CHARACTERIZATION OF ARABIDOPSIS HEAT SHOCK PROTEIN 70 (HSP70) GENE FAMILY AND MICROARRAY ANALYSIS OF GENE EXPRESSION IN RESPONSE TO TEMPERATURE EXTREMES By DONG YUL SUNG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIV ERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2001
Copyright 2001 by Dong Yul Sung
To my wonderful wife Gwijun
iv ACKNOWLEDGMENTS I am thankful to my advisor, Dr. Charles Guy, for his mentoring. I am inspired by his principles and his ways of coping with adversities in and out of the lab. His guidance better equipped me as a scientist and a person. I also send ma ny special thanks to Dale Haskell for his friendship. I was always able to turn to him when I had problems in and out of the lab. Because of him, I was able to know God better. I would like to express my sincere thanks to my committee members, Dr. Kennet h Cline, Dr. Robert Cohen, Dr. William Gurley, and Dr. Alice Harmon, for their advice and guidance. I would like to extend my gratitude to Drs. Harry Klee, Curtis L. Hannah, Michael Kane, and Jason Grabosky for providing materials and facilities needed for th e experiments. Finally, I am eternally grateful to my wonderful wife, Gwijun, for supporting me the last seven years. I am also thankful to my two kids, Richard and Audrey, for their inspiring smiles and warm hearts that enabled me to go through many diff iculties. I am also grateful that my father can see the completion my studies. I send my last but very special thanks to my mother and brother in heaven.
v TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. iv LIST OF TABLES ................................ ................................ ................................ ............. vii LIST OF FIGURES ................................ ................................ ................................ .......... viii ABSTRACT ................................ ................................ ................................ ........................ xi CHAPTERS 1. INTRODUCTION ................................ ................................ ................................ .......... 1 2. LITERATURE REVIEW ................................ ................................ ............................... 3 Hsp70 Gene Family ................................ ................................ ................................ 3 Structure and Functions of H sp70 ................................ ................................ ........... 5 Cytosolic Hsp70 Proteins ................................ ................................ ...................... 12 ER Luminal Hsp70 Proteins ................................ ................................ ................. 14 Chloroplast Hsp70 Proteins ................................ ................................ .................. 17 Mitochondria Hsp70 Proteins ................................ ................................ ............... 19 3. COMPREHENSIVE GENOMIC AND EXPRESSION ANALYSIS OF ARABIDOPSIS HSP70 GENE FAMILY ................................ ............................. 21 Materials and Methods ................................ ................................ .......................... 23 Results ................................ ................................ ................................ ................... 27 Discussion ................................ ................................ ................................ ............. 44 4. CREATING OVER/UNDEREXPRESSION LINES OF Hsc70 1 and BiP 2 IN ARABIDOPSIS ................................ ................................ ................................ ..... 52 Materials and Methods ................................ ................................ .......................... 54 Results ................................ ................................ ................................ ................... 62 Discussion ................................ ................................ ................................ ............. 84
vi 5. MICROARRAY ANALYSIS OF GENE EXPRESSION IN RESPONSE TO HIGH AND LOW TEMPERATURE SHOCK ................................ ................................ 93 Materials and Methods ................................ ................................ .......................... 96 Results ................................ ................................ ................................ ................... 98 Discussion ................................ ................................ ................................ ........... 118 6. CONCLUSION ................................ ................................ ................................ ........... 123 LIST OF REFERENCES ................................ ................................ ................................ . 126 BIOGRAPHICAL SKETCH ................................ ................................ ........................... 140
vii LIST OF TABLES Table Page 3 1. Arabidopsis Hsp70 gene family ................................ ................................ .................. 26 3 2. Oligonucleotide primers used in RT PCR ................................ ................................ ... 34 4 1. Length of products for PC R confirmation of the constructs. ................................ ...... 58 4 2. Transformation efficiency of Hsc70 1 and BiP 2 constructs. ................................ ..... 63 4 3. Possible disease progression phenotypes. ................................ ................................ ... 85 5 1. Comparison of expression of selected genes obtained from RT PCR and microarray analysis ................................ ................................ .............................. 100 5 2. Microarray data for Arabidopsis hsp70 genes ................................ ............................. 103 5 3. Microarray results for the proprietary array. ................................ ............................... 106 5 4. Co induced genes in response to heat and cold shock. ................................ ................ 117
viii LIST OF FIGURES Figure Page 2 1. Three dimensional structures of hsp70 protein. (A) 44 kDa N terminal fragment of bovine hsc70. (B) 30 kDa C terminal fragment of E. coli DnaK. ....................... 6 2 2. The cycle of binding and release of a substrate by Hsp70. ................................ ......... 8 3 1. Chromosomal location of hsp70 genes in Arabidopsis. ................................ .............. 29 3 2. Sequence alignment of chloroplast and mitochondrial hsp70 proteins. ...................... 31 3 3. A Non rooted Neighbor Joining analysis using the CLUSTAL program in DNASTAR for protein sequences of the Arabidopsis hsp70 gene family. ......... 32 3 4. RT PCR optimization. ................................ ................................ ................................ . 35 3 5. Response of cytosolic hsp70s to temperature extremes. ................................ ............. 37 3 6. Response of organellar hsp70s to temperature extremes. ................................ ............ 38 3 7. Predicted cis elements in the p romoters of Arabidopsis hsp70 genes. ........................ 40 3 8. Expression of cytosolic hsp70s during seed maturation and germination. .................. 42 3 9. Expression of organellar hsp70s during seed maturation and germina tion. ................ 43 3 10. Expression of cytosolic hsp70s in different organs. ................................ .................. 45 3 11. Expression of organellar hsp70s in different organs. ................................ ................ 46 4 1. Transformation const ructs of Hsc70 1 and BiP 2 gene. ................................ .............. 55 4 2. Location of primers for PCR confirmation of the constructs and transgenic lines. .... 58 4 3. A representative PCR screening for the presence of pHKCOR Bi P 2 (AS) construct in the transgenic plants. ................................ ................................ ........ 63 4 4. Western blot analysis on pHK Hsc70 1 (S) in T 0 transgenic lines. ............................ 64
ix 4 5. Western blot analysis of pHK BiP (AS) in T 0 transgenic lines ................................ ... 65 4 6. Adaptor ligated PCR confirmation of T DNA integration in BiP 2 transgenic lines. 67 4 7. Locations of the BiP 2 transgene insertion in the Arabidopsis genome. ..................... 68 4 8. Specificity of hsp70 antibodies. ................................ ................................ ................... 70 4 9. Protein expression in BiP 2 sense transgenic plants. ................................ .................. 71 4 10. Protein expression in BiP 2 antisense transgenic plants. ................................ .......... 72 4 11. The gel images of RT PCR analysis of BiP 2 transgenic lines. ................................ 74 4 12. Graphical presentation of RT PCR analysis of BiP 2 transgenic lines. .................... 75 4 13. Effect of tunic amycin on hypocotyl elongation of BiP 2 transgenic lines. (A) A representative picture of seedlings grown on MS plates with or without tunicamycin. (B) Graphical presentation of hypocotyl elongation on MS plates with or without tunicamycin. ................................ ................................ ..... 77 4 14. Seed thermotolerance of BiP 2 (S) lines. (A) Seeds germinated on wet filter paper after heat treatment. (B) Seed thermotolerance of BiP 2 (S) lines. ..................... 79 4 15. BiP protein level in imbibed seeds be fore heat treatment. ................................ ........ 80 4 16. Whole plant thermotolerance of BiP 2 transgenic lines. (A) Electrolyte leakage analysis during induction of acquired thermotolerance. (B) Electrolyte leakage analysis of BiP 2 (S) lines. ................................ ................................ ..... 82 4 17. Population growth progression of Pseudomonas in BiP 2 transgenic lines. (A) P. syringae pv. maculicola ES4326. (B ) P. syringae pv. tomato DC3000. ............ 86 5 1. Induction of thermotolerance i n Arabidopsis. A) Induction of acquired thermotolerance. B) Induction of acquired freezing tolerance. .......................... 97 5 2. RT PCR verification of microarray results. ................................ ................................ 101 5 3. Global view of gene expres sion during heat shock (HS). ................................ ........... 105 5 4. Functional distribution of heat induced genes in the array ................................ ......... 108 5 5. Functional distribution of heat repressed genes in the array ................................ ....... 109 5 6. Global view of gene expression during cold shock (CS). ................................ ........... 112 5 7. Functional distribution of cold induced genes in the array ................................ ......... 114 5 8. Functional distribution of cold repress ed genes in the array ................................ ....... 115
x 5 9. Functional distribution of co repression genes in the array. ................................ ....... 119 6 1. Knockout locations for two cytosolic hsc70 genes and three BiP genes. .................... 125
xi 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 CHARACTERIZATION OF ARABIDOPSIS HEAT SHOCK PROTEIN 70 (HSP70) GENE FA MILY AND MICROARRAY ANALYSIS OF GENE EXPRESSION IN RESPONSE TO TEMPERATURE EXTREMES By Dong Yul Sung December 2001 Chairman: Charles L. Guy Major Department: Plant Molecular and Cellular Biology Twelve full length and two truncated DnaK type heat sho ck protein 70 (hsp70) genes have been identified in Arabidopsis since the completion of the genome sequencing project. While the biochemical mechanism and structure of hsp70 are highly conserved among organisms, the physiological roles of hsp70s in plants remain largely unknown due to promiscuous substrate specificity and functional redundancy of individual hsp70s. To characterize the roles of hsp70 genes in Arabidopsis, the following objectives were undertaken: 1) a comprehensive RT PCR and genomic analy sis for the hsp70 gene family, 2) transgenic approach to over/underexpress two members of the hsp70 gene family, cytosolic Hsc70 1 and ER luminal BiP 2 , and 3) microarray analysis of gene expression during induction of acquired thermotolerance. I isolated two nuclear encoded, organellar members of the Arabidopsis hsp70 gene family mtHsc70 2 (AF217458) and
xii cpHsc70 2 (AF217459). Genomic analysis revealed several unique characteristics of the subfamily of Arabidopsis hsp70 gene family such as gene structure, sequence relatedness, C terminal sequence motifs. RT PCR analys is established the complex differential expression pattern for the hsp70s in Arabidopsis that suggests specialized functions even among members localized to the same subcellular compartment. Transgenic analysis showed that the tight regulation of cytosolic Hsc70 1 expression is essential to viability of plants and that increased BiP 2 protein levels enhanced the survival of seed after heat stress. Microarray analysis identified genes that ar e commonly induced or repressed in response to temperature extremes. The expression patterns of the genes in the microarray also suggested that the membrane components and the photosynthetic apparatus are primary targets for temperature stresses.
1 CHAPTER 1 INTRODUCTION Heat shock protein 70 (hsp70) is a molecular chaperone that helps newly synthesized polypeptides to obtain correct conformations as well as assisting stress denatured proteins to assume their original folded conformations. Hsp70 consists o f two major domains: a 44 kDa ATPase domain and a 30 kDa peptide binding domain. The structure and biochemical properties of hsp70 protein have been well characterized in many organisms (Flaherty et al., 1990; Flynn et al., 1991; Wang et al., 1998; Zhu et al., 1996). Hsp70s are involved in nearly every aspect of protein biogenesis during normal growth development of an organism as well as under various stresses (Brodsky, 1996; Bush and Meyer, 1996; Gaitanaris et al., 1990; Gao et al., 1991; Glover and Lin dquist, 1998; Goloubinoff et al., 1999; Lee et al., 1995; Nelson et al., 1992; Morishima et al., 2000; Sheffield et al., 1990). Hsp70 comprises a highly conserved gene family in every organism. Arabidopsis has 14 or more hsp70 genes (Sung et al., 2001). Hsp70s are present in almost every subcellular compartment of the cell. In many cases, multiple members of hsp70 gene family are present in the same subcellular compartment. Despite its well characterized structure and biochemical properties, specific functions of individual hsp70 in cellular metabolism are poorly characterized. This is mainly because of the broad substrate specificity of hsp70 and functional overlap among individual members.
2 The primary objective of this project was to define some of the whole organism biological processes that rely upon involvement and proper function of the hsp70 chaperones. I attempted to characterize the roles of hsp70s in Arabidopsis by taking three major approaches: 1) genomic analysis of hsp70 gene family, 2) g ene expression analysis of hsp70 gene family by RT PCR, and 3) transgenic studies of two members of the Arabidopsis hsp70 gene family. The results indicated that the Arabidopsis hsp70 gene family has several characteristics that are unique to the subfamil y and a complex differential expression pattern for the hsp70s that portends specialized functions even among members localized to the same subcellular compartment. Transgenic analysis showed that the tight regulation of cytosolic Hsc70 1 expression is es sential to viability of plants and that overexpression of BiP 2 gene increased thermotolerance of seed. Currently, I have identified a total of 10 knockout lines for two cytosolic and three BiP genes. More focused and targeted analysis will be feasible with these lines. Together with knockout analysis, the results obtained in this study will be a basis for further studies on hsp70 in Arabidopsis.
3 CHAPTER 2 LITERATURE REVIEW Hsp70s comprise one subset of heat shock proteins that are induced by a rapid increase of temperature. In eukaryotes, hsp70s are encoded by a highly conserved multi gene family whose proteins function in all major subcellular compartme nts of the cell. Numerous studies have elucidated hsp70 chaperone functions under stress conditions and in protein metabolism. Hsp70s bind and release unfolded/non native proteins, thereby help ing polypeptides undergo productive folding in the highly con centrated protein environment of the cell. H sp70s can prevent aggregation of denatured proteins (Sheffield et al., 1990) and refold stress denatured proteins (Gaitanaris et al., 1990; Glover and Lindquist, 1998; Goloubinoff et al., 1999; Lee et al., 1995) . They are also involved in translation (Nelson et al., 1992), translocation processes (Brodsky, 1996; Bush and Meyer, 1996; Gao et al., 1991) and steroid receptor function (Morishima et al., 2000). In addition, cytosolic hsp70s act as negative repressor s of heat shock factor (HSF) mediated transcription either by themselves or in an h sp90 associated multi chaperone complex (Bonner et al., 2000; Shi et al., 1998; Zou et al., 1998). Hsp70 Gene Family Plant hsp70 genes are encoded by a highly conserved mult i gene family. Fourteen genes encode hsp70s in Arabidopsis and there are at least 12 in spinach (Guy and Li, 1998; Sung et al., 2001). Sequence analyses show that hsp70 gene families consist of four major subgroups and each of which shares the same subce llular
4 localization (Boorstein et al., 1994; Guy and Li, 1998; Sung et al., 2001): cytosol, ER, plastids, or mitochondria. In watermelon, an hsp70 encoded by the same gene is localized to glyoxysome as well as plastid (Wimmer et al., 1997). Conservation o f hsp70 genes in plants is astonishing in that the least conserved plant cytosolic member among 23 full length plant hsp70 sequences analyzed shares 75% amino acid identity (data not shown) with the rest of hsp70 sequences. Mitochondrial and plastid membe rs are more similar to the bacterial hsp70 homolog, DnaK, while the ER luminal BiP and hsp70s targeted to other compartments are more closely related to cytosolic hsp70s (Boorstein et al., 1994; Wimmer et al., 1997). A unique feature of the plant hsp70 gen e family is the presence of multiple members of ER luminal BiP. Five genes encode for BiP in tobacco, four in soybean, three in Arabidopsis and at least two in maize (Cascardo et al., 2000; Denecke et al., 1991; Sung et al., 2001) while only a single gene encodes BiP in yeast (Normington et al, 1989). Another feature of plant hsp70s is the distinctive subcellular localization motifs present at the C termini. The C termini of each subgroup are unique and highly conserved within each group and diagnostic f or their subcellular locations. The conserved motif for cytosolic group is EEVD, for ER it is HDEL, for mitochondria it is PEAEYEEAKK and for plastids it is PEGDVIDADFTDSK (Guy and Li, 1998). HDEL motif in BiP is an ER retention signal. The EEVD motif i n cytosolic hsp70 is essential for interaction with hsp70 interacting proteins such as Hip/p48 and p60/Hop (Liu et al., 1999). Deletion of EEVD from a rat hsc70 resulted in a dramatic increase in ATPase activity suggesting a role for EEVD in modulating AT Pase activity of hsc70 (Boice and
5 Hightower, 1997). It is suspected that the conserved C terminal sequence motifs of mitochondria and plastid hsc70 may have similar roles as EEVD. Gene structure of plant hsp70s is also unique within four subgroups of the gene family. Genes encoding cytosolic hsp70s usually have one or no introns while genes encoding organellar hsp70 have multiple introns. The ER group has 4 to 5 introns, the mitochondrial group has 5 and the plastid group has 7 in Arabidopsis, in contras t to 7 introns for the ER group, 4 for the mitochondrial group and 7 for the plastid group in spinach (Guy and Li, 1998; Sung et al, 2001). It is intriguing how organellar hsp70s with presumed prokaryotic origins have more complex intron exon structure th an their cytosolic counterparts. Although gene structures of organellar hsp70s are distinctly different and characteristic of each subgroup, they are not highly conserved among plants. Structure and Functions of Hsp70 An hsp70 protein can be divided int o two major structural domains: a 44 kDa ATPase domain and a 30 kDa peptide binding domain ( Figure 2 1 ). The crystal structures of the N terminal ATPase domain of bovine hsc70 (Flaherty et al., 1990) and that of C terminal peptide binding domain of DnaK f rom E. coli are now available (Zhu et al., 1996). Hsc70 (heat shock cognate 70) designates the members in the hsp70 gene family that are expressed in normal conditions as well as under stress. The N terminal ATPase domain contains two structural lobes th at form a deep cleft where the binding site for ATP is located (Zhu et al., 1996). The C terminal peptide binding domain contains a ? strand and five ? helical segments. The first two of the five ? helical segments serve as a lid on top of the peptide bi nding site formed by the ? strand
6 A B A B A B A B Figure 2 1. Three dimensional structures of hsp70 protein. (A) 44 kDa N terminal fragment of bovine hsc70. (B) 30 kDa C terminal fragment of E. coli DnaK. Three dimensional structures were adapted from NCBI protei n structure database. The 44 kDa ATPase fragment (PDB ID: 3HSC) is shown with ATP in the nucleotide binding cleft. The 30 kDa C terminal peptide binding fragment (PDB ID: 1DKX) is shown with a substrate peptide in the peptide binding cleft (Arrow). Thre e dimentional structures were viewed with Cn3D 3.0.
7 domain (Zhu et al., 1996). A domain swapping experiment with chimeric proteins containing all combinations of the ATPase, peptide binding, and variable domains of yeast cytosolic hsp70 proteins, Ssa1 a nd Ssb1, proved that the functional differences between Ssa1 and Ssb1 are not determined by either peptide binding or variable domains (James et al., 1997). This suggested that an additional factor(s) is required to confer functional specificity to hsp70. A NMR study revealed that the C terminal tail of DnaK binds to its own peptide binding site (Wang et al., 1998). The ? helical segment bound to peptide binding site was forced to an extended structure reflecting interaction of DnaK with its putative sub strate. The role of the C terminal tail associated with the peptide binding site is not yet known, but the C terminus is implicated in modulation of substrate binding affinity of DnaK (Wang et al., 1998). The binding and release of substrate polypeptide by hsp70 are well demonstrated in the binding cycles of DnaK with its co chaperones, DnaJ and GrpE ( Figure 2 2 ). A substrate polypeptide brought by DnaJ is bound in the narrow cavity of the ? strand peptide binding domain of DnaK ATP. Only selected poly peptides can be bound in this region. The best candidate substrate for this region is an extended heptameric polypeptide enriched with aliphatic amino acids (Flynn et al., 1991). This hydrophobic and extended structure of the candidate peptide is very si milar to those of unfolded regions of denatured proteins. Hydrolysis of ATP causes conformational change of DnaK that shifts the open state of the ? helical lid segment to a closed state. Transition from the open to the closed state of the ? helical lid in the peptide domain of DnaK stabilizes the substrate binding to DnaK, which prevents early release and unproductive folding of substrate polypeptide. Another DnaK interacting co chaperone, GrpE, then
8 ATP ADP Pi ADP, GrpE GrpE ATP ATP ATP DnaJ DnaJ ADP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP HIP BAG 1 ADP, BAG 1 CHIP Stable ATP ATP ATP bound DnaK or hsc70 Peptide substrate for DnaK or hsc70 Figure 2 2. The cycle of binding and release o f a substrate by Hsp70. This figure was modified from the figure published by Rassow et al. (1997). The solid lines indicate the cycle in E. coli and the dotted lines indicate the cycle specific to eukaryotic system. In eukaryotic system, no GrpE homolo g is found in the cytosol. Instead, Ydj1 carries out the combined function of DnaJ and GrpE in yeast. Hsp70 interacting protein (HIP) stabilizes ADP bound hsp70 and BAG 1 facilitates the nucleotide exchange in mammalian system. BAG 1 also acts as a nega tive regulator of hsp70 chaperone activity by competing HIP binding to hsp70. C terminus Hsc70 interacting protein (CHIP) negatively regulates hsp70 activity by inhibiting hsp40 stimulated ATPase activity.
9 interacts with the ATPase domain of DnaK and e xchanges ADP with ATP. The binding of ATP causes a conformational change that leads to the opening of the C terminal ? helical lid region and thereby facilitating the release of substrate polypeptide from peptide binding domain. A new cycle of binding an d release ensues by recruitment of a new substrate polypeptide by DnaJ ( Figure 2 2 ). In the binding and release cycle of mammalian hsp70, cytosolic hsc70s interact with a tetratricopeptide repeat protein known as Hip. Hip is an oligomeric protein that b inds to the ATP binding domain of Hsc70 (Hohfeld et al., 1995) through interaction with its tetratricopeptide repeats (Irmer and Hohfeld, 1997). Hip acts to stabilize the ADP state of Hsc70, maintain high peptide affinity and is an activator of folding ac tivity (Luders et al., 1998). The Arabidopsis genome contains two sequences (BAB02710 and CAA16552) that encode possible Hip homologs. BAG 1 is a negative regulator of Hsc70 (Bimston et al., 1998; Shinichi et al., 1999) that interacts with the ATPase dom ain of Hsc70s through a 45 amino acid region (BAG domain) near its C terminus with high affinity. Its interaction is with the ATPase domain of Hsc70, which stimulates ATP hydrolysis and accelerates ADP release similar to GrpE with DnaK (Hohfeld and Jentsc h, 1997). In fact, based on the crystal structure solution of the Bag domain complex with the ATPase domain of Hsc70 from bovine, it is concluded that Bag1 performs the same nucleotide exchange function as GrpE even though it is structurally different. T his appears to have occurred through convergent evolutionary processes (Sondermann et al., 2001). BAG 1 also inhibits chaperone activity, and acts as a competitive antagonist to the binding of Hip. The Arabidopsis genome contains one sequence (BAB11054) that encodes a possible BAG 1 homolog. BAG 1 contains a ubiquitin like domain and
10 interacts with 26S proteasom (Luders et al., 2000). Another tetratricopeptide repeat containing Hsc70 regulator has been recently identified in mammalian systems (Ballinger et al., 1999). This protein known as CHIP for C terminus Hsc70 interacting protein, is present in the cytosol and acts as a negative regulator of Hsc70 by inhibiting Hsp40 (DnaJ) stimulated ATPase activity of Hsc70 peptide binding and refolding of non na tive substrate. The Arabidopsis genome contains one sequence (AAF02162.1) that may encode a CHIP homolog. CHIP is also known as a E3 ubiquitin ligase (Jiang et al., 2001). BAG 1 and CHIP protein may guide misfolded proteins for degradation when producti ve folding is difficult to achieve (Demand et al., 2001). Hsp70 proteins also participate in a multi chaperone complex in yeast (Ziegelhoffer et al., 1995). However, unlike the DnaK chaperone complex, there is no known GrpE homolog in yeast. Instead, Ydj 1 (a yeast DnaJ homolog) carries out the combined function of DnaJ and GrpE. Ydj1 not only binds to unfolded proteins but also stimulates the ATPase activity of Ssa1p and accelerates the hydrolysis of ATP and the release of ADP from Ssa1p (Ziegelhoffer et al., 1995). The mammalian multi chaperone complex containing hsp70 is more complex than that of E. coli and yeast systems. Hsc70 directs co translational folding of nascent polypeptides with the help of subsequent chaperoning from TriC (a mammalian homo log of GroEL). Mammalian hsp70 also participates in the multi chaperone complex that contains hsp90, Hip (hsc70 interacting protein), Hop (hsp70/90 organizing protein), peptidylprolyl isomerases (PPI), p23, and BAG 1. The multi chaperone complex carries out chaperone functions in the conformational change of the progesterone receptor via intricate and sequential
11 interactions of chaperones in the complex (Freeman and Morimoto, 1996; Schwarz and Grossman 1998). Much evidence shows that plants also have hs p70 containing multi chaperone complexes. For example, reactivation of heat denatured luciferase by hsp70 in conjunction with pea hsp18.1 and a DnaJ homolog indicates the presence of multi chaperone complexes in plants (Lee and Vierling, 2000). Furthermo re, activation of the glucocorticoid promoter by dexamethasone in transgenic plants indicates that plants may have a multi chaperone complex that activates a plant progesterone receptor (McNellis et al., 1998; Nara et al., 2000). Hsp70 proteins fulfill t hree major roles as a molecular chaperone. First, hsp70 proteins bind to nascent polypeptides emerging from the ribosome and prevent aggregation until the synthesis of the polypeptide reaches the point for proper folding. The SSA subfamily of yeast cytos olic hsp70 is involved in initiation of translation. A member of the SSA subfamily, Ssa1, interacts with a hsp40 homolog (Sis1) and a poly A binding protein (Pab1) in translating ribosomes (Horton et al., 2001) and the SSB subfamily is involved more in el ongation (Nelson et al., 1992). Second, hsp70 proteins help the folding of substrate proteins and/or maintain loose conformations of precursor proteins during the translocation process. The SSA subfamily participates in this process (Bush and Meyer, 1996 ; Kim et al., 1998). Third, hsp70 proteins regulate HSF activity either by forming a stable complex with HSF (Bonner et al., 2000) or by modulating HSF conformation after heat induction (Shi et al., 1998). The SSA subfamily performs the latter function b y refolding activated HSF after heat shock while the SSB subfamily
12 fulfils the former by forming a stable and ATP sensitive complex independent of heat shock (Bonner et al., 2000). However, yeast cytosolic hsp70s appear to have functional redundancy in tra nslocation of precursor proteins. Translocation of precursor proteins into ER and mitochondria was blocked in ssa1ssa2ssa4 mutant, but was normal in ssa1SSA2SSA4 and ssa1ssa2SSA4 (Deshaies et al., 1988). SSA4 is heat inducible and not expressed under nor mal conditions indicating SSA4 is replacing missing functions of SSA1 and SSA2 in these mutants (Deshaies et al., 1988). It also appears that there is no specificity among the SSA members on targeting precursor proteins to respective organelles. The memb ers of the SSA subfamily perhaps provide a common pool of cytosolic Hsp70 proteins to bind precursor proteins. Arabidopsis has approximately 80 J domain containging proteins (personal comm. Jan Miernk). The members of the DnaJ family are more likely cand idates that confer specificity on targeting. Cytosolic Hsp70 Proteins In Arabidopsis, five genes for cytosolic Hsp70s have been identified so far. They are Hsc70 1, Hsc70 2, Hsc70 3, Hsp70 and Hsp70b . Hsc70 1 and other members of cytosolic hsp70s in Ar abidopsis are most like the SSA subfamily in yeast. There is no apparent SSB type of hsp70 known in plants. Hsc70 1 and Hsc70 2 are tandemly located 1.5 kb apart on chromosome 5 (Wu et al., 1988). Northern blot analysis showed only Hsc70 1 is present at a significant level during both heat shock and normal conditions in root, leaf, stem, flower, but at lower levels in green and yellow siliques (Wu et al., 1988; Wu et al., 1994). The level of mRNA of Hsc70 1 was increased five fold after heat shock at 37 ? C for 2 hours. The transcript for Hsc70 2 was not detected at all and the
13 transcript for Hsc70 3 was not detected before heat shock but detected at a very low level after heat shock (Wu et al., 1988). In yeast, over expression of SSA1 or SSA4 showed n o effect on thermotolerance, however, ssa1 or ssa2 single mutants showed slow growth at 30C (Craig and Jacobsen, 1984). Single mutants for ssb1 or ssb2 showed slow growth at low temperature (Craig and Jacobsen, 1985). These results suggest specific role s for the SSA and the SSB subfamily at different temperature environments. A double mutant of ssa1ssa2 displayed: 1), increased transcription of a yeast DnaJ homolog SIS1 (Zhang et al., 1996), 2), increased expression of other heat shock proteins (Hallada y and Craig, 1995), 3), poor growth (Halladay and Craig 1995), and 4), deficiency in luciferase folding (Unno et al., 1997). The double mutant, ssa3ssa4, and the triple mutant, ssa1ssa2ssa4, were found to be lethal in yeast (Werner Washburne et al., 1987) . In Drosophila , over expression of a hsp70 protein resulted in enhanced thermotolerance, but was also shown to be cytotoxic for the growth and development of larvae (Feder et al., 1992; Feder et al., 1996). Over expression of a rat hsp70 (Hsp70c) confer red enhanced tolerance to high temperature, H 2 O 2 and reactive oxygen species (Chong et al., 1998). Knockout mutants of these genes in Arabidopsis are not available yet. However, when Lee and Schffl (1996) tried to block the expression of hsp70s in Arabid opsis upon heat shock by introducing anti sense mRNA of a tobacco hsp70 with a soybean heat shock promoter, the transgenic plants showed reduction in protein abundance of both hsc70 and hsp70 pools. Consequently, the transgenic plants lost induced thermot olerance and showed prolonged auto repression of heat shock response (Lee and Schffl, 1996).
14 ER Luminal Hsp70 Proteins Hsp70 proteins located in endoplasmic reticulum (ER) are also known as binding proteins (BiPs). BiP is found associated with nascent polypeptides and involved in translocation and folding of newly synthesized proteins in yeast (Sanders et al., 1992; Simons et al., 1995; Vogel et al., 1990). An apparent role in secretory biogenesis was confirmed by accumulation of secretory precursor pr oteins when a temperature sensitive mutant of KAR2 gene was incubated at a non permissive temperature (Nguyen et al., 1991; Vogel et al., 1990). Yeast BiP (Kar2p) interacts with the Sec complex on the lumen side of ER for translocation of newly synthesize d proteins from the cytosol. BiP and ATP are required for the release of a precursor protein from the Sec complex on the cytosolic face of the ER (Lyman and Schekman, 1997). Kar2p binds to the J domain like peptide loop of Sec63p in the presence of ATP, which allows the secretory precursor protein to pass through a membrane channel formed by Sec61p into the ER lumen (Lyman and Schekman, 1996). Not every secretory protein interacts with Kar2p. Over expression of a BiP defective in ATPase activity prevent ed secretion of a coagulation factor, factor VIII, while the secretion of monocyte/macrophage colony stimulating factor was not affected (Hendershot et al., 1996). In mammalian cells, the phenotypes of BiP mutants are complex. Grp78 antisense mutants we re less tolerant to cytotoxic compounds (Liu et al., 1997). Over expression of defective BiPs that do not release the substrate proteins resulted in reduced secretion of proteins (Hendershot et al., 1995; Hendershot et al., 1996; Morris et al., 1997) and showed ER disruption (Hendershot et al., 1995). Contrary to these observations, secretion of a tissue plasminogen activator (tPA) variant protein actually increased when functional hamster BiP was underexpressed in Chinese hamster ovary
15 cells (Dorner et a l., 1988). These results suggest that the association with BiP is the rate limiting step for secretion of certain proteins, but not every protein. Plant BiPs are presumed to function in a similar way to that of other eukaryotic BiPs since a tobacco BiP partially complemented a yeast temperature sensitive kar2 mutant (Denecke et al., 1991). However, there are still significant differences between plant BiPs and other eukaryotic BiPs since multiple members of BiP proteins are present in many plants (Denec ke et al., 1991; Kalinski et al., 1995; Wrobel et al., 1997). It still warrants further study to determine whether there is functional specificity or redundancy among the members of the BiP family in plants. The expression of plant BiP is induced by vari ous stresses as well as during developmental processes. Expression of BiP mRNA increases during leaf expansion and seed maturation in soybean (Kalinski et al., 1995 ). Similarly, the amount of BiP protein is increased when storage proteins are accumulated in pumpkin cotyledons (Hatano et al., 1997). Degradation of storage proteins in pumpkin cotyledon was also accompanied by a rapid increase of BiP protein (Hatano et al., 1997). The expression of BiP was in duced when maize cells were treated with protein unfolding agents, azetidine 2 carboxylic acid and tunicamycin (Wrobel et al., 1997). In addition to the analysis of BiP mRNA expression, the protein analysis of BiP also indicates that BiP is involved in t he folding of seed storage proteins and formation of protein bodies in plants. BiP has been shown to associate with nascent polypeptides of prolamin in rice (Li et al., 1993). BiP has also been shown to bind to full length polypeptide of ? conglycinin in soybean by reciprocal co immunoprecipitation experiment using antibody against either hsp70 or ? conglycin (Gillikin et al., 1995). From bean cotyledon, BiP was precipitated by anti phaseolin antibody after tunicamycin
16 treatment. (D'Amico et al., 1992). Similarly, it was shown that BiP formed a complex with an assembly defective phaseolin in transgenic tobacco plant (Pedrazzini et al., 1994). Further, it was shown that BiP associates with the monomer of phaseolin, but not with assembed homotrimeric phas eolin (Vitale et al., 1995 ). In maize, BiP was associated with the ER derived protein body in endosperm mutants (Boston et al., 1991; Fontes et al., 1991; Marocco et al., 1 991). When maize storage proteins ( ? zeins) are expressed in transgenic Arabidopsis plants, they are co localized with BiP in reticular/amorphous structures in cells (Geli et al., 1994). Three genomic clones for BiP were sequenced in Arabidopsis ( BiP 1 (D 89341), BiP 2 (D89342), and BiP 3 (AC000106)), but only one full length cDNA clone corresponding to BiP 2 has been isolated (D84414). It is not clear whether these three proteins are all functional in Arabidopsis. There is 98% identity at the amino acid level between BiP 1 and BiP 2 protein, but the amino acid identity drops to 76% between these two BiPs and BiP 3 protein. BiP 1 and BiP 2 may carry out a similar function in Arabidopsis and may even functionally replace each other while BiP 3 may carry ou t a distinct functions from those of BiP 1 and BiP 2. The expression analysis shows that heat shock and protein denaturing agents such as tunicamycin and castanospermin induce the expression of Arabidopsis BiP gene(s) (Koizumi, 1996), but it is not clear which of three BiPs were studied since a non specific probe was used in the Northern blot analysis. Recently, over expression of tobacco BiP was successfully carried out in tobacco and over expression of BiP alleviates ER stress imposed by treatment of tu nicamycin, a protein glycosylation inhibitor (Alvim et al., 2001; Leborgne Castel et al., 1999). During tunicamycin tr eatment, the synthesis of a amylase was significantly reduced in wild type
17 tobacco plants. BiP over expression in tobacco plants reversed the negative effect of tunicamycin treatment on a amylase synthesis (Leborgne Castel et al., 1999). In response to wo unding or exposed elicitation, several of the molecular chaperones of the endoplasmic reticulum are strongly induced showing elevated mRNA levels. Among these, BiP is particularly responsive to wounding (Guy and Li, 1998; Kalinski et al., 1995 ), salicylic acid, culture filtrate elicitors (Denecke et al., 1995), or plant cell wall degrading enzymes (Jelitto Van Dooren et al., 1999). This molecular chaperone response is likely to be a consequence of the host response that results in the elaboration of defense related genes, some of whose products are secreted from the cell. Chloroplast Hsp70 Proteins Hsp70s in chloroplast play roles in translocation of numerous nuclear encoded precursor proteins. Several chloroplast hsp70s were initially identified during the process of isolating components of protein import machinery. Cross linking with radiolabelled precurso r proteins and immuno detection with Hsp70 antibody revealed the presence of Hsp70 proteins on the envelope membranes (Kourtz and Ko 1997; Marshall et al., 1990; Schnell et al., 1994) and in the stroma (Marshall et al., 1990). An hsp70 protein embedded in the outer envelope membrane was identified with anti hsp70 antibodies (Marshall et al., 1990; Schnell et al., 1994). This hsp70 was also identified as a component of the import complex and immunogenically different from cytosolic hsp70 in pea (Schnell et al., 1994). The presence of a stromal hsp70 in pea also was detected by using tomato hsp70 antibody (Marshall et al., 1990). In spinach, a chloroplast outer envelope hsp70 ( Com70/SCE70 ) similar to cytosolic hsp70 was isolated and sequenced (Ko et al., 1 992). Com70/SCE70 was localized in close physical proximity with
18 precursor proteins on the outer envelope membrane (Kourtz and Ko, 1997). Cross linking of Com70 with arrested precursor proteins by using anti Com70 antibody suggests the involvement of Com 70 in early stages of import of precursor proteins into chloroplast (Kourtz and Ko, 1997). Recently, it was shown that a chloroplast precursor protein forms a protein complex with a cytosolic hsp70 and a dimeric 14 3 3 protein. The precursor in the compl ex was imported three to four times better than a free precursor (May and Soll, 2000). Sequence analyses of the transit peptides of chloroplast precursors also corroborate a role of hsp70 in protein transport in chloroplast by localizing hsp70 binding sit es in the transit peptides (Ivey et al., 2000; Rial et al., 2000). The vigorous metabolic activities in chloroplast probably require hsp70s for proper maintenance of the photosynthetic apparatus that is damaged during photosynthesis as well as under stre ss conditions. The induction of stromal hsp70s in spinach ( Chsp70 ) by high temperature and in Chlamydomonas reinharditii ( Hsp70B ) by light and high temperature is concordant with the above hypothesis (Drzymalla et al., 1996; Wang et al., 1993). The prese nce of DnaJ and GrpE homologues in pea chloroplast (Schlicher and Soll, 1997) and the fact that Chsp70 is more similar to bacterial DnaK than cytosolic homologues (Wang et al., 1993) suggest that stromal hsp70s carry out their functions as molecular chaper ones in concert with plant homologues of DnaJ and GrpE . Hsp70 is also required for proper packaging of storage proteins in maize amyloplast (Yu et al., 1998). The abundance of chaperonin 60 (Cpn60; alpha and beta) and heat shock protein 70 (Hsp70) greatl y increased during chromoplast differentiation (Bonk et al., 1996). This plastid hsp70 is a functionally active form and associated with a soluble entity containing phytoene desaturase suggesting an active role
19 of hsp70 in the conversion of chloroplast to chromoplast (Bonk et al., 1996). Overexpression of a chloroplast localized hsp70 (HSP70B) in Chlamydomonas reinhardtii enhances the reactivation of photosystem II after photoinhibition while underexpression of HSP70B resulted in less reactivation of PSII after photoinhibition (Schroda et al., 1999). The cells overexpressing HSP70B recover much faster after photoinhibition than wild type cells. Photoinhibition partly results from damage in the components of PSII such as D1, CP43, OEE2 and OEE3 protein. The data in the study indicate that HSP70B helps reactivation of these proteins after photoinhibition either by directly interacting with unfolded domains of these proteins or by facilitating import of chloroplast proteins required for repairing these prot eins. To date, two stromal hsp70s have been sequenced in Arabidopsis (AL078637 and AF217459). Mitochondria Hsp70 Proteins Initiation of the import process of precursor proteins into the matrix of mitochondria requires matrix ATP and cytosolic hsp70 prote in (Stuart et al., 1994). Matrix hsp70 protein mediates import, folding, assembly and degradation of mitochondrial proteins (Stuart et al., 1994). Yeast matrix hsp70 protein interacts with a component of the inner membrane translocation complex (for revi ew, see Neupert, 1997). It is shown that approximately 10% of total matrix hsp70 protein associates with Tim44 in an ATP dependent manner. Matrix hsp70 utilizes ATP to drive precursor proteins into the matrix either by a molecular ratchet (Neupert et al. , 1990) or a translocation motor mechanism (Glick, 1995). A yeast mitochondria GrpE homolog, Yge1p/Mge1p, is localized in the matrix (Nakai et al., 1994). Yge1p/Mge1p interacts with a matrix hsp70 and facilitates the
20 release of nucleotide from the hsp70 (Miao et al., 1997). Mge1p also facilitates formation of hsp70 Tim44 complex (Schneider et al., 1996). Mge1p can functionally substitute for GrpE in various biological functions such as complex formation with DnaK , stimulation of ATPase activity of DnaK , and refolding of denatured luciferase (Deloche and Georgopoulos, 1996). Two mitochondrial DnaJ homologs, Mdj1p and Mdj2p, have been identified in yeast, and this suggests the presence of a DnaK like chaperone complex in mitochondria (Prip Buus et al., 19 96; Westermann and Neupert 1997). However, Mdj1P was not found in the translocation machinery of mitochondria. A mitochondria encoded protein, var 1, was shown to aggregate more in Mdj1p mutants (Westermann et al., 1996). This study indicates Mdj1P part icipates in folding of proteins but probably not in the import process. Single mutants of either Mdj1p or Mdj2p did not show growth defects at high temperature, but a double mutant showed severe growth defect at high temperature suggesting considerable ov erlap in functions of the two DnaJ homologs in yeast mitochondria (Westermann and Neupert, 1997). In addition to the matrix hsp70, another hsp70 was localized on the outer membrane of mitochondria (Mooney and Harmey, 1996). Like its chloroplast counterpa rt Com70 , this mitochondrial hsp70 is more similar to cytosolic hsp70 than other mitochondria matrix hsp70s. This novel mitochondria hsp70 is presumably involved in recruiting precursor proteins to the import complex on the outer membrane of mitochondria.
21 CHAPTER 3 COMPREHENSIVE GENOMIC AND EXPRESSION ANALYSIS OF ARABIDOPSIS HSP70 GENE FAMILY Partial genomic sequences for three cytosolic members of the Arabidopsis hsp70 gene family were first described more than 10 years ago (Wu et al., 1988). Since then, several additional hsp70 sequences have been added to the gene database (http://www.ncbi.nlm.nih.gov/Genbank/). With the completion of genome sequencing, twelve full length Arabidopsis hsp70 sequences are available in the database, five genes encoding cytosolic p roteins, three encoding ER luminal members, and two each for plastid or mitochondrion localized proteins. Although each member of the hsp70 gene family shares a highly conserved structure and action mechanism, there is accumulating evidence that various m embers of the hsp70 family play distinct roles in growth and development of plants. First, they are targeted to various subcellular compartments where vastly different metabolic processes occur. Second, sequence analysis classifies hsp70s into subfamilie s that may be linked to different functions. Third, expression profiles of individual members of the hsp70 family differ under various conditions and stimuli. Important questions yet to be resolved include how different functions are allocated to each me mber and to what extent members of the family within a single subcellular compartment are functionally distinct and/or redundant. In plants, comprehensive expression analysis of hsp70s has been limited (Denecke et al., 1991; DeRocher and Vierling, 1995; D uck et al., 1989; Dudley et al., 1997; Li et al., 1999; Wang and Lin, 1993) because the entire complement of hsp70
22 genes has not been available or known for any plant. Expression of only three Arabidopsis cytosolic hsp70s has been examined while most rece nt sequence information indicates that Arabidopsis has five cytosolic hsp70s, (Wu et al., 1988; Wu et al., 1994). At Hsc70 1/Hsp70 1 was shown to be expressed in leaves at normal temperature and further induced by heat shock. The mRNA for Hsp70 2, whose corresponding gene is located 1.5 kb downstream from At Hsc70 1/Hsp70 1, was not detected at normal temperatures or during heat shock. At Hsc70 3/Hsp70 3 mRNA was found to be present at very low levels, and showed no induction by heat shock. In addition, At Hsc70 1/Hsp70 1 was also highly expressed at normal temperatures in root, stem and flower, but not detected in green or yellow siliques (Wu et al., 1994). RT PCR is a powerful method for expression analysis of gene families because amplification from mR NAs can be highly specific and quantification of expression signals can be rapidly performed (McDowell et al., 1996; Wang et al., 1999). In order to better define the physiological roles of hsp70s in plant growth and development, mRNA levels for 11 Arabid opsis hsp70 genes were quantified using RT PCR. The data reveal that several members of the Arabidopsis hsp70 family show distinct expression patterns, allowing predictions of when and where function of each hsp70 is expected to become physiologically imp ortant. These data are necessary to devise experimental strategies to assess phenotypes of loss of function mutants and transgenic plants that over/underexpress individual hsp70s.
23 Materials and Methods Plant Growth and Harvest Arabidopsis seed (ecotype Co lumbia) were sown on water soaked Whatman filter paper (No. 1) for germination experiments. For other experiments, plants were grown in a commercial soil mix (Fafard mix No. 2) containing Canadian sphagnum peat, perlite and vermiculite and watered every third day and fertilized once a week with a commercial fertilizer (Peter's 20 20 20). Plants were grown at 20C with a photoperiod of 15 hr light/9 hr dark in growth cabinets. The irradiance was approximately 150 molm 2 s 1 at canopy height and was pro vided by incandescent bulbs and cool white fluorescent tubes. Samples were harvested and flash frozen in liquid nitrogen and stored at â€“ 80C until RNA extraction. Isolation of cDNA Clones for Arabidopsis Hsp70s A full length cDNA clone for a plastid hsp70 ( cpHsc70 2 ) was isolated from Arabidopsis cDNA libraries (obtained from Arabidopsis Biological Resource Center, Ohio) by PCR amplification. In these libraries, Arabidopsis cDNAs were size fractionated (1 2 kb, 2 3 kb, 3 6 kb) and inserted at the EcoR 1 si te in Lambda Zap II phagemid. A forward primer (CG202; 5' CCCAGTCACGACGTTGTAAAA 3') was generated for the Lambda Zap II phagemid vector. A reverse primer (CG205; 5' GCTGCCAACAAATCA CATTA 3') was generated from a partial EST clone (T43623) of a plastid hs p70 that was identified by a conserved C terminal motif. Amplified cDNAs were cloned into pCR2.1 vector (Invitrogen). The cDNA was sequenced in both directions and confirmed in its entirety. Additionally, I employed a PCR strategy to obtain genomic frag ments of all related genes to identify an Arabidopsis homolog of
24 plant mitochondria hsp70. Primers were designed to encode sequences with exact identity between E. coli DnaK (K01298) and the Sacchromyces cerevisiae mitochondria hsp70, SSC1 (M27229). The forward primer was derived from NGDAWV (aa 98 to 103) of DnaK and the reverse primer from EAAEKA (aa 264 to 269). Primers were fully redundant and had additional 5â€™ restriction sites. PCR was performed using Arabidopsis (ecotype Columbia) genomic DNA. T wo distinct 586 bp genomic fragments having ends with an exact match to the primers were amplified. Sequencing revealed both fragments encoded 166 amino acids, interrupted at amino acid 114 by an 88 bp intron having both consensus donor and acceptor splic e sites. With these genomic fragments, I screened a cDNA library prepared from Arabidopsis heat shock RNA (Helm and Vierling, 1989), and obtained a full length cDNA clone of a mitochondria hsp70 ( mtHsc70 2 ). A detailed screening method was described prev iously (Schirmer et al., 1994). RNA Isolation and RT PCR Organ specific expression of hsp70s was analyzed for roots, stems, leaves, flowers (0 day after pollination: DAP), siliques at 3 DAP from 4 week old plants, siliques at 7 DAP from 5 week old plants and siliques at 14 DAP from 6 week old plants. For changes in hsp70 expression during germination, intact seedlings at 0 to 96 hours after imbibition were collected in liquid nitrogen and stored at 80C. Hsp70 expression in response to temperature extr emes was also examined in plants that were exposed to 4C for 12 hours and 48 hours, 40C for 30, 60, and 90 minutes. Control plants were kept at 20C. Temperature treatment was initiated 2 hours after the onset of the light period in order to harvest al l the samples within the light period. Samples were ground in liquid nitrogen and total RNA isolated according to the manufacturerâ€™s protocol using Trizol ? (Gibco BRL). The amount of total RNA was
25 determined by UV spectrophotometry. Total RNA (1 g) was treated with one unit of DNase I (Sigma) for 15 minutes at room temperature prior to RT PCR to remove residual DNA contamination. Using commercial RT PCR beads (Amersham Pharmacia Biotech), aliquots of total RNA were reverse transcribed into cDNA with ran dom primer, d(N) 6, then amplified with gene specific primers ( Table 3 1 ) in the same tube. When resuspended in 25 l, each RT PCR bead generated a reaction solution containing 2.0 units of Taq DNA polymerase, 10 mM Tris HCl (pH 9.0), 60 mM KCl, 1.5 mM MgC l 2 , 200 M of each dNTP and 1 unit of Moloney murine leukemia virus reverse transcriptase. The cDNAs produced by reverse transcription were amplified with a pair of gene specific primers (10 pmoles for each primer) for each gene. For each RT PCR reaction , a plant 18S rRNA internal standard (Ambion Inc.) was included as a loading control. With this standard, a pair of 18S rRNA specific primers and a pair of competitive primers were mixed at the ratio of 2:8 (18S rRNA primers: competitive primers) in order to generate unsaturated RT PCR signals over the concentration range of total RNA used in this experiment. PCR reactions for all genes were subjected to 25 cycles at 95C (30 sec), 52C (45 sec), and 72C (90 sec) with GeneAmp PCR System 2400 (Perkin Elm er). For the analysis of temperature response and organ specific expression, three rounds of RT PCR were conducted with three independently isolated total RNA samples. For the analysis of differential expression during seed maturation and germination, 2 rounds of RT PCR were conducted with two independently isolated total RNA samples. Twenty l from each PCR reaction was fractionated by 1.5% agarose gel in TAE buffer and stained with 0.5% ethidium bromide. The ethidium bromide stained gels were digitall y photographed with an IS 1000 Digital Imaging
26 Table 3 1. Arabidopsis Hsp70 gene family Proposed name Original name Predicted subcellular location Introns GenBank acession number Clone type Reference Hsc70 1 At Hsc70 1 Hsp70 1 Cytosol 1 AL162971 X74604 BAC cDNA Bevan et al., 2000b a Wu et al., 1994 Hsc70 2 Hsp70 2 Cytosol 1 AL162971 BAC Bevan et al., 2000b a Hsc70 3 At Hsc70 3 Hsp70 3 Cytosol 1 AC011436 Y17053 BAC cDNA Lin et al., 1999 a Hsieh et al., 1998 Hsp70 Hsp70 Cytosol 1 AP0020 55 AJ002551 BAC cDNA Nakamura, 2000 a Hinderhofer et al., 1998 a Hsp70b Cytosol 0 AC010924 BAC Liu et al., 1999 a BiP 1 ER lumen 5 AF262043 D89341 BAC Genomic Wilson, 2000 a Koizumi and Sano, 1997 BiP 2 ER lumen 5 AB017067 D89342 BAC Geno mic Nakamura, 1999 a Koizumi and Sano, 1997 BiP 3 ER lumen 4 AC000106 BAC Osborne et al., 1997 a mtHsc70 1 Mitochondrion matrix 5 AL035538 BAC Bevan et al., 1999b a mtHsc70 2 Hsc70 5 Mitochondrion matrix 5 AL353994 AF217458 BAC cDNA Bevan e t al., 2000a a Vierling et al., 2000 a cpHsc70 1 Plastid stroma 7 AL078637 BAC Bevan et al., 1999a a cpHsc70 2 Hsc70 7 Plastid stroma 7 AB024032 AF217459 TAC cDNA Nakamura, 1999 a Sung et al., 2000 a Hsp70t 1 Unknown 1 AC058785 BAC Lin et al ., 2000 a Hsp70t 2 Unknown 0 AC006223 BAC Lin et al., 2000 a a Indicates direct submissions to GenBank. These entries are not included in the literature cited. Note : The sequences for BiP 3, Hsp70t 1 and Hsp70t 2 were identified from database search es after RT PCR analyses were completed. The nucleotide sequence for BiP 3 is quite divergent from BiP 1, BiP 2 . However, their amino acid sequences showed remarkably high homology. Hsp70t 1 (617 aa) and Hsp70t 2 (563 aa) are truncated at their C termin al ends and their subcellular localizations have not been determined
27 System (Alpha Innotech Corporation). Scion Image for Windows (Scion Corporation http://www.scioncorp.com) program was used to quantify the intensity of the ethidium bromide stained DNA bands from the negative images of the gels. Results Arabidopsis Hsp70s Are Encoded by a Gene Family Arabidopsis contains genes encoding five cytosolic hsp70s, three BiPs (hsp70 homologs in the ER), two plastid hsp70s and two mitochondrial hsp70s. Includ ing the sequences of two organellar hsp70s cloned in this study (see Materials and Methods), full length sequences for 12 Arabidopsis hsp70 genes are now available in the database either from cDNA or genomic sequence, in addition to two truncated hsp70 seq uences ( Table 3 1 ). One of the two truncated sequences, Hsp70t 1 (AC058785), has no corresponding EST clone in the database while the other sequence, Hsp70t 2 , has a corresponding EST clone (AI996202). This suggests that Hsp70t 1 may not be expressed or expressed under conditions not included in the construction of EST libraries. Although Neighbor Joining analysis suggests that Hsp70t 1 belongs to the cytosolic group 3), there is not enough information to predict subcellular localization for either Hsp70 t 1 or Hsp70t 2 . From the genome sequencing database, I identified a mitochondrial hsp70 and refer to it as mtHsc70 1 (AL035538) and a chloroplast hsp70 as cpHsc70 1 (AL078637). I cloned cDNAs for a second mitochondrial hsp70 and a second chloroplast hsp 70 and named them mtHsc70 2 (AF217459) and cpHsc70 2 (AF217458). I propose a new nomenclature for hsp70 genes in Arabidopsis in order to clarify and establish consistency for this gene family. All gene names used in this study are listed in Table 3 1 alon g with accession numbers. For the remainder of this dissertation, I
28 will use the gene names proposed in Table 3 1 . Hsp70 genes are found on all five chromosomes ( Figure 3 1 ). Six hsp70s representing members localized to the major subcellular compartment s; two cytosolic ( Hsc70 1 and Hsc70 2 ), two ER ( BiP 1 , BiP 2 ), one chloroplast ( cpHsc70 2 ), and one mitochondrial member ( mtHsc70 2 ) are found on chromosome 5. Of these six hsp70 genes on chromosome 5, only two cytosolic members ( Hsc70 1 and Hsc70 2 ) are present in tandem. Chromosome 2 harbors only one hsp70 gene, Hsp70t 2 . The intron exon structure of the hsp70 genes in Arabidopsis is distinctive and different for genes encoding proteins targeted to different subcellular locations ( Table 3 1 ). Arabidop sis hsp70 genes encoding protein targeted to the same subcellular compartments are highly conserved in the number of introns and the length of exons ( Table 3 1 , and data not shown) indicating they are likely products of gene duplication events. For exampl e, four cytosolic hsp70 genes ( Hsc70 1, Hsc70 2, Hsc70 3, Hsp70 ) have one intron each and their corresponding exons are the same size. The fifth cytosolic member, Hsp70b , has no intron like many of the strongly heat inducible hsp70 genes in other organism s. A new hsp70 member for the ER, BiP 3, has four introns while the other two BiP genes ( BiP 1 and BiP 2 ) have three introns each. There is no conservation in the length of exons between the first two BiP genes and BiP 3 indicating that BiP 3 probably ar ose from a different evolutionary lineage. All organellar members have more introns than cytosolic members. Plastid and Mitochondrial Hsp70s Are Highly Conserved The two mitochondrial hsp70s, mtHsc70 1 and mtHsc70 2 , encode proteins of 666 and 682 amino acids, respectively with predicted isoelectric points of 5.17 and 5.60,
29 CIC2H3RE V IV III II I BiP 3 Hsp70b 0846A ATEAT1 phyA NCC1 g15785 PAI1 PAI3 mi342 GAPB P4 nga280 mi353 m315 PAB5 ADH mi320 mi310 mi398 mi139 GPA1 m283 ve016 m429 m336 mi79a nga32 m262 CIC3F3LE CIC4C5LE KG17 m105 CIC12H6RE GAPA AIG2 CIC2B7LE CIC3A10LE m249 CIC4E6RE AtEm1 CIC12G4RE CIC8D10LE CIC9B7LE mi51 mi233 mi87 CIC9G5 mi465 mi260 yUP19G1 mi422 mi123 mi431 CIC3H2 g3715 mi97 mi174 mi322 mi138 mi219 mi125 CIC4B3 CIC4E7 mi194 mi61 CIC11F10LE CIC6C5 mi184 mi335 Hsc70 3 Hsp70 cpHsc70 1 mtHsc70 1 cpHsc70 2 BiP 1 BiP 2 Hsc70 1/ Hsc70 2 mtHsc70 2 Hsp70t 1 Hsp70t 2 CIC2H3RE V IV III II I BiP 3 Hsp70b 0846A ATEAT1 phyA NCC1 g15785 PAI1 PAI3 mi342 GAPB P4 nga280 mi353 m315 PAB5 ADH mi320 mi310 mi398 mi139 GPA1 m283 ve016 m429 m336 mi79a nga32 m262 CIC3F3LE CIC4C5LE KG17 m105 CIC12H6RE GAPA AIG2 CIC2B7LE CIC3A10LE m249 CIC4E6RE AtEm1 CIC12G4RE CIC8D10LE CIC9B7LE mi51 mi233 mi87 CIC9G5 mi465 mi260 yUP19G1 mi422 mi123 mi431 CIC3H2 g3715 mi97 mi174 mi322 mi138 mi219 mi125 CIC4B3 CIC4E7 mi194 mi61 CIC11F10LE CIC6C5 mi184 mi335 Hsc70 3 Hsp70 cpHsc70 1 mtHsc70 1 cpHsc70 2 BiP 1 BiP 2 Hsc70 1/ Hsc70 2 mtHsc70 2 Hsp70t 1 Hsp70t 2 Figure 3 1. Chromosomal location of hsp70 genes in Arabidopsis. Hsp70 genes were positioned on a physical map created by the Arabidopsis Genome Initiative (AGI; http://www.ar abidopsis.org/agi.html). This map contains molecular markers as well as genetic markers.
30 respectively. The two plastid hsp70s, cpHsc70 1 and cpHsc70 2 encode proteins of 718 amino acids each with predicted isoelectric points of 5.03 and 4.96, respecti vely. These organellar hsp70s were aligned with full length organellar hsp70s from other plant species, revealing strong amino acid sequence conservation. The Arabidopsis plastid hsp70 proteins are 88% identical to each other and 81 to 85% identical with chloroplast hsp70 proteins from other plants. The two mitochondrial hsp70s are 78% identical to each other and 76 to 86% identical to those of other species. Sequence alignment also revealed that mtHsc70 1 differs from mtHsc70 2 and other plant mitochon dria hsp70 proteins in the N and C termini ( Figure 3 2 ). Arabidopsis mtHsc70 1 has deletions of several amino acids in the N terminal signal peptide region and an insertion of three amino acids in the C terminal end compared to other plant mitochondrial hsp70 proteins. This unique C terminus of mtHsc70 1 may indicate alternative suborganellar localization or specialized co chaperone interaction that is different from that of mtHsc70 2 and other known mitochondrial hsp70s. The C terminus of organellar hs p70s is highly conserved and can be used as a predictive localization motif for organellar hsp70 proteins (Guy and Li, 1998). The C termini of the Arabidopsis plastid hsp70s and the mitochondrial hsp70s also contain these conserved motifs (underlined resi dues in Figure 3 2 ). As noted in Table 3 1 , the proposed subcellular localization for the 12 full length Arabidopsis hsp70s was consistent with Neighbor Joining analysis ( Figure 3 3 ) and C terminal sequence motifs. The general branching pattern of this d endrogram is also in agreement with previous phylogenic analyses of hsp70s in yeast, plants and other organisms (Boorstein et al., 1994; Guy and Li, 1998).
31 A) N termini of chloroplast and mitochondrial hsp70 proteins cpHsc70 1 M ASSA A QIH I LG G IG F PNSFC SS STK N LD NK TNSIP R SVFFG N R T ---SPF S TP T SAF L R MGRRNNNAS R YT VGPV R VVNEK V V GID 84 cpHsc70 2 M ASSA A QIH V LG G IG F ASSSS S KRNL N GK GG TFM P R S A FFG T R T ---G PF S TP T SAF L R MGTRNGGGASRYA VGPV R VVNEK V V GID 84 Pea M ASSA QIH G LG T AS F ----SS L K K -P SS ISGNS KTL FFGQR LNSN H SPF T RA AF PK L SSKTFKK ---G FTL R VV S EK V V GID 74 Spinach M ASSA T QIH V LG ATP F --TT SS S K P -S SS RPN -SVFFGQ KMSPS TA CI G N P K SAF L RL KKSSGGRR R SAG VGPV R VVNEK V V GID 81 Cucumber M G ASTA QIH G LG A PS F -AAA S MR K S N NV SS ----R SVFFGQ KLGNS S A F -P AA AF L N L RSNTS -R R NSS V R P L R I VNEK V V GID 77 mtHsc70 1 M A ---------------S V SA F K S V SA N G K N S MFG ----K LGY LARPF C S R P V GND V I GID 42 mtHsc70 2 M ATAALLRSI R RR E VV SS P FSA Y R C L S S S G K A S L NSS Y LG Q NFRSFS R A FSS K PAGND V I GID 62 Potato M ATAALLRSL R RR E F A T SS I SA Y R T L A SN TKPS WC PSLVGA K WAG LARPFSS K PAGN EI I GID 62 Bean M A -A V LRSL R RR D V AS A T FSA Y RSL T G S TKP ---A Y VA QK WSC LARPFSS R PAGND V I GID 57 Pea M A A T LLRSL Q RR N L S SSS V SA F RSL T G S TK T S YATH ---K LAS L T RPFSS R PAGND V I GID 58 Spinach M AT AL R R C L R PEQ F -R SF P A F K SL A G N AS PS L SSP Y MA Q RLAS L V RPFSS R PAGND V I GID 60 B) C termini of chloroplast and mitochondrial hsp70 proteins cpHsc70 1 L Q E L KE K I AS GSTQ EIK D T M AA L N QE V MQ IG QSLYNQP QP G G A DS ----P P G GEAS S S S DT S S S A K G G DNG GDVIDADFTDS N 718 cpHsc 70 2 L Q E L K DK I GS GSTQ EIK DAM AA L N QE V MQ IG QSLYNQP -G AG GP G AGPS P G GEGA S SG D S S S S K G G D GD DVIDADFTDS Q 718 Pea LGE L KE A I T G GSTQ T IK DA L AA L N QE V MQ L G QSLYNQP -G A AG QA G PT P P G ---SESGPS E S SG K E G PEGDVIDADFTDSK 706 Spinach LGE L K D A I N G G E TQ A IK DAM AA L N QE V MQ L G QSLYNQP -G AG GEPG AG P G PTPGA ESGPS D S TS K G PEGDVIDADFTDSK 715 Cucumber LGE L KE A I S G GST EA IK E AM AA L N QE V MQ L G QSLYNQP -G AG AA ----P G PGAS SESGPS E S TG K G PEGDVIDADF S DSK 707 mtHsc70 1 V F D F R T A M A G E DVED IK AK VE AA N KA V SK IG EHMS K G F GSSG SDGSF G EGTS G TE Q T PEAE F EE ASGSR K 666 mtHsc70 2 V A D L R S A SS G DDLN EI K AK IE AA N KA V SK IG EHMS G GS G GG SAP G G G S E G G S DQAPEAEYEEV --KK 682 Potato I SD L R A A M GT EN I D D IK AKLD AA N KA V SK IG EHM A G GSSGG A S G G G GA QG G D Q P PEAEYEEV --KK 682 Bean VSD L R K A M S G D N V D EIK S KLD AA N KA V SK IG EHMS G GSSGG S S A G G SQG G G D Q APEAEYEEV --KK 675 Pea VSD L R T A M A G EN A D D IK AKLD AA N KA V SK IG Q HMS G GSSGG P S E G G SQG G E QAPEAEYEEV --KK 675 Spinach VSD L R S A M QND N LE EIK AK T D AA N KA V SK IG EHMS G G Q G GG S S SS G ---G A D Q T PEAEYEE A --KK 675 Figure 3 2. Sequence alignment of chloroplast and mitochondrial hsp70 proteins. Arabidopsis plastid hsp70s ( cpHsc70 1, cpHsc70 2 ) were aligned with chloroplast hsp70s from pea, spinach and cucumber. Arabidopsis mitochondrial hsp70s ( mtHsc70 1, mtHsc70 2 ) were aligned with mitochondrial hsp70s from potato, bean, pea and spinach. Black shading indicates consen sus residues common to both chloroplast and mitochondrial hsp70s. Gray shading indicates consensus residues specific in either chloroplast or mitochondrial hsp70s. Brackets indicate the beginning of the N terminal highly conserved ATP binding motif. Und erlined residues are C terminal signature motifs for organelle localization. Dashes indicate gaps introduced to maxmize alignment agreement.
32 Hsc70 2 Hsp70t 1 Hsc70 1 Hsc70 3 Hsp70 Hsp70b BiP 1 mtHsc70 2 mtHsc70 1 BiP 3 BiP 2 Hsp70t 2 cpHsc70 1 cpHsc70 2 0 Cytosol ER Mitochondrion Chloroplast 35 35.6 30 25 20 15 10 5 35 35.6 30 25 20 15 10 5 Hsc70 2 Hsp70t 1 Hsc70 1 Hsc70 3 Hsp70 Hsp70b BiP 1 mtHsc70 2 mtHsc70 1 BiP 3 BiP 2 Hsp70t 2 cpHsc70 1 cpHsc70 2 0 Cytosol ER Mitochondrion Chloroplast 35 35.6 30 25 20 15 10 5 35 35.6 30 25 20 15 10 5 Figure 3 3. A Non rooted Neighbor Joining analysis using the CLUSTAL program in DNASTAR for protein sequen ces of the Arabidopsis hsp70 gene family. Full length protein sequences of 12 Arabidopsis hsp70 proteins and two truncated sequences ( Hsp70t 1, Hsp70t 2 ) were used in this analysis. The scale at the bottom represents the branch distance as the number of changes in character states between neighbors. Each shaded area represents subcellular localization; from the top, cytosol, ER, mitochondrion, and chloroplast.
33 Optimization of RT PCR Robust analysis of hsp70 expression required gene specific primers for the eleven hsp70s examined in this study ( Table 3 2 ). Primers were designed to produce PCR products with different lengths in order to conduct multiplex RT PCR. However, multiplex RT PCR could not be used because of unequal and biased amplification of di fferent sequences. Therefore RT PCR reactions for each gene were analyzed individually. The conditions for RT PCR were optimized to produce unsaturated PCR product accumulation that retained a linear relationship with the original transcript levels in al l samples. A range of 1 to 256 ng of total RNA was tested and 16 ng and 64 ng of total RNA were found to generate unsaturated RT PCR product accumulation for each gene through 25 cycles of PCR. As an example, RT PCR signals of Hsc70 1 over a range of 1 to 256 ng of total RNA are shown ( Figure 3 4 ). Two RT PCR signals were generated for each sample; one for the individual hsp70 gene and one for 18S rRNA as an internal loading control. When the two signals were not saturated in the sample, the ratios of the two signals over a range of total RNA concentrations were reasonably constant. Total RNA concentrations of 16 ng and 64 ng consistently yielded the same ratio and also gave stoichiometric increases of RT PCR signals ( Figure 3 4 ). For the present anal ysis, 16 ng of total RNA was used for all reactions. RNA samples were treated with DNase I to eliminate DNA contamination. However, even without DNase I treatment, no amplification products from DNA contamination were detected for any of the genes (data not shown). The gene specificity of RT PCR was confirmed by sequencing all RT PCR products.
34 Table 3 2. Oligonucleotide primers used in RT PCR Gene Primer Sequence CG256F 5â€™ TGCCTACGGTCTTGACAA 3â€™ Hsc70 1 CG257R 5â€™ ACCTGGATCAACACACCG 3â€™ C G301F 5â€™ TGGCCTTTCACTATCATC 3â€™ Hsc70 2 CG302R 5â€™ TAGAAGTCAGCTCCACCA 3â€™ CG268F 5â€™ CCCTTCACGCTCAAATCT 3â€™ Hsc70 3 CG269R 5â€™ TCCTCCAGCGGTTTCAAG 3â€™ CG258F 5â€™ TCAAGCGGATAAGAGTCACT 3â€™ Hsp70 CG259R 5â€™ CTCGTCCGGGTTAATGCT 3â€™ CG284F 5â€™ TGTCGGAGTTTGGATG AAT 3â€™ Hsp70b CG285R 5â€™ CTGTCTCAAGTCCAAGGCTA 3â€™ CG260F 5â€™ ACTAAGATGAAGGAGACAGCT 3â€™ BiP 1 CG261R 5â€™ ACTTGGTGCTGACTACTTAGA 3â€™ CG262F 5â€™ ACTAAGATGAAGGAGACGACC 3â€™ BiP 2 CG263R 5â€™ TTGGTGCTGACTGCTTAAG 3â€™ CG234F 5â€™ GCTGCTGCACTATCATATGG 3â€™ mtHsc70 1 CG23 5R 5â€™ CACGGAGGATACCACCTT 3â€™ CG266F 5â€™ CGTTTCCTCTCCTTTCTCA 3â€™ mtHsc70 2 CG267R 5â€™ TTTGGCTAGGTCTATTCCC 3â€™ CG247F 5â€™ GGTGATCCTTGTTGGTGG 3â€™ cpHsc70 1 CG213R 5â€™ ATCTCAACGCTTGTCTGTC 3â€™ CG264F 5â€™ AGTGCCTTCTTCGGTACA 3â€™ cpHsc70 1 CG265R 5â€™ GGACACTCA AGCTTAACATTATT 3â€™ F: Forward primer, R: Reverse primer.
35 0.2 1.0 0.4 1.2 0.6 0.8 1ng 4ng 16ng 64ng 256ng * ** Hsc70 1 /18S rRNA Figure 3 4. RT PCR optimization. Equivalent increase of duplex Hsc70 1 and 18S rRNA signal in the range of 1 to 256 ng of total RNA was tested. The 16 ng total RNA was selected fo r subsequent experiments. * RT PCR band for Hsc70 1. ** RT PCR band for 18S rRNA.
36 Diverse Responses of Arabidopsis Hsp70s to Temperature Extremes Expression profiles for hsp70 genes during heat shock or cold acclimation were determined on plants that w ere exposed to 40C for 30, 60, and 90 minutes of heat shock treatment and to 4C for 12 and 48 hours of low temperature treatment. Control plants were kept at 20C. The most specific response to temperature extremes was that of Hsp70b ( Figure 3 5 ). The Hsp70b transcript was detectable only during heat treatment and was not detected during any other treatment or developmental stage or in any organ. Except for mtHsc70 1 and cpHsc70 1, all members of the family showed induction of 2 to 20 fold by 30 minu tes at 40C ( Figures 3 5 & 3 6 ). The induction of Hsc70 1 and BiP genes expression in response to heat shock was in good agreement with previous findings by other laboratories using hybridization based techniques (Wu et al., 1988; Wu et al., 1994; Koizumi , 1996). Despite the strong and nearly universal induction by heat shock, repression kinetics of hsp70s were quite diverse. Three classes of repression kinetics could be discerned: rapid, within 30 to 60 minutes at 40C ( Hsc70 2 ); moderate, 60 to 90 min utes at 40C ( Hsc70 1, Hsc70 3, Hsp70b ) and slow, 90 minutes or more at 40C ( Hsp70, cpHsc70 2, BiP 1, BiP 2, mtHsc70 2 ) ( Figures 3 5 & 3 6 ). Several hsp70s were also induced during low temperature treatment, but responsiveness to cold was limited to cyto solic and mitochondrial hsp70s. Hsc70 1 and Hsc70 3 were induced 3 to 5 fold within 12 hours at 4C while Hsc70 2 and Hsp70 were induced 10 fold or more by 48 hours at 4C. mtHsc70 1 and mtHsc70 2 showed about a two fold increase after 48 hours at 4C, while the transcript levels of both ER and chloroplast members ( BiP 1, BiP 2, cpHsc70 1 and cpHsc70 2 ) showed little or no change ( Figures 3 5 & 3 6 ). Contrary to the responses of BiP genes in tomato and spinach (Li et al., 1999), the expression of
37 5 10 15 20 25 5 10 15 20 25 5 10 15 20 25 Arbitrary unit (hsp70/18S rRNA ) Hsc70 1 Hsp70 Hsc70 2 20C 4C 40C 12 48 30 90 60 20C 4C 40C 12 48 30 90 60 Hsc70 3 Hsc70 3 Hsp70b Hsp70b Figure 3 5. Response of cytosolic hsp70s to temperature extremes. Two week old Arabidopsis plants were subjected to 4C for 12 and 48 hours for low temperature treatment or 40C for 30, 60 and 90 minutes for heat shock treatment. Control plants (20C) we re incubated at 20C simultaneously. Signal values obtained from each gene were normalized with the 18S rRNA signal value and the resulting values were presented as relative units. Error bar represents standard deviation.
38 5 10 15 20 25 5 10 15 20 25 5 10 15 20 25 Arbitrary unit (hsp70/18S rRNA ) BiP 1 BiP 1 BiP 2 BiP 2 cpHsc70 2 cpHsc70 2 mtHsc70 1 mtHsc70 1 20C 4C 40C 12 48 30 90 60 20C 4C 40C 12 48 30 90 60 cpHsc70 1 cpHsc70 1 mtHsc70 2 mtHsc70 2 Figure 3 6. Response of orga nellar hsp70s to temperature extremes. Empty space indicated by a white line in the gel pictures of cpHsc70 1 and mtHsc70 2 was cut out to achieve uniform spatial arrangement of the images.
39 BiP 1 and BiP 2 at low temperature did not show much change. T wo interesting aspects of the temperature response of Arabidopsis hsp70s are the absence of heat induction of mtHsc70 1 and the induction of Hsp70b exclusively by heat shock. The absence of heat inducible expression in mtHsc70 1 appears to be due to the absence of heat shock elements in the promoter, as the promoters for 11 Arabidopsis hsp70 genes were examined for the presence of two major temperature responsive cis elements, heat shock element (HSE) and C repeat or dehydration responsive element (CRT/DR E) ( Figure 3 7 ). HSE has been linked with the heat inducible expression of many heat shock genes (Czarnecka et al., 1989). CRT/DRE is known to be associated with drought and cold inducible expression of many genes (Yamaguchi Shinozaki and Shinozaki, 19 94). Overall, the results indicate expression profiles of hsp70 genes and the presence of cis elements in the promoters are in good agreement. Hsp70s that showed strong induction by heat shock contain multiple HSE elements ( Figure 3 7 ), but no functional HSE was found in the promoter of mtHsc70 1 ( Figure 3 7 ). One or more CRT/DRE were found in the promoters for strongly cold inducible members such as Hsc70 3 , Hsp70 , and mtHsc70 2 . In contrast there are exceptions for the presence of the cis elements and induction of hsp70s by temperature stress. An HSE was not found in BiP 1 and BiP 2 where heat induction was clearly observed, and CRT/DREs were not found in the promoter of the strongly cold inducible member, Hsc70 2 . Conversely, a CRT/DRE was found in cpHsc70 2 , yet cold induction was not detected. HSE and CRT/DRE are the best characterized cis elements for heat and cold induction of hsp70 genes, but heat and cold induction of hsp70 genes results from the function of a complex array of cis
40 500 400 300 200 100 0 Hsc70 1 Hsc70 2 Hsc70 3 Hsp70 Hsp70b BiP 2 BiP 1 cpHsc70 1 cpHsc70 2 mtHsc70 1 mtHsc70 2 Rd29A/Cor78 Hsp17.4 Heat Cold 500 400 300 200 100 0 Hsc70 1 Hsc70 2 Hsc70 3 Hsp70 Hsp70b BiP 2 BiP 1 cpHsc70 1 cpHsc70 2 mtHsc70 1 mtHsc70 2 Rd29A/Cor78 Hsp17.4 Heat Cold 500 400 300 200 100 0 Hsc70 1 Hsc70 2 Hsc70 3 Hsp70 Hsp70b BiP 2 BiP 1 cpHsc70 1 cpHsc70 2 mtHsc70 1 mtHsc70 2 Rd29A/Cor78 Hsp17.4 Heat Cold 500 400 300 200 100 0 Hsc70 1 Hsc70 2 Hsc70 3 Hsp70 Hsp70b BiP 2 BiP 1 cpHsc70 1 cpHsc70 2 mtHsc70 1 mtHsc70 2 Rd29A/Cor78 Hsp17.4 Heat Cold Figu re 3 7. Predicted cis elements in the promoters of Arabidopsis hsp70 genes. Promoter sequences for 11 Arabidopsis hsp70 genes, a cold inducible gene (Rd29A/Cor78), and a heat inducible gene (Hsp17.4) were analyzed. The numbers at the bottom indicate the number of nucleotides upstream to the translation initiation codon, ATG. Induction fold of each gene in response to heat and cold are indicated as solid diamonds, one diamond; less than 5 fold, two diamonds; 5 to 10 fold, three diamonds; more than 10 fold . Solid circle; perfect HSE (nTTCnnGAAnnTTCn or nGAAnnTTCnnGAAn). Open circle; imperfect HSE. Solid triangle; the core sequence of CRT/DRE (CCGAC). Solid box; TATA box.
41 elements. For example, heat induction of BiPs without an HSE can be explained by the presence of multiple C1 elements that are critical for the unfolded protein response (Wooden et al., 1991). Similarly, cold induction of Hsc70 2 without CRT/DRE could possibly be explained by the presence of ABA responsive elements (ABRE). Specific M embers of Arabidopsis Hsp70s Are Induced during Seed Maturation and Germination Transcript levels of hsp70 genes in green silique (7 DAP), yellow silique (14 DAP) and dry seed were analyzed. Hsp70 showed the greatest induction (8 fold) of the family durin g seed maturation and desiccation ( Figure 3 8 ). Transcript levels of mtHsc70 2 also rose during this period, but to a lesser extent ( Figure 3 9 ), while the transcript levels for Hsc70 1, Hsc70 2, Hsc70 3, BiP 1, BiP 2, cpHsc70 2 and mtHsc70 1 were diminis hed. The transcripts of Hsp70b and cpHsc70 2 were not detectable during this stage of development ( Figures 3 8 & 3 9 ). Previous analyses from our lab showed induction of hsp70 genes around two days of imbibition (data not shown). When samples were take n at 6, 12, 24, 48, 96 hours of imbibition and analyzed, transcripts of Hsp70 were found to disappear within 24 hours after the onset of imbibition. Hsp70b was not detected at any time point during imbibition and germination ( Figure 3 8 ). Depending on th e timing of peak expression during germination, members of the family could be divided into three classes; early, intermediate and late. The early class showed peak expression at 6 hours after imbibition and Hsc70 2 is indicative of this class. The inter mediate class showed peak expression between 6 and 24 hours after imbibition, and the two mitochondrial members ( mtHsc70 1, mtHsc70 2 ) belong to this class. The late class showed peak expression at 24 and 96 hours of imbibition and Hsc70 1, Hsc70 3, BiP 1 , BiP 2, cpHsc70 1 and cpHsc70 2
42 Hsc70 1 Hsc70 2 Hsc70 3 Hsp70 Hsp70b 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 S1 S2 Sd 6 12 24 48 96 Imbibition S1 S2 Sd 6 12 24 48 96 Imbibition Arbitrary unit (hsp70/18S rRNA ) Figure 3 8. Expression of cytosolic hsp70s during seed maturation and germination. Silique samples (S1, S2) were harvested from 5 week and 6 week old plants. S1; silique at 7 DAP, S2; silique at 14 DAP, Sd; mature d ry seed. Samples were also harvested at 6, 12, 24, 48 and 96 hours after imbibition.
43 BiP 1 BiP 2 cpHsc70 1 cpHsc70 2 mtHsc70 1 mtHsc70 1 S1 S2 Sd 6 12 24 48 Imbibition S1 S2 Sd 6 12 24 48 Imbibition 96 96 1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11 Arbitrary unit (hsp70/18S rRNA ) Figure 3 9. Expression of organellar hsp70s during seed maturation and germination.
44 belong to this class. Noteworthy of this class was the very low expression le vel of BiP 1 and BiP 2 in mature seed, which was followed by very strong induction at 48 and 96 hours of imbibition. Hsp70s Are Differentially Expressed in the Organs of Arabidopsis under Basal Conditions The expression of Arabidopsis hsp70 genes was analy zed to determine whether individual members of the family were expressed in particular organ(s). Hsp70 transcripts were abundant in root, but barely detectable in other organs ( Figure 3 10 ). Hsc70 3 and mtHsc70 1 were also detected at higher levels in ro ot. In contrast, Hsp70b was not detected in any organ in the absence of heat shock. Transcripts for cpHsc70 1 and cpHsc70 2 were detected at higher levels in leaves than other organs ( Figures 3 10 & 3 11 ). Two ER members, BiP 1 and BiP 2 were abundantly present in all organs tested. However, BiP transcript levels were slightly increased in floral tissues ( Figure 3 11 ). Unfortunately, transcript levels of mtHsc70 2 were too low to resolve organ specific expression. Transcript levels for all members app eared to diminish in young silique at 3 DAP ( Figures 3 10 & 3 11 ). Discussion Temperature Response Previous expression studies for plant hsp70s demonstrated induction in 2 hours of heat shock either at 37C or 40C (DeRocher and Vierling, 1995; Koizumi, 19 96; Li et al., 1999; Wu et al., 1988; Wu et al., 1994). Most Arabidopsis hsp70s reached peak induction within 30 minutes of heat shock exposure, and the rapid response of hsp70
45 5 10 15 20 25 5 10 15 20 25 5 10 15 20 25 Arbitrary unit (hsp70/18S rRNA ) Rt Lf St Fl Si Rt Lf St Fl Si Hsc70 1 Hsc70 1 Hsc70 2 Hsc70 3 Hsp70b Hsp70 Hsp70 Figure 3 10. Expression of cytosolic hsp70s in different organs. Rt; roo t, Lf; leaf, St; stem, Fl; flower, Si; silique at 3 DAP
46 5 10 15 20 25 5 10 15 20 25 5 10 15 20 25 Arbitrary unit (hsp70/18S rRNA ) Rt Lf St Fl Si Rt Lf St Fl Si BiP 1 BiP 2 BiP 2 cpHsc70 2 mtHsc70 1 cpHsc70 1 mtHsc70 2 Figure 3 11. Expression of organellar hsp70s in different organs. Rt; root, Lf; leaf, St; stem, Fl; flower, Si; silique at 3 DAP
47 genes to heat shock is not limited to Arabidopsis, as the induc tion of spinach hsp70 genes was detected as early as 5 minutes after heat shock (Li and Guy, unpublished data). After initial heat induction Arabidopsis hsp70s showed rapid and diverse repression profiles. The repression of some hsp70 genes ( Hsc70 2, mtH sc70 2 ) starts as early as 60 minutes after the onset of heat shock while others ( BiP 1, BiP 2, cpHsc70 2 ) remained at an induced level 90 minutes after heat shock. Repression patterns for Arabidopsis BiP previously analyzed (Koizumi, 1996) were consisten t with the present results and demonstrate that BiP genes have a relatively slower repression system than for other plant hsp70s. In contrast, induction by low temperature treatment was limited to cytosolic and mitochondrial members of hsp70s in Arabidopsi s. Except for Hsp70b , all cytosolic hsp70s showed strong induction by low temperature treatment. The reason why the cytosolic members are strongly induced at low temperature is not clear, but it may be related to increased demand for molecular chaperone function at low temperature. The temperature responses suggest that cytosolic and ER hsp70 genes are responsible for molecular chaperone activity under heat stress and mainly cytosolic hsp70s are required under low temperature stress in Arabidopsis. Rol es of Hsp70s in Seed Maturation Induction of BiP expression in Arabidopsis seems to occur earlier during seed development than in pumpkin, rice and wheat (Hatano et al., 1997; Muench et al., 1997). BiPs in all three plants were induced 1 to 2 weeks after pollination where rapid cell expansion and accumulation of seed storage proteins occurs, but decreased rapidly towards the end of seed maturation (Hatano et al., 1997; Muench et al., 1997). Two Arabidopsis BiP genes were induced at or before the time of pollination. The major seed
48 storage protein in Arabidopsis (12S protein) begins to accumulate around one week after pollination (Wehmeyer et al., 1996). Whether the high transcript levels of the two BiPs in flowers were a prerequisite for the rising flux of seed storage proteins through the ER of floral organ cells requires further investigation. I have shown the expression of all hsp70 genes except for Hsp70 and mtHsc70 2 was repressed in siliques during the later stages of seed development (7, 14 DAP) . This implies diminishing roles of hsp70 genes during seed development. Based on the overall low levels of hsp70 expression, siliques (including developing seeds) may be one of the most sensitive organs to heat stress. Alternatively, the decline in the expression of hsp70 genes in seeds may be compensated for by the increased expression of other chaperones and stress proteins during late seed maturation, including the small Hsps, Hsp101 and LEA proteins. Hsp70 showed a striking expression pattern during seed maturation and germination where it was absent in flowers and young siliques, but present at high levels in dry seed. Hsp70 transcripts accumulated during the later stages of seed development and/or during desiccation. Subsequently, Hsp70 transcrip ts rapidly disappeared during germination. Similar expression patterns were observed for cytosolic hsp70s of pea and a mungbean hsp70 (DeRocher and Vierling, 1995; Wang and Lin, 1993). This pattern of Hsp70 expression also closely follows the expression patterns of small heat shock proteins (Wehmeyer et al., 1996) and LEA genes (Raynal et al., 1999) in Arabidopsis, suggesting that Hsp70 expression may be regulated by a common mechanism that regulates these classes of genes during development (Almoguera et al., 1998; Galau and Hughes, 1987).
49 The presence of Hsp70 transcripts in mature dormant seed makes it a preserved mRNA (Harris and Dure, 1978). The reason for the presence of Hsp70 transcript as a preserved mRNA remains unclear, but there are two possibi lities. During imbibition and germination, the resumption of protein synthesis may require immediate production of the cytosolic chaperone for efficient protein biogenesis. Alternatively, the Hsp70 mRNA could serve as an immediate source of Hsp70 for tra nslation if imbibition occurred during a period of high temperature exposure. Roles of Hsp70s in Seed Germination In the early stages of germination, disaggregation of protein bodies and utilization of storage proteins has to be efficiently maintained t o cope with increased demand for amino acids and energy for organogenesis. In the latter stages, a substantial transformation takes place in seedlings during the formation of the photosynthetic apparatus and conversion of plastids to chloroplasts, which i s manifested by the greening of seedlings around 48 to 96 hours after imbibition. The chaperone activities of hsp70s may be needed in two important aspects of protein metabolism during germination, which explains the induction of many hsp70s. First, prot eins that are unfolded or misfolded during seed desiccation could be susceptible to aggregation during seed imbibition (i.e. rehydration of proteins), and hsp70 chaperones may need to be present in every compartment of the cell as soon as the cells become rehydrated in order to minimize the toxic effects of protein aggregation. Second, the initiation of active synthesis and translocation of proteins must be protected to ensure optimal function of metabolic processes during germination. The copious amounts of Hsp70 transcript in dry seed may serve as a reservoir for rapid access to molecular chaperone activity during the initial stages of storage protein utilization. In the same context, induction of Hsc70 2
50 within the first 6 hours of imbibition appears j ust as important as Hsp70 for the initial stages of germination since Hsp70 transcripts rapidly disappeared in the first 12 hours of imbibition. Five hsp70s, Hsc70 1, Hsc70 3, BiP 1, BiP 2, cpHsc70 2, were strongly induced by 96 hours after imbibition in this study. The induction of these genes coincides with the greening of cotyledons and could be involved in the utilization of storage proteins in cotyledons, formation of photosynthetic apparatus and the developmental conversion of plastids to chloroplas ts. In the case of BiP induction, specific roles for BiP proteins during germination have been elucidated in pumpkin where induction accompanied the degradation of seed storage proteins (Hatano et al., 1997). It was concluded that BiP participated in the degradation of seed storage proteins by assisting in folding and assembly of newly synthesized hydrolytic enzymes responsible for the degradation of seed storage protein (Hatano et al., 1997), and the same may be true for Arabidopsis. Organ Specific Expre ssion of Hsp70 Genes in Arabidopsis High expression in root and very low or no expression of Hsp70 in other organs suggest a specific role of Hsp70 in root growth or function. Other hsp70s also expressed in root are Hsc70 3, BiP 1, BiP 2 and mtHsc70 1 . W hen BiP gene expression was reduced in tobacco with an antisense approach, root formation of transgenic shoot cuttings was compromised, suggesting a role of BiP in root formation (Leborgne Castel et al., 1999). In fact, the expression of the two BiP genes analyzed in this study was ubiquitous indicating vital roles of BiP genes in whole plant cellular metabolism. The signal of cpHsc70 1 was higher in leaf and very low elsewhere suggesting a specific role of cpHsc70 1 in chloroplast. Transcripts of cpHsc7 0 2 were higher in all organs compared to cpHsc70 1 suggesting a general role(s) of cpHsc70 2 in all forms of plastids.
51 In summary, I have shown that the expression patterns of hsp70 genes, one of the most highly conserved gene families, are distinct and in many cases, differential expression pattern(s) can be linked to major physiological or developmental processes occurring in plants. Based on the expression patterns, a role for Hsp70 can be ascribed to seed maturation and germination, Hsp70b exclusivel y to heat stress, and the cytosolic/ mitochondrial hsp70s to cold stress. It will be challenging to investigate how the chaperone activities of hsp70 proteins translate into specific physiological roles in the context of plant cell function and in the lar ger context of plant growth and development. I will use the information obtained from this study to devise experimental strategies to assess the major phenotypes of transgenic plants that over/underexpress individual hsp70s.
52 CHAPTER 4 CREATING OVER/UNDEREXPRESSION LINES OF Hsc70 1 and BiP 2 IN ARABIDOPSIS Arabidopsis has five full length hsp70 genes localized in the cytosol ( Hsc70 1, Hsc70 2, Hsc70 3, Hsp70, Hsp70b ) and three BiP genes, BiP 1, BiP 2, and BiP 3 (Sung et al., 2001). T he five cytosolic hsp70 genes share 87 to 89% identity at the amino acid level, and they are more responsive to temperature changes than the genes for organellar hsp70s in Arabidopsis (Sung et al., 2001). Hence, the cytosolic members have been implicated most often with a function in acquired thermotolerance. The presence of multiple members of hsp70 in the cytosol and the overlap of expression patterns among members during temperature treatment, development, and in different organs suggest at least some degree of functional redundancy (Sung et al., 2001). Nonetheless, I also observed expression patterns that were specific to individual members. Unfortunately, any attempt to modify the expression of an individual member of the gene family runs the risk o f falling short of pinpointing the exact consequences of altered expression of individual members in such a redundant background. However, until knockout mutants for all hsp70 genes are available, transgenic approaches to modify the expression of individu al hsp70 genes represent a reasonable methodology in an attempt to decipher physiological roles in plants. I have made four constructs for cytosolic Hsc70 1 , two under the control of a constitutive promoter (a figwort mosaic virus (FMV) promoter) and two under the control of an inducible promoter (an Arabidopsis Cor78/RD29A promoter). The transgene was
53 placed in a sense or antisense orientation for each type of promoter. I have attempted to test the consequences of over and underexpression of cytosolic Hsc70 1 for thermotolerance and monitored growth and development of the transgenic plants. BiP 1 and BiP 2 share 98% identity at the amino acid level, but BiP 3 is quite different from BiP 1 and BiP 2 . I hypothesized that BiP 1 and BiP 2 make up the major portion of the total BiP pool in the ER and BiP 3 makes up the remainder with presumably a more specific function. BiP is known to bind to unfolded proteins in the ER lumen and to participate in ER quality control during protein biogenesis of secretory p roteins. However, it is not well characterized how the molecular chaperone function of BiP in the ER contributes to overall plant metabolism. BiP is an abundant ER resident protein suggesting important housekeeping roles. Consequently, it was reported t hat overexpression of tobacco BiP protein was quite feasible, but underexpression was not (Leborgne Castel et al., 1999). Although primary tobacco transformants containing BiP antisense construct showed apparently a lower level of BiP protein, the reducti on of BiP proteins became less obvious in the next generation (Leborgne Castel et al., 1999). The authors also stated that transforming tobacco calli with BiP antisense construct also produced transformed calli with minor levels of antisense transcript su ggesting the reduction in BiP level was not permissible and possibly detrimental to tobacco cells (Leborgne Castel et al., 1999). I investigated whether altering the abundance of BiP protein by transgenic modification of BiP 2 in Arabidopsis is possible and if so, whether altered levels of BiP protein are sufficient enough to change protein trafficking through ER during stress conditions as well as under normal conditions. Four transformation constructs for BiP 2
54 were generated in the same manner that ge nerates Hsc70 1 constructs. In a series of experiments, I characterized the responses of BiP 2 sense and antisense transgenic plants after imposing three very different ER stresses; heat shock, tunicamycin treatment, and pathogen challenge. Materials and Methods Vector Construction The proprietary transformation vector (pHK1001), obtained from Dr. Harry Klee with consent from Monsanto, was used to generate constitutive expression of Hsc70 1 and BiP 2 constructs. The vector pHK1001 has two major advantage s over the pBI121 vector commonly used in agrobacterium mediated transformation. It is relatively small (7 kb) compared to pBI121 (15kb), and its multi cloning site is near the right border of T DNA, which enhances the transfer of intact transgene sequenc es into the plant genome. Conventional transformation vectors have their multi cloning sites near the left border after the NPTII gene (kanamycin resistant gene), which raises the risk of truncation or deletion of transgene during T DNA transfer. The str ucture of T DNA in the original pHK1001 is as follows; the right border FMV promoter ACC gene NOS terminator NOS promoter NPTII gene NOS terminator the left border ( Figure 4 1 ). FMV promoter is from figwort mosaic virus. It is as strong a constitutive pr omoter as CaMV 35S promoter, and drives expression of transgene well in reproductive tissues and vegetative tissues (Maiti et al., 1997). A full length cDNA for BiP 2 obtained from Dr. Nozomu Koizumi (Nara Institute of Science and Technology, Japan) repla ced the ACC gene to generate constitutive sense or antisense transformation vector. Then the FMV promoter
55 RB Hsc70 1 (S) P NPTII 128 P P NOS LB X N X N FMV NOS NOS RB NPTII 128 P P NOS LB X N N FMV NOS NOS Hsc70 1(AS) P X RB Hsc70 1(S) P NPTII 128 P P NOS LB X N X N COR78 NOS NOS RB NPTII 128 P P NOS LB X N N COR78 NOS NOS Hsc70 1(AS) P X pHK Hsc70 1 (S) pHK Hsc70 1 (AS) pHKCOR Hsc70 1 (S) pHKCOR Hsc70 1 (AS) RB BiP 2 (S) P NPTII 128 P P NOS LB X N X N K FMV NOS NOS RB NPTII 128 P P NOS LB X N N FMV NOS NOS BiP 2 (AS) P K RB BiP 2 (S) P NPTII 128 P P NOS LB X N X N K COR78 NOS NOS RB NPTII 128 P P NOS LB X N N COR78 NOS NOS BiP 2 (AS) P K pHK BiP 2 (S) pHK BiP 2 (AS) pHKCOR BiP 2 (S) pHKCOR BiP 2 (AS) Figure 4 1. Transformation constructs of Hsc70 1 and BiP 2 gene. RB: right border of T DNA, FMV: Figwort Mosaic Virus promoter, BiP 2 (S): BiP 2 cDNA in sense orientation, BiP 2 (AS): BiP 2 cDNA in antisense orientation, NOS: NOS terminator or promoter, NPTII 128: a NPT gene. Green box is the region that was truncated in BiP 2 antisense constructs. It contains a total of 208 bp including 128 b p of N terminus coding sequence. K: Kpn1, N: Not1, P: Pst1, X: Xba1.
56 was replaced with an inducible promoter (Cor78/RD29A) to produce two inducible transgene vectors. The Cor78/RD29A promoter is from Arabidopsis and drives the expression of the Cor78/ RD29A gene in response to osmotic stress, low temperature, dehydration, and ABA (Yamaguchi Shinozaki and Shinozaki, 1993). Altogether, four constructs were generated for BiP 2 gene and four constructs for Hsc70 1 ( Figure 4 1 ). The final eight constructs were then transferred to agrobacterium (ABI) by a tri parental mating method. Agrobacterium was grown on LB plates with kanamycin (50 g/ml) and chlorampenicol (25 g/ml) for two days, helper E. coli (DH5a) with pRK2013 grown overnight on LB plates with kanamycin (50 g/ml), and E. coli (DH5a) with each construct was grown on LB plates with spectinomycin (100 g/ml) then subcultured in 2 ml of liquid LB with respective antibiotics. Aliquots of 50 l of each culture were streaked on LB plates without anti biotics. Bacterial lawns appeared after overnight culture. A loopful of the bacterial lawn was streaked on LB plate with all three antibiotics and incubated at 28C. Colonies of transformed agrobacterium appeared after 2 to 3 days. Plant Growth and Tran sformation by Vacuum Infiltration Seeds of Columbia ecotype were sown in 5" X 5" plastic pots covered with plastic window mesh. Plants were grown under continuous light (150 ? mole/sec ? m2) at 20 ? C, and irrigated once every third day with tap water and fert ilized once a week with a commercial premixed fertilizer (Peter's 20 20 20). For plant transformation, a modified version of the vacuum infiltration method was used (Bent et al., 1994). Infiltrated plants were laid on their side in a plastic flat then co vered with plastic wrap for one day. After one day, plants were set upright and covered with a gallon size freezer bag to maintain
57 humidity and prevent cross pollination. After 3 to 4 weeks, seeds from the same pot were harvested together. Seeds were st erilized and 100 ? l of seeds (2,500 seeds) plated on selection media. After 7 to 14 days, transformants were distinguished as green seedlings. These primary transformants were transferred to a medium with a higher concentration of gelling agent (1.75 % a gar compared to 0.8%). This allows roots of true transformants to grow while it eliminates pseudo transformants. PCR Screening of Transformants Putative transformants obtained after selection on hard agar medium were subjected to PCR screening. Primers s pecific to promoters (FMV, Cor78) and BiP 2 , and Hsc70 1 were designed to amplify the transgenes in kanamycin resistant seedlings ( Figure 4 2 ). A primer pair for each promoter and construct produced a distinct sized PCR product ( Table 4 1 ). Genomic DNA f rom each transformant was extracted as described (Li and Chory, 1998). Three l of genomic DNA was amplified with 10 pmoles of each primer by 35 cycles of 95C for 1 min, 55C for 1 min, and 72C for 1.5 min. Each PCR positive line was identified by gel electrophoresis and designated individually. pHK BiP 2 (S) lines start with #6, pHK BiP 2 (AS) with #2, pHKCOR BiP 2 (S) with #7, and pHKCOR BiP 2 (AS) with #5. Confirmation of T DNA Integration into Arabidopsis Genome A modified adaptor ligation PCR tech nique was used to confirm integration of T DNA into the Arabidopsis genome. This technique yields a different length PCR product for each insertion because the length of the genomic DNA fragment flanking T DNA on a restriction fragment is specific to the integration site. One g of genomic DNA was digested with TaqI. Two adaptor primers (CG336; 5â€™ CTAATACGACTCAC TATAGG GCTCGAGCGGCCGGGCAGGT 3â€™, CG337; 5â€™ GCACCTGCCCAA 3â€™) were ligated at
58 Hsc70 1 (S) RB FMV LB NOS N X P K X N Hsc70 1 (AS) RB FMV LB NOS N X P X N Hsc70 1 (S) RB COR78 LB NOS N X P K X N Hsc70 1 (AS) RB COR78 LB NOS N X P X N CG197 CG197 CG196 CG196 CG194 CG195 CG194 CG195 BiP 2 (S) RB FMV LB NOS N X P K X N BiP 2 (AS) RB FMV LB NOS N K P X N BiP 2 (S) RB COR78 LB NOS N X P K X N BiP 2 (AS) RB COR78 LB NOS N K P X N CG197 CG197 CG196 CG196 CG192 CG193 CG192 CG193 Figure 4 2. Location of primers for PCR confirmation of the const ructs and transgenic lines. Table 4 1. Length of products for PCR confirmation of the constructs . Size of PCR products Constructs (Primer) FMV (CG 197) Cor78 (CG 196) Hsc70 1 sense (CG 194) 461 bp 584 bp Hsc70 1 anti sense (CG 195) 403 bp 526 b p BiP 2 sense (CG 192) 349 bp 472 bp BiP 2 anti sense (CG 193) 289 bp 412 bp
59 80C for 2 min followed by a stepwise cool down over 40 min to produce the adaptor that ligates to TaqI digested genomic DNA. Ligation of genomic DNA and the adaptor was carri ed out as described (Spertini et al., 1999). Adaptor ligated genomic DNA was subject to two rounds of PCR. In the first round of PCR, adaptor ligated genomic DNA was amplified with an adaptor specific primer CG338 (5â€™ ATCCTCTAATACGACTCACTATAGGGC 3â€™) and T DNA left border primer LB1 (5â€™ CGCCTATAAATACGACGGA 3â€™) by 7 cycles of 94C for 30 sec, 52C for 1 min, and 72C for 3 min followed by 32 cycles of 94C for 30 sec, 56C for 1 min, and 72C for 3 min. The PCR product was diluted 50 times and an aliquot o f 1l was subjected to a second round of PCR. In the second round of PCR, the final PCR amplicons were produced with CG339 (5â€™ TATAGGGCTCGAGCGGC 3â€™) and LB2 (5â€™ CGCTGCGGACATCTACAT 3â€™), a nested primer set to CG338 and LB1, by 30 cycles of 94C for 30 sec, 55C for 1 min, and 72 for 3 min. PCR products were separated in a 1.5% agarose gel. The bands on the gel were excised for sequencing using LB3 primer (5â€™ GACCATCATACTCATTGCTG 3â€™) in order to localize the insertion locus on the Arabidopsis chromosome. Preparation of Recombinant BiP 2 Protein and Western Blot Analysis The coding sequence for BiP 2 was amplified from pHK BiP 2 (S) with primer CG212 (5â€™ CCGGAATTCTCTAGAGCTCATCGTGA 3â€™) and CG215 (5â€™ CCGGAATTCATATGGCTCGCTCGTTTG 3â€™). The two primers were des igned with an EcoRI site. PCR amplified BiP 2 cDNA was then digested with EcoRI and cloned into pGEX2T protein expression vector (pGEX2T BiP 2) and E. coli strain BL21 cell was transformed with pGEX2T BiP 2. Expression of recombinant BiP 2 protein was in duced by 0.1 mM IPTG at 37C for 4 hr. Total soluble proteins were extracted as
60 described (Anderson et al., 1994). Total soluble proteins separated by SDS PAGE were transferred onto PVDF membranes using a semi dry blot transfer cell (Trans blot ? SD, Biora d) and immuno blotted with monoclonal antibodies raised against spinach ER luminal hsc70 (1D9). This antibody specifically recognizes spinach BiP protein(s). Hypocotyl Elongation Assay Seeds of transgenic Arabidopsis plants were disinfected with 70% ethan ol for 2 min followed by 5% bleach (0.25% sodium hypochlorite) and 1% SDS for 15 min. Seeds were then washed with sterile water three times. Seeds were plated on 1X MS medium with or without 0.2 g/ml tunicamycin. The plates were wrapped with aluminum f oil and placed vertically in a growth cabinet. The growth temperature was 20C. After one week in the growth cabinet, hypocotyl length was measured. Seed Thermotolerance Assay Seeds were imbibed at 4C for 4 days for uniform germination and subjected to a 10 min heat treatment ranging from 50C to 60C in 2C increments. Seeds were plated on pre soaked filter paper in petri dishes. The dishes were sealed with Parafilm and incubated in a growth chamber with the photoperiod of 15 hrs light/9 hrs dark, a t 20C, and a light intensity of 150 mol/secm 2 . Thermotolerance was evaluated at indicated times by determining germination rate. Whole Plant Thermotolerance Assay Arabidopsis plants were grown in a grid pattern 4x4 (total 16 plants) in sealed clear pla stic delicatessen containers containing commercial soil (Fafard mix No. 2) at 20C for 18 days with a photoperiod of 15 hrs light/9 hrs dark. Thermotolerance assays were conducted by inverting the containers over water baths allowing the aerial portion o f the plants to be immersed at temperatures of 42, 44, 46, 48, and 50C for 10 min.
61 Electrolyte leakage of leaves was measured 3 days after heat treatment. For electrolyte leakage measurements, the aerial portions of the plants were placed in scintil lation vials containing 10 mL distilled water and shaken for 1 hr. After the first conductivity reading, the vials containing tissues were subject to a 2 min microwaving at high setting that raised the temperature of the solution to the boiling point, thu s killing all cells. After cooling to room temperature and shaking for 1 hr, the second readings were taken and the LT 50 determined from electrolyte leakage data. The LT 50 was arbitrarily set at an electrolyte leakage of 50%. Disease Progression Assay Le aves of four week old transgenic and wild type Arabidopsis plants were challenged with a strongly virulent ( P. syringae pv. tomato DC3000 ), and a weakly virulent strain ( P. syringae pv. maculicola ES4326) at a titer level of 10 4 colony forming units (cfu) /mL as described ( Davis et al., 1991 ). Pseudomonas strains were grown in Kingâ€™s B (KB) mediu m in the presence of 100 g/ml rifampicin. Bacterial suspension was prepared from mid to late log phase culture (OD at 600 nm = 0.5 1.5). Bacterial cells were pelleted by centrifugation and resuspended in an equal volume of sterile 10 mM MgCl 2 . The con centration of viable bacterial cells was estimated by measuring the OD at 600 nm and adjusted by diluting with sterile 10 mM MgCl 2 . Leaves were infiltrated with bacteria with a syringe, without a needle, through the underside of fully expanded leaves of i dentical developmental stage. Bacterial populations were monitored in leaf disk samples of 0.6 cm diameter following homogenization in 0.2 ml MgCl 2 solution. Samples (50 l) from serial dilutions of the homogenate were plated on KB medium. The ability o f the bacteria to multiply was monitored over a 7 day period by colony counts.
62 Results Hsc70 1 and BiP 2 Sense/Antisense Transgenic Plants Seeds obtained from infiltrated plants were selected on germination medium containing 30 g kanamycin/ml. Kanamyci n resistant seedlings were further analyzed for the presence of transgenes by PCR ( Figure 4 3 ). In the case of pHKCOR BiP 2 (AS) transgenic lines, the PCR product of 412 bp was amplified as expected ( Table 4 2 ). Quantitative data were collected for all transformation experiments ( Table 4 2 ). Most interestingly, I was not able to obtain any primary transformants for pHK Hsc70 1 (AS) ( Table 4 2 ). Occasional green seedlings of pHK Hsc70 1 (AS) did not survive the two week screening period on kanamycin med ia. The number of primary transformants for pHK Hsc70 1 (S) was also very low. I was able to obtain only 8 lines after screening ca. 50,000 seeds ( Table 4 2 ). The effort to generate homozygous lines for plants containing pHK Hsc70 1 (S) has not been suc cessful. Western blot analysis on T 0 transgenic lines for pHK Hsc70 1 (S) showed that a moderate increase in protein level in several lines was achieved ( Figure 4 4 ). Similarly, only nine transgenic plants containing pHK BiP 2 (AS) were obtained ( Table 4 2 ). Western blot analysis on T 0 transgenic lines for pHK BiP 2 (AS) showed a reduced level of BiP protein in several lines ( Figure 4 5 ). The reduced protein levels observed in those lines were not seen in western blot analysis of T 2 homozygous lines. H owever, I was able to obtain ample an amount of primary transformants for the constructs with the inducible promoter. It is also interesting to observe that not only the constitutive antisense constructs yielded a low percentage or no transformation, but also the inducible antisense constructs rendered
63 + â€“ pHKCOR BiP 2 (AS) lines + : pHKCOR BiP 2 (AS) plasmid DNA â€“ : genomic DNA of wild type Arabidopsis Figure 4 3. A representative PCR screening for the presence of pHKCOR BiP 2 (AS) construct in the transgenic plants. Table 4 2. Transformation efficiency of Hsc70 1 and BiP 2 constructs. Total s eeds screened PCR positive lines Transformation efficiency pHK Hsc70 1 (S) 50,000 8 0.016 % pHK Hsc70 1 (AS) 52,500 0 0.00 % pHKCOR Hsc70 1 (S) 15,000 95 0.63 % pHKCOR Hsc70 1 (AS) 15,000 37 0.25 % pHK BiP 2 (S) 12,500 47 0.38 % pHK BiP 2 (AS) 17,500 9 0.06 % pHKCOR BiP 2 (S) 15,000 123 0.82 % pHKCOR BiP 2 (AS) 17,500 83 0.47 %
64 8 2 8 10 8 8 8 5 8 4 8 11 8 9 8 7 WT Transgenic individuals Figure 4 4. Western blot analysis on pHK Hsc70 1 (S) in T 0 transgenic lines. WT: wild type Arabidopsis. The blots were cross reacted with a monoclonal antibody (5 B7) specific to cytosolic hsc70. The images were modified to bring the lane for wild type Arabidopsis to the left of the image when necessary.
65 WT 2 2 2 22 2 30 2 36 2 20 2 29 Transgenic individuals Figure 4 5. Western blot analysis of pHK BiP (AS) in T 0 transgenic lines. WT: wild type Arabidopsis. T he blots were cross reacted with a monoclonal antibody (1D9) specific to BiP. The images were modified to bring the lane for wild type Arabidopsis to the left of the image when necessary.
66 lower transformation efficiency than the sense constructs. I susp ect that there is a fortuitous expression of the transgenes at room temperature resulting in some degree of antisensing in untreated plants. In the case of pHK BiP 2 (S), 47 primary transformants were obtained. Selected BiP 2 sense/antisense lines were a dvanced to further generations to obtain homozygosity. Homozygosity of a line was confirmed by 100% kanamycin resistance of the subsequent generation. Eight lines for pHK Hsc70 1 (S) were also propagated to subsequent generations, but so far I have not been successful in obtaining homozygous lines for any of the eight lines. Further analysis of transgenic lines for pHK Hsc70 1 (S) was abandoned due to the difficulty in obtaining homozygous lines. Homozygous plants for pHK BiP 2 (S) and (AS) were subjec ted to a modified adaptor ligation PCR to confirm T DNA insertions in the Arabidopsis genome ( Figure 4 6 ). As expected, T3 transgenic lines generated from the same T0 plant harbor T DNA insertion at the same location in their genome. Conversely, transgen ic lines derived from different T0 plants have T DNA insertions in different places in their genome. For example, lanes 2, 3, 4 are progenies of the same T0 plant. The PCR results show identical band patterns among these three lines indicating T DNA inse rtions in these three lines are identical as expected. Lanes 1, 2, 5, 6, and 8 were derived from different T0 plants, hence showing different sized band patterns. Of nine independent transgenic lines, I attempted to sequence six lines in order to determi ne the precise location of T DNA insertions in the genome. The sequences of four lines were successfully obtained and localized on the physical map generated by Arabidopsis Genome Initiative (AGI) ( Figure 4 7 ). A sense
67 1 3 2 4 5 6 7 8 9 10 11 12 13 14 15 1: 2 2 6 2: 2 29 1 3: 2 29 5 8 4: 2 29 6 4 5: 2 20 2 9 6: 2 22 7 7: 2 36 4 8: 2 36 5 9: 6 60 10 10: 6 32 2 5 11: 6 32 6 7 12: 6 61 3 2 13: 6 61 4 3 14: 6 15 2 9 15: Wild type 1 3 2 4 5 6 7 8 9 10 11 12 13 14 15 1 3 2 4 5 6 7 8 9 10 11 12 13 14 15 1: 2 2 6 2: 2 29 1 3: 2 29 5 8 4: 2 29 6 4 5: 2 20 2 9 6: 2 22 7 7: 2 36 4 8: 2 36 5 9: 6 60 10 10: 6 32 2 5 11: 6 32 6 7 12: 6 61 3 2 13: 6 61 4 3 14: 6 15 2 9 15: Wild type Figure 4 6. Adaptor ligated PCR confirmation of T DNA integration in BiP 2 transgenic lines.
68 CIC2H3RE V IV III II I BiP 3 Hsp70b 0846A ATEAT1 phyA NCC1 g15785 PAI1 PAI3 mi342 GAPB P4 nga280 mi353 m315 PAB5 ADH mi320 mi310 mi398 mi139 GPA1 m283 ve016 m429 m336 mi79a nga32 m262 CIC3F3LE CIC4C5LE KG17 m105 CIC12H6RE GAPA AIG2 CIC2B7LE CIC3A10LE m249 CIC4E6RE AtEm1 CIC12G4RE CIC8D10LE CIC9B7LE mi51 mi233 mi87 CIC9G5 mi465 mi260 yUP19G1 mi422 mi123 mi431 CIC3H2 g3715 mi97 mi174 mi322 mi138 mi219 mi125 CIC4B3 CIC4E7 mi194 mi61 CIC11F10LE CIC6C5 mi184 mi335 Hsc70 3 Hsp70 cpHsc70 1 mtHsc70 1 cpHsc70 2 BiP 1 BiP 2 Hsc70 1/ Hsc70 2 mtHsc70 2 Hsp70t 2 Hsp70t 1 6 60 10 2 20 2 9 6 32 3 4 2 36 4 CIC2H3RE V IV III II I BiP 3 Hsp70b 0846A ATEAT1 phyA NCC1 g15785 PAI1 PAI3 mi342 GAPB P4 nga280 mi353 m315 PAB5 ADH mi320 mi310 mi398 mi139 GPA1 m283 ve016 m429 m336 mi79a nga32 m262 CIC3F3LE CIC4C5LE KG17 m105 CIC12H6RE GAPA AIG2 CIC2B7LE CIC3A10LE m249 CIC4E6RE AtEm1 CIC12G4RE CIC8D10LE CIC9B7LE mi51 mi233 mi87 CIC9G5 mi465 mi260 yUP19G1 mi422 mi123 mi431 CIC3H2 g3715 mi97 mi174 mi322 mi138 mi219 mi125 CIC4B3 CIC4E7 mi194 mi61 CIC11F10LE CIC6C5 mi184 mi335 Hsc70 3 Hsp70 cpHsc70 1 mtHsc70 1 cpHsc70 2 BiP 1 BiP 2 Hsc70 1/ Hsc70 2 mtHsc70 2 Hsp70t 2 Hsp70t 1 6 60 10 2 20 2 9 6 32 3 4 2 36 4 Figure 4 7. Locations of the BiP 2 transgene insertion in the Arabidopsis genome.
69 line, 6 60 10, was localized to the intergenic region between gene H2A and C2H2, a zinc finger type protein gene on chromosome 4. Another sense line, 6 32 3 4, was localized to the promoter region of a putative RING finger protein in chromosome 3. An antisense line, 2 20 2 6, was localized in the second exon of a gene of unknown function on chromosome 5. Lastly, another antisense line, 2 36 4, was localized to the intergenic region between two genes of unknown function. Based on these results, lines 6 60 10 and 2 36 4 should have fewer side effects from disrupting Arabidopsis genome function by the i ntegration of transgenes since they inserted into intergenic regions of the genome. Recombinant BiP 2 Protein and Western Analysis of Transgenic Lines. Recombinant BiP 2 protein was expressed in E. coli strain BL21, separated by SDS PAGE, and immuno blot ted with a monoclonal antibody raised against spinach BiP protein (1D9). Total cell extract from wild type Arabidopsis as well as recombinant BiP 2 protein reacted with 1D9 ( Figure 4 8 ). 1D9 did not react with Arabidopsis recombinant cytosolic Hsc70 1 pr otein. With this experiment, I confirmed that the monoclonal antibody raised against spinach BiP protein is highly specific to BiP proteins in Arabidopsis. At this point, it is not clear whether 1D9 antibody reacts only with BiP 2 or reacts with all thre e BiP proteins. The level of BiP proteins in six BiP 2 sense lines and eight BiP 2 anti sense lines was determined by immuno blotting with 1D9. Several sense lines showed significant increase at the protein level ( Figure 4 9 ). Line 6 13 showed about a tw o fold increase, but generation homozygous lines has not been successful, so it was left out from further analyses. Homozygous lines of 6 60 10, 6 32 6 7, and 6 32 2 5 showed 3 to 5 fold increase in protein band intensity. However, anti sense lines did n ot show as much reduction of the protein as primary transformants ( Figure 4 10 ). This result indicates that
70 GST BiP 2 BiP 2 GST Hsc70 1 Hsc70 1 1 2 3 4 5 6 7 8 1. GST Hsc70 1 2. GST BiP 2 3. BiP 2 4. Arabidopsis leaf extract 5. GST BiP 2 6. GST Hsc70 1 7. Hsc70 1 8. Arabidopsis leaf extract 1D9 5B7 Figure 4 8. Specificity of hsp70 antibodies. The first four lanes were cross reacted with a monoclonal antibody specific to BiP (1D9) and the second four lanes with a monoclonal antibody specific to cytosolic hsc70 (5B7).
71 1 2 7 3 4 5 6 BiP LHCP 1: 6 60 10 2: wild type 3: 6 61 4 3 4: 6 13 5: 6 15 2 9 6: 6 32 6 7 7: 6 32 2 5 1 2 3 4 5 BiP/LHCP Figure 4 9. Protein expression in BiP 2 sense transgenic plants. Total cellular protein was extracted with 1:3 w/v ratio with 2X SDS loading buffer and separated i n 12% SDS PAGE. The upper half of the blot was reacted with 1D9 and the bottom half reacted with aLHCP rabbit polyclonal antibody (a gift from Dr. Kenneth Cline).
72 1 8 2 3 4 5 6 7 BiP 9 LHCP 1: 2 2 6 2: 2 20 2 9 3: 2 29 5 8 4: 2 22 7 5: wild type 6: 2 36 4 7: 2 36 5 8: 2 29 1 9: 2 29 6 4 1.5 0.5 1 BiP/LHCP Figure 4 10. Protein expression in BiP 2 antisense transgenic plants. Total cellula r protein was extracted with 1:3 w/v ratio with 2X SDS loading buffer and separated in 12% SDS PAGE. The upper half of the blot was reacted with 1D9 and the bottom half reacted with aLHCP rabbit polyclonal antibody (a gift from Dr. Kenneth Cline).
73 increa sing BiP 2 protein level is tolerable in Arabidopsis, while significantly decreasing BiP 2 levels is not. Further analyses for characterizing the consequences of altered BiP 2 expression by transformation were carried out with selected homozygous lines. Those are 6 15 5 2, 6 32 6 7, 6 60 10, and 6 61 5 4 for sense lines and 2 2 6, 2 20 2 9, 2 29 6 4, and 2 36 4 for anti sense lines. Expression Analysis of BiP 2 Transgenic Lines In order to investigate whether BiP 2 gene expression is indeed increased or d ecreased in transgenic lines and whether altered BiP 2 expression resulted in any changes in gene expression of BiP related and BiP interacting proteins, four BiP 2 sense and four BiP 2 anti sense transgenic lines were subject to RT PCR for eight functiona lly related genes to BiP 2 gene ( Figure 4 11 ). Those eight genes include BiP 1, BiP 2, BiP 3; two ER localized molecular chaperones known to interact with BiP, calreticulin and GRP94; and three hsp70 members from other subcellular compartments, Hsc70 1, c pHsc70 2, and mtHsc70 2. Gene specific primers were designed for each gene. There are three homologs of calreticulin in Arabidopsis, therefore, I designed PCR primers that will amplify all three. RT PCR analysis showed significant increase of BiP 2 in t wo sense lines, 6 32 6 7 and 6 60 10 ( Figure 4 12 ). The increase in transcripts was quantitatively represented at the protein level showing two to five fold increase in BiP 2 protein in sense lines ( Figure 4 9 ). The other two sense lines, 6 61 5 4 and 6 15 5 2 did not show much increase of BiP 2 protein compared to wild type. Two antisense lines, 2 36 4 and 2 29 6 4, show a slight decrease in BiP 2 transcript level, but the protein levels were not much different from that of wild type ( Figure 4 10 ). Tw o other antisense lines, 2 20 2 9 and 2 2 6, showed wild type level of BiP 2 transcripts ( Figure 4 12 ).
74 BiP 1 BiP 3 Grp94 mtHsc70 2 Hsc70 1 cpHsc70 2 Calreticulin BiP 2 Over Under WT Figure 4 11. The gel images of RT PCR analysis of BiP 2 transgenic lines. The upper lane in each image is RT PCR amplicon for the designated gene . The lower lanes in all images are RT PCR amplicon for 18S rRNA. For RT PCR conditions and protocols, refer to materials and methods in chapter 2.
75 6-61-5-4 6-60-10 6-32-6-7 6-15-5-2 2-36-4 2-29-6-4 2-20-2-9 2-2-6 Wild-type Calreticulin Grp94 Hsc70-1 cpHsc70-2 mtHsc70-2 BiP-1 BiP-3 BiP-2 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 Arbitrary unit (gene/18s RNA) Figure 4 12. Graphical presentation of RT PCR analysis of BiP 2 transgenic lines. The average of three replicate experiments is shown in the graph. Standard deviation was calculated for each data point, but left out in the graph for a more clear presentation.
76 Hypocotyl Elongation Assay on BiP 2 Transgenic Lines The hypocotyl elongation was previousl y adopted by Vierling and her colleagues in a thermotolerance assay (Hong and Vierling, 2000). Hypocotyl elongation is very sensitive to stress and is easily scored by measuring hypocotyl length. Strong induction of BiP genes by tunicamycin is a hallmark of the stress response of the BiP gene. Although the induction of BiP by tunicamycin is a result of accumulation of unglycosylated proteins in the ER instead of a direct result of tunicamycin effect on BiP protein, I tested whether altering the BiP 2 pro tein level in Arabidopsis resulted in any changes in the whole plant response to tunicamycin treatment. Sense and antisense transgenic plants were grown on MS plates with or without 0.2 g/ml of tunicamycin for one week in the dark. After one week the hy pocotyl length was scored ( Figure 4 13A ). Tunicamycin treatment inhibited the growth of hypocotyls by 30 to 60% in transgenic line and wild type alike ( Figure 4 13B ). Even the elongation of hypocotyls of the two strong sense lines, 6 32 6 7 and 6 60 10, was not much different from that of wild type. This experiment shows that increasing BiP 2 protein level in Arabidopsis has little effect on the response to tunicamycin at the whole plant growth level. Seed Thermotolerance Assay on BiP 2 Transgenic Lines BiP protein interacts with seed storage proteins in plants: prolamin in rice, zein in maize, and phaseolin in bean. Expression of BiP genes is strongly induced in later stages of germination in Arabidopsis (Sung et al., 2001). I hypothesized that if BiP aids in the translocation of seed proteins during germination processes under normal condition, increased level of BiP protein in seed should help seed to cope better with environmental changes such as heat stress during germination. After 10 min heat tre atment ranging from 50C to 60C in a water bath, seeds were plated on pre wet filter paper in petri dish.
77 A. Tunicamycin Control 6 60 10 2 36 4 B. 0 5 10 15 20 25 Wild type 6-61-5-4 6-60-10 6-32-6-7 6-15-5-2 2-36-4 2-29-6-4 2-20-2-9 2-2-6 Length of hypocotyl (mm) Untreated 0.2 g/mL Tunicamycin Figure 4 13. Effect of tunicamycin on hypocotyl elongation of BiP 2 transgenic lines. (A) A representative picture of seedlings grown on MS plates with or without tunicamycin. (B) Graphical presentation of hypocotyl elongation on MS plates with or without tunicamycin. The error bars indicate the standard deviation of three experiments.
78 After one week, seeds that survive heat treatme nt germinated ( Figure 4 14A ). The performance of seeds was scored by counting germinated seeds of each line ( Figure 4 14B ). Seeds with fully expanded cotyledons were counted as germinated seeds. The effect of heat treatment on seeds was manifested in tw o ways. First, heat treatment up to 54C delayed germination for ca. 48 hours. This is a non discriminating effect on all the lines tested. Second, at 56C and higher, germination is permanently affected by the heat treatment. Up to 54C, seeds from ev ery line showed germination rates close to 100% and at 58C and 60C, virtually all seeds were killed by the heat treatment. However, heat treatment at 56C discriminates lines doing better than others ( Figure 4 14 ). Two strong sense lines, 6 32 6 7 and 6 60 10, showed 50% germination and a moderate sense line, 6 15 5 2, showed 40% germination while a sense line with wild type BiP 2 protein level and non transformed wild type showed ca. 20% germination ( Figure 4 14B ). This indicates that increased BiP 2 protein in transgenic lines resulted in improved germination following heat shock treatment. Western analysis of BiP 2 in transgenic lines just before heat treatment showed a tight correlation between increased protein level and thermotolerance of imbibed seeds ( Figure 4 15 ). Whole Plant Thermotolerance of BiP 2 Transgenic Lines In order to investigate whether enhanced thermotolerance in seed of BiP 2 transgenic lines is also characteristic of whole plants, I analyzed the thermotolerance of BiP 2 transgeni c lines 3 weeks after sowing. Plants were subjected to 10 minute heat treatment in water bath with temperature ranging from 42C to 50C, then incubated for three days in normal growth condition before evaluating thermotolerance. By measuring 50% leakage of total available electrolytes in leaf, I was able to show an increase in
79 A. WT 6 61 5 4 6 60 10 6 32 6 7 6 15 5 2 54C 56C 58C B. Seed thermotolerance 0 10 20 30 40 50 60 70 80 90 100 110 RT 50C 52C 54C 56C 58C 60C Heat treatment Germination (%) 6-61-5-4 6-60-10 6-32-6-7 6-15-5-2 WT Figure 4 14. Seed thermotolerance of BiP 2 (S) lines. (A) Seeds germinated on wet filter paper after heat treatment. (B) Seed thermotolerance of BiP 2 (S) lines.
80 1 2 3 4 BiP Hsc70 Wild type 6 61 5 4 6 60 10 6 32 6 7 6 15 5 2 Wild type BiP /Hsc70 Figure 4 15. BiP protein level in imbibed seeds before heat treatment. Two hundreds seeds from each line were ground in 2x SDS loading buffer and 10 ? l of seed extract were subject to SDS PAGE and western blot analysis. One of two identical gels was cross reacted with BiP specific antibody (the upper panel) and the other was cross reacted with cytosolic hsc70 antibody (the lower panel).
81 thermotolerance of Arabidopsis by 3C after a 90 min pre heat treatment ( Figure 4 16A ). If there is any change in thermotolerance in BiP 2 transgenic whole plants, I should be able to detect it with the electrolyte leakage assay since this assay clearly revealed the induction of acquired thermotolerance in Arabidopsis ( Figure 4 16A ). Electrolyte leakage measureme nt showed no difference in thermotolerance between BiP 2 transgenic lines and wild type plants ( Figure 4 16B ). The result indicates little effect of increased BiP 2 protein level in transgenic lines on thermotolerance of whole plants. Lines 6 61 5 4 and 6 32 6 7 showed slight increase in thermotolerance at 44C. However, 6 60 10, a more strongly overexpressing line than 6 61 5 4, did not show any increase in thermotolerance at 44C. Disease Progression in BiP 2 Transgenic Lines In response to wounding o r elicitation, several of the molecular chaperones of the endoplasmic reticulum are strongly induced showing elevated mRNA levels. Among these, BiP is particularly responsive to wounding (Kalinski et al., 1995; Guy and Li, 1998 ), salicylic acid or culture filtrate elicitors (Denecke et al., 1995). This molecular chaperone response is likely to be a consequence of the host response that results in the elaboration of defense related genes some of whose products are secreted from the cell. Once the presence of a disease agent has been detected and the host response begins, the flux through the secretory pathway o f glucanases, chitinases and other proteins localized to the cell wall area is rapidly increased, which perhaps concomitantly signals the need for increased molecular chaperone capacity to accommodate the rising flux via the unfolded protein signaling path way (Brewer et al., 1997). Since the ultimate outcome of the disease process and the host response is a function, in part, of the dynamics of the
82 A. 0 10 20 30 40 50 60 70 80 90 100 RT 42 44 46 48 50 Temperature (C) % Electrolyte Leakage 0 15 90 0 10 20 30 40 50 60 70 80 90 100 RT 42 44 46 48 50 Temperature (C) % Electrolyte Leakage 0 15 90 B. 0 10 20 30 40 50 60 70 80 90 100 RT 42 44 46 48 50 Temperature (C) % Electrolyte leakage Control 6 61 5 4 6 60 10 6 32 6 7 6 15 5 2 0 10 20 30 40 50 60 70 80 90 100 RT 42 44 46 48 50 Temperature (C) % Electrolyte leakage Control 6 61 5 4 6 60 10 6 32 6 7 6 15 5 2 Figure 4 16. Whole plant thermotolerance of BiP 2 transgenic lines. (A) Electrolyte leakage a nalysis during induction of acquired thermotolerance. (B) Electrolyte leakage analysis of BiP 2 (S) lines. In panel A, plants are incubated at 40C for 0, 15, and 90 min before heat treatment. The error bars represent the standard deviation of two or thr ee experiments.
83 response to suppress the infection process and that of the pathogen to circumvent defenses or overwhelm them, then the ability of the plant to effectively mobilize defenses as rapidly as possible is significant. For those host defense p roteins that must traverse the secretory pathway to their destination of action, being able to complete that journey as quickly as possible is important to the course of disease progression. Therefore, altered expression of BiP may influence the rate of d isease progression upon infection. If BiP function is rate limiting in these processes then changing its abundance could influence the rate of biogenesis of secretory proteins. This has now been clearly demonstrated in yeast where overexpression by the i nsertion of extra copies of the gene increased BiP abundance by 10 fold (Harmsen et al., 1996), which caused a 20 fold increase in the amount of secreted protein. In contrast, overexpression of BiP in Chinese hamster ovary cells delayed the secretion of t hyroglobulin by retaining it in the ER in complexes with BiP and GRP94 (Muresan and Arvan, 1998). So not only may overexpression accelerate flux through the secretory pathway, it may also result in delayed conformational maturation, inappropriate retentio n and reduced flux. When BiP abundance was reduced in yeast to about 5% of wild type, the secretion of three proteins, granulocyte colony stimulating factor, acid phosphatase and pancreatic trypsin inhibitor were all diminished compared to wild type (Robi nson et al., 1996). Interestingly, in contrast to the Chinese hamster ovary cells (Muresan and Arvan, 1998), overexpression of BiP in yeast did not adversely affect secretion so apparently it was not the rate limiting factor at wild type levels for the th ree proteins studied. If BiP is a rate limiting factor in secretory protein biogenesis in
84 Arabidopsis, then several outcomes are possible ( Table 4 3 ). BiP sense lines and wild type (Columbia ecotype) were challenged with a weakly virulent ( P. syringae pv . maculicola ES4326) and a strongly virulent ( P. syringae pv. tomato DC3000) strain at titer level of 10 4 cells/mL as described (Davis et al., 1991). The ability of the inoculant to multiply was determined by dilution plating from a homogenate of a 0.6 c m diameter leaf disk on King's medium. Bacterial populations were monitored over a 7 day period. The weakly virulent ( P. syringae pv. maculicola ES4326) strain progressed very slowly over a 7 day period ( Figure 4 17A ). Overall population growth of ES432 6 was about one order of magnitude. In contrast, the strongly virulent ( P. syringae pv. tomato DC3000) strain progressed very quickly over the first three days showing population increases of three orders of magnitude ( Figure 4 17B ). Over a 7 day period, the populations of DC3000 increased by four orders of magnitude. These results showed that I was successful to elicit the appropriate pathogenicity in Arabidopsis. However, there was no difference in population growth of ES 4326 and DC3000 in wild type and four sense transgenic lines tested ( Figure 4 17 ). Discussion Plant Transformation The efficiency of transformation indicates that lowering the expression of cytosolic Hsc70 1 is apparently lethal to Arabidopsis ( Table 4 2 ). A full length cDNA clone of Hsc70 1 was used to build the antisense constructs of Hsc70 1 . There are five cytosolic hsp70 genes that share 81 to 87% identity at nucleotide level. In this case, the full length cDNA clone used for constitutive anti sensing probably caused a global
85 Table 4 3. Possible disease progression phenotypes. BiP Expression Level Bacterial Population Wild type BiP Effect on Host Response Disease Phenotype Wild type Increases Rate limiting Normal avirulent and virulent processes Overexpression Increases e qual or less than wild type Rate limiting Attenuated virulent process Underexpression Increases equal or more than wild type Rate limiting Accentuated virulent process Wild type Increases Not rate limiting Normal avirulent and virulent processes Overexp ression Increases same as wild type Not rate limiting Normal avirulent and virulent processes Underexpression Increases same as wild type Not rate limiting Normal avirulent and virulent processes
86 A. P. syringae pv. maculicola ES4326 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 Day 0 Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Days after inoculation Log (cfu/cm) 6-61-3-2 6-60-10 6-32-6-7 6-15-2-9 wildtype B. P. syringae pv. tomato DC3000 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 Day 0 Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Days after inoculation Log (cfu/cm) 6-61-3-2 6-60-10 6-32-6-7 6-15-2-9 wildtype Figure 4 17. Population growth progression of Pseudomonas in BiP 2 transgenic lines. (A) P. syringae pv. maculicola ES4326. (B ) P. syringae pv. tomato DC3000.
87 repression of all cytosolic members resulting in death of all strong underexpression transformed cells. In order to test whether elicit ing repression of Hsc70 1 and probably those of the four other cytosolic hsp70 genes was indeed detrimental to cells, I incubated ten cold inducible antisense lines for Hsc70 1 at 5C for two months. All the lines survived low temperature incubation and c ompleted their entire life cycle (data not shown). However, it is too early to conclude that inducing repression of cytosolic hsp70 genes by low temperature did not cause damage to the plants, since it has not been tested yet whether repression of cytosol ic hsp70 genes in those inducible lines actually occurred. This will be attempted in the near future. BiP 2 antisense construct also rendered a very low, but non zero transformation efficiency suggesting that a global repression of the three BiP genes i s not completely lethal. Alternatively, it may have been that BiP 2 specific or at least BiP 1 and BiP 2 specific repression has occurred instead of a general repression of all three BiP genes. This can be tested by confirming expression of antisense RNA and concomitant decrease of endogenous RNA by RT PCR. I obtained nine BiP 2 antisense transformants out of 17,500 seeds screened. All nine primary transformants showed some degree of antisensing at the protein level but showed wild type BiP 2 protein le vels in subsequent generations. This prompted us to suspect that all the severe antisense transformants did not make it through the first screening procedure and that only mild antisense transformants survived the screening procedure and advanced to furth er generations. In order to further test this hypothesis, I need to confirm the presence of strongly inducible BiP 2 antisense lines.
88 Increasing the expression of Hsc70 1 constitutively also seems to be tightly controlled in Arabidopsis since I was able t o obtain only 8 primary transformant lines. Western blot analysis of these primary transformants for Hsc70 1 showed an increase in protein level in several lines but not as strong as those of BiP 2 overexpressing transgenic lines. Most of the constitutiv e Hsc70 1 sense plants are dwarf and very poor in setting seeds (data not shown). It has not been possible to reach homozygosity for any of these lines so far suggesting that doubling the expression level of these eight lines is also detrimental. In orde r to confirm the toxicity of doubling the expression of Hsc70 1 , the lethality of homozygosity of constitutive Hsc70 1 sense lines needs to be confirmed by careful segregation analysis in subsequent generations. Transformation of Arabidopsis for constitut ive BiP 2 overexpression was as efficient as inducible constructs. As to why constitutive BiP 2 overexpression did not affect transgenic plants negatively while constitutive Hsc70 1 overexpression showed adverse effects is not clear. One possible explana tion is a broader range of cellular metabolism that cytosolic Hsc70 1 protein is involved compared to those of BiP 2 proteins. Any negative effect generated from having increased amount of Hsc70 1, for example, increase in retention time of substrate prot eins and thereby causing prolonged protein maturation and/or translocation, would be more global and render serious consequences in case of overexpression of cytosolic Hsc70 1 than that of BiP 2 proteins. Increased Level of BiP 2 Protein in Transgenic Lin es Did Not Change the Response of Plants to Tunicamycin. Tunicamycin inhibits glycosylation of the proteins thereby causing accumulation of unglycosylated and incompletely folded proteins in the ER. A strong induction of BiP by tunicamycin treatment is a result of accumulation of incompletely folded proteins
89 rather than direct action of tunicamycin to BiP protein. However, the recruitment of BiP to incompletely folded proteins generated by tunicamycin treatment might deplete the total cellular pool of fre e BiP and thereby limit the amount of BiP available to other important cellular processes such as translocation of storage proteins and secretion of extracellular proteins. I postulated that transgenic plants should cope with tunicamycin treatment better because increased total cellular BiP pool due to overexpression of BiP 2 would alleviate the depletion of cellular BiP imposed by tunicamycin treatment. Transgenic Arabidopsis seeds were germinated and incubated vertically in the dark for one week on a s olid MS medium with 0.2 g/ml tunicamycin. Major differences in hypocotyl elongation of either sense or antisense lines of BiP 2 compared to that of wild type under tunicamycin treatment were not observed. Contrary to our results, transgenic tobacco BiP overexpression seeds germinated and survived 5 day treatment of 5 g tunicamycin treatment while wild type seed did not (Alvim et al., 2001). However, the experiment by Alvim et al. (2001) is not identical with our experiment. They incubated plants on a medium containg much higher tunicamycin for a shorter period time, then transferred to medium with no tunicamycin and analyzed for plant survival. It will be worthwhile repeating the germination/survival assay used for tobacco seed to test whether our Ara bidopsis transgenic seeds also survive the tunicamycin treatment. Imbibed Seeds of BiP 2 Overexpression Transgenic Lines Showed More Thermotolerance Than Wild type While Whole Plants of Transgenic Lines Did Not. BiP interacts with seed storage proteins in many plants. Through these interactions, BiP probably helps in the accumulation and packaging of seed storage proteins during seed maturation and unpacking and translocation of seed storage proteins during germination. Increased level of BiP protein in t he cells could be beneficial to
90 plants by accommodating increased trafficking of proteins going through the ER during seed maturation and germination and also by helping in the disaggregation of denatured proteins under stresses. In this study, two strong overexpression lines (6 32 6 7 and 6 60 10) and one moderate overexpression line (6 15 5 2) showed better germination than wild type when they were treated for 10 minutes at 56C. A tight correlation between increased protein level and increased thermoto lerance in transgenic lines suggested that increased levels of BiP protein were responsible for increased thermotolerance in transgenic lines. This can be further confirmed if the depletion of cellular BiP proteins in the transgenic plants by other forms of stress such as tunicamycin treatment or by overexpression of permanently unfolded proteins in the transgenic plants resulted in the loss of increased thermotolerance. Thermotolerance observed in transgenic seeds prompted us to ask whether increase in t hermotolerance persists throughout the life cycle of transgenic plants. The identical heat treatment was imposed on 3 week old whole plants with a lower temperature range. No difference in thermotolerance between transgenic and wild type plants was obser ved. Therefore, the thermotolerance I observed in imbibed seed is not permanently installed in plants, but more specific to this stage of plant development. This indicates that the cell damage alleviated by increased BiP in transgenic imbibed seed is spe cific to germination. In a future study, it will be important to determine what aspects of cell metabolism present in imbibed seed, but not in whole plant are protected by increased BiP under heat stress. Increased Cellular BiP Level Did Not Affect Dise ase Progression. I hypothesized that if BiP is a rate limiting factor in the secretory pathway of PR genes and other genes involved in pathogen defense, increased levels of BiP protein will
91 either bolster secretion of the proteins and thereby help cells to restrain pathogen growth. However, it is also possible that increased BiP proteins actually can delay secretion of the proteins by holding on to substrates longer. I challenged transgenic plants with two bacterial pathogens ( P. syringae pv . mauculicola ES4326 and P. syringae pv . tomato DC3000). The pathogens were able to proliferate according to their virulence, but no difference in pathogen progression as determined by population counts was observed between wild type and transgenic lines. This indicat es that BiP was not a rate limiting factor in the secretory pathway during pathogen challenge. Alternatively, the proteins or the processes that are benefited from increased BiP protein level may not be key elements in plant response mechanisms against th ese bacterial pathogens. Even though, the whole plant disease progression assay did not reveal any clear information about the roles of BiP during pathogen attack, I will attempt to determine whether an increased level of BiP protein actually affects secr etion of PR proteins in any manner. First, I plan to perform western blot analysis on extra cellular PR 1 proteins in wild type and transgenic lines to test whether secretion of PR 1 is affected. Second, I will perform reciprocal immunoprecipitation on c ellular BiP and PR 1 proteins to test whether an increase level of BiP results in additional or prolonged interaction between the two proteins. This will shed some light on the roles of BiP at the protein level. I created Hsc70 1 and BiP 2 sense and antis ense constructs under the constitutive or the inducible promoter to investigate the consequences of altered expression of Hsc70 1 and BiP 2 in plant. The absence of transformants for the constitutive antisense Hsc70 1 suggests that either global antisensi ng to all the cytosolic hsp70 genes or Hsc70 1 specific antisensing is detrimental to plants. I need to determine whether the antisensing
92 was specific to Hsc70 1 in future studies. I was able to obtain several putative Hsc70 1 overexpression transgenic p lants. I have tried to obtain homozygous lines for these transgenic lines, but were not successful. Consequently, not much analysis has been done with these lines. In retrospect, I should have tried several analyses, the thermotolerance assay in particu lar, with these heterozygous lines. It is very important to characterize these lines in future studies even as heterozygous lines. Transformation of BiP 2 was more amenable compared to Hsc70 1 . I was able to obtain numerous constitutive BiP 2 antisense transgenic lines. However, I observed lower transformation efficiency in both constitutive and inducible BiP 2 antisense transgenic lines suggesting antisensing BiP protein in the ER is also detrimental to plants, but not as severe as antisensing Hsc70 1 . Overexpression of BiP 2 resulted in significant increase in the level of total cellular BiP protein. Germination assays following heat shock showed increased thermotolerance in imbibed seed of BiP 2 overexpression lines. This trait was tightly associat ed with increased BiP protein in imbibed seed. A similar thermotolerance assays with whole plants of the same transgenic lines also showed some degree of thermotolerance, but was not as tightly associated with increased BiP protein level as in imbibed see ds. Analyses of the effect of tunicamycin treatment on hypocotyl elongation and disease progression did not reveal any concrete differences between the transgenic lines and wild type plants. It is important to determine what aspect(s) of cell metabolism in imbibed seed is(are) protected by increased BiP level in future studies.
93 CHAPTER 5 MICROARRAY ANALYSIS OF GENE EXPRESSION IN RESPONSE TO HIGH AND LOW TEMPERATURE SHOCK Plant responses to temperature stress are complex, and involve various biochemical and physiological mechanisms (Gilmour et al., 2000; Guy, 1999; Nover and Scharf, 1997; Shinozaki and Yamaguchi Shinozaki, 2000; Thomashow, 2001; Weis and Berry, 1988). Hence, tolerance of plants to temperature extremes results from the activities of multiple genes. Molecular approaches have advanced our appreciation of the complexity of s tress responses, and have provided detailed information most frequently on a single gene basis, one gene at a time. Still this has done very little to provide robust insight into the totality of the molecular basis of stress induced tolerance mechanisms ( Thomashow, 1998). To address this challenge, new approaches are needed that will enable us to assess the global nature of gene expression and the regulation of the genes related to stress responses during stress induction of acquired tolerance. Microarra y technology provides one such approach (Schenk et al., 2000; Seki et al., 2001). It is highly probable that stress resulting from one form of temperature extreme is not different from that resulting from the opposite temperature extreme. I already know t hat many stress genes are responsive to more than one type of stress. For example, low temperature exposure leads to increased expression of dehydrins (Zhu et al., 2000). Dehydrins represent a large and diverse class of proteins that were first associate d with water deficits in plants (Close et al., 1989; Zhu et al., 2000). The finding of dehydrins linked with drought and cold stress provided a strong basis for the commonality of
94 mechanisms of osmotic stresses in both these stresses. Similarly, low temp erature leads to the induction of hsp70s in plants (Li et al., 1999; Sung et al., 2001). Given that hsp70s are molecular chaperones with many functions in protein metabolism, it is highly probable that low temperature affects some aspect of protein metabo lism that elicits hsp70 induction (Li et al., 1999). Thus, it is very likely that different stresses exert their deleterious effects on plant cells through a number of mechanisms. Some of these mechanisms may be common to many different stresses. Theref ore, it is important to know those responses that are specific to a given stress, and it is equally important to know those that occur in response to many forms of stress. Microarray technology offers a means to reveal, on a global basis, the transcriptio nal responses that occur in response to a stress as well those that are shared with different forms of abiotic stress. One approach of DNA microarray analysis is based on hybridization of cDNA reverse transcribed from sample RNA to a high density array of immobilized target sequences each corresponding to a specific gene. Two sample mRNAs are each labeled with one of two fluorescent nucleotides (usually Cy3 dCTP or Cy5 dCTP) by reverse transcription and mixed on a one to one basis prior to hybridization to the array. This sets up a competitive and quantitative hybridization of two species of fluorescently labeled cDNAs to the same DNA sequence in the array. This allows quantitative comparison of gene expression in both samples on a large scale that conven tional gene expression analytical methods like Northern blotting can not embrace. However, microarray technology is still early in its development. Many techniques are being developed to advance DNA microarray analysis such as gene specific DNA synthesis or dispensing techniques, developing a better matrix for DNA arraying, and ensuring that all
95 genes present in the genome are represented (van Hal et al., 2000). Alternatively, the expression of a subset of the genome can be analyzed in order to elucidate the regulation of a group of genes that is either highly conserved such as cytochrome P450 family (Xu et al., 2001), or involved in certain aspects of metabolism such as diurnal and circadian regulated genes (Schaffer et al. 2001) with a minimum of cross contamination of signal. One of the early microarray analyses of gene expression was carried out with 45 genes (Schena et al., 1995). Since then, the number of microarray studies reported in the literature has grown substantially, and many microarray stud ies have been conducted with plants. Expression patterns have been monitored in response to drought and cold stress (Seki et al., 2000), defense responses to pathogen challenge (Desikan et al., 2001; Maleck et al., 2000; Schenk et al., 2000;), seed develo pment (Girke et al., 2000), and mechanical wounding and insect feeding (Reymond et al., 2000). All these microarray analyses have been successful in opening new horizons in understanding the role of differential gene expression in complex biochemical and physiological process. In all these cases, a large number of genes co regulated in response to corresponding treatments were identified at a scale that was simply not possible with conventional RNA blot or RT PCR analysis. By identifying large numbers of co regulated genes new avenues of discovery become possible. For example, a novel promoter element was identified from the group of genes commonly regulated during defense response (Maleck et al., 2000). Also, cross talk and overlaps of signal transduct ion pathways have been more easily elucidated (Reymond et al., 2000; Schenk et al., 2000; Seki et al., 2000). It also helped lead to discovery of genes with strong seed specific promoters that will be valuable in crop production (Girke et al., 2000).
96 The primary objective of this research is to identify genes active in both high and low temperature stress induced tolerance. The secondary objective is to deduce the biochemical and molecular processes responsive to temperature changes and those that are inv olved in the development of acquired tolerance for each of the two temperature extremes. Materials and Methods Plant Growth and Harvest Arabidopsis (ecotype Columbia) plants were grown in a commercial soil mix (Fafard mix No. 2) containing Canadian spha gnum peat, perlite and vermiculite. Plants were grown at 20C with a photoperiod of 15 hr light/9 hr dark in growth cabinets. The irradiance was approximately 150 molm 2 s 1 at canopy height and was provided by incandescent bulbs and cool white fluores cent tubes. Plants were watered every third day and fertilized once a week with a commercial fertilizer (Peter's 20 20 20). Based on Q 10 considerations, plants were exposed to 4C for 12 hr, or for 40C for 1 hr. These represent approximately equivalent temperature/time treatments metabolically with respect to the control growth conditions and circadian clock programming of plants. For Arabidopsis, acquired thermotolerance is usually approaching near maximal levels at 40C after 1 hr, while freezing tol erance is just beginning to increase by 12 hr at 4C. Control plants were kept at 20C ( Figure 5 1 ). Temperature treatment was initiated 2 hours after the onset of the light period in order to harvest all the samples within the light period. Samples wer e harvested and flash frozen in liquid nitrogen and stored at â€“ 80C until processed for RNA extraction.
97 A. 42 44 46 48 Hrs at 40C LT 50 C 0 .25 .50 1.0 1.5 2.0 B. 12 10 8 6 4 2 0 Hrs at 5C LT 50 ? C 12 48 96 192 384 0 12 Figure 5 1. Induction of thermotolerance in Arabidopsis. A) Induction of acquired thermotolerance. B) Induction of acquired freezing tolerance. The shaded area indicates the time point when the RNA samples for microarray analysis were harvested. Dale Haskell, in our laboratory, performed the acquired tolerance experiments and the data have been used with his consent.
98 RNA Isolation, Microarray Analysis, and RT PCR Tissue samples were ground in liquid nitrogen and total RNA isolated according to manufacturerâ€™s protocol using Trizol (Gibco BRL). The amount of total RNA was determined by UV spectrophotometry. Three independently isola ted total RNA samples from each treatment were combined to yield average expression levels in the microarray analyses. Two different microarrays were used in this study: a proprietary array that contained 10,000 cDNA clones (Academic Program, Arabidopsis EST Microarray DNA Chip Plant Biology Research, Monsanto Company, St Louis, MO), and a public array (IncyteGenomics, Inc., Palo Alto, CA) that contained 6,000 cDNA sequences. Hybridizations with the proprietary array were done in duplicate and only those results consistent in both sets of hybridizations were used in further analyses. The hybridization to the public array was done only once. Colleagues at Monsanto Company performed the fluorescent labeling of cDNAs, hybridization, and collected initial fl uorescent data. Results Verification of Microarray Data by RT PCR During the exposure to mild temperature stress, plants develop tolerance to subsequent more severe temperature stresses. Such induced tolerances are called acquired thermotolerance and acq uired freezing tolerance. Arabidopsis develops tolerance rather quickly upon exposure to temperature stresses. Figure 5 1 shows the kinetics of development of such tolerances. Within one hour of incubation at 40C, Arabidopsis already engaged developmen t of acquired thermotolerance and showed a 2C increase in LT 50 (temperature that kills 50% of plants). Arabidopsis still continued to
99 develop acquired thermotolerance after two hours at 40C, the longest incubation administered in this experiment. On t he other hand, Arabidopsis began to show acquired freezing tolerance as early as after 6 hours at 4C and reached the maximum freezing tolerance after 96 hours at 4C. Then it lost some of acquired freezing tolerance. To monitor gene expression at an ear ly stage of induction of thermotolerance, the time points, 1 hour at 40C and 12 hours at 4C, were chosen for the microarray study. RT PCR on selected genes was carried out to confirm the efficacy of the temperature treatments that had been given to the p lants, and to determine whether the expression data obtained from the microarrays would properly correlate with the expression data from RT PCR analysis. Strong induction of Hsp70 and Hsp17.6 by heat treatment and strong induction of Cor78 by low temperat ure showed that the temperature treatments were effective in imposing the expected thermal stresses on the plants ( Figure 5 2 ). There are some discrepancies in the data from the two analyses, but expression data obtained for selected genes by RT PCR were r easonably consistent with data obtained from the microarrays ( Table 5 1 ). The discrepancies appear to arise from the lack of specificity of probes for homologous gene cross hydridization, and inaccuracy in annotation of the genes used in the DNA array. E ight different cDNA clones of hsp70 genes were used in the Monsanto DNA array; four clones are listed as Hsc70 1 and three as heat shock protein 70 like protein and one as a DnaK homolog. This indicates that the annotation of the sequences in the microarr ay may not be sophisticated enough to specify properly those genes from highly homologous gene families. Nor were the libraries used to create the microarray free of representational bias to certain expression subsets of the
10 0 Table 5 1. Comparison of expression of selected genes obtained from RT PCR and microarray analysis Heat induction Cold induction Microarray RT PCR Microarray RT PCR Hsc70 1 1.2 a 2.1 b 1.0 2.1 Hsp70 NA 30.6 NA 7.8 Cor78 0.8 1.0 9.3 6.3 Hsp17.6 10 NC 1.0 NC a Induction ratio was calculated as the microarray signal obtained following heat treatment/control or cold treatment/control. The values are the average of all the microarray data with the same annotation. b Induction ratio was obtained from Figure 5 2 by normalizing the signal ratios from 4 and 40C to the signal ratio from 20C. NA: Not Available. Expression data is not available in microarray analysis. NC: Not Calculated. Strong induction of Hsc17.6 was also observed in RT PCR analysis (see Figure 5 2 ), but induction fold was not calculable due to the absence of the signal in the control (20C).
101 Hsc70 1 Hsp70 Cor78 Hsp17.6 5 4 3 2 1 5 4 3 2 1 5 4 3 2 1 5 4 3 2 1 20C 4C 40C Figure 5 2. RT PCR verification of microarray results. The arrows indicate RT PCR amplicons of 18S rRNA. A segment of 18S rRNA was co amplified in the same RT PCR re action tube as an internal loading control. The bar graph next to each gel picture represents the ratio of the intensity of gene specific RT PCR amplicon over 18 rRNA RT PCR amplicon.
102 genome. In fact, the libraries used in the proprietary microarray wer e constructed from root (a Monsanto internal library 22), stem (a Monsanto internal library 23), flower buds (a Monsanto internal library 24), open flower (a Monsanto internal library 25), leaf (a Monsanto internal library 35), green seedling (a Monsanto i nternal library 146), and Euglena gracilus ( a Monsanto internal library 147). Approximately 56% of the cDNA clones in the array came from the two flower libraries showing a strong emphasis on the reproductive phase of the plant. Incyte's Arabidopsis GenB ank microarray contained over 4,500 unique annotated gene clusters and 1,500 unannotated EST clusters. Genes from all major Arabidopsis thaliana tissues are represented: roots, rosettes, and inflorescences. The m icroarray data for these eight hsp70 clone s did not properly represent changes in gene expression of individual hsp70 genes ( Table 5 2 ) as accurately as gene specific RT PCR did (see Figure 3 5 ). This is probably because nonspecific cross hybridization of probes to these eight hsp70 clones in the microarray obscured any changes in expression signals that occurred during the temperature treatments. Therefore, the array system used in this experiment has room for improvement on several counts: annotation, gene representation, gene specificity, and total genic content. Nevertheless, the data from this array experiment resulted in an unprecedented global view of gene expression during high and low temperature shock. Gene Expression during Heat Treatment The proprietary array used in this study contai ns 10,000 clones that would have consisted of approximately 40% the genome if all the clones had been completely independent gene sequences. Yet, only the signals of 5,636 clones were detected
103 Table 5 2. Microarray data for Arabidopsis hsp70 genes M onsanto clone ID Annotation Heat induction Cold induction ARABL1 043 Q1 E1 E5 Hsc70 1 1.4 0.4 LIB25 009 Q1 E1 B5 Hsc70 1 1 1.4 LIB25 010 Q1 E1 H2 Hsc70 1 1 1.1 LIB25 014 Q1 E1 C8 Hsc70 1 1.4 1 LIB22 077 Q1 E1 B6 Heat shock protein 70 like protein 1.2 1 LIB25 056 Q1 E1 H6 Heat shock protein 70 like protein 1.1 1.1 LIB35 053 Q1 E1 E10 Heat shock protein 70 like protein 1.2 1.4 LIB23 017 Q1 E1 B11 DnaK homolog 0.6 0.7
104 following heat treatment. This is little more than half of the array showing det ectable signal under both treatment conditions. Figure 5 3 shows the expression ratio of genes by heat treatment. Each datapoint represents the expression ratio (heat treated to control) of a gene. I arbitrarily used two fold induction/repression lines to separate heat induced or repressed genes from the others; two fold induction/repression is a common limit in many microarray studies. The points above the lines are genes that are induced and below the line are genes that are repressed. The majority o f genes contained in the array did not show obvious changes in gene expression ( Figure 5 3 ) . Ninety six genes were induced during heat shock out of 5,636. That is 1.7% of all the detectable clones in the array, and 1.0% of all the clones in the array. T he number of genes repressed during heat treatment was 307, which is 5.4% of all detectable clones, and 3.1% of all the clones in the array ( Table 5 3 ). Approximately three times as many genes were repressed than induced during heat shock. The genes that were induced or repressed were categorized into 10 arbitrary groups: unknown identity/function, antioxidant, membrane, RNA/protein synthesis, photosynthesis, metabolism, HS/chaperone/protease, transcription factor (TF)/signal transduction, cold shock prote in (CSP), and other representing genes that encode proteins not belonging to any of the other categories. The unknown group represents genes with no functional information. The antioxidant category represents genes related with antioxidant metabolism. T he membrane group represents genes associated with membranes, or that have a membrane related function. The RNA/protein synthesis group represents genes involved in RNA and protein synthesis except for the genes encoding
105 Figure 5 3. Global view o f gene expression during heat shock (HS). Y axis represents the fluorescent signal from heat shock samples. X axis represents the fluorescent signal from control samples. The fluorescent signals from control and heat shock sample designate a single data point in the graph. The data points that fall onto or near the diagonal line represent the genes showing no change during heat shock. The data points in the pink area represent the genes induced two fold or more during heat shock. The data points in bl ue area represent genes repressed two fold or more during heat shock.
106 Table 5 3. Microarray results for the proprietary array. Heat Shock Cold Shock Number Percent a Percent b Number Percent a Percent b Induced 96 1.7 1.0 128 1.9 1.3 Repressed 307 5 .4 3.1 1393 21.0 13.9 Unchanged 5233 92.8 52.3 5113 77.1 51.1 Total 5636 100 56.4 6634 100 66.3 Percent a percent of the detectable cDNA clones in the array. Percent b percent of all the cDNA clones in the array.
107 transcription factors. The photosynthe sis category represents genes involved in photosynthesis. Some of the genes grouped in the photosynthesis category are also membrane proteins. They were placed in the photosynthesis group when they had an obvious function related to photosynthesis. The metabolism grouping includes genes involved in cellular metabolism except for the structural genes encoding actins, tublins, and histones. The HS/chaperone/protease grouping represents genes encoding heat shock proteins, chaperones, and proteases. The TF /signal transduction category represents genes encoding transcriptional factors, protein kinases, protein phosphatases, and receptor proteins that are thought to be involved in any signal transduction pathway. The CSP group represents cold stress proteins . The percentage of heat induced or heat repressed genes for each group is presented in pie charts ( Figure 5 4 & 5 5 ). The metabolism, photosynthesis, TF/signal tansduction groups contained approximately equal percentages of genes that were either induced or repressed. The antioxidant, CSP, HS/chaperone/protease, and membrane group contained three times greater percentage in the total heat induced genes than the percentage in the total heat repressed genes indicating the functions of many genes in these g roups are required to cope with high temperature. On the contrary, the RNA/protein synthesis group was represented by three times greater percentage in the total heat repressed genes than the percentage in the total heat induced genes showing engagement o f transcriptional and/or translational repression of gene expression during heat treatment. The 10 genes most significantly induced by heat shock are a fibrillin homolog (induction fold; 20X), a lipocalin homolog (11X), a 26S proteasome associated pad1
108 Heat Induced (96) Unknown 34% Antioxidant 3% Membrane 11% RNA/Protein synthesis 5% Photosynthesis 3% Metabolism 17% HS/Chaperone / Protease 14% TF/Signal transduction 6% Other 3% CSP 3% Figure 5 4. Functional distribution of heat induced genes in the array. The percentages are based on the total number of heat induced genes.
109 Unknown 39% Antioxidant 0% Membrane 4% RNA/Protein synthesis 17% Photosynthesis 4% Metabolism 18% HS/Chaperone / Protease 4% TF/Signal transduction 6% Other 7% CSP 1% Heat Repressed (307) Figure 5 5. Functional distribution of heat repressed genes in the array.
110 homolog (8X), plasma mem brane intrinsic protein 2A (8X), RNA helicase (8X), Hsp81 2 (8X), an axi 1 like protein (6X) and three proteins of unknown identity or function. Fibrillin is a carotenoid associated protein and known to accumulate during conditions of oxidative stress suc h as high light with low temperature, gamma irradiation or methyl viologen treatment that leads to accumulation of active oxygen species (Langenkamper et al., 2001). Lipocalin is a violaxanthin de epoxidase and a zeaxanthin epoxidase that catalyze the int erconversions between the carotenoids violaxanthin, antheraxanthin and zeaxanthin and thereby dissipates excess light energy that would damage the photosynthetic system (Hieber et al., 2000). Pad1 is a component of 20S catalytic core structure of 26S prot easome and known to interact with an ubiquitin ligase (Farras et al., 2001). Axi 1 like protein activates RUB1, an ubiquitin like protein in conjunction with ECR1 (del Pozo and Estelle, 1999). Considering the protective roles of these proteins against pr otein denaturation in cellular metabolism, it suggests that heat shock may result in two major cellular lesions in plants; protein denaturation and generation of over energization of photosynthetic electron transport that generate active oxygen species. H eat shock probably perturbs either membrane integrity or stability of components of electron transport in the photosystem. Heat shock also denatures heat labile proteins hence resulting in induction of 26S proteasome related proteins along with lipocalin and fibrillin. Gene Expression during Cold Treatment The freezing tolerance assay of Arabidopsis following incubation at 4C showed that Arabidopsis began to acquire induced freezing tolerance as early as 6 hours at 4C ( Figure 5 1 ). Arabidopsis plants were incubated for 12 hours at 4C for this study. Cellular mechanisms involved in developing freezing tolerance would be engaged at this
111 time of microarray analysis. Contrary to gene expression during heat shock, the global view of gene expression durin g cold treatment revealed a broad spectrum up shift off the 1:1 diagonal ( Figure 5 6 ). Whether this up shift reflects an actual induction of gene expression during cold treatment or an artifact of the technique has not been determined. A time course micr oarray analysis would help clarify this. The total number of cold repressed genes made up 21% of all the detectable cDNA (6,634) clones and 13% of all the cDNA clones (10,000) in the array. The cold repressed genes constituted 10 times more genes in the array than cold induced genes ( Table 5 3 ). This suggests that the global repression of transcription is also a major part of the systematic response to cold shock. The CSP group constitutes 10% of the total cold induced genes. Also, four of the ten most strongly induced genes by cold shock belong to this group: cor47 (12X), ERD10 (10X), Cor78 (9X), and Cor15a (8X). These genes encode dehydrin type proteins. The specific cellular functions of these proteins have not been identified, but they are implicat e d in membrane stabilization during a freezing and thaw cycle (personal comm., Tim Close). Another interesting aspect of plant response to cold shock is induction of amylase (13X) and patatin like protein (8X). Beta amylase catalyzes the breakdown of a 1,4 glycosidic linkages in polyglucan at the non reducing end to produce maltose as its product. The purpose of induction of amylase gene by cold shock is not known, but it is suspected that its product, maltose, may act as a compatible solute in the ce ll during cold shock (personal comm., Fatma Kaplan). Patatin is a storage protein initially found in potato tuber and it constitutes 40% of total soluble potato tuber protein. Unlike other storage proteins, patatin has lipid acyl hydrolase and acyl trans ferase activities (Racusen,
112 Figure 5 6. Global view of gene expression during cold shock (CS).
113 1984; Andrews et al., 1988; Rosahl et al., 1987) and is implicated in membrane degradation during stresses such as wounding or pathogen attack (Senda et al., 1996). Patatin like protein induced by cold shock may participate in a salvage process of damaged membrane components during cold shock. The metabolism and TF/signal transduction group constituted 29% and 9% of the total cold induced genes and 20 %, 6% of the total cold repressed genes, respectively. This result supports what has long been known, that some aspects of metabolism and signal transduction pathways are actively involved in acclimation and developing freezing tolerance. The membrane an d RNA/protein synthesis groups make up a greater percentage in the total cold repressed genes than in cold induced genes reflecting a downward adjustment of RNA/protein synthesis and membrane biogenesis during cold treatment. The antioxidant, HS/chaperone /protease, photosynthesis groups contained approximately equal percentage of genes in the cold induced and the cold repressed categories ( Figures 5 7 & 5 8 ). Some of the most strongly cold induced genes were amylase, cold stress proteins (Cor78/RD29A, Cor15A, ERD10, Cor6.6), ascorbate/glutathione pathway related proteins (phospholipid hydroperoxide glutathione peroxidase), photosynthesis (Lhb1b2) and ribosomal proteins (L17, L7A). This re sult suggests that carbohydrate metabolism, ascorbate/glutathione pathway, photosynthesis, and protein translation were strongly affected by low temperature. It seems that the photosynthesis group was equally represented in the cold induction and cold rep ression profile. However, the listing reveals that different genes were represented in the two profiles suggesting the presence
114 Cold Induced (128) Unknown 35% Antioxidant 1% Membrane 2% RNA/Protein synthesis 7% Photosynthesis 2% Metabolism 29% HS/Chaperone / Protease 4% TF/Signal transduction 9% Other 2% CSP 10% Figure 5 7. Functional distribution of cold induced genes in the array.
115 Unknown 40% Antioxidant 0% Membrane 5% RNA/Protein synthesis 11% Photosynthesis 3% Metabolism 20% HS/Chaperone / Protease 4% TF/Signal transduction 6% Other 10% CSP 0% Cold Repressed (1393) Figure 5 8. Functional distribution o f cold repressed genes in the array.
116 of specific and differential regulation of the genes related to photosynthesis in response to cold shock. Comparison of Gene Expression in Response to Heat and Cold A major objective in the microarray study was to sea rch for components of heat and cold stress induced tolerance responses that are common for both forms of stress as well as components that are unique to each form of stress. Obviously, the HS/chaperone/protease grouping makes up a greater percentage in th e total heat induced genes than in the total cold induced genes while the CSP group constitutes a greater percentage in the total cold induced genes than in the total heat induced genes ( Figures 5 4 & 5 7 ). However, more interestingly, the membrane group constitutes a five times greater percentage in the total heat induced genes than it does in the total cold induced genes. This result suggests that the membrane components are affected more by heat shock or are perhaps more susceptible to heat stress. Th e metabolism group constitutes twice as the percentage in the total cold induced genes than in the total heat induced genes. This may reflect an upward adjustment of cellular metabolism to counteract thermodynamic constraints imposed during cold treatment . The RNA/protein synthesis group makes up a slightly greater percentage in the total heat repressed genes (17%) than in the total cold repressed genes (11%). Other than that, the heat and cold repression profiles were very similar ( Figures 5 5 & 5 8 ). T his result indicates that the mechanisms for transcriptional repression in response to heat and cold shock may be similar. Co regulation of the Genes by Heat and Cold Treatment Twelve genes were co induced by heat and cold shock ( Table 5 4 ). This represen ts 12% of the total heat inducible genes and 9% of the total cold inducible genes.
117 Table 5 4. Co induced genes in response to heat and cold shock. Heat Induction Cold Inductio n Category Gene Description Accession Number 5 12 CSP Cold regulated prote in cor47 X59814 5 10 CSP Dehydrin ERD10 D17714 2 2 HS Putative RNA binding protein with RNP2 region AF000657 11 6 Mem Lipoprotein BLC precursor (Lipocalin) U21727 20 3 Photo Fibrillin AF075598 4 2 Photo Lhcb6 AF134130 5 3 R/P Putative ribosomal prot ein L17 AC002332 5 8 UK Putative protein AL030978 4 10 UK MtN19 homolog Y15376 3 2 UK GAST1 homolog NA 3 6 UK NA NA 4 2 UK BAC F5A8 complete sequence AC004146 CSP: cold stress proteins, HS: heat shock proteins/chaperones/proteases, Mem: membrane rel ated proteins, Photo: photosynthesis related proteins, R/P: RNA/protein synthesis related genes, UK: genes with unknown function
118 Approximately, one out of ten heat or cold induced genes was induced by the other temperature shock. This indicates that the re is a relatively small overlap or convergence in cellular mechanisms that are induced during heat and cold shock. Co induction profile of gene expression suggests that there is little overlap between the two temperature signal pathways leading to gene i nduction ( Table 5 4 ). However, the gene repression by both heat and cold shock suggests the opposite prospect. A total of 274 genes were co repressed by heat and cold shock ( Figure 5 9 ). This constitutes 89% of the total heat repressed genes and 20% of the total cold repressed genes. This shows that most of heat repressed genes are also cold repressed while the majority of cold repressed genes are not heat repressed. Since the majority of the heat repressed genes are represented in the co repressed gen es, the composition of the co repressed genes is quite similar to that of the heat repressed genes. Discussion Conservative Response of the Cell to Temperature Stresses In response to temperature shock, Arabidopsis exhibited a greater repression of gene ex pression than induction. These relatively fewer induced genes may play critical roles in helping the plant cope with temperature stresses. However, the repression of the genes is a more prevalent response to both forms of temperature stress. The prevale nt repression in response to heat and cold shock may be specific to temperature stresses. When the growth temperature of a human pathogen (Group A Streptococcus ) was altered from 37C to 29C, 57% of the total genes in microarray were repressed (Smoot et al., 2000). However, different types of treatments related to plant disease response such as a pathogen, salicylic acid, methyl jasmonate, and ethylene all resulted in a smaller number
119 Unknown 38% Antioxidant 0% Membrane 4% RNA/Protein synthesis 18% Photosynthesis 4% Metabolism 18% HS/Chaperone / Protease 4% TF/Signal transduction 7% Other 8% CSP 1% Co Repressed (274) Figure 5 9. Functional distribution of co repression genes in t he array.
120 of repressed genes than that of induced genes in a microarray study (Schenk et al., 2000). It seems that the normal rate of cellular metabolism itself becomes inhibiting or detrimental to cell survival under stressful conditions probably by g enerating more harmful by products. Hence, cells slow down metabolism to make less stress labile components or harmful by products of metabolism. Nevertheless, during cold treatment, the cell induces many genes involved in metabolism in order to cope wit h metabolic constraints imposed by low temperature. Possible Temperature Labile Components of the Cell High temperature denatures proteins, and destabilizes membrane structures by increasing fluidity of the membrane (Weis and Berry, 1988). Low temperature decelerates cellular metabolism and restricts movement and interaction of membrane proteins by decreasing membrane fluidity (Quinn, 1988). These two forms of temperature stress seemingly affect cellular metabolism in opposite ways. However, microarray g ene expression data suggest that the cellular stresses generated by two opposite forms of temperature stress affect common cellular components. The two forms of temperature stress induced expression of the genes encoding lipocalin and fibrillin ( Table 5 4 ). These two genes are strongly induced by accumulation of reactive oxygen species (Hieber et al., 2000; Langenkamper et al., 2001; Manach and Kuntz, 1999). This suggests that the destabilization of membrane structures, either by increasing or decreasing the fluidity of the membrane, resulted in perturbation of electron transport thereby generating reactive oxygen species. It can be extrapolated from this result that the membrane is an important site of reception of temperature stress signals, and the pr oduction of reactive oxygen species. This is perhaps an early and possibly major cellular lesion caused by temperature stresses.
121 The genes related to photosynthesis constitute fairly small and equivalent portion in every gene expression chart rendered in this study. Nonetheless, the components of the photosynthetic apparatus are very sensitive to temperature stresses because volatile photochemical reactions that can easily be affected by temperature change are taking place in the photosystems and because most of the photosystem components are embedded in the membranes which are susceptible to temperature stress. A closer look at the temperature responsive genes in photosynthesis suggests a bi modal regulation of gene expression under non optimal conditions for photochemical reactions rendered by temperature stress. The cell downsizes the photosynthetic systems by gene repression and either protects the photosystems or replaces the damaged components by gene induction. During both heat shock and cold shock , the expression of the following genes was repressed: a chlorophyll biogenesis related gene (protochlorophyllide reductase), the components of the oxygen evolving center (33 kDa polypeptide of oxygen evolving complex, a putative protein 1 photosystem II o xygen evolving complex), plastocyanin, and accessory proteins (Lhb1b1, Lhcb2). On the other hand, the genes encoding fibrillin and Lhcb6 were induced during the both heat and cold shock. The expression of approximately one quarter of the Arabidopsis gen ome was detected and analyzed in this study. A biased library preparation and poor annotation of the array in this study warrants further improvement. Nonetheless, the global gene expression analysis using this microarray revealed much new information. The expression patterns in response to heat and cold shock revealed the small overlap between heat and cold signal transduction pathway leading to gene induction, but a large
122 commonality in the two signal transduction pathways leading to gene repression. Strong induction of the genes related to photosystem protectants such as lipocalin and fibrillin by heat and cold shock suggest the membrane structure, photosystems in particular, is a common and primary target for temperature stresses. New information fr om this microarray study also opens an avenue of research on the gene expression of amylase in response to heat and cold shock.
123 CHAPTER 6 CONCLUSION Fourteen members of the Arabidopsis hsp70 gene family have been identified by the genome sequencing project. High sequence homology and promiscuous substrate affinities of hsp70s make it difficult to assign specific physiological roles to ind ividual members. Comprehensive genomic analysis revealed characteristics unique to each subfamily. A subfamily is defined as members of the same cluster determined by sequence analysis. The same subcellular location was predicted for the members in the same subfamily. Gene structures and C terminal motifs were also found to be unique to the subfamily. RT PCR analysis on 12 members of the gene family revealed the complex expression patterns of individual hsp70 genes. Most notably, the expression of Hs p70b was only induced by heat shock and was not detected in all the developmental stages, heat, and cold stress conditions I tested. Expression patterns of individual members were different even among the members of the same subfamily, for example, among the five cytosolic members. Cytosolic and mitochondrial members are more responsive to temperature stress than the other subfamilies. The expression data along with data from genomic analysis established the basis for assigning more detailed and defined roles of individual hsp70s in plant metabolism. A transgenic approach was used to alter the expression of two members, Hsc70 1 and BiP 2, in order to assess their roles in plant metabolism. Constitutive
124 underexpression of Hsc70 1 is not permitted in Ara bidopsis. However, constitutive overexpression of Hsc70 1 is permitted, but the recovery of the transgenic plants was very difficult. The regulation of ER luminal BiP is less stringent yielding several antisense and numerous sense lines. These results i ndicate that the regulation mechanism for each hsp70 is different. Constitutive overexpression of BiP 2 resulted in increased thermotolerance of seed, but did not change thermotolerance of the whole plant indicating the role of BiP 2 is accentuated during germination. In addition to the hsp70 studies, microarray analysis was conducted to monitor changes in gene expression in response to temperature extremes. The results showed that 12 genes were co induced and 274 genes were co repressed during heat and cold stress. Strong induction of the genes involved in protection of the photosystems and salvage of membranes indicate that the membrane components and photosynthetic apparatus are primary targets for injury from heat and cold stress. I also have iden tified a total of 10 T DNA inserted knockout lines for two cytosolic hsc70 and three BiP genes ( Figure 6 1 ). Analyzing the phenotypic changes resulting from the knockout of individual hsp70 genes will help elucidate the function of the individual hsp70 pr oteins in Arabidopsis.
125 BiP 1 BiP 2 BiP 3 Hsc70 2 2269 2783 2376 Hsc70 1 2279 2609 BiP 1 BiP 2 BiP 3 Hsc70 2 2269 2783 2376 Hsc70 1 2279 2609 Figure 6 1. Knockout locations for two cytosolic hsc70 genes and three BiP genes. The black and gray triangles indicate the region of T DNA insertion in the gene. Plants containing T DNA insertions indicated by the black triangles were further pursued and homozygous lines are being developed from them.
126 LIST OF REFERENCES Almoguera C, Prieto Dapena P, Jordano J (1998) Dual regulation of a heat shock promoter during embryogenesis: stage dependent role of heat shock elements. Plant J 13: 437 446 Alvim FC, Carolino SM, Cascardo JC, Nunes CC, Martinez CA, Otoni WC, Fontes EP (2001) Enhanced accumulation of BiP in transgenic plants confers tolerance to water stress. Plant Physiol 126: 1042 1054 Anderson JV, Li QB, Haskell DW, Guy CL (1994) Structural organization of the spinach endoplasmic reticulum luminal 70 kilodalton heat shock cognate gene and expression of 70 kilodalton heat shock genes during cold acclimation. Plant Physiol 104: 1359 1370 Andrews DL, Beames B, Summers MD, Park WD (1988) Characterization of the lipid acyl hydrolase activity of the maj or potato ( Solanum teberosum ) tuber protein, patatin, by cloning and abundant expression in a baculovirus vector. Biochem J 252: 199 206 Bent AF, Kunkel BN, Dahlbeck D, Brown KL, Schmidt R, Giraudat J, Leung J, Staskawicz BJ (1994) RPS2 of Arabidopsis tha liana : a leucine rich repeat class of plant disease resistance genes . Science 265: 1856 1860 Bimston D, Song J, Winchester D, Takayama S, Reed JC, Morimoto RI (1998) BAG 1, a negative regulator of Hsp70 chaperone activity, uncouples nucleotide hydrolysis from substrate release. EMBO J 17: 6871 6878 Boice JA, Hightower LE (1997) A mutational study of the peptide binding domain of Hsc70 guided by secondary structure prediction. J Biol Chem 272: 24825 24831 Bonk M, Tadros M, Vandekerckhove J, Al Babili S, B eyer P (1996) Purification and characterization of chaperonin 60 and heat shock protein 70 from chromoplasts of Narcissus pseudonarcissus. Involvement of heat shock protein 70 in a soluble protein complex containing phytoene desaturase. Plant Physiol 111: 931 939 Bonner JJ, Carlson T, Fackenthal DL, Paddock D, Storey K, Lea K (2000) Complex regulation of the yeast heat shock transcription factor. Mol Biol Cell 11: 1739 1751 Boorstein WR, Ziegelhoffer T, Craig EA (1994) Molecular evolution of the hsp70 mu ltigene family. J Mol Evol 38: 1 17
127 Boston RS, Fontes EB, Shank BB, Wrobel RL (1991) Increased expression of the maize immunoglobulin binding protein homolog b 70 in three zein regulatory mutants. Plant Cell 3: 497 505 Brewer J, Cleveland J, Hendershot L (1997) A pathway distinct from the mammalian unfolded protein response regulates expression of endoplasmic reticulum chaperones in non stressed cells. EMBO J 16: 7207 7216 Brodsky JL (1996) Post translational protein translocation: not all hsc70s are cre ated equal. Trends Biochem Sci 21: 122 126 Bush GL, Meyer DI (1996) The refolding activity of the yeast heat shock proteins Ssa1 and Ssa2 defines their roles in protein translocation. J Cell Biol 135: 1229 1237 Cascardo JCM, Almeida RS, Buzeli RAA, Carol ino SMB, Otoni WC, Fontes EPB (2000) The phosphorylation state and expression of soybean BiP isoforms are differentially regulated following abiotic stresses. J Biol Chem 275: 14494 14500 Chong KY, Lai CC, Lille S, Chang C, Su CY (1998) Stable overexpress ion of the constitutive form of heat shock protein 70 confers oxidative protection. J Mol Cell Cardiol 30: 599 608 Close TJ, Kortt AA, Chandler PM (1989) A cDNA based comparison of dehydration induced proteins (dehydrins) in barley and corn. Plant Mol Bio l 13: 95 108 Craig EA, Jacobsen K (1984) Mutations of the heat inducible 70 kilodalton genes of yeast confer temperature sensitive growth. Cell 38: 841 849 Craig EA, Jacobsen K (1985) Mutations in cognate genes of Saccharomyces cerevisiae hsp70 result in reduced growth rates at low temperatures. Mol Cell Biol 5: 3517 3524 Czarnecka E, Key JL, Gurley WB (1989) Regulatory domains of the Gmhsp17.5 heat shock promoter soybean. Mol Cell Biol 9: 3457 3463 D'Amico L, Valsasina B, Daminati MG, Fabbrini MS, Nitt i G, Bollini R, Ceriotti A, Vitale A (1992) Bean homologs of the mammalian glucose regulated proteins: induction by tunicamycin and interaction with newly synthesized seed storage proteins in the endoplasmic reticulum. Plant J 2: 443 455 Davis KR, Schott E, Ausubel FM (1991) Virulence of selected phytopathogenic pseudomonads in Arabidopsis thaliana . Mol Plant Microbe Interact 4: 477 488 Deloche O, Georgopoulos C (1996) Purification and biochemical properties of Saccharomyces cerevisiae's Mge1p, the mitoch ondrial cochaperone of Ssc1p. J Biol Chem 271: 23960 23966
128 del Pozo JC, Estelle M (1999) The Arabidopsis cullin AtCUL1 is modified by the ubiquitin related proein RUB1. Proc Natl Acad Sci USA 96: 15342 15347 Demand J, Alberti S, Patterson C, Hohfeld J (20 01) Cooperation of a ubiquitin domain protein and an E3 ubiquitin ligase during chaperon/proteasome coupling. Curr Biol 11: 1569 1577 Denecke J, Carlsson LE, Vidal S, Hoglund AS, Ek B, van Zeijl MJ, Sinjorgo KMC, Palva ET (1995) The tobacco homolog of mam malian calreticulin is present in protein complexes in vivo . Plant Cell 7: 391 406 Denecke J, Goldman MHS, Demolder J, Seurinck J, Botterman J (1991) The tobacco luminal binding protein is encoded by a multigene family. Plant Cell 3: 1025 1035 DeRocher A , Vierling E (1995) Cytoplasmic HSP70 homologues of pea: differential expression in vegetative and embryonic organs. Plant Mol Biol 27: 441 456 Deshaies RJ, Koch BD, Werner Washburne M, Craig EA, Schekman R (1988) A subfamily of stress proteins facilitate s translocation of secretory ahd mitochondrial precursor polypeptides. Nature 332: 800 805 Desikan R, A H Mackerness S, Hancock JT, Neill SJ (2001) Regulation of the arabidopsis transcriptome by oxidative stress. Plant Physiol 127: 159 172 Dorner AJ, K rane MG, Kaufman RJ (1988) Reduction of endogenous GRP78 levels improves secretion of a heterologous protein in CHO cells. Mol Cell Biol 8: 4063 4070 Drzymalla C, Schroda M, Beck CF (1996) Light inducible gene HSP70B encodes a chloroplast localized heat s hock protein in Chlamydomonas reinhardtii. Plant Mol Biol 31: 1185 1194 Duck N, McCormick S, Winter J (1989) Heat shock protein hsp70 cognate gene expression in vegetative and reproductive organs of Lycopersicon esculentum . Proc Natl Acad Sci USA 86: 3674 3678 Dudley P, Wood CK, Pratt JR, Moore AL (1997) Developmental regulation of the plant mitochondrial matrix located HSP70 chaperone and its role in protein import. FEBS Lett 417: 321 324 Farras R, Ferranoo A, Jasik J, Kleinow T, Okresz L, Tiburcio A, S alchert K, del Pozo C, Schell J, Koncz C (2001) EMBO J 20: 2742 2756 Feder JH, Rossi JM, Solomon J, Solomon N, Lindquist S (1992) The consequences of expressing Hsp70 in Drosophila cells at normal temperatures. Genes Dev 6: 1402 1413
129 Feder ME, Cartano NV , Milos L, Krebs RA, Lindquist SL (1996) Effect of engineering Hsp70 copy number on Hsp70 expression and tolerance of ecologically relevant heat shock in larvae and pupae of Drosophila Melanogaster . J Exp Biol 199: 1837 1844 Flaherty KM, DeLuca Flaherty C , McKay DB (1990) Three dimensional structure of the ATPase fragment of a 70K heat shock cognate protein. Nature 346: 623 628 Flynn GC, Pohl J, Flocco MT, Rothman JE (1991) Peptide binding specificity of the molecular chaperone BiP. Nature 353: 726 730 Fontes EBP, Shank BB, Wrobel RL, Moose SP, OBrian GR, Wurtzel ET, Boston RS (1991) Characterization of an immunoglobulin binding protein homolog in the maize floury 2 endosperm mutant. Plant Cell 3: 483 496 Freeman BC, Morimoto RI (1996) The human cytosol ic molecular chaperones hsp90 hsp70 (hsc70) and hdj 1 have distinct roles in recognition of a non native protein and protein refolding. EMBO J 15: 2969 2979 Gaitanaris GA, Papavassiliou AG, Rubock P, Silverstein SJ, Gottesman ME (1990) Renaturation of den atured ? repressor requires heat shock proteins. Cell 61: 1013 1020 Galau GA, Hughes DW (1987) Coordinate accumulation of homologous transcripts of seven cotton Lea gene families during embryogenesis and germination. Dev Biol 123: 213 221 Gao BC, Biosca J, Craig EA, Greene LE, Eisenberg E (1991) Uncoating of coated vesicles by yeast hsp70 proteins. J Biol Chem 266: 19565 19571 Geli MI, Torrent M, Ludevid D (1994) Two structural domains mediate two sequntial events in gamma zein targeting: protein endopla smic reticulum retention and protein body formation. Plant Cell 6: 1911 1922 Gillikin JW, Fontes EP, Boston RS (1995) Protein -protein interactions within the endoplasmic reticulum. Methods Cell Biol 50: 309 323 Gilmour SJ, Sebolt AM, Salazar MP, Everard JD, Thomashow MF (2000) Overexpression of the Arabidopsis CBF3 transcriptional activator mimics multiple biochemical changes associated with cold acclimation. Plant Physiol 124: 1854 1865 Girke T, Todd J, Ruuska S, White J, Benning C, Ohlrogge J (2000) M icroarray analysis of developing Arabidopsis seeds. Plant Physiol 124: 1570 1581 Glick BS (1995) Can Hsp70 proteins act as force generating motors? Cell 80: 11 14 Glover JR, Lindquist S (1998) Hsp104, hsp70 and hsp40: A novel chaperone system that rescue s previously aggregated proteins. Cell 94: 73 82
130 Goloubinoff P, Mogk A, Zvi AP, Tomoyasu T, Bukau B (1999) Sequential mechanism of solubilization and refolding of stable protein aggregates by a bichaperone network. Proc Natl Acad Sci USA 96: 13732 13737 Guy C (1999) Molecular responses of plants to cold shock and cold acclimation. J Mol Microbiol Biotechnol 1: 231 242 Guy CL, Li QB (1998) The organization and evolution of the spinach stress 70 molecular chaperone gene family. Plant Cell 10: 539 556 Hal laday JT, Craig EA (1995) A heat shock transcription factor with reduced activity suppresses a yeast HSP70 mutant. Mol Cell Biol 15: 4890 4897 Harmsen MM, Bruyne MI, Raue HA, Maat J (1996) Overexpression of binding protein and disruption of the PMR1 gene synergistically stimulate secretion of bovine prochymprochymosin osin but not plant thaumatin in yeast. Appl Microbiol Biotechnol 46: 365 370 Harris B, Dure L (1978) Developmental regulation in cotton seed germination: polyadenylation of stored messenger RNA. Biochem 8: 3250 3256 Hatano K, Shimada T, Hiraiwa N, Nishimura M, Hara Nishimura I (1997) A rapid increase in the level of binding protein (BiP) is accompanied by synthesis and degradation of storage proteins in pumpkin cotyledons. Plant Cell Physio l 38: 344 351 Helm KW, Vierling E (1989) An Arabidopsis cDNA clone encoding a low molecular weight heat shock protein. Nucl Acids Res 17: 7995 Hendershot LM, Wei JY, Gaut JR, Lawson B, Freiden PJ, Murti KG (1995) In vivo expression of mammalian BiP ATPas e mutants causes disruption of the endoplasmic reticulum. Mol Biol Cell 6: 283 296 Hendershot L, Wei J, Gaut J, Melnick J, Aviel S, Argon Y (1996) Inhibition of immunoglobulin folding and secretion by dominant negative BiP ATPase mutants. Proc Natl Acad S ci USA 93: 5269 5274 Hieber AD, Bugos RC, Yamamoto HY (2000) Plant lipocalins: violaxanthin de epoxidase and zeaxanthin epoxidase. Biochem Biophys Acta 1482: 84 91 Hohfeld J, Jentsch S (1997) GrpE like regulation of the hsc70 chaperone by the anti apopto tic protein BAG 1. EMBO J 16: 6209 6216 Hohfeld J, Minami Y, Hartl FU (1995) Hip, a novel cochaperone involved in the eukaryotic Hsc70/Hsp40 reaction cycle. Cell 83: 589 598
131 Hong SW, Vierling E (2000) Mutants of Arabidopsis thaliana defective in the acqu isition of tolerance to high temperature stress. Proc Natl Acad Sci U S A 97: 4392 4397 Horton LE, James P, Craig EA, Hensold JO (2001) The yeast hsp70 homolgue Ssa1 required for translation and interacts with Sis1 and Pab1 on translating ribosomes. J Bio l Chem 276: 14426 14433 Hsieh K, Wang YC, Lin BL (1998) At hsc70 3 encodes a cytosolic hsp70 in Arabidopsis thaliana . Plant Physiol 117: 1525 Irmer H, Hhfeld J (1997) Characterization of functional domains of the eukaryotic co chaperone hip. J Biol Chem 272: 2230 2235 Ivey RA 3rd, Bruce BD (2000) In vivo and in vitro interaction of DnaK and a chloroplast transit peptide. Cell Stress Chaperones 5: 62 71 James P, Pfund C, Craig EA (1997) Functional specificity among hsp70 molecular chaperones. Science 27 5: 387 389 Jelitto Van Dooren EP, Vidal S, Denecke J (1999) Anticipating endoplasmic reticulum stress. A novel early response before pathogenesis related gene induction. Plant Cell 11: 1935 1944 Jiang J, Ballinger CA, Wu Y, Dai Q, Cyr DM, Hohfeld J, Patt erson C (2001) CHIP is a U box dependent E3 ubiquitin ligase: identification of Hsc70 as a target for ubiquitylation. J Biol Chem 16: 42938 42944 Kalinski A, Rowley DL, Loer DS, Foley C, Buta G, Herman EM (1995) Binding protein expression is subject to te mporal, developmental and stress induced regulation in terminally differentiated soybean organs. Planta 195: 611 621 Kim S, Schilke B, Craig EA, Horwich AL (1998) Folding in vivo of a newly translated yeast cytosolic enzyme is mediated by the SSA class of cytosolic yeast Hsp70 proteins. Proc Natl Acad Sci USA 95: 12860 12865 Ko K, Bornemisza O, Kourtz L, Ko ZW, Plaxton WC, Cashmore AR (1992) Isolation and characterization of a cDNA clone encoding a cognate 70 kDa heat shock protein of the chloroplast env elope. J Biol Chem 267: 2986 2993 Koizumi N (1996) Isolation and responses to stress of a gene that encodes a luminal binding protein in Arabidopsis thaliana . Plant Cell Physiol 37: 862 865 Koizumi N, Sano H (1997) Isolation of two genes (Accession nos. D89341 and D89342) encoding luminal binding protein from Arabidopsis thaliana . Plant Physiol 113: 664
132 Kourtz L, Ko K (1997) The early stage of chloroplast protein import involves Com70. J Biol Chem 272: 2808 2813 Langenkamper G, Manach N, Broin M, Cuine S, Becuwe N, Kuntz M, Rey P (2001) Accumulation of plastid lipid associated proteins (fibrillin/CDSP34) upon oxidative stress, ageing and biotic stress in Solanaceae and in response to drought in other species. J Exp Bot 52: 1545 1554 Leborgne Castel N, J elitto Van Dooren EP, Crofts AJ, Denecke J (1999) Overexpression of BiP in tobacco alleviates endoplamic reticulum stress. Plant Cell 11: 459 470 Lee GJ, Pokala N, Vierling E (1995) Structure and in vitro molecular chaperone activity of cytosolic small he at shock proteins from pea. J Biol Chem 270: 10432 10438 Lee GJ, Vierling E (2000) A small heat shock protein cooperates with heat shock protein 70 systems to reactivate a heat denatured protein. Plant Physiol 122: 189 198 Lee JH, Hubel A, Schffl F (199 5) Derepression of the activity of genetically engineered heat shock factor causes constitutive synthesis of heat shock proteins and increased thermotolerance in transgenic Arabidopsis . Plant J 4: 603 612 Lee JH, Schffl F (1996) An Hsp70 antisense gene a ffects the expression of HSP70/HSC70, the regulation of HSF, and the aquisition of thermotolerance in transgenic Arabidopsis thaliana . Mol Gen Genet 252: 11 19 Li J, Chory J (1998) Preparation of DNA from Arabidopsis . In Martinez Zapater, J. M. and Salina s, J. eds, Methods in Molecular Biology: Arabidopsis Protocols Humana Press, Totowa, NJ, pp 55 60 Li QB, Haskell DW, Guy CL (1999) Coordinate and non coordinate expression of the stress 70 family and other molecular chaperones at high and low temperature in spinach and tomato. Plant Mol Biol 39: 21 34 Li X, Wu Y, Zhang DZ, Gillikin JW, Boston RS, Franceschi VR, Okita TW (1993) Rice prolamine protein body biogenesis: a BiP mediated process. Science 262: 1054 1056 Liu FH, Wu SJ, Hu SM, Hsiao CD, Wang C (19 99) Specific interaction of the 70 kDa heat shock cognate protein with the tetratricopeptide repeats. J Biol Chem 274: 34425 34432 Liu H, Bowes RCI, van de Water B, Sillence C, Nagelkerke JF, Stevens JL (1997) Endoplasmic reticulum chaperones GRP78 and ca lreticulin prevent oxidative stress, Ca 2+ disturbances, and cell death in renal epithelial cells. J Biol Chem 272: 21751 21759
133 Luders J, Demand J, Hohfeld J (2000) The ubiquitin related BAG 1 provides a link between the molecular chaperones Hsc70/Hsp70 a nd the proteasome. J Biol Chem 275: 4613 4717 Luders J, Demand J, Schonfelder S, Frien M, Zimmermann R, Hohfeld J (1998) Cofactor induced modulation of the functional specificity of the molecular chaperone Hsc70. Biol Chem 379: 1217 1226 Lyman SK, Schekm an R (1996) Polypeptide translocation machinery of the yeast endoplasmic reticulum. Experientia 52: 1042 1049 Lyman SK, Schekman R (1997) Binding of secretory polypeptides to a translocon subcomplex is regulated by BiP. Cell 88: 85 96 Maiti IB, Gowda S, Kiernan J, Ghosh SK, Shepherd RJ (1997) Promoter/leader deletion analysis and plant expression vectors with the figwort mosaic virus (FMV) full length transcript (FLt) promoter containing single or double enhancer domains. Transgenic Res 6: 143 156 Maleck K, Levine A, Eulgem T, Morgan A, Schmid J, Lawton KA, Dangl JL, Dietrich RA (2000) The transcriptome of Arabidopsis thaliana during systemic acquired resistance . Nat Genet 26: 403 410 Manach N, Kuntz M (1999) Stress induction of a nuclear gene encoding f or a plastid protein is mediated by photo oxidative events. Plant Physiol Biochem 37: 859 868 Marocco A, Santucci A, Cerioli S, Motto M, Di Fonzo N, Thompson R, Salamini F (1991) Three high lysine mutations control the level of ATP binding HSP70 like prot eins in the maize endosperm. Plant Cell 3: 507 515 Marshall JS, DeRocher AE, Keegstra K, Vierling E (1990) Identification of heat shock protein hsp70 homologues in chloroplasts. Proc Natl Acad Sci USA 87: 374 378 May T, Soll J (2000) 14 3 3 proteins form a guidance complex with chloroplast precursor proteins in plants. Plant Cell 12: 53 64 McDowell JM, An YQ, Huang S, McKinney EC, Meagher RB (1996) The Arabidopsis ACT7 actin gene is expressed in rapidly developing tissues and responds to several external stimuli. Plant Physiol 111: 699 711 McNellis TW, Mudgett MB, Li K (1998) Glucocorticoid inducible expression of a bacterial avirulence gene in transgenic Arabidposis induces hypersensitive cell death. Plant J 14: 247 257 Miao B, Davis JE, Craig EA (199 7) Mge1 functions as a nucleotide release factor for Ssc1, a mitochondrial Hsp70 of Saccharomyces cerevisiae. J Mol Biol 265: 541 552
134 Mooney B, Harmey MA (1996) The occurence of hsp70 in the outer membrane of plant mitochondria. Biochem Biophys Res Comm 2 18: 309 313 Morishima Y, Murphy PJ, Li DP, Sanchez ER, Pratt WB (2000) Stepwise assembly of a glucocorticoid receptor.hsp90 heterocomplex resolves two sequential ATP dependent events involving first hsp70 and then hsp90 in opening of the steroid binding p ocket. J Biol Chem 275: 18054 18060 Morris JA, Dorner AJ, Edwards CA, Hendershot LM, Kaufman RJ (1997) Immunoglobulin binding protein (BiP) function is required to protect cells from endoplasmic reticulum stress but is not required for the secretion of se lective proteins. J Biol Chem 272: 4327 4334 Muench DG, Wu Y, Zhang Y, Li X, Boston RS, Okita TW (1997) Molecular cloning, expression and subcellular localization of a BiP homolog from rice endosperm tissue. Plant Cell Physiol 38: 404 412 Muresan Z, Arva n P (1998) Enhanced binding to the molecular chaperone BiP slows thyroglobulin export from the endoplasmic reticulum. Mol Endocrinol 12: 458 467 Nakai M, Kato Y, Ikeda E, Toh, Endo T (1994) Yge1p, a eukaryotic Grp E homolog, is localized in the mitochondr ial matrix and interacts with mitochondrial Hsp70 . Biochem Biophys Res Commun 200: 435 442 Nara Y, Kurata H, Seki M, Taira K (2000) Glucocorticoid induced expression of a foreign gene by the GVG system in transformed tobacco BY 2 cells. Biochem Eng J 6: 1 85 191 Nelson RJ, Ziegelhoffer T, Nicolet C, Werner Washburne M, Craig EA (1992) The translation machinery and 70 kDa heat shock protein cooperate in protein synthesis. Cell 71: 97 105 Neupert W (1997) Protein import into mitochondria. Annu Rev Biochem 6 6: 863 917 Neupert W, Hartl FU, Craig EA, Pfanner N (1990) How do polypeptides cross the mitochondrial membranes? Cell 63: 447 450 Nguyen TH, Law DTS, Williams DB (1991) Binding protein BiP is required for translocation of secretory proteins into the end oplasmic reticulum in Saccharomyces cerevisiae . Proc Natl Acad Sci USA 88: 1565 1569 Normington K, Kohno K, Kozutsumi Y, Gething MJ, Sambrook J (1989) S. cerevisiae encodes an essential protein homologous in sequence and function to mammalian BiP. Cell 57 : 1223 1236
135 Nover L, Scharf KD (1997) Heat stress proteins and transcription factors. Cell Mol Life Sci 53: 80 103 Pedrazzini E, Giovinazzo G, Bollini R, Ceriotti A, Vitale A (1994) Binding of BiP to an assembly defective protein in plant cells. Plant J 5: 103 110 Prip Buus C, Westerman B, Schmitt M, Langer T, Neupert W, Schwarz E (1996) Role of the mitochondrial DnaJ homologue, Mdj1p, in the prevention of heat induced protein aggregation. FEBS Lett 380: 142 146 Quinn PJ (1988) Effects of temperature on cell membranes. Symp Soc Exp Biol 42: 237 258 Racusen D (1984) Lipid acyl hydrolase of patatin. Can J Bot 62: 1640 1644 Rassow J, von Ahsen O, Bmer U, Pfanner N (1997) Molecular chaperones: towards a characterization of the heat shock protein 70 family . Trends in Cell Biol 7: 129 133 Raynal M, Guilleminot J, Gueguen C, Cooke R, Delseny M, Gruber V (1999) Structure, organization and expression of two closely related novel Lea (late embryogenesis abundant) genes in Arabidopsis thaliana . Plant Mol Biol 40 : 153 165 Reymond P, Weber H, Damond M, Farmer EE (2000) Differential gene expression in response to mechanical wounding and insect feeding in Arabidopsis. Plant Cell 12: 707 720 Rial DV, Arakaki AK, Ceccarelli EA (2000) Interaction of the targeting sequ ence of chloroplast precursors with Hsp70 molecular chaperones. Eur J Biochem 267: 6239 6248 Robinson AS, Bockhaus JA, Voegler AC, Wittrup KD (1996) Reduction of BiP levels decreases heterlogous protein secretion in Saccharomyces cerevisiae . J Biol Chem 2 71: 10017 10022 Rosahl S, Schell J, Willmitzer L (1987) Expression of a tuber specific storage protein in transgenic tobacco plants: demonstration of an esterase activity. EMBO J 6: 1155 1159 Sanders SL, Whitfield KM, Vogel JP, Rose MD, Schekman RW (1992 ) Sec61p and BiP directly facilitate polypeptide translocation into the ER. Cell 69: 353 365 Schaffer R, Landgraf J, Accerbi M, Simon VV, Larson M, Wisman E (2001) Microarray analysis of diurnal and circadian regulated genes in Arabidopsis. Plant Cell 13: 113 123 Schena M, Shalon D, Davis RW, Brown PO (1995) Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270: 467 470
136 Schenk PM, Kazan K, Wilson I, Anderson JP, Richmond T, Somerville SC, Manners JM (2000) C oordinated plant defense responses in Arabidopsis revealed by microarray analysis. Proc Natl Acad Sci USA 97: 11655 11660 Schirmer EC, Lindquist S, Vierling E (1994) An Arabidopsis heat shock protein complements a thermotolerance defect in yeast. Plant Ce ll 6: 1899 1909 Schlicher T, Soll J (1997) Chloroplastic isoforms of DnaJ and GrpE in pea. Plant Mol Biol 33: 181 185 Schneider HC, Westermann B, Neupert W, Brunner M (1996) The nucleotide exchange factor MGE exerts a key function in the ATP dependent c ycle of mt Hsp70 Tim44 interaction driving mitochondrial protein import. EMBO J 15: 5796 5803 Schnell DJ, Kessler F, Blobel G (1994) Isolation of components of the chloroplast protein import machinery. Science 266: 1007 1012 Schroda M, Vallon O, Wollman FA, Beck CF (1999) A chloroplast targeted heat shock protein 70 (HSP70) contributes to the photoprotection and repair of photosystem II during and after photoinhibition. Plant Cell 11: 1165 1178 Schwarz R, Grossman AR (1998) A response regulator of cyanob acteria integrates diverse environmental signals and is critical for survival under extreme conditions. Proc Natl Acad Sci USA 95: 11008 11013 Seki M, Narusaka M, Abe H, Kasuga M, Yamaguchi Shinozaki K, Carninci P, Hayashizaki Y, Shinozaki K (2001) Monito ring the expression pattern of 1300 Arabidopsis genes under drought and cold stresses by using a full length cDNA microarray. Plant Cell 13: 61 72 Senda K, Yoshioka H, Doke N, Kawakita K (1996) A cytosolic phospholipase A2 from potato tissues appears to b e patatin. Plant Cell Physiol 37: 347 353 Sheffield WP, Shore GC, Randall SK (1990) Mitochondrial precursor protein. Effects of 70 kilodalton heat shock protein on polypeptide folding, aggregation, and import competence. J Biol Chem 265: 11069 11076 Shi Y, Mosser DD, Morimoto RI (1998) Molecular chaperones as HSF1 specific transcriptional repressors. Genes Dev 12: 654 666 Shinichi T, Xie Z, Reed JC (1999) An evolutionarily conserved family of Hsp70/Hsc70 molecular chaperone regulators. J Biol Chem 274: 7 81 788 Shinozaki K, Yamaguchi Shinozaki K (2000) Molecular responses to dehydration and lwo temperature: differences and cross talk between two stress signaling pathways. Curr Opin Plant Biol 3: 217 223
137 Simons JF, Ferro Novick S, Rose MD, Helenius A (199 5) BiP/Kar2p serves as a molecular chaperone during carboxypeptidase Y folding in yeast. J Cell Biol 130: 41 49 Smoot LM, Smoot JC, Graham MR, Somerville GA, Sturdevant DE, Migliaccio CA, Sylva GL, Musser JM (2001) Global differential gene expression in r esponse to growth temperature alteration in group A Streptococcus. Proc Natl Acad Sci USA 98: 10416 10421 Sondermann H, Scheufler C, Schneider C, Hhfeld J, Hartl F U, Moarefi I (2001) Structure of a Bag/Hsc70 complex: Convergent functional evolution of H sp70 nucleotide exchange factors. Science 291: 1553 1557 Spertini D, Beliveau C, Bellemare G (1999) Screening of transgenic plants by amplification of unknown genomic DNA flanking T DNA. BioTechniques 27: 308 314 Stuart RA, Cyr DM, Craig EA, Neupert W ( 1994) Mitochondrial molecular chaperones: their role in protein translocation. Trends Biochem Sci 19: 87 92 Sung DY, Vierling E, Guy CL (2001) Comprehensive expression profile analysis of the Arabidopsis hsp70 gene family. Plant Physiol 126: 789 800 Thom ashow MF (1998) Role of cold responsive genes in plant freezing tolerance. Plant Physiol 118: 1 7 Thomashow MF (2001) So what's new in the field of plant cold acclimation? Lots! Plant Physiol 25: 89 93 Unno K, Kishido T, Hosaka M, Okada S (1997) Role of Hsp 70 subfamily, Ssa, in protein folding in yeast cells, seen in luciferase transformed ssa mutants. Biol Pharm Bull 20: 1240 1244 van Hal, NLW, Vorst O, van Houwelingen AMML, Kok EJ, Peijnenburg A, Aharoni A, van Tunen AJ, Keijer J (2000) the applicatio n of DNA microarrays in gene expression analysis. J Biotech 78: 271 280 Vitale A, Bielli A, Ceriotti A (1995) The binding protein associates with monomeric phaseolin. Plant Physiol 107: 1411 1418 Vogel JP, Misra LM, Rose MD (1990) Loss of BiP/GRP78 funct ion blocks translocation of secretory proteins in yeast. J Cell Biol 110: 1885 1895 Wang C, Lin BL (1993) The disappearance of an hsp70 species in mung bean seed during germination: purification and characterization of the protein. Plant Mol Biol 21: 317 329
138 Wang H, Goffreda M, Leustek T (1993) Characteristics of an Hsp70 homolog localized in higher plant chloroplasts that is similar to DnaK, the Hsp70 of prokaryotes. Plant Physiol 102: 843 850 Wang H, Kurochkin AV, Pang Y, Hu W, Flynn GC, Zuiderweg ER ( 1998) NMR solution structure of the 21 kDa chaperone protein DnaK substrate binding domain: a preview of chaperone protein interaction. Biochemistry 37: 7929 7940 Wang SM, Khandekar JD, Kaul KL, Winchester DJ, Morimoto RI (1999) A method for the quantitat ive analysis of human heat shock gene expression using multiplex RT PCR assay. Cell Stress Chaperones 4: 153 161 Wehmeyer N, Hernandez LD, Finkelstein RR, Vierling E (1996) Synthesis of small heat shock proteins is part of the developmental program of lat e seed maturation. Plant Physiol 112: 747 757 Weis E, Berry JA (1988) Plants and high temperature stress. Symp Soc Exp Biol 42: 329 346 Werner Washburne M, Stone DE, Craig EA (1987) Complex interactions among members of an essential subfamily of hsp70 ge nes in Saccharomyces cerevisiae . Mol Cell Biol 7: 2568 2577 Westermann B, Gaume B, Herrmann JM, Neupert W, Schwarz E (1996) Role of the mitochondrial DnaJ homolog Mdj1p as a chaperone for mitochondrially synthesized and imported proteins. Mol Cell Biol 16 : 7063 7071 Westermann B, Neupert W (1997) Mdj2p, a novel DnaJ homolog in the mitochondrial inner membrane of the yeast Saccharomyces cerevisiae. J Mol Biol 272: 477 483 Wimmer B, Lottspeich F, Van der Kle I, Veenhuts M, Gietl C (1997) The glyoxysomal an d plastid molecular chaperones (70 kDa heat shock protein) of watermelon cotyledons are encoded by a single gene. Proc Natl Acad Sci USA 94: 13624 13629 Wooden SK, Li LJ, Navarro D, Qadri I, Pereira L, Lee AS (1991) Transactivation of the grp78 promoter b y malfolded proteins, glycosylation block, and calcium ionophore is mediated through a proximal region containing a CCAAT motif which interacts with CTF/NF I. Mol Cell Biol 11: 5612 5623 Wrobel RL, OBrian GR, Boston RS (1997) Comparative analysis of BiP g ene expression in maize endosperm. Gene 204: 105 113 Wu CH, Caspar T, Browse J, Lindquist S, Somerville C (1988) Characterization of an hsp70 cognate gene family in Arabidopsis. Plant Physiol 88: 731 740
139 Wu SH, Wang C, Chen J, Lin BL (1994) Isolation of a cDNA encoding a 70 kDa heat shock cognate protein expressed in vegetative tissues of Arabidopsis thaliana . Plant Mol Biol 25: 577 583 Xu W, Bak S, Decker A, Paquette SM, Feyereisen R, Galbraith DW (2001) Microarray based analysis of gene expression in v ery large gene families: the cytochrome P450 gene superfamily of Arabidopsis thaliana . Gene 272: 61 74 Yamaguchi Shinozaki K, Shinozaki K (1993) Characterization of the expression of a desiccation responsive rd29 gene of Arabidopsis thaliana and analysis of its promoter in transgenic plants. Mol Gen Genet 236: 331 340 Yamaguchi Shinozaki K, Shinozaki K (1994) A novel cis acting element in an Arabidopsis gene is involved in responsiveness to drought, low temperature, or high salt stress. Plant Cell 6: 251 264 Yu Y, Mu HH, Mu Forster C, Wasserman BP (1998) Polypeptides of the maize amyloplast stroma. Stromal localization of starch biosynthetic enzymes and identification of an 81 kilodalton amyloplast stromal heat shock cognate. Plant Physiol 116: 1451 1460 Zhang J, Furukawa RD, Fremes SE (1996) The beneficial effects of heat shock for prolonged hypothermic storage. J Surg Res 63: 314 319 Zhu B, Choi DW, Fenton R, Close TJ (2000) Expression of the barley dehydrin multigene family and the development of free zing tolerance. Mol Gen Genet 264: 145 153 Zhu X, Zhao X, Burkholder WF, Gragerov A, Ogata CM, Gottesman ME, Hendrickson WA (1996) Structural analysis of substrate binding by the molecular chaperone DnaK. Science 272: 1606 1614 Ziegelhoffer T, Lopez Bues a P, Craig EA (1995) The dissociation of ATP from hsp70 of Saccharomyces cerevisiae is stimulated by both Ydj1p and peptide substrates. J Biol Chem 270: 10412 10419 Zou J, Guo Y, Guettouche T, Smith DF, Voellmy R (1998) Repression of heat shock transcript ion factor HSF1 activation by Hsp90 (Hsp90 complex) that forms a stress sensitive complex with HSF1. Cell 94: 471 480
140 BIOGRAPHICAL SKETCH Dong Yul Sung was born in Kwangju, Korea, on August 16, 1968. He received a Bachelor of Science degree in horticultureal science from Seoul National University in 1992. He enrolled in the Plant Molecular and Cellular Biology (PMCB) P rogram at the University of Florida in the fall of 1994. He received his Master of Science degree in the PMCB program under the guidance of Dr. Indra K. Vasil in 1996. He then entered the same program as a doctoral student in the fall of 1996. He is mar ried to a wonderful wife, Gwijun Kwon and is the father of two lovely children, Richard and Audrey.