arsenic uptake by P. vittata ...................................................................................................43 3-5 Concentrations of arsenic extracted by different extraction methods in six soils before plant growth..................................................................................................45 3-6 Percent of the arsenic taken up by P. vittata in each extracting method..................46 4-1 Plant biomass, bioconcentration f actor and translocation factor of P. vittata of different ages after 8 weeks of growth in an arsenic-contaminated soil..................54 4-2 Macronutrients (K, Ca, and Mg) (g kg-1) and micronutrients (Fe, Zn, and Mn) content (mg kg-1) in the fronds and roots of P. vittata of different ages..................61 51 Selected properties of the soils used in this study....................................................66 5-2 The root biomass (dw), r oot length, and root area of P. vittata and N. exaltata after 8 weeks of growth in a As-contamin ated soil (As-soil) and control soil.........71 5-3 The frond biomass (dw), arsenic accumulation, and bioconcentration and translocation factors of P. vittata and N. exaltata ....................................................76 5-4 The root uptake efficiency (RUE) as measured by As and P uptake (mg plant-1) by the fronds and roots of P. vittata and N. exaltata ................................................78 5-5 Nutrients concentrations (g kg-1) in the fronds and roots of P. vittata and N. exaltata after 8 weeks of growth in an As-c ontaminated soil and a control soil.....80 5-6 Arsenic distribution in di fferent fractions in the bulk soil after 8 weeks of plant growth.......................................................................................................................81

PAGE 10

x 5-7 Arsenic distribution in different fractions in the rhizosphere of P. vittata (PV) and N. exaltata (NE) after 8 weeks of plant growth.................................................82 6-1 Selected properties of th e soils used in this study....................................................92 6-2 Comparison of the arsenic concentration (mg kg-1), translocation factor (TF) and bioconcentration factor (BF) of P. vittata and P. biaurita .......................................96 6-3 The macronutrients (P, K, Ca, Mg) (g kg-1) and micronutrients (Fe, Zn, Mn) content (mg kg-1) in the fronds and the roots of P. vittata and P. biaurita ............106

PAGE 11

xi LIST OF FIGURES Figure page 2-1 Factors that influence the efficiency of phytoextraction..........................................11 2-2 Conceptual response of hyperaccumulato r, indicator and excluder plants for metal concentrations in the aboveground tissues.....................................................15 3-1 Fond biomass of Pteris vittata after first, second and th ird harvests in six soils.....34 3-2 Arsenic accumulation by P. vittata after first, second and third harvests in six soils.......................................................................................................................... .38 3-3 Arsenic removed from the soil by Pteris vittata after three harvests.......................40 3-4 Arsenic concentrations (mg kg-1) in different fractions in soils before and after plant growth..............................................................................................................41 4-1 Pteris vittata of 1.5, 4, 10, and 16 old months before transplant.............................51 4-2 Arsenic accumulation in the fronds and roots of P. vittata of different ages after 8 weeks of growth in an arsenic-contaminated soil.................................................54 4-3 The frond biomass of P. vittata of different ages before and after 8 weeks of growth in an arsenic-contaminated soil....................................................................57 4-5 Effect of plant age on phosphorus an d arsenic distribution in the fronds of P. vittata after 8 weeks of gr owth in an arsenic-contaminated soil..............................59 4-6 Effect of plant age on the phosphorus a nd arsenic distributio n in the roots of P. vittata after 8 weeks of gr owth in an arsenic-contaminated soil..............................60 5.1 Cross section of the rh izopot used in the study........................................................67 5-2 The root length density (LV) of P. vittata and N. exaltata as a function of root diameter in a control soil (a) and an As-contaminated soil (b)................................74 5-3 The root surface area density (AV) of P. vittata and N. exaltata as a function of root diameter in a control soil (a) and an As-contaminated soil (b).........................75 5-4 Comparison of water-soluble arsenic in the bulk and rhizosphere soil of P. vittata and N. exaltata after 8 weeks of growth in an arsenic-contaminated soil.....86

PAGE 12

xii 5-5 Comparison of the soil pH in the bulk and rhizosphere soil of P. vittata and N. exaltata after 8 weeks of growth in an arsenic-contaminated soil...........................88 5-6 Comparison of dissolved organic carbon (DOC) in the bulk and rhizosphere soil of P. vittata and N. exaltata ......................................................................................89 6-1 Pteris biaurita and Pteris vittata growing in an arsenic-contaminated soil.............95 6-2 The frond and root biomass (mg kg-1 DW) of P. vittata and P. biaurita after 8 weeks of growth in two arsenic contamin ated soils (Soil 1: 153; Soil2: 266).........99 6-3 The bulk soil and rhizosphere pH of P. vittata and P. biaurita after 8 weeks of growth in two arsenic contaminated soils..............................................................101 6-4 The bulk soil and rhizosphere dissolved organic carbon (DOC) of P. vittata and P. biaurita after 8 weeks of growth in tw o arsenic contaminated soils.................102 6-5 The water-soluble P concentrations in the bulk soil and rhizosphere of P. vittata and P. biaurita ........................................................................................................103 6-6 The water-soluble As concentrations in the bulk soil and rhizosphere of P. vittata and P. biaurita .............................................................................................104

PAGE 13

xiii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF SOIL AND PLANT ON AR SENIC ACCUMULATION BY ARSENIC HYPERACCUMULATOR Pteris vittata L By Maria Isidoria Silva Gonzaga May 2006 Chair: Lena Q. Ma Major Department: Soil and Water Science Phytoextraction using Pteris vittata L., an arsenic hyperacc umulator plant, could be potentially applied to remediate ar senic-contaminated sites worldwide. Pteris vittata grew well and took up large amounts of arsenic in six arsenic-contaminated soils with different properties and differe nt sources of arsenic contamination. However, efficiency of the ferns in continually taking up arseni c from the soils decreased with time, and varied with growing season. The arsenic concen trations in different chemical fractions before and after plant uptake showed that P. vittata took up arsenic from the most available and also from the less available pools in all soils. The change in arsenic availability in the rhizosphe re was evaluated by comparing P. vittata with a non-arsenichyperaccumulator Nephrolepis exaltata. As expected, P. vittata removed more arsenic from the soil than the non-arsenic-hyperaccumulator fern. Besides the efficient detoxification mechanisms of P. vittata , its more extensive and finer root systems, and its greater capacity to change the soil pH and produce root exudates in the rhizosphere likely

PAGE 14

xiv contributed to the difference. Fu rthermore, the experiments with P. vittata of different physiological ages revealed that younger plants (because of their higher metabolic and growth rate) were more efficient than older plants in taking up arsenic. Therefore, younger plants are recommended for phytor emediation. Comparing two arsenic hyperaccumulators in the genus Pteris showed that arsenic hyperaccumulator ferns diffred in their ability to take up arse nic when growing under the same soils and environmental conditions. The concentrations of arsenic in both plants increased with the soil arsenic concentration, however, P. vittata performed better, regardless of the arsenic level in soils. After 4 weeks of growth, P. biaurita showed signs of stress and probably would not have survived longer under those conditions, while P. vittata showed no toxicity symptom. Our results showed that, in implementing an arsenic phytoextraction project, many aspects related to the plant must be considered.

PAGE 15

1 CHAPTER 1 INTRODUCTION Arsenic contamination in the environmen t from both anthr opogenic and natural sources occurs in many parts of the world a nd is a global problem. Arsenic-contaminated soils, sediments, and sludge are the major s ources of arsenic contamination in food chain, surface water, groundwater, and drinking wate r (Frankenberger and Arshad, 2002). Other potential sources of arsenic contaminati on are the chemicals used extensively in agriculture such as pesticid es, insecticides, defoliants, wood preservatives, and soil sterilants (Azcue and Nriagu, 1994). In many areas, arsenic levels in the envi ronment have exceeded the safe threshold for human health. Epidemiologi cal studies have documented various adverse effects of arsenic on humans and animals. Conventional remediation technologies have been used to clean up metal-contaminated sites because they are relatively insensitive to the heterogeneity in contaminated matrix, and can function over a wide range of oxygen, pH, pressure, temperature, and osmotic potenti als (Cunningham et al., 1997). However, they are expensive and time-consuming, often hazardous to workers, and produce secondary wastes (Lombi et al., 2000). Of the disadva ntages of conventional remediation methods, cost is the primary driving force behind the search for alternative remediation technologies, such as phytoremediation. Phytoremediation is the use of plants and their associated microbes to reduce, remove, degrade, or immobilize environmenta l contaminants in soil and water systems and can be applied for both organic and inorganic pollutants (S alt et al., 1998).

PAGE 16

2 Phytoextraction, one of the st rategies of phytoremediation, ha s attracted much attention as an environmentally-friendl y low-input remediation techni que that uses plants that extract heavy metals from the soil and accumu late them in the harvestable aboveground biomass (McGrath et al., 2002). However, the effectiveness of a phytoremediation plan is plant-dependent. Phytoextraction can be accomplished by using either tolerant high biomass plant species or hyperaccumulator plant species. Plants native to the target area should be considered, since they are adapted to the local climate, in sects, and diseases. Hyperaccumulators are plants that can ta ke up and concentrate more than 0.1% of a given element in their tissue (Brooks, 1998). Metal hyperaccumulation is a rare phenomenon in terrestrial higher plants. The first arsenic hyperaccumulator, Pteris vittata L., was identified by Ma et al. (2001), followed by Pityrogramma calomelanos L. (Francesconi et al., 2002) a nd many other species of the Pteris genus such as P. cretica L., P. longifolia L., P. umbrosa L., and P. argyraea L. (Zhao et al., 2002), and P. quadriaurita L., P. ryiunkensis L. and P. biaurita (Srivastava et al., 2006). Pteris vittata is a fast-growing fern and has th e potential to produce relatively high biomass while accumulating a large amount of arsenic in its aboveground tissue. These characteristics and its ability to surviv e in different environments make this plant an exceptional candidate for use in phytorem ediation of arsenic-contaminated soils. To increase the efficiency of phytoextrac tion of arsenic contaminated soils using P. vittata , besides understanding its ar senic tolerance and detoxi fication mechanisms, it is important to learn more about the biologica l processes involved, su ch as plant arsenic uptake, plant nutritional requirements, rh izosphere processes and mobilization and

PAGE 17

3 bioavailability of arsenic in the soil. Experiments included in our study had the following objectives: 1. Evaluate the efficiency of P. vittata in extracting arsenic from different arseniccontaminated soils 2. Evaluate the importance of the physiological age of P. vittata on its arsenic uptake and plant growth 3. Determine the characteristics of the root systems and changes in the rhizosphere of P. vittata in comparison with a non-hyperaccumulator 4. Compare the efficiency of P. vittata in arsenic phytoextract ion in comparison with Pteris biaurita , another arsenic hyperaccumulator.

PAGE 18

4 CHAPTER 2 LITERATURE REVIEW Arsenic Occurrence, Availability and Toxicity Arsenic concentrations range from below 10 mg kg in non-contaminated soils (Adriano, 1986) to as high as 30,000 mg kg-1 in contaminated soils (Vaughan, 1993). In rocks, arsenic is concentrated in magmatic sulphide minerals and iron ore. The most important arsenic ores include arsenical pyr ite or arsenopyrite (FeA sS), realgar (AsS), and orpiment (As2S3). Arsenic has been identified as a majo r toxic contaminant in many countries. Various human activities have elevated the ar senic levels in soils such as the production and the use of arsenical pestic ides (fungicides, herbicides, an d insecticides). In addition, manufacture of arsenic-based compounds, sm elting of arsenic-containing ores, and combustion of fossil fuels have also contribute d to arsenic contamination in soils, water, and atmosphere (Azcue and Nriagu, 1994). An in crease in industrializ ation has also lead to an increase in the amount of arsenic pres ent in biosolids. Arsenic deposits from the atmosphere, runoff, and effluent s of industries often increase the concentration of arsenic in biosolids. Woolson (1983) re ported a range of 0 to 188 mg kg-1 As dry weight of biosolids. Biosolids are often disposed on land and may then increase arsenic concentrations in the top 20 cm of soil by up to 0.15% (O’Neill, 1990). Sorption onto soil particles is an import ant process for arsenic immobilization in soils. Many studies have been devoted to ar senic sorption on well-characterized minerals

PAGE 19

5 or soil particles (clay, oxides of Al, Fe, and Mn, calcium carbonates, and/or organic matter) (Sadiq, 1997; Dobran and Zagury, 2005). The high affinity of arsenic for oxide surfaces is well known and is affected by se veral biogeochemical factors such as pH, redox potential, and competing ions (Adriano, 2001). The activity of arsenic in soil solution is mostly controlled by surface co mplexation reactions on oxides/hydroxides of Al, Mn, and especially Fe (Inskeep et al., 2002) . The mobility of ar senic in soil is pH dependent as arsenate is prefer entially sorbed onto hydrous oxides in the pH range of 4 to 7; whereas, arsenite is prefer entially sorbed for in the pH range of 7 to 10 (Pierce and Moore, 1982). Frost and Griffin (1977) have shown that arseni c sorption onto kaolinite and montmorillonite is pH dependent, while Lin and Puls (2000) confirmed that arsenate is more strongly sorbed onto these minerals than arsenite. Furthermore, arsenate adsorption onto humic substances reaches a maximum at pH 5.5, while arsenite adsorption peaks at much higher pH of 8.5 (Thanabalasingam and Pickering, 1986). The bioavailability, toxicity, and chemi cal behavior of arsenic compounds are largely influenced by the form and speciation of arsenic. In natural systems, arsenic can exist in four oxidation states: (-3), (0), (+3), and (+5). Ar senate [As(+5)] and arsenite [As(+3)] are the main forms present in soils (Harper and Haswell, 1988). Arsenate prevails under aerobic conditions and is somewhat less toxic and also less mobile than arsenite (dominant form under anaerobic conditions) because arsenate sorbs more strongly than arsenite onto minerals (Pie rce and Moore, 1982). Generally, inorganic arsenicals are more toxic than organic arseni cals while the trivalent oxidation state is more toxic than the pentavalen t oxidation state (Adriano, 2001).

PAGE 20

6 Remediation of Arsenic Contaminated Soils Many remediation techniques are available to address the problems in contaminated sites. However, the relatively high capital expenditure, unsuitability for large areas, and environmental disruption are some of the shortcomings of those techniques. No single soil-reme diation technique is suitable for all situations. Careful investigation of the contaminated site char acteristics, contaminant problem, treatment options and treatment timefr ame must be considered. Current remediation methods for arseniccontaminated soils include soil removal and washing, physical stabilizat ion, and/or the use of chemic al amendments, all of which are expensive and disruptive, with an aver age cost of $ 404,700 per ha (Raskin et al., 1997). Following are some selected curre nt remediation technologies for arseniccontaminated soils (USEPA, 2002a). a. Capping: a hard cover is placed on the surface of a contaminated soil. It is a simple method to reduce contamin ant exposure. However, it does not remove contaminants from the soil. b. Solidification and stabilization: the contaminated soil is mixed with stabilizers to reduce the arsenic mobil ity in a soil. It can be relatively costly. c. Vitrification: arsenic is chemically bonded inside a glass matrix forming, silicoarsenates. d. Soil flushing: uses water, chemicals, or organics to mobilize arsenic in a soil and then flush it from the soil.

PAGE 21

7 e. Phytoremediation/phytoextraction: uses plants to take up arsenic from soil. Ex situ remediation methods f. Excavation: physical removal and disposal of a contaminated soil in a designated landfill. Even though it produces rapid results, excavation is often expensive because of the opera tion, transport, a nd special landfill requirements. g. Soil washing/Acid extraction: based on suspension or dissolution of arsenic in a water-based solution to concentrate the contaminant. Phytoremediation Phytoremediation includes several methods that use plants to either remove contaminants or render them harmless in soil and water systems. It can be applied to both organic and inorganic contaminants in soil and water (Salt et al., 1998). This practice has been growing in popularity, b ecause of its overall cost-effectiveness (Watanabe, 1997; Salt et al., 1998; Kabata -Pendias and Pendias, 2001). The term phytoremediation includes the following strategies. Phytoextraction Phytoextraction is the use of contaminant-accumulating plants, which are able to extract and translocate contaminan ts to the harvestable parts. Phytoextraction is the most effective strategy of phytoremediation, although technically the most difficult one. It uses tolerant plants that concentrate soil contaminants in their aboveground biomass and the contaminant-enriched biomass can th en be properly disposed (Kramer, 2005).

PAGE 22

8 Phytostabilization Phytostabilization refers to the use of contaminant-tolerant plants for mechanical stabilization of contaminated soils to pr event soil erosion and to reduce air-borne transport and leaching of contaminants. It is used to provide a vegetation cover for a contaminated site, thus preventing wind and water erosion (Kramer, 2005). Plants that are suitable for phytostabili zation have an extensive root system, provide a good soil cover, are tolerant to the contaminants, a nd ideally immobilize the contaminants in the rhizosphere. Arsenic-tolerant plants that may be potential ly used for phytostabilization purposes have been known for a long tim e (Rocovich and West, 1975; Porter and Peterson, 1977; Benson et al., 1981). Phytoimobilization Phytoimmoblization is the use of plants to decrease the mobility and bioavailability of contaminants by altering soil factors that lower contaminant mobility by formation of precipitates and insoluble compounds and by sorption onto the roots. Based on the chemical similarities betw een arsenic and phosphorus, there may be formation of arsenic/lead compounds as s hown for phosphorus/lead precipitates in the rhizosphere of Agrostis capillaris (Cotter-Howells et al., 1999). Other plant-mediated processes of arsenic immobiliz ation at the soil-root interf ace involve pH reduction and oxidation of the root environment by O2 release from roots. Doyle and Otte (1997) found accumulation of arsenic onto the iron plaque in the oxidized rhizosphere of salt marsh plants, which may provide an effective imm obilization and detoxification mechanism for the plants.

PAGE 23

9 Phytovolatilization Phytovolatilization is the use of plants to volatilize contaminants and has been demonstrated for Hg and Se. In the case of Hg, this was achieved by genetic manipulation of plants (Rugh et al., 1996); whereas phytovolatilization of Se occurs naturally in plants (Terry and Zayed, 1994). De Souza et al. (1999) demonstrated that rhizosphere bacteria can enhance the Se volatilization and accumulation in plants. Volatilization of arsenic is also known to occur in natura l environments (Frankenberger and Arshad, 2002), but rhizosphere studies have not been repo rted for the formation of gaseous arsenicals enhancement at the soil -root interface. Available information on arsenic volatilization for soil s uggests that in the absence of plant roots, volatile compounds account only for small proportions of total arsenic (Turpeinen et al., 1999). Phytoextraction of Arsenic Arsenic is a nonessential element for plan ts, and inorganic arsenic species are generally phytotoxic. Under normal conditions , arsenic concentrati ons in terrestrial plants are usually less than 10 mg kg-1 (Matschullat, 2000). Di fferent plants contain different levels of arsenic. The following pl ants contain arsenic in an increasing order: cabbage (0.020 to0.050 mg kg-1) < carrots (0.040 to 0.080) < grass (0.020 to 0.160) < potatoes (0.020 to 0.200) < lettuce (0.020 to 0.250) < mosses and lichens (0.26) < ferns (1.3) (Matschullat, 2000). An average toxicity threshold of 40 mg kg-1 was established for crop plants (Sheppard, 1992). The chemical behavior of arsenic is largely similar to that of phosphorus in soils and plants. In all plant species tested so far, arsenate is taken up via the phosphate transport systems (Asher and Reay, 1979; Lee, 1992; Meharg and Macnair, 1992). As a phosphate analog, arsenate can replace phosphate in many biochemical

PAGE 24

10 processes, thus disrupt phos phate metabolism in plants. For example, arsenate can disrupt mitochondrial oxidat ive phosphorylation, thus the production of nucleotide adenosine triphosphate, the main energy sour ce for cells. This process is known as arsenolysis, or the hydrolytic pr ocess whose first step is th e replacement of arsenate for phosphate (Oremland and Stolz, 2003). Unlike arsenate, arsenite reacts with sulfhydryl groups of enzymes and tissue proteins, leading to inhibition of cellular function and death (Meharg and Hartley-Whitaker, 2002). The transfer of arsenic from soils to plants is low for most plant species. This is due to several factors includi ng: i) the restricted uptake by plant roots, ii) the limited translocation of arsenic from roots to shoot s, iii) arsenic phytotoxicity even at low concentrations in plant tissues , and iv) the low bioavailability of arsenic in soil (Wang et al., 2002). Among phytoremediation methods, phytoextr action is the most suitable for arsenic contaminated soils. It represents one of the largest economic opportunities for phytoremediation. This is due to the si ze and scope of environmental problems associated with arsenic contaminated soils , minimal environmental disturbance, public acceptance and the competitive advantage offered by the plant-based remediation technology (Raskin et al., 1997; Susarla, 2002). After harvesting arsenic-enriched plants, the weight and volume of contaminated material can be further reduced, transported, an d disposed off site as hazardous material (Salt et al., 1998; Ma et al., 2001). Successful application of phytoe xtraction to arsenic contaminated soils depends on many fact ors, among which the concentration and

PAGE 25

11 bioavailability of arsenic, the nutrient and wate r status of the soils a nd the capacity of the plant to access the arsenic (Figure 2-1). The arsenic bioavailability in soils is regulated by vari ous physical, chemical and biological processes and their interactions (E rnst, 1996). The physical characteristics of soils have a great influence on th e bioavailability of arsenic. For instance, soils with very fine texture offer a physical re sistance that may demand a lot of energy from the plant in order to penetrate deeper layers. In this si tuation, arsenic may be inaccessible for plant uptake. Figure 2-1. Factors that influence the efficiency of phytoextraction Arsenic speciation and chemical cond itions such as soil acidity and redox potential can largely determine arsenic bioava ilability (Wenzel et al., 2002). The soil biota can strongly modify the chemical and physical conditions and processes, which determine arsenic bioavailability. Rhizosphere interactions plays a ke y role in controlling nutrient bioavailability to crop plants (Hinsinger, 2001) as well as in the understanding of the processes occurring during phytoremediation (Lombi et al., 2001). The availability of

PAGE 26

12 arsenic in a soil around the roots is str ongly affected by root exudates and root depositions (mucilage and borde r cells) but also by microbial activities (Lombi et al., 2000). The processes occurring in the rhizosphere such as pl ant uptake, changes of pH and redox potential, root exudation and etc. a lter the chemical compositions of the soil/ root interface. It may also influence the bioavailability of arsenic in the soil and consequently the efficiency of phytoextrac tion. For instance, unde r oxic soil conditions, an increase in the rhizosphere pH could favor mobilization of labile and exchangeable As(V), enhancing plant uptake (Wenzel et al ., 2002). Also, nitrog en fertilization as nitrate (NO3 -) can potentially increase the rhizos phere pH, and thus possibly enhance arsenic accumulation in plant tissues. The symbiotic association of plant root s with mycorrhizal fungi in arsenic contaminated soils can influence its bioava ilability by exploiting a greater soil volume and by solubilizing the arsenic. Therefore, the interactions of arsenic with soil matrix and the ability of plants to conti nually accumulate and detoxify arsenic in their shoot system are essential to the phytoremediation concept. Phytoextraction can be accomplished by us ing either tolerant high biomass plant species or hyperaccumulator plant species. The growth and remediation potential has been assessed for various pl ants including cottonwood ( Populus deltoides Bartr.), cypress ( Taxodium distichum L.), eucalyptus ( Eucalyptus amplifolia Naudin, E. camaldulensis Dehnh., and E. grandis Hill), and leucaena ( Leucaena leucocephala L.), which are all potential high biomass species. However, the use of hyperaccumulator plants has the advantage of producing a more concentrated residue, reducing the final disposal of the contaminant-rich biomass.

PAGE 27

13 Unfortunately, most metal hyperaccumula tor plants grow slowly with low biomass, while plants that produce a high biomass are usua lly sensitive to high metal concentrations. The energy costs of metal to lerance mechanisms are partially responsible for this phenomenon (trade-off hypothesis). There are, however, exceptions (e.g. the nickel hyperaccumulator Berkheya coddii and arsenic hyperaccumulator ferns), indicating the capacity of a plan t to accumulate and tolerate high metal concentrations in shoots and to produce high amounts of dry ma tter is not always mutually exclusive (Robinson et al., 1997; Ma et al., 2001). Plant species used in the phytoextraction of arsenic should be ab le to thrive in a contaminated site and at the same time accu mulating a substantial amount of arsenic in the aboveground tissues. They should also be responsive to ag ricultural practices designed to enhance arsenic accumulation and to allow repeated pl anting and harvesting of arsenic-rich biomass (Tu and Ma, 2002) . However, continuous phytoextraction depends on the natural ability of a plant to accumulate, translocate and tolerate high concentrations of arsenic over the entire growth cycle (Garbisu and Alkorta, 2001) as well as the arsenic bioavailability in the soil. Hyperaccumulator Plants Plants show several response patterns when growing in the presence of toxic metals, such as tolerance, indicator and hype raccumulation (Figure 22). Most plants are sensitive even to low concentrations, othe rs have developed resistance and a reduced number behave as hyperaccumulators of a metal (Baker and Brooks, 1989; Schat et al., 1999). Tolerant species are those that can grow in a soil with metal concentrations toxic to most other plants. However, they are not necessarily indicators or hyperaccumulators,

PAGE 28

14 as tolerant non-accumulators can exclude metals from entering the root tissue. Both indicator species and hyperaccumulators are al so tolerant (Assuno et al. 2001; Bert et al. 2003). On the other hand, hyperaccumulator s take up particularly high amounts of a toxic element in their shoots during normal growth and reproduction cycle ( Baker and Whiting 2002) . Metal resistance in species with excl usion strategy is frequently based on reduced metal uptake into the roots, preferenti al storage of metals in the root vacuoles and restricted translocation into the shoots. Hyperaccumulators, in contrast, take up more metals, store a lower proportion of them in the root vacuoles, and export higher amounts to the shoots. Hyperaccumulators are plants that can ta ke up and concentrate greater than 0.1% of a given element in their tissue (Br ooks, 1998). Metal hyperaccumulation is a rare phenomenon in terrestrial higher plants. To date, over 400 plant species have been identified as metal hyperaccumulators, representing < 0.2% of all angiosperms (Brooks, 1998; Baker et al., 2000; Ma et al.’ 2001) . Approximately two-thirds of the known hyperaccumulators are Ni accumulators. This is because of the widespread occurrence of Ni-rich ultramafic (serpentine) soils and the long history of ge obotanical studies of ultramafic floras. Plant species that are able to hyperaccumulate Cd, Co, Cu, Pb, Zn, and As are much less numerous. Hyperaccumulator plants have evolved inte rnal mechanisms that allow them to take up and tolerate high concen trations of metals that woul d be toxic to other organisms (Lasat, 2002). They are adapted to the particular environmenta l conditions of their habitat and high metal accumulation may have contribut ed to their defense against herbivores and fungal infections (Martens and Boyd, 2002).

PAGE 29

15 The definition of a hyperaccumulator has to take into consideration not only the metal concentration in the aboveground biomass, but also the metal concentration in the soil. Both bioaccumulation factor (BF) and translocation factor (TF) have to be considered while evaluating whether a particul ar plant is a hyperaccumulator (Ma et al., 2001). The term BF, defined as the ratio of meta l concentration in plan t biomass to that in the soil, has been used to determine the eff ectiveness of plants in removing metals from soil. The term TF, defined as the ratio of me tal concentrations in the shoots to those in the roots of a plant, has been used to determin e the effectiveness of plants in translocating metals from the roots to the shoots (T u and Ma, 2002). Therefore, an arsenic hyperaccumulator plant should have BF > 1 and TF > 1 as well as total accumulation > 1,000 mg kg-1 arsenic in plant biomass. Figure 2-2. Conceptual response of hyperaccumu lator, indicator and excluder plants for metal concentrations in the aboveground tissues in relation to increasing metal concentrations in a soil. (Ada pted from Ghosh and Singh (2005). While some plants can survive in an e nvironment containing high concentrations of metals, they may not show an ability of hyperaccumulati ng the metals. For example, Agrostis tenuis concentrated 3, 470 mg kg-1 of arsenic when growing in a soil containing

PAGE 30

16 as high as 26, 500 mg kg -1 of arsenic. Even though the plant arsenic c oncentration was very high, it can not be characterized as an ar senic hyperaccumulator. This is because the BF and TF were both lower than one. Arsenic Hyperaccumulation by Pteris vittata Arsenic hyperaccumulators were discovered only recently and the majority of them are fern species in the Pteris genus. Ma et al. (2001) reported the first known arsenic hyperaccumulator plant, Pteris vittata L, commonly known as Chinese Brake fern. The plant was found on a site contamin ated with the wood preservative chromated copper arsenate in Central Florida. Afterwards, other arsenic hyperaccumulator ferns have been identified in Florid a and elsewhere (Francesconi et al., 2002; Zhao et al., 2002; Srivastava et al., 2005). Other non fern plants have also been reported as “arsenic hyperaccumulators” (>1,000 g g-1 As) growing on mine wastes from various sites in the United Kingdom (Porter and Peterson, 1977) and on smelter wa stes in northeast Portugal (De Koe, 1994). However, though accumulating large amounts of ar senic, these plants do not concentrate arsenic, i.e. arsenic concentra tions in the plant are lower than those in the soil. Hence, they should not be classified as arsenic hyperaccumulators. Although there are a number of ferns known to accumulate arsenic, P. vittata is by far the most studied arsenic hyperaccumulator plant. It is nativ e from China (Nelson, 2000) but is widespread in the old world, occurring from Europe to Asia. In the U.S., this fern grows in the southeast and southern California (Ma et al., 2001). In Florida, P. vittata is one of the only three natura lized exotic ferns (Nelson, 2000). Pteris vittata ferns are very diverse and have survived in gr eat numbers and adapted to both a vastly

PAGE 31

17 changing environment and competition from seed plants. They are widespread thriving in both temperate and tropical climates (Mat schullat, 2000). Also, the distribution of P. vittata is controlled by its requirement of a well-drained alkaline s ubstrate exposed to abundant sunshine (Ma et al., 2001). Pteris vittata has an extraordinary capacity not on ly to tolerate and take up arsenic but also translocate large amounts of arseni c to the fronds, with frond concentrations reaching levels up to 100-fold greater than th e soil concentrations. In addition, this plant, with its high biomass, is an easy-to-grow, vi gorous perennial that is resistant to disease and pests, and exhibits a high arsenic accumulation rate. While natural arsenic concentrations in most plants seldom exceed 1 mg kg-1 (Porter and Peterson, 1977), after 6 weeks of growth P. vittata accumulated 438-755 mg kg-1 As in a uncontaminated soil and 3,525 to 6,805 mg kg-1 As in a contaminated soil (Ma et al., 2001). It also removed 26% of the added arsenate after 18 weeks of growth (Tu and Ma, 2002). Liao et al. (2004) showed that, within seven months, P. vittata could extract up to 7.8 % of the arsenic from a soil containing 64 mg kg-1 As. The ability of P. vittata to take up high concentrations of arsenic and sequester it into aboveground portions when grown on eith er uncontaminated or arsenic-enriched soils implies that the fern has highly eff ective arsenic scavenging mechanisms. Another interesting characteristic of the ferns is their perenialit y, which allows for successive cuttings of the aboveground biomass while growing in contaminated soils. An intriguing observation was made by Tu et al. (2002). In a pot experiment, they observed that addition of 50 mg kg-1 arsenate to a sandy soil increased the fern biomass by 107%. After 12 weeks of growth, P. vittata produced more aboveground biomass in

PAGE 32

18 soils containing 50 and 100 mg kg-1 arsenic compared to the control. The increase in plant biomass may be due to the arsenic-enha nced phosphate concentrations in the soils. However, when the soil arsenic concentration reached 200 mg kg-1, there was a slight decrease in fern biomass. Another interesting fact is that arsenic ha s been shown to leach from P. vittata fronds as they senesce (Tu et al ., 2003). This may pose a potential drawback to the use of P. vittata in phytoremediation of arseni c contaminated soils, as the arsenic may be returned to the soil. To date, the only nonPteris fern to exhibit the ability of arsenic hyperaccumulation is Pityrogramma calomelanos (Francesconi et al., 2002). Its fronds accumulated 2,760 to 8,350 mg kg-1 arsenic when growing in soil containing 135 to 510 mg kg-1 arsenic. Interestingly, the fronds with the greatest ar senic concentration were collected from the ferns growing in the lowest arsenic concentration in the soil (135 mg kg-1 As) (Francesconi et al. ( 2002). In the case of P. vittata, studies have shown that the arsenic concentration in the fronds increase with increasing arsenic concentration in the soil. Arsenic Species in Pteris vittata The form of arsenic accumulated by plan ts is important in determining its suitability to remediate arsenic contaminated sites, because the arsenic-rich fronds will need to be disposed properly to avoi d further contamination (Ma et al., 2001). Pteris vittata has been shown to be capable of taki ng up both inorganic and organic arsenic species including arsenate, arse nite and monomethylarsonic acid (MMA), concentrating up to 93% of the arsenic in the fronds (Ma et al., 2001; Kertulis et al., 2005). Research on arsenic hyperaccumulation by P. vittata showed that arsenic exists in the plant mostly as inorganic species, and up to 94 % of the arse nic in the fronds is present as arsenite. Similar results were observed when arsenic wa s supplied to the ferns in several different

PAGE 33

19 forms (Zhang et al., 2002; Tu et al., 2003; Kert ulis et al., 2005). Regardless of the arsenic species supplied to the fern, >90% of the total arsenic in the roots is present as arsenate, versus approximately 94% arsenite in the fr onds. In both studies, low concentrations of organic arsenic were found in the fern. Similarly, in a study involving P. calomelanos , most of the arsenic in its fronds was pres ent as arsenite. Only trace amounts of monomethylarsonate (MMA) and dimethylarsi nate (DMA) were found in a few samples (Francesconi et al., 2002). The uptake rates of arsenite and arsenate by P. vittata roots may not be equal, because the species are taken up through differe nt mechanisms. For instance, Wang et al. (2002) found that arsenate was taken up more quickly by P. vittata than was arsenite, especially in the absence of phosphate. The auth ors suggest that this is due to arsenate being assimilated via phosphorus-suppre ssible uptake system in the roots. Arsenic Tolerance and Detoxification Mechanisms Plants often contain trace c oncentrations of contaminants of concern. At low levels, plants can usually metabolize or dispose of these compounds without significant injury. Generally, at high concentrati ons in soil or water, plants are not able to metabolize contaminants. However, some plants can survive and even grow well when they accumulate high concentrations of toxic elem ents as is the case of the hyperaccumulator plants. In ferns like P. vittata, arsenic is taken up at high rates and at concentrations proportional to the arsenic concentrations in the growth media at least up to a certain point where arsenic availability becomes a li miting factor and its arsenic concentration exceeds its detoxification ability (Ma et al., 2001; Zhang et al., 2002; Wang et al., 2002; Kertulis et al., 2005).

PAGE 34

20 The fact that P. vittata could survive in a soil spiked with 1,500 mg kg-1 arsenic and concentrate 2.3 % of arsenic in its biomass i ndicates that it is e quipped with efficient mechanisms for detoxifying accumulated arse nic. Mechanisms employed by plants to detoxify metals include chelation, compartm entalization, biotransformation and cellular repair (Salt et al., 1998). Fo r example, heavy metals are generally transported and deposited in a vacuole as me tal chelates. Baker et al. ( 2000) reported that the solution concentration of free metal ions taken up by plants into thei r tissues is reduced greatly when they are chelated by specific high-affi nity ligands (like oxyge n-donor, sulfur-donor, or nitrogen-donor ligands). Sulfur-donor ligands (like metallothioneins and phytochelatins) form highly stable complexe s with heavy metals, because sulfur is a better electron donor than oxygen. In fact, the reduction of arsenate to arsenite is catalyzed by glutathione (GSH) in microbes (Rosen, 2002). However, this has not been demonstrated in P. vittata . The formation of arsenite-tris-thiolate complexes has been demonstrated both in vivo and in vitro by electrospray ionization mass sp ectroscopy (Schmoger et al., 2000) and x-ray absorption spectroscopy (Pickering et al., 2000) with the th iolate derived from either GSH or phytochelatins (PCs). The majo rity of the reports on arsenic toxicity in plants show a clear role for PCs in the detoxification of arsenic (Schmoger et al., 2000; Hartley-Whitaker et al., 2001, 2002; Schat et al., 2002). Reina et al. (2005) demonstrated that PCs were the most abundant thiols in white lupin under higher arsenic exposure levels than other plants could tolerate. Toge ther, GSH and PCs were able to complex the majority of arsenic in the shoot s. However, the role of PCs seems to be minor in arsenic hyperaccumulator ferns (Zhao et al., 2003; Raab et al., 2004; Zhang et al., 2004).

PAGE 35

21 As for arsenic hyperaccumulators P. vittata and P. calomelanos the reduction of arsenate to arsenite inside pl ant cells occurs as well (Ma et al., 2001; Francesconi et al., 2002). This reduction of arsenate inside the pl ant cells is intriguing because arsenite is more toxic than arsenate. Additionally, P. vittata was shown to have only 4.5% of its arsenic complexed with phytochelatins, as a glutathione-arsenitephytochelatin complex (Zhao et al., 2003). In a study by Raab et al. (2004), the arsenic hyperaccumulator, P. cretica, had only 1% of its arsenic complexe d with phytochelatins. The conclusion reached in both studies was that the phytochela tins may act as shuttles for transporting arsenic in a non-toxic form through the cyt oplasm and into the vacuoles. However, arsenic complexation with phytochelatins by its elf does not account for the high efficient detoxification mechanism in arsenic hype raccumulator ferns, suggesting a novel mechanism of arsenic tolerance in P. vittata . Arsenic detoxification in microor ganisms includes methylation and biotransformation. Some bacteria enzymatically reduce arsenate to arsenite by Ars C, and the arsenite is then pumped out by the me mbrane protein Ars B (Cai and Ma, 2003). None of these mechanisms were identified in th e ferns. Therefore, an important gap in the arsenic hyperaccumulation mechan ism is how the ferns rapidly translocate arsenic from the roots to the fronds, and still able to su rvive the exceedingly high concentrations of arsenite in the fronds, which may perturb the cellular function by disrupting the sulfhydryl groups of proteins. Effects of Heavy Metals on Arsenic Hyperaccumulation Many contaminated sites contain a variety of contaminants. So another important point to be considered in phytoremediation is the ability of arsenic hyperaccumulator plants to grow in soils contaminated with other heavy meta ls. The ability of P. vittata to

PAGE 36

22 survive in soil contaminated with Cu and Cr is confirmed by the fact that this fern was first discovered in a soil contaminated with CCA (chromated-copperarsenate). The total arsenic in that soil was 131 mg kg-1, Cr was 40.6 mg kg-1, and Cu was 8.30 mg kg-1. Fayiga et al. (2004) studied the effects of Cd, Ni, Pb, and Zn on the arsenic accumulation by P. vittata . They observed the concentratio ns of heavy metals (50 and 200 mg kg-1) negatively affect the growth of the fern. Still total plant biomass increased by 12-fold after 8 weeks of growth. However, the efficiency of arsenic accumulation by the fern depended on the concentration of the metals in the soil (Fayiga et al., 2004; An et al., 2006). In the study of Fayiga et al. (2004), ar senic uptake decreased with an increase in the metal concentration except in Pb-treated soils . Though effective in taking up arsenic, P. vittata had a limited capability to take up other metals. Arsenic transfer factors ranging from 15.8 to 46.1, indicating that P. vittata was able to effectively translocate arsenic in the presence of other metals. Caille et al. (2004) reported that P. vittata grew poorly in a soil heavily contaminated with As, Cu, Pb and Zn, probably as a result of Zn and Cu toxicity. The concentrations of Zn and Cu in the plants were well above the toxicity thresholds of 100-500 mg kg-1 Zn and 20 mg kg-1 Cu, respectively (Kabata-Pe ndias and Pendias, 2001). Phytotoxicity of Zn and Cu not only decreased plant growt h, but also arsenic uptake by P. vittata , resulting in a negligib le phytoextraction of arsenic from the soil. Contamination of multiple metals or metalloids, particularly at high concentrations, thus presents a challenge for phytoremediation. Effect of Phosphate on Arsenic Hyperaccumulation The competitive effect of arsenate and phosphate in soils (Livesey and Huang, 1981; Manning and Goldberg, 1997; Smith et al., 2002; Tu and Ma, 2003) and their

PAGE 37

23 interactions in plants (Ash er and Reay, 1979; Meharg a nd Macnair, 1992; Fourquerean and Cai, 2001) have been demonstrated. Ba sically, the uptake of arsenate and phosphate by plants has been reported to be competitive (Tu and Ma, 2003). Moreover, arsenate interferes with phosphate me tabolism causing toxicity in plants. On the other hand, phosphate may be able to alleviate arsena te toxicity by improv ing phosphate nutrition (Sneller et al., 1999). Thus, the presence of phos phorus in contaminated soils plays a role in the phytoextraction process. Tu and Ma (2003) studied the infl uence of phosphate on the arsenic phytoextraction by P. vittata . They observed that, at low and medium arsenate levels (50 to 200 mg kg-1, respectively), phosphate had slight effect on arsenate accumulation by and growth of P. vittata. However, phosphate substantially increased plant biomass and arsenate accumulation by alleviating arsenate phytotoxicity at high ar senate levels (400 mg kg-1). The authors suggested a minimum P/As molar ratio of 1:2 in soil solution for effective removal of arsenic by the plant. Fayiga et al. (2006) reported that additi on of phosphate rock in a soil spiked with arsenic and other metals increased plan t arsenic uptake, from 608 to 1,046 mg kg-1 in the fronds. This shows the ability of phosphate rock to aid in arsenic accumulation by P. vittata in a multi-metal system by reducing the t oxic effects of the metals on the fern; thereby, enhancing plant arsenic uptake. Boisson et al. (1999) also reported an increase in arsenic uptake by plants ( Zea mays cv. Volga and Phaseolus vulgaris cv. Limburgse vroege) after applying hydroxyapa tite to a soil contaminated with Zn, Pb, Cu, Cd and As. They suggested that the increased phosphate concentration in the so il solution might be

PAGE 38

24 responsible for increased plan t uptake, since phosphate can displace adsorbed arsenate from soils and increase the availabili ty of arsenate (Smith and Naidu, 1998). In a chromated-copper-arsenate contamin ated sandy soil, addition of a large amount of rock phosphate (15 g kg-1) also increased arsenate uptake by P. vittata (Cao et al., 2003). Therefore, phosphorus additions, in the first place, enhanced plant growth, and secondly, mobilized exchangeable arsenic result ing in an increased to tal arsenic uptake. Despite the appeal of increased arsenic availability from application of soil amendments such as phosphorus fertilizer s for the purpose of phytoremediation, it is environmentally paradoxical in that the bene fit of enhanced arsenic removal from soils may be offset by the risk of increased arse nic leaching into ground water. Therefore, the rate of arsenic uptake should be higher than the rate of arseni c release in the soil. In this regard, a very extensive root system is a va luable characteristic in a hyperaccumulator plant. Mycorrhizae Association and its Role in Arsenic Hyperaccumulation Mycorrhizae have a well-documented role in increasing the plant uptake of phosphorus (Smith and Read, 1997) and other poorly mobile elements, and are recognized as an important component of bioremediation strategies for soils contaminated with heavy metals (Khan et al ., 2000). In fact, mycorrhizae are integral, functional components of plant root systems a nd the fungi involved can play an important role either in alleviating metal toxicity in plants or enhancing meta l uptake. For instance, Weissenhorn and Leyval (1995) repo rted higher concentrations of heavy metals in plants, even resulting in toxic levels, due to ar buscular mycorrhizae (AM) colonization, while Heggo et al. (1990) observed reduced metal concentrations in the shoots due to

PAGE 39

25 mycorrhizal colonization. On the other hand, Galli et al. (1995) found no effects exerted by AM fungi on metal concentrations in plant shoots and roots. Since arsenic and phosphorus are chemi cal analogues, it is expected that mycorrhizal symbiosis may be involved in arsenic uptak e as well as P uptake by arsenic hyperaccumulator plants. However, few studies have been done in this regard. Ying et al. (2004) investigated the coloni zation and diversity of AM fu ngi associated with common pteridophytes and found no defined relationshi p between mycorrhizae and plants under field condition. Out of all arsenic hyper accumulator ferns, only association of P. vittata with mycorrhizae has been investigated (Al agely et al., 2005; Leung et al., 2006). In a study on the effect of increasing levels of arsenic and phosphorus on P. vittata infected with mycorrhizae, Al Agely et al. (2005) observed that the AM fungi not only tolerated arsenic amendment, but its presence increased the frond biomass at the highest arsenic application rate (100 mg kg-1). The AM fungi also increased arsenic uptake across a range of phosphorus levels, while phosphorus uptake was generally increased only when there was no arsenic amendment, suggesting an important role of AM fungi in arsenic accumulation by P. vittata . The percentage of mycorrhizal infection in P. vittata increased as soil arsenic concentrations increased (Leung et al., 2006). The authors also reported an increase in arsenic accumulation by P. vittata infected with mycorrhizae and improvement of the nutrient status of the plants. The presence of mycorrhizae likely increased the amount of P transporters at hyphae leve l and, consequently, the amount of arsenic taken up by the plant. The presence of mycorrhizae also seems to be related to the formation of thiol like glutathione (Schutzendubel and Andrea, 2002), a phenolic defense system. It is

PAGE 40

26 hypothesized that the production of thiols by P. vittata, induced by the increased arsenic concentration, alleviated the toxicity of arseni c (Cai et al., 2004) and the plant can readily retain arsenic with the aid of mycorrhizae (Diaz et al., 1996). Therefore, it is possible that P. vittata may derive some benefit from the symb iosis by forming low molecular weight thiols.

PAGE 41

27 CHAPTER 3 ARSENIC ACCUMULATION BY Pteris vittata L FROM ARSENIC CONTAMINATED SOILS Introduction Arsenic contamination in soils, groundw ater, and drinking water is a serious concern as it can affect both human and anim al health. The extensive use of arsenic compounds such as pesticides, insecticides , defoliants, wood preservatives, and soil sterilants in the past (Azcue and Nriagu, 1994) ha s left a legacy in the history of arsenic contamination in the environment and a burden for the future generations. For instance, pesticides cont aining arsenic trioxide were largely used in cattle dip vats in the southeastern USA to eradicat e cattle fever tick resulting in many soils contaminated with arsenic (Thomas et al. , 2000). The production of the wood preservative containing chromated-copper-a rsenate (CCA) represented 67 % of the arsenic used in 1992. Besides th e contamination of soils in the sites of operation, CCAtreated wood can potentially contaminate so ils through the leaching of arsenic from the wood. Smelting and mining activities repr esent significant sources of arsenic contamination because pyrometallurgical produc tion processes lead to large emissions of Pb, Zn, Cu, Cd and As (Boisson et al. , 1999). Moreover, herbicides containing monosodium methane arsonate (MSMA) are stil l used for weed control in Florida (Cai et al. , 2002). The use of plants to remove arsenic from soils is a fairly new technology. It is an efficient and less costly way to treat contam inated soils. Hyperaccumulator plants possess

PAGE 42

28 highly efficient mechanisms to acquire and c oncentrate arsenic in the plants. However, not all contaminated sites are suitable to be treated by means of phytoextraction. This is because the arsenic concentra tion in a soil, its availabil ity and the depth of arsenic contamination can be limiting factors. Moreover , the plant to be used for this purpose must be tolerant to arsenic, adapted to the local soil and climate characteristics and take up a large amount of arsenic. During phytoextraction process, several cr ops of hyperaccumulator plants may be needed to sequentially reduce soil arsenic c oncentration (Raskin et al., 1994). However, continuous phytoextraction depends on the natu ral ability of a plant to accumulate, translocate and tolerate high concentrations of metals over the whole growth cycle (Garbisu and Alkorta, 2001). Pteris vittata is a fern species identified by Ma et al. (2001) as an efficient arsenic-hyperaccumulator plant. Because it hyp eraccumulates an extr emely high level of arsenic (up to 2.3%) in its aboveground biomass and it is easy to grow in a variety of soil environments, this plant has the potentia l to clean up arsenic contaminated sites nationwide, and potentially worldwide. Furthermore, the perennial nature of P. vittata make the phytoextraction process even more cost-effective since no replanting after harvest is needed. However, practical issues such as the time required to achieve a given target level, the long term effi ciency of the process, and th e arsenic pools depleted by the plant still need to be addressed. The aim of this study was thus to 1) assess the efficiency of repeated harvests of the fronds of P. vittata growing in soils with different sources of arsenic contamination, and 2) investigate the effects of plant arsenic uptake on arse nic redistribution in soils.

PAGE 43

29 Materials and Methods Soil Collection and Characterization Five arsenic contaminated soils plus a so il with naturally high arsenic (Marl soil) were used for this study. The soils were: c ontaminated with arseni cal insecticide (Avon soil), arsenical wood preservative (CCA soil) , arsenical pesticide (CDV soil), arsenical herbicide (EDS soil), and mining activities (Min ing soil). Selected pr operties of the soils are shown in Table 3-1. The soils were co llected, air-dried, a nd passed through 2 mm sieve. They were analyzed for: CEC by the ammonium acetate method (Thomas, 1982), organic matter content by the Walkley Bl ack method (Nelson and Sommers (1982), soil texture by the pipette method (Day, 1965) and soil pH in 1:2 soil: water. Concentrations of P, Ca, Mg, and trace elements we re determined using the EPA Method 3050. Experimental Design a nd Statistical Analyses Six soils with different sources of arse nic contamination were used to grow P. vittata . Each treatment was replicated four times and pots without plants were included as controls. The pots were arranged in a completely randomized design. SAS software was used for all statistical analyses (SAS Institute, 1987). Means and standard deviations of arsenic concentrations were calculated for different treatments. Analysis of variance was used to assess significant differences among treatments. Experimental Set up Pteris vittata used for the experiments were of similar age and size. One-year-old ferns were transferred (1 per pot) to 2gall on-size plastic pots filled with 4 kg of arsenic contaminated soil. The soils were thor oughly mixed with Osmocote, extended timerelease base fertilizer (N-P-K = 18-6-12) at a rate of 2 g kg-1 soil. The fertilization was

PAGE 44

30 applied annually. After transplanting, the ferns were watered to 60 % of the field capacity. Table 3-1. Selected chemical and physical ch aracteristics of six arsenic contaminated soils Soil characteristic Marl Avon CCA CDV EDS Mining pH 7.85 6.70 7.00 6.766.70 6.75 CECa 26.6 4.50 4.40 16.8 22.8 12.0 OM (g kg-1) 1.80 19.6 11.0 26.5 28.0 4.20 Total As (mg kg-1) 22.2 26.5 110 211 640 214 Extractable Asc (mg kg-1) 22.7 14.8 58.8 45.9 499 22.5 Extractable P (mg kg-1) 0.36 31.9 24.8 19.2 96.0 8.07 Extractable Ca (mg kg-1) 21000 1892 2960 2040 1602 4920 Extractable Mg (mg kg-1) 120 28.8 130 64.8 336 63.6 Fe oxides (mg kg-1)b 3.96 509 1322 5577 1117 1626 Al oxides (mg kg-1)b 5.23 632 884 1163 1056 179 Sand (g kg-1) 401 866 882 840 851 807 Silt (g kg-1) 410 89 91 140 117 152 Clay (g kg-1) 189 45 27 20 32 41 a:CEC: cation exchange capacity (Cmolc kg-1) b: Amorphous Fe and Al extracted by 0.2 M oxalic acid + ammonium oxalate solution c: Extractable elements were obtained using Mehlich III solution. The first harvest was performed in the m onth of October, 2003, four months after the transplant. The second (April, 2004) and third (October, 2004) harvests were performed in a six months interval from the fi rst harvest and from each other. The plants aboveground biomass was analyzed for dr y biomass weight and total arsenic concentration.

PAGE 45

31 Sampling, Digestion and Analysis The plant samples were dried in a 65oC oven for approximately 48 hours, weighed and then ground into powder through a 1-mm mesh screen using a Wiley Mill for digestion. Also, soil samples co llected together with the plan t samples were air-dried and analyzed. The soil samples were digested accordi ng to the EPA Method 3051a and the plant samples a modified EPA Method 3051 (Che n and Ma, 1998; Komar et al., 1999). One blank, one Standard Reference Material fr om National Institute of Standards and Technology (NIST), one duplicate and one spik ed sample were included for every 20 samples. Soil and plant samples were digest ed with nitric acid using a Hot Block digestion system (Environmental Expre ss, Mt. Pleasant, SC; EPA Method 3051). Approximately 0.5-1.0 g of air-dried soil or 0.1.5 g of dry plant sample were mixed with 1:1 HNO3 and allowed to set for approximately 24 hours. They were heated at 105oC for 2 hours and then cooled for 3 minutes. The samples were mixed with 1 ml of 30% H2O2 and placed on the block digester for 15 additional minutes. After the second heating, the samples were cooled completely and diluted to a 50 mL volume with distilled water. A filter cartridge was placed in the bottom of the digestion tube. The digested samples were analyzed for As concentration with a SIMMA 6000 graphite furnace atomic absorption spectrophotometer (GFAAS, Perkin-Elmer, Norwalk, CT ) using the EPA SW 846 method 7060 A. Arsenic Fractionation Since it is assumed that plant arsenic upt ake will substantially reduce soil arsenic availability, it is important to determine how plant arsenic uptake affects arsenic

PAGE 46

32 redistribution in a soil. A sequential extrac tion procedure was used to account for the changes in different soil arse nic-pools as a function of plan t uptake. According to Wenzel et al. (2001), soil samples were fractionated in to five arsenic fractions with decreasing availability: (N) non-specif ically bound, (S) specifica lly bound, (A) amorphous hydrous oxide-bound, (C) crystalline hydr ous oxide-bound and (R) residua l (Table 3-2). Samples from each fraction, with the exception of th e residual fraction, were centrifuged at 3,500 rpm for 15 min and 20C after each extracti on and/or wash. The supernatants were collected and filtered through Whatman 42 filter paper and analyzed for arsenic concentration using GFAAS. Table 3-2. Arsenic sequen tial extraction procedure Fractions Extracting solution Extraction condition SSR* NNon-specificallybound(NH4)2SO4 0.05M 4 h shaking, 20oC 1:25 SSpecifically-bound (NH4)H2PO4 0.05M 16 h shaking, 20oC 1:25 A-Amorphous hydrous oxide-bound NH4-oxalate buffer (0.2 M) pH 3.25 4 h shaking, 20oC 1:25 C-Crystalline-hydrousoxide-bound NH4-oxalate buffer (0.2 M) + ascorbic acid (0.1M) pH 3.25 30 min in a water basin at 96 3 oC in the light 1:25 R-Residual HNO3/H2O2 Hot block digestion 1:50 *SSR=Soil solution ratio Single Extractions Water-extractable arsenic was obtained us ing a 1:10 soil to deionized water ratio, shaking for two hours and centrifuging at 3500 rpm for 15 min (Olsen and Sommers, 1982). Mehlich III method (Mehlich, 1984) was commonly used as a standard procedure in soil testing to assess the amount of P av ailable in the soil for growing plants. As a phosphorus analogue, this method was chosen to correlate plant arseni c availability in

PAGE 47

33 soil as well. Mehlich III extractable arsenic was obtained using Mehlich III extracting solution (0.2 M CH3COOH, 0.25 M NH4NO3, 0.015 M NH4F, 0.013 M HNO3 and 0.001 M EDTA) after shaking for 5 min. Arsenic was al so extracted with 1.6 mM organic acids (phytic acid -1.4mM + oxalic acid 0.2 mM), in a 1:20 soil to solution ratio, shaking for 24 hours and centrifuging at 3500 rpm for 15 min. The concentrations of organic acids used in this study were similar to those repor ted by Tu et al. (2004) in the root exudates of P. vittata . All the extracts were filtered through Whatman 42 filter paper and analyzed for arsenic concentration using GFAAS. Results and Discussion Plant Growth The aboveground biomass production and th e re growth capacity are important factors in the phytoextraction of arsenic using perennial pl ants since multiple harvests will be needed in order to remove the arsenic from the soil (Fayiga and Ma, 2005). Despite the different soil properties and arsenic concen tration among the six soils, the ferns grew well in all six soils during the first four months of the experiment (JuneOctober 2003), without showing any toxicity symptom (Figure 3-1). The plants biomass was uniform in all soils ranging from 24.8 to 33.5 g plant-1, greater than the biomass data reported for P. vittata growing in pot experiments (Tu and Ma, 2002; Fayiga et al., 2004). It should be pointed out that the 12-month-old plants used in this study were small, with 5-6 fronds. There was a significant reduction in the frond biomass production in the subsequent harvests, except for the ferns grow ing in the Marl and Mining soils in the third harvest. For all soils, there was a signi ficant decrease in the frond biomass in the

PAGE 48

34 second harvest (40, 81, 82, 84, 79, and 71 % fo r the Marl, Avon, CCA, CDV, EDS and Mining soils, respectively) , likely due to the seasonal effect (cooler climate from October to April). For all soils, October harvests yielded larger plant biomass than the April harvest, except for the Marl and EDS soils. Figure 3-1. Fond biomass of Pteris vittata after first, second a nd third harvests in six soils. Means followed by the same letter within the same soil are not different by the Duncan test at p < 0.05. Due to the alkaline nature of the Marl soil, condition that favors the fern growth, dry matter yield was comparable in all th ree harvests. On the other hand, plants drastically failed to regrow in the EDS soil af ter the first harvest, probably because of the extremely high solubility of arse nic in this soil (Table 3-1). Even though P. vittata has shown to be very tolerant to arsenic exposition, Tu and Ma (2002) reported that addition of 500 mg kg-1 As to a sandy soil reduced the fern

PAGE 49

35 biomass by 64 %, a common symptom of arseni c toxicity (Kabata Pendias and Pendias, 2001). However, in their study, addition of 200 mg kg-1 As did not affect biomass production, suggesting a much higher tolerance of P. vittata as compared to normal plants. The difference in arse nic concentrations in the Avon, CCA and CDV soils (Table 3-1) did not influence the dry matter yield si nce all had the same trends among different harvests (Figure 3-1). However, the better pe rformance of the CCA soil was likely due to the higher pH and Ca concentration, and also due to the fact that this was the soil where P. vittata was found naturally gr owing (Ma et al., 2001). Overall, the regrowth capacity of the plan ts was related to the temperature effect. That is, the difference in plan t performance between harvests is due to colder temperature prevailing during December to Apr il. According to Jones (1998), P. vittata prefers warmer climate. In a field experiment, Ke rtulius (2005) also re ported greater biomass production when P. vittata plants were harvested in the months of October and December. Another factor that , associated with the temperature, could also explain the poor regrowth capacity of the plants after the first harvest was the c lipping procedure. All the fronds, including the coiled young fronds (f iddleheads), were removed in the first harvest. This did not happen in the second harvest. The effect of successive fern cultivati on on plant biomass and arsenic removal by P. vittata is relatively scarce and inconsistent . For instance, pot experiments using P. vittata have shown that the amount of biomass harvested decreased after two or three successive cuttings as well as the amount of arse nic extracted from a so il (McGrath et al., 2002). However, Kertulius (2005) f ound positive results when growing P. vittata under field conditions. Likely, the exploitation of a greater soil volume and the amount of

PAGE 50

36 residue produced at field condition may account for some of the differences between the studies in pot (McGrath et al., 2002 ) and field condition (Ketulius, 2005). Arsenic Concentrations in the Frond Biomass The total arsenic concentrati on in the soils used in this study varied from 22 mg kg-1 in the Marl soil to 640 mg kg-1 in the EDS soil (Table 3-1). The arsenic concentration in the EDS soil was greater than the arsenic level of 500 mg kg-1 used by Tu and Ma (2002), which was spiked as arsenate. Despite the high arsenic level, the plants grew well and took up arsenic from all soils. The frond arsenic concentrations ranged from 166 to 6,151 mg kg-1 in the first harvest, from 110 to 3,056 mg kg-1 in the second harvest, and from 162 to 2,139 mg kg-1 in the third harvest (Table 3-3). The highest arsenic con centrations in plants were observed in the EDS (from 2139 to 6151 mg kg-1), CDV (477 to 1872 mg kg-1) and Mining (423 to 1079 mg kg-1) soils (higher soil arsenic concentrations) and the lowest in the Marl (110 to 166 mg kg-1) soil (the lowest soil arsenic concentration). Even though the Marl and Avon soils ha d similar arsenic concentrations, the plants took up more arseni c from the Avon soil. There was no significant difference in the frond arsenic concentrations of P. vittata in the first and third harvest for all soils, except for the CDV and EDS soils. However, cooler climate conditions not only affected plant bioma ss production but also plant arsenic accumulation. Fern frond arsenic concentrations were 34, 46, 51, 75, 61 and 50% lower in the second harvest compared to those in the first harvest for the Marl, Avon, CCA, CDV, Mining and EDS so ils, respectively (Table 3-3).

PAGE 51

37 Table 3-3. The frond arsenic concentrations of Pteris vittata after the first, second and, third harvest in six soils. Soil As levels Frond arsenic (mg kg -1) mg kg -1 Harvest 1 Harvest2 Harvest 3 Marl 22 166 a 110 b 162 a Avon 27 336 a 181 b 280 a CCA 110 659 a 325 b 747 a CDV 211 1872 a 477 b 715 b Mining 215 1079 a 423.4 b 962 a EDS 638 6151 a 3056 b 2139 b Means followed by the same letter in a row a nd within the same so il are not different by the Duncan test at P < 0 : 05 within the same soil. The frond arsenic concentrations increased with soil arsenic con centrations, in all three harvests with a correlation coefficient of 0.97-0.98. However, the ability of the plants to remove arsenic from the soils reduced with time. The highest arsenic concentration (6,151 mg kg-1) was observed in the plants growing in the EDS soil, which also had the highest soil arse nic concentration, both tota l and extractable (638 and 498 mg Kg-1 As, respectively; Table 3-1). The high organic matter content of the EDS soil likely plays a role in the availability of arsenic. However, little in formation is known on arsenic reactions with organic matter. In fact, it is suggested that soil organic matter contributes very little to the arsenic sorption in soils due to their anioni c nature (Livesey and Huang, 1981; Wenzel et al., 2001a). Therefore, considering the sandy na ture and arsenic con centration of the EDS soil, phytoextraction might not be a good alternative for this soil, especially if there is any

PAGE 52

38 risk of arsenic exposure. Besides the long time required to cleaning up, th ere is the risk of potential arsenic leaching. Plant Arsenic Removal from Soils Pteris vittata had significantly higher arsenic ac cumulation (arsenic concentration X plant biomass) in the first harvest as comp ared with the second and third harvest from the Avon, CDV and EDS soils (Figure 3-2).Therefore, the first harvest accounted for most of the arsenic removed from these soil s. In the Marl, CCA and Mining soils, the accumulation of arsenic in the first harvest wa s similar to that of the third harvest. 0 40 80 120 160 200 Marl 22Avon 27CCA 110CDV 211Mining 214EDS 640Total Arsenic (mg kg-1)As uptake (mg pot-1) Harvest 1 (October 2003) Harvest 2 (April 2004) Harvest 3 (October 2004) Figure 3-2. Arsenic accumulation by P. vittata after first, second and third harvests in six soils. Values represent mean standard deviation (n=4). If the Mining soil is not considered, arsenic accumulation followed the trend arsenic concentrations in the soils. The tota l arsenic removed by the plants varied from 1.71 to 4.54 mg pot-1 in the Marl soil and from 17.0 to 175 mg pot-1 in the EDS soil. More espicificaly, the plants removed 10, 13.5, 42.7, 70.9, 53.3, and 210 mg arsenic per

PAGE 53

39 pot from the Marl, Avon, CCA, CDV, Mining and EDS soils, respec tively (Figure 3-2) after the three harvests. Ev en though arsenic hyperaccumulation did occur in the ferns growing in the EDS soil, the rate of arseni c removal was low rela tive to the amount of soluble and extractable arsenic in this soil. The percentage of arsenic removed from each soil by the fern uptake was used as a parameter of relative comparison of th e soils with a wide range of arsenic concentration. For instance, the plants re moved 8.2 % of arsenic from the EDS soil (higher arsenic content) after the three harvests, while the plants growing in the Marl and Avon soils (soils with the lowest arsenic content) removed approximately 12 % of the soil arsenic. The lower percent of arsenic removal from the EDS soil as compared to the Marl and Avon soils was likely due to the failure of the plant in regr owing after the first harvest. Similarly, the percent of arsenic removed from the CCA soil (9.5 %) was greater than from the Mining soil (5.8 %). In this case , despite the similar grow th of the plants in both soils, most of the arse nic in the Mining soil is unavailable to the plants. In general, the accumulation of arsenic by P. vittata translates to arsenic reduction of 6.4 to 13 % (Figure 3-3). In spite of the greater amounts of arsenic extracted from the first and third harvest as compared to the s econd harvest, it was still difficult to project the time required for soil arsenic reduction to meet the required limits because the plants failed to regrow after the third harvest. Soil Arsenic Distribution Despite all the criticisms to the fracti onation procedures, it is important in environmental research because it may provi de useful information on the reactivity, mobility and availability of the elements (Wenzel et al., 2001).

PAGE 54

40 0 2 4 6 8 10 12 14 16Marl 22 Avon 27CCA 110CDV 211Mining 215EDS 638SoilArsenic removed (%) Figure 3-3. Arsenic removed from the soil by Pteris vittata after three harvests. Bars represent mean standa rd deviation (n=4). Figure 3-4 shows arsenic di stribution in the five fr actions based on sequential extraction before and after plant growth. Among the five fractions, the N and S fractions are considered to be the most plant-available, whereas R the least plant-available. The sum of N and S fractions constituted 17.5, 22.0, 32.1, 21.7, 4.33 and 46.3 % of the arsenic in the Marl, Avon, CCA, CDV, Mini ng and EDS soils, respec tively (Figure 3-4). As expected, arsenic in these soils was prim arily associated with the A fraction, ranging from 40.0 to 58.7 %, except for the Marl so il which had a substa ntial amount of the arsenic (59.1%) associated with the C fraction and only 9.28 % associated with the N+S fraction. In the Mining soil, the A and C fractio ns constituted 84.3 % of the total arsenic. In the EDS soil, the majority of the arsenic was distributed among the first three fractions in the order A > S > N, further explaining the hi gh arsenic availability in this soil (499 out of 640 mg kg-1 total As was extractable As).

PAGE 55

41 Figure 3-4. Arsenic concentrations (mg kg-1) in different fractions in soils before and after plant growth. Values represent mean standard deviation (n=4). N= non-specifically bound, S= specifically bound, A= amorphous hydrous oxidebound, C= crystalline hydrous oxide -bound and R= residual fraction. Wenzel et al. (2001) fractionated arsenic in 20 arsenic contaminated soils and also showed that most of the arsenic was a ssociated with the amorphous fraction. The different soils with different sources of arsenic contam ination used in this study confirmed that, regardless of the source or fo rm of arsenic that enter the soil, arsenic reacts with soil components and becomes less available. It seems that the predominant reactions occur with hydrous Fe and Al oxides which mostly coat soil particles (Pierce and Moore, 1980; Smith et al., 1998).

PAGE 56

42 Plant growth influenced arsenic mobilizati on from all five arsenic fractions in the soils, except the residual fraction, which is th e most stable one. Among the fractions, the A fraction contributed the most for the arsenic reduction (45.3, 48.0, 72.0 and 59.0 %) for the Avon, CCA, CDV and mining soils, respective ly (Table 3-4). This was not observed for the Marl and EDS soils (Figure 3-4; Table 3-4). For instance, in the Marl soil, the highest arsenic mobilization o ccurred in the C fraction, which was also the fraction with the highest arsenic concentration. Even though the greatest arsenic mobilization in the EDS soil occurred in the S fraction (46.4 %), bot h the N and A fractions also contributed to a great extent. The treatments without plan ts showed that the ar senic mobilization was primarily a result of plant growth, although so me arsenic mobilization was also observed in the absence of plants. Th is was probably due to wetting and drying cycles during the experimental period. Although arsenic has a high affinity to soil, when the environmental condition changes, the mobility and even speciation of arsenic can change. For instance, the presence of plants leads to the exudation of some organi c anions into a soil, which directly impacts the mobilizat ion of arsenic in the soil. Total arsenic concentration is not a good pr edictor of its bioavailability because only arsenic dissolved in water can be transp orted to the roots and taken up by plants (Ritchie and Sposito, 1995). Plants tend to fi rst take up the most available fraction of arsenic from the soils and as this pool become s smaller, some of the arsenic from other fractions will be slowly transformed to water-soluble fraction to reestablish their equilibrium (McGrath et al., 2000).

PAGE 57

43 Table 3-4. Arsenic reduction (%) in each fraction in six soils as a result of plant arsenic uptake by P. vittata . Soil As removed (mg pot-1) N1 S A C R -------------------------% As --------------------------Marl 9.93 13.9 9.65 3.47 62.32 10.6 Avon 13.5 20.9 27.8 45.3 4.71 1.28 CCA 41.9 15.0 16.3 47.9 16.2 4.60 CDV 72.1 5.60 18.9 71.6 1.85 2.03 Mining 50.0 1.17 7.18 58.7 26.9 5.98 EDS 210 28.0 46.4 21.2 3.26 1.56 1 N= non-specifically bound, S= specifi cally bound, A= amorphous hydrous oxidebound, C= crystalline hydrous oxidebound and R= residual fraction. 2 the fraction with the highest arsenic re duction for a given soil was in bold font. As plants take up more arsenic, there is less available arsenic present in the soils, resulting in reduced available arsenic in soils and reduced plant arsenic uptake. Therefore, it is reasonable to assume that, on ly arsenic in N fracti on can be readily taken up by the plant and arsenic in other fractions has to convert to N fraction before being taken up by the plant. In other words, N fracti on acts as both a source of arsenic for the plant and a sink for arsenic in the soil, thus linking arsenic in the soil with the arsenic taken up by the plant. High proportion of the total arsenic in histor ically polluted soils is mostly present in the residual fraction (Voi gt et al., 1996; Kavanagh et al., 1997). Thus, most of the schemes commonly used tend to underestimate ar senic availability (G leyzes et al., 2001). However, the present results show that the pr esence of arsenic in the residual fraction was low, even in the soils where the presence of ar senic is of natural origin, such as Marl and Mining soils. This could indicate that the scheme we used seems to be more suitable for the study of arsenic fractiona tion in these polluted soils.

PAGE 58

44 Comparison of Extractable Arsenic Using Single Extraction Methods In this study, three single ex traction methods were used in order to relate arsenic availability in soils and plant uptake. The arsenic concentrations extracted with these methods are shown in Table 3-5. Generall y, the extractable arsenic followed the descending order of Mehlich III > organic aci ds > water-soluble in all soils. When calculated in percent of the total arsenic determined using the EPA 3050 method, the water soluble arsenic, Mehlich III and organi c acid extractable arsenic represented 1.62, 102 and, 28.2 % (Marl), 5.81, 54.8 and, 49.3 % (Avon), 5.32, 53.4, 12.1% (CCA), 0.06, 21.7 and, 18.9 % (CDV), 1.46, 10.5 and, 3.96 % (Mining) and 21.3, 78 and 38 % (EDS), respectively. The Mehlich III was expected to extract more arsenic than the water and the organic acids, irrespective the soil because of its chemi cal composition, a dilute acidfluoride-EDTA solution at pH 2.5. The result showed that the presence of low molecular weight organic acids in soils can be potential ly dangerous as it increase the mobility and availability of arsenic in the soil. For in stance, in the Avon and CDV soils, the organic acid extractable arsenic was comparable to the Mehlich III extractable arsenic. Therefore, the presence of typical plants that do not hyperaccumulate arsenic in arsenic contaminated soils has the potential to mob ilize arsenic from the so il since the exudation of organic acids is a common phenomenon in the rhizosphere of the plants. Plant uptake and Arsenic Availability in So ils as Measured by Single Extractants Arsenic phytoextraction from contaminat ed soils depends, among other factors, on arsenic bioavailability and the capacity of a plant to access arsenic. Arsenic bioavailability is regulated by physical, chem ical and biological processes and their interactions, which makes the simulation of elem ent availability to plants with chemical extractants a difficult task (Carter, 1993). On the other hand, the total concentration of an

PAGE 59

45 element is only important for evaluating the toxicity potential of soils, but they say little about elemental bioavailability. Table 3-5. Concentrations of arsenic extracted by different extraction methods in six soils before plant growth. Arsenic extracting solutions Soil Water Mehlich III Organic acids Total -------------------------------As (mg kg-1)----------------------------------Marl 0.36 0.15 (1.6)a 22.7 1.70 (102) 6.26 0.89 (28) 22.2 4.11 Avon 1.57 0.05 (5.9) 14.8 0.88 (55) 13.3 0.05 (49) 27.0 1.00 CCA 5.85 0.29 (5.3) 58.8 5.02 (54) 13.3 1.03 (12) 110 1.73 . CDV 0.12 0.03 (0.1) 45.9 20.5 (22) 39.9 1.00 (19) 211 2.65 Mining 3.13 0.15 (1.5) 22.5 1.00 (11) 8.49 0.97 (4.0) 214 16.5 EDS 136 4.04 (21) 498 57.5 (78) 242 30.1 (38) 640 5.52 Results are mean standard deviations (n=3). a: Results in the brackets are percentage of the total. In this section, the relationship between the soil arsenic extracted by three different methods and the arsenic ta ken up by the plants was evaluated. Water-soluble arsenic grossly underestim ated the plant arsenic uptake by 242% (based on the average of five soils, excl uding the CDV soil), whereas the Mehlich III over-estimated by 72%. Therefore, neither water nor Mehlich III solution was good indicators of the plant available arseni c. On the other hand, arsenic taken up by P. vittata constituted 74% of the arsenic extracted us ing organic acids, only 26% underestimation. Based on arsenic extracted with water, the arsenic taken up by P. vittata implies that a greater amount of ar senic was mobilized from othe r pools over time. This was particularly true for the CDV soil in which the amount of arsenic taken up by the plants was over 15,500 times greater than the amount of water-extractable ar senic (Table 3-6).

PAGE 60

46 Because low molecular weight organic acid s used in this study are products of root exudates, microbial secr etions, and plant and animal residue decomposition in soils (Stevenson, 1986; Zhang et al., 1997), and they are found in th e rhizosphere of P. vittata (Tu et al., 2004), we hypothesized that the or ganic acids extractable arsenic would be a good predictor of plant available arsenic in soils . Phytic acid and oxalic acid were the two main low molecular weight organic acids identified in the root exudates of P. vittata (Tu et al., 2004). Furthermore, the role of oxalic acid in mobilizing mineral elements has been widely reported (Jones, 1998). In fact, in this study, the organic acid extractable arse nic was the best method to predict plant arsenic uptake by P. vittata (Table 3-6). Arsenic bioavailability in soils has be en found to be enhanced by plant root exudates (Dinkelaker et al., 1989; Ohwaki and Hirata, 1992), suggesting that root exudates could be important in the mobilizat ion of soil arsenic and arsenic accumulation by a plant. Table 3-6. Percent of the arsenic taken up by P. vittata in each extracting method. Soil Water-Soluble Mehlich III Organic acids Marl 836* 11 40 Avon 216 23 95 CCA 183 18 81 CDV 16475 45 46 Mining 435 60 159 EDS 39 11 22 Average 342# 28 74 *Obtained by dividing the am ount of arsenic taken up by P. vittata by the amount of water-soluble arsenic. A valu e of 836% indicates that P. vittata took up 736% more arsenic than the amount of water soluble arsenic. # excluding the CDV soil.

PAGE 61

47 The percent of ar senic taken up by P. vittata according to the organic acids extraction method varied with soils. The best correlation was found for Avon (95 %), CCA (81 %) and Mining (159 %). In the Mining soil, P. vittata took up 59 % more arsenic than predicted by the organic acid method. However, the plants took up 60, 54 and 78 % less arsenic than predicted for th e Marl, CDV and EDS soils, respectively. Tu et al. (2004) reported that P. vittata exuded 40% more phytic acid and 300% times more oxalic acid under arsenic stress than a non arsenic hyperaccumulator, N. exaltata fern. It also mobilized si gnificantly greater amounts of arsenic from both arsenic mi nerals and a CCA soil (data not shown). Under the same concentrations, phytic acid mob ilized more arsenic from the soils than oxalic acid, likely due to its stronger complexation capability and greater acidity (Tu et al., 2004). Oxalic acid has been widely used as an extractant for plant available nutrients from soil, including phosphate (Fransson, 2001; Sagoe et al., 1998) due to its typical presence in plant root exudates (Pinton et al., 2001) and its capa bility of both proton donation and ion complexation. Arsenic extraction with water underestimated the availability of soil arsenic to P. vittata in all soils except the ED S soil. This was because 21% of the arsenic was water soluble in the EDS soil (Table 3-5) whereas the plant took up 39% wa ter-soluble arsenic, i.e., the amount of arsenic taken up by th e plant was less than the amount of watersoluble arsenic in the EDS soil. On the ot her hand, the concentrati ons of water-soluble arsenic were low in all other soil s. The amount of arsenic taken up by P. vittata after three harvests was consistent with the fact that more arsenic was solubilized over time from other pools (Figure 3-3).

PAGE 62

48 Because Mehlich III extract s a large proportion of total soil arsenic, it overestimated plant available arseni c. The amount of arsenic taken up by P. vittata was lower than the amount of arsenic extracted us ing Mehlich III extractant. Only Mehlich III extractable arsenic in the Mining soil was cl ose to the amount of arsenic taken up by the plant (60 %). Results from this study suppor t the conclusion of McLaughlin (2002) that indices of metal phytoavailability in soils th at are based on the determination of the most available metal pools (water-soluble and or ganic acids), while not perfect, appear to better predict risk from metal contamination than do total metal concentrations or metals removed from soils by strong extractants (Mehli ch III). Therefore, organic acids might be more suitable than water and Mehlich III to assess arsenic availability to P. vittata in the soils used in this study.

PAGE 63

49 CHAPTER 4 EFFECTS OF PLANT AGE ON THE AR SENIC ACCUMULATION BY THE HYPERACCUMULATION Pteris vittata L. Introduction Arsenic is a ubiquitous trace constituent in soil and plants. However, natural and anthropogenic activities have resulted in many areas polluted with ar senic. The extensive contamination of soil and water by arsenic wo rldwide has renewed the interest in more cost-effective remediation t echnologies such as phytor emediation/phytoextraction. The potential use of hyperaccumulator plants such as Pteris vittata L. for the phytoremediation of arsenic-contaminated so ils has been reported (Komar et al., 1998 ; Ma et al., 2001; Tu and Ma, 2002) . Several desirable characteristi cs, such as the ability to concentrate arsenic in the fronds, large biomass, fast growth, easy reproduction, resistance to adverse soil characteri stics and its perennial nature, make P. vittata suitable for phytoremediation (Ma et al., 2001). It is known that during the sigmoidal phase of a plant growth cycle, dry matter and nutrient accumulation occur at a maximal rate, although the rate of nutrient uptake varies with developmental conditions of a pl ant and is different for different nutrients (Garcia et al., 2003). There exis ts a set of plant physiological characteristics that change with plant age, which either isolated or combined with ot her factors influence biomass production, and nutrient and contaminant accu mulation. For instance, young roots are generally considered to have higher nutrient uptake activity than old roots (Eshel and Waisel, 1972; Mengel and Barber, 1974: Ve gh, 1991; Yanai, 1994) and the root systems

PAGE 64

50 at different growth stages may not have th e same specific nutrient uptake activity (Chen and Barber, 1990; Smethurst and Comerford, 1993). This is important because the accumulation ability of a plant is highly in fluenced by the effectiveness of its root system. Little information is available about arsenic hyperaccumulation in plants at different ages. Furthermore, no study has been reported for P. vittata of different ages growing in soil conditions. Therefore, identifying the plant age that allows maximal arsenic accumulation may play an important role in successful phytoextraction. The aim of this study was to assess the influence of the physiological ages of P. vittata on its arsenic removal from a contaminated soil. Materials and Methods Plant Propagation Pteris vittata plants used in this study were propagated in our laboratory. An assembly of cool-white fluorescent lamps s upplied an 8-h photoperiod to the plants with an average photon flux of 825 mol m-2 s-1. Soil Characterization The soil used in this study (sandy, sili ceous, hyperthermic grossarenic paleudult) was collected from an abandoned chromatedcopper-arsenate (CCA) wood preservation site in north central Florida. The soil had pH (1:2 soil:H2O) 7.30; cation exchange capacity (Thomas, 1982) 4.4 Cmol(-) kg-1; soil organic matter (Nelson and Sommers, 1982) 11 g kg-1; total As 153 mg kg-1; 880 g kg-1 sand; 90 g kg-1 silt, and 30 g kg-1 clay.

PAGE 65

51 Experimental Design This greenhouse study, set up as completely randomized design, assessed the influence of four physiological ages of P. vittata (1.5, 4, 10 and 16 months after transplant from the sporophytic phase) (F igure 4-1) on the arsenic removal of P. vittata from a contaminated soil. The plants of different ages were chosen based on their availability. Each treatment had four rep lications. Three contro l pots, soil without plants, were also included. Air-dried soil (2.5 kg) was weighed into each pot and thoroughly mixed with 3.0 g of Osmocote, extended time-release fertilizer (18-6-12) (Scotts-Sierra Horticultural Products Co., Marysville, OH). After one w eek of equilibrium under field capacity, one plant was transplanted into each pot. The pre-experimental frond and root biomass was taken before transplantation. The plants were allowed to grow for 8 weeks. All plants were watered throughout the study to keep the soil at approximately 70% of field capacity. Figure 4-1. Pteris vittata of 1.5, 4, 10, and 16 old months before transplant.

PAGE 66

52 After 8 weeks of growth, plants were harvested and separated into roots and fronds. Plant tissues were washed thoroughly with tap water, and then rinsed with deionized water. The fronds and ro ots were oven-dried for 3 d at 65o C, weighed and ground with a Wiley mill to pass through a 1mm mesh screen for chemical analysis. Soil samples from each pot were collected fr om the rhizosphere and bulk soil. The rhizosphere was defined as the soil attached to the roots. The rhizosphere soil was gently shaken from the roots and sieved through 2 mm me sh screen to separate it from the roots. Chemical Analysis Plants were digested using a modified version of the EPA Method 3050A for the Hot Block Digestion System (Environmental Expr ess, Mt. Pleasant, SC). Analysis was performed with a transversely heated, Z eeman background correction equipped graphite furnace atomic absorption spectrophotometer (Perkin Elmer SIMMA 6000, Norwalk, CT). The analysis of K, Ca, Mg, Zn, Cu, Fe and Mn was performed with a flame atomic absorption spectrophotometer (V arian model FS 220, Australia) with the same digestate. Quality control of arsenic analysis was assured by including the Standard Reference Material 1547 (Peach Leaves). Total P an alysis was carried out using a modified molybdenum blue method (Carvalho et al ., 1998). This involved reduction of the As in digestates from As (V) to As (III) with L-cy steine to minimize its interference with phosphate analysis. Tightly capped test tubes are incubated at 80oC for 5 min to allow complete reduction of arsenate into arse nite. P was determined, after cooling, by a double-beam spectrophotometer (Shimadzu UV I60U, Shimadzu Corp., Columbia, MD).

PAGE 67

53 Data Analysis All results were expressed as an average of four replicates. Treatment effects were determined by analysis of varian ce according to the General Linear Model procedure of the Statistical An alysis System (SAS Institute Inc., 1987). Duncan test at a 5% probability was used for post-hoc comp arisons in order to separate treatment differences. Results and Discussion Arsenic Uptake Arsenic uptake in P. vittata ranged from 3.98 to 6.18 mg plant-1 in the fronds and from 0.94 to 2.84 mg plant-1 in the roots (Figure 4-2). Th e As uptake in the fronds of 1.5month old P. vittata was 36% greater than those in the 4and 16-month old plants and similar to that of the 10-month old plants. On the other hand, the arsenic accumulation in the roots of the 1.5-month old ferns was lower th an the rest. Interesting to note that, even though the 10-month old plants accumulated as much arsenic in the fronds as those in the 1.5-month old ferns, they accumulated more arsenic in the roots as compared to the plants of other ages. Transfer factor (TF) (As in fronds/As in root s) was calculated to evaluate the relative effectiveness of plants in moving arsenic from the roots to the fronds. This characteristic of P. vitatta in accumulating high arsenic concentrat ions in its fronds makes this plant suitable for phytoextraction. . The TF valu es for 1.5, 4, 10 and 16 months’ old plants were 3.20, 2.05, 1.63 and, 1.60, respectively (data not shown). The TF reduced with plant age. The 1.5-month old plants were the more efficient in translocating arsenic from the roots to the fronds.

PAGE 68

54 Figure 4-2. Arsenic accumulation in the fronds and roots of P. vittata of different ages after 8 weeks of growth in an arseniccontaminated soil. Means with the same letter for the fronds or the roots are no t significantly different (P < 0.05) based on Duncan’s multiple range test. Table 4-1. Plant biomass, bioconcentrati on factor and transl ocation factor of P. vittata of different ages after 8 weeks of growth in an arsenic-contaminated soil. Biomass DW (g pot-1) Bioconcentration Factor Plant growth stage (months) Frond Root Frond Root Translocation Factor 1.5 12.7 b1 6.8 b 2.9 a 0.9 b 3.0 a 4 12.0 b 8.8 b 1.9 b 1.1 ab 1.8 b 10 20.7 a 16.0 a 1.9 b 1.3 a 1.6 b 16 13.1 b 8.9 b 2.0 b 0.5 c 3.8 a 1 Means (3 reps.) followed by the same lette r within the column are not significantly different (P < 0.05) based on Duncan’s multiple range test. Tu et al. (2004) also report ed that young fern plants were more efficient in removing arsenic from a hydroponic system than older fern s. Arsenic levels in the growth medium seem to determine the extent of arsenic accumulation by plants of different age. For a

PAGE 69

55 given plant, arsenic tends to be preferentially stored in young fronds at low arsenic level and in old fronds at high arsenic level in the media (Tu and Ma, 2002; Tu and Ma, 2005). The ability of young P. vittata to accumulate more arsenic may be related to several factors. The concentration of glutathione , a sulfur-containing tripeptide thiol and a precursor of phytochelatins, a very important antioxidant involved in plant detoxification (Scott et al., 1993), seems to decrease with the plant age (Hatton et al. 1996). In P. vittata , the glutathione concentration te nds to increase with the arsenic concentration in the plant (Cao et al., 2004). Elsewhere, it has also been reported that plant contaminant absorption decreases as plant is more mature, hence tolerance increase with age (Wilcut et al., 1989; Leah et al ., 1995). Knuteson et al. (2001) reported that four–week-old parrot feather a nd canna were more tolerant of simazine than two-weekold plants. Additionally , the root system at different ages has different nutrient uptake activities. The average uptake rate per unit of root decreases as a plant matures (Barber, 1984). The fact that P. vittata ferns are very diverse, easy to adapt to different environments and are widespread thriving in both temperat e and tropical climates (Matschullat, 2000) raises the concerns on using these ferns fo r phytoremediation of contaminated sites as they are classified as type II invasive species. One aspect that contri butes to this is their life cycle and breeding system. The large quantities of spores produced and released by the ferns are easily dispersed to long distances favoring colo nization of wide ecological niches (Bondada and Ma, 2002). Therefore, th e use of younger plants besides being more efficient in arsenic uptake c ould also minimize the risk of uncontrolled ferns propagation since younger plants could grow longer up to the phase of spore maturation in a

PAGE 70

56 contaminated site. The arsenic rich bioma ss could be harvested before the spores maturation phase as compared with older plan ts, which produce spores earlier after being transplanted in a field. Plant Biomass Arsenic hyperaccumulation efficiency de pends on both plant biomass production and arsenic concentration in the plant tissue (Baker et al., 1991; Salt et al., 1998; Ma et al., 2001). Therefore, the biomass of P. vittata was determined for all plant ages. After 8 weeks of growth, the frond and root biomass of the 10-month old plants was greater than the rest (Figures 4-3 and 4-4). Because the initial differences in the frond and root biomass of the treatments, the similarity of final root and frond biomass among plants of different ages indicated that plant biomass output occurred at different rate s. Therefore, the net biomass (final minus initial biomass) for each plant age and for each plant part (frond and root) was calculated. Even though the final biomass of the 10-month old plants was greater than the plants of other ages, the rate of increase in bioma ss was greater in the 1.5-month old plants, suggesting a higher metabolic activity. Phosphorus Distribution in the Plants As a P chemical analogue, arsenic is taken up by plants via the phosphate transport system (Meharg and Hartkey-Whita ker, 2002). Phosphate concentrations in P. vittata of different ages were determined to examine its relationship with arsenic concentrations. During vegetative growth phase, P contents in plants usually ranges from 97 to 161 mM of plant dry matter (Marschner, 2003).

PAGE 71

57 Figure 4-3. The frond biomass of P. vittata of different ages before and after 8 weeks of growth in an arsenic-contaminated so il. Means with the same letter are not significantly different ( P < 0.05) based on Duncan’s multiple range test. Figure 4-4. The root biomass of P. vittata of different ages before and after 8 weeks of growth in an arsenic-contaminated so il. Means with the same letter are not significantly different ( P < 0.05) based on Duncan’s multiple range test.

PAGE 72

58 In this study, P concentrations ranged fr om 140 to 183 mM in the fronds and from 97 to 124 mM in the roots, which were typical of most plants. Phosphate concentrations in the fronds and roots increased with plant age (Figures 4-5 and 4-6). As for arsenic, its concentrations in the roots increased w ith plant age whereas those in the fronds decreased. Therefore, accumulation of both ar senic and P in the roots was the highest in the 16-month old plants. On the other hand, the tissues of older P. vittata had lower arsenic concentrations in the fronds and higher arsenic concen trations in the roots, as compared with the younger plants. It is interesting to note th at the molar ratios of P/As stayed approximately one in the roots for all plant age (Figure 4-6) whereas those in the fronds increased with plant age (Figure 4-5). The effectiveness in arsenic hyperaccumulation by P. vittata is related to a series of detoxification mechanisms which are important to mitigate arsenic toxicity in the roots and fronds (Lombi et al., 2002). One of th em is the ability to maintain lower concentrations of arsenic in the roots a nd higher in the fronds, thereby minimizing the potential damage to the roots. The lower ar senic concentrations in the roots of the younger plants suggest a more eff ective arsenic removal process by P. vittata . Another important factor is the ability of P. vittata to manipulate phosphate in the plant biomass. Tu and Ma (2003c) reported that P. vittata maintained greater amount of phosphate in the roots than Nephrolepis exaltata , a non arsenic-hyperaccumulator fern. It is hypotesized that the ability of P. vittata in maintaining higher concentrations of phosphorus in the roots may constitute one of its mechanisms of arsenic tolerance (Tu and Ma, 2003c).

PAGE 73

59 0 40 80 120 160 200 1.541016Plant growth stage (Month)Frond phosphorus (mM)0 1 2 3 4 5 6 7Frond arsenic (mM) P As Figure 4-5. Effect of plant age on phosphorus and arsenic distributi on in the fronds of P. vittata after 8 weeks of growth in an arsenic-contaminated soil. Bars represent standard deviations of four replicates. Concentrations of Other Nutrients in Plants The K concentrations in the fronds and in the roots of P. vittata were within the normal range for most plants (Havlin et al., 2005). However, the K concentration in the fronds of the 1.5-month old plants was higher th an those in other ages (Table 4-2).The K distribution in the fronds and the roots in plants of different ages was similar to that of arsenic suggesting synergetic relationship of K with arsenic in the plants. Tu and Ma (2005) reported that arsenic and K had similar distribution in the fronds of P. vittata and speculated that K may function as a c ounterbalancing ion in these plants.

PAGE 74

60 0 20 40 60 80 100 120 140 160 1.541016Plant growth stage (Month)Root phosphorus (mM ) 0 0.5 1 1.5 2 2.5 3Root arsenic (mM ) P As Figure 4-6. Effect of plant age on the phosphorus and arsenic distribu tion in the roots of P. vittata after 8 weeks of growth in an arsenic-contaminated soil. Bars represent standard deviati ons of four replicates. The Ca and Mg distribution in the fronds and roots of P. vittata were similar to that of P, and increased with plant age (Table 4-1). Also, the plants that accumulated more arsenic (1.5and 4-month old plants) in the fronds had lower Ca. Tu and Ma (2005) reported that Ca concentrations in the fronds were inversely related to those of arsenic and were higher in older fronds growing in a soil spiked w ith up to 30 mg kg-1 As. Due to the very low mobility of calcium in the phloe m, the growth of the young parts of the plant is dependent upon the concurrent uptake of calcium. Among other roles of calcium in plants, th e ionic form of calcium is required in the vacuole of plant cells as a couter-cation for inorganic and organi c anions (Marschner, 2001). As arsenic forms compounds with calcium in soil and arsenic is hypotesized to be

PAGE 75

61 stored in the cell vacuoles, it would be reasonable to assume that the calcium concentration in the plant would increase with increasing arsenic concentrations in the plant tissues. However, based on the the results of this st udy and the studies of Tu and Ma (2005) and Fayiga and Ma (2005), that does not seem to be the case. In fact, the fact that calcium concentration in the older plants increased while arsenic concentrations decreased suggests a limited ro le of calcium in arsenic detoxification by the plants. Table 4-2. Macronutrients (K, Ca, and Mg) (g kg-1) and micronutrients (Fe, Zn, and Mn) content (mg kg-1) in the fronds and roots of P. vittata of different ages after 8 weeks of growth in an arsenic contaminated soil. Frond growth stage (Month) Root growth stage (Month) Elemental content 1.5 4 10 16 1.5 4 10 16 Potassium 18.8 a 14.2 b 15.6 b 15.3 b 8.70 ab10.9 a 10.0 a 5.50 b Calcium 13.5 b 15.5 b 20.4 a 20.7 a 21.5 a 20.0 a 23.4 a 23.8 a Magnesium 2.80 b 3.40 b 4.10 a 4.20 a 2.10b 1.70 b 2.90 b 16.6 a Iron 69.2 ab 75.0 a 58.0 bc 45.0 c 1235 b 1397 b1669 b 5627 a Zinc 43.6 a 37.0a 39.5 a 44.3 a 114 a 131 a 123 a 122 a Manganese 15.1 bc 28.6 a 20.3 b 13.9 c 41.3b 40.5b 39.5b 70.4a Means followed by the same letter in a row and within each plant pa rt are not different by the Duncan test at P < 0.05 The concentrations of Fe, Mn and Zn we re within the rang e of most plants (Havlin et al., 2005). While the Fe concentra tions in the fronds of the 1.5-and 4-months old plants were higher, the Fe concentrations in the roots of the 16-month old plants were higher than those in other plan t ages. Tu and Ma (2005) obs erved an inverse correlation between arsenic and Fe in P. vittata, but they offered no explanation. Chen et al. (2003) reported an unusual high concentrat ion of Fe in the epidermis of Pteris nervosa roots, indicating the formation of Fe plaques. Other studies with non hyperaccumulator plants

PAGE 76

62 also showed the formation of Fe plaques a nd the inhibition of arsenic uptake in roots (Otte et al., 1991; Colleen et al., 2002). .

PAGE 77

63 CHAPTER 5 ARSENIC ACCUMULATION AND ROOT CHARACTERISTICS OF Pteris vittata L. AND Nephrolepis exaltata L Introduction For successful application of phytoextra ction, the hyperaccumulator plants must be metal-tolerant, adapt to soil and climat e environment of their intended use, and effectively absorb the metal from soils (Kel ler et al., 2003; Schwartz et al., 1999). Since heavy metals are present in soils predominantly as insoluble forms, their bioavailability is generally low. Therefore, the efficacy of phytoextraction largely depends on the development of extensive plant root system s in a contaminated soil (Schwartz et al., 1999). Plant root systems perform many importan t functions includin g water and nutrient uptake, anchorage to the soil, and the esta blishment of biotic interactions in a rhizosphere. The characteristics of a root system are important in plant nutrient acquisition, particularly for elements of lo w mobility such as phosphorus and arsenic. Plant root systems can be characterized based on length, surface area, biomass and their relationship to plant shoots (Van Noordwijk and Brouwer, 1991; Jungk, 2002). The efficiency of roots to transfer metals from a soil to the shoots can be characterized by net influx per unit root length, and root length pe r unit of shoot biomass. Structural root properties, such as diameter of the roots and th e formation of root hairs, may also exert an influence on nutrient uptake from the immediat e vicinity of a root cylinder (Waisel and Eshel, 2002).

PAGE 78

64 According to Ernst (1996), most hyperaccu mulators have limited root systems. However, in a number of studies, unexplaine d growth enhancements were observed in plants subjected to a mild stress from toxi c metals. For example, the root growth of Betula pendula seedlings was induced by low concen trations of cadmi um (Gussarsson, 1994). Also, enhanced root elongation, root bi omass production, and root hair formation were found in plants of a tolerant population of Silene vulgaris treated with lead chloride (Wierzbicka and Panufnik, 1998). Jiang and Liu (1999) reported stimulated root growth of Brassica juncea plants with low concentr ations of lead nitrate. Schwartz et al (1999) and Whiting et al (2000) reported a prefer ential root development in the Zn hyperaccumulator Thlaspi cearulences in the presence of Zn. Little information is available on the root system of P. vittata . The process by which P. vittata mobilizes and takes up arsenic is not well known. Hyperaccumulator species may release root exuda tes containing chelators to enhance heavy metal uptake, translocation and resistance (Wenzel et al., 2003 ). The ability of a plant to exude large quantities of dissolved organic carbon (DOC ) and to change the rhizosphere pH may enhance the arsenic bioavailabil ity in soils, thereby increasing its arsenic uptake (Tu et al., 2004). The fate and bioavailability of arse nic in the rhizosphere can be different from that of the bulk soil (Fitz et al., 2003). De pending on plant and soil factors, rhizosphere pH can be up to two units different fro m the bulk soil (Marschner, 2003). Factors affecting the rhizosphere, such as pH, plant nutritional status, organi c acids excretion, and CO2 production by roots and rhizosphere micr oorganisms, are accountable for the differences (Marschner, 2003).

PAGE 79

65 Several attempts have been made to measure the bioavailability of metals in soils. Tessier at al. (1979) used a sequential extraction technique for determining the labile metal in soils. This approach is based on th e fact that metal associated with various geochemical phases varies in its chemical r eactivity and bioavailability. The drawback of the method is its non-selectiv ity and metal redistribution among geochemical phases (Howard and Brink, 1999). Despite its limitati on, fractionation provides an understanding of the relative mobility and bioavailability of metals in soils (Fitz and Wenzel, 2002; Krishnamurti and Naidu, 2002), because plant meta l uptake or metal toxi city is related to those fractions (Chlopecka and Adriano, 1996; Guo and Yost, 1998). Water-extractable and exchangeable forms of metals are usuall y considered to be the most available to plants (Petruzzelli, 1989). This study was conducted aiming to evaluate and compare two fern species, an As-hyperaccumulator ( P. vittata ) and a non As-hyperaccumulator ( Nephrolepis exaltata ), in terms of: 1) root system characteristics (b iomass, length, surface area and diameter); 2) plant biomass and arsenic uptake; and 3) root uptake efficiency of arsenic; and 4) effects of plant arsenic uptake on arse nic distribution and bioavailab ility in the rhizosphere and bulk soils. The information obtained from th is study should be useful for determining the plant density and effective root depth of P. vittata for phytoextraction purpose as well as for providing the critical linkage between the ability of P. vittata in solubilizing soil arsenic and arsenic hyperaccumulation.

PAGE 80

66 Materials and Methods Soil Characterization The soils used in this study were collected (0-20 cm) from an abandoned chromated-copper-arsenate (CCA) wood preser vation site (As-contaminated soil) and from a non-contaminated site (control soil) in Florida. The site was contaminated with arsenic from using CCA wood preservativ e between 1951-1962 (Komar, 1999). The soil analyses were performed as described in th e Chapter 4 and summarized in Table 5-1. Table 51. Selected properties of the soils used in this study Property As-contaminated soil Control soil pH (1:2 soil/water ratio) 7.30 0.01* 6.81 0.02 Organic matter content (g kg-1) 11.0 0.30 13.2 0.42 CEC** (cmol(+) kg-1) 4.40 0.02 5.60 0.04 Total As (mg kg-1) 101 5.40 0.41 0.01 Water-soluble As (mg kg-1) 0.2 0.02 0.003 0.001 Mechlic III extractable P (mg kg-1) 21.8 0.30 87.2 0.77 Sand (g kg-1) 882 4.32 892 4.76 Silt (g kg-1) 91 0.40 75 0.30 Clay (g kg-1) 27 0.01 33 0.02 Bulk density (g cm-3) 1.29 1.39 * Values represent mean standard deviation; ** Cation exchange capacity Experimental Set up The study employed a completely randomized design in a 2 x 2 split plot scheme. Two fern species ( P. vittata and N. exaltata ) were used as the main plot and the chemistry of the rhizosphere and bulk soil as the sub-plot . Each treatment was replicated four times.

PAGE 81

67 A known weight (2.5 kg) of air dried a nd sieved (2 mm) soil was poured into plastic bags and thoroughly mixed with 3.0 g of Osmocote extended time-release base fertilizer (18-6-12) (ScottsSierra Horticultural Products Co., Marysville, OH), and then poured into plastic pot s (2.5-L rhizopot). The rhizopot (16 cm in height and 15 cm in diameter Figure 5-1) was used to grow the plants. Plastic frames (13 cm in height and 7 cm in diameter) covered with nylon mesh cloth (mesh size 45 m) were used to separate the rhiz osphere from the bulk soil in the rhizopot. Figure 5-1. Cross section of th e rhizopot used in the study. Pteris vittata was propagated in our laboratory whereas N. exaltata was procured from a nearby nursery (Milestone Agriculture, Inc., Apopka, FL). Efforts were made to ensure visual uniformity across all plants. The average biomass per plant before the transplant was 0.70 g for the fronds and 0.45 g for the roots.

PAGE 82

68 Root growth was limited within the nylon cloth in the central compartment (500 g soil). One week before the study, the soil was set to equilibrate at water holding capacity. One healthy fern of similar age with five to six fronds was transplanted into each pot. The plants grew for 8 weeks in a greenhouse with an average night/day temperature of 14/30oC and an average photosynthetica lly active radiation flux of 825 mol m-2 s-1. The plants were watered throughout the study to keep the soil at approximately 70% of its field capacity. At the end of the experi ment, the ferns were harvested and each plant was separated into roots and fronds. Rhizosphere and bulk soil were collected and used for analysis of ar senic, soil pH and dissolved organic carbon (DOC). Rhizosphere soil was defined as th e soil attached to the roots, which was removed from the roots by shaking gently. The rhizosphere soil was sieved to remove the roots while keeping the roots in tact as much as possible. Th e bulk soil was defined as the soil outside the central compartment. The plants were washed thoroughly with tap water, and then rinsed with distilled water. The fronds were ove n-dried for 3 days at 65oC, weighed and ground using a Willey mill to 60-mesh fineness for chemical analysis. The roots were immediately separated and scanned to prev ent dehydration. After scanning, the roots were also ovendried, weighed and ground for chemical analysis. Root Measurement The roots were washed with water, spread out on a transparent tray and then the image was digitalized using a scanner (E nvisions88005) at 400 dpi resolutions with Vistascan software (UMAX Data system, In c). When necessary, root samples were divided into sub-samples for more accurate measurement. The digitalized images were

PAGE 83

69 processed by the program GSROOTS (G uddanti and Chambers, 1993), which was programmed to yield root length data of pr e-defined diameter cl asses (<0.10; 0.10 < d < 0.25; 0.25 2.0 mm). Root length density (Lv, root length per unit vol ume of soil) and root area density (AV, root surface area per unit volume of so il) were calculated. Unit root length and unit root area were used in all discussions unless otherw ise specified. The root biomass was also determined. Chemical Analysis Plants and soils digestion and analyses were performed as described in the Chapter 4. Rhizosphere and bulk soil samples were evaluated for water-soluble arsenic and DOC in 1:4 soil to water ratio (Olsen and So mmers, 1982), obtained after shaking (1 h), centrifuging (15 min at 3500 g) and filtering (0.45 m syringe filter). Arsenic in solution was determined by the same method descri bed above. The concentration of DOC was measured using a TOC-5050A TOC analyzer (Shimadzu, Japan). Quality control of arsenic analysis was included using Standa rd Reference Materials 1547 (Peach Leaves) and 2710 (soil) (US NIST, MD). Arsenic Fractionation The improved sequential extraction proce dure (Wenzel et al. 2001) was used to fractionate arsenic into five operationally-d efined fractions, incl uding non-specifically bound (N), specifically bound (S), amorphous hydrous oxide-bound (A), crystalline hydrous-oxide-bound (C), and residual (R), as described in the Chapter 3.

PAGE 84

70 Data Analysis All results were expressed as an average of four replications. Treatment effects were determined by analysis of variance accord ing to General Linear Model procedure of the Statistical Analysis System (SAS Institute Inc., 1987, Cary, NC). The Duncan test at 5% of probability was used for post-hoc comp arisons to separate treatment differences. Pearson correlation coefficients were calcu lated between the root characteristics and different plant parameters. Results and Discussion Root Biomass, Root Length, R oot Area, and Root Diameter Plant roots are in direct contact with, and are affected by soil constituents in addition to regulating many plant proce sses. Measurement of below ground plant productivity may provide information on root growth, distribution, water and nutrient supply and plant potential of removing contaminant (Van Noordwijk and Brouwer, 1991). Root-soil contact is an especially impor tant factor for the uptake of less mobile elements such as P and arsenic from a soil. However, the available information on the root characteristics of P. vittata has been limited to the measurement of root biomass (Tu at al., 2002; Tu and Ma, 2003; Liao et al., 2004) . Root characteristics including root bi omass, length and su rface areas were determined for both plants. They are impor tant for estimating plant water and nutrient uptake (Zhuang et al. 2001). In both soils, P. vittata had a more extensive root system than N. exaltata (Table 5-2). The root biomass of P. vittata was 3.0 and 2.4 times greater than that of N. exaltata in the As-contaminated soil and in the control soil , respectively.

PAGE 85

71 The root biomass accounted for approximate ly 25% of plant biomass for both plant species (Tables 5-2 and 5-3). Several studies have reported that the presence of contaminants stimulated the growth of plant roots (Gussarsson, 1994; Wi erzbicka and Panufnik, 1998; Jiang and Liu, 1999). Other studies indicate a seve re inhibition of root growt h. For instance, Schwartz et al. (1999) reported that Thlaspi caerulences growing in a soil spiked with Zn had reduced root growth while it explored a large soil volume with enhanced Zn acquisition in an unpolluted soil. However, the root growth of T. caerulescens , a Zn hyperaccumulator, was higher than that of Thlaspi arvense , a non-hyperaccumulator, when grown in a Znamended soil (Schwartz et al., 1999; Whiting et al., 2000). Table 5-2. The root biomass (dw), root length, and root area of P. vittata and N. exaltata after 8 weeks of growth in a As-contamin ated soil (As-soil) and control soil. Fern species Root biomass (g dw pot-1) Root length (cm cm-3) Root area (cm cm-2) As-contaminated soil P. vittata 1.8a 4.3a 1.3 a N. exaltata 0.6b 1.1b 0.4 b Control soil P. vittata 2.2a 6.2 a 1.8 a N. exaltata 0.9b 2.1 b 0.7 b Means followed by the same letter in a column within the same soil are not significantly different at p < 0.05 (n=4). The root length and root area of P. vittata were 3.9 and 3.2 times greater than that of N. exaltata in the As-contaminated soil, and were 2.9 and 2.6 times greater in the control soil, respectively (Table 5-2). In our study, the root length and surface area were lower in the As-contaminated so il than in the cont rol soil for both fern species. Although

PAGE 86

72 the overall chemical and physical characteristics of the two soils (Table 5-1) were similar, the higher P content and low As content of th e control soil must have accounted for the better development of the plant roots in that soil. Root diameter determines the extent of the soil volume that potentially supplies nutrients and contaminants to plants. For both soils, P. vittata had a greater root length for all classes of diameter than N. exaltata (Figure 5-2). The roots of 0.25 to 1.00 mm of diameter comprised a major proportion of P. vittata roots as measured by root length. For both soils and plant species, the roots < 0.50 mm diameter comprised over 40% of the total root length, while thos e of 0.50 to 1.00 mm diameter contributed another 25%. The fine roots (0.25 to 1.00 mm) of P. vittata contributed more to the total root length than did N. exaltata . This is important because fine roots al low the root system to exploit a soil volume more effectively while coarser roots have greater transpor t capacity (Forde and Lorenzo, 2001). For all classes of root diameter, P. vittata had a greater root area than N. exaltata . The roots with diameter >1 mm represente d 67 to 71 % of the total root area for N. exaltata , whereas they only accounted for 59 to 60 % for P. vittata (Figure 5-3). Based on the root characteristics (root biomass, length and area) of the two fern species, P. vittata presented a much greater potential to re move As and nutrients from soils than N. exaltata regardless of the soil characteristics. Frond Biomass, Arsenic Uptake, Bi oconcentration and Translocation Good biomass production and high arseni c accumulation ability are two of many desirable treats of hyperaccumulator pl ants for a successful application of phytoextraction. Although the two ferns species had similar frond biomass

PAGE 87

73 (approximately 0.7 g dw plant-1) before the experiment, P. vittata had a greater biomass (5.5 -6.6 g) than N. exaltata (1.8 – 2.8 g) after 8-week of growth (Table 5-3). The frond biomass of P. vittata was 3.1 and 2.4 times greater than that of N. exaltata in the Ascontaminated soil and in the control soil, respectively. This observation confirms the fast er growing characteristic of P. vittata when compared with other fern species, which is desirable for a plant to be used for phytoremediation purpose. Other studies have shown that P. vittata can grow normally or even have the growth stimulated in th e presence of arseni c (Tu and Ma, 2002). The amount of arsenic taken up by P . vittata (2.51 mg per plant) was 29 times greater than that of N. exaltata (0.09 mg per plant) in the arsenic-contaminated soil (Table 5-3). Tu and Ma (2004) found a 5.8 fold increase in arsenic accumulation of P. vittata over N. exaltata in a hydroponic study. Huang et al. (2004) reported that P. vittata and Pteris cretica , another arsenic hyperaccumulator, removed more arsenic from water than N. exaltata . While 92% of the plant arsenic was accumulated in the fronds of P. vittata , only 37% was in the fronds of N. exaltata (Table 5-3). The ratio between the contaminant concentra tions in the tissue relative to that in the soil (bioconcentration factor, BF) as well as the partitioning of the contaminant between the fronds and roots (tra nsfer factor, TF) were determined aiming to gain insight of the arsenic accumulating capability of the plants.

PAGE 88

74 Figure 5-2. The root length density (LV) of P. vittata and N. exaltata as a function of root diameter in a control soil (a) and an As-contaminated soil (b) after 8 weeks of growth. Bars represent standard deviations of four replicates.

PAGE 89

75 Figure 5-3. The root surface area density (AV) of P. vittata and N. exaltata as a function of root diameter in a co ntrol soil (a) and an As-contaminated soil (b) after 8 weeks of growth. Bars represent standa rd deviations of four replicates

PAGE 90

76 Table 5-3. The frond biomass (dw), arseni c accumulation, and bioconcentration and translocation factors of P. vittata and N. exaltata after 8 weeks of growth in a arsenic-contaminated soil and control soil. Fern species Frond biomass (g plant-1) As (mg plant-1) Bioconcentration factor (BF) Transfer factor (TF) Frond Root Frond Root As-contaminated soil P. vittata 5.5 a 2.30 a 0.21 a 1.7 a 0.5 a 3.6 a N. exaltata 1.8 b 0.03 b 0.06 b 0.1 b 0.4 a 0.2 b Control soil P. vittata 6.6 a 0.003 a 0.00 0.55 a 2.5 a 0.2 a N. exaltata 2.8 b 0.000 b 0.00 0.03 b 0.3 b 0.1 b Means followed by the same letter in a column within the same soil are not significantly different at p < 0.05 (n=4). We assumed that plant root characteris tics play a more important role in determining BF than TF. The fronds (1.7) and roots (0.5) BF of P. vittata was 17 and 1.3 time greater than that of N. exaltata in the arsenic-contaminated soil (Table 5-3).The fact that P. vittata had a more extensive and more effici ent root system may have contributed to its greater BF as compared with N. exaltata . The transfer factor (TF=3.6) of P. vittata was 18 folds greater than that of N. exaltata (Table 5-3). Huang et al. (2004) also reported that N . exaltata was unable to translocate the abso rbed arsenic from the roots to the fronds, which was expected for a non-hyper accumulator plant specie. The ability of P. vittata to translocate arsenic from the roots to the fronds is considered to be one of the tolerance and detoxification mechanisms involved in its arse nic hyperaccumulation. Root Uptake Efficiency of Arsenic and Phosphorus Root uptake efficiency (RUE), which is defined as elemental accumulation in plant tissues (fronds or roots) per unit root biomass, length or area, can be used to

PAGE 91

77 evaluate how efficient a plant is in taking up an element (T able 5-4). The arsenic RUE for the fronds of P. vittata measured by root biomass, root length and root area was 15-23 times greater than that of N. exaltata in both soils. This means that not only did P. vittata have a greater root biomass but also a more efficient root uptake system for arsenic. Because N. exaltata tended to accumulate arsenic in the roots, the differences in the arsenic RUE for the roots between the two plan ts were smaller, 8-10 times in the control soil and similar in the arsenic contaminated soil (Table 5-4). Because arsenic is an analogue to phosphorus, we also determined the P RUE for the fronds and roots of both plants. The P RU E in the control soil and for roots in the arsenic contaminated soil was similar be tween the two plant species (Table 5-4). However, the P RUE for the fronds was greater in N. exaltata than P. vittata in the arsenic-contaminated soil. These RUE results are rather interesting because they indicate that despite the similarity between P and arsenic, both P. vittata and N. exaltata root systems tended to remove P from the soil preferentially and more efficiently than arsenic. Recently, it has been found that some natu ral hyperaccumulators proliferate their roots positively in patches of high metal availability.I n contrast, non-accumulators actively avoid these areas, and this is one of the mechanisms by which hyperaccumulators absorb more metals when grown in the same soil (Schwartz et al., 1999; Whiting et al., 2000). In the greenhouse study it was observe d that most of the roots of P. vittata tended to concentrate in the top 7 cm of the pot (data not shown). The observation gathered in this study points to the need to evaluate the behavior of P. vittata root system under field

PAGE 92

78 conditions in an arsenic-contaminated soil as well as the effective root depth of the plant. The answer to this question w ill help to determine the effective depth that the plant will be able to clean up arsenic -contaminated soil. Table 5-4. The root uptake efficiency (RUE) as measured by As and P uptake (mg plant-1) by the fronds and roots of P. vittata and N. exaltata per unit of root parameters (root dry weight-RB, lengthLv, and surface area-Av) in a As-contaminated soil and control soil. As-contaminated soil Control soil Root characteristics P. vittata N. exaltata P. vittata N. exaltata Frond As-RB* 1277 a 61 b 1.7 a 0.1 b As-Lv** 552 a 31 b 0.6 a 0.04 b As-Av*** 1913 a 85 b 2.1 a 0.1 b P-RB* 5400 b 10500 a 11600 a 9000 a P-Lv** 2300 b 5600 a 3800 a 3600 a P-Av*** 8200 b 14500 a 12500 a 10800 a Root As-RB 116 a 92 b 2.6 a 0.3 b As-Lv 50 a 47 a 0.9 a 0.1 b As-Av 176 a 124 a 3.2 a 0.3 b P-RB 2400 a 1700 a 2900 a 2500 a P-Lv 1100 a 900 a 1000 a 1100 a P-Av 3700 a 2400 a 3200 a 3200 a * : ug plant-1 (As or P)/g root; ** : ug plant-1 (As or P)/cm cm-3 root; *** : ug plant-1 (As or P)/cm2 cm-3 root.Means followed by the same letter in a row within the same soil are not different at P < 0.05 (n=4). Plant Nutrient Uptake Balanced supply of essentia l nutrients is one of the most important factors in increasing crop yields (Fageria, 2001). The frond P (1.8 .5 g kg-1), K (1.28 2.51g kg1), Ca (1.26-1.58 g kg-1), and Mg (2.8-4.8 g kg-1) concentrations of both fern species

PAGE 93

79 (Table 5-5) were within the normal con centration range for most plant species (Marschner, 2003). They were also simila r to those reported by Tu and Ma (2005). For the As-contaminated soil, the P and Mg concentrations in the fronds of N. exaltata were greater than that of P. vittata (Table 5-5). Positive interaction between P and Mg are expected since Mg is an activat or of kinase enzymes and activates most reactions involving phosphatase transfer (Faglia, 2001). In mo st plants, P is generally concentrated in the upper parts or reproducti ve organs (Marschner, 2003). However, in this study, the frond P concentration of P. vittata (1.8 g kg-1) was lower than that in the roots (2.4 g kg-1) in the arsenic-contaminated soil. This result agrees with Tu and Ma (2004) who found similar trend for P c oncentration in fronds and roots of P. vittata and N. exaltata growing in a hydroponic solution. Arsenic hyperaccumulation by P. vittata may be facilitated by its high ar senic influx rate and its high molar P/As ratio in the roots resulting from both high arsenic TF and low P TF (Tu and Ma, 2004). In the control soil, N. exaltata had a greater K and Mg c oncentration in the fronds (1.7-1.9 fold) as well as Mg in the roots than P. vittata . Plant potassium content influences the plant biomass and arsenic content (Komar, 1999). In the present study neither plant biomass nor arsenic uptake of N .exaltata was related with th e status of K in the plant. It is unclear why N. exaltata took up more K in both soils, especially in the fronds. Arsenic Reduction and Distribution in the Soil Whereas they caused no significant change in total arsenic concentration in the bulk soil (Table 5-6), N. exaltata and P. vittata reduced arsenic concentrations in the rhizosphere soil by 28.6 and 40.7% (Table 5-7).

PAGE 94

80 Table 5-5. Nutrients concentrations (g kg-1) in the fronds and roots of P. vittata and N. exaltata after 8 weeks of growth in an As-c ontaminated soil and a control soil. fronds roots Fern species P K Ca Mg P K Ca Mg As-contaminated soil P. vittata 1.8b 13.3a 15.8a 3.2b 2.4a 8.9a 17.3a 3.0a N. exaltata 3.4a 25.1a 14.9a 4.8a 1.7a 7.8a 17.2a 2.8b Control soil P. vittata 3.5a 12.8b 15.0a 2.8b 2.8a 7.6b 16.0a 2.5a N. exaltata 3.0a 21.7a 12.6a 4.4a 2.5a 10.3a 13.3a 2.6a Means followed by the same letter in a column within the same soil are not significantly different at p < 0.05 (n=4). To achieve the arsenic reduction from 105 to 75.0 mg kg-1 for N. exaltata and 105 to 62.3 mg kg-1 for P. vittata in the rhizosphere soil, the so il mass required to contribute the 0.09 and 2.51 mg arsenic (Table 5-3) constituted only 3 and 61g out of 2.5 kg, respectively. Similar to total arsenic, no change was observed for different fractions of arsenic in the bulk soil with or without a plant (Table 5-6). Most of the arsenic was present in the A fraction (61.5%) and C fraction (22.2%), c ontributing a total of 83.7% (Table 5-6). The N and R fractions accounted for only 4.63 and 2.93%. Unlike the bulk soil, the decrease in ar senic concentrations was observed across all five arsenic fractions in th e rhizosphere soil (Table 5-7). Nephrolepis exaltata and P. vittata reduced arsenic concentrations in the A fraction (As-A) from 66.7 to 42.0 and 66.7 to 30.9 mg kg-1, respectively, accounting for 76 a nd 67% arsenic decrease in the rhizosphere.

PAGE 95

81 Table 5-6. Arsenic distribution in different fractions in the bulk soil after 8 weeks of plant growth. No plant P. vittata N. exaltata As fractions -------------mg kg-1 --------------As distribution*** ----%--N** 5.00 a* 5.30 a 5.40 a 4.63 S 9.80 a 9.50 a 8.60 a 9.07 A 66.4 a 63.8 a 63.1 a 61.5 C 24.0 a 24.4 a 22.4 a 22.2 R 3.20 a 3.20 a 3.70 a 2.96 Sum 108 106 103 100 Total As 105 99 97 Recovery (%) 103 107 106 * Means with the same letter in each row are not significantly different according to Tukey – HSD test at 5 % level. ** Fractions N, S, A, C and R stand for non-specifically bound, sp ecifically bound, amorphous Al and Fe hydrous-oxide bound, crystalline hydrous-oxide bound, and residual arsenic fraction; *** Based on the treatment with no plant.

PAGE 96

82Table 5-7. Arsenic distribution in differe nt fractions in the rhizosphere of P. vittata (PV) and N. exaltata (NE) after 8 weeks of plant growth. Soil As concentrations (mg kg-1) % As in each fraction** As reductions*** (mg kg-1) % of total As reductions**** As fractions No plant PV NE No plant PV NE PV NE PV NE N***** 4.30 a 2.0 b4.10 a*14.63 4.18 6.03 2.300.20 4.38 0.62 S 10.7 a 3.9 c6.50 b 9.07 8.16 9.56 6.804.18 13.0 12.9 A 66.7 a 30.9 c42.0 b 61.5 64.6 61.8 35.2 24.6 67.0 76.2 C 16.0 a 9.7 b13.7 ab 22.2 20.3 20.1 6.402.30 12.2 7.12 R 2.80 a 1.3 b1.70 b 2.96 2.09 2.50 1.501.10 2.863.41 Sum 101 a 47.8 c68.0 b 100 100 100 52.5 32.3 100 100 Total As 105 a 62.3 c75.0 b 40.6 28.5 Recovery (%) 95.7 76.7 90.7 * Means with the same letter in each row are not significantly different according to Tukey – HSD test at 5 % level. ** Based on total arsenic of 101, 47.8 and 68.0 mg kg-1. *** Based on no-plant treatment for each fraction. **** Based on total reduction of 52.5 and 32.3 mg kg-1. ***** Fractions N, S, A, C and R stand for non-specifically b ound, specifically bound, amorphous Al and Fe hydrous-oxide bound, crystalline hydrous-oxide bound, a nd residual arsenic fraction;

PAGE 97

83 Compared to N. exaltata , P. vittata removed more arsenic from each fraction. For example, reductions in arsenic concentrati on in the A, S and C fractions were 24.6, 4.18 and 2.30 mg kg-1 for N. exaltata and 35.2, 6.80, and 6.40 mg kg-1 for P. vittata (Table 57). However, plant uptake had little effect on the relative arseni c distribution among the five fractions in the rhizosphere of two pl ant species, with 61.8-64.6% in the A fraction, 20.1-20.3% in the C fraction and 8.16-9.54% in th e S fraction. This was because arsenic in all fractions was reduced, with an average reduction of 52.7% and 27.0% for P. vittata and N. exaltata , respectively (Table 5-7) The results for arsenic dist ribution in different fracti ons obtained in our study were consistent with the da ta reported by Wenzel et al. (2001) where they determined arsenic fractionations in 20 Austria soils with differing levels of arsenic contamination (96,183 mg kg 1). Arsenic in those soils was also mostly present in the A fraction (42.3%) followed by C fraction (29.2%), acc ounting for a total of 71.5% or 12.2% less than that obtained in our study. However, th ey reported concentration of 0.24 and 17.5% in the N (most available) and R (least available) fractions. Comparatively, the soil used in our study had 4.39% more arsenic in the most available fraction and 14.5% less arsenic in the least available fraction, indicating that arsenic in the so il used in our study had greater availability than in the soils from their study. The decrease in arsenic concentration in the five arsenic fractions of the rhizosphere soil did not follow arsenic availa bility according to th e sequential extraction, rather it was positively correlated to the arse nic concentrations in each fraction (r = 0.995 for both plants, n=4). This means that the fraction having the highest arsenic concentration, i.e. the A fraction, had the grea test reduction in its co ncentration. The fact

PAGE 98

84 that arsenic concentration in the N frac tion, the most available, was lower with P. vittata (2.0 mg kg-1) than N. exaltata (4.1 mg kg-1) was consistent with the greater ability of P. vittata to take up arsenic. The results also impl y that arsenic transfer from less-available fractions to the N fraction was slow er than the arsenic depletion by P. vittata . Fitz et al. (2003) reported a much lower arsenic transformation in P. vittata rhizosphere in a soil with a higher buffer capacity. The high rate of arsenic mobilization (reduction in the A fraction) observed in the rhizosphere of P. vittata can also be attributed to the sandy nature of the soil (88% sand) (Table 5-1), i. e. low ability to retain arsenic, which was supported by low oxalate extractab le Fe and Al (267 and 260 mg kg-1) (Table 5-1). The mass balance of the total arsenic in the soil, as determined by the sum of the fractions, was evaluated (Table 5-6 and 5-7). Good recoveri es of total so il arsenic (103– 107% in the bulk soil and 7796% in the rhizosphere) indicated that the difference between the two methods ranged from 3 to 24% . Despite all the cri ticism attributed to sequential extraction, the procedure was ab le to provide some insight into the mobilization and availability of arsenic in contaminated soils. Additionally, the sum of the fractions was in good agreement with the total arsenic values. Influence of Plants on Water-Soluble Arsenic, Soil pH and DOC The water soluble arsenic in the rhizosphere of P. vittata (0.7 mg kg-1) was two times lower than that in the N. exaltata (1.4 mg kg-1), which was similar to the no plant treatment (1.7 mg kg-1). Compared to the bulk soil, reduc tion in water-soluble arsenic in the rhizosphere soil by P. vittata and N. exaltata were 58.9 and 17.6% (Fig 5-4). The high water-soluble arsenic in the N. exaltata rhizosphere was consistent with the results found for arsenic in the N fracti on, the most available arsenic.

PAGE 99

85 Concentration of water-soluble metal is a good indicator of bioavailability for plant uptake in soils (McBride, 1994) since pl ants preferentially take up nutrients from soil solution (Linehan et al., 1985). The depletion or the enrichment of an element in the rhizosphere is determined by the capacity of a soil to replenish the soluble or exchangeable forms of the element (Hinsinger, 1998). Concentration of water-soluble metal is a good indicator of bioavailability for plant uptake in soils (McBride, 1994) since pl ants preferentially take up nutrients from soil solution (Linehan et al., 1985). The depletion or the enrichment of an element in the rhizosphere is determined by the capacity of a soil to replenish the soluble or exchangeable forms of the element (Hinsinger, 1998). The supply of elements such as P and As to the plant rhizosphere is limited by diffusion. When plant outpaces the soil suppl y capacity, the depletion occurs in the rhizosphere, as it was the case of P. vittata (Table 5-7). When the ability of the plant removes ar senic from the soil is lower than the arsenic diffusion rate, arsenic accumulates in the rhizosphere, as it was the case of N. exaltata (Table 5-7). Thus, lower water-solu ble arsenic in the rhizosphere of P. vittata was due to the higher ability of the plant to deplete the arsenic from the rhizosphere as compared with N. exaltata . These results are consistent with the more extensive root system of P. vittata , as well as with the arsenic reduction in its rhizosphere. The changes in rhizosphere pH and DOC may have a great effect on soil arsenic bioavailability. The influence of plant gr owth on soil pH was reflected by the pH difference in the bulk and rhizosphere soil. Plan ts had no effect on the bulk soil pH for all treatments (Figure 5-5).

PAGE 100

86 0 0.5 1 1.5 2 2.5 P. vittataN. exaltataControlTreatmentsWater-soluble As (mg kg-1 ) Bulk soil Rhizosphere Figure 5-4. Comparison of wate r-soluble arsenic in the bulk and rhizosphere soil of P. vittata and N. exaltata after 8 weeks of growth in an arsenic-contaminated soil. Bars represent standard de viations of four replicates. However, the rhizosphere pH of P. vittata (7.66) and N. exaltata (7.18) were 0.4 units higher and -0.13 lower, respectively, than that in the no plant treatment. Youssef and Chino (1989) and Luo et al. (2000) also reported an increase of pH in the rhizosphere of Triticum aestivum and Thlaspi cearulences . Yet Fitz et al. (2003) found no change in the rhizosphere pH of P. vittata . Since the soil used in their experiment had high carbonate content and hi gh buffer capacity, their results were not unexpected. The increase in soil pH observed in the rhizosphere of P. vittata in our study was more likely due to the low buffer capacity of the sandy soil as well as the cation/anion balance caused by high excreti on of hydroxyl groups in response to high absorption of arsenic. The pH buffering capacity of the soil and the in itial soil pH are the main factors determining the extent to wh ich plant roots can change rhizosphere pH

PAGE 101

87 (Marschner, 2003). Thus, the high rhizosphere pH measured in the P. vittata rhizosphere not only made HAsO4 2the prevailing form in the system but also increased the negative surfaces charges of soil minerals (iron and aluminum oxides). Similar to pH, plant growth had no effect on DOC concentration in the bulk soil. However, the rhizosphere of P. vittata and N. exaltata had about 44 and 33% higher DOC concentration than in the bulk soil (Fig. 5-6) . The DOC content in th e rhizosphere soil of P. vittata was approximately 9% higher that that of N. exaltata. Some components of DOC are active substances exhibiting strong a ffinities with trace el ements (Phillippe, 1999). Thus, increased DOC content may facilitat e the change of otherwise stable arsenic fractions . The small difference in DOC concentrati on in the rhizosphere between the two ferns suggests that DOC-induced mobilization of arsenic in the rhiz osphere may not play a significant role in the arsenic hyperaccumulation by P. vittata . Fitz et al. (2003) reported an increase of 86% in DOC in the rhizosphere of P. vittata growing in a soil containing 2,270 mg kg-1 As with higher pH and higher cl ay content. Wenzel et al. (2003) also found a significantly higher DOC concentrat ion in the rhizosphere than in the bulk soil of the Ni hyperaccumulator Thlaspi goesingense after growing in soil containing 2,580 mg kg-1 Ni. Tu et al. (2004) repo rted DOC concentration of P. vittata twice as high as N. exaltata in a hydroponic nutrient solution . However, data obtained from hydroponic studies do not account for plant-soil interactions and th e potential role of microorganisms in the rhizosphere (Wenzel et al., 2003).

PAGE 102

88 6.6 6.8 7 7.2 7.4 7.6 7.8 P. vittataN. exaltataControlTreatmentsSoil pH Bulk soil Rhizospheret Figure 5-5. Comparison of the soil pH in the bulk and rhizosphere soil of P. vittata and N. exaltata after 8 weeks of growth in an arsenic-contaminated soil. Bars represent standard deviati ons of four replicates. It has been proposed that hyperaccumulator species may enhance metal solubility in the rhizosphere via root e xudation as it is known for othe r plants (Knight et al., 1997). However, the separation of plant root e xudation from those deri ved from microbial metabolites in the rhizosphere is a difficult task since the root exudates are hard to collect. Moreover, the rates of rhizo-deposition even for a given plant species vary greatly and can be 2-4 times hi gher for soil-grown plants than for plants grown in nutrient solution (Trofymow et al., 1987). This is be cause rhizosphere mi croorganisms increase rhizodeposition (Meharg and Killham, 1991).

PAGE 103

89 0 20 40 60 80 100 120 140 160 180 P. vittataN. exaltataControlTreatmentDOC (mg kg-1 ) Bulk soil Rhizosphere Figure 5-6. Comparison of dissolved organi c carbon (DOC) in the bulk and rhizosphere soil of P. vittata and N. exaltata after 8 weeks of grow th in an arseniccontaminated soil. Bars represent sta ndard deviations of four replicates. The different results from the above mentioned studies are likely due to differences in soil characteristics, plant species, growing medium (soil x nutrient solution), metal concentration in the soil, different microbial population and activity, experimental period and climate conditions, plant age, and stress factors (Uren, 2000). Also, the methods used to extract DOC affect the amount extracted.

PAGE 104

90 CHAPTER 6 ARSENIC ACCUMULATION BY TWO HYPERACCUMULATOR Pteris SPECIES FROM TWO ARSENI C CONTAMINATED SOILS Introduction Arsenic (arsenic) pollution has beco me a major public concern in many countries. Phytoextraction is an in situ , low cost and environmentally sustainable emerging technique (McGrath, 1998). Th is technology relies on the use of hyperaccumulator plants that can produce high biomass and achieve sufficiently high metal concentrations in the shoots (Raskin and Ensley, 2000). Many studies have been conducted to identify plant species capable of accumulating undesirable toxic elements. For instance, Reeves and Baker (2000) compiled an exhaustive list of plant species that hyperaccumulate metals. Ma et al., (2001) reported that Pteris vittata L, an arsenic-hyperaccumulator fern, accumulated concentrations as high as 23 000 mg kg-1 As in the fronds. Later research has identified several othe r arsenic-hyperaccumula tor fern species belonging to the order Pteridales, including Pteris cretica , Pteris longifolia and Pteris umbrosa (Zhao et al., 2002). A screening experiment carried out in a hydroponic system revealed that another fern species Pteris biaurita was capable of accumulating as much arsenic as P. vittata when growing in solution containing 50 mg L-1 of arsenic (Srivastava et al., 2006). The ability of this new fern specie in accumulating arsenic from contaminated soils with different arsenic levels is still

PAGE 105

91 unknown as well as its growth and developm ent. This could be important when choosing a given plant specie to phyt oremediate contaminated soils. The geographical distribution, grow th and physiology of pteridophytes species such as P. biaurita and P. vittata are highly influenced by abiotic factors such as sunlight, available water, substrate conditions, soil fertility and etc (Smith et al., 1997). Some ferns are found only in sunny and disturbed habitats, others are in shady regions (Greer et al., 1997). Shade l eaves are thinner with more leaf surface area, have higher chlorophyll contents and less dry mass per unit leaf area than sunny leaves (Brach et al., 1993). Hill (1972 ) reported that fern species from open, sunny habitats have higher maximum photos ynthesis rates than the shade species. Due to the diversity of ecosystems that contamination occurs, effort has to be made not only to identify new plants with the ability to accumulate the contaminant but also to compare the perf ormance of hyperaccumulator plants in different environments. In the pres ent study we selected two arsenic hyperaccumulators, P. vittata , a plant adapted to wide ra nge of ecosystem, specially in sunny, alkaline and well drained environment, and P. biaurita ., a plant adapted to low temperature, high moisture and sh ady conditions (Jones, 1997). The objective of this study was to compare the two plants in terms of: i) plant arsenic accumulation and distribution, ii) plant biom ass; iii) chemical changes in the rhizosphere; and iv) plan t elemental composition. Materials and Methods Soil Collection and Characterization Two soil samples were collected from the CCA site as described in the Chapter 5. Surface soil (20 cm) was sample d, air-dried and sieved (2 mm mesh).

PAGE 106

92 The soil analyses were performed as desc ribed in the Chapter 4. The soil samples had different arsenic le vels (153 and 266 mg kg-1), and thereafter they are referred to as soil-153 and soil-266 (Table 6-1). In addition, water extractable arsenic was measured using a 1:4 soil to water ratio after shaking for 2 h at 25 1oC. Extractable P was extracted with the Mehl ich-3 extractant, and determined using a modified molybdenum blue method to mini mize arsenic interfer ence (Carvalho et al., 1998). Select physical and chemical prop erties of the soil are summarized in Table 6-1. Table 6-1. Selected properties of the soils used in this study Property Soil-153 Soil-266 pH (1:2 soil/water ratio) 7.32 + 0.02* 7.37 + 0.04 Organic matter content (g kg-1) 10 + 0.08 13 + 0.40 Cation exchange capacity(cmol(-) kg-1 4.40+ 0.02 5.20 + 0.03 Total As (mg kg-1) 153 + 5.30 266 + 5.45 Mehilch III-As (mg kg-1) 43.7 + 2.60 61.9+ 3.60 Mehilch III-P (mg kg) 28.8 + 0.3 26.9 + 0.7 Total Zn (mg kg-1) 0.88 + 0.3 0.90 + 0.5 Total Cd (mg kg-1) 0.08 + 0.01 0.05 + 0.01 Total Cu (mg kg-1) 8.50 + 2.2 14.0 + 1.2 Total Cr (mg kg-1) 40.8 + 4.2 70.5 + 3.3 Sand (g kg-1) 882 + 4.32 884 + 5.76 Silt (g kg-1) 91+ 0.04 90 + 0.07 Clay (g kg-1) 27+ 0.01 26 + 0.01 * Means + standard deviations, n = 3 Experimental Design This greenhouse experiment was set up as a completely randomized design in a 2 by 2 factorial scheme consistin g of two arsenic levels (153 mg kg-1 and 266

PAGE 107

93 mg kg-1) with two fern species ( P. vittata and P. biaurita ) (Figure 6-1). Each treatment was replicated three times. Air-dried soil (2.5 kg) was weighed into each pot and thoroughly mixed with 3.0 g of Osmocote, an extended time-re lease fertilizer (186-12) (Scotts-Sierra Horticultural Products Co., Marysville, OH). After one week of equilibrium at field capacity, one plant was transplanted into each pot. The rhizopot used in this study was 16 cm in height and 13 cm in diameter and composed of two compartments separa ted by a plastic frame covered with a nylon mesh cloth (mesh size 45 m). Root growth was limited to the central compartment of 7 cm of diameter (Figure 5-1). Pre-experimental biomass of fronds and roots of the two fern species was al so determined. The average dry frond and root biomass were 5.60 g plant-1 and 1.85 g plant-1, respectively. The plants were allowed to grow for eight weeks in a gr eenhouse (average temperature was 14oC at night to 30oC during the day, with an average photosynthetically active radiation flux of 825 mol m-2 s-1). The plants were watered throughout the study to keep the soil at approximate ly 70% of field capacity. After 8 weeks of growth, the plants we re harvested and separated into roots and fronds. Plant tissues were washed thor oughly with tap water, and then rinsed with deionized water. The fronds a nd roots were oven-dried for 3 d at 65o C, weighed and ground with a Wiley mill to pass through a 1 mm mesh screen for chemical analysis. Soil samples from each pot were collected from the rhizosphere and bulk soil. The rhizosphere is define d as the soil attached to the roots.

PAGE 108

94 Rhizosphere soil was shaken gently from the roots and passed through a 2 mm sieve to separate it from the roots. Chemical Analysis Plants and soils digestion and analyses were performed as described in the Chapter 4. Rhizosphere and bulk soil were evalua ted for water-soluble arsenic and phosphorus, and dissolved orga nic carbon (DOC) from the supernatant of a 1:4 soil to water ratio solution (Olsen & Somme rs, 1982), obtained after shaking (1 h), centrifuging (15 min at 3500 g) and filt ering (0.45 m syringe filter). The determination of water-soluble arsenic was performed by the same procedure described above. Concentrations of DOC were measured using a TOC-5050A Total Organic Carbon analyzer (Shimadzu, Japan). The pH in the rhizosphere and in the bulk soil was also measured us ing a 1:2 soil to water ratio. Data Analysis Plant and soil effects were determined by analysis of variance according to the linear model procedure of the Statistical Analysis System (SAS Institute Inc. 1987). Treatment means were separated by Duncan’s multiple range tests using a level of significance of P <0.05. Results and Discussion Arsenic Accumulation and Distribution in the Ferns Plant arsenic concentration and biomass are two important traits in selecting an ideal plant species to be used for phytoextraction purpose. We compared the arsenic accumulation ability of P. biaurita , a newly reported arsenic

PAGE 109

95 hyperaccumulator (Srivastava et al., 2006) with P. vittata , one of the most studied arsenic hyperaccumulators. Regardless of th e soil arsenic levels (153 and 266 mg kg-1), P. vittata was more efficient in taking up arsenic from the two soils than P. biaurita (Table 6-2). Figure 6-1. Pteris biaurita and Pteris vittata growing in an arsenic-contaminated soil. Frond and root arsenic concentrations of the two fern species were proportional to the soil arsenic concentra tion. The correlation between the soil and frond arsenic concentrations has been well documented (Tu and Ma, 2003; Luongo and Ma, 2005; Al Agely et al., 2005). After 8 weeks of growth, with soil ar senic concentrations increasing from 153 to 266 mg kg-1, arsenic concentrations in the fronds of P. vittata increased from

PAGE 110

96 758 to 3759 mg kg-1 (5 times greater) while those in P. biaurita increased from 468 to 2549 mg kg-1 (5.4 times greater). The arseni c concentration in the fronds of P. biaurita was 1.6 and 1.5 times lower than that of P. vittata in the soil-153 and soil266, respectively (Table 6-2). Even though P. biaurita accumulated large amount of arsenic in its fronds, P. vittata was more efficient. Table 6-2. Comparison of the arsenic concentration (mg kg-1), translocation factor (TF) and bioconcentration factor (BF) of P. vittata and P. biaurita after 8 weeks of growth in tw o As-contaminated soils. Variable P. biaurita P. vittata P. biaurita P. vittata As in soil 153 266 As in fronds 468 b* 758 a 2549 b 3759 a As in roots 284 a 162 b 1024 a 378 b BF in fronds 3.05 b 4.95 a 9.60 b 14.1 a BF in roots 1.85 a 1.06 b 3.85 a 1.42 b TF 1.64 b 5.11 a 2.51 b 9.98 a * Means followed by the same letter in a row for the same soil are not statistically different by the Duncan test at P < 0 : 05. On the other hand, regardless of the so il arsenic concentration, the arsenic concentrations in the roots of P.biaurita were much greater th an in the roots of P. vittata, further confirming the greater effectiveness of P. vittata in translocating arsenic from the root to the fronds. The arsenic concentration in the fronds of P. biaurita was only 1.6 and 2.5 times greater than the arsenic concentration in the roots in soil-153 and soil-266, respectively, while the arsenic concentration in the frond of P. vittata was 4.7 and 10 times greater than in the roots in the soil-153 and

PAGE 111

97 soil266, respectively. The large arseni c concentration in the roots of P. biaurita may be a limiting factor for its use in phytoextraction. Arsenic accumulation by P. vittata in in the soil-153 was lower than previously reported (Ma et al., 2001; Tu and Ma, 2002; Wang et al., 2002; and Fayiga et al., 2004), suggesting that P. vittata was not functioning in its full potential. Using the same CCA soil containing 131 mg kg-1 As, Fayiga et al. (2004) reported arsenic concentrations in th e fronds and roots of 4,200 and 130 mg kg-1 after 8 weeks of growth. Tu et al. (2002) reported 1,280 and 22.3 mg kg-1 after growing in the same CCA soil containing 98 mg kg-1 for 8 weeks. However, in the four studies carried out in this research using CCA soil, the highest arsenic concentr ation in the fronds of P. vittata was 3,759 mg kg-1 after growing in a CCA soil containing 266 mg kg-1 (Table 6-3). In the other three studies, the arsenic concentration in the fronds ranged from 325 mg kg-1 to 747 mg kg-1. Even though the CCA soil samples were taken from the same CCA contaminated site for this study and for the studies of Fayiga et al. (2004) and Tu et al (2002), the arsenic concen trations were different fo r all samples due to the heterogeneity of the site. Based on the results gathered in this research, it is suggested that the differences observed among the different st udies are likely due to plant age and seasonal effects, among others. It is not po ssible to make considerations about age of the plants, seasonal effects etc, for someonelse’s studies, however, in this research, the studies of the chapters 4, 5 a nd 6 were carried out from late January to March. Therefore, it is possible that the te mperature and light affected the growth

PAGE 112

98 and arsenic accumulation by P. vittata by the same way as described in the chapter 3. Based on bioaccumulation factor (BF), defined as the ratio of arsenic concentration in the plant tissue to that in the soil (Tu and Ma, 2002), both plant species were efficient in accumulating arse nic from the soil-266 (BF>1) (Table 62). However, the BF in P. biaurita frond was approximately 1.4 times lower than that of P. vittata in the soil-266 and 1.6 times lower in the soil-153. In this regard , P. biaurita was not as efficient as P. vittata in accumulating arsenic from either soil. Plants used for phytoextraction must be able to transport most of the contaminant to the fronds, which facilit ates contaminant removal (Raskin and Ensley, 2000). In both plants we observed pr eferential partitioning of arsenic to the fronds, with transfer factor (TF) (the ra tio of contaminant c oncentration in the fronds to that in the roots) also > 1, typical for hyperaccumulator plants (Baker, 1981; McGrath and Zhao, 2003). However, the ability of P. vittata in transferring arsenic from the root to the frond was much greater than that of P. biaurita , in both soils (Table 6-2). According to Ma et al. (2001), the high effi ciency of arsenic accumulation by arsenic-hyperaccumulator fern species is associated with their ability to rapidly translocat e absorbed arsenic from root s to shoots, although they differ in efficiency. Plant Biomass of P. biaurita and P. vittata Total dry matter is another key compone nt in determining the effectiveness of contaminant removal from a site by a hyperaccumulator plant. After eight weeks

PAGE 113

99 of growth, there were no significant diffe rences between the two fern species in terms of biomass production (F igure 6-2). Frond biomass of P. biaurita and P. vittata increased nearly 3-fold as compared to their original biomass, with the exception of the frond biomass of P. biaurita in the soil-266, which increased approximately two-fold. In contrast, th e root dry matter accretion of both fern species was slightly smaller in the so il-266 (Figure 6-2). Many hyperaccumulator ferns, incl8uding P. vittata , have shown a substantial biomass production either in greenhouse or field conditions (Komar et al . 1998; Francesconi et al. 2002). For instance, Fayiga et al. (2 004) reported a much higher increase (12 times) in the plant biomass of P. vittata after eight weeks of growth in arsenic contaminated soil (131 mg As kg-1). 0 2 4 6 8 10 12 14 16 18 20FrondRootFrondRoot Soil 1 Soil 2 Plant biomass (g plant-1 dw ) P.vittata P. biaurit a Figure 6-2. The frond and root biomass (mg kg-1 DW) of P. vittata and P. biaurita after 8 weeks of growth in two arse nic contaminated soils (Soil 1: 153; Soil2: 266). Results are means + standard deviations (n = 3). The final frond and root biomass were similar for the two fern species growing in the soil-153. The small root bi omass of both ferns, especially at the

PAGE 114

100 higher arsenic level, can be a limiting factor for both plant species in plant arsenic uptake, since a smaller volume of soil was exploited. Rhizosphere Chemistry Plant cultivation effects can be unders tood by observing the changes in the rhizosphere such as soil pH and DOC, wh ich may be important for the mobilization of soil arsenic. Therefore, we measured the concentrations of the water-soluble arsenic and P in the bulk soil and rhiz osphere of the two plant species. The rhizosphere soils were more alkaline than the bulk soils, with the difference for P. vittata being 0.28 in the soil-153 and 0.51 units in the soil-266. The pH difference for P. biaurita was 0.22 and 0.47 in the two soils (Figure 6-3). No change in pH was observed in the bulk so il. This clearly show ed that plant roots induced the changes in soil pH. The pref erential uptake and exudation of either cations or anions by plant roots occurs in order to maintain their charge balance (Hinsinger et al., 2003), with consequent changes of rhizosphere pH of up to 1-2 units. Similar results were observed in the chapter 5 but with greater differences between bulk and rhizosphere soils after growth of P. vittata. The fact that both P. vittata and P. biaurita increased the rhizosphere pH in this study while N. exaltata even reduced the rhizosphere pH in the ch apter 3 study suggests that soil pH may play a role in the arsenic acumulation by hyperaccumulator ferns.

PAGE 115

101 6.6 6.8 7 7.2 7.4 7.6 7.8 8 8.2 P.vittata P. biauritaP.vittata P. biaurita Soil 153control153 Soil 266control266 Soil pH Bulk soil Rhizospheret Figure 6-3. The bulk soil and rhizosphere pH of P. vittata and P. biaurita after 8 weeks of growth in two arsenic cont aminated soils. Results are means + standard deviations (n = 3). The DOC concentrations in the rhiz osphere of both fern species were approximately twice as high as those in the bulk soil (Figure 6-4), with the exception of P. biaurita in the soil-266, where the DOC content was nearly three times higher than the bulk soil. It has been reported that P. vittata releases organic acids of low molecular weight such as phytic and oxalic aci d as well as small quantities of ascorbic, succinic and fumaric acid via root exudation (Tu et al. 2003). The presence of these acids may have a significant influence on arsenic mobilization in the plant rhiz osphere. It is known that r oot exudation varies with species, age and relative st ress (Uren, 2000). For example, in the soil-153, P. biaurita had 15% less DOC in its rhizosphere than P. vittata . But in the soil-266

PAGE 116

102 the situation reversed, with P. biaurita ’s rhizosphere having 30% more DOC than P. vittata’s. The change in rhizosphere DOC concentration may have further implications with respect to the availa bility of nutrients and soil microbial population (Kullberg et al. 1993). In gene ral, DOC readily forms both aqueous and surface inner-sphere complexes with cationic metals and metal oxides, which, in turn, may associate strongly with othe r dissolved anions by metalbridging mechanisms (Thanabalasingam and Pickeri ng, 1986), diminishing the tendencies of such anions to form surface complexes. T hus, the reaction of DOC with arsenic is expected and has the potential to influence on arsenic sorption and mobility in soils. 0 50 100 150 200 250 300Bulk soilRhizospherecontrolBulk soilRhizospherecontrol Soil-153 Soil-266 DOC (mg kg-1 ) P. vittata P.biaurita Figure 6-4. The bulk soil and rhizosphe re dissolved organic carbon (DOC) of P. vittata and P. biaurita after 8 weeks of growth in two arsenic contaminated soils. Results are means + standard deviations (n = 3).

PAGE 117

103 Water-soluble (WS) metal is a good indi cator of bioavailability for plant uptake from a soil (McBride, 1994), since plants preferential ly take up their nutrients from a soil solution (Linehan et al., 1985). The concentrations of watersoluble arsenic and P were significantly hi gher in the rhizosphe re than in the bulk soil for both ferns (Figure 6-5 and 6-6). Th e increase in the rhizosphere WS-P and WS-As (greater than the control and bulk so il) indicated that mobilization of P and arsenic was plant driven. Pteris biaurita presented 27 % less WS-P in its rhizosphere in the soil-153 than P. vittata . In contrast, the WSP concentration in P. biaurita rhizosphere was 37 % gr eater than that in P. vittata rhizosphere in the soil 266. It is very likely that higher DOC concentration in P. biaurita rhizosphere in the soil-266, compared to that in the soil-153, may have cont ributed to the high P as well as high arsenic mobilization in the soil. 0 1 2 3 4 5 6 7 8Bulk soilRhizospherecontrolBulk soilRhizospherecontrol Soil-153 Soil-266 Water-soluble P (mg kg-1) P. vittata P.biaurita Figure 6-5. The water-soluble P concentratio ns in the bulk soil and rhizosphere of P. vittata and P. biaurita after 8 weeks of growth in two arsenic contaminated soils. Results are means + standard deviations (n = 3).

PAGE 118

104 The concentrations of WS-ars enic in the rhizosphere of P. biaurita in the soil-266 was greater than that of P. vittata , likely due to both lower arsenic uptake and higher DOC exudation, showing that P. biaurita was not as efficient in arsenic uptake as P. vittata . In general, it is hypothesized that, as plant uptake depletes WS-arsenic, the plants are capable of rel easing more arsenic in to the soil solution, one of the characteristics of hyperaccumulat ors. This did not seem to apply in our study since the fern with more WS-arsenic in its rhizosphere was not the one that took up more arsenic, suggesting that othe r external factors contributed to the higher WS-arsenic in the rhizosphere. 0 5 10 15 20 25 Bulk soilRhizospherecontrolBulk soilRhizospherecontrol Soil-153 Soil-266 Water-soluble As (mg kg-1 ) P. vittata P.biaurita Figure 6-6. The water-soluble As concentratio ns in the bulk soil and rhizosphere of P. vittata and P. biaurita after eight weeks of growth in two arsenic contaminated soils. Results are means + standard deviations (n = 3). A higher WS concentration of low mobility elements (As and P) in the rhizosphere is an indicator of either lo w uptake and/or high solubilization of the elements by plants. It is possible that the arsenic was being sol ubilized at a faster

PAGE 119

105 pace than the capacity of the plants to ta ke it up, which resulted in accumulation of arsenic in the rhizosphere. The combined data of WS-P and WS-A s may partially expl ain the low plant arsenic uptake in the soil-153 (Figs. 6-5 and 6-6). Compared to the soil-266, the WS-P was greater but its WS-As was lowe r in the soil-153. In other words, the molar ratio of WS-P to WS-As in the so il-153 was 8.5 compared to 0.69 in the soil 266. The high WS-P in the soil-153 may have severely reduced the plant’s ability to take up arsenic via competition. However, the greater P concentration in the soil153 did not translate into high P accumulati on in the fronds nor in the roots of P. vittata (Table 6-3). The P con centration in the fronds of P. vittata was 0.46% in the soil-153 compared to 1.08% in the soil-266 though the root P concentration was the same (0.15%). Elemental Composition Proper plant nutrition may enhance arsenic accumulation and the potential for soil remediation by phytoextraction. The P concentrations in the fronds of both species were 2.5 ( P. vittata ) and 1.6 ( P. biaurita ) times higher in the soil-266 than in the soil-153 (Table 6-3). Soil arse nic concentrations influenced frond P concentrations but not root P in both fern species. Phosphorus concentration in the fronds were within the normal range for most plants (0.1 to 0.5 %) (Marschner, 2003) in the soil-153, which is consistent with the results of Tu and Ma (2005). However, significantly greater P concentr ation was observed in the fronds in the soil-266 (1.08 % for P. vittata and 0.87 % for P. biaurita ).

PAGE 120

106Table 6-3. The macronutrients (P, K, Ca, Mg) (g kg-1) and micronutrients (Fe, Zn, Mn) content (mg kg-1) in the fronds and the roots of P. vittata and P. biaurita after 8 weeks of growth in tw o arsenic-contaminated soils. P. vittata P. biaurita P. vittata P. biaurita Elemental content Frond Root Frond Root Frond Root Frond Root -----------------Soil 153 ------------------------------------------------Soil 266 --------------------Phosphorus 4.60 A 1.50 a 5.50 A 1.30 a 10.8 A 1.50 a 8.70 A 1.70 a Potassium 24.8 A 4.90 a 27.6 A 2.70 a 33.5 A 2.80 a 23.2 B 2.70 a Calcium 7.10 A 46.9 a 5.50 B 3.20 b 6.20 A 32.6 a 5.90 A 21.3 a Magnesium 3.40 A 2.50 b 3.30 A 3.70 a 4.20 A 3.00 a 3.50 A 1.80 a Iron 160 A 1828b 72.0B 2625 a 75.0 A 2005 a 80.0 A 483 b Zinc 52.3 A 89.0 a 44.0 A 84.0 a 53.7 A 131 a 38.3 B 117 a Manganese 23.7 A 39.3 b 27.0 A 53.0 a 26.3 A 51.7 a 22.3 A 41.0 b Numbers with same uppercase lette r or lowercase letter in a row are not statistically diffe rent by the Duncan test at P < 0 . 05 for the same soil or plant, respectively.

PAGE 121

107 Phosphate is critical for pl ant growth especially when arsenic is at toxic levels (arsenic concentration was 73% greater in the soil-266 than the soil-153). Both fern species had the same P concentrations when they were grown in the same soil. However, the frond P concentrations in both ferns were gr eater in the soil containing more arsenic. Luongo and Ma (2005) studied the characteristics of several Pteris and nonPteris ferns and observed that, for all ferns except P. vittata , with increasing arsenic concentrations, phosphate concentrations in the fronds and roots d ecreased. In their study, phosphate concentrations in P. vittata increased by 91.8 % in the fronds and 455 % in the roots when growing in a solution containing 10 mg L-1 As, as compared to the control. Phosphate plays an important role in ar senic detoxification in ferns. Arsenate interferes with phosphate metabolism by replac ing phosphate in ATP synthesis and /or in various phosphorolysis reactions (Tu and Ma, 2003b). Therefore, it may exist a proportion of P in ferns that may be repl aced by arsenic without causing significant changes in plant metabolism. Another interesting observation from these re sults is that the P concentrations in the roots were lower than those in the fronds (Table 6-3). Compared to non arsenic hyperaccumulators, more P tends to be con centrated in the roots in hyperaccumulator plants (Tu and Ma, 2003c). Luongo and Ma (200 5) also reported that higher level of arsenic in solution reduc ed P translocation in P. vittata , which was able to keep much more P in the roots than other ferns. The same results were observed in Chapter 5 of this research when P. vittata and N. exaltata were grown in the control soil (0.4 mg kg -1 As). Both plants concentrated more P in the fr onds than in the roots. On the other hand, P. vittata concentrated more P in the roots than in the fronds in the arsenic contaminated

PAGE 122

108 soil, agreeing with the results of Tu and Ma (2003c). Therefore, the greater accumulation of P in the fronds of P. vittata and P. biaurita instead of in the root s likely had a negative effect on the arsenic uptake by these plants by interfering with the arsenic tolerance mechanisms. The slow growth of the plants during the experimental period and the yellowish of the leaves after 6 weeks of growth, mostly in P. biaurita , were signals that indicated the poor conditons of the plants. The ability of P. vittata in maintaining high concentra tions of P in its roots is reported to be one of its mechanisms of ar senic tolerance (Tu and Ma, 2003c). This is because arsenate uptake is intimately linke d to phosphate nutrition, with increased phosphate uptake reducing arsenate uptak e, via suppression of a high-affinity phosphate/arsenate uptake system (Meharg and Macnair, 1992). The phosphate/arsenate transporter has a higher affinity for phosphate, and if external phosphate status is high, phosphate wi ll be taken up preferentially to arsenate. This is the case for both plants. Take the so il-266 for example, the molar ratios of WS-P to WS-As were 0.69 and 0.81 for P. vittata and P. biaurita , respectively (Figures 6-5 and 6-6). On the other hand, the molar ratio of P to As in the fronds of P. vittata and P. biaurita were 7.0 and 8.0, respectively (Tables 6-2 and 6-3). This indicates that the plants preferentially take up more P than arsenic though the WS-As concentrations were greater than WS-P concentrations in the soil. Compared to P. biaurita , P. vittata not only took up more arsenic (Table 6-2) but also more P (Table 6-3). It is possible that the P uptake system in P. vittata was more efficient than P. biaurita . In this study, the plants absorbed more phos phate to cope with the greater level of arsenate in the soils. Considering the abil ity of a plant to take up P as one of the

PAGE 123

109 mechanisms of arsenic detoxification, both Pteris ferns behaved the same way in this study, that is, absorbed more P from the high arsenic soil. However, they concentrated more P in the fronds than in the roots, which was unexpected since more P in the roots is hypothesized to be required to minimize the arsenic toxicity effect by improving phosphate nutrition in the roots (Sneller et al., 1999; Tu and Ma, 2003b; Luongo and Ma, 2005). Al Agely et al. (2005) reporte d similar results when growing P. vittata in the presence and absence of mycorrhizal fungi. They observed that phosphorus concentrations were greater in the fronds than in the roots of plants. The phosphorus concentrations in the fronds of P. vittata grown in arsenic -c ontaminated soils (As concentrations ranging from 69 to 28, 522 mg kg-1), under field conditions, were also higher than those in the roots (Liao et al., 2004). Therefore, the interaction between P and arsenic still remains unclear in term of arsenic hyperaccumulation and detoxification by these plants. The nutritional status of th e two plants was similar. However, in the soil-153, P. vittata accumulated more Ca in the fronds than P. biaurita (Table 6-3). Furthermore, the greater influx of Ca in P. vittata could be explained by the need for balancing the anions phosphate and arsenate taken up by the plants. It is hypothesized that Ca is related to the arsenic hyperaccumulation in ferns (Chen et al. 2003). The oxidative stress caused by metal toxicity can increase Ca concentrati ons in plant cells, where Ca stimulates the induction of free radical scavenge rs (Price et al. 1994) to main tain the cellular unity of a plant. In this study, frond Ca contents of both fern species did not differ and were in the normal range for most plants ( 0.2 to 1.0 %) (Marschner, 2003).

PAGE 124

110 Our results are similar to those of Li et al. (2005) , who observed Ca concentrations in the fronds of P. vittata ranged from 0.28 to 0.32 %. However, Ca concentrations of both fern species were much higher in the roots than in the fronds. We observed Ca enrichment in the roots of both ferns species, especially in the roots of P. vittata in the soil-153 (4.69 %). U nder high arsenic conditions, the roots accumulated Ca to maintain the cellular integrity. Liao et al . (2004) also reported Ca enrichment in the roots of P. vittata under field conditions. In contrast, P. biaurita had more root Mg than P. vittata. In addition, results for Fe, Zn and, Mn we re obtained for both species (Table 6-3). Zinc and Mn concentrations did not differ between the two fern species, but frond Fe concentrations of P. vittata were significantly higher (2.2 fold) than that of P. biaurita in the soil-153. However, no difference was f ound in the soil-266, the values were still within the normal range fo r plants (50 to 250 mg kg-1; Marschner, 2003). The iron content in the roots of both fe rn species was greater than thos e in the fronds. In the soil153, P. biaurita had 30% more Fe in the roots than those in P. vitatta . In contrast, in the soil-266, P. biaurita had 76 % less Fe as compared to P. vittata . Wei et al. (2005) reported greater Fe concentrations in P. vittata and P. cretica when compared with other plants. Wenzel et al. (2003) observed an incr ease in Fe concentrations in the rhizosphere of P. vittata , which they related to the dissolution of soil minerals caused by root activity. Under these experimental condi tions, both acidification and the release of DOC into the rhizosphere may have contributed to the m obilization of iron, which was taken up by the roots.

PAGE 125

111 CHAPTER 7 CONCLUSIONS Phytoextraction is a new technology and its success depends on both plant and soil factors such as soil suitability for plant grow th, depth of contamination, depth of the plant root system, level of contamination, and ur gency in cleaning up. Fu rthermore, there is need for a better understanding of the physiolo gy, biochemistry, and metal uptake of the plants. The ability of P. vittata to grow and take up arseni c from different soils with different levels of arsenic was investigated. Th e ferns grew well in all soils showing little toxicity symptom. However, both the bioma ss production and arsenic concentrations in the plants depended on the growing season, w ith best results observed during warmer months. After three harvests, the plants removed 6.4 to 13 % arsenic from the soils, but failed to re-grow in all soils. Plant growth in fluenced arsenic mobilization from different fractions of the soils as evaluated by a se quential extraction procedure, showing that plants took up arsenic from the le ss available arsenic pools. It is likely that the exudation of organic acids in the rhizosphere played a role in this process, as evidenced by the amounts of arsenic extracted from the soils using the organic acids and its positive relationship with the amounts of arsenic take n up by the plants . In almost all soils studied, arsenic uptake by P. vittata was similar to the amounts of arsenic extracted by the organic acids. Therefore, in evalua ting the bioavailabil ity of arsenic to P. vittata, the use of organic acids may be a good approach than the total and Mehlich III arsenic.

PAGE 126

112 The influence of P. vittata age on the arsenic removal from a contaminated soil was also investigated. Younger plants accumulated more arsenic in the fronds as a result of their higher metabolic activity, higher rate of biomass accretion and lower P concentration in the fronds and roots. On the other hand, plants with lower biomass accretion, thus, lower nutritional requirement, had higher arsenic and P concentration in the roots, and lower arsenic concentration in the fronds. Therefore, in establishing a program of arsenic phytoextraction using P. vittata, the use of young plants has the benefits of higher rate of arsenic uptake, higher arsenic transloca tion to the fronds and consequently higher efficiency of arseni c removal. Furthermore, the use of younger plants offers the possibility for managing harv ests before spores maturation, therefore avoiding the risks of uncont rolled ferns propagation. The differences between the root system of a hyperaccumulator and nonaccumulator species may account for some of its elemental scavenging ability of the species. This was observed through a comparis on of the root system characteristics (biomass, length, area and diameter), plant biomass and arsenic uptake and root uptake efficiency of P. vittata with a non-arsenic accumulator fern species N. exaltata. The study revealed that P. vittata had a more extensive root system (root biomass, length and area) and possessed a greater proportion of fine roots (root diameter) than N. exaltata. Pteris vittata was more efficient in removing arsenic from an arsenic-contaminated soilthan N. exaltata . Also, plant growth significantly imp acted the total arsenic, water soluble arsenic, as well as arsenic di stribution in different fractions in the rhizosphere soil, but not the bulk soil. Pteris vittata had a greater effect on arsenic mobilization from different pools than N. exaltata . The greater decrease in th e soil pH and increase in DOC

PAGE 127

113 concentrations in the rhizosphere of P. vittata may also have enhanced its ability to take up more arsenic. Pteris vittata was compared with another arse nic hyperaccumulator fern of the same genus, P. biaurita, in terms of its arsenic ac cumulation, biomass production and changes in the rhizosphere in two soils with different arsenic conten ts. Regardless of the levels of arsenic contamination, P. vittata removed more arsenic than P. biaurita from both soils. Although the two ferns have differe nt growth requiremen ts, in general, both ferns seemed to have similar arsenic hypera ccumulation and growth characteristics. However, P. biaurita was more sensitive when exposed to higher soil arsenic level, showing some arsenic toxicity symptoms in the fronds. Therefore, P. vittata continues to be a better choice for the application to arsenic phytoextraction in subtropical ecosystems. A relatively small group of plants are capab le of sequestering ar senic in their shoot tissues at high concentrations with the majo rity of them belonging to the Pteridophytes. Efficient reproduction of these plants for a large scale phytore mediation can be a challenge. Cultivation of arsenic hyperaccumulators has to be optimized in order to provide an effective alterna tive to common engineering-base d remediation technologies. In recent years, major scientific prog ress has been made in understanding the physiological mechanisms of arsenic uptake and transport in these plants. However, little is known about the molecular basis of arsenic hyperaccumulation. Therefore, future research should focus on molecular genetic t echnologies and, possibly, transgenic plants with increased resistance and uptake not onl y of arsenic but also other heavy metals. Discovering the molecular mechanisms underlying arsenic tolerance and accumulation in

PAGE 128

114 the arsenic hyperaccumulator ferns can be facilitated by identifying genes that are both necessary and sufficient for these properties. Although phytoextraction is a promising reme diation technology, it is still in its early stage of development and little informa tion is available related to site cleanup from start to finish. This is probably due to the fact that few, if any, site s that have yet to be completely remediated using this technology. It is the case for arseni c phytoextraction as the first arsenic hyperaccumulator plant was only discovered in 2001.

PAGE 129

115 LIST OF REFERENCES Adriano, DC. 1986. Trace Elements in the Terrestrial Environment . New York, NY, USA: Springer. Adriano DC. 2001. Trace Elements in the Terrestrial Environment. New York, NY, USA: Springer. Al Agely A, Sylvia DM, Ma LQ. 2005. Mycorrhizae increase arsenic uptake by the hyperaccumulator chinese brake fern ( Pteris vittata L.). Journal of Environmental Quality 34 : 2181-218. An ZZ, Huang ZC, Lei M, Liao XY, Zheng YM, Chen TB. 2006. Zinc tolerance and accumulation in Pteris vittata L. and its potential for Phytoremediation of Znand As-contaminated soil. Chemosphere 62: 796-802 Asher CJ, Reay PF. 1979. Arsenic uptake by barley hordeum-vulgare cultivar zephyr seedlings. Australian Journal of Plant Physiology 6 : 459-466. Assunao AGL, Da Costa Martins P, De Folte r S, Vooijs R, Schat H, Aarts MGM. 2001. Elevated expression of metal transporte r genes in three accessions of the metal hyperaccumulator Thlaspi caerulescens . Plant Cell Environment 24 : 217-226. Azcue JM, Nriagu JO. 1994. Arsenic: historic al perspectives. In: Nriagu JO, ed. Arsenic in the environmen t . New York, NY, USA: Jonh Wiley and Sons, 1-15. Baker AJM. 1981. Accumulators and excluders – st rategies in the response of plants to heavy metals. Journal of Plant Nutrition 3 : 643 – 654. Baker AJM, Brooks RR. 1989. Terrestrial highe r plants which hyperaccumulate metallic elements– a review of their dist ribution, ecology and phytochemistry. Biorecovery 1 :81-126. Baker AJM, McGrath SP, Reeves RD, Smith JAC. 2000. Metal hyperaccumulator plants: a review of the ecology and physiolo gy of a biochemical resource for phytoremediation of metal-polluted soils. In: Terry N, Bauelos G, eds. Phytoremediation of contaminated soil and water . Boca Raton, FL, USA: Lewis Publishers, 85. Baker AJM, Reeves RD, Mc Grath SP.1991. In situ decontamination of Heavy metal polluted soils using crops of heavy metal accumulating plants-a feasibility study . Boston, MA, USA: Columbus, Butterworth-Heinemann.

PAGE 130

116 Baker AJM, Whiting SN. 2002. In search of the Holy grail-a further step in understanding metal hyperaccumulation? New Phytologist 155 : 1-7. Barber SA. 1984. Soil nutrient bioavailabili ty: a mechanistic approach. , New York, NY, USA: John Wiley & Sons. Benson LM, Porter EK, PETERSON PJ. 1981. Arsenic accumulation, tolerance, and genotypic variation in plants on arseni cal mine wastes in S.W. England . Journal of Plant Nutrition 3 : 655-666. Bertrand I, Hinsinger P. 1997. Dissolution of an iron oxyhydroxi de in the rhizosphere of various crop Greer, G. K., R. M. Lloyd, and B. C. McCarthy. Factors influencing the distribution of pteridophytes in a southeastern Ohio hardwood forest. Journal Torr. Botany Society 124 : 11. Boisson J, Mench M, Vangronsveld J, Koppone n P, de Koe, T. 1999. Immobilization of trace metals and arsenic by different so il additives: evaluation by means of chemical extractions. Communications in Soil Sc ience and Plant Analysis 30 , 365– 387. Bondada B, Ma LQ. 2002 . Tolerance of heavy metals in vascular plants: arsenic hyperaccumulation by Chinese brake fern ( Pteris vittata L.), In: Nayar BK, Chandra S, Srivastava M, eds. Pteridology in New Millennium, Kluwer Academy Publishers, 397-420. Brach AR, NcNaughon SJ, Raynal DJ. 1993. Phot osynthetic adaptability of two fern Species of a northern hardwood forest. American Fern Journal 83 : 47. Brooks RR. 1998. Plants that hyperaccumulate heavy metals . Cambridge: University Press, 380p. Brooks, RR. 1998. Plants That Hyperaccumulate Heavy Metals . Cambridge, UK, University Press, p 1-53. Cai, Y, Ma LQ. 2003. Metal tolerance, Accumu lation, and Detoxificati on in Plants with Emphasis on Arsenic in Terrestrial Plants. ACS Symposium Series 835. Biogeochemistry of Environmentally Im portant Trace Elements. Yong Cai, O C Braids Eds. Cai Y, Su JH, Ma LQ. 2004. Low molecular we ight thiols in arsenic hyperaccumulator Pteris vittata upon exposure to arsenic and other trace elements. Environmental Pollution, 129 : 69-78. Caille S, Swanwick S, Zhao FJ, McGrath SP. 2004. Arsenic hyperaccumulation by Pteris vittata from arsenic contaminated soils a nd the effect of liming and phosphate fertilization. Environmental Pollution 132 :113-120.

PAGE 131

117 Cao X, Ma, LQ, Shiralipour A. 2003. Effects of compost and phosphate amendments on arsenic mobility in soils and ar senic uptake by the hyperaccumulator Pteris vittata L. Environmental Pollution , 126 :157-167. Cao X, Ma LQ, Tu C. 2004. Antioxidative responses to arsenic in the arsenic hyperaccumulator Chinese Brake fern ( Pteris vittata L.). Environmental Pollution 128: 317-325. Carter MR. 1993. Sampling and Methods of Analysis . Boca Raton, FL, USA. Lewis Publishers. Carvalho LHM, De koe T, Tavares PB. 1998. An improved molybdenum blue method for simultaneous determination of inorganic phosphate and arsenate. Ecotoxicology and Environmental Restoration 1: 13-19. Casarin V, Plassard C, Hinsinger P, Arvi eu JC. 2004. Quantificati on of ectomycorrhizal fungal effects on the bioavailability and mob ilisation of soil P in the rhizosphere of Pinus pinaster. New Phytologist 163 : 177– 185. Chen JH, Barber SA. 1990. Soil pH and phosphorus and potassium uptake by maize evaluated with an uptake model. Soil Science Society of the American Journal 54 , 1032-1036. Chen M, Ma LQ. 1998. Comparison of four EPA digestion methods for trace metals using certified and Florida soils. Journal of Environmental Quality 27 : 1294-1300. Chen M, Ma LQ, Harris WG. 1999. Baseline co ncentrations of 15 trace elements in Florida surface soils. Journal of Environmental Quality 28 : 1173-1181. Chen TB, Huang ZC, Huang YY, Xie H, Liao XY. 2003. Cellular distribution of arsenic and other elements in hyperaccumulator Pteris nervosa and their relations to arsenic accumulation. Chinese Science Bulletin 48 : 1586-1591. Chlopecka A, Adriano DC. 1996. Mimicked in s itu stabilization of metals in a cropped soil: bioavailability and chemical forms of zinc. Environmental Science and Technology 30: 3294. Colleen MH, Matthew JLF, Scott F. 2002. Spatia l and temporal association of as and Fe on aquatic plant roots. Environmental Scie nce and Technology . 36: 1988-1994. Cotter-Howells JD, Champness PE, Charnock JM. 1999. Mineralogy of Pb/ P grains in the roots of Agrostis capillaris L. by ATEM and EXAFS. Mineralogy Magazine 63 :777-789 Cunningham SD, Shann JR, Crowley DE, Anderson TA. 1997. Phytoremediation of contaminated water and soil. In. Krug er E L, Anderson TA, Coats JR eds. Phytoremediation of Soil and Water Contaminants . Washington, DC, USA. American Chemical Society , 2-17.

PAGE 132

118 Day PR. 1965. Pippete method of particle size analysis. In : Black CA, ed. Methods of soil analysis . Part 1. Ist edn. Agronomy Monogra phy. 9. ASA, Madison, WI, USA: 552-562. De Koe T. 1994. Agrostis castellana and Agrostis delicatula on heavy metal and arsenic enriched sites in NE Portugal. Science of the Total Environment 145 :103-109. De Souza MP, Chu D, Zhao M, Zayed AM, Ruzin SE, Schichness D, Terry N. 1999. Rhizosphere bacteria enhance selenium accumulation and volatilisation by Indian mustard. Plant Physiology 119 : p.563/ 573. Diaz G, Azcon-Aguilar C, Honrubia M. 1996. Influence of arbuscular mycorrhizae on heavy metal (Zn and Pb) uptake and growth of Legeum spartum and Anthyllis cytisoides . Plant and Soil 180 : 241-249. Dinkelaker B, Rmheld V, Marschner H, 1989. Citric acid excretion and precipitation of calcium citrate in the rhiz osphere of white lupin ( Lupinus albus L.). Plant Cell Environment 12 : 285. Dobran S, Zagury G. 2005. Arsenic specia tion and mobilization in CCA-contaminated soils: Influence of organic matter content. The Science of the Total Environment , in press, a vailable online 1 August 2005 ( http://www.sciencedirect.com/science ). Doyle MO, Otte ML. 1997. Organism-induced accumulation of iron, zinc and arsenic in wetland soil. Evironmental pollution 96 :1/ 11. Ernst WHO.1996. Bioavailability of heavy metals and decontamination of soils by plants. Applied Geochemistry 11: 163. Eshel A, Waisel Y. 1972. Variations in s odium uptake along primary roots of corn seedlings. Plant Physiol ogy 49: 585-589. Fageria VD. 2001. Nutrient interactions in crop plants. Journal of Plant Nutrition 24: 1269-1290. Fayiga AO, Ma LQ, Cao RX. 2004. Rathinasabapat hi, d. B. Effects of Cd, Ni, Zn, and Pb on plant growth and arsenic uptake of hyperaccumulator Pteris vittata in a contaminated soil. Environment pollution 132 : 289-296. Fayiga AO, Ma LQ. 2006. Using phosphate rock to immobilize metals in soil and increase arsenic uptake by hyperaccumulator Pteris vittata . Science of the Total Environment 359: 17-25 Fitz WJ, Wenzel WW. 2002. Arsenic transf ormations in the soil-rhizosphere-plant system: fundamentals and potential application to phytoremediation. Journal of Biotechnology 99 : 259-278.

PAGE 133

119 Fitz WJ, Wenzel WW, Zhang H, Nurmi J, Stipek K, Fischerova Z, Schweiger P, Kollemsperger G, Ma LQ, Stingeder G. 2003.Rhizosphere characteristics of the Arsenic Hyperaccumulator Pteris vittata L. and Monitoring of phytoremoval efficiency. Environmental Science and Technolog 37 : 5008-5014. Forde B, Lorenzo H. The nutritional control of root development. Plant and soil 232 : 5168. Fourqurean JW, Cai Yong. 2001. Arsenic and phos phorus in sea grass leaves from the Gulf of Mexico. Aquatic Botany 71 : 247. Francesconi K, Visoottiviseth P, Sridokchan W, Goessler W. 2002. Arsenic species in an Arsenic hyperaccumulatingfern, Pityrogramma calomelanos: a potential phytoremediator of arsenic-contaminated soils. Science of the Total Environment 284 : p27/ 35. Frankenberger WT, Arshad M. 2002. Volatilisatio n of arsenic. In: Frankenberger WT Jr ed. Environmental Chemistry of Arsenic . New York, NY, USA: Marcel Dekker, 363-380. Fransson AM, 2001. Evaluation of oxalate/oxa lic acid for extracting plant available phosphorus in unferti lized acid soils. Communications in Soil Science and Plant Analysis 32 : 2469-2484. Frost RR, Griffin RA. 1977. Effect of pH on adsorption of arsenic and selenium from landfill leachate by clay minerals. Soil Science Society of American Journal 41 : 53-57. Galli U, Schuepp H, Brunold C. 1995. Thiols of Cu-treated maize plants inoculated with the arbuscular-mycorrhizal fungus Glomus intraradices . Physiology Plantarum 94 : 247-253. Garbisu C, Alkorta I. 2001. Phytoextraction: a cost-effective plan t-based technology for the removal of metals from the environment. Bioresource and Technology 77 : 229236. Gleyzes C, Tellier S, Sabrier R, Astruc M. 2001. Arsenic characterisationm in industrial soils by chemical extractions. Environmental Technology 22 : 27. Ghosh M, Singh SP. 2005. A review on phytoremediation of heavy metals and volatilization of its byproducts. Applied Ecology and Environmental Research 3 : 118. Greer GK, Lloyd RM, McCarthy B C. 1997. Fact ors Influencing the distribution of pteridophytes in a southeastern Ohio hardwood forest. Journal Torr Botany Society 12 4: 11.

PAGE 134

120 Guddanti S, Chambers J. 1993. GSROOT-Automated Root Length Measurement Program. Version 5.0. Guo FM, Yost RS. 1998.Partitioning soil phosphorus into three discrete pools of differing availability. Soil Science 163 : 822-833. Gussarsson M. 1994. Cadmium-induced altera tions in nutrient composition and growth of Betula pendula seedlings: The significance of the fine roots as a primary target for cadmium toxicity. Journal of Plant Nutrition 17 : 2151-2163. Harper M, Haswell SJ. 1988. A comparison of copper, lead and ar senic extraction from polluted and unpolluted soils. Environmental Technology Letters 9 :1271. Hartley-Whitaker J, Ainsworth G, Vooijs , R, Ten Bookum W, Schat H, Meharg AA . 2001. Phytochelatins are involved in di fferential arsenate tolerance in Holcus lanatus . Plant Physiology 126 : 299. Hatton PJ, Cole DJ, Edwards R. 1996. Influence of plant age on glutathione tranferases involved in herbicide detoxifi cation in corn (Zea mays L. ) and giant foxtail (Setaria faberi Herrm) Pesticide Biochemistry and Physiology 54 :199-209 Havlin JL, Beaton JD, Tisdale SL, Nelson WL. Soil Fertility and Fertilizers: An Introduction to Nutrient management . 7th ed. New Jersey, NJ, USA: Prentice hall, 515p. Heggo A, Angle A, Chaney RL. 1990. Effects of vesicular arbuscular mycorrhizal fungi on heavy metal uptake by soybeans. Soil Biology and Biochemistry 22: 865-869. Hill RH. 1972. Comparative patterns of CO2 exchange of three fern species of southeastern Michigan. Michigan Academy 4 : 303. Hinsinger P. 1998. How do plant roots acquire mineral nutrients? Chemical processes involved in the rhizosphere. Advances in Agronomy 24 : 225. Hinsinger P. 2001. Bioavailability of soil inor ganic P in the rhizosphere as affected by root-induced chemical changes: a review. Plant and Soil 237: 173-195. Hinsinger P, Plassard C, Tang C, Jaillard B. 2003. Origins of root-induced pH changes in the rhizosphere and their responses to environmental constraints: a review. Plant Soil 248 :43– 59. Howard JL, Brink VWJ. 1999. Sequential extr action analysis of heavy metals in sediments of variable composition using nitr iltriacetic acid to counteract resorption. Environmental Pollution 106: 285-292. Huang JWW, Poynton CY, Kochian LV, Elless MP. 2004. Phytofiltration of arsenic from drinking water using As-hyperaccumulating ferns. Environmental Science and Technology 38 : 3412-3417.

PAGE 135

121 Inskeep WP, McDermott TR, Fendorf S. 2002. Arsenic (V)/(III) cycling in soil and natural waters: chemical and microbiologi cal processes. In: Frankenberger WTJr, Ed. Environmental Chemistry of Arsenic. New York, NY, USA: Marcel Dekker, 183/ 215. Jiang W, Liu D. 1999. Effects of Pb+2 on root growth, cell division, and nucleolus of Brassica juncea L. Plant Science 47: 153-156. Jones DL. 1997. Encyclopaedia of Ferns .. Portland, Oregon, USA: Timber press; 433p. Jones DL. 1998. Organic acids in the rhizosphere – a critical review. Plant and Soil 205 : 25-44. Jones DL, Darrah PR.1994. Role of root deri ved organic acids in the mobilization of nutrients from the rhizosphere. Plant and Soil 166 : 247. Jungk AO. 2002. Dynamics of nutrient movement at the soil-root interface. In: Waisel Y, Eshel A, kafkafi U, eds. Plant Roots: the hidden half . New York, NY, USA: Marcel Dekker Inc. Kabata-Pendias A, Pendias H. 2001. Trace elements in soils and plants ., Boca Raton, FL, USA:CRC Press, 413p. Kavanagh PJ, Farago ME, Thornton I, Braman RS. 1997. Biovailability of arsenic in soil and mine wastes of the Tamar valley, SW England. Chemical Speciation and Bioavailability 9 : 77. Keller C, Hammer D, Kayser A, Richner W, Brodbeck M, Sennhauser M. 2003. Root development and heavy metal phytoextracti on efficiency: comparison of different plant species in the field. Plant and Soil 249 : 67 – 81. Kertulius GM. 2005. Arsenic hyperaccumulation by Pteris vittata L and its potential for phytoremediation of arsenic-contaminated soils .PhD dissertation, University of Florida, USA. Kertulis GM, Ma LQ, MacDonald GE, Chen R, Winefordner JD, Cai Y. 2005. Arsenic speciation and transport in Pteris vittata L. and the effects on phosphorus in the xylem sap. Environmental and Experimental Botany 54: 239. Khan AG, Kuek C, Chaudhry TM, Khoo CS, Hayes WJ. 2000. Role of plants, mycorrhizae and phytochelators in heavy metal contaminated land remediation. Chemosphere 41 :197-207. Knight K, Zhao FJ, McGrath SP, Shen ZG. 1997. Zinc and cadmium uptake by the hyperaccumulator Thlaspi caerulescens in contaminated soils and its effects on the concentration and chemical specia tion of metals in soil solution. Plant and Soil 197 : 71.

PAGE 136

122 Knuteson SL, Whitwell T, Klaine S J. 2002. In fluence of plant age and size on simazine toxicity and uptake. Journal of the Envi ronmental Quality 31 : 2096-103. Komar KM. 1999. Phytoremediation of arsenic contami nated soils: plant identification and uptake enhancement . M.S. thesis, University of Florida, USA. Komar KM, Ma LQ, Rockwood D, Syed A. 1998. Identification of ar senic tolerant and hyperaccumulating plants from arsenic contaminated soils in Florida. Agronomy Abstract 343 p. Kot A, Namiesnik J. 2000. The role of speciation in analytical chemistry. Trends in Analytical Chemistry 19 : 69-79. Kramer, U. 2005. Phytoremediation: novel appr oaches to cleaning up polluted soils. Current Opinion in Biotechnology 16 :133-141. Krishnamurti GSR, Naidu R. 2002. Solid-solut ion speciation and phyt oavailability of copper and zinc in soils. Environmental Science and Technology 36 : 2645-2651. Kullberg A, Bishop KA, Hargeby A, Jan sson M, Peterson RC. 1993. The ecological significance of dissolved organic carb on in acidified lakes. Ambio; 22: 331. Kumar PBA, Dushenkov V, Motto H, Raskin I. 1995. Phytoextraction: The use of plants to remove heavy metals from soils. Environmental Science and Technology 29 : 1232-1238. Lasat MM. 2002. Phytoextraction of toxic meta ls: a review of biological mechanisms. Journal of the Environmental Quality 31 , 109-120. Leah JM, Caseley JC, Riches CR, Valverde B. 1995. Age-related mechanisms of propanil tolerance in jungle-rice, Echinochloa colona . Pesticide Science 43: 347-354. Lee RB. 1982. Selectivity and kinetics of ion uptak e by barley plants following nutrient deficiency. Annals of Botany 50 :.429. Leung HM, Ye ZH, Wong MH. 2006. Interact ions of mycorrhi zal fungi with Pteris vittata (As hyperaccumulator) in As-contaminated soils. Environmental Pollution 139 :1-8. Li WX, Chen TB, Chen Y, Lei M. 2005. Role of trichome of Pteris vittata L. in arsenic hyperaccumulation. Science China Ser. Chinese-Life Science 48: 148. Liao XY, Chen TB, Lei M, Huang ZC, Xiao XY, An ZZ. 2004. R oot distribution and elemental accumulations of Chinese brake (Pteris vittata L) from As-contaminated soils. Plant and Soil 261: 109-116. Lin Z, Puls RW. 2002. Adsorption, desorption, a nd oxidation of arsenic affected by clay minerals and aging process. Environmental Geology 39 :753.

PAGE 137

123 Linehan DJ, Sinclair AH, Mitchell MC. 1985. M obilization of Cu, Mn and Zn in the soil solution of barley rhizospheres. Plant and Soil 86 : 147. Livesey NT, Huang PM. 1981. Adsorption of arsena te by soils and its re lation to selected chemical properties and anions. Soil Science 134: 88-94. Lombi E, Wenzel WW, Gobran GR, Adria no DC. 2001. Dependency of phytoavailability of metals on indigenous and induced rhizos phere processes: a review. In: Gobran GR, Wenzel WW, Lombi E, eds. Trace Elements in the Rhizosphere . Boca Raton, Fl, USA: CRC Press, pp 3-24. Lombi E, Zhao FJ, Dunham SJ, McGrath SP. 2000. Cadmium accumulation in populations of Thlaspi caerulenses and Thlaspi goesingense . New Phytologist 145 : 11-20. Luo YM, Christie P, Baker AJM. 2000. Soil solution Zn and pH dynamics in nonrhizosphere soil and in the rhizosphere of Thlaspi cearulences grown in a Zn/Cd contaminated soil. Chemosphere 41 : 161-164. Luongo T, Ma LQ. 2005. Characteristics of arsenic accumulation by Pteris and nonPteris ferns. Plant Soil 277 : 117-126. Ma LQ, Komar KM, Tu C, Zhang W, Cai Y, Kennelley E D. 2001. A fern that hyperaccumulates arsenic. Nature 409 :579. Manning BA, Goldberg S. 1997. Arsenic(II I) and arsenic(V) adsorption on three California soils. Soil Science 162 :886-895. Marschner H. 2003. Mineral Nutrition of Higher Plants . New York, NY, USA: Academic Press, 889 p. Martens SN, Boyd RS. 2002. The defensive ro le of Ni hyperaccumulation by plants: a field experiment. American Journal of Botany 89 : 998-1003. Matschullat J. 2000. Arsenic in the geosphere-a review. Science of the Total Environment 249 : 297– 312. McBride MB. 1994. Environmental Chemistry of Soils . New York, NY, USA: Oxford University Press, 406p. McGrath SP. 1998. Phytoextraction for soil remediation. In : Brooks RR, ed. Plants that Hyperaccumulate Heavy Metals . Wallingford: CAB International, 261. McGrath SP, Zhao FJ. 2003. Phytoextraction of metals and metalloids from contaminated soils. Current Opinion in Biotechnology 14 : 277.

PAGE 138

124 McGrath SP, Zhao FJ, Dunham SJ. 2000. Long-te rm changes in the extractability and bioavailability of zinc and cadmium after sludge application. Journal of the Environmental Quality 29 : 875-883. McGrath SP, Zhao FJ, Lombi E. 2002 . Phytoremediation of metals, metalloids, and radionuclides. Advances in Agronomy 75 : 1. McLaren RG, Naidu R, Smith J, Tiller KG. 1998. Fractionation and distribution of arsenic in soils contaminated by cattle dip. Journal of the Environmental Quality 27 : 348. McLean JE, Bledsoe BE. 1992. Behavior of Metals in Soils , Ground Water Issue. U.S.EPA. EPA/540/S-92/018. Meharg AA, Hartley-Whitaker J. 2002. Arsenic uptake and metabolism in arsenicresistant and nonresistant plant species. New Phytologist 154 : 29. Meharg AA, Killham K, 1991. A novel m.thod of quantifying root exudation in presence of soil microflora. Plant and Soil 92 : 111-116 Meharg AA, Macnair MR. 1992. Suppression of the high-affinity phosphate uptake system: a mechanism of arsenate tolerance in Holcus lanatus L. Journal of Experimental Botany 43 : 519. .Mehlich A, 1984. Mehlich-3 soil test extracta nt: a modification of Mehlich-2 extractant. Communications in Soil Science and Plant Analysis 15 : 1409-1416. Mengel DB, Barber SA. 1974. Rate of nutrient uptake per unit of corn root under field conditions. Agronomy Journal 66 : 399-402. Nelson DW, Sommers LE. 1982. Total carbon, or ganic carbon, and organic matter. In: Page AL, Miller RH, Keeney DR, ed. Methods of Soil Analysis , Part 2. 2nd ed. Agronomy Monography 9. AS A. Madison, WI, USA: 539-579. Nelson G. 2000. The ferns of Florida: a reference and field guide . Sarasota, FL, USA: Pineapple Press, 200p. Ohwaki Y, Hirata H, 1992. Differences in carboxylic acid and exudation among P-tarved leguminous crops in relation to carboxylic acid contents in plant tissues and phospholipid level in roots. Soil Science and Plant Nutrition 38 : 235. Olsen SR, Sommers LE. 1982. Phosphorus. In: Page AL, Miller RH, Keeney DR, eds. Methods of Soil Analysis , Part 2. Chemical and Microbiological Properties. American Society of Agronomy & So il Science Society of America. p.403. O’Neil P. 1990. Heavy Metals in Soils . New York, NY, USA: John Wiley and Sons, 8399.

PAGE 139

125 Oremland RS, Stolz JF. 2003. The ecology of arsenic. Science 300 : 939-944. Otte ML, Dekkers MJ, Rozema J.1991. Uptake of arsenic by Aster tripolium in relation to rhizosphere oxidation. Canadian Journal of Botany 69: 2670-2677. Petruzzelli G. 1989. Recycling waste in ag riculture: heavy metal bioavailability. Agriculture, Ecosystems and Environment 27 : 493-503. Philippe H. 1999. Bioavailability of trace elem ents as related to root-induced chemical changes in the rhizosphere. In: Wenzel WW, Adriano DC, Alloway B, Doner HE, Keller C, Lepp NW, Mench M, Naidu R, Pierzyski GM, Eds. Proceedings of Extended Abstracts of 5th International conference on the Biogeochemistry of Trace Elements, Vienna, Austria, 152-153. Pickering IJ, Prince RC, George MJ, Smith RD, George GN, Salt DE. 2000. Reduction and coordination of arse nic in Indian mustard. Plant Physiology 122: 1171. Pierce ML, Moore CB. 1982. Adsorption of ar senite and arsenate on amorphous iron hydroxide. Water Residue 16: 1247. Pinton R, Varanini Z, Nannipieri P, 2001. The Rhizosphere: Biochemistry and Organic Substances at the Soil-plant Interface .New York, NY, USA.Marcel Dekker, Inc. 424 pp. Porter EK, Peterson PJ. 1977. Arsenic tolera nce in grasses growing on mine waste. Environmental Pollution 14 :255-265. Price AH, Taylor A, Ripley SJ, Griffiths A, Trewavas AJ, Knight MR. 1994. Oxidative signals in tobacco increase cytosolic calcium. Plant Cell 6 : 1301. Raab A, Feldmann J, Meharg AA. 2004. The na ture of arsenic-phytochelatin complexes in Holcus lanatus and Pteris cretica . Plant Physiology 134 :1113-1122. Raskin I, Ensley BD. 2000. Phytoremediation of Toxic Metals . New York, NY, USA. John Wiley and Sons. Raskin I, Kumar PBAN, Dushenkov S, Salt DE. 1994. Bioconcentration of heavy metals by plants. Current Opinion in Biotechnology 5 : 285-290. Raskin I, Smith RD, Salt DE. 1997. Phytoremed iation of metals: using plans to remove pollutants from the environment. Current Opinion on Biotechnology 8: 221-226. Reeves RD, Baker AJM. 2000. Metal-accumulating pl ants. In: Raskin I, Ensley BD, eds. Phytoremediation of Toxic Metals: Usi ng Plants to Clean Up the Environment . New York, NY, USA: John Wiley & Sons, 193-229. Reina SV, Esteban E, Goldsbrough P. 2005. Ar senate-induced phytochelatins in white lupin: influence of phosphate status. Physiologia Plantarum 124 : 41.

PAGE 140

126 Ritchie GSP, Sposito G. 1995. Speciation in soils. In: Ure, and Davidson (eds.) Chemical speciation in the environment. New York, NY, USA. Blachie Academic & Professional, 201-233. Robinson BH, Brooks RR, Howes AW, Kirkma n JH, Gregg PEH. 1997. The potential of the high-biomass nickel hyperaccumulator Berkheya coddii for Phytoremediation and phytomining. Journal of Geochemical Exploration 60 :115-126 Rocovich SE, West DA. 1975. Arsenic to lerance in populations of the grass Andropogon scoparius . Science 188 :187-188. Rosen B. 2002. Biochemistry of arsenic detoxification. FEBS Letters 529 :86-92. Rugh CL, Wilde D, Stack NM, Thompson DM, Summers AO, Meagher RB. 1996. Mercuric ion reduction and re sistance in transgenic Arabidopsis thaliana plants expressing a modified bacterial merA gene. Proceedings of th e National Academy of Science 93: 3182/ 3187. Sadiq M. 1997. Arsenic chemis try in soils: an overview of thermodynamic predictions and field observations. Water, Air, and Soil Pollution 93: 117. Sagoe CI, Ando T, Kouno K, Nagaoka T, 1998 Relative importance of protons and solution calcium concentration in phospha te rock dissolution by organic acids. Soil Science and Plant Nutrition 44 : 617Salt DE, Smith RD, Raskin I. 1998. Phytoremediation. Annual Review in Plant Physiology and Plant Molecular Biology 49 :643-668. Salt DE, Kato N, Krmer U, Smith RD, Raskin I 2000. The role of root exudates in nickel hyperaccumulation and tolerance in accumulator and nonaccumulator species of Thlaspi . In: Terry N, Bauelos G, eds. Phytoremediation of Contaminated Soils and Waters . Boca Raton, FL, USA: CRC Press, 171-188. SAS Institute. 1987. SAS User’s Guide: Statistics . 6th ed.; Statistical Analysis Institute, Inc.: Cary, NC. Schat H, Llugany M, Bernhard R. 1999. Metal specific patterns of to lerance, uptake, and transport of heavy metals in hyperaccumulating and non-hyperaccumulating metallophytes. In: Terry N, Bauelos G, eds. Phytoremediation of Contaminated Soils and Waters . Boca Raton, FL, USA: CRC Press, 171-188. Schat H, Llugany M, Vooijs R, Hartley-Whitaker J, Bleeker PM. 2002.The role of phytochelatins in constitutive and ad aptive heavy metal tolerances in hyperaccumulator and non-hyperaccumulator metallophytes. Journal of Experimental Botany 53 : 2381-2392. Schmoger ME, Oven M, Grill E. 2000. Detoxifi cation of arsenic by phytochelatins in plants. Plant Physiology 122 :793-801.

PAGE 141

127 Schutzendubel A, Andrea P. 2002. Plant respon ses to abiotic stresses: heavy metalinduced oxidative stress and protection by mycorrhization. Journal of Experimental Botany 53 :1351-1365. Schwartz C, Morel JL, Saumier S, Whiting SN, Baker AJM. 1999. R oot architecture of the Zn-hyperaccumulator plant Thlaspi caerulescens as affected by metal origin, content and localization in soil. Plant Soil 208 : 103. Scott N, Hatlelid KM, MacKenzie NE, Carter D E. 1993. Reactions of arsenic (III) and arsenic (V) species with Glutathione. Chemistry Residue Toxicology 6: 102-106. Sheppared BS, Caruso JA, Heitkemper DT, Wolnik KA. 1992. Arsenic speciation by ion chromatography with induced couple d plasma mass spectrometry detection. Analyst 117 : 971-975. Smethurst P, Comerford NB. 1993. Simulati ng nutrient uptake by single or competing and contrasting root system. Soil Science Society of the American Journal 57: 1361-1367. Smith E, Naidu R, Alston A M. 1998. Arse nic in the soil environment: A Review. Advances in Agronomy 64 :149-195. Smith AH, Lopipero PA, Bates MN, Stei nmaus CM. 2002. Public health arsenic epidemiology and drinking water standards . Science 296: 2145-2146. Smith SE, Read DJ. 1997. Mycorrhizal symbiosis . San Diego: Academic Press. 605p. Smith WK, Vogelmann TC, Delucia EH, Be ll DT, Shepherd KA. 1997. Leaf form and photosynthesis. BioScience 47 : 785-793. Sneller FEC, Van Heerwaarden LM, Kraaije veld-Smit FJL, Ten Bookum WM, Koevoets PLM, Schat H, Verkleij JAC. 1999. Toxicity of arsenate in Silene vulgaris , accumulation and degradation of arsenate-induced phytochelatins. New Phytologist 144 : 223-232. Srivastava M, Ma LQ, Santos JAS. 2006. Three new arsenic hyperaccumulating ferns. Science of the Total Environment (in press) Available online 20 December Stevenson FJ. 1986. Cycle of soil . New York, NY, USA. John Wiley and Sons. Susarla S, Medina VF, McCutcheon SC. 2002. Phytoremediation: An ecological solution to organic chemical contamination. Ecological Engineering 18 :647-658. Terry N, Zayed AM. 1994. Selenium volatilization by plants. In: Frankenberger Jr WT, Benson S, eds. Selenium in the Evironment . NY, New York, USA. Marcel Dekker, 343-67.

PAGE 142

128 Tessier A, Campbell PGC, Bisson M. 1979. Sequential extraction procedure for the speciation of particulate trace metals. Analytical Chemistry 51 : 844-851. Thanabalasingam P, Pickering WF. 1986. Arsenic sorption by humic acids. Environmental Pollution 12 : 233-246. Thomas GW. 1982. Exchangeable cations. In: Page AL, Miller RH, Keeney DR, eds. Methods of Soil Analysis . Part 2. Chemical and Microbiological Properties. American Society of Agronomy & Soil Science Society of America. 159-165. Trofymow JA, Coleman DC, Cambardella C. 1987. Rates of rhizodeposition and ammonium depletion in the rhiz osphere of axenic oat roots. Plant and Soil 97: 333344. Tu C, Ma L Q. 2002. Effects of arsenic c oncentrations and forms on arsenic uptake by the hyperaccumulator ladder brake. Journal of the Environmental Quality 31 : 641647. Tu C, Ma LQ. 2003. Effects of arsenate and phosphate on their accumulation by an arsenic-hyperaccumulator Pteris vittata L. Plant and Soil 249 : 373-382. Tu C, Ma LQ. 2003c. Comparison of arseni c uptake and distribution in arsenic hyperaccumulator Pteris vittata L. and non-hyperaccumulator Nephrolepis exaltata L. Journal of Plant Nutrition 27 :1227-1242. Tu C, Ma LQ. 2005. Effects of As hyperaccumulation on nutrient content and distribution in fronds of the hyperaccumulator Chinese brake. Environmental Polluttion 135 : 333-340. Tu C, Ma LQ, Bondada B. 2002. Arsenic a ccumulation in the hyperaccumulator Chinese Brake ( Pteris vittata L. ) and its utilization pote ntial for phytoremediation. Journal of the Environmental Quality 31 :1671-1675. Tu C, Ma LQ, Zhang W, Cai Y, Harris WG. 200 3. Arsenic species and leachability in the fronds of the hyperaccumulator Chinese brake ( Pteris vittata L.). Environmental Pollution 124 : 223-230. Tu S, Ma LQ. 2003b. Interactive effect of pH , As and P on growth and As/P uptake in hyperaccumulator Pteris vittata . Environmental and Experimental Botany 50 : 243251. Tu S, Ma L, Luongo T. 2004. Root exudate s and arsenic accumulation in arsenic hyperaccumulating Pteris vittata and non–hyperaccumulating Nephrolepis exaltata. Plant and Soil 258 : 9-19. Turpeinen R, Pantsar-Kallio M, Haggblom M, Kairesalo T. 1999. Influence of microbes on the mobilisation, toxicity and biomethylation of arsenic in soil. Science of the Total Environment 236 :173-180.

PAGE 143

129 Tye AM, Young SD, Crout NMJ, Zhang H, Da vison W, McGrath SP, Paton GI, Kilham K. 2002. Predicting arsenic solubility in c ontaminated soils using isotopic dilution techniques. Environmental Science and Technology 36: 982-988. United States Environmental Protection Agency (USEPA). Arsenic treatment technologies for soil waste and wate r. Report EPA-542-R-02-004. Washington, DC. 2002a. Uren NC. 2000. Types, amounts and possible fu nctions of compounds released into the rhizosphere by soil-grown plants. In: Pinton R, Varanini Z, Nannipiero P, eds. The Rhizosphere: Biochemistry and Organic Substa nces at the Soil-Plant Interface. New York, NY, USA: Marcel Dekker, 19-40. Van Noordwijk M, Brouwer G. 1991. Review of quantitative root length data in agriculture. In: McMichael BL, Persson H, eds. Plant Roots and their Environment . Amsterdam: The Netherlands: Elsevier, 515. Vaughan, GT. The environmental chemistry and fate of arsenical pesticides in cattle tick dip sites and banana land plantations. Au stralia: CSIRO Division of Coal Industry, Center for Advanced Analytical Chemistry, NSW, 1993. Vegh K R. 1991. Effect of soil water and nut rient supply on root characteristics and nutrient uptake of plants. In: Persson BLM, ed. Plant roots and their environment. Amsterdam, The Netherlands: Elsevier. Voigt DE, Brantle SL, Hennet RJC. 1996. Chemi cal fixation of arseni c in contaminated soils. Applied Geochemistry 11 : 633. Waisel Y, Eshel A. 2002. Functional diversity of various constituents of a single root system. In: Waisel Y, Eshel A, kafkafi U, eds. Plant Roots: The Hidden Half . New York, NY, USA:Marcel Dekker Inc. Wang JR, Zhao FJ, Meharg AA, Raab A, Feldmann J, McGrath SP. 2002. Mechanisms of arsenic hyperaccumulation in Pteris vittata. Uptake kinetics, interactions with phosphate, and arsenic speciation. Plant Physiology 130 :1552 -1561. Watanbe ME. 1997. Phytoremediation on the brink of commercialization. Environmental Science and Technology 31: 182-186. Weissenhorn I, Leyval C. 1995. Root coloniza tion of maize by a Cd-sensitive and a Cdtolerant Glomus mosseae and cadmium uptake in sand culture. Plant and Soil 175 : 233-238. Wenzel WW, Bunkowski M, Puschenr eiter O, Horak O. 2003. Rhizosphere characteristics of indigenously growing nickel hyperaccumulator and excluder plants on serpentine soil. Environmental Pollution 123 : 131-138.

PAGE 144

130 Wenzel WW, Kirchbaumer N, Prohaska T, Stingeder G, Lombi E, Adriano DC. 2001. Arsenic fractionation in soils using an improved sequential extraction procedure. Analytica Chimica Acta 436 : 309-323. Whiting SN, Leake JR, McGrath SP, BAKER A JM. 2000. Positive responses to Zn and Cd by roots of the Zn and Cd hype raccumulator Thlaspi caerulescens. New Phytologist 145 : 199-210. Wierzbick M, Panufnik D. 1998. The adaptati on of silene vulgaris to growth on a calamine waste heap (S. Poland). Environmental Pollution 101 : 415-426. Wilcut JW, Wehtje GR, Patterson MG, Cole TA, Hicks TV.1989. Absorption, translocation and metabolism of foliar-app lied chlorimuron in soybeans (Glycine max), peanuts (Arachis hypogaea), and selected weeds. Weed Science 37: 175-180. Woolson EA. 1983. Emissions, cycling, and e ffects of arsenic in soil ecosystems. Amsterdam, The Netherlands: Elsevier, 52-125. Yanai RD.1994. Steady-state model of nutrient up take accounting for newly grown roots, Soil Science Society of American Journal 58 , 1562-1571. Ying Z, Guo LD, Liu RJ. 2004. Arbuscular myco rrhizal fungi associated with common pteridophytes in Dujiangyan, southwest China. Mycorrhiza 14 :25-30. Yang-Wei C, Chen T. 2005. Arsenic accumulation by two brake ferns growing on an arsenic mine and their pote ntial in phytoremediation. Chemosphere, in press, available online 17 November 2005 Youssef RA, Chino M. 1989. Root-induced chan ges in the rhizosphere of plants II. Distribution of heavy metal ac ross the rhizosphere in soil. Soil Science and Plant Nutrition 35 : 609. Zhang FS, Ma J, Cao YP. 1997. Phosphorus defi ciency enhances root exudation of lowmolecular weight organic acids and util ization of sparingl y soluble inorganic phosphates by radish (Raphanus sativus L.) and rape (Brassica napus L.) plants. Plant and Soil 196: 261 -264. Zhang W, Cai Y, Downum KR, Ma LQ. 2004. Thiol synthesis and arsenic hyperaccumulation in Pteris vittata (Chinese brake fern). Environmental Pollution 131 : 337-345. Zhang W, Cai Y, Tu C, Ma LQ. 2002. Arsenic speciation and distribution in an arsenic hyperaccumulating plant. Science of the Total Environment 300 : 167-177. Zhao FJ, Wang JR, Barker JHA, Schat H, Bleeker PM, McGrath SP. 2003.The role of phytochelatins in arsenic tole rance in the hyperaccumulator Pteris vittata . New Phytologist 159 : 403-410.

PAGE 145

131 Zhu B, Alva AK. 1993. Effect of pH on growth and uptake of copper by swingle citrumelo seedlings. Journal of Plant Nutrition 16: 1837-1845. Zhuang J, Nakayama K, Urushisaki T. 2001. Estimation of r oot water uptake of maize: an ecophysiological perspective. Field Crops Research 69: 201-213.

PAGE 146

132 BIOGRAPHICAL SKETCH Maria Isidoria Silva Gonzaga (named after her grandmother) was born on December 14, 1970, in Sao Felipe, Bahia, Brazil, to Mr. Henrique F. Carvalho and Mrs. Ines A. Silva. After receiving her high school diploma in 1989, Maria attended the School of Agronomy at the Federal University of Bahia, Brazil, receiving a degree in agronomy in 1995. During her study, Maria also worked there, in the Soil Physics Laboratory, lecturing and doing soil physical analyses. In 1999, Maria received her Master of Scien ce degree in agrarian sciences from the Federal University of Bahia, Bahia. Her mast er’s thesis was titled “Effects of industrial residues from cellulose production on Calopogonium mucunoids growth and on its soil physical and chemical attributes”. In the Fall 2002, Maria join ed the Soil and Water Science Department at the University of Florida, Gainesville, as a Ph.D. candidate to study the use of ferns to phytoremediate arsenic-contaminated soils.