Chapter 4: Expression of Heat Shock Genes and HSPs

4.1 Introduction
Since Ritossa’s discovery of the heat shock response, and Sorger and Pelham’s isolation of the first HSF, there has been a significant accumulation of literature on the heat shock response, how heat shock genes are regulated and the functions of heat shock proteins (Nover et al., 2001; Forreiter and Nover, 1998; Boston et al., 1996). The sequencing of the Arabidopsis genome has revealed that there are at least 200 genes involved in the heat shock response and still more may have yet to be identified. In plants, low molecular weight HSPs are the most abundant, although in terms of gene number, analysis of the Arabidopsis genome indicates that the HSP 40 family is the largest (Table 1.1).

HSPs are expressed under control conditions and a variety of abiotic stresses including heat stress. The relative activates of these HSPs can differ significantly between tissue types and organisms. Depending on the stress, a select number of HSP families or specific HSP family members may only be responsive. This gene selective nature of HSP activation under non-heat stresses again emphasizes the questions of; what is the master control switch for non-heat stresses and if it is the HSF, how does the family act in a gene selective manner?

Under low temperature stress, specific members of the HSP 70, HSP90 and the low molecular weight HSP families are known to be up-regulated. Anderson et al., (1993) have identified clones corresponding to HSP 70 and HSC 70 that are up-regulated during cold acclimation, cold and heat shock in spinach. Similar results were observed in soybean (Neven et al., 1997; Cabané et al., 1993) and associations between HSP 70 with cold labile proteins were noted in spinach (Guy et al., 1998b). Other studies have found low temperature up-regulation of several HSP members from the low molecular weight HSP family, specifically HSP 17.5 in chestnut (Soto et al, 1999), HSP 17.6 in tomato and Arabidopsis (Cheong et al., 2002; Kadyrzhanova et al., 1998), HSP 18.1 in tomato (Sabehat et al., 1998), HSP 21 in tomato (Sabehat et al., 1998) and other low molecular weight HSPs in potato (Van Berkel et al., 1994). A member from the HSP 90 family is also observed to be up-regulated under cold stress in Brassica napus (Krishna et al., 1995).

In alfalfa, only a handful of HPS genes have been isolated, specifically, a HSP 86 (Jeff Volenec, Purdue University, unpublished data, not submitted to GenBank), a HSP 70 (GenBank accession number X99197), an HSP18 (GenBank accession number X58710 and X58711) and a HSP 17 (GenBank accession number X98617). These HSPs are limited to direct submissions to GenBank and almost no characterization has been done.

This chapter details experiments that characterize the gene expression of alfalfa HPS 86, HSP 18 and protein accumulations of class I low molecular weight HSPs. It is hypothesized that the aforementioned genes and proteins will accumulate under controlled heat and cold stress and during fall field conditions. The general goal of these experiments was to follow the results described in Chapter 3 and generate a more holistic view of the heat shock response in alfalfa. More simply, to observe a transcription factor, and find a connection between the stress induced downstream products the transcription factor might activate.

4.2 Materials & Methods

4.2.1 DIG Probe Preparation
A clone for alfalfa HSP 18 (specifically HSP 18.2) was obtained by PCR using primers (Appendix A, MsHSP-1a and MsHSP-1b) designed based on the ORF of the published sequence in Genbank (X58711). The clone for the alfalfa HSP 86, was obtained from Dr. Jeff Volenec, Purdue University. The full HSP 86 cDNA was amplified through PCR using the M13 forward and M13 reverse universal primers (Appendix A). DIG probes were prepared using the aforementioned clones, corresponding primers and according to the methods described in chapter 3, 3.2.10. To confirm probe quality, the DIG labeled DNA probes were hybridized to random RNA samples from heat and cold stressed tissues and resolved by northern blot hybridization (Appendix G).

4.2.2 Rescreening of RNA Macroarrays
The macroarray membranes used in the chamber and field experiments (Chapter 3, 3.2.12) were reprobed for this experiment. The DIG probe was stripped off by pouring boiling stripping solution (0.1 % SDS) over each membrane. The membranes were transferred to a medium sized Pyrex glass dish, a further 700 ml of boiling stripping solution were added and allowed to shake at room temperature for 10 min. The membranes were then resealed in a hybridization bag and a new probe was hybridized and detected as described in Chapter 3, 3.2.11. Following hybridization the membranes were screened and analyzed as described in Chapter 3, 3.2.13. This procedure was repeated for each additional DIG probe used.

4.2.3 Protein Tissue Preparation
Protein samples were obtained from the organic phase separated during the RNA extractions described in Chapter 3, 3.2.7. Thus the tissue treatments were identical to the chamber and field experiments described in Chapter 3 (3.2.5 and 3.2.6).

4.2.4 Protein Extraction
Tubes were briefly vortexed to resuspend the organic and interphase layers isolated during the RNA extractions described in Chapter 3, 3.2.7. Into a 96 well 2 ml microplate, 250 ul aliquots were transferred and combined with 0.15 ml of 100% ethanol, vortexed and incubated at room temperature for 3 min to precipitate the DNA. The DNA was sedimented by spinning at 2000xg for 5 min at 4°C. Using a robotic pipetter (Biomec 2000, Beckman Coulier) the supernatant was transferred to a new microplate and combined with 0.75 ml of isopropanol to precipitate the protein. The protein was pelleted by centrifugation at 12 000xg for 10 min at 4°C. The pellets were washed 3 times in 0.3 M guanidine-HCl solution (prepared in 95 % ethanol). For each wash, the pellets were incubated in the wash solution for 20 min then centrifuged at 7500xg for 5 min at room temperature. A final wash of 1.5 ml of 100% ethanol were added (as above) to remove any remaining guanidine-HCl salt. The protein pellets were air dried for 10 min at room temperature. The proteins were resuspended by adding 0.5 ml of 1% SDS and shaking in a 50°C incubator for 2 hrs. Any undissolved material was pelleted by centrifugation at 10 000xg for 10 min. Unused protein was stored at -20°C.

4.2.5 Protein Quantification
Protein was quantified using the Dc protein assay kit for samples containing SDS (
BioRad) and a SpectraMax plus384 spectrophotometer (Molecular Devices). Using a robotic pipetter, 200 ul of assay solution were aliquoted into a 96 well microplate. To each well 5 ul of protein sample were added and mixed by pipetting. The samples were incubated at room temperature for 15 min and the absorbance was read at 750 nm. Sample concentration was calculated from the absorbance and all protein samples were diluted to a concentration of 1 ug/ul.

4.2.6 Western Blot Immuno-detection
The antibody to class 1 low molecular weight HSPs was raised in rabbit against rice HSPs, and provided by Dr. Chu-Yung Lin, National Taiwan University, Taiwan. Protein samples (10 ug) were diluted with equal volumes of SDS loading buffer and boiled for 5 min. Samples were separated on a 15% SDS polyacrylamide gel at 200 volts until the loading dye reached the bottom of the gel. The protein was then electro-blotted onto a Immobilon-P PVDF membrane (Millipore) using a mini Trans-Blot Cell (
BioRad) containing western transfer buffer (Appendix B) at 14 volts overnight. After transfer, the membrane was submerged in western blocking solution 1 (Appendix B) gently shaking for 3 hours. The membrane was then transferred to western blocking solution 2 (Appendix B) and was agitated for 5 min to remove the milk powder residue. The blocking solution was poured off and the primary antibody solution was added (1:10 000 dilution in western blocking solution 2) and incubated gently shaking at room temperature for 1.5 hrs. The membrane was transferred to a new dish and washed 3 times for 30 min each in western blocking solution 2. The blocking solution was poured off and the secondary antibody solution was added (1:10 000 dilution of anti-rabbit IgG conjugated to an alkaline phosphatase enzyme, (Sigma) to western blocking solution 2 and incubated while gently shaking at room temperature for 2 hrs. The membrane was then washed 2 times, 15 min each, in western Mg2+ wash solution (Appendix B). The wash solution was then discarded, and western detection solution (Appendix B) was added and incubated in the dark for 5 to 10 min or until pink bands appeared. To stop the detection reaction, the membrane was transferred to a large volume of water and then sealed in acetate. Membranes were then scanned and analyzed as described in Chapter 3, 3.2.13.

4.2.7 Preparation of Protein Macroarrays
A 10 ug aliquot of each protein sample was diluted in 200 ul of western transfer buffer. Using a 96 well plate (Fisher), the diluted samples were transferred (using the Biomec 2000 pipetting robot), vacuum blotted onto a PVDF membrane and rinsed with 200 ul of western transfer buffer with the 96 well Bio-Dot microfiltration device (
BioRad). The membranes were then transferred to western blocking solution 1 and incubated with gentle shaking overnight.

4.2.8 Screening of Protein Macroarrays
All membranes were processed at the same time to minimize experimental variations. Detection and screening was performed as according to the method described in Chapter 4, 4.2.6. The volumes of the solutions were scaled up to accommodate 36 macroarray membranes. Developed protein macroarray membranes were scanned into digital images and values for relative video intensity were determined using the Array-Pro Analyzer software as was describes for northern macroarrays (Chapter 3, 3.2.13 and Figure 3.8).

4.2.9 Statistical Analysis
In the absence of stress, HSP 18 mRNA, HSP 86 mRNA and the low molecular weight HSPs all exhibited a diurnal effect (data not shown). Thus in the chamber experiment, values obtained under heat and cold stress were normalized to the values obtained in the diurnal control prior to statistical analysis. Furthermore, cultivars with the most similar fall dormancy were grouped together to facilitate the comparison. For definition of cultivars and their dormancy group see Chapter 3, 3.2.3.

In the field experiment, the data for the CUF 101 cultivar at the Elora field location were removed from the analysis as there were not enough field plants to complete the experiment. Similarly, all field data from NewLiskeard location were also excluded due to the low sampling number and improper processing of field samples.

All statistical analyses were performed using SAS (statistical analysis software, version 8.0, Cary California). All data were subjected to a repeated measures variance analysis using the “proc mixed” procedure of SAS (Statistical Analysis Software). For this analysis the variance was partitioned into block, variety, time, block x variety and variety x time. The blocks and block x variety were assumed to be random effects, with varieties and time considered to be fixed effects. A log transformation of the data was required to obtain homogeneity and normal distribution of error variance. Treatment and environmental effects were assumed to be additive, while experimental errors were random, independent and normally distributed about a mean of zero with a common variance. The type I error rate was set at 0.05.

4.3 Results

4.3.1 Transcript Expression of HSP 18 mRNA, HSP 86 mRNA and Protein Accumulations of Low Molecular Weight HSPs Under Heat Stress 41°C Chamber Experiment - HSP 18 mRNA
There were significant (at 95% confidence) variety x time interactions in HSP 18 levels under heat stress at 41°C. HSP 18 mRNA was induced similarly over all 10 cultivars with the exception of dormancy group 3 (cv. 5246) (Figure 4.1). The heat stress induced a sharp spike (almost 5 fold increase) after 1 hr of exposure which gradually declined over the remainder of the 12 hr experiment. Several dormancy groups exhibited a secondary response peak at 5 hrs of exposure. On comparing dormant to non-dormant cultivars, the height of the primary response peak was greater in the dormant cultivars as compared to the non-dormant cultivars (Figure 4.1, A and B). As well, at the tail end of the experiment (after 8 hrs), dormant cultivars exhibited an upward trend that was not observed in the non-dormant cultivars. This suggested that dormant cultivars exhibited a secondary and or possibly a tertiary response to the stress as compared to non-dormant cultivars which exhibited a single response. Chamber Experiment - HSP 86 mRNA
Under heat stress at 41°C, HSP 86 exhibited fluctuations in transcript levels which followed similar response trends across all cultivars (Figure 4.2). After 1 hr of exposure transcript levels in all cultivars peaked (~2 fold increase) only to dramatically decrease (~4 fold decrease) within 3 hrs. Transcript levels then gradually increased throughout the rest of the exposure. Laboratory and Chamber Experiment - lmwHSPs
To observed HSP proteins accumulations in alfalfa, western blot analysis using antibodies raised to class 1 low molecular weight HSP (lmwHSP) in rice, was employed. In response to heat shock (as described in Chapter 3, 3.2.2) the lmwHSP antibody produced one very strong band and two lower faint bands that were responsive to heat stress (Figure 4.3). The lmwHSP antibody also produced a constitutive band, but this bands intensity was significantly lower than the heat induced band (Figure 4.3). Further analysis of this western indicated a linear increase of lmwHSPs from 0 to 4 hrs of 41°C heat stress (Figure 4.3).

In the chamber experiment, alfalfa low molecular weight (lmw) HSPs exhibited significant increases in protein accumulation, at 41°C, in all dormancy groups with a variety of expression patterns (Figure 4.4). Most notable were the two response peaks observed in dormancy group 1&2 and the single peak observed after 5 hrs of heat stress in dormancy group 11.

4.3.2 Transcript Expression of HSP 18 mRNA, HSP 86 mRNA and Protein Accumulations of Low Molecular Weight HSPs Under Cold Stress at 4°C Chamber Experiment - HSP 18 mRNA
Under cold stress at 4°C, in all cultivars HSP 18 transcript levels exhibited fluctuations over the 12 hr exposure (Figure 4.5). The response curves followed no detectable pattern and were not as dramatic as was observed under heat stress. However, as was observed under heat stress, at the tail end of the experiment, dormant cultivars exhibited an upward trend that was not observed in the non-dormant cultivars. Again, this suggests that dormant cultivars were possibly exhibiting some type of secondary or tertiary response to the stress. Chamber Experiment - HSP 86 mRNA
Under cold stress at 4°C, HSP 86 transcripts exhibited very limited stress response with the exceptions of dormancy group 3 and 11 (Figure 4.6). In contrast, dormancy group 3 (cv. 5246) transcript levels significantly decreased after 1 hr of exposure and remained depressed throughout the experiment. In dormancy group 11 (cv. Wadi Qurayat) transcript levels decreased after 1 hr of exposure, peaked at 5 hrs, again decreased after 8 hrs and began to increase by the end of the experiment. Chamber Experiment - lmwHSPs
All dormancy groups responded positively to cold stress at 4°C with variations in the time of the response peak (Figure 4.7). Dormancy group 1&2 exhibited two response peaks at 1 and 5 hrs of exposure. Dormancy group 3 exhibited its response peak at 3 hrs while dormancy group 7&8 and 9&10 exhibited their response peaks at 1 hr. Dormancy group 11 had a maximum response after 3 hrs.

4.3.3 Transcript Expression of HSP 18 mRNA, HSP 86 mRNA and Protein Accumulations of Low Molecular Weight HSPs Under Field Fall Conditions in 2001 HSP 18 mRNA - Root and Bud Expression
All cultivars exhibited statistically significant fluctuating patterns of HSP 18 mRNA expression in both bud and root tissue throughout the fall (Figure 4.8 and 4.9). Expression patterns in both bud and root followed similar patterns up to the end of October. However, through November and December patterns between cultivars diverged from each other, indicating the cultivars were responding differently to the stresses. Expression patterns in the root and bud were affected by temperature, where a decrease in ambient temperature (below 0°C) correlated to a increase in expression of HSP 18 mRNA in filed buds. HSP 86 mRNA - Root and Bud Expression
HSP 86 mRNA expression in bud tissue followed similar expression patterns to HSP 18 mRNA in that the fluctuations in transcript levels during September and October cultivars followed similar patterns which then significantly diverged during November and December (Figure 4.10). Expression in bud tissue fluctuated opposite to decreases in temperature which was most noticeable in Wadi Qurayat which exhibited significant increases in transcript levels in the end of November. This trend was also observed in the other cultivars but to a lesser degree. In root tissue all cultivars followed the same fluctuating expression patterns throughout the fall (Figure 4.11). These patterns seem to fluctuate with soil temperature (Figure 4.8, B). As soil temperature decreased HSP 86 transcript levels increased. Low Molecular Weight HSPs - Root and Bud Expression
An analysis of protein content per g of dry weight tissue was performed to ascertain if protein levels in alfalfa were changing throughout the fall (Appendix M). Varying total protein levels would have required that the data be adjusted so that protein accumulations would not be misjudged as fluctuations in total protein content. Protein levels did not significantly change throughout the fall of 2001, and thus data was used directly.

In the bud tissue, accumulation of lmwHSPs varied among all four cultivars (Figure 4.12). The more dormant cultivars (Beaver and Saranac) exhibited increasing accumulations during November and December while non-dormant cultivars (Sutter and Wadi Qurayat) exhibited a significant decrease during November and December. In the roots, all cultivars exhibited similar fluctuating expression patterns with a subtle increasing trend throughout the fall (Figure 4.13). This subtle increase was more noticeable in the dormant cultivars (Beaver and Saranac), which suggests a correlation between the progressing fall stresses and increases of lmwHSPs.

4.4 Discussion
Characterization of the heat shock response in the chamber experiment with the high number of sampling events and replications produced extremely reliable and significant results. The experiment revealed concrete patterns of HSP 18 mRNA and HSP 86 mRNA inductions and accumulations of class 1 lmwHSPs.

As alfalfa tends to have low nucleic acid sequence conservation (Figure 2.1), it was important to use native HSP cDNA clones to generate probes used in the northern macroarray analysis. However, in the literature there are only a few HSP gene sequences available from alfalfa. Attempts to PCR out new native sequences using primers designed from homologous sequences proved futile. The sequences that were available allowed me to observe a heat and cold responsive nature of alfalfa HSP genes and some of the expressed proteins.

It was interesting to observe that HSP 18, HSP 86 and the lmwHSPs all exhibited a significant diurnal expression patterns in the chamber experiment (data not shown). These results are consistent with the findings of Merquiol et al., (2002) who observed daily and seasonal diurnal expression in HSP 18 and HSP 90 transcripts in the legume Retama raetam. Thus, including the diurnal control was important as the diurnal data was significantly different from the raw unadjusted data. I was, therefore, able to standardize the diurnal data and gain a much clearer picture of the true nature of the stress response.

Under heat stress, HSP 18, HSP 86 transcripts and the lmwHSPs all exhibited increases and generally followed similar patterns of expression among cultivars. Analysis of HSP18 expression, revealed that some dormancy groups exhibited a second smaller response peak and dormancy group 3 (cv. 5346) appeared to have a different response curve with its primary peak at 3 hrs. Similarly, HSP 86 expression in all groups was similar with the exception of dormancy group 3 which appeared to have a higher basal level of expression and a small response peak. However the small response peak could be attributed to the peak occurring between sampling points. At the protein level, all groups showed heat induced accumulations of HSPs but the expression patterns over the 12 hrs was very different among the groups.

Among HSP 18, HSP 86 transcript induction and lmwHSP protein accumulation, the primary response peaks in the dormant cultivars were significantly higher than the primary peaks in the non-dormant cultivars. This suggests that the non-dormant cultivars may be better adjusted to warmer climates and thus require less induction of HSPs as compared to the dormant cultivars whose physiology is adapted to colder climates. Nguyen et al., (1994) also observed cultivar differences in wheat HSP (low molecular weight) expression at the RNA and protein level in which heat tolerant cultivars had higher and unique levels of expression as compared to heat susceptible cultivars. Again this is suggesting that physiological adaptation to a niche reduces the dependency on the heat shock response.

Under cold stress, expression patterns of both HSP 18 and HSP 86 mRNAs exhibited much lower levels as compared to heat stress, but the genes were responsive nonetheless. The possibility that the cold induced response curves were only background noise was ruled out as the adjusted data was still statistically significant even after the data was normalized to the diurnal control. At the protein level all groups were strongly responsive yet all groups exhibited a variety of expression patterns. This indicates that the accumulations of lmwHSPs are affected by genotype and probably at a number of important regulatory steps.

Several studies of HSP gene expression under low temperature have observed similar (yet less comprehensive) activities of HSP 90 (Krishna et al., 1995), HSP 70 (Guy et al., 1998) and specific low molecular weight HSPs. Sabehat et al., (1998) demonstrated upregulation of tomato HSP 21 homologue under low temperature stress at 2°C for two days. This data confirms the accumulation of specific HSPs under low temperature stress with the potential for them to act as molecular chaperones during cold stress, possibly protecting cells from deleterious effects.

The results from the chamber experiment illustrate the importance of genotype sampling when observing gene expression in a given species. The expression differences, which were most noticeable under cold stress would have provided dramatically different results had only a single cultivar been tested. This is exemplified by the expression patterns of dormancy group 3 which consistently, across all the genes surveyed in this experiment, exhibited a significant difference in expression.

Interpreting the data from the field experiment proved considerably more difficult than the chamber experiment. In the chamber experiment, temperature was easily isolated as the only stress applied to the plants. In the field, plants were exposed to a myriad of stresses and environmental cues that could have influenced the expression of these genes.

In general, the field experiment revealed a variety of fluctuating expression patterns among all surveyed genes, with an apparent divergence in cultivar expression patterns in the latter part of the fall. The most noted response occurred under HSP 86 mRNA expression in the bud tissue. The end of November exhibited the first temperatures in the fall to reach below 0°C. Cultivars Beaver and Saranac, the dormant cultivars, exhibited no significant increase in expression where as cultivars Sutter and even more dramatically Wadi Qurayat, the non-dormant cultivars, exhibited a significant boost in expression. This was opposite to what was observed in the chamber experiment which showed very low levels of expression under both heat and cold. It is possible that under low temperature stress HSP 86 is most responsive at freezing temperatures and is more required by varieties that cannot cope with cold.

Observations of the lmwHSPs showed a steady increase in protein accumulations over the fall in the root, but not in the bud. This trend correlated with the gradual decrease in temperature, suggesting that that in root tissue lmwHSPs are accumulating and possibly protecting against low temperature damage. Numerous studies have shown field HSP accumulation but most of the studies were localized to the warmer months or were performed in seasonally warm regions (Hernandez and Vierling, 1993; Burke et al., 1985; Kinple and Key, 1985). However, Ukaji et al., (1999) observed increases in field low molecular weight HSPs expression throughout monthly sampling over the fall and winter in mulberry (Sapporo, Japan), data consistent with the results of this experiment.

The use of the macroarray analysis revealed that HSP 18 mRNA, HSP 86 mRNA and the lmwHSPs are clearly responsive to heat and cold stress. However, the nature of their expression patterns is dramatically different. The results revealed that differences in heat induced expression are related to a cultivar’s adapted physiology. Specifically, warm-climate adapted cultivars expressed lower levels of the heat shock response because their global physiology is pre-adapted to a hotter climate, a theory supported by Nquyen et al. (1994). Although it was not observed, the opposite may be true for low temperature stress. The clue came from the field observations of HSP 86 in which the non-dormant cultivars exhibit a dramatic increase in transcript levels, especially once the temperature reached freezing conditions. The dormant cultivars (Beaver and Saranac) did not need the extra protection of HSP 86 because their physiology was possibly pre-adapted to cope with the freezing stress.
In the literature, studies on HSP 18 and HSP 86 widely vary in topic from general species specific studies such the HSP 90 family in rice (Pareek at al., 1998) to very specific localizations of gene expression studies such as HSP 18 activity in maize during microsporogenesis (Atkinson et al., 1993; Bouchard et al., 1993). Despite the tremendous volume of literature on the heat shock response, comprising more than 12 000 references, there are large gaps in the literature which prevent a holistic view of how all the active heat shock genes are functioning in concert to achieve stress tolerance. Most of the studies that deal with the heat induction of these specific genes, assessed the heat shock versus control nature of the these genes expression. To my knowledge, this is the first study to semi-quantitatively profile these genes’ expression under different stresses within the same experiment. Clearly, with advent of microarray technology and the completion of more sequenced genomes the holistic, functional and evolutionary story of the stress response will be solidified.

4.5 Conclusion
Expression of HSP 18 mRNA, HSP 86 mRNA and the lmwHSPs in alfalfa were actively expressed in response to heat and cold stress under controlled environmental conditions. Expression patterns were significantly different between all the cultivars surveyed and a relationship between dormant and non-dormant cultivars was observed. HSP 18 mRNA, HSP 86 mRNA and the lmwHSPs are also accumulated in root and bud tissue and were responsive to field fall environmental conditions.