From Epigenetic Mechanisms to the Expressed Phenotype

Unraveling Heat Acclimation Memory:

Thesis submitted for the degree of

―Doctor of Philosophy‖

Anna Tetievsky

Submitted to the Senate of the Hebrew University of Jerusalem

June 2012

This work was carried out under the supervision of:

Proffesor Michal Horowitz

Abstract

Heat acclimation (AC) is a reversible ‘within lifetime‗ phenotypic adaptation to long-term elevations in environmental temperature that evolves via a continuum of temporally varying processes. Successful AC is characterized by enhanced thermal tolerance manifesting as improved endurance and resistance to temperature extremes, collectively delaying the onset of heat injury. Heat acclimation can reinforce or interfere with the ability to combat novel acute stressors. In this way, AC was found to confer protection to a variety of stressors (namely, cross-tolerance) with impaired oxygen supply/oxygen demand ratios. Among these, cross-tolerance between ischemia-reperfusion insult in the heart and hyperoxia in the brain have been extensively studied. The available evidence substantiates that a reprogramming of gene expression and translational processes are essential events in the pathway to heat acclimation. The important hallmarks of AC and AC-mediated cross-tolerance identified to date include enhanced reserves of heat shock protein 70 (HSP70), heat shock protein 90 (HSP90), anti-apoptotic and anti-oxidative pathways, accelerated (vs. nonacclimated) stress-mediated transcriptional activation of these genes and in turn, the heat shock response (HSR). Although the benefits of acclimation are considered transitory, the few published investigations on the time course of AC loss indicate that reacquisition of the acclimated phenotype [re-acclimation-(ReAC)] is more rapid upon return to the hot environment than the time required to acclimate initially. Our previous findings show that the cardiac acclimated phenotype has a molecular memory and regains both acclimatory enhanced performance and cardioprotection after loss of acclimation within only 2 days of exposure to the acclimating conditions. Interestingly, the protein levels of hsp70 and anti-apoptotic bcl-xL were upregulated even after 30 days of deacclimation (DeAC) conditions. Given that the evolvement of AC depends on a reprogramming of gene expression, the dichotomy between the molecular and physiological phenotypes led us to conclude that AC induces a long-lasting, transcriptional program that enables individuals who have undergone an initial AC session to achieve faster ReAC. We hypothesized that AC memory involves upstream epigenetic information that predisposes to rapid reacclimation and cytoprotective memory.

In a broad sense, the transfer of epigenetic information is associated with chromatin remodeling via molecular and biochemical processes that maintain the chromatin- DNA package in active or silent states. Post-translational modifications (e.g. lysine acetylation, serine phosphorylation) of the N-terminal tails of histone proteins H3 and H4 that protrude from the nucleosome are the most common forms of chromatin remodeling. Such modifications can be controlled by intracellular signaling and are likely to be important in selective epigenetic tagging when environmental stressors are involved.

My goal in this study was twofold:

 To test our hypothesis that “acclimation memory” stems from epigenetic adaptations and to unravel the mechanisms involved.  To exploit the DeAC/ReAC model and discover “core gene clusters” and their master regulators that are possibly involved in the generation of “acclimation memory”.

To achieve these goals my specific aims are:

1) To study the transcriptional kinetics of genes of interest: hsp70, hsp90 and hsf1 to validate the hypothesis that transcriptional dynamics is a part of the acclimatory memory repertory. 2) To focus on the molecular and biochemical processes that maintain the chromatin-DNA package in an active state, allowing accessibility and binding of transcription factors to DNA recognition sites. 3) To take a genome-wide approach using a whole rat genome array, to screen for alterations in the expression of genes involved with chromatin remodeling and transcriptional regulation following AC, DeAC, and ReAC.

All experiments were conducted on male rats, Rattus norvegicus, divided into five experimental groups:

1. Control rats (C) maintained at normothermic conditions (24 ºC). 2. Heat acclimated for 2 days (AC2d) at 34ºC. 3. Heat acclimated for 30 days (AC) at 34ºC. 4. Deacclimated (DeAC) at normothermic conditions for 30 days (after AC for 30 days). 5. Reacclimated for 2 days (ReAC) after DeAC for 30 days.

Because faster transcriptional activation in response to stress is a hallmark of successful acclimation, we studied the kinetics of hsp70 and hsp90 transcription following acute heat stress (HS) at 41º C in all experimental groups. Our results here show that both hsp70 and hsp90, similar to the AC group, exhibit faster (than Control) transcriptional dynamics in the DeAC and ReAC groups, with an mRNA peak of hsp70 at 40 min post-HS and mRNA peak of hsp90 immediately after the heat shock (HS) treatment (0 min post-HS). This is the first evidence that the altered ―acclimated‖ HS response is retained after 30 days at normothermic conditions (DeAC), and also exists at ReAC conditions. This finding supports our hypothesis that faster transcriptional dynamics is part of the memory repertory.

Interestingly, the DeAC phenotype is characterized by a mismatch between the greater HSP70 basal mRNA/protein levels and the loss of cardioprotection. One explanation for the dichotomy between the phenotypic-physiological and the genotypic response at DeAC group could stem from the decrease in HSP90 levels in this group. In contrast to the elevated protein HSP70 levels, the DeAC hearts did not display high HSP90 protein or mRNA levels. Given that HSP90 is an essential component in the HS response and the duration of its upregulation is critical to cellular integrity we hypothesized that an inverse relationship between HSP70 and HSP90 in DeAC causes the loss of cardioprotection in DeAC group but does not interfere with predisposition to fast restoration of cardioprotection upon ReAC.

As HSPs are essential in the AC process and the HS response, we used these genes as a prototype model for proving the concept that epigenetic mechanisms contribute to cytoprotective memory.

Histone acetylation, referred to as euchromatin modification, is associated with active transcription. Hence, to substantiate our hypothesis that ReAC involves the activation of the epigenetic machinery, we first measured the levels of acetylated H4 and phosphorylated H3 (Serine10) in the promoter regions of all the experimental treatments. The rationale for measuring H4 acetylation was based on the report of Thomson et al., (2004), which demonstrated the involvement of H4 acetylation in transcription of HSP70 during heat stress in a mammalian species. We screened H3 acetylation/phosphorylation because of the vast body of evidence supporting its involvement in the HS response in non-mammalian species and its role in stress memory. To confirm the binding of transcription factors to the euchromatin, as an

indication of the initiation of transcriptional events, we measured the binding of HSF1 to the heat shock element (HSE) on the HSP70 and HSP90 genes (an essential step in HSP transcription).

Our results revealed a profile of chromatin remodeling at the HSE of the promoter site of hsp70 and hsp90 and HSF1 binding, which provides a conceptual model of the cytoprotective memory:

(i) At the onset of AC, in an ambient-temperature dependent manner, histone H3 phosphorylation by MSK1 kinase switches on HSF1 binding at the HSE of hsp70 and hsp90, with subsequent histone H4 acetylation by a specific acetyl transferase TIP60 at the HSE of both genes, (ii) The acetylation persists throughout DeAC and ReAC, resulting in constitutive HSF1 binding to the hsp70 promoter, irrespective of the transitions in ambient temperatures (from acclimating temperature (34ºC) to 24ºC during DeAC and again returning to 34ºC during ReAC). In contrast, HSF1 binding to hsp90 is temperature dependent. No HSF1 binding occurs in the DeAC state, despite the maintenance of histone H4 acetylation in the HSE in the promoter area of this gene. HSF1-hsp90 binding requires elevated ambient temperatures. The maintenance of elevated histone H4 acetylation in the hsp90 promoter and constitutively elevated HSP70 reserve during DeAC may facilitate the rapid resumption of HSF1 binding to hsp90 HSE, hsp90 transcript translation and the reformation of a cytoprotective milieu upon ReAC.

This investigation delineates, for the first time, the whole-genomic response in a mammalian species during the AC process and also at DeAC and ReAC regimes.

From our analyses, we 1) outline the dynamics of the genomic response of different sets of genes in all the experimental groups, thus allowing some perception of the global acclimatory molecular strategy underlying heat acclimation and, 2) discuss the likely pathways leading to the ―molecular memory‖ formation conferred by AC.

Our data demonstrate that from a total of 27,342 distinct probes, 651 genes showed a significantly changed transcriptional behavior (either upregulated or downregulated > 1.5 - fold) at least in one of the experimental groups. The clustering by a bioinformatics tool revealed five gene clusters characterized by a significantly identical transcriptional behavior of the genes in each cluster. Among these clusters, we can clearly distinguish between the cluster in which the genes are

affected only by short-term heat acclimation (STHA) for 2 days (AC2d) and the clusters in which the gene transcription significantly changed in response to the DeAC and ReAC regimes, both or each. The magnitude of change in expression of these gene groupings shows temporal variation over the course of heat acclimation- deacclimation-reacclimation.

The majority of 18 transcripts increased in expression upon AC2d (STHA), reflecting a basic paradigm of exposure to high environmental temperatures: that organisms frequently compensate for the stress effects of high temperature by transcribing more cytoprotective genes. Functional annotation showed that 24% of the genes in this cluster were associated with cytoprotective pathways related to apoptosis, metabolism, and mitochondrial function. The interesting finding is that some genes (for example, Prim1) assigned to this category are associated with maintaining DNA integrity. Activating these genes could provide an initial line of defense for alleviating any genomic damage resulting from the strain associated with the onset of the acclimation process.

The genes clustering analysis revealed four clusters comprising groups of genes that were simultaneously either up-regulated or down-regulated after DeAC and ReAC or both.

The four clusters can be divided into:

1. Two clusters that showed upregulation at DeAC and/or ReAC. 2. Two clusters that showed downregulation at DeAC and/or ReAC. The results of the present study call our attention to transcriptional and chromatin- regulation associated genes that have not yet been linked with AC in mammalian species. The functional annotation analysis of the significantly expressed genes in the DeAC and ReAC periods revealed a group of genes that may act as upstream regulators of histone modifications and transcriptional activation and confer the molecular memory.

We discovered a group of histone-modifications regulators that are annotated to the p38 MAPK/ERK/MSK signaling pathway. MAPK (Mitogen activated protein kinase) pathways are key pathways in regulating stress responses and transducing extracellular signals to cytoplasmic and nuclear effectors. At least one of the roles of the p38 MAPK pathway is to activate MSK1 and to lead to epigenetic changes by phosphoacetylation. Given that we have shown that at the onset of AC (AC2d group), MSK1 recruitment and further histone H3 phosphorylation switches on HSF1

binding at the HSE with subsequent histone H4 acetylation at the HSE of both hsp70 and hsp90 genes, we can argue that the upstream regulatory pathway that activates this signaling cascade may be the p38 MAPK pathway.

Among the genes annotated to the p38 MAPK pathway in our array analysis, the CREB (cAMP response element-binding) was found. MSK1 is a potent CREB kinase, and CREB phosphorylation is indeed taking place in the cell's response to environmental stressors. Under such circumstances, P-CREB may recruit one of the coactivator CREB-binding proteins (e.g. CBP, P300, pCAF) whose histone acetyl transferase (HAT) activity has been shown, at least in vitro, to acetylate histone H4.

One of the interesting findings in the array results was the significant and reversed expression of CITED1 and CITED2 genes in DeAC and ReAC groups. The CITED [cAMP-responsive element-binding protein (CBP)/p300-interacting transactivator with glutamic acid/aspartic acid-rich tail] proteins belong to a family of transcriptional cofactors. Our special interest in this protein was caused by the fact that CITED2 was shown to function as a repressor of hypoxia-inducible factor-1 (HIF-1). Our finding showed a marked upregulation of CITED2 in the DeAC group. Given that CITED2 is a functional repressor of HIF-1α and that we previously showed a loss of cardioprotection in DeAC (the rat heart tissue have had a significantly larger infarct area after the exposure to ischemia/reperfusion treatments), we can hypothesize that one of the reasons for the loss of cardioprotection in this group was the HIF1-1α repression following CITED2 upregulation.

Another interesting finding in our study was the significant transcriptional downregulation of genes encoding histones in the DeAC and ReAC experimental groups.

The marked downregulation of histone H1 transcripts in DeAC and ReAC, particularly in reacclimation conditions, may stem from either the role of histone H1 as a chromatin conformation regulator or for the need to change histone H1 levels as a result of DNA damage in stressful conditions.

The finding of nine miRNA transcripts significantly expressed in the array, six of which are regulators of genes whose expression was significantly influenced by our experimental conditions, indicate that miRNA-epigenetic regulation of genes may be involved in cytoprotection and memory acting by different mechanisms.

Taken together, we can conclude that AC activates different cellular pathways that induce long-lasting histone modifications in the chromatin structure of large group of cytoprotective and chromatin regulation genes. These changes may allow constitutive binding of transcription factors (as HSF1) even under ―non-stressful‖ conditions (such as DeAC), promoting cytoprotective ―memory‖. This research is of relevance in the present era, which is characterized by more frequent heat waves and greater instability along climatic border zones, causing changes in the composition of ecological communities, species diversity, and interfering with quality of life. Understanding adaptive limitations is, therefore, of prime importance among the populations at risk.

The published articles based on this work (Attached to the Supplements):

1. Tetievsky A., Cohen O., Eli-Berchoer L., Gerstenblith G., Stern MD., Wapinski I., Friedman N., and Horowitz M., Physiological and molecular evidence of heat acclimation memory: a lesson from thermal responses and ischemic cross-tolerance in the heart Physiol. Genomics (2008) 34; 78 – 87.

2. Tetievsky A, Horowitz M. Posttranslational modifications in histones underlie heat acclimation- mediated cytoprotective memory. J Appl Physiol. (2010) Nov;109(5):1552-61.

List of Abbreviations

Activatin transcription factor ATF

B cell lymphoma x long BCL-xl

cAMP Binding Protein CREB

cAMP Response Element CRE

Chromatin immunoprecipitation ChIP

cAMP response element Cre

Deacclimation DeAC

Dihydrofolate reductase DHFR

Colonic Temperature Tc

Heat acclimation HA

Heat Stress HS

Heat Shock Element HSE

Heat Shock Factor-1 HSF1

Heat Shock Proteins HSP

Heat Shock Proteins with molecular weight 70 kDA HSP70

Heat Shock Proteins with molecular weight 90 kDA HSP90

Histone Acetylases HATs

Histone Deacteylases HDACs

Hypoxia inducible factor 1α HIF- 

Ischemic Contracture IC

Ischemia/Reperfusion I/R

Long Term Heat Acclimation LTHA

Mitogen-activated protein kinase MAPK

MiRNA MicroRNA

Mitogen- and stress-activated protein kinase1 MSK1

Non Heat Stress N-HS

Reverse Transcriptase Polymerase Chain Reaction RT-PCR

Relative quality RQ

Short Term Heat Acclimation STHA

Stroke volume SV

60-kDa HAT-interactive protein TIP60

Table of Contents

Abstract

List of abbreviations

Table of contents…………………………………………………ix-xiv

1. Introduction…………………………..…1-23

1.1 Heat Acclimation – the general concept……………………….…………1

1.1.1 Heat Acclimation – mediated cross tolerance…………………3

1.1.2 Heat acclimation and the heart…………………5

1.1.3 Heat acclimation molecular cytoprotection…………….5

1.1.3.1 Heat shock response and Heat shock proteins…………………….6

1.2 The loss of heat acclimation and its re-induction: a lesson from experimental animal model………………………9

1.2.1 Deacclimation and reacclimation: physiological evidences of acclimatory memory……………………………………10

1.2.2 Deacclimation and reacclimation: molecular evidences of acclimatory memory…………………………13

1.2.3 Physiological and molecular evidence of heat-acclimation memory in the rat heart - a summary of the previous results………….15

1.3 Acclimatory memory and epigenetics…………………………16

1.3.1 Epigenetics, chromatin remodeling and histone modifications..……..……16

1.3.2 Epigenetics – a link between the environment and gene functions…………...19

1.3.2.1 Histones acetylation and phosphorylation as a transcriptional regulator of heat shock proteins…………..…20

1.3.2.2 Epigenetic mechanisms and molecular memory formation of environmental stress…………………………….21

1.3.4 MicroRNA – epigenetic regulators of gene expression………..……...22

1.4 Working hypothesis…………………………………23

2. Aims of Study……………………….24-25

3. Methods…………………………………………26-38

3.1 Animal and maintenance………………………26

3.2 Experimental groups…………………………………26

3.3 Heat Stress (HS)………………………….27

3.4 Tissue collection………………………………28

3.5 Evaluations of protein levels………………………….28

3.6 Evaluation of mRNA levels by quantitative real-time

reverse-transcription polymerase chain reaction (Real-time PCR)……..….….…....29

3.6.1 Total RNA isolation………………………………30

3.6.2 Double-stranded cDNA synthesis…………………………………30

3.6.3 Real-Time (RT) PCR reaction…………………31

3.7 Chromatin immunoprecipitation (ChIP) analysis…………………31

3.8 Rat whole genome mRNA microarray experiment……34

3.8.1 Isolation, quantitation and qualification of RNA…35

3.8.2 Microarray processing…………………………………………36

3.8.3 Data analysis and bioinformatics…………………37

3.9 Statistical analysis……………………………………37

4. Results…………………………………………39-80

4.1 Characterization of hsp70, hsp90 and HSF1 transcript levels………………39

4.1.1 hsp70 and hsp90 transcription levels in normothermic

conditions and during acute heat stress………………………………39

4.1.2 Transcriptional activity of hsf1 in the experimental groups……………….…42

4.2 Chromatin modifications and transcriptional control

of hsp70 and hsp90 genes……………………………………43

4.2.1 Bioinformatic characterization of HSF1 binding sites at hsp70 and hsp90 promoters……………………………………43

4.2.2 Histones H3 and H4 acetylation at the HSE sites of the hsp70 and hsp90 promoters and TIP60 acetyltransferase recruitment………..45

4.2.2.1 Histone H3 and histone H4 acetylation at the HSE hsp70 and hsp90 promoters………………………………………45

4.2.2.2 The level of histone acetyltransferase

(HAT) TIP60 recruitment to the hsp70 and hsp90 promoters…………….…….47

4.2.3 Histone H3Ser10 phosphorylation at the HSE sites of the hsp70

and hsp90 promoters, MSK1 recruitment and protein level…………50

4.2.3.1 Histone H3Ser10 phosphorylation levels at hsp70 and hsp90 promoter………………………………50

4.2.3.2 The level of MSK1 recruitment to the hsp70 and hsp90

promoters and activated MSK1 protein levels measurement……………51

4.2.4 Transcription factor HSF1 binding to the HSE of the hsp70 and hsp90……………………………………53

4.2.5 Summery of the ―Chromatin regulation and transcriptional control of hsp70 and hsp90 genes‖ chapter……………….56

4.3 Rat whole transcriptome screening and identification of pathways involved in the generation of

xii

heat-acclimation molecular ―memory‖………………57

4.3.1 Microarray analysis of gene clusters…………..………………57

4.3.2 Identification of transcriptional regulators of the significantly expressed genes……………………………………71

4.3.2.1 Phospho-CREB protein level……………………72

4.3.3 Evidence for MicroRNA involvement in AC- associated gene expression changes and memory…………………73

4.3.4 Real Time RT-PCR verification of specific genes significantly expressed in array……….……………………77

4.3.5 Summery of the ―Rat whole transcriptome screening and identification of pathways involved in the generation of heat-acclimation molecular memory‖ chapter…………………………79

5. Discussion…………………………………………81-108

5.1 Characterization of hsp70, hsp90 and hsf1 transcript levels and the link to memory formation………………………………………81

5.2 Posttranslational modifications in histones underlie heat acclimation-mediated cytoprotective memory……………………85

5.3 Identification of pathways involved in the generation of heat-acclimation molecular ―memory‖……………………..……………92

xiii

5.3.1 Short-term heat acclimation (AC2d group) switching on an altered expression of genes encoding DNA damage repair…………...... 93

5.3.2 Identifications of the core genes significantly expressed in DeAC and ReAC groups………………………94

5.3.3 Heat deacclimation (DeAC) and reacclimation (ReAC)

linked to significant changes in histone transcription………………….99

5.3.4 miRNA may be involved in transcriptional regulation

of target genes in DeAC and ReAC groups……………………………..102

5.4 General conclusions……………………………105

5.5 Significance and future possible directions……………………106

5.5.1 The significance of the study…………………106

5.5.2 Future directions……………………………107

References…………………………109-116

Abstract in Hebrew…………………………………117-125

Supplementary material ……………………………126-181

xiv

1. Introduction

Heat Acclimation (AC) is a reversible ―within lifetime‖ mechanism of phenotypic adaptation to long – term elevations in environmental temperature that evolves via a continuum of temporally varying processes (44). Successful AC is brought into play by enhanced thermal endurance and resistance to temperature extremes, collectively delaying the onset of heat injury. AC may also reinforce or interfere with the ability to combat other environmental stressors, among which are ischemia-reperfusion in the heart (58) and hyperoxia in the brain (4). Although several studies have dealt with the loss of AC, only few have explored the processes of its reinduction. The reinduction of AC is markedly faster than the initial/original AC session, and the enhanced physical fitness promotes slow de-acclimation (De-AC) and rapid re-acclimation (Re-AC) (82, 83) . Given that animal studies substantiated that the underlying acclimatory processes are molecular (40), we hypothesized that reinduction of AC involves molecular processes as well. My goal in this study, was to test the hypothesis that faster reinduction of AC implicates ―molecular memory‖ and involves long-lasting histones modifications and whole-genome transcriptional changes.

1.1 Heat Acclimation – the general concept

Heat acclimation (AC) is a conserved phenotypic adaptive response to a prolonged transfer to higher ambient temperatures that confers protection against acute heat stress, delays thermal injury, characterized by decreased heat production, core temperature, and heart rate (37, 44). AC is a biphasic process (Fig. 1.1) involving cross-talk between the peripheral effector organs and central, autonomic control. In homeotherms, which adjust constant body temperature independently of the surrounding temperature, an apparent acclimated state (e.g., reduced heart rate, enhanced evaporative cooling, and increased thermal endurance) emerges shortly (2–5 days) after exposure to acclimating conditions. The development of

the new acclimatory homeostasis, however, is a prolonged process (3–4 weeks) (34, 37, 44).

Fig. 1.1: Phases of heat acclimation in the rat acclimation model. Heat acclimation (achieved by continuous exposure to 34⁰C) is a bi-phasic process. STHA—Short Term Heat Acclimation (Phase I); LTHA — Long Term Heat Acclimation (Phase II). Adapted from Horowitz M.,1998 (33).

At the onset of AC (Phase I — Short-Term Heat Acclimation [STHA]), an increased excitability of the autonomic nervous system compensates for the impaired cellular performance resulting from altered signaling pathways. This phase is dominated by changes in the cell membrane, leading to a desensitization of G-protein-coupled receptors (49), their associated signaling pathways including Ca2+ signals, and target genes sensitivity (48). When acclimatory homeostasis has been attained (Phase II — Long Term Heat Acclimation [LTHA]), metabolic alterations improve cellular function leading to enhanced efficiency — namely an increased effector-organ-output/excitation-signal ratio, suggesting decreased neural excitability (33, 47). During the second acclimatory phase, enhanced integrative physiological mechanisms determined by temperature-adaptive shifts in gene expression, expand the dynamic thermoregulatory range (34). Conceptually, the process of AC (in homeotherms) represents a transition from an early transient, ‗‗inefficient‘‘ to very ‗‗efficient‘‘ cellular performance when acclimatory homeostasis has been reached (37). Figure 1.2 illustrates the conceptual model of AC in the mammalian acclimation model (rats).

Fig. 1.2: Heat acclimation: A conceptual model in rat. During the initial phase of heat acclimation (STHA – short term heat acclimation) accelerated autonomic excitability compensates for impaired effector responsiveness. This phase triggers several molecular and metabolic pathways leading to adaptive changes in gene expression during the next phase – the long-term heat acclimation (LTHA). Tcore – Core temperature, EO – Effector/Organ. Adapted from Horowitz M., 1998 (33).

1.1.1 Heat Acclimation – mediated cross tolerance

An inseparable feature of AC is the protection conferred against a diversity of novel stressors (namely, stressors to which the subject/animal, was not previously exposed). AC has been shown to convey improved cardiac mechanical and metabolic performance and reduced injury upon ischemic-reperfusion insult in the heart. Levi et al. (58) showed that

following total ischemia, acclimated hearts displayed enhanced preservation of the cardiac ATP pool and a delayed decline in intracellular pH in contrast to the non-acclimated hearts. AC has also been shown to induce cross tolerance toward exposure to traumatic brain injury (103, 105), hyperoxia (3), and ionized irradiation (43). Several studies in our laboratory, suggest that acute novel stressors exploit the protective networks enhanced by AC (36, 39). Given the diversity of AC cross-tolerance events, however, in terms of both stressors and organs, signaling pathways are likely to be shared by all stressors and combined with stress-specific (and/or organ-specific) protective pathways, provide cross-tolerance (36). This phenomenon is demonstrated in Figure 1.3, which describes stress-associated gene profiling of rat heart tissue subjected to ischemic reperfusion insults or to a whole-body heat stress. Nine percent (9%) of the genes responded to both heat stress and ischemic reperfusion insults, mostly anti-apoptotic, anti- oxidative and heat shock chaperone genes. Two additional clusters were stress specific; 38% and 13% for ischemia/reperfusion and heat stress, respectively. The cross talk between the shared and the stress-specific genes provided the cross-tolerance event (36).

Fig. 1.3: The genomic basis of AC cross tolerance. Signaling pathways, shared by all stressors and combined with stress specific (and/or organ specific) protective pathways. HS, heat stress; I/R, ischemic reperfusion insult; NC, nonsignificant response. Adapted from Horowitz et al. 2007 (36).

1.1.2 Heat acclimation and the heart

In the heart, AC produces favorable adaptations in mechanical and metabolic performances. At the systemic level, this adaptation is manifested by a lowered heart rate and greater compliance and stroke volume (SV), which together lead to increased cardiac work efficiency. The greater SV matches increased venous return resulting from the AC- induced plasma volume expansion (126). Using isolated perfused hearts and cardiomyocytes, in tandem with molecular tools, Kaspler et al. (47) have shown that intrinsic changes leading to greater Ca2+ transients and contractile force take place. A reprogramming of genes associated with excitation– contraction coupling and calcium regulatory proteins underlie these changes (50). An important beneficial effect of AC is the development of ―cross-tolerance‖ against oxygen supply/oxygen demand mismatching and its consequences. During global ischemia, there is a delay in the onset of ischemic contracture, attenuated drop in ATP reserves and in pH. Upon reperfusion insult, AC cross-tolerance is manifested by improved mechanical and metabolic performance and reduced tissue injury (36, 59). In the single isolated myocyte, an imbalance between glycogen stores and the rate of anaerobic glycolytic energy consumption, leading to ATP depletion, results in a sudden rigor (106). This discrete event enabled to measure the time to onset of ATP-depletion rigor and use the result as an additional (metabolic) criterion for acclimation status and cross- tolerance. We have demonstrated that the average time to contracture of the AC myocytes population after exposure to anoxic conditions was profoundly longer than in control myocytes (114).

1.1.3 Heat acclimation molecular cytoprotection

Our knowledge, particularly in homeotherms, of genomic responses associated with AC is sparse. Accruing data on the changes in cellular proteins suggest to us that the induction of the AC phenotype involves the transcription of genes encoding both constitutive proteins and stress-inducible molecules (42, 67, 74, 102). The cellular functions associated with these genes represent predominantly (a) anti-apoptosis and the maintenance of protein integrity networks, (b) maintenance of DNA and chromatin integrity, (c) genes linked to metabolic processes, (d) reactive oxygen species (ROS) scavengers, and (f) molecular

stress chaperons, among which, those encoding the heat-shock proteins (HSP‘s) family have been the most extensively studied in mammalian species (76).

1.1.3.1 Heat-shock response and heat-shock proteins

The heat-shock response (HSR) phenomenon was originally discovered in the salivary glands and other tissues of Drosophila melangoster by Ritossa in 1962 (92). This discovery led to the finding that transient sublethal heat shock, which increases body temperature by about 5⁰C above normal, leads to the synthesis of new proteins called – the ―heat shock proteins‖ (HSP). HSPs belong to the functional class of molecular chaperones, which assist to correct the non-covalent assembly of other polypeptide-containing structures in vivo, but are not components of these assembled structures when performing their biological functions (76, 77). The chaperones recognize and selectively bind non-native proteins under physiological and stress conditions and inhibit incorrect interactions and the formation of nonfunctional structures (40, 43). Chronic insults, such as long-term/continuous exposure to heat, result in a constitutive elevation of cellular HSP, in the redistribution of HSP isoforms, and in an alteration of their response rate to other stresses (43). The HSPs are grouped into families based on molecular weight and amino acid sequence (77). In mammalian cells, the five major HSP families are HSP 110, 90, 70, 60, and 27 (76, 77). Among the HSPs, the inducible isoforms of the 70-kDa family (HSP70) are considered the most responsive to heat stress (as well as to ischemia) and to a variety of pharmacological stresses (40). A number of studies have shown that whole body heat stress 24 h before the onset of myocardial ischemia HSP70 accumulation is protective against ischemia/reperfusion injury (40, 66, 67). In addition, adenovirus-mediated transfer of HSP70 might protect against ischemia, hyperthermia, and septic shock both in cell culture and at the whole organism level (73). HSP70 can interfere with the apoptotic program by the binding and localization regulation of pro- and anti-apoptotic factors and inhibition of the DNA-binding activity of the transcription factor NFkB involved in pro-oxidative and pro- inflammatory related apoptosis (123). A study by Maloyan A. (67) from our laboratory, revealed that AC significantly increases the cellular reserves of the inducible HSP70 in rat

hearts (Fig. 1.4). In addition, elevated HSP70 was detected in brains, salivary glands, and the evaporative cooling organs of heat-acclimated rats (3, 43).

Fig. 1.4: HSP72 levels in left ventricle of hearts excised from control (C) and AC rats (1, 2, and 30 days). Top: a representative set of blots of various groups (enhanced chemiluminescence detection) is presented. Bottom: bar graph showing HSP protein levels in different groups of rats, normalized to commercial HSP control sample. Adapted from Maloyan et al., 1999 (67).

The role of HSP70 as a protein chaperone is often associated with another important chaperone – Heat shock protein 90-kDA (HSP90). Both chaperones share co-chaperones through which they can transitorily associate. HSP90 assures the stability of hundreds of proteins, known as HSP90 client proteins. The broad clientele of HSP90 include structurally and functionally different proteins that include a growing range of protein kinases, a variety of nuclear hormone receptors, cell surface receptors, such as Her2/neu, transcription factors, such as Hypoxia Inducible Factor (HIF)-1α, and many others (53). In addition, HSP90 participates in the conformational regulation of signal transduction molecules, such as tyrosine kinase and steroid hormone receptors, intermediate filaments, microtubules, and actin microfilaments. The protein is also implicated in the maturation and activation of cystic fibrosis regulator (CFTR), nitric oxide synthase, and telomerase. Queitsh et al. (88) have hypothesized that HSP90 functions as a

buffer for protein conformational diversity in the cell. This hypothesis ascribes to the HSP90 system a pivotal role in evolution: allowing the accumulation of agenetic diversity that will permit the emergence of adaptive changes in the face of selective pressure. Heat shock genes hsp70 and hsp90 expression is regulated on the transcriptional level by the activation of the pre-existing transcriptional activator – the heat shock factor (HSF). In vertebrates, four members of the HSF family have been identified, and of these, HSF1 is ubiquitously expressed and conserved (76, 77). In the resting state, HSF1 seem to be bound by HSP70 or HSP90. When the cell is stressed, the HSF1 monomer is released from the HSP70/HSP90 – HSF1 complex, trimerizes in the cytoplasm, activated by phosphorylation and then migrates to the nucleus. In the nucleus HSF1 binds to the heat shock response element (HSE) found in the promoters of hsp70 and hsp90 and other heat shock genes initiating gene transcription (Fig. 1.5) (77). Several factors can lead to HSF1 activation: 1.The accumulation of stress-mediated unfolded proteins shifts the equilibrium from the binding of HSF1 to HSP70 or HSP90 toward the binding of HSP70 or HSP90 to the unfolded proteins. This binding leads to release of HSF1 and further activation. 2. HSF1 can be activated by protein kinase C and tyrosine kinase. 3. In the heart, the increase of diastolic intracellular calcium, heart-wall stretching, and alpha-adrenoreceptor stimulation lead to HSF1 activation and enhances gene transcription. In addition, HSF1 can be activated into an intermediate state, in which it binds to HSE sequences without stimulating gene transcription [Maloyan and Horowitz unpublished].

HSF1-HSP70/HSP90 HSP70/90 – unfolded proteins Stress factors

HSP70 HSP90 HSF1 HSP70

P HSP70 mRNA

Trimerization P HSF1 HSE HSP's gene

Fig. 1.5: Schematic representation of the heat shock response (HSR). Different extracellular and intracellular stress factors activate HSF1 and subsequent HSPs gene expression. The central process in HSF1 activation is the equilibrium between the binding of HSP70/HSP90 to HSF1 and to stress-mediated unfolding proteins. Any increase in the presence of unfolding proteins shifts the equilibrium towards HSP70/90 – unfolded protein complexes, thereby releasing free HSF1 monomers , which can be subsequently activated. HSE – heat shock element, HSF – Heat shock factor, HSP- heat shock protein, P – phosphorylation. Taken from Morimoto et al.1998 (77).

1.2 The loss of heat acclimation and its reinduction: a lesson from experimental animal model

If not maintained by repeated challenge, it‘s believed, that AC either gradually decays or is lost. Although multiple studies have disclosed the physiological and molecular processes underlying AC in mammals, few have dealt with the decay and loss of AC, and even fewer have explored the processes of its reinduction. The current study is a direct continuation of the research conducted as a part of my master‘s degree project, named ―Does heat acclimation have a memory?‖ (114). The study characterized the physiological and molecular phenotypes of acclimatory decay and reinduction and revealed the phenomenon of ―Acclimatory memory‖ in the rat heart model. The following sections provide a brief summary of the main results from the previous, basic

study together with the literature background, needed for the understanding of the basis and the starting point of the current study.

1.2.1 Deacclimation and reacclimation: physiological evidences of acclimatory memory

The early pioneering studies, dealing with the decay and loss of AC, were confined to characterizing the physiological phenotype. The AC phenotype, when defined by consensus acclimatory criteria, such as decreased heart rate and rectal temperature, was shown to persist during a long period of 7–20 days of deacclimation (DeAC), depending on whether humid or dry heat acclimation was applied [reviewed by Pandolf (82)]. The author hypothesized that the physiological AC phenotype decays at a rate of a 1 day loss of acclimatized status for every 2 days spent without heat exposure. Reinduction of AC after the DeAC period occurs markedly faster than during the initial/original AC session (4). Interestingly, a similar phenomenon was documented for cold temperature (32) and altitude reacclimation (ReAC) (62), implying that slow DeAC/rapid ReAC is not an AC- specific feature and might be a conserved adaptive response. Taking the Tc (colonic TºC) - plateau, the hyperthermic plateau on which body temperature is regulated during heat stress as an acclimatory signature, we validated that the reinduction of the acclimated phenotype is apparent after only 2 ReAC days following both 1 and 2 months of DeAC at normothermic conditions (24⁰ C) (Figure 1.6A). Moreover, the heating rates of the AC, ReAC(1) and ReAC(2) groups were significantly faster than those of the control, acclimated for 2 days (AC2d) or DeAC groups (114). We used the cross-tolerance mechanism as an additional criterion to characterize de- /reacclimation events. Ischemic and anoxic cross-tolerance (22, 65) upon ReAC was demonstrated at both organ and cellular levels. Figure 1.6B shows infarct area measurements and the infarct-risk area ratio in heart slices from the experimental groups. The area of infarction was 3 and 2.4 times smaller in the AC and 2d reacclimation groups than in the control group, respectively. In contrast, the infarct sizes of hearts from AC2d rats and DeAC groups were not significantly different from the control group average (114). In the single isolated myocyte, an imbalance between glycogen stores and the rate of anaerobic glycolytic energy consumption, leading to ATP depletion, results in a sudden rigor (106).

This discrete event enabled us to measure the time to onset of ATP-depletion rigor and use the result as an additional (metabolic) criterion for acclimation status and cross-tolerance. Similarly to the previous results, a ReAC period of only 2 days was sufficient to reestablish the acclimated phenotype, displayed in cardiomyocytes as an enhanced contractile response and a greater endurance under anoxic conditions, whereas AC2d and DeAC for 30d myocyte populations did not differ from controls (Fig. 1.6C). The physiological criteria used in our previous research provided substantial evidence that reinduction of the AC phenotype (even after 2-mo of deacclimation) only takes 2 d, rather than the 30 d initially required to achieve acclimatory homeostasis, implying that AC involves a memory.

A C

(a)

Figure 1.6 (A-C): Heat acclimation memory – physiological evidences. (A) Tc before and during heat stress (HS) at 41°C. C, controls; AC2d, heat acclimation (at 34°C) for 2 days; AC, heat acclimation for 30 days; DeAC, deacclimation at normothermic conditions for 1 or 2 mo. ReAC (1) and ReAC (2), reacclimation for 2 days following 1 or 2 mo of DeAC. Each symbol represents the average value for the particular time point. The repeated measures 1-way ANOVA model results in significant differences between groups (P = 0.001). *Significant difference of AC, ReAC(1), and ReAC(2) vs. C P = 0.05, Dunnett‘s; n = 5–10 rats per group. (B) Infarct size, expressed as the percentage of infarct size/area at risk (AAR) in isolated hearts subjected to global ischemia (30 min) followed by (40 min). The effect of treatment time point (TTP) on AAR was significant (1-way ANOVA, P = 0.001), *Significant difference of AC and ReAC from C (P = 0.05, Dunnett‘s); n = 5–10 rats per groups. (C) The time elapses from the onset of anoxia to rigor contraction in individual cardiomyocytes. Each panel depicts the distribution curve of time to rigor in cardiomyocytes population from rats subjected to one of the assigned treatments and the average onset time to rigor contraction (means SE, min); n = 6 rats per group. The effects of TTP on the time to rigor was significant (1-way ANOVA, P =0.001). *Significant difference of the time to rigor of AC and ReAC from the C group. P < 0.05, Dunnett‘s. All the graphs taken from Tetievsky et al. 2008 (114).

1.2.2 Deacclimation and reacclimation: molecular evidence of acclimatory memory

Memory in response to a stimulus is considered a reversible regulatory system in which a self- propagating ―on‖ state is generated from common signaling molecules, such that once a system has been ―sensitized‖ by a strong initiating stimulus, a second exposure elicits a markedly faster response (28, 75, 91, 96, 114). Given that underlying heat-acclimatory responses involve molecular processes e.g. upregulated transcription and translation of proteins associated with cytoprotective networks (35), we hypothesize that the faster reinduction of the physiological AC phenotype after the DeAC period may implicate a molecular memory. To test this hypothesis, in our preliminary study (114), we used a stress-cDNA atlas array analysis of altered gene expression in rat heart tissue. The cluster analysis highlighted a very important finding, namely that ~ 37% of the visible stress genes, the majority of which were transcriptionally upregulated, did not resume their preacclimation levels after 1 month of DeAC, despite a return of the physiological phenotype to its preacclimation state (114). Among the stress genes assigned to these clusters were heat shock response (HSR) assemblage genes e.g., the heat shock proteins - hsp70, hsp90 and the anti-apoptotic factor bcl-xl. As described before, the promoter of these genes contains sequences resembling the classical heat shock element (HSE), allowing the transcription factor HSF1 to bind the promoter and to regulate the transcription of the genes during thermal stimuli. Following the results obtained by the stress- array analysis, we have measured the protein levels of HSP70, HSP90, and Bcl-xL genes in de- and reacclimated groups. Both HSP70 (Fig. 1.7A) and Bcl-xL (Fig. 1.7B) maintained this elevated level and did not return to their preacclimation levels throughout one month of deacclimation and 2 days of reacclimation. In contrast, HSP90 (Fig 1.7C), returned to preacclimation levels in the DeAC group but was upregulated significantly following 2 days of reacclimation. In view of the upregulation of the hsp gene transcripts in the DeAC and ReAC groups, we also studied heat shock factor 1 (HSF1), the consensus transcription factor of the hsp genes. Both the array analysis and the protein measurements revealed hsf1 upregulation in AC, DeAC, and ReAC groups (Fig. 1.7D) as compared to the control.

A B

* * * * * *

C D

* * * * *

Figure 1.7 (A-D): HSP70, HSP90, Bcl-xL and HSF1 protein profiles at rat heart. Both HSP70 (A), Bcl- xL (B), and HSF1 (D) maintained an elevated level and did not return to their preacclimation levels throughout one month of deacclimation (DeAC) and 2 days of ReAC. In contrast, HSP90, (C) an essential cytoprotective component shows a return to preacclimation levels in the DeAC group but was upregulated significantly following 2 d ReAC. Protein levels are presented as protein/actin ratio. Each bar represents mean SE, n = 5–10 rats per group. *Significant difference (P = 0.05, Dunnett‘s) of the protein level vs. basal C level. Taken from Tetievsky et al. 2008 (114).

In conclusion, our findings showed that the DeAC phenotype is characterized by a mismatch between the greater HSP70, BCL-xL and HSF1 reserves and the loss of cardioprotection (114).

1.2.3 Physiological and molecular evidence of heat-acclimation memory in the rat heart - a summary of the previous results

In my previous work (114), aimed to substantiate the phenomenon of acclimatory memory in the rat heart at the integrative physiological level, and to provide cues for the underlying mechanisms, the main conclusions were: 1. Reinduction of the AC phenotype, even after 2-mo of deacclimation in normothermic conditions, takes only 2 days, rather than the 30 days initially required to achieve acclimatory homeostasis, implying that AC involves memory. As in human studies, in our rat model the rapid return to the AC phenotype was noted in both thermoregulatory responses to heat stress and cytoprotective mechanisms (cross-tolerance). 2. Despite a return of the physiological phenotype to its preacclimation state, after 1 mo of the DeAC period, the gene transcripts did not resume their preacclimation levels, suggesting a dichotomy between physiological and molecular phenotypes in this system. HSP70, BCL-xL and HSF1 protein profiles followed the observed dichotomizes genomic response. In contrast, HSP90, an essential cytoprotective component, mismatched protein levels upon DeAC. The time interval that occurs between the priming event and the subsequent stress exposure in which the altered organism response is manifested is of considerable interest. There appears to be a mechanism for storing information from previous exposure indicating that some parts of the organism responses to stress are more complicated than mere signaling cascades set off by stress signals. The uniform activation of the similarly responding gene clusters upon De-/ReAC implied that ReAC phenotypic plasticity may be associate with an epigenetic transfer of information - remodeling of chromatin in response to heat acclimation and predisposing to cellular “acclimatory memory”, which we define as “a genetic or biochemical modification occuring after stress exposure that causes future responses to future stresses to be different”.

1.3 Acclimatory memory and epigenetics

1.3.1 Epigenetics, chromatin remodeling, and histone modifications

Epigenetics is defined as "the study of stable genetic modifications that result in changes in gene expression and function without a corresponding alteration in DNA sequence". Epigenetic information can mean instructions e.g. ―do not transcribe this gene‖ or “transcribe this gene if…” and is associated with chromatin remodeling via molecular and biochemical processes that maintain the chromatin-DNA package either in active or silent states (119). Chromatin is the combination of DNA and proteins that make up the contents of the nucleus of a cell. The primary functions of chromatin are: to package DNA into a smaller volume to fit in the cell, to strengthen the DNA to allow mitosis and meiosis and prevent DNA damage, and to control gene expression and DNA replication. The protein components of chromatin are histones. Histones are highly alkaline proteins found in eukaryotic cell nuclei that package and order the DNA into structural units called nucleosomes. Five major classes of histones exist: H1/H5, H2A, H2B, H3, and H4. Histones H2A, H2B, H3 and H4 are known as the core histones, whereas histones H1 and H5 are known as the linker histones (Fig. 1.8). Histones undergo post-translational modifications that alter their interaction with DNA and nuclear proteins (118- 120) .

Posttranslational modification of the NH2-terminal tails of histone proteins H3 and H4 that protrude from the nucleosome is the most common form of chromatin remodeling (119). Modifications of the tails include methylation, acetylation, phosphorylation, ubiquitination, SUMOylation (Small Ubiquitin-like Modifier protein), citrullination, and ADP-ribosylation. Histone modifications act in diverse biological processes such as gene regulation, DNA repair, chromosome condensation (mitosis) and spermatogenesis (meiosis) and can regulate chromatin condensation state and in turn, the transcription factors binding to the gene promoter and its further transcription. Other factors, however, such as the association and dissociation of transcription factor complexes with chromatin can affect chromatin condensation also (56, 119).

Figure 1.8: Schematic representation of the assembly of the core histones into the nucleosome. The basic unit of chromatin is the nucleosome (a), which is folded through a series of successively higher order structures to eventually form a chromosome. Nucleosomes are composed of 146 bp of DNA around a histone octamer, consisting of two copies each of the core histones H2A, H2B, H3, and H4 with a linker H1. (b) Crystal structure of the nucleosome depicting the interaction of DNA with histones. Adapted from Levinson et al. (56).

DNA coding genes that are actively transcribed ("turned on") are more loosely packaged (referred to as euchromatin) and are found associated with RNA polymerases and transcription factors whereas DNA coding inactive genes ("turned off") are found associated with structural proteins and are more tightly packaged (heterochromatin) usually not allowing the binding and TFs transcription (Fig. 1.9) (57, 120).

Figure 1.9: Histone modifications and chromatin remodeling. Eukaryotic genes are located on multiple linear chromosomes packed into a complex with histone proteins to form chromatin. Histones are the primary regulator of the chromatin structure, which is dynamic, existing in either a heterochromatin (condensed) or euchromatin (open) state. Whether chromatin is active or inactive is largely dependent on the post-transcriptional modifications of the histone tails, including acetylation, methylation, and phosphorylation. Taken from Sigmaaldrich.com.

Histone methylation is generally associated with transcriptional repression, whereas histone acetylation and phosphorylation are believed to be associated with chromatin opening and the transcriptional initiation (119, 120) . Histone acetylation: The acetylation of the N-terminal tails is the most well studied modification of the core histones. The histones are reversibly acetylated and deacetylated on lysine or arginine residues in the N-terminal tail as part of gene regulation. The acetylation of histone H3 lysines 9, 14, 18, and 23 and histone H4 lysines 5, 8, 12, and 16 is associated with dynamic transcriptional regulation and controlled by intracellular signaling. Typically, this reaction is catalyzed by enzymes with "histone acetyltransferase" (HAT) or "histone deacetylase" (HDAC) activity (119). One mode of thinking is that the tendency of acetylation to be associated with "active" transcription is biophysical. Because lysine normally has positively charged nitrogen at its end, it can bind the negatively charged phosphates of the DNA backbone. The acetylation event converts the positively charged amine group on the side chain into a neutral amide linkage. This reaction removes the positive charge, thus loosening the DNA from the histone. As a result, transcriptional factors can bind to the DNA and allow transcription to occur. This is the "cis" model of epigenetic function. In other words, changes to the histone tails have a direct effect on the DNA itself. Another model of epigenetic function is the "trans" model. In this model, changes to the histone tails act indirectly on the DNA. For example, lysine acetylation may create a binding site for chromatin modifying enzymes (and basal transcription machinery as well). This ―chromatin remodeler‖ can then cause changes to the state of the chromatin. Acetylation could act in this and the previous way to aid in transcriptional activation (56, 119, 120) . Histone phosphorylation: Several histones are subject to phosphorylation, and this modification is associated with a large-scale chromatin reorganization during such processes as mitosis, apoptosis, and DNA repair. Recent studies have outlined a strong possibility that histone H3 phosphorylation plays a role in the dynamic regulation of transcription. Particular attention has been focused on histone H3 phosphorylation on serine 10 (Ser 10) because it is found to be connected with chromatin opening and transcriptional activation (17). Histone H3Ser10 phosphorylation is found to rapidly increase in quiescent cells during mitogenic stimulation (9) as well as during immediate–early gene induction via the epidermal growth factor (EGF)-signaling pathway (63). In addition, in vitro experiments have suggested that EGF-stimulated H3 phosphorylation may act as a signal for histone acetyltransferase binding and the subsequent acetylation of a particular locus

during transcription initiation, implying crosstalk between post-translational-histone acetylation in transcriptional gene regulation. Phosphorylation of histone H3Ser10 is mediated by mitogen- and stress-activated protein kinase1 (MSK1), which is positioned downstream of both ERK and mitogen-activated protein kinase 1 (MAPK) p38 pathway (14).

1.3.2 Epigenetics – a link between the environment and gene functions

Rather than acting as static landmark, epigenetic changes can function as dynamic and transient operational marks supporting specific steps in diverse processes throughout the mammalian genome. The effect of specific environmental factors on the epigenetic status of adult organisms has been reported [reviewed in (1)]. The environmental factors that could affect epigenetic status during adult life can be divided mostly into three categories: diet, environmental stressors (such as temperature), diseases and pharmacological treatments. We must also take into account that because higher organisms are composed of multiple tissues, the effects of environmental factors on the epigenotype of an organism can be tissues specific (1). Previous studies have shown that the amount of dietary folate, methionine, and selenium intake is associated with the epigenetic status of the organism. Recently, other substances occurring naturally in some foods, like butyrate in cheeses, diallyl disulphide in garlic, and sulphoraphane in broccoli, have been identified as HDAC inhibitors, and a putative role for some of these compounds has been proposed in cancer chemoprevention through the disruption of the uncontrolled progression of cell cycle or by the induction of apoptosis via increased acetylation and derepression of specific genes. The effect of diet on the epigenetic status of an organism can be so important that it has even been described that a high-fat diet can be associated with promoter DNA hypermethylation of specific tumor-suppressor genes (1). Epigenetic factors may also be susceptible to alteration by pharmacological substances. One example is diethylstilbestrol. This drug has been shown to alter the expression of DNA methyltransferases and methylation of genomic DNA. Another example is sodium valproate, an anticonvulsant classically used to treat some mental disorders that is now a well-known and potent inhibitor of histone deacetylases. Several antibiotics and cannabinoids, heroin and cocaine also induce significant epigenetic alterations (1).

The organism‘s living environment can determine the level of exposure to many environmental factors and environmental stresses that are potentially able to alter epigenetic status. For instance, the environmental pollutants chromium, cadmium, and nickel reduce methylation levels by inhibiting the activity of DNA methyltransferases (1). Likewise, oxidative and osmotic stresses and UV radiation can influence epigenetic gene regulation by inducing chromatin modifications (1). In mammals, the heart of any circadian clock lies in the transcription– translation feedback loop. The most salient phase-resetting environmental stimulus is light, and pulses of light induce changes in the transcription of several genes that comprise the molecular clock. Epigenetic mechanisms seem to be associated with this regulation, as discrete pulses of light induce increases in acetylation of histones H3 and H4 associated with the promoters of the circadian clock genes (56).

Collectively, this compendium of evidence suggests that epigenetic mechanisms bridge environmental influences and gene expression, thereby reversibly affecting physiological responses to changing environmental conditions.

1.3.2.1 Histones acetylation and phosphorylation as a transcriptional regulator of heat shock proteins

Environmental stressors induce distinct cellular responses of particular stress-responsive genes, which play pivotal roles in the stress response of eukaryotic cells. Upon stress, the expression patterns of the stress genes are altered, redirecting cellular energy and metabolic resources from rapid growth toward efficient stress protection. Dynamic regulators of reprogramming gene expression are essential in these processes. Histone acetylation and phosphorylation play a central role in the epigenetic regulation of stress- response genes and genes from the HSP family, due to the relatively dynamic character and reversibility of their expression. Nowak et al. (81) demonstrated an elevated H3Ser10 phosphorylation in transcriptionally activated heat shock loci in Drosophila‘s chromosome puffs after heat shock and in mammalian hsp70 promoter, where it was essential for transcriptional activation. Additional evidence, showed that hsp70 promoter phosphorylation may act as a recognition site for subsequent acetylation and transcriptional activation. An important early study by Thomson et al. (116) focused on hsp70 in mouse fibroblasts exploring whether hsp70 is regulated by histone H3 phosphorylation and histones H3 and H4 acetylation in a stimulus- dependent way. The results revealed that heat shock and sodium arsenite both strongly induce

hsp70, with differing requirements for p38 MAPK activation and with distinct histone modifications at the gene. Heat shock produces acetylation of histone H4 at hsp70 chromatin, whereas arsenite produces both H4 acetylation and H3 phosphorylation at this gene. In addition, the study used Hsf1-/-cells to show that this factor is responsible for targeting histone H4 acetylation to hsp70 chromatin.

1.3.2.2 Epigenetic mechanisms and molecular memory formation of environmental stress

Several classic examples illustrate the importance of epigenetic mechanisms in information storage at the cellular level. These studies indicate that epigenetic mechanisms are not only a dynamic transcription regulation but also can be widely used for the formation and storage of cellular information in response to transient environmental signals. Plants are known to be adept at altering their physiology and metabolism in response to prior experience. The process of ―priming‖ or ―hardening‖ involves prior exposure to a biotic or an abiotic stress factor making a plant more resistant to future exposure. Sung et al. (112) proposed that epigenetic changes are responsible for this phenomenon. In mammals, Reul and Chandramohan (91) discovered that in rats and mice, elevated and long- lasting phospho-acetylation of histone H3 occurs in neurons following exposure to physiological stress in a functional context, i.e., in those neurons involved in stress-related memory formation. The authors observed that the increase in histone H3 phosphoacetylation in dentate neurons after physiological stress causes a distinct and transient transcriptomic response. These observations indicated that epigenetic tagging of the genome occurs during consolidation of long-term memory. Interestingly, a different form of long-term memory — latent inhibition — has been associated with an altered acetylation of histone H4 but not histone H3 (55). In contrast with those findings, Hall et al., (28) demonstrated that C. elegance exposed to stressful conditions during their early life history show genome-wide histone tail modifications, leading to post-stress chromatin remodeling and gene expression. Those findings indicate that modifications of the genome as a consequence of an early experience may contribute in part to the stress memory and associated with specific patterns of histone modification.

1.3.4 MicroRNA – epigenetic regulators of gene expression

The miRNAs are ~22 nucleotides-long RNA molecules encoded in the genome that can have a profound effect in controlling gene expression. The RNA is transcribed by RNA polymerase II (Pol II) into primary miRNAs, and then processed in the nucleus by the RNase III Drosha and DGCR8 (microprocessor complex) into the precursor miRNAs (70) Precursor miRNAs are structured as imperfect stem-loops, and they are exported into the cytoplasm. When binding to its target mRNA with complete complementarity, the miRNA can lead to degradation of the target. miRNAs can also bind to their targets with incomplete complementarity, often in the 3' UTR regions, which leads to the translational suppression of their target genes by a mechanism that has yet to be completely explained. Each miRNA is predicted to have many targets, and each mRNA may be regulated by more than one miRNA. Currently, there are more than 460 human miRNAs known (15, 70, 72). Diverse miRNAs have been gaining considerable attention as regulators of gene expression during cell stress as well as in neuronal physiology and plasticity, including excitability, synapse maturation, synaptic function and neurodegeneration (72). Recent evidence also suggests that miRNAs affect histone modifications. Maison et al. (64) showed that RNase treatment can abolish the localization of methylated H3 lysine 9 and HP1 to pericentromeric chromatin. Fukagawa et al. (26) showed that Dicer-related RNAi machinery is necessary for the formation of heterochromatin structure. Because siRNAs (small interfering RNAs) and miRNAs are closely related, miRNAs could also play important roles in controlling DNA methylation and histone modifications by regulation of key chromatin- modifying enzymes such as the DNA methyltransferases (24), histone deacetylases (80) and lysine methyltransferases (124). Taken together, miRNAs can be considered important players in the epigenetic control of gene expression.

1.4 Working hypothesis

Our recent work (114) on AC characterized, substantiated, and defined the phenomenon of ―acclimatory memory‖. The dichotomy between the return of the physiological vs. the molecular phenotype to pre-acclimation status upon a DeAC period of 1 and 2 months suggests that the molecular acclimation program has selective and long-lasting effects that can rapidly re-activate the pathways linked to the protective physiological responses and cross-tolerance mechanisms. Given our preliminary findings and the concept that chromatin is considered the ―physiological template of eukaryotic cells‖, I suggest that chromatin remodeling regulated by histone modifications may play a central role in adaptive memory and transcription during adjustments to long – term and fluctuating environmental perturbations.

2. Aims of Study

The outcomes of our previous results (114) are the awareness of physiological ―acclimatory memory‖ associated with molecular processes. Consequently, the major aims of present study as follows:

1) To test our hypothesis that acclimation memory stems from epigenetic adaptations and to unravel the mechanisms involved. 2) To exploit the deacclimation/reacclimation (DeAC/ReAC) model and discover the “core gene clusters” and their master regulators that are possibly involved in generating of acclimation memory.

To achieve these goals my specific aims were as follows:

For aim 1: (1)1 To study the transcriptional kinetics of genes of interest: hsp70, hsp90 and hsf1 to support the hypothesis that transcriptional dynamics is a part of the acclimatory memory repertory.

(1)2 To focus on the molecular and biochemical processes that maintain the chromatin-DNA package in an active state, allowing the accessibility and binding of transcription factors to DNA recognition sites. Based on relevant information in the scientific literature, we conducted the following analyses to achieve this goal. :  In silico identification of the transcription factor HSF1 binding site (the heat shock element – HSE) to the promoters of hsp70 and hsp90.  The levels of histones H4 and H3 acetylation and histone H3 phosphorylation at the HSE of hsp70 and hsp90.  The levels of specific acetyltransferase TIP60 and kinase MSK1 recruitment to the promoter area of hsp70 and hsp90.  To confirm binding of transcription factors to the euchromatin, as an indication of chromatin opening and initiation of transcriptional events, the measuring of the binding of HSF1 to the heat HSE on the hsp70 and hsp90 genes was performed.

For aim 2: (2)1 To take a genome-wide approach using a whole rat genome array, to screen for alterations in the expression of genes involved with chromatin remodeling and transcriptional regulation following acclimation, DeAC, and ReAC. The following actions will be conducted:  Identification of genes that significantly change their transcriptional behavior in the experimental groups.  In silico clustering of the ―visible‖ genes based on similar expression levels across treatments.  Specific functional analyses of pathways associated with broad scale chromatin remodeling, cellular memory and transcription initiation.  Verification of the mRNA levels of specific genes of interest.  Analyzing possible epigenetic and co-transcriptional regulation of groups of genes by MicroRNA (miRNA) molecules.

Collectively, the aim of this study was to attempt to explore the upstream regulatory events (epigenetic and transcriptional), leading to acclimatory memory, and to provide further insight into the molecular template during physiological states.

3. Materials and Methods

3.1 Animal and maintenance All studies were conducted according to the guidelines of the Institutional Animal Care Committees of the Hebrew University and complied with the guidelines of the National Research Council Guide for the Care and Use of Laboratory Animals (National Institutes of Health Publication no. 85–23, 1996. Males Rattus norvegicus (Sabra strain, albino var.) initially weighing 60-90 gram were fed Ambar laboratory chow with water ad libitum and held under light-dark cycled conditions (12:12 h).

3.2 Experimental groups The animals were randomly assigned to the following experimental groups: heat acclimated for 2 (AC2d) and 30 days (AC); deacclimated for 1 mo (DeAC); reacclimated for 2 days after the 1- mo deacclimation (ReAC), and control, normothermic (C). C and DeAC rats were held at an ambient temperature of 24 ± 1°C; AC2d, AC and ReAC were attained by continuous exposure to 34 ± 1°C and 30%–40% relative humidity, as described in Fig. 3.1 (114).

Fig. 3.1, Protocol bar illustrating experimental plan. C - controls (at 24°C); AC - long-term heat acclimation (30 days at 34°C); DeAC - deacclimation for 1 mo at 24°C; ReAC - reacclimation for 2d at 34°C. Taken from Tetievsky et al. (114, 115). 26

During the different experimental regimes, body temperature was monitored continuously every two days (Table 3.1). The colonic temperature (Tc) profile of the animals during the AC period is a reliable criterion for successful acclimation (40, 41). The Tc was measured using a digital colonic thermistore (Eutech - Instruments, ITS - 90 Singapore), inserted 6 cm deep beyond the anal sphincter.

3.3 Heat Stress (HS) The animals in all experimental groups were randomly subdivided into animals undergoing no further treatments (basal) and those subjected to an acute Heat Stress (HS) treatment. This allows us to test the 2d/AC/DeAC/ReAC status effect both on basal properties and on the superimposed (novel) stress response. For HS treatment, rats were placed in a temperature controlled incubators at 41 C for 2 hrs. During the HS, colonic temperature of the animals was measured every 20 minutes, as previously described, to follow up their correct heating rate (40, 114). After HS, rats were allowed to recover at 24 ± 1°C for 0, 20, 40, or 60 min before collecting the tissues.

Table 3.1: Basal and hyperthermic plateau temperatures of rats undergoing heat acclimation (AC2d, AC),deacclimation of 1 mo (DeAC), and reacclimated for 2 days (ReAC)

Control AC2d AC DeAC ReAC Basal Tc 37.68±0.08 37.88±0.03 37.96±0.05* 37.64±0.09 38.04±0.02* Tc Plateau 40.0±0.10 39.92±0.02 40.78±0.03* 40.19±0.09 40.68±0.10*

Values are means ± SE and were derived from representative groups of 6–12 animals in each treatment group. Control (C) and deacclimated (DeAC) animals were maintained at 24 ± 1°C. Heat-acclimated (AC) and reacclimated (ReAC) rats were maintained at 34°C and 35% relative humidity. Tc, basal colonic temperature before heat stress; Tc plateau, average Tc at time points 80, 100, and 120 min exposure to heat stress at 41°C. * Significant difference vs. C -P < 0.05. For abbreviations see Fig. 3.1.

In our experiments, during HS the Tc rises to a hyperthermic plateau (Tc-plateau) (Table 3.1). The normal thermoregulatory response of the rat thus provides a reliable criterion for the acclimation status.

3.4 Tissue collection Immediately upon the termination of each basal treatment or basal treatment + HS, the animals were weighed (table 3.2) and euthanized by ketamine-xylazine (8.5 mg/100g body wt ketamine in 0.5% xylazine ip) anesthesia, followed by cervical dislocation. The isolated hearts were quickly weighed (table 3.2) and then retrogradely perfused for 2 min, with Krebs-Henseleit buffer containing (in mM) 120 NaCl, 4.7 KCl, 1.2 MgSO4, 1.2 KH2PO4, 1.25 CaCl2, 25 NaHCO3, and 11 glucose, at pH 7.4, and aerated with a mixture of 95% O2-5% CO2 at 37°C, with a perfusion pressure of 100 cmH2O. The left ventricle was then excised and stored at ~70°C until further processing.

Table 3.2: Rat body weight and heart tissue at the end of each experimental regime Control AC2d AC DeAC ReAC

Body weight 276±5.0 267±19.0 242±10.0* 316±5.5 320±7.4 (gr)

Heart tissue 1.14±0.02 1.25±0.05 1.05±0.01* 2.0±0.08 1.96±0.07 weight (gr)

The data are presented as means ±SE (n=10). The measurements were performed at the end of each experimental regime. The data presented in this table show that the AC rats body weight and their heart tissue mass were lower than in the control group (* Significant difference vs. C -P < 0.05). The results are consistent with our earlier studies on AC rats (40, 51). For abbreviations see Fig. 3.1.

3.5 Evaluations of protein levels In this work, the levels of Phospho-MSK1 (mitogen- and stress-activated protein kinase 1) and Phospho-CREB (cAMP response element-binding) proteins were measured. The protein levels were measured by Western immunoblotting similar to the protocols previously described in our publications (40, 114).

Cellular fractionation For cytoplasmic and nuclear fractions separation, 100 mg rat heart left ventricle tissue, collected from the different treatment groups animals, was homogenized in 1 ml of buffer A containing (in mM): 500 Sucrose, 10 HEPES (pH 7.9), 1.5 MgCL2, 1 EDTA, 1DTT, 1PMSF, 1.2 Na3 VO4, and 10% glycerol. The homogenates were then centrifuged at 5000 rpm for 10 minutes at 4⁰ C. The

supernatant (cytoplasmic fraction) was removed and frozen at -80⁰ C. The pellets were washed twice in detergent – free buffer A and dissolved in buffer B containing (in mM): 20 HEPES (pH

7.9), 420 NaCl, 0.2 EDTA, 1 DTT, 0.5 PMSF, 1.2 Na3 VO4, and 25% glycerol. After 30 minutes of incubation on ice, nuclear extracts were separated by centrifugation at 15000 rpm for 20 minutes at 4⁰ C and frozen at - 80⁰ C. Western immunoblotting Protein samples (50µg) were diluted in 3 x loading buffer containing 200 mM Tris-HCL, pH 6.8, 5% SDS, 30% glycerol, 0.04% bromphenol blue, 1% β-mercaptoethanol, vortexed and heated at 95⁰C for 2-3 minutes. Samples were separated on 9% polyacrylamide gels at 80 V. After the separation process, proteins were transferred onto a nitrocellulose membrane at 190-220 mA, 1 h at 4⁰ C. The membranes were then blocked for 2 h in a blocking solution containing 5% non- fat milk in 1x PBS/Tween-20 and probed overnight at 4⁰C with the respective primary antibody (see antibodies list). Membranes were then washed in PBS/Tween-20 and incubated at room temperature with the suitable secondary antibody (Jackson Lab., West Grove, PA) at 1:10000 dilutions. Finally, the membranes were washed, and antibody binding was detected by the enhanced chemiluminescence (ECL) reaction (EZ ECL, Beit Haemek Biological industries, Israel). The membranes were exposed to X-ray film (Kodak) and the density of the bands was analyzed by ―TINA‖ software (Raytest, Straubenhardt, Germany).

Antibodies list

1. Rabbit monoclonal to MSK1 (phospho Ser360) – (Abcam, England, ab81294) – diluted 1:500 2. Rabbit monoclonal to CREB (phospho Ser133) – (Cell Signaling Technology, Beverly, MA, #9198) – diluted 1:1000 3. Rabbit polyclonal to β-Actin (Santa Cruz Biotechnology, Santa Cruz, CA, SC-81178) – diluted 1:10000 was used as a housekeeping gene for normalization.

3.6 Evaluation of mRNA levels by quantitative real-time reverse-transcription polymerase chain reaction (Real-time PCR) mRNA expression was evaluated in several cases: (i) The kinetics of hsp70 and hsp90 transcription following HS and basal levels, in the

hearts of rats from all the experimental groups was studied. The aim of this section was to validate the hypothesis that transcriptional dynamics is a part of the acclimatory memory repertory.

(ii) hsf1 transcription levels in all the experimental groups were measured.

(iii) For array hybridization confirmatory analysis and transcriptional dynamics, atf3 and ucp3 mRNA detection was chosen.

All procedures in this section were performed according to the protocols published previously (114, 115).

3.6.1 Total RNA isolation Total RNA was extracted from rat hearts after the basal or basal + HS treatments. Left ventricle tissue (100 mg) was homogenized with a Polytron homogenizer in 1 mL of TRI-reagent (Molecular Research Center Inc., Cincinnati, OH). Homogenates were mixed with 0.5 volumes of chloroform, and incubated on ice for 10 minutes. After centrifugation at 9500 rpm for 15 min, the RNA phase was precipitated by 0.2 volumes of isopropanol, incubated for 10 minutes at room temperature and pelleted for 10 minutes in a centrifuge at 8000 rpm. The pellets were washed twice in 75% ethanol, and suspended in 50 µl of double-distilled water (DDW). RNA purity was assessed using a NanoDrop 2000 spectrophotometer (Thermo Scientific,

Wilmington, DE, USA), by determining the OD260/OD280 ratio – a ratio > 1.8 being indicative of relatively pure RNA. RNA was quantified by spectrophotometry at OD260 using the following formula: OD260 * dilution/1000 * final volume (where 1000 µl is the total volume of probe using for spectrophotometer reading). To assess the integrity of probes, each was separated by gel electrophoresis on 1.5% agarose (Promega, Madison, WI) gel in 1X TAE buffer and visualized by UV.

3.6.2 Double- stranded cDNA synthesis Total RNA (1 µg) was reverse transcribed in a 20 µl reaction mixture containing 0.5 µg of oligo(dT) as primer and 200 U of Moloney murine leukemia virus reverse transcriptase, according to the manufacturer‘s instructions (Fermentas, RevertAid™). For Real- Time PCR, 1 µl of the cDNA was used for a 20 µl volume of Real- Time PCR reaction mixture 30

3.6.3 Real-Time (RT) PCR reaction mRNA levels were measured using quantitative real-time reverse-transcription PCR (qRT-PCR; ABI Prism 7000 Sequence Detection System, Applied Biosystems). Reaction volumes of 20 µl contained 10 µl of SYBR Green Master Mix (Applied Biosystems), 500 nM each of the forward and reverse primer, and 5 µl of diluted cDNA. The appropriate cDNA dilution was determined using calibration curves established for each primer pair. The thermal profile for SYBR Green real-time RT-PCR was 95°C for 10 min, followed by 40 cycles at 95°C for 15 s and 60°C for 1 min. The primers (Table 3.3) for real-time RT-PCR were designed using Primer Express software (Applied Biosystems). Relative quantity (RQ) values were analyzed according to the ΔCt method which reflects the difference in threshold for each target gene relative to that of the housekeeping gene - β Actin. Actin is less prone to dynamic changes, and previous experiments in our laboratory showed that β-actin expression is unaltered by our AC protocols (40, 67).

Table 3.3: The gene primers that were used in mRNA analysis by Real-Time PCR

Gene Symbol Sense Antisense hsp70 5-caagaatgcgctcgagtccta-3 5-ctctttctcagccagcgtgtta-3 hsp90 5-gccaagtctgggaccaaagcgttcat-3 5-gtgtctgtcctcactgtgaatgatccc3 hsf1 5-tcggtgaccatgcccgacatgag-3 5-cagttcactgctccctgtgtccac-3 atf3 5-ccctcctagggaagatggag-3 5-tagcacaacaccaggctcag-3 ucp3 5-gcactgaccagatgggaaat-3' 5-aaggcgtcgactctggtttt-3' β-actin 5-tgtggcatccatgaaactac-3 5-atttgcggtgcacgatggag-3 (housekeeping gene used for normalization)

3.7 Chromatin immunoprecipitation (ChIP) analysis Chromatin immunoprecipitation (ChIP) is a widely used method to identify specific proteins associated with a region of the genome (for example, transcription factors), or in reverse, to identify regions of the genome associated with specific proteins. These proteins can be isoforms of histones modified at a particular amino acid or other chromatin associated proteins (120). When employed with antibodies that recognize histone modifications, ChIP can be used to ―measure‖ the amount of the modification. An example of this would include measurement of the

amount of histone acetylation and phosphorylation associated with a specific gene promoter region under various conditions that might alter expression of the gene. The ChIP assay was performed as described in our latest publishing Tetievsky et al.(2010) (115). Tissue collection was performed as described in section 3.4. Isolation and fractionation of chromatin was performed according to the protocol of Umlauf et al. (122), modified for cardiac tissue. Briefly, 250 mg of frozen heart muscle were minced on dry ice. Chromatin was solubilized and extracted using detergent lysis. The chromatin underwent further digestion using 10 units of micrococcal nuclease (MNase I, Worthington) added to 500 µl of soluble chromatin at 37°C for 20 min, to obtain fragments of 200–1,000 bp. The reaction was stopped by adding EDTA to a final concentration of 20 mM at 4°C. Fragment size was validated on an agarose gel. Chromatin fractions were pre-cleared by adding 60 µl of protein G agarose/salmon sperm DNA (Upstate) to 1 ml of chromatin immunoprecipitation dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris·HCl, pH 8.1, and 167 mM NaCl) and incubated for 30 min at 4° C on a rotating platform. After agarose removal by centrifugation, 1% of the precleared chromatin was saved and used as input DNA. Antibodies directed against the relevant histone modification or a specific protein (Table 3.4) were added to the chromatin samples and incubated for 12 h at 4° C on a rotating platform. For mock immunoprecipitation (background), normal rabbit IgG (4 µl/sample, Cell Signaling Technology) was used. Sixty microliters of protein G agarose/salmon sperm DNA were added to each immunoprecipitate and incubated at 4°C with rotation for ~2 h, followed by three washes with low-salt buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris·HCl, pH 8.1, and 150 mM NaCl). The preparations were then washed again for 15 min with rotation with the following buffers: high salt (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris·HCl, pH 8.1, and 500 mM NaCl), LiCl [0.25 M LiCl, 1% Igepal-CA630, 1% deoxycholic acid (sodium salt), 1 mM EDTA, and 10 mM Tris, pH 8.1], and two separate washes with Tris-EDTA buffer (10 mM Tris·HCl, 1 mM EDTA, pH 8.0).

Table 3.4: A list of antibodies used in ChIP experiments

Antibody name Serial number, company and the units used acetyl H3K9, acetyl H3K18, acetyl H3 antibodies kit C-9927, Cell Signaling Technology, H3K23 5 µl/sample phosphor(P)- histone /H3Ser10 H3 antibodies kit C-9927, Cell Signaling Technology, 5 µl/sample acetyl lysines 5, 8, 12, and 16 of 06–598, Upstate Biotechnology, 5 µg/sample histone H4 anti-HSF1 ab2923, Abcam, 10 µl/sample anti-TIP60 SC-5727, Santa Cruz, 5 µg/sample anti-phospho(P) MSK1 ab81294, Abcam, 10 µl/sample

Different units for the antibodies are presented, as the companies used different quantification methods.

The chromatin-DNA complexes were then eluted (0.1 M NaHCO3, 1% SDS) from agarose beads, and the DNA was purified using phenolchloroform and precipitation in ethanol. Input and immunoprecipitated DNA amplification was detected by real-time PCR (qPCR; ABI Prism 7300 Sequence Detection System, Applied Biosystems). The reaction was carried out in triplicate, using SYBR Green PCR Master Mix (Invitrogen). The primers (Table 3.5) were designed using Primer Express 2.0 software (Applied Biosystems) to have an annealing temperature (melting temperature) between 58 and 60°C. The thermal profile of the real-time PCR reaction was as follows: 50° C for 2 min, 95°C for 10 min, followed by 40 cycles at 95° C for 15 s (denaturation) and 60° C for 60 s (annealing and extension). The results were analyzed using the ΔCt (change in cycle threshold) method, which reflects the difference in threshold for the target gene relative to that of housekeeping gene in each sample. The amount of immunoprecipitated DNA in each sample is represented as signal relative to 1% of the total amount of input chromatin.

Table 3.5: Real-Time PCR primers used in ChIP experiments

Gene Symbol Sense Antisense hsp70 promoter region 5-ctccctggtctgattcccaaa-3 5- tgtccagaactctccagacgg-3

hsp70 3` UTR region 5-ttgcatgttctttgcgtttaatct-3 5-aggtgttcgcaggaaggaaa-3 hsp90 promoter region 5-ggcccaccctgctctgtac-3 5-cggagaccccaggaagaag-3

hsp90 3` UTR region 5-cccaaccctgctattctgt-3 5-ttctatcctgtccttgtgccttaa-3 dhfr 5-ccaccaaggcctcaaatcta-3 5-gccgtacgagagcctatgaa-3 β-actin 5-tcgctgacaggatgcagaag-3 5-ccaccaatccacacagagtacttg-3

(housekeeping gene used for normalization)

The primers were designed using Primer Express 2.0 software (Applied Biosystems). Transcription factor binding site prediction was performed with TFSearch version 1.3 (http://www.cbrc.jp/ research/db/TFSEARCH.html)T Genomic analyses were performed using the University of California Santa Cruz (UCSC) genome browser: http://genome.ucsc.edu. The results of the bioinformatic identification of HSF1 binding to hsp70 and hsp90 promoters are shown in Results chapter 4.2.1.

3.8 Rat whole genome mRNA microarray experiment We used the GeneChip Rat Gene 1.0 ST Array (Affymetrix, Santa Clara, CA), which interrogates 27,342 well–annotated genes with 722,254 distinct probes. The GeneChip gene 1.0 ST Array system offers a complete and accurate view of total transcription activity at each genomic locus and provides an effective solution for whole-transcript coverage. The design of the array was based on a November 2004 rat genome sequence with comprehensive coverage of RefSeq, putative complete CDS Gen Bank transcripts, all Ensembl transcript classes and syntenically mapped full-length mRNAs and RefSeq NMs from human and mouse. The Rat Gene 1.0 ST Array has 99.98 percent coverage of RefSeq NMs sequences present in RefSeq database.

3.8.1 Isolation, quantitation and qualification of RNA For microarray analysis we used an isolated total RNA, from rat hearts, as described in section 3.6.1. All samples were diluted to the concentration of ~100-150 µl and 2 RNA pools were prepared, each comprising 5 RNA samples from different animals in each experimental group. RNA quantitation, purification, and quality of all the pools were evaluated using a Bioanalyzer (2100, Agilent Technologies, USA). The assessment of RNA integrity is a critical ﬁrst step in obtaining meaningful gene expression data. The quality measurements included: 1) The RIN number-The RNA Integrity Number is a measure of the quality of the RNA. A RIN of 7 or greater is necessary for a sample to be used for microarray. 2) OD 260/280 and OD 260/230 ratios determine the purity of the RNA sample. Nucleic acid is detected at 260 nm, whereas protein, salt, and solvents are detected at 230 and 280 nm. High OD 260/280 and OD 260/230 ratios therefore indicate that we extracted RNA devoid of any of these contaminants. It is recommended that samples with ratios <1.8 for either ratio not be labeled for microarray. The concentration and the quality measurements data for each RNA pool are presented in Table 3.6.

Table 3.6: Quantitation and qualification of RNA from rat hearts tissues of all experimental groups. Sample Experimental Concentration 260/280 260/230 RIN number group ng/µL 1 Control 1 146 1.97 1.8 8.8 2 Control 2 155 1.97 2.04 9 3 AC2d 1 187 1.98 1.95 8.6 4 AC2d 2 163 1.95 1.95 8.6 5 AC 30d 1 111 2.02 1.7 8.7 6 AC 30d 2 121 2.04 1.8 8.7 7 DeAC 1 224 1.95 2.26 8.5 8 DeAC 2 193 1.98 2 8.8 9 ReAC 1 100 1.91 1.99 8.6 10 ReAC 2 100 1.96 2.00 8.6

2 RNA pools were prepared from each experimental group, each pool comprising 5 RNA samples from different animals. RIN- RNA Integrity Number. OD 260/280 and OD 260/230 ratios determine the purity of the RNA sample. 35

3.8.2 Microarray processing A 10 µg aliquot of previously isolated total RNA (as described in previous sections) from all the experimental heart tissue samples was used for the microarray experiment. The Affymetrix gene expression assay protocol for GeneChip Rat Gene 1.0 ST Array, was carried out in the microarray unit of the Weizmann institute. A brief summary of the microarray processing protocol is presented in Fig. 3.2.

Figure 3.2: Workflow for GeneChip Rat Gene 1.0 ST Array assay. The basic concept behind the use of GeneChip arrays for gene expression is: labeled cDNA or cRNA targets derived from the RNA of an experiments sample are hybridized to nucleic acid probes attached to the solid support. By monitoring the amount of label associated with each DNA location, it is possible to infer the abundance of each mRNA species represented. (The figure taken from http://www.vbi.vt.edu/index.php)

3.8.3 Data analysis and bioinformatics

An initial signal filtration and background subtraction were performed at the Bioinformatics unit of the Weizmann institute. For analysis, in brief, we standardized different recordings (radioactivity level at 5 time points) of the same hybridization experiment via a linear regression of the separate time points for each probe and averaged their regressed values to the recorded intensity at the middle time point. We then rescaled each measured expression value by the average expression of all probes in that hybridization, thereby normalizing all hybridizations to the same radioactive intensity, and transformed these values to make the control conditions a ―baseline‖ for estimating the relative changes in expression. A base-2 log was used, so that a relative expression of 1 or -1 would indicate a twofold change over the average control conditions. We received a list of detectable ―valid‖ expressed probes. From this list we selected 651 genes whose expression was up-regulated or down-regulated > 1.5 - fold in comparison with the control (C) group, in at least one of the experimental conditions. For further analysis, clustering, pathway finding, and specific gene searching, we used Bioinformatics tools provided in the internet:

1. Expander 6.0 - http://acgt.cs.tau.ac.il/expander/. We used this tool to perform the Gene ontology (GO) clustering of detected genes into GO categories. 2. GeneCoDis2.0 – http://genecodis.dacya.ucm.es/ for biological and molecular functional analysis of detected genes by GO categories and providing data about specific genes. 3. Onto Express - http://vortex.cs.wayne.edu/ontoexpress/ for biological functional pathway analysis. 4. GeneCards V3 database - www.genecards.org for specific genes data and relevant publications. 5. MicroRNA database Sanger institute - http://www.mirbase.org/ for miRNA data

3.9 Statistical analysis For statistical analyses, we used commercially available computer software (SigmaStat 2.03). The treatment time points (TTP: C, AC2d,AC, DeAC, ReAC) and the acute stressors (HS) were taken as the independent categorical variables, and individual animals or hearts were considered a random sample from the population. The following analyses were conducted. To

test the effects of TTP as the only factor on the basal values or on the superimposed HS of the dependent expressed variables (for variables see Fig. 3.1), we used a one-way ANOVA. For comparing continuous on line temperature measurements, we used the repeated-measures one-way ANOVA. For post hoc pairwise comparisons between the control (nonacclimated) and the various treatment time point groups, Dunnett‘s test was applied unless otherwise specified. To test whether TTP affect the response to superimposed HS, the two-way ANOVA was used. If significant interaction (with HS and TTP as the two factors) was indicated, a multiple- comparisons test to detect if origin of significance stems from HS vs. basal (Tukey) was conducted. Comparisons between HS and basal groups within matched TTP groups were also conducted using two-sample Student‘s t-test. Additional details are specified in the figure legends. Data are expressed as means ± SE. Values of P < 0.05 were considered statistically significant.

4. Results

4.1 Characterization of hsp70, hsp90 and HSF1 transcript levels

4.1.1 hsp70 and hsp90 transcription levels in normothermic conditions and during acute heat stress

HSP70 and HSP90 are important molecular chaperones. Their induction during heat stress enhances the ability to tolerate lethal temperature, ischemic and oxidative stress (39). We previously showed that AC significantly increases the cellular reserves of the inducible HSP70 and HSP90 in rat hearts, contributing to the establishment of the acclimated phenotype (67, 114). In this series of experiments, my goal was to reveal the transcription profile of hsp70 and hsp90 to support the hypothesis that transcriptional dynamics is a part of the acclimatory memory repertory. Because faster transcriptional activation in response to stress is a hallmark of successful acclimation, the kinetics of hsp70 and hsp90 transcription following acute heat stress (HS) at 41⁰ C was studied in all experimental groups. The measurements of mRNA levels were done in non-stressed (basal levels), 0, 20, 40, and 60 minutes post-HS, due to previous results showing that the peak of heat shock proteins transcription takes place within 1h post-HS recovery. The results (Fig. 4.1) show that both the treatment time point (TTP) and heat stress (HS) contributed to hsp70 and hsp90 dynamics with a statistically significant interaction between the two (two-way ANOVA P< 0.001). The control (C) transcriptional phenotype of hsp70 and hsp90 demonstrated peak levels at 60 min post-HS and 40 min post-HS, respectively. The hsp70 and hsp90 transcript kinetics of DeAC and ReAC groups, and the AC group, were similar, with peak hsp70 mRNA levels at 40 min post-HS and hsp90 mRNA peak immediately after the HS session (0 min post-HS). HS blunted the transcription of hsp70 in the AC2d group, as previously demonstrated (67), but not of hsp90 which showed at AC2d a similar to the control group transcriptional peak– 40 min post-HS. Basal (non-stressed) transcript levels of hsp70 differed significantly according to the TTP groups (one-way ANOVA P < 0.006) with AC2d, DeAC, and ReAC significantly upregulated and AC down-regulated compared with the nonacclimated (C) group (Fig. 4.1[a] inset). In contrast, the only group that showed significant upregulation of hsp90 basal transcript level, was the AC group (Fig. 4.1 [b] inset).

An additional interesting observation from this experiment is the difference between the total transcriptome rates of hsp70 vs. hsp90 in the DeAC and ReAC groups: whereas hsp70 showed very high levels of transcription at these groups, the hsp90 transcription was significantly lower at DeAC and ReAC as compared with the mRNA levels of hsp70 in the same groups.

Fig. 4.1: hsp70 (a) and hsp90 (b) mRNA 0, 20, 40, and 60 min post-HS at 41⁰ C in heart and hsp70 and hsp90 mRNA levels at each acclimation time point relative to basal C value (insets). Each bar represents mean ± SE, n = 4 rats per group. TTC showed significant effects on both hsp transcriptional dynamics post-HS (2-way ANOVA P value < 0.001). A statistically significant interaction was found between the 2 factors (P value ≤ 0.001). To explore interactions in mRNA transcription dynamics among the groups, 1-way ANOVA followed by multiple pairwise comparisons (Tukey test) within each individual acclimation time point was conducted. # Significance (P value < 0.05) of peak transcript level within the group. To test for significant effects of TTP within basal mRNA levels (inset), one-way ANOVA was applied. (P value < 0.001). * Significant difference vs. C (P value < 0.05, Dunnett‘s). C-control, AC2d- acclimated 2d, AC30d – acclimated 30d, DeAC – deacclimated 30d, ReAC – reacclimated 2d.

4.1.2 Transcriptional activity of hsf1 in the experimental groups

The binding of activated transcription factor HSF1 to the heat shock element (HSE) at the promoter of heat shock protein family genes initiates their transcription. Our previous findings showed that the HSF1 protein level (total as well as phosphorylated) was significantly upregulated in DeAC and ReAC groups similar to the acclimated phenotype (AC) (114). In the present study, hsf1 transcription was measured in all experimental groups. The results (Fig. 4.2) revealed a marked upregulation of hsf1 transcript in AC2d, AC, DeAC, and ReAC groups (matching the protein level profile). Although 30 days of acclimation increases the mRNA levels of hsf1 as compared with the control, the higher levels of hsf1 transcription were measured at 30 days of DeAC in normothermic conditions and the same levels were measured after 2-days of ReAC. Another striking outcome of this experiment is a high correlation between the hsf1 and hsp70 non-stressed transcriptional profiles (Fig. 4.1[a] inset and Fig. 4.2 respectively) in experimental groups, besides the AC group. Interestingly, hsp90 (Fig. 4.1[b] inset) mRNA profile did not match the hsf1 transcriptional behavior (Fig.4.2).

Fig. 4.2: hsf1 transcription levels in the experimental groups. n= 5-10 rats per group.*Significant difference from the C group (P value < 0.05 Dannett‘s test). All experimental groups expressed high transcriptional rate, as compared with C, with extremely high levels of transcription in the DeAC and ReAC groups. Abbreviations: C-control, AC2d-acclimated 2d, AC30d – acclimated 30d, DeAC – deacclimated 30d, ReAC – reacclimated 2d.

4.2 Chromatin modifications and transcriptional control of hsp70 and hsp90 genes

We have previously demonstrated in rats that the reinduction of the heat-acclimation physiological phenotype takes 2 days rather than the 30 required for acclimatory homeostasis (114). Nevertheless, the results we obtained from a series of experiments aiming to detect the molecular aspects of DeAC/ReAC emphasized that an altered gene-expression program underlies acclimatory responses, their decline and reinduction indicating molecular ―acclimatory memory‖. To address our hypothesis that the epigenetic machinery is involved in acclimatory memory formation and predisposes to faster ReAC, we performed chromatin immunoprecipitation (ChIP) assays to detect histone modifications at the promoters of hsp70 and hsp90 genes and the recruitment of a specific histone acetyl transferase (TIP60) and a specific histone kinase (MSK1) to the area of interest. To validate the open chromatin state of the promoter region of the hsp70 and hsp90 genes, HSF1 binding to the promoters of both genes was measured.

4.2.1 Bioinformatic characterization of HSF1 binding sites at hsp70 and hsp90 promoters hsp70 and hsp90 transcription is regulated by an activated HSF1 transcription factor that binds to conserved regulatory sequences known as heat shock response elements (HSE) - repeats of nGAAn element that forms stable binding sites for HSF1 (77). In order to further analyze the histone modifications, we examined its influence on the chromatin opening and the accessibility to the HSF1 binding, the identification of the HSE sites at the hsp70 and hsp90 promoter region. Fig. 4.3 shows the location of nGAAn elements clustered within 500 base pairs of transcription start site at the promoter regions of hsp70 (a) and hsp90 (b). The Real-time PCR - amplified fragment of the hsp70 gene contains 8 nGAAn elements (a), whereas hsp90 contains 3 HSE elements (b), identified by bioinformatic tools.

(A) gi|3413498|emb|X74271.1| Rattus norvegicus hsp70 kDa heat shock protein gene,

promoter region (1321 – 1924)

gcgaataaactatactgcaagatctcttctctttccctatttaaacctaaaatggagggagtggggggcagacacagacaggcgagc attccacaggcggcccccccacgctgtcacttccaggcaggacccaatcacagacttcttagccaagcgttatccctcccgttttgaga aactttctgcgtccgccatcctgtaggaagaatttgtacaccttaaactccctccctggtctgattcccaaatgtctctcaccgcccagcac tttcaggagctgacccttctcagcttcacatacagagaccgctaccttgcgtcgccatggcaacacttgtcacaaccggaacaagcact tcctaccaccccccgcctcaggaatccaatctgtccagcgaagcccagatccgtctggagagttctggacaagggcggtaccctcaa

CAT box catggattactcatgggaggcggagaagctctaacagacccgaaactgctggaagattcctggccccaaggcctcctcccgctcgct gattggccc atgggagggtgggcggggccggaggaggctccttaaaggcgcagggcggcgcgcaggacaccag

TATA box

(B) gi|91234897|gb|DQ022068.2| Rattus norvegicus hsp90 kDa heat shock protein

gene, promoter region (1-280)

tggcagccactcctttaaggcggagggatccaagggcggggctagggctgtgcttcgccttatatagggcggtcgggggcgtttgg

CAT box TATA box gagctctcttgagtcacccccgcgcagcctaggcttgccgtgcgagtcggacgtggtccgggcccaccctgctctgtactactactcgg ctttctcgtcaaggtaaggccgcgctctcctgtacttggcggctccgtgggtcttcttcctggggtctccgccaacatctcatctgaactaaa ggctccttgttg

Fig. 4.3: HSE sequence identification at the hsp70 (a) and hsp90 (b) promoter region. HSE sequence from human hsp70 promoter) (A) and HSE sequence from human hsp90 promoter (B) clustered within 500 base pairs of transcription start site of the genes. CAT box and TATA box are identified (in boldface). Arrows denote the Real-time PCR primers (forward and reverse). The hsp70 amplified region consist 8 HSE sequences, and hsp90 – 3 HSE sequences (sequences in frames).

4.2.2 Histones H3 and H4 acetylation at the HSE sites of the hsp70 and hsp90 promoters and TIP60 acetyltransferase recruitment

Because of the vast body of evidence that supports acetylated histones H3 and H4 involvement in transcriptional regulation and stress memory formation, we screened for H3 and H4 acetylation at the previously identified HSE sites of hsp70 and hsp90 promoters. Following the acetylation screening, histone H4 acetyltransferase recruitment to the hsp70 and hsp90 promoter area, was measured.

4.2.2.1 Histone H3 and histone H4 acetylation at the HSE hsp70 and hsp90 promoters

The results show significantly elevated levels of acetylated lysines 5, 8, 12, and 16 of histone H4-associated DNA at the HSE site of both hsp70 (Fig. 4.4a) and hsp90 promoter regions (Fig. 4.4b) as compared with the control. This elevation was seen not only in the long-term AC group, but also in the DeAC and ReAC groups, suggesting a preservation of the remodeled chromatin state of the studied genes established during acclimation throughout DeAC when the animals were no longer exposed to elevated environmental temperatures. In contrast to the acetylation of histone H4 at hsp70 promoter, the level of acetylation of lysines 9, 18, and 23 of histone H3, although presenting high levels, did not differ among the groups (Fig 4.5). Interestingly, no detectable acetylation of histone H3 was seen at HSE of hsp90 in all experimental groups (data not shown).

a Acetylation levels of histone H4 at the HSE of hsp70

Acetylation levels of histone H4 at the HSE of hsp90 b

Fig. 4.4: Acetylation levels of histone H4 (lysines 5, 8, 12, and 16) at the HSE of hsp70 (a) and hsp90 (b). The acetylation levels of histone H4 associated with the HSE of hsp70 and hsp90 were significantly elevated in AC, DeAC, and ReAC groups. The PCR product levels were normalized using β- actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ± SE of 4–7 experimental sets. *Significant differences from untreated C (one-way ANOVA followed by Dunnett‘s P< 0.05). For symbols, abbreviations, and experimental conditions see Fig. 3.1 in Methods. Published in Tetievsky and Horowitz 2010 (115). 46

Fig. 4.5: Acetylation of histone H3 (lysines 9, 18, and 23) at the HSE of hsp70 in AC, DeAC, and ReAC rats. No significant differences were found between the experimental groups. The PCR product levels were normalized using β -actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ≤SE of 4–7 experimental sets. *Significant differences from untreated C (one-way ANOVA followed by Dunnett‘s P < 0.05). For symbols, abbreviations, and experimental conditions, see Fig. 3.1 in Methods. Published in Tetievsky and Horowitz 2010 (115).

4.2.2.2 The level of histone acetyltransferase (HAT) TIP60 recruitment to the hsp70 and hsp90 promoters

Our initial analysis of the levels of histone acetylation in all experimental groups showed a significant elevation in the acetylation of histone H4 in AC, DeAC, and ReAC groups at the hsp70 and hsp90 promoters. Hence, we measured the levels of a specific histone H4 acetyltransferase recruitment - the TIP60 (60-kDa HAT-interactive protein) to the promoters of hsp70 and hsp90. TIP60 is a catalytic subunit of the NuA4 histone acetyltransferase complex that is involved in transcriptional activation of select genes principally by the acetylation of nucleosomal histones H4 at the promoter region (110). The results presented in Fig. 4.6a show a marked increase in TIP60 acetyltransferase recruitment to the hsp70 promoter in the AC, DeAC, and ReAC groups (Δ 95%-133%), whereas the recruitment of TIP60 to 3‘-UTR (3‘ un - translated region) was negligible (Fig. 4.6b).

Fig. 4.6: Recruitment of TIP60 (60-kDa HAT-interactive protein), a catalytic subunit of the NuA4 HAT complex to hsp70 promoter. a: TIP60 recruitment to the HSE of hsp70 was significantly elevated in AC, DeAC, and ReAC groups. b: no significant binding of TIP60 recruitment was detected at 3‘ untranslated region (UTR) of hsp70. The PCR product levels were normalized using β-actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ± SE of 4–7 experimental sets. * Significant differences from untreated C (one-way ANOVA followed by Dunnett‘s P <0.05). For symbols, abbreviations, and experimental conditions, see Fig. 3.1 in Methods. Published in Tetievsky and Horowitz 2010 (115).

Similarly, TIP60 acetyltransferase recruitment to the hsp90 promoter was markedly increased in the AC2d, AC, DeAC, and ReAC groups (Δ ~130% – 80%) ( Fig. 4.7a), whereas TIP60 binding at the 3‘-UTR showed negligible levels (Fig. 4.7b).

Fig. 4.7: Recruitment of TIP60 (60-kDa HAT-interactive protein), a catalytic subunit of the NuA4 HAT complex to hsp90 promoter. a: TIP60 recruitment to hsp90 promoter was significantly elevated at HSE of hsp90 in cardiac tissue in the AC2d, AC, DeAC, and ReAC groups. b: no significant TIP60 recruitment was detected at3‘-UTR of hsp90. The PCR product levels were normalized using β-actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ± SE of 4–7 experimental sets. *Significant differences from untreated C (one-way ANOVA followed by Dunnett‘s P < 0.05). For symbols, abbreviations, and experimental conditions, see Fig. 3.1 in Methods. Published in Tetievsky and Horowitz 2010 (115).

4.2.3 Histone H3Ser10 phosphorylation at the HSE sites of the hsp70 and hsp90 promoters, MSK1 recruitment and protein level

As described in the introduction chapter, several studies have demonstrated a correlation between an elevated histone H3Ser10 phosphorylation and transcriptionally activated heat shock loci in mammalian hsp70 promoter (81). Moreover, some evidences shows that HSP genes promoter phosphorylation may act as a recognition site for subsequent acetylation and chromatin opening in response to heat stress (116).

4.2.3.1 Histone H3Ser10 phosphorylation levels at hsp70 and hsp90 promoter

The results obtained from histone H3Ser10 phosphorylation ChIP measurements are presented in Fig. 4.8 (a and b). Histone H3Ser10 phosphorylation levels hsp70-promoter-associated were significantly elevated following each short-term heating session, i.e., AC2d and ReAC (Δ 35% and Δ 25%, respectively) as compared with the control (Fig. 4.8a). The phosphorylation of histone H3Ser10 at the HSE of hsp90 was significantly higher only at the onset of the acclimation session, i.e. in the AC2d (Fig. 4.8b).

Phosphorylation levels of histone Phosphorylation levels of histone H3Ser10 at the HSE of hsp70 H3Ser10 at the HSE of hsp90

a b

Fig. 4.8: Phosphorylation levels of histone H3Ser10 at the HSE of hsp70 (a) and hsp90 (b). Phosphorylation levels of histone H3Ser10 at the HSE of hsp70 were significantly elevated in the AC2d and ReAC groups, whereas the phosphorylation levels of histone H3Ser10 at the HSE of hsp90 were significantly elevated only in the AC2d group. The PCR product levels were normalized using β-actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ± SE of 4–7 experimental sets. *Significant differences from untreated C (one-way ANOVA followed by Dunnett‘s (P <0.05). For symbols, abbreviations, and experimental conditions, see Fig. 3.1 in Methods. Published in Tetievsky and Horowitz 2010 (115).

4.2.3.2 The level of MSK1 recruitment to the hsp70 and hsp90 promoters and activated MSK1 protein levels measurement

Due to the high levels of phosphorylated histone H3Ser10, it was interesting to determine the levels of recruitment of one of the major histone H3 kinases, the mitogen- and stress-activated protein kinase (MSK1) to the promoters of hsp70 and hsp90. MSK1 is a serine/threonine kinase that resides almost exclusively in the nucleus and once activated by various environmental stressors, phosphorylates Ser-10 on histone H3 (104). MSK1 is also known to be required for the phosphorylation of CREB, ATF1 and HMG14 in response to mitogens and stress (104). ChIP analysis showed a significant (almost 4 - fold) elevation in the recruitment of MSK1 to the hsp70 HSE area after 2 days of initial exposure to AC conditions (AC2d) and 2 days of reacclimation (ReAC), as compared with the control (Fig. 4.9a), whereas MSK1 recruitment the hsp90 was significantly slightly higher only at 2 days of acclimation (2d) (Fig. 4.9b). The recruitment of MSK1 to 3‘ UTR (un-translated region) of both genes was almost undetectable (data not shown). 51

a b

* *

Fig. 4.9: Recruitment of MSK1 (mitogen- and stress-activated protein kinase), to hsp70 and hsp90 promoters. a: MSK1 recruitment to hsp70 promoter was significantly elevated at 2d, and ReAC groups. b: MSK1 recruitment to hsp90 promoter was significantly elevated at 2dgroup.The PCR product levels were normalized using β-actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ± SE of 4 experimental sets. *Significant differences from untreated C (one- way ANOVA followed by Dunnett‘s P < 0.05). For symbols, abbreviations, and experimental conditions, see Fig. 3.1. Published in Tetievsky and Horowitz 2010 (115).

Notably, the levels of MSK1 recruitment to the promoters of hsp70 and hsp90 correlated to the levels of histone H3Ser10 phosphorylation at hsp70 and hsp90 genes (Fig. 4.8). This finding may suggest an involvement of MSK1 in direct phosphorylation of histone H3Ser10 at both genes when AC conditions are applied. To further validate the correlation between the histone H3Ser10 phosphorylation levels of hsp70 and hsp90 and an activation of MSK1, we used Western blot analysis to measure the levels of activated MSK1 protein. Several works reported that once MSK1 is phosphorylated on Ser360 by ERK1/2 and SAPK2/p38, MSK1 autophosphorylates on at least one of the five sites and becomes activated (104). Hence, we probed the whole cell lysates from rat hearts with an activated - phosphorylated on Ser360 MSK1 antibody. Our measurements showed that only 2 days of heat acclimation (AC2d) leads to a significant ~ 2 –fold elevation of phospho-MSK1, compared with control group (Fig. 4.10). This finding indicates a correlation between the active phospho-MSK1, the level of MSK1 recruitment to the hsp70 and hsp90 genes promoters and the rate of phosphorylation of histone H3Ser10.

Phospho-MSK1 levels in Rat heart

Fig. 4.10: Western blot analysis of whole cell extracts from left heart ventricle of rats in all the experimental groups probed with an antibody against MSK1 (Mitogen- and stress-activated protein kinase 1) phosphorylated on Ser360. Significantly high levels of phosphor-MSK1 were detected only in the AC2d group as compared to control. n=5 rats per group. *Significant difference from the C group (P < 0.05 Dunnett‘s). For symbols, abbreviations, and experimental conditions, see Fig. 3.1.

4.2.4 HSF1 binding to the HSE of the hsp70 and hsp90

To confirm binding of transcription factors to the euchromatin, to confirm the chromatin opening and initiation of transcriptional events, we measure the binding of HSF1 to the heat shock element (HSE) on the hsp70 and hsp90 genes (an essential step in heat shock proteins transcription). As a negative control, qPCR analyses of immunoprecipitated HSF1 was performed using primers that amplified a sequence within the 3‘-untranslated region (3‘-UTR) of hsp70 and hsp90 genes. Analysis of HSF1 binding to the promoter region of rat dihydrofolate reductase (DHFR), a gene that lacks HSE promoter elements and does not appear to bind HSF family, was performed as an additional negative control. The results show that HSF1 binding to the HSE of the hsp70 gene (Fig. 4.11a) was elevated in the AC2d, AC30, DeAC, and ReAC groups (Δ95-133%). As seen in Fig. 4.11b and 4.11c, HSF1 binding to the 3‘-UTR of hsp70 and to promoter of DHFR gene was negligible

b a

Fig. 4.11: HSF-1 transcription factor binding to the HSE of hsp70. Binding level is expressed as a percentage of input DNA. a: HSF-1 binding was elevated in the AC, DeAC, and ReAC groups, confirming transcriptional events. b: no significant binding of HSF-1-hsp70 was detected at 3‘-UTR region of hsp70. c: no significant binding of HSF-1 was detected at the dihydrofolate reductase (DHFR) promoter. The PCR product levels were normalized using β-actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ± SE of 4–7 experimental sets. *Significant differences from untreated C (one-way ANOVA followed by Dunnett‘s P < 0.05). For symbols, abbreviations, and experimental conditions, see Fig. 3.1 in Methods. Published in Tetievsky and Horowitz 2010 (115).

Binding of HSF-1 to the HSE of the hsp90 gene was elevated in the AC2d, AC30, and ReAC groups (Fig. 4.12a), confirming constitutive chromatin opening in these groups. Interestingly, no significant binding was measured in the DeAC group. Similarly to hsp70, binding to 3‘-UTR of the gene (Fig. 4.12b) was insignificant, thus confirming the specificity of HSF-1 to HSE at the hsp90 promoter. In addition, no detectable binding of HSF-1 was detected at the dihydrofolate reductase (DHFR) promoter (data not shown).

a HSF2 binding to the HSE at the hsp90 promoter

b HSF2 binding to the HSE at the hsp90 3’ - UTR

Fig. 4.12: HSF-1 transcription factor binding to the HSE of hsp90. Binding level is expressed as percentage of input DNA. a: HSF-1 binding to the HSE of hsp90 was elevated in the AC2d, AC, and ReAC phases, confirming transcriptional events, but not in DeAC. b: no significant binding of HSF-1 was detected at 3‘-UTR of hsp90. The PCR product levels were normalized using β-actin. Each bar (PCR abundance relative to 1% of the total amount of input chromatin) represents mean ± SE of 4–7 experimental sets. *Significant differences from untreated C (one-way ANOVA followed by Dunnett‘s P < 0.05). For symbols, abbreviations, and experimental conditions, see Fig. 3.1 in Methods. Published in Tetievsky and Horowitz 2010 (115).

4.2.5 Summery of the “Chromatin regulation and transcriptional control of hsp70 and hsp90 genes” chapter

 The results show markedly elevated levels of acetylated H4 associated DNA at the hsp70 and hsp90 promoter region in the long term acclimation for 30 days (AC), DeAC, and ReAC groups. Histone acetyltransferase recruitment of TIP60, the specific acetyltransferase of histone H4, was used to validate the acetylation of histone H4.  Elevated phosphorylation levels of histone H3 at the serine10 (H3Ser10) site associated with hsp70 were measured following AC2d and ReAC. Phosphorylation of histone H3Ser10 associated with hsp90 was only measured following AC2d, implying that these modifications were only found when the stress was greater than imposed by long term acclimation. The specific histone H3Ser10 kinase, the MSK1, recruitment to the promoters of both genes and the protein levels of an activated phospho-MSK1 were found to be in correlation with phosphorylation levels of the histone H3Ser10.  HSF-1-HSE binding was detected in the hsp70 gene throughout AC-DeAC-ReAC. The hsp90 gene lacked HSF-1 binding during DeAC, but resumed a high binding level upon ReAC.

4.3 Rat whole transcriptome screening and identification of pathways involved in the generation of heat-acclimation molecular “memory”

The purpose of this section in my research was twofold: 1) to discover upstream pathways and ―core-gene-clusters‖ of de-/re-acclimation; 2) pending on the results obtained, to test the hypothesis that upstream epigenetic regulatory events leading to formation of acclimatory memory. The Affymetrix Whole Genome Array Rat 1.0 ST system offers the most advanced gene expression profiling option for whole-transcript coverage. This array interrogates 27, 342 distinct probes. The design of the array is based on the November 2004 rat genome sequence (UCSC rn4, Baylor HGSC build 3.4) with comprehensive coverage of RefSeq, putative complete CDS GenBank transcripts and all Ensemble transcript classes. RNA samples from all the experimental groups were used for this experimental series. The array experiment and the initial results normalization were carried out by the special unit of Weizmann institute, as described in the ―methods‖ chapter. Following an initial signal filtration and background subtraction we received a list of detectable ―valid‖ expressed probes. From this list we have selected the probes that were up-regulated or down-regulated > 1.5 - fold in comparison to the control (C) group, in at least one of the experimental conditions.

4.3.1 Microarray analysis of gene clusters

An overall of 651 known genes changed their transcriptional rate > 1.5 –fold in at least one of the experimental conditions as compared to control (Table1, supplements). Assuming that groups of genes showing orchestrated expression profiles may relate to the same biological condition or process, we clustered the resulting data-set using ―EXPANDER 6.0‖ – a gene expression analysis and visualization software tool (100) http://acgt.cs.tau.ac.il/expander/. The Expander tool uses an algorithm in order to perform the gene clustering on the basis of mutual expression profiles. The algorithm is based on the calculation of two parameters: homogeneity and separation. - Overall average homogeneity: calculated as the average value of similarity between each element and the center of the cluster to which it has been assigned, weighted according to the size of the cluster. In our work we used the program default value for this parameter which is the estimated homogeneity of the true clustering.

- Overall average separation: calculated as the average similarity between mean patterns of different clusters, weighted according to their sizes. Mean Patterns of all clusters with error bars (±1 STD). This method iteratively assembles the genes such that the resulting set of clusters optimizes the likelihood that its genes come from the same distribution. The 651 genes were clearly grouped into 5 different clusters according to their transcriptional behavior as compared with the control:

1. Up-regulated after AC2d

2. Down-regulated after DeAC and ReAC

3. Down-regulated after ReAC

4. Up-regulated after DeAC and ReAC

5. Up-regulated after DeAC

Biological and molecular functions of genes in cluster #1 Genes categorized under cluster 1 encompass 18 genes (Fig. 4.13a), significantly inducible by short-term heat acclimation for 2 days (2d), then, stabilizing a lower expression level on reaching acclimation homeostasis (AC30) and keeping the similar to control level of transcription toward the DeAC and ReAC stages (Fig. 4.13b).

A B 1 2 3 4

Fig. 4.13: Cluster #1 of whole genome array (Affymetrix Whole Genome Array Rat 1.0 ST) analysis in rat heart. The transcription of the genes in the cluster was upregulated after AC2d, compared to control. A: tree view of the genes included in the cluster. Red, upregulation; green, downregulation. Each row represents 1 gene, a gene accession number (left), and gene symbol (right); columns represent treatment groups (1 – 2d/C, 2 – AC/C, 3 – DeAC/C, 4 – ReAC/C). B: a graph representing the cluster genes transcriptional behavior.

The majority of transcripts increased in expression upon AC2d, reflecting, a basic paradigm of short-term heat acclimation: switching on an altered expression of genes encoding networks of cytoprotective proteins. And indeed, according to their GO annotation for biological processes: the majority of the genes assigned to this cluster (for the list of the genes and their biological roles, see Table 2 in supplements chapter) are linked with functions such as, stress regulators and effectors (24%), mitochondria function (21%), metabolism (10%), and apoptosis (17%) (Fig.4.14).

Fig. 4.14: Major molecular functions of the genes comprised Cluster #1. The majority of the genes significantly (> 1.5- fold) upregulated by 2 days of AC are acting in the stress-related pathways as apoptosis, metabolic functions, calcium homeostasis and others. Each gene may precipitate in more than one functional pathway.

One of the interesting findings from this cluster is the upregulation of Prime1 (Primase1) gene at 2d group. The role of this gene is to synthesize small RNA primers for the Okazaki fragments and its activity is markedly engaged in double-stranded DNA repair and DNA synthesis. This finding may imply that the onset of AC leads to enhanced DNA synthesis possibly because of the double-stranded DNA damage. Another interesting finding is the upregulation of microRNA 297 (rno-mir-297), suggesting, for the first time, a role played by microRNA- in regulation of acclimatory responses at the onset of acclimation.

Due to the reason that the current work is dealing with the ―acclimation memory‖ phenomena and its main goal is to reveal the upstream molecular pathways of this process, we were particularly interested to see whether the clustering would uncover a groups of genes with similar behavior at DeAC and ReAC series. Indeed, four clusters revealed groups of genes that were simultaneously either up-regulated or down-regulated after DeAC and ReAC or after ReAC alone. Cluster number 2 includes a large group of 232 genes (Table 3, supplements), down-regulated after both de-acclimation for 30 days period (DeAC) and re-acclimation for 2 days (ReAC) (Fig. 15A). Cluster number 3 comprises 91 genes (Table 4, supplements) downregulated only at ReAC group (Fig. 4.15B).

A B 1 2 3 4 1 2 3 4

C D

Fig. 4.15: A: Cluster #2(A) and #3(B) of whole genome array analysis in rat heart. A: A tree view of cluster #2, the genes in this cluster were down-regulated at DeAC and ReAC groups. B: a tree view of cluster #3, the genes in this cluster were significantly down-regulated only at ReAC group. Red, upregulation; green, Indownregulation order to reveal. Each the rowbiological represents processes 1 gene ,and a gene molecular accession functions number of (leftsuch), anda large gene groups symbol of (right); columns represent treatment groups (1 – 2d/C, 2 – AC/C, 3 – DeAC/C, 4 – ReAC/C). C: cluster #2 genes genestranscriptional in these behavior clusters. D: cluster we have #3 genes used transcriptional the GeneCoDis2.0 behavior . http://genecodis.dacya.ucm.es/ database. GeneCodis is a grid-based tool that integrates different sources of biological

information to search for biological features (annotations) that frequently co-occur in a set of genes and rank them by statistical significance. Biological and molecular functions of genes in cluster #2 As shown in figure number 4.16, the main biological processes, in which the genes from cluster # 2 involved, are: DNA-dependent transcriptional regulation, ion transport, cell adhesion and more specific processes such as heart and brain development.

Fig.4.16: Biological processes in which genes from cluster #2 are involved. The analysis performed with GeneCoDis database. The total number of genes are representing all the genes in the annotation database.

When looking at the molecular functions of the genes comprising cluster # 2 (Fig. 4.17), several main functional groups are noticeable: Protein binding activity (68 genes), ion binding (35 genes), hydrolase activity (22 genes), signal transduction activity (20 genes), transmembrane transporter activity (18 genes) and substrate transporter activity (18 genes). Interestingly, additional large groups of genes with transcriptional down-regulation at DeAC/ReAC from cluster #2 were related to nucleotide binding (17 genes) or nucleotide acid binding (17 genes). Among these, several genes encoding histones H1 and H3, epigenetic regulators or genes involved in chromatin remodeling and a group of microRNAs were found (Table 4.1).

Fig. 4.17: Molecular functions of genes from cluster #2. The analysis performed with GeneCoDis database. The total number of genes is representing all the genes in the annotation database.

Table 4.1: A representative group of genes, involved in chromatin and transcription regulation, down-regulated at DeAC and ReAC (Cluster #2). Gene symbol Main Function Hist1h2bl A member of the Histone 1 family Hist2h2bb A member of the Histone H2b family Hist2h3c2 A member of the Histone H3 family H2afz A member of the histone H2a family

Nap115 (nucleosome assembly p.) Nucleosome assembly protein (NAP) family, Interact and regulates chromatin. Cited1 Transcriptional co-activator. Associates with (Cbp/p300-interacting transactivator 1) chromatin to the estrogen-responsive TGF-alpha promoter region.

Aebp1 (AE binding protein) DNA-binding transcriptional repressor. Stimulate complex which methylates 'Lys-9' and 'Lys-27' residues of histone H3.

Smarca1 SWI/SNF related complex. Facilitate the (SWI/SNF related, matrix associated, perturbation of chromatin structure in an ATP- actin dependent regulator of dependent manner. chromatin) CREB3L2 Transcriptional activator. Regulates the (cAMP responsive element transcription of unfolded protein response target binding protein) genes. Connected to histone H3 phosphorylation.

rno-mir-212 Micro-RNA 212 rno-mir-224 Micro-RNA 224 rno-mir-328 Micro-RNA 328 rno-mir-490 Micro-RNA 490 For a complete list of genes, their symbols and accession numbers see table 3 in Supplements chapter. 63

Biological and molecular functions of genes in cluster #3 As shown in Fig. 4.18, a marked down-regulation of genes from the DNA elongation/repair pathway occurred at the ReAC group (24%). Other large functional groups in cluster #2 are: translational regulation (15%), metabolism (13%), transport and cytoskeleton (11%), membrane proteins and receptor (11%), chromatin remodeling (9%), apoptosis (2%), and MicroRNAs (2%).

Fig. 4.18: Major molecular functions of the genes found in Cluster #3. The genes in cluster #3 were down-regulated by 2 days of re-acclimation (ReAC). The majority of the genes participating in cell-cycle and DNA elongation and repair pathways, translational regulation. 9% related to chromatin remodeling. Each gene may precipitate in more than one functional pathway.

Closer examination revealed a group of serine/threonine kinases that are active in (or associated with) chromatin remodeling-related processes (DNA repair/elongation and transcriptional regulation) and that belong to the MAPK family, among these kinases: PBK (PDZ binding kinase), encodes a serine/threonine kinase that phosphorylates MAPK p38 and activates its pathway (the p38 MAPK pathway is also responsible for several histone modifications, active in DNA damage repair; PLK1 (Polo-like kinase 1) belongs to a family of four serine/threonine protein kinases that are critical regulators of the DNA damage response. Additional genes, founded in cluster #3 related to chromatin remodeling were as follows: AURKB (Aurora kinase B) is a serine/threonine kinase that plays a critical role in histone H3 phosphorylation and regulation of transcription; ANKRD41 (ankyrin repeat domain-containing protein41), recruits HDACs (Histone de-acetylases) to the coactivators/nuclear receptor 64

complex to inhibit the ligand-dependent transactivation of several genes; a group of 7 genes encoding parts of histone H1 (Hist1h1a, Hist1h1b), histone H2a (H2a. Hist1h2ai, Hist1h2ak, Hist2h2ab), and histone H2b ( Hist1h2bc, Hist1h2bn); microRNAs (rno-mir-208, rno-mir-375).

Two additional gene clusters revealed in our study are of a special interest due because the genes in these clusters are transcriptionally up-regulated in both DeAC and ReAC groups. Cluster #4 is a largest cluster, includes 243 genes (Table 5, supplements), up-regulated after both DeAC for 30 days and ReAC for 2 days (Fig. 4.19A). Cluster #5 comprises 53 genes (Table 6, supplements) up-regulated primarily at the DeAC group (Fig. 4.19B).

Biological and molecular functions of genes in cluster #4 As shown in figure number 4.20, the main biological functions revealed by the GeneCoDis2.0 database of the genes in cluster #4 (genes that are up-regulated in DeAC and ReAC) are as follows: primary, cellular, and macromolecule metabolic processes (~80%), regulation of biological and cellular processes (~53%), response to chemical stimulus (~20%), the nitrogen compound metabolic processes (~19%), and various biosynthetic processes (~18%). When analyzing the molecular functions of the genes in this cluster (Fig. 4.21), the major common response includes genes with regulatory and binding ability: primarily protein and nucleotide binding (90 genes), transferase (18 genes) and oxidoreductase (17 genes) activity, carbohydrate (11 genes) and cofactor (10 genes) binding, enzyme inhibitory activity (6 genes), and pattern binding (6 genes).

A 1 2 3 4 B 1 2 3 4

c D

Fig. 4.19: Cluster #4(A) and #5(B) of whole genome array analysis in rat heart. A: A tree view of cluster #4, the genes in this cluster were significantly up-regulated in the DeAC and ReAC groups. B: a tree view of cluster #5, the genes in this cluster were up-regulated only in the ReAC group. Red, upregulation; green, downregulation. Each row represents 1 gene, a gene accession number (left), and gene symbol (right); columns represent treatment groups (1 – 2d/C, 2 – AC/C, 3 – DeAC/C, 4 – ReAC/C). C: cluster #4 genes transcriptional behavior. D: cluster #5 genes transcriptional behavior.

Fig. 4.20: Biological processes in which genes from cluster #4 are involved. The analysis was performed with GeneCoDis database. The total number of genes (gray sector) is representing all the genes in the annotation database.

Fig.4.21: Molecular functions of genes from cluster #4. The analysis was performed with GeneCoDis database.

Close examination of the genes comprising this cluster revealed genes involved in cytoprotective pathways, such as HSPa41 – a chaperone from the HSPs family and Fkbp5

(FK506 binding protein 5), a protein involved in folding and trafficking processes that interacts with HSP90. Further individual functional examination of the genes in this cluster revealed a group of 6 genes belonging to the MAPK signaling pathway. The names and the functional roles of these genes are presented in Table 4.2.

Table 4.2: A list of genes from Cluster #4 (up-regulated at DeAC and ReAC) which are a part of MAPK signaling pathway. For a complete list of genes, their symbols and accession numbers see table 5 in Supplements chapter.

Gene symbol Main Function

MAPK12 Responds to activation by environmental stress (mitogen-activated protein kinase 12 and pro-inflammatory cytokines by phosphorylating gene) downstream targets. ATF3 Transcriptional regulation (activating transcription factor 3 Gene) Cnksr1 Connector enhancer of kinase suppressor of Ras 1 gene Nfkbia Inhibits the activity of dimeric NF-kappa-B

Dusp14 and Dusp15 dephosphorylation and diphosphorylation synthetic (Dual specificity phosphatases 14/15) MAPK peptides

An additional interesting group of genes associated with transcriptional regulation by means of chromatin remodeling or genes connected to chromatin modifications and founded in this cluster presented in Table 4.3.

Table 4.3: A list of genes from Cluster #4 (up-regulated at DeAC and ReAC) associated with chromatin remodeling and transcriptional regulation.

Gene symbol Main Function ANKRD 6 May recruit HDACs to the nuclear receptor complex (ankyrin repeat domain 6) and to inhibit ligand-dependent transactivation Cited2 Inhibits transactivation of HIF1A (Hypoxia inducible (Cbp/p300-interacting transactivator 2) factor 1 A)-induced genes by competing with binding of HIF1a to a chromatin of target genes Csdc2 RNA-binding factor which binds specifically to the (cold shock domain containing C2) very 3'-UTR ends of both histone H1 and H3 encompassing the polyadenylation signal important for the negative regulation of histone synthesis Mettl7a methyltransferase methyltransferase 7A Mir29b-2 Micro-RNA29b-2

For a complete list of genes, their symbols and accession numbers see table 5 in Supplements chapter.

Interestingly, another members of ANCRD and Cited family – ANCRD41 and Cited 1, was found to be down-regulated (Cluster #2 and Cluster #3, respectively) at DeAC and ReAC, implying different behavior of these related genes in response to De-and Re-acclimation.

Biological and molecular functions of genes in cluster #5

GeneCoDis2.0 functional analysis of the 53 genes assigned to the cluster #5 (Fig. 4.22), revealed that similar to the genes in cluster #4, the majority of the genes were assigned to the functional group of metabolic processes (~40%). To a lesser extent, the genes in this cluster are involved in the following biological pathways: multicellular (~24%) and anatomical structure (~20%) development, cell cycle (~13%), regulation of cell proliferation (~13%), response to endogenous stimulus (~13%), and others.

Fig. 4.22: Biological processes in which genes from cluster #5 are involved. The analysis was performed with GeneCoDis database.

When looking at the molecular functions of the genes in cluster #5 (Fig. 4.23) , a high similarity is noticed between the main molecular functions of the genes in cluster #4 and cluster #5: protein binding (15 genes), nucleotide binding (11 genes), hydrolase activity (9 genes), transferase activity (8 genes). Interestingly, this cluster contains two genes that play a role in pre-mRNA processing: Snrpn (small nuclear ribonucleoprotein polypeptide N) and related to it, Snurf (SNRPN upstream reading frame). Additionally, this cluster contains gene encoding the Micro-RNA-22 (rno-mir-22).

Fig.4.23: Molecular functions of genes from cluster #5. The analysis was performed with GeneCoDis database.

4.3.2 Identification of transcriptional regulators of the significantly expressed genes

The clustering analysis revealed that the major group among the 651 genes in the array comprises transcription factors and their targets. This finding led us to ask who are the main transcriptional regulators of the significantly expressed genes in the array. Manual examination of all 651 genes with the GeneCards V3 database (www.genecards.org) (Fig. 4.24) revealed a large group of 182 genes (~30% from the total) which are transcriptionally regulated by one of the several important TFs, functionally related to the stress response cascade:  CREB (cAMP response element-binding) - ~38%  C-Fos or/and C-jun - ~35%  ATF (activating transcription factors) - ~12%  HSF1/2 - Heat shock factor - ~12%  HIF1α/β – Hypoxia inducible factor - ~0.5%

Fig. 4.24: Transcription factors, related to the stress response, which regulate 30% (182 genes) of the genes significantly expressed in at least one experimental group (>1.5 fold) in the array. The percentages were calculated taking 182 as 100%.

CREB is an important cellular transcription factor that binds to certain DNA sequences called cAMP response elements (CRE), thereby increasing or decreasing the transcription of the downstream genes. Interestingly, CREB is a transcriptional regulator of c-fos and c-jun, which 71

belong to an immediate early genes family of transcription factors and closely related in structure and function to ATF. Moreover CREB, ATF, c-fos, and c-jun are a part of the MAPK p38 pathway situated upstream to the histone H3 acetylation and the histone phosphorylation sites by MSK1. The CREB and ATF genes are also significantly expressed in the array (clusters #2 and #4, respectively).

4.3.2.1 Phospho-CREB protein level Our findings show that (i) CREB showed a significant down-regulation of transcription in DeAC and ReAC groups (cluster #2 in array); (ii) CREB is a transcriptional regulator of a large group of genes, significantly expressed in our array results; (iii) CREB is an important part of the p38 MAPK pathway, responsible also for histone acetylation and phosphorylation. Hence we used Western blot analysis to measure the protein level of an activated Phospho-CREB (Ser133) in the lysates from rat hearts in all experimental groups. Figure 4.25 represents the Phospho-CREB protein levels in all experimental groups. The figure shows a significant increase in protein levels at 2d and 30d acclimation groups (Pvalue<0.05) and a return to the Control levels in both DeAC and ReAC groups. Notably, the protein levels matching the mRNA profile seen in array results, which showed a marked down-regulation of CREB transcription at DeAC and ReAC groups.

* *

Fig. 4.25: Western blot analysis of whole cell extracts from left heart ventricle of rats in all experimental groups probed with an antibody against CREB phosphorylated on Ser133. Significantly high levels of phosphor-CREB were detected only in the AC2d and AC groups as compared with the control. n=5 rats per group. *Significant difference from the C group (P < 0.05 Dunnett‘s). For symbols, abbreviations, and experimental conditions, see Fig. 3.1 in Methods.

4.3.3 Evidence for MicroRNA involvement in AC- associated gene expression changes and memory

Analysis of the whole genome rat array showed that each experimental treatment leads to a rapid change in the profile of gene expression. The finding of 9 genes encoding miRNAs significantly expressed in at least one of the experimental groups implies an important role of miRNA in these responses. The summery of the miRNAs genes found in the array results are presented in Table 4.4.

Table 4.4: The list of detected MicroRNAs, significantly changing in at least one of the experimental groups in the rat transcriptome.

MicroRNA symbol Cluster number Transcriptional behavior (>1.5 fold) rno-mir-297 1 Upregulation at 2d rno-mir-212 2 Downregulation at DeAC and ReAC rno-mir-224 2 Downregulation at DeAC and ReAC rno-mir-328 2 Downregulation at DeAC and ReAC rno-mir-490 2 Downregulation at DeAC and ReAC rno-mir-208 3 Downregulation at RaAC rno-mir-375 3 Downregulation at ReAC rno-mir29b-2 4 Upregulation at DeAC and ReAC rno-mir-22 5 Upregulation at DeAC

The results show that the major MicroRNA group belongs to cluster #2: Downregulated at DeAC and ReAC.

miRNAs play roles in stress memory formation (71, 72) as it often acts to attenuate or silence gene translation in a transient mood of operation. There is great interest in the miRNAs as ―fine tuners‖ of gene expression and protein translation in an ever-growing range of cellular and physiological functions (72, 89). This paradigm fits well with the possibility that miRNAs are involved in the dynamic epigenetic regulation of gene transcription in changing environments. To determine whether a relationship exists between the miRNAs that were significantly changed by one of the experimental treatments and the genes with expression ―visible‖ in the array, we used in silico analysis with the GeneCoDis2.0 database to search for the target genes of 9 miRNAs found in the array, among the significantly expressed genes also found in the array. The search revealed that only 6 out of 9 miRNAs have target genes (an amount of 48 genes) among the 651 used for our analysis.

The next step in our analysis was the identification of the transcriptional behavior of these target genes per each miRNA. The results presented in Table 4.5 show the following data:  Rno-mir-29b-2, which belongs to cluster #4 (Upregulated at DeAC and ReAC) has the largest group of target genes (15 genes). Among the targets, 1 gene belongs to cluster #5 74

(Upregulated at DeAC), 2 genes belong to cluster #3 (Downregulated at ReAC) and the major group of 12 targets from cluster # 2 (Downregulated at DeAC and ReAC). This means that 93% of the rno-mir-29b-2 gene targets are transcriptionally downregulated whereas rno-mir-29b itself is upregulated.  Rno-mir-224, from cluster # 2 (Downregulated at DeAC and ReAC) regulates 10 genes from array. 3 genes - from cluster #3 (Downregulated at ReAC) and 7 (70% ) belong to the same cluster as the miRNA-224 itself (Downregulated at DeAC and ReAC).  Rno-mir-212, belongs also to cluster #2. Seven targets of this miRNA are all from cluster #4 (Upregulated at DeAC and ReAC). Here we can see that downregulation of miRNA leads to upregulation of 100% of the target genes.  Rno-mir-328, from cluster #2 regulated nine possible targets. Two of the genes are from cluster #1 (Upregulation at 2d) and seven genes from the cluster similar to the one miRNA-328 belongs to, cluster #2 (Downregulation at DeAC and ReAC).  Rno-mir-208, from cluster #3 (Downregulated at ReAC), is a target regulator of four genes from the array. One of them belongs to cluster #1 (Upregulated at 2d) and three to cluster #5 (Upregulated at DeAC).  Rno-mir-375, belongs to cluster #3 and can regulate three genes from the same cluster - #3.

Table 4.5: MiRNAs and their target genes from the whole rat genome array.

miRNA symbol Target Target Target Target genes Target genes from genes from genes from genes from cluster#1 cluster #2 from cluster#4 cluster#3 cluster#5 Eln Aurkb Ckb Gap43 Nrep Stmn2 Tubb2b Tubb6 Hist1h2b1 rno-mir29b-2 Upk1b Agpat4 cluster #4 Col3a1 Alcam Slc30a3 Gpc3 Nov Nid1 Id2 Cenb1 Kcna2 Casc5 rno-mir-224 Sqle Upk1b cluster #2 Fn1 Gos2

Ms4a6b H2afz Hmox1 rno-mir-212 Rps27 Pfkfb2 cluster#2 Cyp2e1 Actr3b

Zfhx2 Aqp5 Cish Tubb3 rno-mir-328 Ica1 Agpat4 cluster #2 Tubb6 Slc36a2 Dhrs9 Aldh3a1 Bdh1 rno-mir-208 Bloc1s2 Acta1 cluster #3 rno-mir-375 Cuzd1 cluster #3 Psat1 Adra1d The left column contains the name of miRNA and the cluster to which it‘s belongs. Other columns represent the five clusters with the list of genes founded to be a targets of the miRNA by the GeneCoDis2.0 database. For a full names of the genes look at the gene lists of each cluster in the supplements chapter.

The main finding in this part of the investigation is that out of 48 target genes, 21 genes (~43%) show a transcriptional profile opposite to its miRNA transcriptional profile, and 24 target genes (50 %) show the same transcriptional profile as its regulatory miRNA. Three genes do not show any clear significant correlation between the miRNA and the gene transcriptional profile. Among the detected miRNAs are some (eg, miRNA -29b-2 and miRNA 212) that are clearly (almost 100%) transcribed in the opposite way to their targets. In contrast, miRNA -224, miRNA-328, and miRNA-375 predominantly have the same transcriptional profile as their targets. When looking specifically at the gene groups, the striking finding is that 75% of the target genes of the miRNAs whose expression level had been influenced by one of the treatment groups are connected to the p38 MAPK pathway, while regulated by one of the TFs from this pathway: CREB, ATF, or c-jun/c-fos.

4.3.4 Real Time RT-PCR verification of specific genes significantly expressed in array

To confirm the results obtained from the arrays, we subjected the RNA samples used for gene chip array analysis to a separate Real Time RT-PCR detection system to measure the mRNA of a specific genes significantly expressed in array analysis. 1) ATF3 – a transcription factor that belongs to the MAPK p38 family, which was upregulated in DeAC and ReAC groups (Cluster number 4) due to the array results, showed similar transcriptional profile at Real Time – PCR analysis (Fig. 4.26).

Figure 4.26: ATF3 transcription levels in the experimental groups. n= 4 rats per group.*Significant difference from the C group (P value < 0.05 Dannett‘s test). Similar to the transcriptional profile of ATF3 in the array results (inset), DeAC and ReAC group expressed high transcriptional rate, as compared with control group. Abbreviations: 2d-acclimated for 2d, AC – acclimated for 30d, DeAC – deacclimated for 30d, ReAC – re acclimated for 2d.

2) UCP3 (Uncoupling protein 3) – a mitochondrial gene, that participates in thermogenesis and energy balance, belongs to cluster #4 – upregulated in the DeAC group in array results. As presented in Fig. 4.27, UCP3 mRNA transcription, detected by means of Real-Time PCR matched to the array results, shows in both cases a marked upregulation only in the DeAC group as compared with the control.

Figure 4.27: UCP3 (Uncoupling protein 3) transcription levels in the experimental groups. n= 4 rats per group.*Significant difference from the C group (P value < 0.05 Dannett‘s test). Similar to the transcriptional profile of UCP 3 in the array results (inset), only DeAC group expressed high transcriptional rate, as compared with control group. Abbreviations: 2d-acclimated for 2d, AC – acclimated for 30d, DeAC – deacclimated for 30d, ReAC – re acclimated for 2d.

4.3.5 Summary of the “Rat whole transcriptome screening and identification of pathways involved in the generation of the heat-acclimation molecular memory” chapter

 Our results revealed an overall of 651 known genes that changed their transcriptional rate > 1.5 –fold in at least one of the experimental conditions as compared with the control group.  The 651 genes were clearly grouped into five different clusters, by bioinformatics tool, according to their transcriptional behavior.

 The majority of transcripts increased in expression upon 2 days of AC reflecting, a basic paradigm of short-term heat acclimation (STHA): switching on an altered expression of genes encoding networks of cytoprotective proteins.  Among the large clusters comprising genes significantly either up-or down-regulated at DeAC and ReAC groups, we have identified the transcripts of chromatin-related genes histones H1, H2b, H2a and H3, large group of MicroRNA (miRNA) transcripts, groups of chromatin remodeling genes, and regulators of histone modifications.  Among the chromatin-related genes we identified a group of genes that belong to the p38 MAPK signaling pathway, among these genes, CREB transcript showed a significant down-regulation in the DeAC an ReAC groups. The phosphor-CREB protein levels analysis revealed upregulated levels in the AC group but downregulation at DeAC and ReAC. ~38% of the genes significantly expressed in the array were found to be regulated by CREB.

5. Discussion

Heat-acclimation (AC) is a reversible ―within life-time‖ phenotypic adaptation to long-term shifts in environmental conditions. The underlying mechanisms include transient short- and long-term genomic responses that reprogram gene expression and enlarge cytoprotective protein reserves. Successful AC improves endurance and resistance to temperature extremes and also can reinforce or interfere with the ability to combat novel stressors. Novel and powerful outcomes of our previous studies and results (114) are the awareness of acclimation plasticity leading to the induction of ―acclimatory memory‖. These findings inspired us to further investigate to what extent and how within-life thermal history (environmental) is involved in shaping the phenotypic features associated with AC memory. Specifically, we focused here on AC mediated cytoprotective memory.

Our present findings show that epigenetic features play a role in ―within-life‖ adaptation of the species by allowing reversible phenotype variability in adulthood and expand our understanding of the impact of environmental extremes on epigenetic markers in general. The attempt to explore the upstream epigenetic regulatory events and molecular pathways involved, leading to acclimatory memory, if successful, will provide further insight into the broad role of chromatin as a physiological or pathophysiological (e.g. toxicant stress, drug addiction) states and may advance the notion of epigenomic therapy.

5.1 Characterization of hsp70, hsp90 and HSF1 transcript levels and the link to memory formation.

The available evidence substantiates that a reprogramming of gene expression is an essential event in the pathway to AC (AC) (39, 40, 42, 50, 67).  A cytoprotective network comprising HSP70, HSP90 and HSF1 plays a pivotal role in the generation of the mammalian acclimated phenotype (39, 67).  Reinduction of the physiological heat-acclimated phenotype needs only 2 days rather than the 30 days required for achieving initial acclimatory homeostasis, thus confirming that AC has a memory (114).

These findings led us to hypothesize that acclimation may induce persistent, selective transcriptional changes, predisposing individuals who have undergone an initial AC session to reacclimate (ReAC) to elevated temperatures more rapidly. To test our hypothesis we focused on measuring the following: 1. Transcriptional dynamics profile of hsp70 and hsp90 post exposure to acute heat stress (HS) 2. Basal mRNA levels of hsp70 and hsp90 3. Transcriptional activity of the HSF1, the HSPs family transcriptional activator.

hsp70 and hsp90 transcription dynamics is part of the memory repertory

Because faster transcriptional activation in response to stress is a hallmark of successful acclimation (39, 67), the kinetics of hsp70 and hsp90 transcription following acute heat stress (HS) at 41⁰ C was studied in all the experimental groups. Our results here show that both hsp70 and hsp90, similar to the AC group, exhibit faster (than Control) transcriptional dynamics in the DeAC and ReAC groups (Fig. 4.1 a and b in Results chapter), with an mRNA peak of hsp70 at 40 min post-HS and an mRNA peak of hsp90 immediately after the HS treatment (0 min post-HS). Our current data fit with earlier findings (67), showing that the AC phenotype is characterized by faster transcriptional dynamics of the heat shock proteins (HSPs) in response to HS. Nonetheless, this study is the first to provide evidence that the altered ―acclimated‖ mode of the HS response is retained after 30 days at normothermic conditions (DeAC), and reappears promptly upon reacclimation (ReAC) conditions. This finding supports our hypothesis that faster transcriptional dynamics is part of the memory repertory.

Inverse relationships between HSP70 and HSP90 in DeAC causes the loss of cardioprotection in DeAC group but do not interfere with predisposition to fast restoration of cardioprotection upon ReAC

In this study we also measured the non-stressed-basal mRNA levels of hsp70 and hsp90 (Figure 4.1 a and b insets) in all experimental groups. Table 5.1 presents the hsp70 and hsp90 basal mRNA analysis results and protein levels with the cardioprotection phenotype analyzed in

our previous study (114). The summary of all results promotes our understanding of the interplay between transcription-translation and the physiological phenotype.

Table 5.1: hsp70 and hsp90 behavior at transcriptional and translational levels, with correlation to cardioprotection.

HSP70 HSP90 AC2d AC30d DeAC ReAC AC2d AC30d DeAC ReAC mRNA ↑ ↓ ↑ ↑ ↔ ↑ ↓ ↑ Protein* ↔ ↑ ↑ ↑ ↔ ↑ ↓ ↑ Cardio- ↔ ↑ ↓ ↑ ↔ ↑ ↓ ↑ Protection* The arrows indicate up/down-regulation or unchanged levels compared with the control group. * The protein levels and cardioprotection data were taken from my previous publication (114). The mRNA levels presented here are basal levels (the data taken from insets in graphs 4.1a and 4.1b in the Results chapter).

In the current study, low levels of hsp70 mRNA together with augmented HSP70 protein reserves were detected at AC group. Constitutive overexpression of inducible HSP70 in the AC phenotype was also detected by Maloyan et al. (67) who proposed that the high endogenous level of this cytoprotective protein confers both thermotolerance and cardio-protection from acute stress (e.g. ischemic stress and heat stress) without the need for de novo protein synthesis. Interestingly, in the current study, the DeAC phenotype is characterized by a mismatch between the greater HSP70 mRNA and protein reserves and the loss of cardioprotection (Table 5.1). This finding agrees with the results of Xi et al. (127), who demonstrated that heat shock and HSP70 induction fail to protect the heart against ischemic injury. One explanation for the dichotomy between the phenotypic-physiological and the molecular response at DeAC group could stem from the decrease in HSP90 levels in this group. In contrast to the elevated HSP70 levels, the DeAC hearts did not display high HSP90 protein or mRNA levels (Table 5.1). The inverse relationship between HSP70 and HSP90 has already been reported by other investigators who showed that the attenuated induction of HSP70 during STHA was accompanied by HSP90 overexpression (61). Additionally, the same work showed that treating

rat cardyomyocytes with an HSP90 inhibitor – geldanamycin (GA) leads to an overexpression of other HSPs, including HSP70. HSP90 is an essential component in the heat shock response and the duration of its upregulation is critical to cellular integrity (20). Furthermore, the role of HSP90 in cardioprotection is also attributed to enhanced NO signaling, and the targeted overexpression of hsp90 reduces two important facets of infarct size and myocardial dysfunction (53). HSP90 has been identified directly either as the effector allowing the activation of the pro-survival PI3K/Akt pathway (13) or as an obligatory chaperone for involved receptors (7). Hence, our conclusion from this part of the study is that the low HSP90 mRNA and subsequent protein levels is perhaps, one reason for the failure of cytoprotection, even in the presence of a large HSP70 reserve in the DeAC state. Environmental effects result in a global genomic/proteomic responses (27). Therefore, a change in HSP90 is likely to be only one of many factors involved in shaping the DeAC phenotype.

The upregulation of hsf1 transcript and its encoded protein provides one mechanism for rapid ReAC and cytoprotection

The marked upregulation of the hsf1 transcript (graph 4.2 in the Results chapter) and its encoded protein (114), as well as the enhanced phosphorylated HSF1 fraction in the DeAC and ReAC hearts described by us (114), may imply one mechanism for rapid ReAC and cytoprotection recovery. The binding of HSF1 to the HSE initiates the transcriptional activation of the HSP chaperone family, thus, upregulation of hsf1 may contribute to the rapid induction of hsp70 and hsp90 transcription following heat stress. Additionally, the high levels of hsf1 in the DeAC and ReAC groups could facilitate the faster conversion of inactive HSF1 monomers to active trimmers and significantly accelerate constitutive HSP70 even in normothermic DeAC conditions, as we saw in our experiments. Sorger et al. (109), Høj and Jakobsen (31), and Hashikawa and Sakurai (30) demonstrated that HSF1 binding causes a conformational change required to overcome promoter and chromatin repression in a stress-induced and promoter-specific manner. Concomitantly, Duncan (20) proposed that post-transcriptional changes in HSF1 in yeasts, due to hyperphosphorylation, may also confer chromatin changes. Interestingly, a preliminary finding in our laboratory provided evidence for a constitutive HSF1 binding to the HSE following an

acclimation period of 30 d (Maloyan and Horowitz unpublished). The hsf1 high mRNA and phosphor-HSF1 levels together with the occurrence of constitutive HSE-HSF1 binding in stress- related genes promoters in the DeAC group, may imply a preconditioned chromatin state, allowing rapid chromatin changes and enhanced transcription and post-translational changes (including that of hsp90 because hsf1 is responsible for its transcriptional activation) upon reexposure to heat.

To summarize this part of our study:

 We have demonstrated a faster mRNA dynamics of hsp70 and hsp90 genes in the DeAC and ReAC groups. The transcriptional profile post-HS of both genes were similar to those seen in the AC group, indicating that faster transcriptional dynamics is part of the memory repertory.  The DeAC phenotype, however, is characterized by a mismatch between the greater HSP70 mRNA and protein reserves and the loss of cardioprotection, which can be explained by a decrease in hsp90 (an important cellular chaperone) mRNA and protein levels in this group.  The high content of mRNA, seen in this work, and the high levels of phosphorylated hsf1 protein in the DeAC group (114), may imply the existence of a predisposition to the rapid switch of the HSR transcriptional system and the further cardioprotection in the ReAC group, even after it is lost at normothermic conditions.

5.2 Post-translational modifications in histones underlie heat acclimation-mediated cytoprotective memory

Our next step in this research was (i) to test the hypothesis that ―acclimatory memory‖ stems from epigenetic transfer of information, and (ii) to unravel the upstream mechanisms involved. To test our assumptions, we used Chromatin immunoprecipitation (ChIP) analysis combined with qRT-PCR, to directly detect changes in chromatin modifications and TF‘s binding to an open chromatin of genes of interest. This section of the work provides substantial evidence of chromatin modifications and TF and chromatin regulators recruitment throughout AC, DeACand ReAC, as summarized in Table 5.2.

These modifications include the maintenance of active chromatin states at specific sequences, and probably facilitate the immediate transcription of cytoprotective genes upon re-acclimation, as discussed in Tetievsky et al. 2010 (114).

Table 5.2: Histone modifications and HSF1, MSK1 and TIP60 recruitment to hsp70 and hsp90 promoters with correlation to mRNA, protein and cardioprotection levels.

Cardio- P H3Ser10 MSK1 Ac H4 TIP60 HSF1 mRNA Protein* protection*

AC2d ↑ ↑ ↑ ↑ ↑ ↑ ↑

AC ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑

DeAC ↑ ↑ ↑ ↑ ↑ ↓ ↑ ↓ ↑ ↓ ↓

ReAC ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑ ↑

The arrows indicate up/down-regulation compared to the control group. Empty cell indicate not significantly different from control level *The protein levels and cardioprotection data from Tetievsky et al. 2008 (114), mRNA, histone modifications amd MSK1, HSF1 and TIP60 recruitment taken from Tetievsky et al. 2010 (115). The mRNA levels presented here are basal levels (the data taken from insets in graphs 4.1.1a and (4.1.1b in Results chapter). P H3Ser10 – phosphorylation on Serine 10 histone H3; MSK1 - Mitogen- and stress-activated protein kinase 1; Ac H4 – acetylation of histone H4 on lysines 5, 8, 12, and 16; TIP60 - 60-kDa HAT-interactive protein; HSF1 – heat shock factor1. For experimental conditions see Fig. 3.1 in Methods.

Histone modifications at the hsp70 promoter

Histone modifications, such as acetylation and phosphorylation, determine the functional state of chromatin (118, 119), and, therefore, these posttranslational changes found in specific histones may be a sign of transcriptional activity. Nowak and Corces (81) demonstrated in Drosophila that an elevated H3Ser10 has been found in transcriptionally activated heat shock loci in the chromosome puffs of heat-shocked larval Drosophila salivary glands. In addition, they showed that the loci, which were histone H3 phosphorylated after heat shock, were histones H3- and H4-acetylated before heat shock, suggesting that a gene locus can be acetylated, even if it is not actively transcribed. Labrador and Corces (54) provided evidence that P-H3Ser10 is essential in the transcriptional induction of hsp70, irrespective of the acetylation state, raising the hypothesis that P-H3Ser10 is required to establish a recognition site (conserved arginine 164 in GCN5) for subsequent acetylation. In mammalian cells, Solomon et al. (108), and Thomson et al. (116) using immunoprecipitation of in vivo cross-linked histone H4-containing hsp70 chromatin fragments, showed that only histone H4 (and not H3) acetylation of hsp70 chromatin is associated with the heat shock-induced changes in the hsp70 gene. As discussed above, the two histone modifications presented here, histone H3 phosphorylation (Serine 10) and histone H4 acetylation (lysines 5, 8, 12, and 16), may play a role in the regulation of heat stress-induced hsp70 transcription. Hence, using the available data we can examine the histone modifications detected in our protocol and determine whether these combined modifications are involved in acclimatory cytoprotective memory. In this investigation, we found elevated histone H3 phosphorylation at the serine 10 site in both the AC2d and ReAC groups; i.e., this modification occurred at the onset of each AC regimen, when the animal experiences maximum strain (90). In contrast, high acetylation levels were detected in histone H4 after 30 days of AC, as well as in the DeAC and ReAC groups (Table 5.2). The detection of elevated histone H4 acetylation at lysines 5, 8, 12, and 16 in the AC, DeAC, and ReAC groups and elevated histone H3-phosphorylation at Ser10 (both associated with the hsp70 promoter region) following AC2d and ReAC, but not during DeAC, implies that the histone H4 modification is ambient-temperature independent, whereas histone H3-phosphorylation modification occurs only upon exposure to the higher acclimating temperature. Histone H3 acetylation levels, in contrast, were similar in all experimental groups.

Our present data not only confirm the induction of histone H3 phosphorylation in hsp70 promoter upon transfer to a hot environment but also imply a constitutive histone H4 acetylation effect in the production (via transcription) of sustained elevated HSP70 reserves, independent of ambient temperature (67, 114). Because histone H3 phosphorylation is a transient response, its contribution to the subsequent histone H4 acetylation seen in our study is irrefutable. Furthermore, this conclusion agrees with other reports, demonstrating that histone acetylation may require a phosphorylated histone H3 at the inducible gene locus (17). To confirm the findings of high histone H4 acetylation in the AC, DeAC, and ReAC groups, we measured the recruitment of a specific histone H4 histone acetyltransferase (HAT) (TIP60, the catalytic subunit of NuH4) to the hsp70 promoter (93, 129). Previous studies provided evidence that this complex is involved in the activation of transcriptional programs associated with repairing DNA strand breaks and apoptosis upon environmental stress (110). We found a strong correlation between the levels of histone H4 acetylation and the recruitment of TIP60 to the hsp70 promoter. Similarly, to confirm the findings of high histone H3 phosphorylation in the AC2d and ReAC groups, we measured the recruitment of a specific histone H3Ser10 kinase MSK1 (Mitogen- and stress-activated protein kinase 1) to the hsp70 promoter. Stress-induced phosphorylation of histone H3 is mediated by mitogen- and stress-activated protein kinases (MSK) 1 serine/threonine kinases that are activated via both the ERK1/2 and p38 MAPK pathways (19). We found a significant correlation between the levels of histone H3Ser10 phosphorylation and the recruitment of MSK1 to the hsp70 promoter, with high levels of phosphorylation and MSK1 recruitment in the AC2d and ReAC groups. Consistent with this finding, our measurements of activated phosphor-MSK1 protein levels showed significant elevation in AC2d group. Several studies show that MSK1 is activated by extracellular stimuli in non-dividing cells, leading eventually to the phosphorylation of Ser-10 in histone H3 and well as Ser-133 in CREB and Ser- 63 in ATF1 (5) in correlation with rapid transcriptional induction of a subset of genes (87).

Validation of hsp70 chromatin accessibility to TFs binding

Histone modifications cannot provide a full understanding of transcriptional activity. In our experiment, designed to address HSF1 binding to the HSE of hsp70, the detection of constitutive HSF1 binding to HSE in the ―nonstressed‖ DeAC group, possibly allowing high transcription and translation 30 days post-AC (114), agrees with several studies demonstrating

that HSF1 can bind constitutively to the hsp70 promoter under long-term non-stressful conditions. Similar mechanisms induce high levels of the hsp70 transcriptome in species evolutionarily adapted to desert conditions by maintaining large HSP70 protein reserves vs. matched species inhabiting temperate areas with limited HSP70 reserves (21, 130). The question remains is how do histone H3 phosphorylation and histone H4 acetylation at the HSE site of the hsp70 promoter work together to promote transcriptional memory. According to Cheung et al. (17), the synergistic coupling of histone H3 phosphorylation and acetylation suggests that histone H3 phosphorylation can affect the efficiency of subsequent acetylation. Therefore, our data imply that the increased p-H3Ser10 during AC2d might define a particular recognition locus that is primed for HSF1 binding and further hsp70 transcription, as described in the literature (132). Thomson et al. (116), using HSF1 -/- fibroblasts, demonstrated that HSF1 is an essential factor for directing inducible histone H4 acetylation at the hsp70 promoter upon heat stress. This finding led us to suggest that HSF1 binding during AC2d and AC recruits histone H4 acetylation to the locus during DeAC. Our observation of a significant correlation between HSF1-histone H4 acetylation seen only during AC, DeAC, and ReAC confirms constitutive chromatin opening during the 30-d DeAC and 2-day ReAC periods, resulting in elevated transcriptional events. Hence, long-term histone H4 acetylation might promote an open chromatin state and constitutive binding of HSF1 to the hsp70 promoter, leading to the high transcription and translation of the gene long after the initial AC period, as well as when AC conditions are resumed.

Histone modifications at the hsp90 promoter

HSF1-HSE binding controls other HSP species involved in the HSR, including hsp90 (77). HSP90 is not only a critical component of the HSR and protein quality control, it is also a master regulator, controlling hubs in homeostatic signal transduction and regulating chromatin structure and gene expression (85). Because HSP-90 protein levels in our AC-DeAC-ReAC protocol dropped during DeAC (loss of AC), whereas HSP70 levels remained elevated (114), we examined chromatin modifications in the HSE of hsp90. We found increased H3Ser10 phosphorylation (together with MSK1 recruitment) only in the AC2d group, yet the higher levels of histone H4 acetylation established during AC2d persisted throughout the AC, DeAC, and ReAC protocol. Surprisingly, elevated HSF1 binding to HSE of hsp90 was detected in AC2d, AC, and ReAC, but not in the DeAC phase, despite histone H4 acetylation at that phase. These

data are congruent with Zhao et al. (132) and Uffenbeck and Krebs (121), who used an HSF1- HSP82 activation model in yeasts and showed that, following heat stress, rapid transient histone H4 acetylation is required for subsequent HSF1 binding, induction of the DNase I hypersensitive site, and nucleosomal displacement. Although we examined only constitutive features, the temperature effects on chromatin remodeling described by Zhao et al. (132) allow certain deductions. The data suggest the importance of constitutive histone H4 acetylation when exposed to acclimating temperatures, as well as HSF1-HSE binding in the hsp90 gene. The lack of HSF1-HSE binding during DeAC may partially explain the decrease in HSP-90 levels at this phase and the subsequent loss of cardioprotection (114), as hypothesized above.

Heat AC predisposes to cytoprotective memory: support of concept

In summary, the profile of chromatin remodeling at the HSE of the promoter site of two hsp genes provides a conceptual model (Fig. 5.1) of the evolution of AC memory. We suggest the following chain of events: 1) At the onset of AC (AC2d), MSK1 recruitment and histone H3 phosphorylation switches on HSF1 binding at the HSE with subsequent histone H4 acetylation at the HSE of both hsp70 and hsp90 genes. This reaction occurs in a temperature dependent manner. 2) The acetylation persists throughout DeAC and ReAC, resulting in constitutive HSF1 binding to the hsp70 promoter, irrespective of transitions in ambient temperatures. In contrast, HSF1 binding to hsp90 is temperature dependent. HSF1 binding to hsp90 does not occur in DeAC, despite elevated TIP60 activation and histone H4 acetylation. Because HSF1-hsp90 binding requires elevated temperatures, the increased histone H4 acetylation may facilitate the rapid resumption of HSF1 binding, hsp90 transcript translation, and the reformation of a cytoprotective milieu upon ReAC.

Figure 5.1: Conceptual model of the evolution of AC memory with cross-talk between the transcriptional activation of hsp70 and hsp90 genes. Figure 5.1: Conceptual model of the evolution of AC memory with cross-talk between the transcriptional activation of hsp70 and hsp90 genes. Chromatin panel (top) presents post-translational H3 and H4 modifications and transcription factor binding, and the protein panel (bottom) presents the encoded proteins. Top and bottom panels represent the treatment protocol and cytoprotection, respectively. At the onset of heat AC (AC2d), histone H3 phosphorylation by MSK1 switches on HSF1 binding to the HSE with subsequent histone H4 acetylation at the HSE of both hsp70 and hsp90 genes. The acetylation persists throughout DeAC and ReAC, resulting in constitutive HSF1 binding to the hsp70 promoter throughout AC2d, AC, DeAC, and ReAC. Concomitantly, HSF1 binding to hsp90 and the encoded protein are absent in DeAC. The constitutively elevated histone H4 acetylation at the hsp90 gene promoter, however, may facilitate the rapid resumption of HSF1 binding, hsp90 translation, and the formation of a cytoprotective milieu upon ReAC. This is displayed by constitutively greater HSP70 reserves throughout AC-ReAC, whereas HSP-90 was elevated only in AC and ReAC hearts. Cytoprotection during these phases is displayed by reduced cardiac infarct size. Notably, whereas H3 phosphorylation is ambient-temperature dependent, H4 is ambient- temperature independent. H3, histone H3; H4, histone H4; P, phosphorylation; AC, acetylation; TA, ambient temperature effect. For other symbols, see Fig. 3.1 in MethodsTaken from Tetievsky et al.2010 (115). 91

5.3 Identification of pathways involved in the generation of heat-acclimation molecular “memory”

To-date, our knowledge of genome-wide responses of the heat-acclimated phenotype is limited to changes in transcription levels of specific genes, during the acclimation process. This investigation delineates, for the first time, the whole-genomic response in a mammalian species during the AC process and also in the DeAC and ReAC regimes. From our analyses, we 1) outline the dynamics of the genomic response of different sets of genes in all the experimental groups, thus allowing some perception of the global acclimatory molecular strategy underline heat acclimation and, and 2) discuss the likely pathways leading to the ―molecular memory‖ formation conferred by heat acclimation. Understanding the mechanism of acclimation memory and the signature of ―core-gene clusters‖ upon DeAC/ReAC can thus be a means to identify new candidate genes, inducible and non- inducible, that are essential for temperature acclimation. The identification of such genes and discovering whether they are comparable with those that signify evolutionary heat adaptation will expand our knowledge of acclimation phenotypic plasticity across species. The attempt to explore the upstream epigenetic regulatory events leading to acclimation memory, if successful, will provide further insight into the broad role of chromatin as a physiological template during ―within lifetime‖ adaptations.

From a total of 27,342 distinct probes, 651 genes, showing a significantly changed transcriptional behavior (either upregulated or downregulated > 1.5 - fold) at least in one of the experimental groups, were selected. The clustering revealed five gene clusters characterized by a significantly identical transcriptional behavior of the genes in each cluster. Among these clusters we can clearly distinguish between the cluster in which the genes are affected only by short-term heat acclimation for 2 days (AC2d) and the clusters in which the gene transcription significantly changed in response to the DeAC and ReAC regimes, either both or each. The magnitude of change in expression of these gene groupings shows temporal variation over the course of AC/DeAC/ReAC.

5.3.1 Short-term heat acclimation (AC2d group) switching on an altered expression of genes encoding DNA damage repair

We previously provided evidence that STHA induces molecularly based physiological and biochemical adaptations (34). Given the biphasic acclimation profile, during which temporal perturbations in cellular homeostasis take place (34), correlative changes in stress-gene dynamics are likely to occur. The results of the present investigation confirm this view and support our hypothesis that STHA is a critical acclimatory checkpoint. The majority of 18 transcripts increased in expression upon AC2d (STHA), reflecting a basic paradigm of exposure to high environmental temperatures - that organisms frequently compensate for the stress effects of high temperature by transcribing more cytoprotective genes (39). Functional annotation showed that 24% of the genes in this cluster were associated with cytoprotective pathways related to apoptosis, metabolism, and mitochobdria function (Fig. 4.14 in Results chapter). The interesting finding is that some genes assigned to this category are associated with maintaining DNA integrity. The enhanced expression of Prime1 (Primase1), a gene that synthesizes small RNA primers for the Okazaki fragments, engaged in double-stranded DNA repair and DNA synthesis (68), implies that the onset of AC leads to possible double-stranded DNA damage. This finding is supported by Horowitz et al. (39), which showed a significant upregulation of topoisomerase II and ERCC1 (genes involved in double-stranded DNA repair and DNA synthesis pathways) at STHA. Additionally, Kultz et al. (52), reported that the onset of adaptation to hypertonic stress is accompanied by an upregulation of genes encoding proteins associated with the maintenance of DNA integrity to allow different sets of adaptations to develop. Activating these genes could provide an initial line of defense for alleviating any genomic damage resulting from the strain associated with the onset of the acclimation process.

Another surprising finding is the presence of miRNA 297 (miR-297) upregulated transcripts at AC2d group. This finding demonstrates, for the first time, the involvement of miRNA- regulation in the response to the onset of HA conditions. Currently not much is known about miRNA 297 specific targets or functional pathways. A recent study by Zhang et al. (131) showed that miR- 297 is capable of regulating MLLT3 (Myeloid/lymphoid or mixed-lineage leukemia translocated to chromosome 3 protein) gene expression negatively by binding to its 3'-untranslated region. MLLT3 is a critical regulator of cell proliferation or carcinogenesis. In addition, overexpression of

miR-297 also led to the altered expression of p27(Kip1), an important regulator of cell cycle progression (131). Interestingly, previous evidence showed that chaperone complex HSP90-HSP70-p60/Hop directly and specifically interacts with MLLT3 (myeloid/lymphoid or mixed-lineage leukemia) (60). Lin et al. suggest that the interaction of HSP90, HSP70 with MLLT3 is necessary for the proper subnuclear localization and activity of Milt3 (60). Our finding showing a marked upregulation of miR-297, an important MLLT3 regulator, suggests that significant upregulation of miR-297 by STHA may confer protection from an abnormal expression of MLLT3 and further carcinogenesis. This finding may be a first clue to a connection between heat acclimation and cross-tolerance to an abnormal cell proliferation, and may open a new and exciting therapy and prevention possibilities in the future.

5.3.2 Identifications of the core genes significantly expressed in DeAC and ReAC groups.

The bi-phasic sequence of events-AC/DeAC as well as memory testing (ReAC) allows us to examine the ―on call‖ protective pathways and the outcome. Moreover, we can study the transcriptional dynamics of the whole rat genome throughout the maintenance of the predisposing state (DeAC) and memory testing (ReAC). Indeed, four clusters revealed groups of genes that were simultaneously either up-regulated or down-regulated after DeAC and ReAC or both. The four clusters can be divided into: 3. Two clusters that showed upregulation at DeAC and ReAC (cluster #4) and at DeAC alone (cluster #5). 4. Two clusters that showed downregulation at DeAC and ReAC (cluster #2) and at ReAC alone (cluster #3). In this section, I chose to focus on discussing the molecular pathways that connect to a chromatin/histone regulation and can reveal the upstream mechanism responsible for transcriptional predisposition and ―molecular memory‖. The results of the present study call our attention to transcriptional and chromatin-regulation associated genes that have not yet been linked with AC in mammalian species. The functional annotation analysis of the significantly expressed genes in the DeAC and ReAC periods revealed a group of genes that may act as upstream regulators of histone modifications and

transcriptional activation and confer molecular memory. Table 5.3 represents the detected genes linked to chromatin-regulating pathways for which transcription was significantly influenced by DeAC and ReAC.

Table 5.3: The summary of genes that belongs to the chromatin-remodeling pathways and significantly changed their expression in DeAC or ReAC groups.

Down-regulated Up-regulated

Cluster 2 Aebp1 MAPK12 Cluster 4 Creb3 ATF3 Smarca1 Cnksr1 Cited1 Cited2

Cluster 3 PBK Nfkbia PLK Dusp14 an Dusp15 AURKB Csdc2 ANKRD41 ANKRD23 and ANKRD6 Mettl7a

Snrpn Cluster 5 Snurf The left column representing genes that were down-regulated at DeAC and ReAC groups (#2) and ReAC only (#3). The right column representing genes that were upregulated at DeAC an ReAC groups (#4) and DeAC only (#5). Nine genes (marked in red) are related to the P38 MAPK/ERK/MSK signaling cascade. For genes biological functions see the tables 2-6 in the Supplements chapter.

MAPK pathway – an upstream regulator of histone phosphorylation and acetylation involved in molecular acclimatory memory

We found that 10 of 19 histone-modifications regulators are annotated to the P38 MAPK/ERK/MSK signaling pathway (Table 5.3). The MAPK (Mitogen activated protein kinase) pathways are key in regulating stress responses and transducing extracellular signals to cytoplasmic and nuclear effectors (18, 101). The MAPK superfamily consists of three main protein kinase families: extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase and the p38 MAPK pathway (18). 95

P38 MAPK are a class of protein kinases that are responsive to stress stimuli, such as cytokines, ultraviolet irradiation, heat shock, and osmotic shock, and are involved in cell differentiation and apoptosis (18, 101). Several studies showed that this pathway is activated by stress-induced Ca2+ -dependent or K+ -channel dependent mechanisms that are known to occur at the onset of AC (98, 101).

Stress stimuli

Figure 5.2: P38 Mitogen activated-protein kinase (MAPK) signaling pathway. MSK1, CREB, ATF, and others are activated by phosphorylation by p38. The CREB, ATF1, and NF-kB regulative unit Nfkbia was found significantly expressed in the microarray result. MSK1 phosphorylates Histone H3Ser10 in response to stress stimuli and p38 MAPK pathway activation. The figure was taken from http://www.sabiosciences.com.

The ChIP experiments, described in this work, showed significant elevation in the phosphorylation of histone H3Ser10 in promoters of hsp70 and hsp90 genes in AC2d and in hsp70 in both AC2d and ReAC groups. In addition, our results revealed a correlation between phosphor-H3Ser10- MSK1 recruitment and the protein level of an activated MSK1 in these groups. At least one of the roles of p38 kinase pathway is to activate MSK1 and to lead to epigenetic changes by phosphoacetylation (18, 104). Previous studies have created MSK1/2 knock-out mice, which are viable (107). These mice show marked reduction of H3Ser10 phosphorylation

and dampened responses to stress but continue to show changes in gene expression, as well as increases in histone H3K14 acetylation. Our results and the findings with the knock-out mice for MSK1 support a role for the phosphorylation of Ser-10 on H3 in modulating gene expression levels in the stress response, particularly in the beginning of AC (AC2d) and ReAC. Given that we have shown that at the onset of AC (AC2d), MSK1 recruitment and histone H3 phosphorylation occur coincidentally with HSF1 binding to the HSE (probably due to a higher chromatin accessibility at the promoter region) with subsequent histone H4 acetylation at the HSE of both hsp70 and hsp90 genes, we can argue that the upstream regulatory pathway that activates this signaling cascade may be the p38 MAPK pathway.

Another gene from this pathway, for which transcription was significantly expressed in our array results, was the CREB (cAMP response element-binding) gene. We showed that: (i) Phospho-CREB protein was significantly upregulated at the AC group and decreased in the DeAC and ReAC groups (Fig. 4.25); (ii) Significant downregulation of CREB transcription in DeAC and ReAC groups (array cluster#2 and (114); (iii) CREB is a transcriptional regulator of a large group of genes, significantly expressed in the array results (Fig. 4.24); (iiii) CREB is an important component of the MAPK P38 pathway, responsible also for histone acetylation and phosphorylation (5, 104, 107) . Given that MSK1 is a potent CREB kinase and that CREB phosphorylation is indeed taking place in the cellular response to environmental stressors (11), P-CREB in our experimental paradigm, may recruit one of the coactivator CREB-binding proteins (e.g. CBP, P300, pCAF) whose histone acetyl transferase (HAT) activity has been shown, at least in vitro, to acetylate histone H4 (97). How is the H4 acetylation of phosphor-histone H3Ser10 may be brought about? Based on the following observations: (i) physiological challenge, such as onset of AC (AC2d) causes an enhanced histone H3Ser10 phosphorylation by an activated MSK1; (ii) MSK1 apart from being a H3S10 kinase, is also a CREB kinase; (iii) the hsp70 and hsp90 gene promoters contain a CRE site (125) and (iv) pCREB is able to recruit the CREB-binding protein (CBP/p300, proteins with histone acetyl transferase (HAT) activity) to the promoters and to acetylate histone H4 at later stages: in AC, DeAC, and ReAC groups. Nevertheless, the exact signaling mechanisms underlying the acetylation of phosphorylated histone H3 are presently still unclear. It is possible that both CREB-mediated HATs and TIP60

target histones independently and their effects on the opening of the chromatin at the genes promoters and transcriptional output are additive or redundant. Precisely what other signaling pathways may be involved and how their relative contributions in histone regulation are integrated to produce a maintenance of histone H4 acetylation even 30 days post-AC, which allows HSF1 binding and transcriptional activation, remains to be seen.

CITED1 AND CITED2, involved in HIF1α regulation, may confer the loss of cardioprotection in the DeAC group

Our results from the array study can shed a light on additional AC/DeAC/ReAC-induced molecular pathways. One of the interesting findings in the array results was the significant and reversed expression of CITED1 (Cluster 2) and CITED2 (Cluster 4) genes in the DeAC and ReAC groups. The CITED [cAMP-responsive element-binding protein (CBP)/p300-interacting transactivator with glutamic acid/aspartic acid-rich tail] proteins belong to a family of transcriptional cofactors. The biological properties of CITED proteins include modulating a variety of cellular and developmental processes and responding to diverse biological and environmental stimuli (10, 79, 128). To date, four different CITED homologs have been reported in vertebrates. CITED2, which can function as an activator and a repressor depending on the tissue, is the most extensively studied of the four (10). Our special interest in this protein caused by the fact that CITED2 was shown to function as a repressor of hypoxia-inducible factor-1 (HIF-1) through competition for binding to the CH1 domain of CBP/p300 at the promoters of HIF-1α target genes (79). Genetic evidence indicates that loss of CITED2 is associated with the increased activation of HIF-1 target genes, supporting the hypothesis that CITED2 is a negative regulator of HIF-1α(10). The important hallmark of AC and AC-mediated cross-tolerance identified to date include enhanced reserves of HIF-1α. The presence of HIF-1α in unstressed AC hearts can mediate several metabolic changes, including a high pre-ischemic glycogen level and slowed glycolysis. These metabolic changes are aimed at improving the ability to cope with oxygen deprivation, and allow facing with ischemic/hypoxic insults in the heart tissue (38, 65, 117). Our finding showed a marked upregulation of CITED2 in the DeAC group. Given that CITED2 is a functional repressor of HIF-1α (10) and that we previously showed a loss of cardioprotection in

DeAC (the rat heart tissue had a significantly larger infarct area after exposure to ischemia/reperfusion treatments) (114), we may hypothesize that one of the reasons for the loss of cardioprotection in this group is HIF1-1α repression following the significant CITED2 upregulation. Notably, studies on cell lines demonstrated that HIF-1 is required for HSF1 function during hypoxia (65). Hence, upregulation of CITED2 and decreased HSF1 binding might be also connected. CITED1 showed an opposite (downregulated) transcriptional profile to that of CITED2 in DeAC group. Although no publications has indicated a direct involvement of CITED1 in HIF-1α regulation, CITED1 and CITED2 have an overlapping expression pattern and a dynamic expression in different conditions (12). We can assume that the upregulation of CITED2 may compensate for the loss of CITED1 in the DeAC group to maintain normal function of other pathways in which this protein is involved (12, 79). The interplay between these related genes is still not understood completely and may open a new and exciting direction for further study.

5.3.3 Heat deacclimation (DeAC) and reacclimation (ReAC) linked to significant changes in histone transcription

An extremely surprising finding in our Microarray analysis is the significant transcriptional up- down regulation of genes encoding histones in both DeAC and ReAC experimental groups (Table 5.3).

Table 5.3: The list of histone gene which transcription significantly (> 1.5 fold) up-down-regulated in DeAC and ReAC groups.

Cluster number Histone gene name Function 2 Hist1h2bl A member of the Histone 1 family 2 Hist2h2bb A member of the Histone H2b family 2 Hist2h3c2 A member of the Histone H3 family 3 Hist1h1a A member of the Histone 1 family 3 Hist1h2ai A member of the Histone 1 family 3 Hist1h2a A member of the Histone 1 family 3 Hist2h2ab A member of the Histone 2a family 3 Hist1h2bc A member of the Histone 1 family 3 Hist1h2bn A member of the Histone 1 family 3 H2afz A member of Histone H2A family member Z 3 H2a A member of Histone H2a 3 Hist1h1b A member of the Histone 1 family Cluster #2 - downregulated at DeAC and ReAC; Cluster #3 - downregulated only at ReAC.

The histones are highly positively charged proteins that package DNA, which is negatively charged, into the nucleus in the form of chromatin, the substance of chromosomes. Packaging was initially thought to be the only function of the histones, but it is now clear that they also participate in gene regulation via both classical and epigenetic mechanisms. Most of the packaging of genomic DNA by core histones occurs primarily during the S phase of the cell cycle, when DNA is being actively replicated; stoichiometric ally appropriate levels of histone proteins are required to bind DNA immediately following replication (69). Nevertheless, the regulatory mechanisms appear to act primarily by controlling the extent to which DNA is made accessible to regulatory factors, i.e., by controlling packaging (119). This activity is possible due to a post-transcriptional modification of histone proteins, as described in previous chapters. Very few studies are dealing with histone transcription in ―adult‖ non-dividing tissues (29). To our knowledge, this is the first time transcriptional regulation of histone genes has been linked to environmental temperature fluctuations and heat acclimation. Our results showed that Histone H1 genes are the most transcriptionally influenced by our model, 7 of 12 the significantly altered histone transcripts belong to the Histone H1 coding

100

sequence (table 5.3). Another finding is that 11 of the 12 histone transcripts are found in clusters #2 and #3 which were down-regulated during the DeAC and ReAC periods. Among these, the majority were down-regulated only in the ReAC group. It was previously shown that DNA damage induced by ionizing radiation in mammalian cells results in the down-regulation of histone transcription, which is caused by a possible decrease in histone mRNA half-life and the DNA replication arrest (111). Based on our results, revealing the upregulation in DNA-damage/repair genes in the AC2d group [(39) and discussion chapter 5.3.1], we may hypothesize that STHA for 2 days may cause some double-stranded DNA damage and influence the histones mRNA levels. Another possible explanation may be associated with the marked transcriptional expression of the group of genes encoding histone H1. The linker histone H1 sits at the base of the nucleosome near the DNA entry and exit sites to stabilize two full turns of DNA. Histone H1 participates in nucleosome spacing and the formation of the higher-order chromatin structure. In addition, H1 seems to be actively involved in the regulation of gene expression. Histone H1 in mammals is a family of closely related, single-gene encoded proteins, including five somatic subtypes (from H1.1 to H1.5) and a terminally differentiated expressed isoform (H1.0). Whether the different variants have distinct roles or if they regulate specific promoters is not well known (95). Previous studies on the effect of H1 depletion on global gene expression have reported changes in the expression of small groups of genesrather than affecting the vast majority of cellular genes (95). Overexpression experiments have also contributed to challenge the concept of H1 as a general repressor of chromatin activity. In Xenopus embryos, over-expression of the somatic H1 variant repressed oocyte- 5S rRNA genes or other Pol III transcripts (99). H1 depletion could also have an effect on the formation and stability of chromosomes. In the absence of H1, chromosomes assembled and replicated in Xenopus egg extracts failed to compact properly, leading to segregation anomalies. Nonetheless, the decrease in H1 leads to a general increase in telomere length (78). Chromatin misorganization could also cause changes in the expression of sensitive promoters that may include genes relevant for cell proliferation. Alternatively, H1.2 may be a specific regulator of certain particular promoters (99). Given our findings that large gene groups were simultaneously up/down-regulated in the DeAC and ReAC groups [(114) and Discussion chapter], possibly histone H1 might have differential functions in some of its nuclear roles, including nucleosome spacing and in turn, gene expression control. These regulatory mechanisms could contribute to establishing or

101

maintaining specialized regions of chromatin in open or close states necessary for TFs binding and further transcription in the DeAC and ReAC groups.

5.3.4 miRNA may be involved in the transcriptional regulation of target genes in both DeAC and ReAC groups

MicroRNAs (miRNAs) are a class of non-protein-coding RNA transcripts that regulate gene expression at the post-transcriptional level. miRNAs control gene expression by binding to complementary sequences (miRNA response elements; MREs) in the 3′ untranslated region (3′ UTR) of target mRNA transcripts to facilitate their degradation and/or inhibit their translation. A single miRNA can affect mRNA and protein levels of a large numbers of genes, and several miRNAs can bind to the same mRNA target, allowing for a multifaceted pattern of gene expression regulation (6, 8). Regulation of distinct miRNAs has been well established as part of the typical stress response; to date, thousands of publications have documented their role in a multitude of cell functions and include research articles on the roles of miRNAs in ischemic injury, cardiac hypertrophy, muscle atrophy, neurodegeneration, cancer, osmotic responses, cellular proliferation and apoptosis. ischemic injury, cardiac hypertrophy, muscle atrophy, neurodegeneration, cancer, osmotic responses, cellular proliferation and apoptosis [reviewed in (2, 71, 72)].

One of the interesting outcomes of the current study was the finding of 8 genes encoding miRNAs that significantly changed their transcription in the DeAC and ReAC periods. The summary of the miRNAs genes detected in our analysis is displayed in Table 4.4 in the Results chapter. The majority of miRNA genes (six of eight) were downregulated in the DeAC and ReAC groups: four were downregulated at DeAC and ReAC both (miRNA-212, miRNA-224, miRNA-328, miRNA-490), and two were downregulated only at ReAC (miRNA-208, miRNA-375). Two additional miRNAs (miRNA-29b and miRNA22) were upregulated in the DeAC and ReAC groups. Significant expression of miRNA especially in the DeAC and ReAC groups could stem from the need to regulate a large groups of genes simultaneously [among them such stress- related genes as hsp70 and anti-apoptotic bcl-xL (114)], without conferring cardioprotection but probably predisposing to faster ReAC in these regimes. Moreover, the miRNA activity in these

102

groups may play a role in stress memory formation as it often acts to attenuate or silence gene translation in a transient mode of operation (89). The miRNAs as ―fine tuners‖ of gene expression and protein translation in an ever-growing range of cell and physiological function has generated much interest. This paradigm fits well with the possibility that miRNAs are involved in the dynamic epigenetic regulation of gene transcription in changing environments (23, 24) and contribute to AC ―molecular memory‖ formation. miRNA are able to regulate target genes expression by different post-transcriptional mechanisms in DeAC and ReAC groups

The results obtained from this study show that six of nine miRNAs have corresponding target genes (a total of 48 target genes). If degradation of miRNAs were the predominant mechanism responsible for changes in gene expression in our study, then one would expect first that a downregulation of miRNA would be accompanied by an upregulation of gene expression; and, conversely, an upregulation of miRNA would lead to a downregulation of gene expression. We therefore examined the relationship between the specific directional change in miRNA and their targets gene expression. Our results show that out of 48 target genes, 21 genes (~43%) showed the a transcriptional profile opposite to their miRNA transcriptional profiles, and 24 target genes (50%) showed the same transcriptional profile as their regulatory miRNA. Three genes did not show any clear significant correlation between the miRNA and the gene transcriptional profile. However, some miRNA were clearly (almost 100%) transcribed in opposite way to their targets, as miRNA -29b-2 and miRNA 212, and they predominantly have the same transcriptional profile as their targets: miRNA -224, miRNA-328, and miRNA-375. A possible explanation for this mismatch between some miRNA transcriptional profile and their target genes mRNA, as revealed in the DeAC and ReAC groups, could stem from the different mode of operation by which miRNA is able to regulate target genes expression. Although miRNAs have primarily been demonstrated to mediate dependent decay of messages through partially complementary miRNA target sites in target mRNA (71, 72), an emerging assortment of studies, revealed that miRNAs and their associated protein complexes (microribonucleoproteins or microRNPs) can additionally function to posttranscriptionally stimulate gene expression by direct and indirect mechanisms (86, 89). These reports indicate that miRNA‐mediated effects can be selective, are regulated by the RNA sequence context, and are associated with cellular

103

conditions. Repression, translation and upregulation by miRNAs ranges from fine‐tuning effects to significant alterations in expression without necessary levels correlation between miRNA and the targets gene transcription. For example, miRNAs are capable of silencing gene expression in the cytoplasm by mechanisms that can result in translational repression in which degradation of mRNA is not the primary mechanism (86). Additionally, some investigators suggest that degradation of mRNAs occurs only when a perfect match is found between the miRNA and its mRNA target, whereas others claim that even perfect base pairing in the seed region does not always guarantee the down-regulation [reviewed in Fabian et al. (25)]. These studies uncover remarkable, new abilities of miRNAs in the control of gene expression and underscore the importance of regulation, in cis and trans, in directing appropriate trancriptome responses.

The influence of miRNA on signaling pathways in DeAC and ReAC groups

When looking specifically at the miRNA target gene groups, the surprising revelation is that 75% of the target genes of the miRNAs whose expression level had been influenced by one of the treatment groups are connected to the p38 MAPK pathway, while regulated by one of the TFs from this pathway: CREB, ATF or c-jun/c-fos. MAPK/ERK cascades detect, amplify and integrate diverse external signals to generate responses, such as changes in protein activity and gene expression, and may provide a conduit for a rapid response in stress-responsive miRNA expression. The results together with the transcriptional activation of a large group of genes that are involved in the MAPK/ERK pathway in DeAC (as shown in this work), provide an indication that MAPK/ERK signaling could yield regulatory effects on the miRNA signaling pathways through miRNA-influenced biological responses. Indeed, a recent study found that the expression of miR-221 and miR-222 was induced by nerve growth factor stimulation and that this induction was dependent on sustained ERK1/2 activation (113). A similar study, further examining the role of MAPK influence on miRNA expression, showed that the ERK pathway was also capable of enhancing the stability of the miRNA-processing complex (84). Not surprisingly, miRNAs are also important in regulating skeletal and heart muscles growth and differentiation (16). These miRNAs include miRNA-29b that is transcriptionally upregulated in the DeAC and ReAC groups. miRNA-29b provides a mechanism for atrophy protection in myocytes (94). Safdar et al. (94) have shown that miRNA-29b also represses pro-atrophy genes (such as Hox-A11, PI3-K/Akt, MuRF1, MAFbx, and utrophin) and favors the maintenance of muscle during hibernation. Another interesting finding is that among the targets of miRNA-29b,

104

is the Aurkb (Aurora kinase b) gene, a serine/threonine kinase responsible also for the phosphorylation of histone H3 and chromatin remodeling (45). This finding may provide a link between miRNA-29b and chromatin regulation of the genes responsible for myocyte protection and reveal some possible miRNA-29b non-direct way of gene transcriptional regulation. MicroRNAs may also provide a mechanism for the expression of critical genes immediately after stress removal by sequestering selected transcripts into stress granules or P-bodies known to store miRNA-targeted mRNAs during stress (133). The number of P-bodies increases under certain environmental stress conditions, including nutrient deprivation and osmotic stress (46). This knowledge provides a useful basis for miRNAs to establish rapid biological controls, which regulate metabolic rate depression during periods of environmental stress. Similarly we can hypothesize that miRNA may promote stress-related mRNA during the post-heat acclimation period (DeAC). By temporarily storing mRNA transcripts in stress granules or P-bodies, miRNAs provide a mechanism by which animals can rapidly emerge from a suspended condition and reinstate normal cellular activity. In summary, our data presented here demonstrate that alterations in miRNAs may play a role in regulating the transcription of large groups of genes in DeAC and ReAC groups by a possible involvement of the MAPK/ERK pathway. The mismatch between the miRNA levels and transcriptional profile of their target genes could imply a various post-transcriptional mechanisms that miRNA uses to regulate gene transcription.

5.4 General conclusions

Our investigation has demonstrated that:

1. Faster transcriptional dynamics of hsp70 and hsp90 is a part of the memory repertory. 2. The mRNA and protein levels decrease of HSP90 in DeAC, despite the high HSP70 levels, may cause the loss of cardioprotection in the DeAC group but do not interfere with predisposition to fast restoration of cardioprotection upon ReAC. 3. Post-translational histone tail modifications (Histone H4 acetylation and Histone H3Ser10) that lead to an open chromatin state at the heat shock element (HSE) of the hsp90 and hsp70 gene throughout the AC, DeAC, and ReAC regimen are associated with: i) a constitutive binding of HSF-1 to the hsp70 promoter and greater HSP70

105

reserves in normothermic temperatures (DeAC), and ii) a rapid resumption of the acclimated cytoprotected phenotype, when re-exposure to acclimation conditions replenishes HSP90 stores. 4. The whole rat genome profiling delineated in this investigation showed a marked difference among the experimental groups. Four large gene clusters comprising genes maintained an up-/downregulated state upon DeAC/ReAC, despite the return of the preacclimated physiological phenotype, were identified. 5. A large number of the significantly responding transcripts upon DeAC/ReAC are assigned to the p38 MAPK signaling pathway, which is responsible for the phosphor- acetylation of histones by activating the MSK1 kinase and CREB. From these findings, we can argue that the stress-inducible p38 MAPK pathway may be the upstream activator of the epigenetic events leading to the cytoprotective memory formation. 6. The marked change in histone H1 gene isoforms, particularly in ReAC conditions, may stem from either the role of histone H1 as a chromatin conformation regulator or the need to change histone H1 levels as a result of DNA damage in stressful conditions. 7. The finding of nine miRNA transcripts significantly expressed in the array, six of which are regulators of 48 genes whose expression was significantly influenced by our experimental condition, may indicate a miRNA-epigenetic regulation of genes involved in cytoprotection and memory.

5.5 Significance and future possible directions

To the best of our knowledge, no other study has demonstrated that exposure to persistent moderate environmental temperatures (AC) has an imprinted effect on the epigenome of a mammalian species in early adulthood, and induces stress-memory.

5.5.1 The significance of the study

Plasticity of the acclimated phenotype determines the upper limits of an organism‘s ability to cope with stressful environments. Unfortunately, our understanding of the molecular mechanisms involved is meager. Here, our findings that epigenetic features play a role in this within-life adaptation of the species by allowing for reversible phenotype variability in adulthood 106

expands our understanding of the adaptive span of this particular species, as well as of the impact of environmental extremes on epigenetic markers in general. Furthermore, the attempt to explore the upstream epigenetic regulatory events leading to acclimation memory, if successful, will provide further insight into the broad role of chromatin as a physiological template during physiological states. This research is of relevance in th present era, which is characterized by more frequent heat waves and greater instability along climatic border zones, causing changes in the composition of ecological communities, species diversity, and interfering with quality of life. Understanding adaptive limitations is, therefore, of prime importance among the populations at risk.

5.5.2 Future directions

This research may set the stage for a further understanding of the molecular programs that are involved in the AC ―memory‖ phenomenon. Future experiments might focus on multiple areas. The following are some directions that can be studied:  To establish an experimental model allowing pharmacological manipulations that ablate or induce rapid ReAC memory.  To study Heat shock factor (HSF) 1 and 2 interactions that control heat shock protein (HSP) induction: to unravel whether HSF2 is another player in acclimatory memory and cytoprotection (the study in this direction is already being conducted by Horowitz and Ioffe in our laboratory).  To assess the role played by Hypoxia-inducible transcriptional factor 1α (HIF1α) in the molecular AC memory formation.  To further explore the miRNA involvement in the AC phenotype and its transcriptional regulation in both DeAC and ReAC groups.  To continue studying the MAPK signaling pathway involvement in regulating histones modifications in response to a stress stimulus, in particular CREB and ATF involvements in this process.  To analyze whether the cytoprotective memory can be genetically passed through generations, and participate in evolutionary adaptations to changing environments.  To study the dynamic molecular and physiological processes taking place at 30 days of the DeAC period.

107

Being multidisciplinary, our study may become a basis to multiple and different scientific research directions. We believe that our findings will enhance the now flourishing area of environmental genomics in aspects associated with heat acclimation in homeotherms.

108

1. Aguilera O, Fernandez AF, Munoz A, and Fraga MF. Epigenetics and environment: a complex relationship. Journal of applied physiology 109: 243-251, 2010. 2. Ambros V. The functions of animal microRNAs. Nature 431: 350-355, 2004. 3. Arieli Y, Eynan M, Ertracht O, and Arieli R. Protection against CNS oxygen toxicity induced by heat acclimation involves increased levels of HSP72 and CuZnSOD. Faseb J 19: A671-A672, 2005. 4. Arieli Y, Eynan M, Gancz H, Arieli R, and Kashi Y. Heat acclimation prolongs the time to central nervous system oxygen toxicity in the rat - Possible involvement of HSP72. Brain research 962: 15-20, 2003. 5. Arthur JSC, and Cohen P. MSK1 is required for CREB phosphorylation in response to mitogens in mouse embryonic stem cells. Febs Lett 482: 44-48, 2000. 6. Baek D, Villen J, Shin C, Camargo FD, Gygi SP, and Bartel DP. The impact of microRNAs on protein output. Nature 455: 64-U38, 2008. 7. Bamberger CM, Wald M, Bamberger AM, and Schulte HM. Inhibition of mineralocorticoid and glucocorticoid receptor function by the heat shock protein 90-binding agent geldanamycin. Molecular and cellular endocrinology 131: 233-240, 1997. 8. Barbato C, Arisi I, Frizzo ME, Brandi R, Da Sacco L, and Masotti A. Computational challenges in miRNA target predictions: to be or not to be a true target? Journal of biomedicine & biotechnology 2009: 803069, 2009. 9. Barratt MJ, Hazzalin CA, Cano E, and Mahadevan LC. Mitogen-Stimulated Phosphorylation of Histone H3 Is Targeted to a Small Hyperacetylation-Sensitive Fraction. P Natl Acad Sci USA 91: 4781- 4785, 1994. 10. Bhattacharya S, and Ratcliffe PJ. ExCITED about HIF. Nat Struct Biol 10: 501-503, 2003. 11. Bilang-Bleuel A, Rech J, De Carli S, Holsboer F, and Reul JMHM. Forced swimming evokes a biphasic response in CREB phosphorylation in extrahypothalamic limbic and neocortical brain structures in the rat. Eur J Neurosci 15: 1048-1060, 2002. 12. Boyle S, Shioda T, Perantoni AO, and de Caestecker M. Cited1 and cited2 are differentially expressed in the developing kidney but are not required for nephrogenesis. Dev Dynam 236: 2321-2330, 2007. 13. Brouet A, Sonveaux P, Dessy C, Balligand JL, and Feron O. Hsp90 ensures the transition from the early Ca2+-dependent to the late phosphorylation-dependent activation of the endothelial nitric- oxide synthase in vascular endothelial growth factor-exposed endothelial cells. J Biol Chem 276: 32663- 32669, 2001. 14. Chadee DN, Hendzel MJ, Tylipski CP, Allis CD, Bazett-Jones DP, Wright JA, and Davie JR. Increased Ser-10 phosphorylation of histone H3 in mitogen-stimulated and oncogene-transformed mouse fibroblasts. J Biol Chem 274: 24914-24920, 1999. 15. Chang TC, and Mendell JT. microRNAs in vertebrate physiology and human disease. Annu Rev Genom Hum G 8: 215-239, 2007. 16. Chen PY, and Meister G. microRNA-guided posttranscriptional gene regulation. Biol Chem 386: 1205-1218, 2005. 17. Cheung P, Tanner KG, Cheung WL, Sassone-Corsi P, Denu JM, and Allis CD. Synergistic coupling of histone H3 phosphorylation and acetylation in response to epidermal growth factor stimulation. Mol Cell 5: 905-915, 2000. 18. Cowan KJ, and Storey KB. Mitogen-activated protein kinases: new signaling pathways functioning in cellular responses to environmental stress. J Exp Biol 206: 1107-1115, 2003. 19. Deak M, Clifton AD, Lucocq JM, and Alessi DR. Mitogen- and stress-activated protein kinase-1 (MSK1) is directly activated by MAPK and SAPK2/p38, and may mediate activation of CREB. Embo J 17: 4426-4441, 1998. 109

20. Duncan RF. Inhibition of Hsp90 function delays and impairs recovery from heat shock. Febs J 272: 5244-5256, 2005. 21. Evgen'ev MB, Garbuz DG, Shilova VY, and Zatsepina OG. Molecular mechanisms underlying thermal adaptation of xeric animals. J Biosciences 32: 489-499, 2007. 22. Eynan M, Knubuvetz T, Meiri U, Navon G, Gerstenblith G, Bromberg Z, Hasin Y, and Horowitz M. Heat acclimation-induced elevated glycogen, glycolysis, and low thyroxine improve heart ischemic tolerance. Journal of applied physiology 93: 2095-2104, 2002. 23. Fabbri D, Baravelli V, Chiavari G, and Prati S. Dimethyl carbonate as a novel methylating reagent for fatty acids in analytical pyrolysis. J Chromatogr A 1065: 257-264, 2005. 24. Fabbri M, Garzon R, Cimmino A, Liu Z, Zanesi N, Callegari E, Liu S, Alder H, Costinean S, Fernandez-Cymering C, Volinia S, Guler G, Morrison CD, Chan KK, Marcucci G, Calin GA, Huebner K, and Croce CM. MicroRNA-29 family reverts aberrant methylation in lung cancer by targeting DNA methyltransferases 3A and 3B. Proc Natl Acad Sci U S A 104: 15805-15810, 2007. 25. Fabian MR, Sonenberg N, and Filipowicz W. Regulation of mRNA Translation and Stability by microRNAs. Annual Review of Biochemistry, Vol 79 79: 351-379, 2010. 26. Fukagawa T, Nogami M, Yoshikawa M, Ikeno M, Okazaki T, Takami Y, Nakayama T, and Oshimura M. Dicer is essential for formation of the heterochromatin structure in vertebrate cells. Nat Cell Biol 6: 784-791, 2004. 27. Gasch AP, Spellman PT, Kao CM, Carmel-Harel O, Eisen MB, Storz G, Botstein D, and Brown PO. Genomic expression programs in the response of yeast cells to environmental changes. Mol Biol Cell 11: 4241-4257, 2000. 28. Hall SE, Beverly M, Russ C, Nusbaum C, and Sengupta P. A cellular memory of developmental history generates phenotypic diversity in C. elegans. Curr Biol 20: 149-155, 2010. 29. Halsell SR, Ito M, and Maxson R. Differential Expression of Early and Late Embryonic Histone Genes in Adult Tissues of the Sea-Urchin Strongylocentrotus-Purpuratus. Developmental biology 119: 268-274, 1987. 30. Hashikawa N, and Sakurai H. Phosphorylation of the yeast heat shock transcription factor is implicated in gene-specific activation dependent on the architecture of the heat shock element. Mol Cell Biol 24: 3648-3659, 2004. 31. Høj A, Jakobsen, B.K. A short element required for tuning off heat shock transcription factor: evidence that phosphorylation enhances deactivation. Embo J 13: 2617-2624, 1994. 32. Hori K, Ishigaki T, Koyama K, Kaya M, Tsujita J, and Hori S. Adaptive changes in the thermogenesis of rats by cold acclimation and deacclimation. The Japanese journal of physiology 48: 505-508, 1998. 33. Horowitz M. Do Cellular Heat Acclimation Responses Modulate Central Thermoregulatory Activity? News in physiological sciences : an international journal of physiology produced jointly by the International Union of Physiological Sciences and the American Physiological Society 13: 218-225, 1998. 34. Horowitz M. From molecular and cellular to integrative heat defense during exposure to chronic heat. Comparative biochemistry and physiology Part A, Molecular & integrative physiology 131: 475-483, 2002. 35. Horowitz M. Genomics and proteomics of heat acclimation. Frontiers in bioscience 2: 1068- 1080, 2010. 36. Horowitz M. Heat acclimation and cross-tolerance against novel stressors: genomic- physiological linkage. Progress in brain research 162: 373-392, 2007. 37. Horowitz M. Matching the heart to heat-induced circulatory load: heat-acclimatory responses. News in physiological sciences : an international journal of physiology produced jointly by the International Union of Physiological Sciences and the American Physiological Society 18: 215-221, 2003. 110

38. Horowitz M, and Assadi H. Heat acclimation-mediated cross-tolerance in cardioprotection: do HSP70 and HIF-1alpha play a role? Annals of the New York Academy of Sciences 1188: 199-206, 2010. 39. Horowitz M, Eli-Berchoer L, Wapinski I, Friedman N, and Kodesh E. Stress-related genomic responses during the course of heat acclimation and its association with ischemic-reperfusion cross- tolerance. Journal of applied physiology 97: 1496-1507, 2004. 40. Horowitz M, Maloyan A, and Shlaier J. HSP 70 kDa dynamics in animals undergoing heat stress superimposed on heat acclimation. Annals of the New York Academy of Sciences 813: 617-619, 1997. 41. Horowitz M, and Meiri U. Thermoregulatory activity in the rat: effects of hypohydration, hypovolemia and hypertonicity and their interaction with short-term heat acclimation. Comparative biochemistry and physiology A, Comparative physiology 82: 577-582, 1985. 42. Horowitz M, Oron Y, and Atias E. Amylase activity, glycoprotein and electrolyte concentration in rat's submaxillary salivary gland during heat acclimation. Comparative biochemistry and physiology B, Comparative biochemistry 60: 351-354, 1978. 43. Horowitz M, and Robinson SD. Heat shock proteins and the heat shock response during hyperthermia and its modulation by altered physiological conditions. Progress in brain research 162: 433-446, 2007. 44. Horowitz M, Shimoni Y, Parnes S, Gotsman MS, and Hasin Y. Heat acclimation: cardiac performance of isolated rat heart. Journal of applied physiology 60: 9-13, 1986. 45. Hsu JY, Sun ZW, Li XM, Reuben M, Tatchell K, Bishop DK, Grushcow JM, Brame CJ, Caldwell JA, Hunt DF, Lin RL, Smith MM, and Allis CD. Mitotic phosphorylation of histone H3 is governed by IpI1/aurora kinase and Glc7/PP1 phosphatase in budding yeast and nematodes. Cell 102: 279-291, 2000. 46. Jud MC, Czerwinski MJ, Wood MP, Young RA, Gallo CM, Bickel JS, Petty EL, Mason JM, Little BA, Padilla PA, and Schisa JA. Large P body-like RNPs form in C. elegans oocytes in response to arrested ovulation, heat shock, osmotic stress, and anoxia and are regulated by the major sperm protein pathway. Developmental biology 318: 38-51, 2008. 47. Kaspler P, and Horowitz M. Heat acclimation and heat stress have different effects on cholinergic-induced calcium mobilization. American journal of physiology Regulatory, integrative and comparative physiology 280: R1688-1696, 2001. 48. Kaspler P, and Horowitz M. Heat acclimation and heat stress have different effects on cholinergic muscarinic receptors. Annals of the New York Academy of Sciences 813: 620-627, 1997. 49. Kloog Y, Horowitz M, Meiri U, Galron R, and Avron A. Regulation of submaxillary gland muscarinic receptors during heat acclimation. Biochimica et biophysica acta 845: 428-435, 1985. 50. Kodesh E, and Horowitz M. Soleus adaptation to combined exercise and heat acclimation: physiogenomic aspects. Medicine and science in sports and exercise 42: 943-952, 2010. 51. Kodesh E, Nesher N, Simaan A, Hochner B, Beeri R, Gilon D, Stern MD, Gerstenblith G, and Horowitz M. Heat acclimation and exercise training interact when combined in an overriding and trade- off manner: physiologic-genomic linkage. American journal of physiology Regulatory, integrative and comparative physiology 301: R1786-1797, 2011. 52. Kultz D, and Chakravarty D. Maintenance of genomic integrity in mammalian kidney cells exposed to hyperosmotic stress. Comparative biochemistry and physiology Part A, Molecular & integrative physiology 130: 421-428, 2001. 53. Kupatt C, Dessy C, Hinkel R, Raake P, Daneau G, Bouzin C, Boekstegers P, and Feron O. Heat shock protein 90 transfection reduces ischemia-reperfusion-induced myocardial dysfunction via reciprocal endothelial NO synthase serine 1177 phosphorylation and threonine 495 dephosphorylation. Arteriosclerosis, thrombosis, and vascular biology 24: 1435-1441, 2004. 54. Labrador M, and Corces VG. Phosphorylation of histone H3 during transcriptional activation depends on promoter structure. Gene Dev 17: 43-48, 2003. 111

55. Levenson JM, O'Riordan KJ, Brown KD, Trinh MA, Molfese DL, and Sweatt JD. Regulation of histone acetylation during memory formation in the hippocampus. J Biol Chem 279: 40545-40559, 2004. 56. Levenson JM, and Sweatt JD. Epigenetic mechanisms in memory formation. Nat Rev Neurosci 6: 108-118, 2005. 57. Levenson JM, and Sweatt JD. Epigenetic mechanisms: a common theme in vertebrate and invertebrate memory formation. Cell Mol Life Sci 63: 1009-1016, 2006. 58. Levi E, Vivi A, Hasin Y, Tassini M, Navon G, and Horowitz M. Heat acclimation improves cardiac mechanics and metabolic performance during ischemia and reperfusion. Journal of applied physiology 75: 833-839, 1993. 59. Levy E, Hasin Y, Navon G, and Horowitz M. Chronic heat improves mechanical and metabolic response of trained rat heart on ischemia and reperfusion. The American journal of physiology 272: H2085-2094, 1997. 60. Lin JJ, and Hemenway CS. Hsp90 directly modulates the spatial distribution of AF9/MLLT3 and affects target gene expression. J Biol Chem 285: 11966-11973, 2010. 61. Lu A, Ran R, Parmentier-Batteur S, Nee A, and Sharp FR. Geldanamycin induces heat shock proteins in brain and protects against focal cerebral ischemia. Journal of neurochemistry 81: 355-364, 2002. 62. Lyons TP, Muza SR, Rock PB, and Cymerman A. The effect of altitude pre-acclimatization on acute mountain sickness during reexposure. Aviation, space, and environmental medicine 66: 957-962, 1995. 63. Mahadevan LC, Willis AC, and Barratt MJ. Rapid Histone H3 Phosphorylation in Response to Growth-Factors, Phorbol Esters, Okadaic Acid, and Protein-Synthesis Inhibitors. Cell 65: 775-783, 1991. 64. Maison C, Bailly D, Peters AHFM, Quivy JP, Roche D, Taddei A, Lachner M, Jenuwein T, and Almouzni G. Higher-order structure in pericentric heterochromatin involves a distinct pattern of histone modification and an RNA component. Nat Genet 30: 329-334, 2002. 65. Maloyan A, Eli-Berchoer L, Semenza GL, Gerstenblith G, Stern MD, and Horowitz M. HIF- 1alpha-targeted pathways are activated by heat acclimation and contribute to acclimation-ischemic cross-tolerance in the heart. Physiological genomics 23: 79-88, 2005. 66. Maloyan A, and Horowitz M. beta-Adrenergic signaling and thyroid hormones affect HSP72 expression during heat acclimation. Journal of applied physiology 93: 107-115, 2002. 67. Maloyan A, Palmon A, and Horowitz M. Heat acclimation increases the basal HSP72 level and alters its production dynamics during heat stress. The American journal of physiology 276: R1506-1515, 1999. 68. Marini F, Pellicioli A, Paciotti V, Lucchini G, Plevani P, Stern DF, and Foiani M. A role for DNA primase in coupling DNA replication to DNA damage response. Embo J 16: 639-650, 1997. 69. Marino-Ramirez L, Hsu B, Baxevanis AD, and Landsman D. The histone database: A comprehensive resource for histones and histone fold-containing proteins. Proteins 62: 838-842, 2006. 70. Mendell JT. MicroRNAs - Critical regulators of development, cellular physiology and malignancy. Cell Cycle 4: 1179-1184, 2005. 71. Mendell JT. MicroRNAs: critical regulators of development, cellular physiology and malignancy. Cell Cycle 4: 1179-1184, 2005. 72. Mendell JT, and Olson EN. MicroRNAs in Stress Signaling and Human Disease. Cell 148: 1172- 1187, 2012. 73. Mestril R, Giordano FJ, Conde AG, and Dillmann WH. Adenovirus-mediated gene transfer of a heat shock protein 70 (hsp 70i) protects against simulated ischemia. Journal of molecular and cellular cardiology 28: 2351-2358, 1996.

112

74. Mirit E, Gross C, Hasin Y, Palmon A, and Horowitz M. Changes in cardiac mechanics with heat acclimation: adrenergic signaling and SR-Ca regulatory proteins. American journal of physiology Regulatory, integrative and comparative physiology 279: R77-85, 2000. 75. Molinier J, Ries G, Zipfel C, and Hohn B. Transgeneration memory of stress in plants. Nature 442: 1046-1049, 2006. 76. Morimoto RI. Cells in Stress - Transcriptional Activation of Heat-Shock Genes. Science 259: 1409- 1410, 1993. 77. Morimoto RI. Regulation of the heat shock transcriptional response: cross talk between a family of heat shock factors, molecular chaperones, and negative regulators. Gene Dev 12: 3788-3796, 1998. 78. Murga M, Jaco I, Fan Y, Soria R, Martinez-Pastor B, Cuadrado M, Yang SM, Blasco MA, Skoultchi AI, and Fernandez-Capetillo O. Global chromatin compaction limits the strength of the DNA damage response. Journal of Cell Biology 178: 1101-1108, 2007. 79. Ng PKS, Chiu SK, Kwong TFN, Yu RMK, Wong MML, and Kong RYC. Functional characterization of two CITED3 homologs (gcCITED3a and gcCITED3b) in the hypoxia-tolerant grass carp, Ctenopharyngodon idellus. Bmc Mol Biol 10: 2009. 80. Noonan EJ, Place RF, Pookot D, Basak S, Whitson JM, Hirata H, Giardina C, and Dahiya R. miR- 449a targets HDAC-1 and induces growth arrest in prostate cancer. Oncogene 28: 1714-1724, 2009. 81. Nowak SJ, and Corces VG. Phosphorylation of histone H3 correlates with transcriptionally active loci. Genes Dev 14: 3003-3013, 2000. 82. Pandolf K. Time course of heat acclimation and its decay. Int J Sporta Med 19: s157-s160, 1998. 83. Pandolf KB, Burce, RL., Goldman, RF.,. Role of physical fitness in heat acclimatisation, decay and reinduction. Ergonomics 20: 399-408, 1977. 84. Paroo Z, Ye XC, Chen S, and Liu QH. Phosphorylation of the Human MicroRNA-Generating Complex Mediates MAPK/Erk Signaling. Cell 139: 112-122, 2009. 85. Pearl LH, Prodromou C, and Workman P. The Hsp90 molecular chaperone: an open and shut case for treatment. Biochem J 410: 439-453, 2008. 86. Petersen CP, Bordeleau ME, Pelletier J, and Sharp PA. Short RNAs repress translation after initiation in mammalian cells. Mol Cell 21: 533-542, 2006. 87. Prigent C, and Dimitrov S. Phosphorylation of serine 10 in histone H3, what for? Journal of cell science 116: 3677-3685, 2003. 88. Queitsh C, Sangster, T.A., Lindquist, S., . Hsp90 as a capacitor of phenotypic variation. Nature 417: 618-624, 2002. 89. Radom-Aizik S, Zaldivar F, Jr., Oliver S, Galassetti P, and Cooper DM. Evidence for microRNA involvement in exercise-associated neutrophil gene expression changes. Journal of applied physiology 109: 252-261, 2010. 90. Reul JMHM, and Chandramohan Y. Epigenetic mechanisms in stress-related learning and memory. Eur Neuropsychopharm 17: S1-S2, 2007. 91. Reul JMHM, and Chandramohan Y. Epigenetic mechanisms in stress-related memory formation. Psychoneuroendocrino 32: S21-S25, 2007. 92. Ritossa FM. A new puffing pattern induced by a temperature shock and DNP in Drosophila. Experentia 18: 571-573, 1962. 93. Roth SY, Denu JM, and Allis CD. Histone acetyltransferases. Annual review of biochemistry 70: 81-120, 2001. 94. Safdar A, Abadi A, Akhtar M, Hettinga BP, and Tarnopolsky MA. miRNA in the regulation of skeletal muscle adaptation to acute endurance exercise in C57Bl/6J male mice. Plos One 4: e5610, 2009. 95. Sancho M, Diani E, Beato M, and Jordan A. Depletion of human histone H1 variants uncovers specific roles in gene expression and cell growth. Plos Genet 4: e1000227, 2008. 113

96. Satake A, and Iwasa Y. A stochastic model of chromatin modification: Cell population coding of winter memory in plants. J Theor Biol 302: 6-17, 2012. 97. Schiltz RL, Mizzen CA, Vassilev A, Cook RG, Allis CD, and Nakatani Y. Overlapping but distinct patterns of histone acetylation by the human coactivators p300 and PCAF within nucleosomal substrates. J Biol Chem 274: 1189-1192, 1999. 98. Schwimmer H, Eli-Berchoer L, and Horowitz M. Acclimatory-phase specificity of gene expression during the course of heat acclimation and superimposed hypohydration in the rat hypothalamus. Journal of applied physiology 100: 1992-2003, 2006. 99. Sera T, and Wolffe AP. Role of histone H1 as an architectural determinant of chromatin structure and as a specific repressor of transcription on Xenopus oocyte 5S rRNA genes. Mol Cell Biol 18: 3668-3680, 1998. 100. Shamir R, Maron-Katz A, Tanay A, Linhart C, Steinfeld I, Sharan R, Shiloh Y, and Elkon R. EXPANDER - An integrative program suite for microarray data analysis. Bmc Bioinformatics 6: 2005. 101. Shaul YD, Gibor G, Plotnikov A, and Seger R. Specific phosphorylation and activation of ERK1c by MEK1b: a unique route in the ERK cascade. Genes Dev 23: 1779-1790, 2009. 102. Shein NA, Doron H, Horowitz M, Trembovler V, Alexandrovich AG, and Shohami E. Altered cytokine expression and sustained hypothermia following traumatic brain injury in heat acclimated mice. Brain research 1185: 313-320, 2007. 103. Shein NA, Horowitz M, and Shohami E. Heat acclimation: a unique model of physiologically mediated global preconditioning against traumatic brain injury. Progress in brain research 161: 353-363, 2007. 104. Shimada M, Nakadai T, Fukuda A, and Hisatake K. cAMP-response element-binding protein (CREB) controls MSK1-mediated phosphorylation of histone H3 at the c-fos promoter in vitro. J Biol Chem 285: 9390-9401, 2010. 105. Shohami E, Novikov M, and Horowitz M. Long term exposure to heat protects against brain damage induced by closed head injury in the rat. Restorative neurology and neuroscience 6: 107-112, 1994. 106. Silverman HS, Wei S, Haigney MC, Ocampo CJ, Stern MD. Myocyte adaptation to chronic hypoxia and development of tolerance to subsequent acute severe hypoxia. Circ Res 80: 699-707, 1997. 107. Soloaga A, Thomson S, Wiggin GR, Rampersaud N, Dyson MH, Hazzalin CA, Mahadevan LC, and Arthur JSC. MSK2 and MSK1 mediate the mitogen- and stress-induced phosphorylation of histone H3 and HMG-14. Embo J 22: 2788-2797, 2003. 108. Solomon MJ, Larsen PL, and Varshavsky A. Mapping protein-DNA interactions in vivo with formaldehyde: evidence that histone H4 is retained on a highly transcribed gene. Cell 53: 937-947, 1988. 109. Sorger PK, Lewis MJ, and Pelham HRB. Heat-Shock Factor Is Regulated Differently in Yeast and Hela-Cells. Nature 329: 81-84, 1987. 110. Squatrito M, Gorrini C, and Amati B. Tip60 in DNA damage response and growth control: many tricks in one HAT. Trends in cell biology 16: 433-442, 2006. 111. Su C, Gao G, Schneider S, Helt C, Weiss C, O'Reilly MA, Bohmann D, and Zhao J. DNA damage induces downregulation of histone gene expression through the G1 checkpoint pathway. Embo J 23: 1133-1143, 2004. 112. Sung SB, and Amasino RM. Vernalization and epigenetics: how plants remember winter. Curr Opin Plant Biol 7: 4-10, 2004. 113. Terasawa K, Ichimura A, Sato F, Shimizu K, and Tsujimoto G. Sustained activation of ERK1/2 by NGF induces microRNA-221 and 222 in PC12 cells. Febs J 276: 3269-3276, 2009.

114

114. Tetievsky A, Cohen O, Eli-Berchoer L, Gerstenblith G, Stern MD, Wapinski I, Friedman N, and Horowitz M. Physiological and molecular evidence of heat acclimation memory: a lesson from thermal responses and ischemic cross-tolerance in the heart. Physiological genomics 34: 78-87, 2008. 115. Tetievsky A, and Horowitz M. Posttranslational modifications in histones underlie heat acclimation-mediated cytoprotective memory. Journal of applied physiology 109: 1552-1561, 2010. 116. Thomson S, Hollis A, Hazzalin CA, and Mahadevan LC. Distinct stimulus-specific histone modifications at Hsp70 chromatin targeted by the transcription factor heat shock factor-1. Mol Cell 15: 585-594, 2004. 117. Treinin M, Shliar J, Jiang H, Powell-Coffman JA, Bromberg Z, and Horowitz M. HIF-1 is required for heat acclimation in the nematode Caenorhabditis elegans. Physiological genomics 14: 17-24, 2003. 118. Turner BM. Defining an epigenetic code. Nat Cell Biol 9: 2-6, 2007. 119. Turner BM. Histone acetylation and an epigenetic code. Bioessays 22: 836-845, 2000. 120. Turner BM. Introduction: Chromatin - a target for intracellular signalling pathways. Semin Cell Dev Biol 10: 165-167, 1999. 121. Uffenbeck SR, and Krebs JE. The role of chromatin structure in regulating stress-induced transcription in Saccharomyces cerevisiae. Biochem Cell Biol 84: 477-489, 2006. 122. Umlauf D, Goto Y, and Feil R. Site-specific analysis of histone methylation and acetylation. Methods in molecular biology 287: 99-120, 2004. 123. Umschwief G, Shein NA, Alexandrovich AG, Trembovler V, Horowitz M, and Shohami E. Heat acclimation provides sustained improvement in functional recovery and attenuates apoptosis after traumatic brain injury. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism 30: 616-627, 2010. 124. Varambally S, Cao Q, Mani RS, Shankar S, Wang X, Ateeq B, Laxman B, Cao X, Jing X, Ramnarayanan K, Brenner JC, Yu J, Kim JH, Han B, Tan P, Kumar-Sinha C, Lonigro RJ, Palanisamy N, Maher CA, and Chinnaiyan AM. Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science 322: 1695-1699, 2008. 125. Vilaboa NE, Calle C, Perez C, Deblas E, Garciabermejo L, and Aller P. Camp Increasing Agents Prevent the Stimulation of Heat-Shock Protein-70 (Hsp70) Gene-Expression by Cadmium Chloride in Human Myeloid Cell-Lines. Journal of cell science 108: 2877-2883, 1995. 126. Windham CH, Rogers, G.G., Senay, L.C., Mitchell, D. Acclimatization in a hot, humid, environment: cardiovascular adjustments. Journal of applied physiology 40: 779-785, 1976. 127. Xi L, Chelliah J, Nayeem MA, Levasseur JE, Hess ML, and Kukreja RC. Whole body heat shock fails to protect mouse heart against ischemia/reperfusion injury: role of 72 kDa heat shock protein and antioxidant enzymes. Journal of molecular and cellular cardiology 30: 2213-2227, 1998. 128. Yahata T, Shao WL, Endoh H, Hur JY, Coser KR, Sun HP, Ueda Y, Kato S, Isselbacher KJ, Brown M, and Shioda T. Selective coactivation of estrogen-dependent transcription by CITED1 CBP/p300- binding protein. Gene Dev 15: 2598-2612, 2001. 129. Yang XJ. Lysine acetylation and the bromodomain: a new partnership for signaling. Bioessays 26: 1076-1087, 2004. 130. Zatsepina OG, Ulmasov KA, Beresten SF, Molodtsov VB, Rybtsov SA, and Evgen'ev MB. Thermotolerant desert lizards characteristically differ in terms of heat-shock system regulation. J Exp Biol 203: 1017-1025, 2000. 131. Zhang T, Luo Y, Wang T, and Yang JY. MicroRNA-297b-5p/3p target Mllt3/Af9 to suppress lymphoma cell proliferation, migration and invasion in vitro and tumor growth in nude mice. Leukemia & lymphoma 2012.

115

132. Zhao J, Herrera-Diaz J, and Gross DS. Domain-wide displacement of histones by activated heat shock factor occurs independently of Swi/Snf and is not correlated with RNA polymerase II density. Mol Cell Biol 25: 8985-8999, 2005. 133. Zhao S, and Liu MF. Mechanisms of microRNA-mediated gene regulation. Science in China Series C, Life sciences / Chinese Academy of Sciences 52: 1111-1116, 2009.

116

תקציר

המוטיבציה של העבודה

מחקרים רבים עסקו בנסיון לחקור את השינויים הפיזיולוגיים והמולקולאריים המתרחשים במהלך האקלום לחום והמאפשרים את הקניית העמידות מפני העקות מייד בתום תקופת האקלום. אך מתי מעט עסקו בשאלות כגון "מהו משך השמירה על הפנוטיפ המאוקלם, בתום תהליך האקלום?", ו-"איזה פנוטיפ פיזיולוגי ומולקולארי נראה כתוצאה מאקלום חוזר (Reacclimation), במידה וקצר יותר מאשר תקופת האקלום הראשונית?". העבודה עתידה לעסוק בתופעה שהוגדרה על ידנו כ- "זיכרון של האקלום לחום" ומטרתה לבחון אותה על היבטיה הפיזיולוגיים והמולקולאריים. מחקר זה, העוסק בחשיפת תופעת "הזיכרון של האקלום לחום" במודל יונקים, הינו חלוצי ויוביל בעתיד להבנה מעמיקה יותר של המנגנונים המולקולאריים - תאיים העומדים בבסיס תהליכי האקלום לעקות סביבתיות, אליהן נחשף כל אורגניזם במהלך חייו לאורך זמן.

הקדמה

האקלום לחום

האקלום לחום (Heat Acclimation) הינו תהליך דינמי בו מתפתחת הסתגלות לעקת חום

קיצונית. האקלום מפעיל מנגנונים מולקולאריים ופיזיולוגיים ועקב כך הופעת פגיעות חום

"נדחית" ע"י הרחבת האזור התרמורגולטורי, בין השאר, כתוצאה משינויים ארוכי טווח

בהתנהגות חלבונים, מסלולים מטבוליים והגנה תוך תאית. לחלבונים ממשפחת חלבוני עקת

חום Heat Shock Proteins( HSP‘s( תפקיד מרכזי בהקניית ההגנה לתאים פרוקריוטים

ואוקריוטים מפני חשיפה לטמפרטורות גבוהות ועקות אחרות כגון חשיפה לכמיקלים רעילים,

חוסר חמצן, מחמצנים, אתנול ועוד.

המלווה המולקולארי (HSP70 (chaperon הינו החלבון המרכזי בבואנו לחקור את התגובה

המולקולארית למצב של היפרתרמיה. המחקרים רבים מצאו כי ביטוי ביתר של הגן hsp70

מקנה הגנה מפני עקת חום, עקה מטבולית ואיסכמיה/רפרפוזיה בכל הרקמות שנבדקו.

החלבון HSP90, בדומה לHSP70- מתפקד גם הוא כמלווה מולקולארי. ידוע כי הוא

משתתף בשמירה על מבנה ה- Signal transduction molecules כגון טירוזין קינאז,

הורמונים סטרואידיים, מרכיבי הציטוסקלטון כגון Intermediate filaments, מיקרוטובולי

ופילמנטים של אקטין. נמצא כי HSP90 מעורב במנגנון ההובלה התאי של פקטור השיעתוק 117

HIF-1α לו תפקיד חשוב בהגנה על רקמת הלב מפני עקת היפוקסיה. ביטוי החלבונים

ממשפחת חלבוני עקת החום מבוקר ברמת השיעתוק ע"י פקטור השיעתוק – Heat Shock

HSF1)( Factor-1). בתא שאיננו חשוף לתנאי עקה, HSF1 מצוי בצורה של מונומר

אינרטי בגרעין או בציטופלזמה וקשור למלווים HSP70 ו- HSP90. בחשיפת התא לעקה,

HSF1 משתחרר מהקומפלקס חלבון-פקטור שעתוק, עובר זירחון וטרימריזציה ונודד

לגרעין, שם נקשר לרצף ספציפי הנמצא בפרומוטר של הגנים ממשפחת HSP's והנקרא

(Heat Shock Element (HSE מה שמאפשר את תחילת שיעתוק הגנים.

דהאקלימציה (Deacclimation) ורהאקלימציה (Reacclimation)

דהאקלימציה (Deacclimation) הינו תהליך בו האורגניזם שעבר אקלום לחום חוזר לשהות

בסביבה נורמותרמית. רהאקלימציה (Reacclimation) הינו תהליך בו אותו האורגניזם

נחשף שוב לתנאי האקלום לחום בתום השהות בתנאי נורמותרמיה.

מעט מאד מחקרים עסקו בנושא דעיכת התכונות שנרכשו בתקופת האקלום לחום, או בחקר

האינדוקציה המחודשת של האקלום לחום. המחקרים היחידים שעסקו בנושא בוצעו כבר

בשנות ה- 40 של המאה הקודמת. בין השנים 1443 ו1463- בוצעו 8 מחקרים שניסו לבחון

את נושא הזיכרון של האקלום לחום בבני האדם. .Pandolf et al הגדירו את המצב

המאוקלם כשינויים אדפטיביים בקצב הלב, טמפרטורת הגוף ומאפיינים קרדיווסקולריים

נוספים. לאחר תקופת האקלום של 4 ימים, נמדדה הדעיכה בביטוי של הפנוטיפ המאוקלם

הנרכש. התוצאות הראו כי גם 18 יום לאחר תקופת האקלום אחוזי השוני מהפנוטיפ

המאוקלם )שנצפה מייד בתום תקופת האקלום( היו נמוכים ביותר. חשיפה חוזרת של

קבוצת הניסוי לתנאי אקלום ראשוניים למשך יומיים בלבד, החזירה את הפנוטיפ המאוקלם

במלואו. בנוסף, נמצא כי בבני האדם קיימת רהאקלימציה מהירה לתנאי קור ותנאי

היפוקסיה בגובה רב, לאחר שהות ממושכת בהרים וחזרה לזמן מסוים לגובה פני הים,

עדויות אלה מהוות הוכחה כי תופעת ה"זיכרון" האקלימטורי אינה מוגבלת רק לאקלום

לחום. במעבדתינו השתמשו בתופעת העמידות הצולבת כקריטריון לאפיון פנוטיפ

הדה/רהאקלימציה. בתאי קרדיומיוציטים שבודדו מלב חולדה שהתאקלמה לחום במשך 30

יום נמצאה ירידה משמעותית במספר התאים הנכנסים למצב של Rigor Contracture

כתוצאה מחשיפה לתנאי אנוקסיה, לעומת תאי הביקורת. נמצא שבתום דהקאלימציה במשך

21 יום דעך הפנוטיפ המאוקלם, אך יומיים של רהאקלימציה הספיקו על מנת לשחזרו. 118

אקלום לחום במשך יומיים בלבד )טיפול שהיווה ביקורת לרהאקלימציה במשך יומיים( הציג

פנוטיפ שלא היה שונה מהביקורת. תופעה זהה אופיינה בצמחים אשר נחשפו לעקה ביוטית

ואביוטית מה שהקנה להם עמידות טובה יותר בפני אותה עקה בחשיפה חוזרת.

האם בקרה אפיגנטית אחראית לתופעת "הזיכרון של האקלום לחום?"

קיימת הסכמה על כך ש: 1( תהליכי שיעתוק ובעקבות כך התרגום הם הבסיס לתגובות פיזיולוגיות. 2( היכולת לשנות את הפנוטיפ בצורה הפיכה בזמן האקלום לחום, מהווה אסטרטגית הסתגלות לתנאי סביבה משתנים. הגמישות הפנוטיפית המתקיימת בזמן חייהם של האורגניזמים השונים מתאפשרת, בין השאר, הודות למודיפיקציות אפיגנטיות המתרחשות בתא.

מודיפיקציות אפיגנתיות כמנגנון לבקרת ביטוי גנים

שינויים אפיגנטיים הינם שינויים בביטוי גנים שלא ניתן להסביר ע"י שינוי ברצף ה- DNA.

בדרך כלל מדובר בשינויים מבניים של ה- DNA המשפיעים על בטוי גני. בקצרה, ה-DNA

הגנומי בכל התאים האוקריוטיים ארוז במבנה קומפקטי גמיש הנקרא כרומטין אשר מלופף

מסביב לאוקטמר של זוגות החלבונים ההיסטוניים מסוגים H3 ,H2B ,H2A וH4- ויוצר את

מבנה הנוקלאוזום. היסטונים הינם חלבונים בסיסיים קטנים המורכבים מדומיין גלובולארי

וקצוות NH2 גמישים )"זנבות ההיסטונים"(. זנבות אלה עוברים אינטרקציה עם אנזימים

וקומפלקסים חלבוניים המשרים מודיפיקציות שונות כגון אצטילציה, מתילציה, פוספורילציה

ויוביקוינציה האחראיות על השינויים במבנה הכרומטין. המודיפיקציות של הכרומטין

מערבות תהליכים מולקולאריים וביוכימיים הגורמים להפיכתו לאוכרומטין (Euchromatin)

– כרומטין "פתוח" המאפשר נגישות של פקטורי שיעתוק לאתרים ספציפיים של הגנים, או

מצב של הטרוכרומטין (Heterochromatin) – כרומטין צפוף ו"סגור" ההופך את רצף הגן

ל"מושתק". אצטילציה וזירחון היסטונים מסעיים בד"כ בפתיחת מבנה הכרומטין, מה

שמאפשר קשירה של פקטורי שעתוק לרצף הגן ותחילת שיעתוק.

אצטילציה וזירחון של היסטונים בתגובה לעקות וגירויים סביבתיים 119

ידוע כי Inducible genes מרכיבים קבוצת גנים המאופיינת בהפעלה מהירה אך ברת חלוף

כחלק משרשראות של אותות תוך תאיים המופעלות לעיתים בתגובה לעקה סביבתית או

גירוי. ביטוי מוגבר של הגנים הללו נמצא לעיתים באסוציאציה עם אצטילציה של שיירי הליזין

בהיסטונים H3 ו- H4 וזירחון של שייר סרין 10 בהיסטון H3. אנליזת Chromatin

ChIP) immunoprecipitation) הראתה כי חלה עליה ברמת הזירחון של סרין 10

בהיסטון H3 בפרומוטרים של גנים אנידוסיביליים המופעלים ע"י עקה ומבוקרים ע"י -NF kB ו- MAP-kinase. בנוסף, נמצא כי זנב היסטון H3 יכול לעבור אצטילציה וזירחון ביחד –

פוספו-אצטילציה. לאחרונה הראו כי חומרים מוטוגניים או תנאי עקה גורמים לפוספו-

אצטילציה של היסטון H3 בפרומוטרים של פרוטואונקוגנים c-fos ו- c-jun בתאי יונקים.

נמצא קשר בין פוספואצטילציה של היסטון H3 בפרומטרים של הגנים הנ"ל ושעתוקם

בתאי פיברובלסט של עכברים לאחר חשיפה לעקת חום. הוכח כי פוספואצטילציה של

היסטון H3 מתרחשת כתוצאה מהפעלה של מסלולי הולכת אותות תאיים

NMDA/MAPK/ERK/MSK. מחקר חשוב הקושר את הפעלת השיעתוק של הגן hsp70

כתוצאה מעקת חום ואצטילציה של היסטון 4 בוצע ע"י Thomson S. et al. מחקר זה

מראה לראשונה תוך שימוש בטכניקת Chromatin immunoprecipitation) ChIP) כי

עקת חום משרה אצטילציה של היסטון H4 אך לא של H3 בפרומוטר של hsp70. החוקרים

העלו השערה כי האצטיליציה של H4 )ובעקבות כך, הפעלת השיעתוק( באיזור הפרומוטר

של hsp70 הינה תגובה ספציפית לעקת חום מכיוון שטיפול ברעל ארסן (arsenite) השרה

אצטילציה ופוספורילציה של היסטון H3 אך לא של היסטון H4 בפרומוטר של hsp70.

החוקרים מצאו כי אצטילציה של H4 הינה הכרחית לקשירת פקטור השיעתוק HSF1

לפרומוטר של הגן hsp70 ועל כן לתחילת שעתוקו.

ה"זיכרון" של האקלום לחום במחקר שבוצע על ידי במסגרת עבודת המאסטר השתמשתי במודל חולדה מאוקלמת על מנת לבחון מאפיינים פיזיולוגיים ומולקולאריים בתום 30 ימי דהאקלימציה ויומיים רהאקלימציה (Tetievsky et.al. 20081) .התוצאות הראו כי 30 ימי דהאקלימציה גרמו לאיבוד הפנוטים המאוקלם, המקנה הגנה )המתבטאת באזור אוטם קטן משמעותית בלב החולדה שנחשפה לתנאי איסכמיה/רפרפוזיה, מזה שבקבוצת ביקורת(, אבל יומיים בלבד של רהאקלימציה 120

הספיקו על מנת לשחזר את הפנוטים המאוקלם המקנה הגנה בלב החולדה. בסדרה של ניסויים מולקולאריים, הראתי כי רמת החלבונים hsp70 ו-bcl-xl נשמרת גבוהה גם לאחר 30 ימי דהאקלימציה. קיים מצב של "דיכוטומיה" בין הפנוטיפ הפיזיולוגי למולקולארי לאחר 30 ימי דהאקלימציה אך ניתן לשחזר את ההגנה הפיזיולוגית לאחר חשיפה קצרה חדשה לתנאי אקלום )רהאקלימציה(. התוצאות הביאו אותנו למסקנה שכנראה קיים מנגנון הגנה מולקולארי המאפשר התניה מוקדמת לחזרה מהירה של הפנוטיפ המאוקלם המקנה הגנה למרות איבודו. הנחנו כי מנגנון אפיגנטי )שיכול לשמר מצב של כרומטין פתוח המאפשר שיעתוק( המופעל ע"י האקלום לחום, אחראי על ההתניה המולקולארית לשיחזור ההגנה בחשיפה חדשה וקצרה למרות איבודה בשהיה בתנאי נורמותרמיה.

מטרות המחקר

1. לחקור האם מודיפיקציות אפיגנטיות ברמת ה- DNA מעורבות בהקניית ה"זיכרון" האקלימטורי. 2. להשתמש באנליזת גנום שלם (Global genomic responses array) על מנת לחשוף מנגנונים מולקולאריים ו"צברי גנים" אשר משתתפים ביצירת ה"זיכרון" של האקלום לחום. בהתבסס על אנליזת התגובה הגנומית הגלובלית, לבחון מסלולי התמרה נבחרים המשתתפים ביצירת הזיכרון התאי-מולקולארי של האקלום לחום. 3. על סמך כלל התוצאות לנסות ולהעלות מנגנון אינטגרטיבי-מולקולארי אשר עומד בבסיס ה"זיכרון" של האקלום לחום.

שיטות עבודה עיקריות כל הניסויים בוצעו בחולדה ממין זכר זן Rattus norvegicus. החיות חולקו בצורה רנדומלית ל5- קבוצות ניסוי: 1. חיות ביקורת (C) ששהו בתנאים נורמותרמיים ב-24ºC. 2. חיות שעברו אקלום לחום (heat acclimation) ב-34ºC במשך יומיים (AC2d). 3. חיות שעברו אקלום לחום ב-34ºC במשך 30 יום (AC). 4. חיות מאוקלמות לחום שעברו דה-אקלימציה במשך 30 יום בתנאים נורמותרמיים (DeAC). 5. חיות שעברו רה-אקלימציה במשך יומיים ב-34ºC לאחר חודש בתנאי דה- אקלימציה (ReAC). בעבודה זו נעשה שימוש במגוון טכניקות: קביעת רמות החלבונים הספציפיים בדוגמאות על ידי Western immunoblotting; קביעת רמות mRNA של גני עניין בדוגמאות על ידי -RT

121

PCR ו- qReal Time RT-PCR; אנליזה של ביטוי כלל הגנים Whole genome gene) (expression באמצעות -Affymetrix whole genome array של גנום החולדה; אנליזת אצטילציה וזירחון של היסטונים וקשירת פקטורי שיעתוק לאתרי קשירה בגנים שונים בוצעה באמצעות .Chromatin immunoprecipitation (ChIP) שימוש בתוכנות ביואינפורמטיות לאנליזת התוצאות ופיענוח ניסוי Microarray. מבחנים לבדיקת מובהקות התוצאות וקיום ההבדלים בין הקבוצות השונות והטיפולים השונים נעשו באמצעות תוכנה סטטיסטית SigmaStat 2.03.

ממצאים עיקריים

1. קצב השיעתוק של hsp70 וhsp90- בקבוצות ה- DeAC וה-ReAC , לאחר חשיפה לעקת חום קיצונית של 41ºC היה מהיר מבקבוצת הביקורת וזהה לזה שבקבוצת ה-AC. הממצא מצביע על כך שקצב שיעתוק הגנים הללו בתגובה לעקת חום הינו חלק מה"זיכרון" של האקלום לחום שכן 30 ימי דהאקלימציה לא שינו את פנוטיפ השיעתוק כפי שנצפה לאחר האקלום לחום. רמות מאד גבוהות של mRNA בקבוצות DeAC, מרמזות כי רמות גבוהות של תעתיקי hsp70 אינם מספיקים על מנת להקנות הגנה מבחינה פיזיולוגית, תופעה אשר דווחה בעבר. hsp90 .2 התנהג שונה מhsp70-: רמתו חזרה לרמת הביקורת בקבוצה DeAC. ייתכן והעלמות ההגנה בתום 30 ימי דהאקלימציה נובעת בין השאר מירידה ברמת החלבון HSP90 הידוע כמשחק תפקיד מרכזי במסלולי הגנה תאית תגובה לעקת חום. 3. קיימת עליה משמעותית ברמת האצטילציה של היסטון H4 בפרומוטרים של hsp70 ו- hsp90 בתום האקלום לחום במשך 30 יום - רמה גבוהה שנשמרה גם בקבוצות DeAC ו- ReAC, מה שיכול להצביע על שמירת המבנה של כרומטין פתוח גם בתנאי נורמותרמיה לטווח זמן של לפחות חודש ימים. רמות אצטילציה של H3Lys10 בשני הגנים לא הראו שוני משמעותי בין קבוצות הניסוי השונות. אנליזת רמות הזירחון של היסטון H3Ser10 הראתה עליה משמעותית ברמות הזירחון לאחר יומיים של אקלום לחום וירידה בתום האקלום הארוך טווח במשך 30 יום בשני הגנים. רהאקלימציה של יומיים לא גרמה לעליה חוזרת ברמת הזירחון של H3Ser10 בגן hsp90. בגן hsp70, יומיים של רהאקלימציה גרמו לעליה משמעותית ברמת הזירחון של H3Ser10. 4. נמצאה התאמה משמעותית בין רמות האצטילציה של היסטון H4 וגיוס של אצטילאזה ספציפית להיסטון זה –TIP60 לאזור הפרומוטר של hsp70 וhsp90-, וכמו כן קיימת התאמה בין הזרחון של היסטון H3Ser10 וגיוס MSK1, הקינאז של היסטון H3.

122

5. קיימת עליה בקשירת HSF1 לאזור ה-HSE של hsp70 בתום 30 ימי האקלום לחום. בתום 30 ימי דהאקלימציה ויומיים רהאקלימציה עדיין ניתן לראות רמות גבוהות של קשירת ה- HSF1. ניתן לשער שקשירה קונסטיטוטיבית של HSF1 לפרומוטר של הגן בתום 30 ימי אקלום לא בהכרח תגרום לשעתוק פעיל, אלא תיתן התניה מוקדמת לתחילת השעתוק כתלות ברמת החלבון - באמצעות הפעלה של מנגנון המשוב השלילי בין רמות החלבון וה- mRNA. אנליזת קשירה בין HSF1 לפרומוטר של hsp90 הראתה קשירה מוגברת בתום 30 ימים של האקלום לחום וירידה בקשירה בקבוצות DeAC ו- ReAC. 6. ניתוח תוצאות של ה- Whole genome rat array הראה ש:  651 גנים השתנו בצורה משמעותית לפחות פי 1.5< , באחת מקבוצות הניסוי. הגנים חולקו ל5- צברי גנים ברורים.  קבוצת גנים אשר הראו עליה משמעותית בקבוצת AC2d מתאימים לתגובה תאית לחשיפה לטמפרטורה סביבתית הגבוהה מהנורמה. בין הגנים שעלו נמצאו גנים השייכים למסלולי הגנה נפני עקה וכמו כן גנים שקשורים לתיקון נזק בדנ"א דו גדילי.  4 צברים גדולים של גנים הראו שינוי משמעותי בקבוצות ה- DeAC וה-ReAC. מבין הגנים שהשתנו נמצאה קבוצה של גנים השייכים למסלול p38 MAPK האחראית בין השאר לפוספורילציה של היסטון H3Ser10, ביניהם גם פקטורי שיעתוק CREBו- ATF. 35% מהגנים המשתנים בצורה משמעותית נקשרים לפקטורי שיעתוק אשר, בין השאר מעורבים במסלול p38 MAPK.  גנים שאחראים על שיעתוק היסטונים H1, H2b, H2a ו- H3 נמצאו כמשתנים משמעותית בקבוצות ה-DeAC וה-ReAC. רוב הגנים היו אחראיים על שיעתוק גן של היסטון H1 והראו ירידה בקבוצות ה-DeAC וה-ReAC .  נמצאו טרנסקריפטים של תשעה miRNA שהשתנו בקבוצות ה- DeAC וה-ReAC. שישה מביניהם, מבקרים סך של 48 גנים מבין אלה שנמצאו כמשתנים בצורה משמעותית באחת מקבוצות המחקר. בבדיקת התאמה בין השינוי בטרנסקריפט של miRNA וגני המטרה שלהם נמצא כי ההשתנות אינה אחידה ו-כמעט חצי מה- miRNA משועתקים בצורה הפוכה לגני המטרה שלהם.

דיון

במחקרינו הקודם הגדרנו את תופעת "הזיכרון של האקלום לחום" שקיימת ברמת הגוף השלם )מנגנוני ויסות חום מרכזיים(, האיבר )הלב( ורמת תעתיקי גנים וחלבוני עקות בתא. הפנוטיפ הפיזיולוגי חוזר למצב הביקורת בתום דהאקלימציה של חודש ימים לעומת פנוטיפ 123

מולקולארי שלרוב שומר על הפרופיל המאוקלם, המקנה הגנה, גם בתום תקופת הדהאקלימציה של חודש ימים. יחד עם זאת הראנו כי ניתן לשחזר את הפנוטיפ ההגנה הפיזיולוגי באמצעות חשיפה חוזרת קצרה לתנאי האקלום הראשוניים. התוצאות במחקר הנוכחי, הראו שגם שיעתוק הגנים בתגובה לחשיפה לעקת חום קיצונית הוא חלק מתופעת ה"זיכרון", שכן נמצא כי פרופיל העליה ברמת השיעתוק של hsp70 ו- hsp90 נשמר זהה לקבוצה המאוקלמת (AC) גם לאחר 30 ימים בתנאים נורמותרמיים .(DeAC) מצאנו שעל אף שמירה על רמה גבוהה של שיעתוק hsp70 בזלי בקבוצת DeAC, נצפתה ירידה של hsp90 בקבוצה זו. הסקנו כי ייתכן והעלמות ההגנה בתום 30 ימי דהאקלימציה נובעת בין השאר מירידה ברמת החלבון HSP90 הידוע כמלווה מולקולרי בעל תפקיד מרכזי בתגובה לעקת חום. בנוסף, השמירה על רמות שיעתוק של hsf1 ורמות חלבון מזורחן גבוהות מצביע גם על מעורבות של hsf1 בשמירה על יכולת לשעתק גנים בצורה מהירה ומשמעותית בחשיפה מחודשת וקצרה לתנאי האקלום. התוצאות הראו כי קיים קשר בין השינויים האפיגנטיים בפרומוטורים של הגנים hsp70 ו- hsp90 להתניה מוקדמת לחזרה מהירה של הפנוטים המאוקלם. להערכתינו ובהתבסס על התוצאות, השינויים במודיפיקציות של היסטונים המושרים ע"י האקלום לחום )תחילה זירחון ולאחר מכן אצטילציה של היסטון H4( מאפשרות קשירה קונסטיטוטיבית של פקטורי השיעתוק, ביניהם גם HSF1 כפי שהראנו, ושיעתוק מוגבר ומהיר של הגנים הקשורים בהתמודדות עם עקת חום גם לאחר שהות ארוכה בתנאי נורמותרמיה ואיבוד הפנוטיפ המאוקלם. ניתוח תוצאות ה- Microarry הראה כי קיימת הפעלה של שיעתוק גנים השייכים למסלול העברת האותות של p38 MAPK. מסלול זה מופעל ע"י עקה תאית ובין השאר הוכח כי גם עקת חום יכולה להפעילו. הקינאז הספציפי שמזרחן את ההיסטון H3 בעמדה Ser10 - ה- MSK1 מופעל ע"י מסלול זה בתא. מצאנו התאמה בין רמות הגיוס של MSK1 לאיזור הפרומוטר של hsp70 ו- hsp90 , רמות הMSK1- המזורחן הפעיל ורמות הזירחון של היסטון H3Ser10. הממצא המעניין הינו שהעליה בשלשת המאפיינים הללו בשני הגנים חלה ביומיים של אקלום (AC2d) ויומיים של רהאקלימציה (ReAC) מה שכנראה מצביע על כך שחשיפה קצרה לתנאי האקלום מפעילים את המסלול p38 MAPK אשר בתורו מפעיל את MSK1 המזרחן את היסטון H3 ו"מסמן" את האתר בגן בו, כפי שהוסבר לעיל, מתרחשת אצטילציה, קשירת פקטור השיעתוק ושיעתוק. גן נוסף אשר שייך למסלול p38 MAPK הינו CREB - פקטור שיעתוק מרכזי המשתתף בתגובה לעקות אשר מפעיל גנים רבים. ידוע כי אחד מתקפידיו הינו הפעלת )ע"י זירחון( MSK1. הממצאים המראים התאמה בין פרופיל השיעתוק של גן זה בקבוצות הניסוי לרמות

124

החלבון המזורחן שלו והקשר בינו לבין הMSK1- גורמים להשערה כי ייתכן ו - CREB מעורב בשרשרת העברת האותות התאיים האחראיים להקניית הזיכרון המולקולארי של האקלום לחום. ממצא חדש בעבודה זו היא מציאת תעתיקים של היסטונים המתבטאים בצורה משמעותית בקבוצות הדה/והרה-אקלימציה. השערתינו היא שייתכן ויש צורך בעליה/ירידה ברמת התעתיקים של היסטונים כתוצאה מתהליכי עקה תאית המערבים שברים ב-DNA דו גדילי או להבדיל, קשורים לתפקיד של היסטון H1 )ההיסטון שרוב התעתיקים שנמצאו מקודדים לו( כמבקר המבנה של הכרומטין מבחינת פתיחתו וסגירתו. לסיכום, עבודה זו מדגימה, כי האקלום לחום מפעיל מסלולי סיגנלים תאיים כגון p38 MAPK אשר משרה פוספו-אצטילציה של היסטונים H4 – H3 באיזור הפרומוטר של גנים חשובים להגנה תאית מפני עקת חום כגון hsp70 וhsp90- . המודיפיקציות של ההיסטונים מאפשרים שמירה על מבנה כרומטין פתוח, קשירה קונסטיטוטיבית של פקטורי שיעתוק ושמירה על רמת שיעתוק גבוהה של צברי גנים גם בתנאי נורמותרמיה לאורך זמן. רמת שיעתוק גבוהה זו מאפשרת חזרה מהירה לפנוטיפ ההגנה הפיזיולוגי בחשיפה מחודשת וקצרה במשך יומיים בלבד.

פירסומים בעקבות עבודה זו )מצורפים כ- Supplements(:

3. Tetievsky A., Cohen O., Eli-Berchoer L., Gerstenblith G., Stern MD., Wapinski I., Friedman N., and Horowitz M., Physiological and molecular evidence of heat acclimation memory: a lesson from thermal responses and ischemic cross-tolerance in the heart. Physiol. Genomics (2008) 34; 78 – 87.

4. Tetievsky A., Horowitz M., Posttranslational modifications in histones underlie heat acclimation- mediated cytoprotective memory. J Appl Physiol. (2010) Nov;109(5):1552-61.

125

Supplementary Material

Unraveling Heat Acclimation Memory:

From Epigenetic Mechanisms to the Expressed Phenotype

Thesis submitted for the degree of

―Doctor of Philosophy‖

Anna Tetievsky

Submitted to the Senate of the Hebrew University of Jerusalem

June 2012

126

חוברת נספחים

זיכרון האקלום לחום: מעורבות המנגנונים האפיגנטיים

והתאיים בביטוי הפנוטיפ המאוקלם

חיבור לשם קבלת תואר דוקטור לפילוסופי

מאת

אנה טטייבסקי

הוגש לסנט האוניברסיטה העברית בירושלים

יוני 2012

127

Table of Contents

Table 1: A list of genes which transcriptional expression change >1.5 fold in at least one of the experimental groups in whole rat genome array analysis……………… …129

Table 2: List of genes in Cluster #1………………………………………………..….….152

Table 3: List of genes in Cluster #2…………………………………………………...….153

Table 4: List of genes in Cluster #3…………………………………………………….…163

Table 5: List of genes in Cluster #4…………………………………….……………....…167

Table 6: List of genes in Cluster #5………………………………………………….……178

128

Probe ID mRNA Accession Gene Symbol Fold- Fold- Fold- Fold- Change Change Change Change ReAC 2d vs. C AC vs. DeAC vs. C C vs. C

10791478 NM_017193 Aadat -1.01106 -1.19026 1.54751 1.73753

10876769 NM_178095 Abca1 1.17106 -1.07083 1.00743 1.51154

10711664 NM_013084 Acadsb 1.09383 1.08456 1.73923 1.49516

10733582 NM_130739 Acsl6 -1.39139 -1.80259 1.28405 -1.10398

10811824 NM_019212 Acta1 -1.46541 -1.61468 -1.31595 -1.45644

10863549 NM_012893 Actg2 -1.41683 1.035 -1.59132 -1.30273

10859919 XM_342611 Actr3b 1.62244 1.39684 1.5166 1.90162

10813694 NM_001106420 Adamts12 -1.33634 -1.68798 -1.16024 -1.74044

10707862 ENSRNOT00000055877 Adamts17 -1.04274 -1.13996 -1.39922 -1.60766

10812834 NM_001108544 Adamts6 1.12729 1.35852 -1.59316 1.02056

10908776 NM_001106811 Adamts8 1.0142 1.41995 -1.71714 -1.43254

10834670 ENSRNOT00000036995 Adamtsl2 1.34977 1.5056 -1.48277 -1.20555

10774120 NM_001107239 Adcy1 -1.60566 -1.60973 1.34516 1.05196

10819379 NM_019286 Adh1 -1.06657 1.13195 1.40076 1.61967

10755112 NM_144744 Adipoq -1.66806 1.28164 -3.11413 -1.91155

10850087 NM_024483 Adra1d -1.0882 -1.20663 -1.32543 -1.52077

10774015 NM_001100970 Aebp1 1.23479 1.53696 -1.09762 1.16602

10843838 NM_001107821 Agpat2 -1.29249 -1.17931 1.51459 1.3181

10829209 NM_001106378 Agpat3 -1.49201 -1.54508 -1.38132 -2.07711

10718001 NM_133406 Agpat4 1.08444 1.43454 -1.52903 -1.14268

10811900 NM_134432 Agt -1.44326 -1.37709 -1.80363 -1.56997

10837424 NM_031349 Agtrl1 -1.11493 -1.30063 -1.2004 -1.59087

10765820 XM_001057147 Aim2 1.09984 1.36988 1.2855 1.5218

10750878 NM_031753 Alcam 1.65199 2.00931 -1.30293 1.1614

10894582 XM_235005 Aldh1l2 -1.09916 -1.05989 -1.24673 -1.58832 129

10734228 NM_031972 Aldh3a1 1.05576 -1.00396 -1.06334 1.51148

10876719 NM_012496 Aldob 1.18426 -1.01667 1.81434 1.27071

10745095 NM_012497 Aldoc -1.21511 -1.32102 -1.40792 -1.60287

10744425 NM_031010 Alox15 -1.99704 -2.44689 -3.07561 -3.73673

10833659 NM_031011 Amd1 -1.00366 -1.06356 1.57153 1.34321

10817943 NM_138876 Ampd1 -1.04056 -1.00387 -1.09714 1.86744

10903529 NM_053546 Angpt1 1.38568 1.41459 1.33099 1.72125

10927361 NM_001108211 Ankrd23 -1.16386 -1.10039 1.18922 3.06187

10787275 XR_008568 Ankrd41 -1.20556 -1.26769 -1.40717 -1.64004

10875837 ENSRNOT00000009571 Ankrd6 -1.1214 -1.26153 1.52329 1.42834

10894814 XM_001076955 Anks1b 1.20411 1.59912 1.25085 1.40762

10863777 NM_001044249 Antxr1 -1.11427 -1.07235 -1.67223 -1.36875

10923432 NM_001008527 Aox3 1.59458 1.91959 1.9091 3.48007

10865329 NM_012907 Apobec1 -1.18887 1.0873 1.21576 1.52191

10719530 NM_138828 Apoe 1.31262 1.44588 -1.07369 1.64524

10905188 ENSRNOT00000031568 Apol11a -1.20994 -1.3683 1.51554 1.72943

10899151 NM_012779 Aqp5 -1.07982 -1.06837 -1.41849 -1.57209

10771545 NM_001024742 Ard1b 1.12048 1.06124 1.50264 1.08674

10848293 ENSRNOT00000010994 Arhgap11a -1.45953 -1.46496 -1.14489 -1.61442

10833784 ENSRNOT00000031047 Armc2 1.31486 1.32443 1.42828 1.9134

10834022 NM_001007797 Arrdc3 -1.07252 -1.12605 1.05249 1.52275

10709330 NM_001107541 Art1 -1.03859 1.531 1.18568 1.79525

10929732 NM_001108231 Asb18 1.01204 1.02756 1.16977 1.68341

10764415 NM_001105955 Aspm -1.84381 -1.54451 -1.50765 -2.10866

10770710 NM_012912 Atf3 -1.10506 -1.23093 1.88978 1.47923

10930616 --- ATP synthase F0 subunit 1.43184 1.76692 2.01053 1.05831 8

10719728 NM_012506 Atp1a3 -1.26961 -2.09493 1.26378 -1.14226

130

10936248 NM_053381 Atp1b4 1.61363 2.22741 -1.81321 -1.20495

10831152 NM_212490 Atp6v1g2 -1.12329 -1.0755 -1.48403 -1.65354

10734853 NM_053749 Aurkb -1.61091 -1.52571 -1.59778 -1.7951

10764495 NM_001109492 B3galt2 -1.65952 -1.92485 -1.18142 -1.65564

10797717 NM_001108880 Barx1 -1.1974 -1.20207 -1.54298 -1.37297

10915981 ENSRNOT00000011386 Barx2 -1.34703 -1.40022 -1.55404 -1.38813

10895394 NM_001109286 Bbs10 1.34495 1.33871 1.20146 1.51718

10706884 NM_022400 Bcat2 1.03901 1.14426 1.55448 1.5561

10719863 NM_012782 Bckdha -1.08853 -1.1636 1.51458 1.4048

10754931 NM_053995 Bdh1 -1.50413 -2.09292 1.93362 -1.06241

10939437 NM_001037365 Bex1 1.46145 1.89043 -1.07954 1.50058

10935038 NM_001037554 Bex4 1.62002 2.08622 -1.29364 1.00899

10715475 EF061226 Bloc1s2 1.40979 1.04454 1.62414 1.3371

10857064 NM_001031824 Bmp10 3.40518 4.65774 -28.7587 1.22246

10911711 NM_001108168 Bmp5 1.29588 1.64003 -1.00158 1.00363

10925636 NM_017312 Bok -1.12828 -1.2168 -1.24844 -1.58964

10745850 ENSRNOT00000038829 Brip1 -1.45293 -1.37596 -1.4091 -1.61685

10849737 NM_001106507 Bub1 -1.61575 -1.57208 -1.25665 -1.55569

10838647 U83666 Bub1b -1.53015 -1.43299 -1.24874 -1.57007

10880734 NM_001008515 C1qa 1.04234 1.12937 1.35052 1.65744

10905307 NM_001034932 C1qtnf6 -1.24899 -1.32069 -1.50278 -1.7727

10865442 NM_138900 C1s 1.32447 1.62886 1.8277 2.12536

10931717 NM_016994 C3 1.15248 1.14064 1.07021 1.75407

10828229 NM_001002805 C4-2 1.3521 1.98233 -1.87407 1.40832

10767422 NM_012516 C4bpa -1.1265 -1.14893 -1.37735 -1.54239

10813253 NM_176074 C6 1.30794 2.17588 1.19084 1.71293

10821666 AF309948 C7 1.39847 1.54773 -1.0353 1.12674

10822242 NM_019292 Ca3 -1.48263 1.85451 -5.18513 -2.19488

131

10737047 NM_019174 Ca4 -1.06724 -2.09581 -1.07162 1.08754

10746327 NM_031601 Cacna1g 1.56791 2.01518 -1.41085 1.02059

10741422 NM_153814 Cacna1h 1.34341 1.66269 1.13893 -1.1669

10789819 NM_175595 Cacna2d3 1.20377 1.55685 -1.38174 1.00494

10730651 ENSRNOT00000036202 Calhm3 -1.15689 -1.12452 -1.19694 -1.52374

10838745 XM_230465 Casc5 -1.79651 -1.72997 -1.74316 -1.97441

10745631 BC058138 Ccl5 1.16804 1.43975 1.4105 1.54675

10745670 NM_001004202 Ccl6 -1.08756 1.18571 1.21233 1.52365

10822852 AY623028 Ccna2 -1.86304 -1.86601 -1.73455 -1.97331

10821016 NM_171991 Ccnb1 -1.49072 -1.35434 -1.47416 -1.69566

10717195 NM_001009470 Ccnb2 -1.70791 -1.6166 -1.46647 -2.00627

10741028 NM_001100474 Ccnf -1.51827 -1.23915 -1.31982 -1.50058

10914614 NM_021866 Ccr2 1.23822 1.4016 1.67822 1.57624

10914618 NM_053960 Ccr5 1.10907 1.01873 1.51772 1.76944

10919712 AF090348 Ccrl1 1.63252 2.09117 -1.1219 1.11541

10911818 ENSRNOT00000037665 Cd109 -1.32672 -1.76063 -1.5312 -1.71907

10858626 NM_001107887 Cd163 1.06139 1.06145 1.33783 1.74734

10712997 NM_001106325 Cd248 -1.31431 -1.33649 -1.36373 -1.52062

10923799 NM_013121 Cd28 -1.02481 1.07689 1.28169 1.52603

10916946 NM_001077646 Cd3g -1.07657 1.43912 1.22805 1.70868

10825809 BC078900 Cd53 1.52076 1.52281 1.76622 2.16406

10865993 NM_134327 Cd69 1.18258 1.2595 1.56938 1.33938

10847572 NM_031797 Cd82 1.22462 1.61048 1.13779 1.24515

10879257 NM_171993 Cdc20 -1.62028 -1.42761 -1.3912 -1.61657

10829761 NM_019296 Cdc2a -1.64093 -1.57904 -1.64584 -2.07115

10775231 NM_001108352 Cdc7 -1.0686 -1.36155 -1.53546 -1.49291

10769657 NM_001012028 Cdca1 -1.69081 -1.69073 -1.76446 -1.81249

10858707 NM_001007648 Cdca3 -1.62516 -1.45991 -1.55855 -1.40835

132

10763225 NM_001012748 Cdh20 1.29237 1.34758 1.63247 1.87385

10828827 NM_080782 Cdkn1a -1.32249 -1.71711 1.34539 1.16229

10779638 NM_001106028 Cdkn3 -2.22624 -1.94965 -1.97157 -2.45458

10804396 NM_052809 Cdo1 -1.31076 1.35751 -2.16161 -1.19573

10850066 ENSRNOT00000002929 Cenpb -1.0033 1.06579 1.50017 1.45192

10819145 ENSRNOT00000034565 Cenpe -1.38214 -1.35614 -1.32352 -1.61478

10770649 ENSRNOT00000004525 Cenpf -2.05499 -1.83358 -2.03691 -2.37101

10810703 NM_001024257 Cenpt /// Tsnaxip1 -1.31359 -1.3736 -1.34445 -1.65617

10806221 NM_133295 Ces3 1.24121 1.484 1.59439 1.78765

10808984 NM_144743 Ces6 -1.06437 -1.13871 -1.15035 -1.79508

10768269 NM_130409 Cfh 1.25595 1.47071 1.02993 1.61521

10918600 NM_001108164 Cgnl1 1.61108 1.60404 -1.25228 1.11921

10764069 NM_053560 Chi3l1 1.12623 1.31934 1.13931 1.77818

10864652 NM_001024333 Cidec -1.61819 1.04232 -2.15148 -1.97305

10796921 ENSRNOT00000031949 Cir1 1.09483 1.08331 1.33904 1.59417

10912908 ENSRNOT00000040056 Cish -1.00197 -1.56788 -1.15658 -1.30828

10938729 NM_172055 Cited1 1.27541 1.53151 -1.51107 -1.28468

10701846 NM_053698 Cited2 -1.63295 -1.10489 1.3484 1.6067

10789152 ENSRNOT00000029266 Ckap2 -1.62395 -1.66863 -1.73252 -1.98608

10892265 NM_012529 Ckb -1.5191 -1.62693 -1.37227 -1.75292

10855035 NM_013147 Clcn1 1.28421 1.5264 -1.11222 1.11926

10751896 NM_031699 Cldn1 1.32382 1.91846 -1.23367 -1.03516

10858553 GENSCAN00000056309 Clec4a1 1.00162 1.17834 1.34974 1.82077

10858573 NM_001005899 Clecsf6 -1.02001 1.11283 1.11654 1.53166

10863679 ENSRNOT00000021103 Cml5 /// Cml3 1.13018 1.03714 1.31116 1.58031

10880472 NM_001039011 Cnksr1 1.017 -1.04835 1.42148 1.80053

10908596 ENSRNOT00000029577 Cnn1 -1.08246 1.89866 -1.59624 -1.03148

10738313 NM_032061 Cntnap1 -1.29196 -1.05073 -1.23757 -2.2378

133

10868881 ENSRNOT00000017217 Col15a1 -1.13324 -1.2792 -1.36764 -1.76817

10737532 NM_053304 Col1a1 -1.14782 -1.15229 -2.09449 -2.27742

10853559 NM_053356 Col1a2 -1.13792 -1.11531 -1.66792 -1.81472

10923052 NM_032085 Col3a1 -1.0762 -1.12869 -1.83757 -1.89389

10789594 ENSRNOT00000057379 Col4a1 -1.10922 -1.48206 -1.3396 -1.52372

10834731 NM_134452 Col5a1 -1.16654 -1.26871 -1.52144 -1.78788

10919864 ENSRNOT00000038871 Col6a6 1.08275 -1.00682 -1.53312 -1.55065

10790238 NM_019274 Colq 1.18858 1.25138 1.29897 1.60614

10772490 NM_182473 Corin 1.32964 1.71349 -1.04116 1.08565

10725806 NM_130411 Coro1a 1.15006 1.34553 1.33404 1.56715

10814430 AF202115 Cp 1.41417 1.62951 1.02916 1.5761

10809303 XM_214630 Cpne2 1.05297 1.26708 1.47878 1.68372

10726255 NM_001106306 Cpxm2 -1.48011 -1.64862 1.19396 -1.17684

10861890 NM_001012188 Creb3l2 1.37626 1.69928 -1.21407 1.05776

10765335 NM_001105966 Creg 1.0626 1.2217 1.29449 1.67744

10898022 ENSRNOT00000007091 Csdc2 -1.14222 -1.05667 1.76815 1.71473

10817429 NM_017320 Ctss 1.00499 1.00016 1.18827 1.66887

10726230 NM_054005 Cuzd1 -1.24435 -1.04925 -1.27226 -1.54188

10771649 NM_182952 Cxcl11 -1.29476 -1.32912 1.75462 1.50081

10775731 NM_001017496 Cxcl13 1.46324 4.03113 1.02224 1.26597

10936899 NM_023965 Cybb -1.07772 -1.08182 1.27426 1.56408

10910376 NM_012540 Cyp1a1 -1.08454 1.07971 1.64293 1.20539

10712090 NM_031543 Cyp2e1 1.38931 1.28287 3.59191 6.59373

10894180 NM_173124 Cyp4f5 1.10387 1.23477 1.49913 1.68345

10827231 NM_031327 Cyr61 -1.0899 1.0447 1.72912 1.29155

10777350 ENSRNOT00000035427 Cytl1 1.45928 1.59591 1.19375 2.07006

10706953 NM_012543 Dbp -1.06759 1.36426 1.2778 1.68016

10838284 XM_230359 Dcdc5 -1.29433 1.1409 -1.1719 -1.62786

134

10858559 NM_001005891 Dcir3 1.15041 1.1825 1.38007 2.01735

10906959 NM_030993 Ddn 1.30158 1.61442 1.45125 1.6379

10791358 ENSRNOT00000036512 Ddx60 1.38895 1.2203 1.12755 1.86437

10819946 ENSRNOT00000012667 Depdc1 -1.55523 -1.43859 -1.62457 -1.59486

10723866 NM_001012345 Dgat2 -1.92618 -1.15621 -1.07026 -1.26351

10734701 ENSRNOT00000036752 Dhrs7c 1.20532 1.28532 1.34286 1.51048

10836588 NM_130819 Dhrs9 1.18454 1.7996 -1.27901 1.29912

10796262 NM_001025720 Dhtkd1 1.10798 1.53863 1.28846 1.34729

10785479 ENSRNOT00000057956 Diaph3 -1.73804 -1.50938 -1.67711 -2.04076

10799187 NM_001107360 Dip2c 1.2496 1.0688 1.42916 1.86918

10724967 NM_138519 Dkk3 1.24146 1.65639 -1.08035 1.08989

10782956 ENSRNOT00000040505 Dlgap5 -1.51259 -1.36908 -1.49852 -1.38666

10708618 NM_022282 Dlgh2 1.32412 2.03829 -1.27422 1.01575

10892706 ENSRNOT00000007233 Dnah11 1.59346 1.54064 1.04045 1.31684

10808620 L07315 Dpep1 1.02123 1.13853 1.53627 1.51866

10810778 NM_001011928 Dpep2 -1.70555 -2.07815 -1.52277 -1.90694

10804245 NM_012934 Dpysl3 -1.15701 -1.13071 -1.58407 -1.67466

10895090 NM_001108088 Dspg3 2.59496 3.14581 -1.48078 1.32132

10770728 ENSRNOT00000005576 Dtl -1.4854 -1.40944 -1.31698 -1.65251

10745702 ENSRNOT00000043148 Dusp14 1.17715 1.33931 1.09626 1.52282

10850841 NM_001108598 Dusp15 -1.17099 -1.11017 1.14802 1.54673

10767503 NM_001024767 Dyrk3 1.16358 1.19985 1.53394 1.18698

10895344 NM_001108092 E2f7 -1.41857 -1.31007 -1.37011 -1.63309

10822558 NM_001108547 Ect2 -1.91965 -1.96435 -1.68347 -2.2339

10934186 ENSRNOT00000004505 Eda 1.44445 1.82169 -1.13224 -1.01122

10774638 NM_001012039 Efemp1 1.53862 2.0402 1.1185 1.66988

10744081 NM_001100980 Efnb3 -1.10486 -1.11102 1.55205 1.14953

10821716 NM_001108938 Egflam -1.24027 -1.56228 -1.18823 -1.14807

135

10889923 NM_019371 Egln3 -1.28329 -1.5029 1.03397 -1.12611

10757726 NM_012722 Eln 1.12819 1.29755 -1.64859 -1.65808

10812922 ENSRNOT00000014074 Elovl7 -1.29048 -1.73408 -1.38894 -1.68886

10731531 NM_001007721 Emp2 1.00107 1.14044 1.27362 1.52213

10931222 NM_001007557 Emr1 1.21599 1.32001 1.35659 1.55435

10812689 NM_001003401 Enc1 1.13278 -1.55906 1.09658 1.27615

10735331 NM_012949 Eno3 1.17891 1.57227 1.1065 1.04301

10752630 NM_031564 Epha3 1.12196 1.15875 1.58142 1.28953

10929211 ENSRNOT00000041689 Epha4 2.01736 3.18993 -2.22404 1.14076

10770342 NM_001034090 Ephx1 1.06428 -1.08558 1.42952 1.64675

10784621 NM_022936 Ephx2 -1.12376 -1.0953 2.12066 1.06606

10928636 NM_021687 Erbb4 1.26687 1.55998 1.05252 1.09914

10884339 NM_001108709 Etv1 1.4893 1.5864 -1.04055 1.17225

10816303 NM_001100992 Etv3l -1.17135 -1.21641 -1.2985 -1.60958

10818708 NM_013057 F3 1.15519 1.11037 1.27065 1.55725

10714106 NM_001109163 Fam111a 1.39491 -1.49158 -1.51215 -1.62482

10911156 ENSRNOT00000012048 Fam148b -1.36624 -1.33467 -1.23905 -1.91364

10873314 ENSRNOT00000055904 Fam43b -1.0366 -1.05461 -1.31356 -1.50109

10931930 NM_182822 Fam70a -1.48621 -1.7643 -1.47241 -1.65339

10841800 NM_001107796 Fam83d -1.14387 -1.19338 -1.50304 -1.24735

10845681 NM_138850 Fap -1.49329 -1.84999 -1.40417 -1.49731

10898315 ENSRNOT00000047234 Fbln1 1.57824 1.66303 1.0719 1.34275

10849327 NM_031825 Fbn1 -1.10275 -1.1634 -1.47268 -1.71178

10918342 NM_001108769 Fbxl22 -1.07888 1.18179 -1.19594 -1.62202

10881590 XM_575958 Fbxo44 -1.2882 -1.52843 1.21683 -1.06127

10717779 NM_001106206 Fbxo5 -1.6402 -1.3841 -1.64702 -1.51287

10769825 ENSRNOT00000029179 Fcer1g 1.06605 1.04156 1.18564 1.53954

10756178 NM_001033924 Fcer2a -1.47962 -1.36727 1.1353 -1.68423

136

10784454 NM_019238 Fdft1 1.10922 1.12753 1.4309 1.77607

10839434 NM_022182 Fgf7 1.35431 1.64299 1.27791 1.66628

10780813 NM_012952 Fgf9 1.12472 1.02962 1.58517 1.40187

10937725 NM_031761 Figf 1.39867 1.50026 1.33794 1.69077

10862643 NM_001013210 Fkbp14 1.11555 1.50362 1.18125 1.23065

10831940 NM_001012174 Fkbp5 -1.11978 -1.24786 1.70287 2.50116

10769361 NM_012792 Fmo1 1.07779 -1.0547 1.61065 1.67002

10769370 NM_144737 Fmo2 1.11968 1.16333 1.78912 2.32168

10769385 NM_053433 Fmo3 1.23768 1.28237 1.56493 2.174

10764050 NM_080698 Fmod 1.2246 1.36516 1.09404 1.66551

10928761 NM_019143 Fn1 -1.01879 -1.0314 -1.72591 -1.47581

10775647 XM_214004 Fras1 1.26418 1.66481 -1.46838 -1.17164

10754218 NM_024369 Fstl1 -1.07162 -1.17414 -1.51102 -1.5036

10932181 NM_001025027 Fundc1 1.19365 1.26497 1.25247 1.52658

10813392 NM_001109176 Fyb 1.14409 1.07127 1.36105 1.59693

10770807 NM_001009632 G0s2 1.13936 1.09319 -1.33729 -1.56201

10859142 NM_001044294 Gabarapl1 1.088 1.17399 1.307 1.67932

10940090 ENSRNOT00000040729 Gabra3 1.37675 1.89239 -1.28128 1.05857

10935811 NM_023091 Gabre 1.29417 1.65043 -1.1056 1.57493

10726269 BC085687 GalNAc4S6ST -1.36941 -1.53379 1.76595 1.09784

10751218 NM_017195 Gap43 -1.13978 -1.11253 -1.54597 -1.45842

10827094 ENSRNOT00000048038 Gbp1 -1.11757 1.16485 2.37328 3.52701

10729209 NM_022276 Gcnt1 -1.09464 -1.05546 1.6925 1.82875

10867821 NM_012960 Ggh 1.8587 2.76699 -1.14442 1.43067

10795520 NM_001007626 Ggps1 1.2769 1.21388 1.39573 1.6259

10893557 NM_001109282 Gipc3 -1.2408 -1.25196 -1.30377 -1.5583

10902313 NM_001011987 Glipr1 1.09971 1.31947 1.12951 1.81099

10797786 NM_057188 Gmpr 1.15674 1.08827 1.41317 1.62105

137

10784747 M15527 Gnrh1 -1.03561 1.00479 1.29254 1.57794

10718954 NM_001013894 Gp49b -1.07898 -1.17432 -1.00832 1.51969

10939764 NM_012774 Gpc3 1.34798 1.77261 -1.36765 1.14494

10899187 NM_022215 Gpd1 -1.60521 -1.41506 1.32513 -1.07573

10855449 NM_133298 Gpnmb -1.16008 -1.00957 1.18221 2.40243

10716939 XM_001071417 Gpr126 1.56771 1.73277 -1.01955 1.26603

10934631 NM_001106940 Gpr23 -1.11042 -1.36243 -1.70499 -1.58044

10867948 NM_001106640 Gpr63 1.34931 1.6784 -1.32653 1.34492

10935023 ENSRNOT00000004260 Gprasp2 1.45336 1.58149 -1.54825 1.14073

10733680 NM_022525 Gpx3 1.36559 1.51705 -1.01621 1.34637

10801794 NM_001014011 Gramd3 -1.08194 -1.00417 -1.01395 1.74887

10733723 ENSRNOT00000003279 Gria1 1.29924 1.51653 1.01946 -1.06099

10787021 ENSRNOT00000060766 Grid1 -1.05856 -1.05978 1.55455 1.54019

10835757 NM_001004080 Gsn 1.32056 1.41142 1.30835 1.5639

10926958 NM_031509 Gsta3 1.03071 1.25246 1.22129 1.56493

10784059 NM_134332 Gzmc 1.19476 -1.32228 1.63066 1.37847

10726991 BC099104 H19 3.41188 2.62144 -1.34584 1.09414

10726995 BC099104 H19 /// Mir675 2.66145 2.17789 -1.18506 -1.03773

10795295 ENSRNOT00000041929 H2a -1.71389 -1.72279 -1.49244 -1.87869

10788858 NM_022674 H2afz -1.2539 -1.70136 -1.28928 -1.24973

10720829 NM_053469 Hamp 3.18726 5.10608 -3.91004 1.87262

10900793 NM_053684 Hcn2 -1.22479 -1.44411 1.50832 1.26728

10910473 NM_021658 Hcn4 1.41518 1.57474 1.64998 1.45166

10862765 ENSRNOT00000031023 Herc6 1.1902 1.02468 -1.09746 1.74288

10809328 NM_053523 Herpud1 1.02089 -1.04251 1.41322 1.58983

10795203 NM_001106113 Hist1h1a -2.08794 -2.10662 -2.04418 -2.89531

10795291 NM_001109417 Hist1h1b -2.78249 -2.27452 -2.16767 -2.80418

10798455 ENSRNOT00000049021 Hist1h2ai -1.59618 -1.44161 -1.56125 -1.70086

138

10798507 NM_001109423 Hist1h2ak -1.83421 -1.61503 -1.74835 -2.22851

10798475 NM_001109400 Hist1h2bc -1.3607 -1.1971 -1.3095 -1.77694

10798488 NM_022647 Hist1h2bl -1.47965 -1.35095 -1.57115 -1.78162

10798494 NM_001106114 Hist1h2bn -1.45196 -1.41601 -1.54192 -1.70235

10817527 ENSRNOT00000048582 Hist2h2ab -1.61659 -1.52683 -1.36039 -1.7993

10817539 ENSRNOT00000042005 Hist2h2bb -1.63841 -1.29385 -1.63512 -1.56123

10817537 NM_001107698 Hist2h3c2 -1.51254 -1.38554 -1.52217 -1.48161

10817759 NM_173094 Hmgcs2 1.33626 1.2918 1.41386 1.73502

10806122 NM_012580 Hmox1 -1.04945 -1.16095 1.36492 1.68467

10758457 NM_017233 Hpd 1.05441 1.1965 1.47526 2.27939

10771389 ENSRNOT00000002983 Hpse -1.67622 -1.57892 1.10365 -1.24255

10781378 NM_024364 Hr -1.3378 -1.25059 -1.22038 -1.53911

10815099 ENSRNOT00000015474 Hspa4l -1.11319 -1.00507 1.507 1.27079

10929566 NM_017250 Htr2b 1.03949 1.24654 -1.01073 1.52694

10861005 L20900 Ica1 1.36332 1.66501 -1.4901 -1.04171

10889379 NM_013060 Id2 1.26112 1.51411 1.00268 1.17535

10834014 XM_227820 Ifi44l 1.01833 -1.13915 -1.15554 1.70034

10714903 NM_001007694 Ifit3 1.039 1.04443 1.2137 1.72596

10753784 NM_001107093 Ift57 1.18482 1.18872 1.16384 1.53616

10778390 NM_012588 Igfbp3 2.04794 1.83285 -1.04945 1.13125

10928837 NM_012817 Igfbp5 -1.01133 1.11539 -1.45836 -1.56542

10887647 ENSRNOT00000047600 IgG-2a 1.52401 -1.01044 -1.0523 1.06402

10892509 ENSRNOT00000044905 Igha /// Ighg 1.55045 -1.01374 -1.13497 1.09646

10739471 XR_009485 Igsf7 1.43088 1.40632 1.60712 1.23273

10924593 NM_012590 Inha -2.8083 -3.48451 -1.54926 -3.12236

10816433 ENSRNOT00000040822 Iqgap3 -1.613 -1.57669 -1.5251 -1.95096

10726758 NM_001033691 Irf7 1.12248 1.25835 1.40621 1.88301

10891910 NM_206846 isg12(b) 1.09797 1.17824 1.07835 1.53858

139

10764626 NM_001047085 Ivns1abp 1.15261 -1.21209 1.38843 2.12344

10747202 NM_001008819 Ka28 -1.26949 -1.11577 -1.62334 -1.49302

10921052 NM_001108195 Kbtbd5 -1.16383 1.19783 1.27312 1.71347

10818245 NM_012970 Kcna2 1.18858 1.53619 1.09131 1.20114

10838326 NM_012971 Kcna4 1.19792 1.55802 1.02756 1.45877

10853905 NM_031730 Kcnd2 1.11069 1.05742 1.4835 1.525

10730475 NM_020094 Kcnip2 1.08405 -1.14831 1.50352 1.39266

10722005 NM_031358 Kcnj11 1.09173 1.10673 1.35002 1.53437

10836271 U09243 Kcnj3 1.45564 1.91368 -1.06094 1.09505

10915998 ENSRNOT00000041038 Kcnj5 1.54562 1.86285 1.00406 1.42355

10801520 NM_019314 Kcnn2 1.39076 1.36953 1.1567 1.50542

10715078 ENSRNOT00000022555 Kif11 -1.70169 -1.69363 -1.63662 -1.97126

10838331 ENSRNOT00000006689 Kif18a -1.53603 -1.34287 -1.38056 -1.27435

10929937 AM180765 Kif1a -1.15978 -1.31493 -1.52991 -1.38545

10800891 NM_001108426 Kif20a -1.6306 -1.57717 -1.76427 -1.79513

10725927 NM_001009645 Kif22 -1.90798 -1.94688 -1.5909 -1.96704

10918031 NM_001108155 Kif23 -2.15476 -1.96576 -1.90156 -2.36634

10878969 NM_001085369 Kif2c -1.53197 -1.34409 -1.4807 -1.61929

10934239 ENSRNOT00000003789 Kif4 -2.45006 -2.29588 -1.93904 -2.70146

10857259 NM_053536 Klf15 1.07199 1.03124 1.29875 1.54755

10866098 XM_001073024 Klra17 1.18997 1.31228 1.48124 1.93797

10866140 XM_578407 Klra7 1.42909 1.89834 1.62494 3.0539

10761810 NM_001107140 Kntc1 -1.61343 -1.50579 -1.53943 -1.77923

10747262 NM_199498 Krt1-19 1.15399 1.69839 -1.34557 1.19365

10768642 ENSRNOT00000036947 Lamc2 -1.32595 -1.37987 -1.00227 -1.58671

10835449 NM_001107830 Lamc3 1.22062 1.79951 -1.0604 1.06863

10936925 ENSRNOT00000005042 Lancl3 1.17219 1.50928 -1.23202 -1.07605

10811160 NM_001008893 Ldhd -1.06603 -1.05044 1.54646 1.28475

140

10908521 NM_175762 Ldlr -1.33009 -1.05556 -1.31533 -1.55308

10779673 NM_031832 Lgals3 -1.00802 1.19968 1.26049 1.59572

10823819 NM_201417 Lgr7 2.4546 3.93859 -2.00827 1.37562

10789670 NM_001106095 Lig4 1.47261 1.05062 1.72389 1.66778

10801821 NM_053905 Lmnb1 -1.62988 -1.54093 -1.50495 -1.79599

10764214 NM_001107179 Lmod1 1.21205 1.50645 -1.15899 1.25732

10864367 ENSRNOT00000047453 Lmod3 -1.05484 -1.05461 1.52983 1.56967

10917883 NM_001012125 Loxl1 -1.10275 -1.03764 -1.37829 -1.5154

10781304 NM_001106047 Loxl2 -1.22963 -1.26276 -1.33837 -1.78936

10721052 NM_053541 Lrp3 -1.24069 -1.22487 -1.29286 -1.57972

10913803 NM_001012001 Lrrc2 -1.15313 -1.33253 1.77046 2.16726

10918004 ENSRNOT00000016456 Lrrc49 1.10798 1.24677 1.13224 1.53704

10759553 NM_001034139 Lrrc8e 1.00936 1.53261 1.03029 1.0735

10850170 ENSRNOT00000040802 Lrrn4 1.0404 1.22106 -1.71709 -1.06516

10877130 NM_138863 Ltb4dh 2.01597 2.44953 1.88208 2.00682

10888424 NM_021587 Ltbp1 1.39102 1.50534 -1.03975 1.13573

10866120 NM_001009497 Ly49si1 1.10057 1.13552 1.2675 1.77365

10866109 NM_001009498 Ly49si2 1.08354 1.01924 1.10508 1.92932

10902547 NM_012771 Lyz 1.08965 1.18276 1.40219 1.87985

10934546 NM_001106941 Magee2 -1.31987 -1.40864 -1.54232 -1.53691

10936717 ENSRNOT00000003910 Maoa -1.2548 -1.30039 1.51147 1.58881

10771267 NM_012806 Mapk10 1.25905 1.35536 -1.52021 -1.35342

10906142 NM_021746 Mapk12 1.14186 1.31697 1.20905 1.7436

10829888 NM_138530 Mawbp 1.0586 1.0697 1.64919 2.83781

10802972 NM_001025289 Mbp 1.25353 1.66342 -1.2666 -1.07629

10804956 NM_013099 Mc4r -1.07005 -1.2769 1.08287 1.83014

10909446 NM_023983 Mcam -1.32954 -1.48903 -1.47987 -1.67672

10767338 NM_017287 Mcm6 -1.50298 -1.3041 -1.35377 -1.41222

141

10850255 U15425 membrane- and -1.73065 -1.10018 3.11582 -3.32218 microfilament-associated protein p58 mRNA, complete cds.

10907793 BC104688 Mettl7a 1.23664 1.11765 1.21106 1.67098

10791474 NM_001012049 Mfap3l -1.0175 -1.05734 1.77568 1.61333

10815308 NM_001106430 Mgst2 -1.0119 -1.03697 1.28555 1.56755

10766880 ENSRNOT00000053760 Mir29b-2 -1.00052 1.25872 1.05337 1.64188

10726408 ENSRNOT00000038176 Mki67 -2.54421 -2.54981 -1.96484 -3.65448

10714103 NM_022617 Mpeg1 -1.12928 1.10416 1.09173 1.50867

10714913 NM_001107609 Mphosph1 -1.53006 -1.50672 -1.50833 -1.45112

10798923 BC085813 Mpp7 -1.33458 -1.59772 -1.17419 -1.48118

10765508 NM_017027 Mpz -1.76721 -1.48539 -2.80016 -2.10428

10728918 ENSRNOT00000032937 Ms4a6a 1.00805 1.03025 1.30661 1.68977

10728924 NM_001006975 Ms4a6b -1.10563 1.00174 1.13609 1.51748

10741486 NM_031658 Msln 1.15682 1.57581 -1.78489 -1.00814

10809392 NM_138826 Mt1a 1.39131 1.59288 -1.04086 1.67461

10827989 ENSRNOT00000047663 Mt2A /// RGD1562885 1.06374 1.41388 -1.02408 2.03067

10885851 NM_138907 Mte1 -1.26044 -1.50748 1.32648 1.04117

10857696 NM_001107880 Mtmr14 1.11819 1.18861 1.25709 1.51474

10930606 ENSRNOT00000044320 MtRNA-Ala 1.68467 1.13273 2.36702 2.08962

10930588 K00160 MtRNA-Arg 1.17757 -1.22747 1.80406 1.72776

10930608 ENSRNOT00000048525 MtRNA-Asn 1.32023 1.23905 1.70693 1.57877

10930576 M25877 MtRNA-Asp 1.10122 1.18919 2.15241 2.34605

10930610 ENSRNOT00000045310 MtRNA-Cys -4.29438 1.0545 3.5226 1.89578

10930604 NC_001665 MtRNA-Gln 1.45138 1.24507 2.24408 1.98292

10930620 ENSRNOT00000049806 MtRNA-Glu 1.50068 1.40011 2.46899 2.07494

10930555 ENSRNOT00000041720 MtRNA-Phe 2.21683 1.54162 3.28775 2.82617

10930614 --- MtRNA-Ser 2.04454 1.54832 1.72687 1.82814

142

10930595 K00241 MtRNA-Ser /MtRNA-Leu 1.24475 1.0622 1.73815 1.60871

10930602 X52757 MtRNA-Thr 1.54139 1.39889 3.76807 3.202

10930612 ENSRNOT00000051860 MtRNA-Tyr 1.40077 1.24053 2.16161 1.97692

10753425 NM_173096 Mx1 1.17783 1.08669 1.65768 2.98232

10818347 NM_001014042 Mybphl 2.62307 4.31919 -6.77988 1.30993

10731247 ENSRNOT00000044317 Myh11 -1.12875 1.32059 -1.51937 -1.21011

10783715 NM_017240 Myh7 2.55345 2.60205 2.11316 3.50825

10778247 NM_001106017 Myl7 2.20282 2.81415 -14.6691 1.3677

10841579 NM_001100885 Myl9 -1.32565 -1.09748 -1.83367 -2.03951

10793002 NM_138893 Myo16 1.20914 1.18733 1.99762 1.18494

10836679 ENSRNOT00000000006 Myo3b /// Sp5 1.17561 1.18225 1.18958 1.59889

10765124 NM_030865 Myoc 2.02602 2.18269 -1.03484 1.73518

10801483 NM_001106148 Myot -1.58824 -1.48783 1.03692 1.07986

10862795 NM_001044293 Nap1l5 1.69409 1.7466 -1.23881 1.1324

10930488 ENSRNOT00000018644 Ndc80 -2.05161 -2.06402 -1.90668 -2.80746

10707649 NM_001008558 Ndn 1.03958 1.13836 -1.61911 -1.53509

10818896 ENSRNOT00000012773 Ndst4 1.39911 1.90695 -1.21569 1.14487

10845143 NM_001107731 Neb 2.33324 3.79804 -1.2484 1.35979

10784809 NM_017029 Nef3 -1.21637 -1.24489 -1.58881 -1.60727

10781266 NM_031783 Nefl -1.17292 1.01057 -2.32245 -2.22595

10925683 NM_001108234 Neu4 -1.13472 -1.2478 -1.32872 -1.57539

10890024 NM_001105720 Nfkbia -1.20033 -1.69754 1.0994 1.15995

10780268 NM_001107263 Ngdn -1.56983 -1.38179 -1.15715 -1.75538

10799017 ENSRNOT00000003349 Nid1 -1.08798 -1.32413 -1.31476 -1.56214

10782387 BC088325 Nid2 -1.19506 -1.4083 -1.39733 -1.66399

10726477 NM_001107558 Nkx6-2 -1.11481 -1.19237 -1.49775 -1.61416

10841637 NM_053601 Nnat -1.2088 1.15757 -1.69462 -1.38618

10807174 NM_053516 Nol3 -1.01082 -1.00436 1.1785 1.51397

143

10803486 NM_001107401 Nol4 -1.11961 -1.22348 -1.20099 -1.5428

10896541 NM_030868 Nov -1.01173 1.93953 -1.78342 1.20561

10873899 NM_012612 Nppa 1.88126 2.06945 -6.81293 1.37764

10821991 NM_012868 Npr3 1.97389 2.27012 -1.46906 1.01401

10880267 NM_057133 Nr0b2 -1.01263 1.08027 1.34965 1.6325

10746955 NM_145775 Nr1d1 1.18896 1.83277 1.69537 1.61503

10884430 NM_013150 Nrcam -1.25474 -1.18959 -1.53873 -1.22448

10803692 NM_178096 Nrep -1.22741 -1.41116 -1.5394 -1.74927

10935087 ENSRNOT00000012063 Nrk -1.03933 -1.26223 -1.25584 -1.53533

10888368 NM_021767 Nrxn1 1.19739 1.19809 1.32508 1.57837

10773247 NM_024128 Nsg1 -1.02296 1.15239 -1.54775 -1.06759

10722864 NM_019248 Ntrk3 -1.41739 -1.19281 -1.50357 -1.55615

10902003 NM_053598 Nudt4 -1.08783 -1.01178 -1.1651 -1.50881

10838843 NM_001107762 Nusap1 -1.53726 -1.50728 -1.19108 -1.27117

10883912 NM_001108016 Oact2 1.31559 1.85269 -1.61777 -1.11195

10758777 NM_138913 Oas1 -1.0217 -1.32212 1.1825 1.64783

10762247 NM_144752 Oas1b 1.19858 1.04097 1.07359 1.73017

10758771 NM_001009680 Oas1i 1.0146 1.05343 -1.02613 1.62842

10762254 NM_001009489 Oas1k -1.27127 -1.31939 1.26771 1.90085

10742989 ENSRNOT00000038125 Obscn 1.02909 1.13849 1.32003 1.58988

10765480 NM_001107195 Olfml2b -1.04075 1.0954 -1.58297 -1.49137

10743164 NM_001000023 Olr1463 -1.10454 -1.09566 -1.50512 -1.0572

10752235 NM_001000043 Olr1567 -1.26992 -1.25336 -1.54475 -1.4698

10830820 NM_214460 Olr1718 -1.08227 -1.38113 -1.51213 -1.15132

10797657 NM_031817 Omd 1.22245 1.64747 1.27366 2.43691

10796673 XM_001074167 Otud1 -1.08512 1.18283 1.67686 1.46726

10815652 NM_012800 P2ry1 -1.09728 -1.1211 1.44829 1.51206

10856474 NM_053289 Pap 1.00664 -1.56075 -1.26178 2.06113

144

10884162 NM_177928 Pbef1 1.17387 1.03399 1.39656 1.72082

10781146 NM_001079937 Pbk -1.91119 -1.8138 -1.7575 -2.75784

10781745 NM_001107279 Pcdh17 -1.20786 -1.63849 -1.49937 -1.70597

10842596 NM_198780 Pck1 -2.18819 1.31774 -2.69026 -2.52083

10822444 ENSRNOT00000043859 Pde7a 1.5419 1.86684 -1.46056 1.29936

10751498 NM_001014125 Pdia5 -1.33685 -1.56903 -1.20656 -1.60849

10860900 NM_053551 Pdk4 -1.75036 -2.42368 1.36259 1.19956

10875300 NM_017139 Penk1 -1.1095 1.5615 2.27858 2.75182

10734882 NM_001034125 Per1 -1.33212 -1.57393 1.01614 1.09987

10881861 NM_023978 Per3 -1.05895 1.22976 -1.02379 1.52562

10717056 NM_001034147 Pex7 1.01601 1.03077 1.56749 1.65156

10932917 NM_012621 Pfkfb1 -1.51469 -1.05698 1.18945 -1.05873

10767444 AB040533 Pfkfb2 -1.56188 -1.57996 -1.00116 -1.3384

10895406 NM_017180 Phlda1 1.39636 1.12194 1.24528 1.51278

10874903 NM_001106917 Pi15 -1.34802 -1.26368 -2.74165 -2.12205

10880872 NM_017174 Pla2g5 -1.2845 -1.22053 -1.00034 -1.52564

10926683 NM_001009353 Pla2g7 -1.12888 -1.55571 1.89077 1.42669

10722973 NM_013094 Plin -1.31098 1.25031 -1.78923 -1.6334

10710627 NM_017100 Plk1 -1.76515 -1.53668 -1.49489 -1.81171

10935064 NM_030990 Plp 1.23075 1.51481 -1.17326 1.05082

10778568 NM_199083 Pno1 1.15644 1.09794 1.71551 1.53456

10898203 ENSRNOT00000015767 Pnpla3 -1.29976 -1.24639 1.62922 -1.56301

10762717 NM_001105752 Pop5 -1.00767 -1.10365 1.58295 1.32823

10815369 NM_001108550 Postn -1.24898 -1.79243 -3.05472 -1.24374

10898359 NM_013196 Ppara 1.08485 1.07989 1.35433 1.50257

10804225 BC078834 Ppp2r2b -1.01251 1.57438 -1.18 -1.00217

10708214 NM_001107529 Prc1 -2.39761 -2.37404 -1.98829 -2.99478

10892939 NM_001008768 Prim1 1.52382 -1.23184 -1.34468 -1.43311

145

10840080 NM_001102431 Prnd -1.18761 -1.58753 -1.60432 -1.98519

10748077 ENSRNOT00000043156 Prp2l1 /// Prp2l2 -1.00599 -1.08873 1.7466 1.67131

10899061 NM_012633 Prph1 -1.32147 -1.14673 -1.79677 -1.33437

10746022 NM_001108287 Prr11 -1.58543 -1.22395 -1.24167 -1.45947

10729096 NM_198738 Psat1 -1.38951 -1.31295 -1.41165 -1.79491

10834005 NM_013115 Ptgfr 1.05827 1.71584 1.41927 1.44313

10768138 NM_138507 Ptprc 1.0727 1.30476 1.16396 1.5413

10887746 NM_031600 Ptprn2 -1.25642 -1.32782 -1.53841 -1.4909

10742194 NM_022391 Pttg1 -1.71247 -1.58064 -1.59532 -2.03447

10816805 NM_001106447 Pygo2 -1.05297 -1.02881 -1.56965 -1.16065

10890734 NM_198749 Rab15 1.73467 1.80558 -1.3949 1.03131

10933945 NM_001008370 Rab1b -1.20074 -1.01379 1.58671 1.37005

10708665 NM_001015012 Rab30 -1.15509 -1.1684 1.13351 1.547

10821243 NM_133536 Rab3c -1.42219 -1.28158 -1.75727 -1.86803

10907178 NM_001108112 Racgap1 -2.0013 -1.86728 -1.54782 -2.01733

10838762 NM_001109204 Rad51 -1.47242 -1.39628 -1.41426 -1.54153

10820335 NM_053721 Rasgrf2 -1.16985 -1.50048 -1.17263 -1.20075

10932422 BC087677 Rbm3 -1.52766 -1.21023 -1.20895 -1.12999

10823941 ENSRNOT00000035846 Rbm46 1.14551 1.56625 1.04153 1.13871

10863410 NM_173097 Reg3g 1.48297 -1.95966 -1.86211 -1.16129

10864918 NM_012643 Ret -1.38644 -1.47482 -1.32073 -1.61079

10753861 NM_181625 Retnlg 1.13043 1.10771 1.1717 2.03158

10901306 XM_576205 Rfx4 1.03615 1.43393 1.53038 1.30115

10768357 NM_001047084 Rgs18 1.1634 1.43295 1.41699 1.82125

10768332 BC061969 Rgs2 1.36851 1.57775 -1.02421 1.30695

10885769 ENSRNOT00000035753 Rgs6 1.2208 1.64591 1.0386 -1.19954

10832772 NM_001107622 Rhobtb1 -1.12108 -1.03035 1.51369 1.85305

10783713 ENSRNOT00000053706 rno-mir-208 1.09556 1.70526 1.30529 -1.44076

146

10735866 XM_001059538 rno-mir-212 -1.21499 -1.10309 -1.30934 -1.50947

10735887 ENSRNOT00000053743 rno-mir-22 -1.72987 -1.12636 1.56887 -1.33304

10935829 ENSRNOT00000053634 rno-mir-224 2.01346 2.70914 -1.01 2.34682

10748763 rno-mir-297 rno-mir-297 1.53298 -1.03335 -1.47543 -1.02523

10810583 ENSRNOT00000054525 rno-mir-328 1.09677 1.02042 -1.43497 -1.54638

10928989 ENSRNOT00000053578 rno-mir-375 -1.34914 -1.21457 -1.42798 -1.75857

10854542 ENSRNOT00000054395 rno-mir-490 1.33524 1.62168 -1.26814 1.00577

10824596 NM_053597 Rps27 1.25942 1.09409 1.37789 1.51497

10883903 NM_001025740 Rrm2 -1.9718 -1.95751 -1.96981 -2.68302

10831105 NM_001008832 RT1-CE1 -1.01229 1.34918 1.27305 1.62872

10833944 NM_001008835 RT1-CE12 1.05294 1.30668 1.22314 1.6108

10831099 NM_001033986 RT1-CE5 -1.01257 1.59172 1.36381 1.77705

10828344 NM_001008847 RT1-Da 1.01459 1.48887 1.44166 1.84062

10827789 NM_001008857 RT1-S2 -1.03853 -1.00975 -1.26509 -1.70482

10827809 NM_001008886 RT1-S3 1.11389 1.24832 1.1907 1.72513

10806314 NM_001107415 Sall1 1.551 1.67384 -1.17004 1.2537

10703922 ENSRNOT00000058956 Sbk2 2.02186 2.70393 -1.6329 1.10701

10706059 NM_001044231 Sbsn 1.43553 2.18568 1.07163 1.46773

10730349 NM_139192 Scd1 -2.33828 -1.31643 -2.68295 -2.85475

10715546 NM_031841 Scd2 1.19353 1.63263 -1.4474 1.12923

10909621 NM_001008880 Scn4b -1.34325 -1.30351 1.86072 1.47051

10931678 NM_031115 Sctr -1.58954 -1.09648 2.12645 1.88727

10860481 NM_017310 Sema3a 1.11582 1.29782 -1.5551 -1.3417

10860499 NM_001104633 Sema3d -1.60831 -1.58368 -1.39019 -1.54194

10761047 NM_012620 Serpine1 -1.03978 -1.17247 1.64571 1.95393

10846854 NM_199093 Serping1 1.22609 1.28427 1.37716 1.5983

10830561 NM_001106396 Sesn1 1.00014 1.03542 1.37018 1.53986

10792344 ENSRNOT00000024128 Sfrp1 -1.21392 1.28033 -1.90883 -1.64374

147

10816144 ENSRNOT00000012658 Sfrp2 2.15633 1.81063 -1.47861 1.90374

10826846 NM_001014043 Sgms2 1.07434 -1.05596 1.79283 -1.06245

10848287 NM_013175 Sgne1 -1.14932 -1.0211 -1.57002 -1.32293

10923385 ENSRNOT00000029526 Sgol2 -1.64978 -1.82632 -1.89389 -1.84985

10823538 NM_013028 Shox2 -1.28057 1.05152 -1.68145 -1.59515

10784172 NM_001108379 Ska3 -1.67515 -1.38453 -1.50707 -1.50652

10772405 ENSRNOT00000049108 Slain2 -1.10393 1.10023 1.51302 1.29729

10765746 NM_001105971 Slamf9 1.03619 1.17585 1.23972 1.51568

10790340 NM_031663 Slc18a3 -1.22815 -1.19733 -1.33162 -1.61585

10703063 NM_019230 Slc22a3 -1.10527 1.01383 1.7407 1.10711

10886429 NM_001108051 Slc24a4 1.5163 1.8092 -1.37065 -1.04018

10881140 NM_001013936 Slc25a34 1.00086 -1.06499 1.68052 1.43436

10902912 ENSRNOT00000052288 Slc26a10 -1.34669 -1.90062 -1.37653 -1.40175

10787313 NM_053580 Slc27a1 1.09653 1.05133 1.71249 1.4345

10883162 BC086513 Slc30a3 1.09925 1.62791 -1.163 -1.08577

10742766 NM_139339 Slc36a2 -1.23285 -1.0263 -1.57305 -1.62306

10811379 ENSRNOT00000038580 Slc38a8 -1.1879 -1.2011 -1.33066 -1.53743

10837366 NM_001107743 Slc43a3 1.2567 1.30037 1.20217 1.59424

10756411 NM_001024968 Slc46a3 1.07203 1.19399 1.46258 1.6683

10747813 NM_012651 Slc4a1 -1.21241 -1.60424 -1.07437 -1.63622

10921208 NM_053521 Slc5a7 -1.60253 -1.35885 -1.62644 -1.77255

10783648 NM_053442 Slc7a8 1.62287 1.70066 1.20962 1.53133

10736802 NM_053687 Slfn3 1.14222 1.21 1.19555 1.60261

10910047 NM_001013247 Sln 3.64707 4.91977 -28.7245 1.8776

10851581 NM_053372 Slpi -1.0535 1.31787 1.04261 1.61762

10918186 NM_001109002 Smad6 1.24099 1.5834 -1.01506 1.14951

10939538 XM_229124 Smarca1 1.43338 1.53095 -1.41558 1.00402

10885680 NM_001002835 Smoc1 1.35382 2.05554 -1.58741 -1.01234

148

10840245 NM_030991 Snap25 -1.41199 -1.66599 -2.17735 -1.86006

10862820 NM_019169 Snca 1.56604 2.60629 -1.96418 -1.02182

10790471 NM_031688 Sncg -1.32611 -1.02272 1.90314 -1.60527

10722489 NM_031117 Snrpn -1.55752 -1.46032 1.09627 -1.49704

10798943 NM_130738 Snurf -1.50113 -1.48488 -1.13962 -1.51362

10855549 NM_001013085 Snx10 1.14354 1.57346 1.14165 -1.03089

10745107 NM_001044224 Spag5 -1.74441 -1.47758 -1.93519 -1.73511

10780873 ENSRNOT00000018407 Spata13 -1.1281 -1.228 1.50915 1.27571

10788053 ENSRNOT00000059511 Spcs3 1.23731 1.41027 1.34763 1.70984

10782777 XM_001055133 Spetex-2D /// Spetex-2C 1.66264 1.04504 1.53188 1.16715 /// Spetex-2B

10779253 XM_001054925 Spetex-2F /// Spetex-2C 1.58172 1.07296 1.29708 1.07231 /// Spetex-2G

10782383 XM_001073572 Spetex-2G 1.64866 1.04867 1.26716 1.03704

10913538 NM_001106860 Spink8 -1.20643 -1.08067 -1.16031 -1.57326

10720479 NM_001082549 Spint2 1.2066 1.54489 -1.0216 1.23947

10765850 NM_001011908 Spna1 -1.59921 -1.55676 1.52088 -1.0112

10829965 NM_001108533 Spock2 1.39299 1.99041 -1.59156 1.1187

10896772 NM_017136 Sqle -1.25369 -1.07842 -1.36305 -1.57995

10839320 ENSRNOT00000000186 Sqrdl 1.20824 1.511 1.5507 1.8039

10709157 NM_001013069 Stard10 2.07506 3.40704 -1.4942 1.2416

10781273 NM_031123 Stc1 -1.36625 -1.98034 -1.64907 -1.57489

10872801 NM_017166 Stmn1 -1.50285 -1.86838 -1.44873 -1.4831

10822358 NM_053440 Stmn2 -1.3869 -1.21636 -2.10886 -2.16533

10781197 BC092646 Stmn4 -1.24591 -1.26112 -1.30143 -1.52459

10844687 ENSRNOT00000025904 Stom 1.1028 1.12846 1.54319 1.24694

10933780 NM_012661 Sts 1.02625 1.08143 1.34689 1.53739

10725782 BC072468 Sult1a1 1.08982 -1.04669 1.3941 1.84914

10926098 NM_133547 Sult1c2 1.61704 1.03479 -1.0354 -1.07036

149

10926125 NM_001013177 Sult1c2a 1.52161 -1.01296 1.00007 -1.12063

10877170 NM_001106650 Susd1 1.57361 2.3547 -1.49274 1.07434

10832496 NM_001106381 Susd2 -1.4225 -1.53158 -1.26422 -1.50643

10937103 NM_012664 Syp -1.0303 1.05903 -1.38888 -1.52269

10902232 NM_001033680 Syt1 -1.38655 -1.30729 -1.78287 -1.75263

10708591 NM_001108492 Sytl2 1.93019 2.54914 1.21846 1.71485

10777770 NM_001004424 Tacc3 -1.53399 -1.53256 -1.62699 -1.528

10917034 NM_031549 Tagln -1.1911 1.73604 -1.60454 -1.14966

10762368 NM_001009964 Tbx5 1.33797 1.84375 -1.18634 1.0042

10872978 NM_001015008 Tcea3 1.3025 1.12993 1.45633 1.53372

10702306 NM_001032397 Tcf21 1.20747 1.59543 1.0098 1.03408

10901357 NM_001017458 Tcp11l2 -1.04937 -1.03627 1.23859 1.55827

10793218 NM_001014005 Tcta -1.01509 1.27084 1.53219 1.67867

10817237 ENSRNOT00000028299 Tdrkh 1.20753 1.63703 1.11663 1.20578

10770252 NM_001008293 Tfb2m -1.03043 -1.05569 1.54472 1.40001

10754785 M58040 Tfrc -1.65676 -1.59548 -1.09696 -2.50263

10711364 ENSRNOT00000027139 Tgfb1i1 -1.17161 -1.14051 -1.4358 -1.53291

10723720 NM_012703 Thrsp -1.98962 1.39693 -3.64193 -2.74846

10909407 NM_012673 Thy1 -1.25625 -1.3183 -1.72838 -1.80522

10751202 ENSRNOT00000059623 Tigit -1.25256 -1.30978 -1.5613 -1.43456

10864732 NM_001109393 Timp4 1.01459 -1.67816 1.13021 1.25598

10933345 NM_001097582 Tlr7 1.00946 1.00417 1.35903 1.63751

10823278 NM_001106434 Tm4sf1 1.45672 1.63833 -1.63748 1.03159

10815503 NM_053785 Tm4sf4 1.04997 1.73528 -1.58018 -1.04987

10854482 NM_001009709 Tmem140 /// -1.27247 -1.15521 1.236 1.55536 LOC500077

10745249 NM_001024992 Tmem199 1.04381 1.09682 1.43777 1.52505

10746976 NM_022183 Top2a -2.84946 -2.52318 -2.31255 -4.47198

150

10876324 NM_001024345 Tpm2 1.06913 1.75907 -2.62896 -1.44582

10840910 NM_001107790 Tpx2 -1.87301 -1.66742 -1.74242 -2.09621

10717359 NM_021666 Trdn 1.64008 1.65225 1.15457 1.43235

10762701 ENSRNOT00000001534 Triap1 1.06457 1.09485 1.35602 1.63234

10888852 NM_001013217 Trim54 1.08975 1.16159 1.45493 1.66347

10733049 ENSRNOT00000003339 Trim7 1.0912 1.23144 1.60998 1.3667

10899073 XM_238523 Troap -1.49196 -1.40514 -1.42999 -1.5444

10768323 NM_001107183 Trove2 1.29698 1.53434 1.02891 1.05964

10817898 NM_022589 Tspan2 -1.01488 1.13138 -1.45773 -1.51334

10905284 NM_012808 Tst 1.16945 1.13529 1.49254 1.63659

10911947 NM_001108172 Ttk -1.82414 -1.40219 -1.90104 -1.96307

10800426 NM_012681 Ttr -1.42679 -1.59172 1.10055 -1.03949

10794824 NM_001013886 Tubb2b -1.16051 -1.19535 -1.41392 -1.55393

10808702 NM_139254 Tubb3 -1.25242 -1.26833 -1.41251 -1.54525

10802422 NM_001025675 Tubb6 -1.32719 -1.25981 -1.58493 -1.62177

10776739 NM_017237 Uchl1 -1.21447 -1.11576 -1.71183 -1.70559

10810334 NM_012682 Ucp1 -3.60554 1.48372 -3.86549 -3.7191

10709093 NM_019354 Ucp2 -1.3138 -1.30071 -1.29491 -1.61855

10709083 NM_013167 Ucp3 -1.33963 -1.53502 1.85763 -1.23846

10931457 NM_001008882 Uhrf1 -1.89407 -1.53277 -1.60889 -1.74828

10819469 NM_199407 Unc5c -1.49952 -1.60482 -1.56 -1.56983

10712720 NM_001108513 Unc93b1 1.01889 1.11107 1.23141 1.52875

10751239 NM_001024253 Upk1b 1.25407 1.47715 -1.74665 1.06007

10761101 ENSRNOT00000037639 Upk3b 1.21508 1.98374 -2.23135 1.02965

10858370 ENSRNOT00000056174 Usp18 1.1214 1.18101 1.44671 1.92571

10826249 NM_012889 Vcam1 1.44412 1.52484 -1.08204 1.18325

10859262 NM_001003403 Verge -1.23731 -1.99591 1.14182 1.00748

10717295 NM_001025623 Vnn1 1.46801 1.57032 1.03718 1.56755

151

10909382 NM_147207 Vof16 1.17102 1.29134 -1.50323 -1.09395

10774898 NM_001099461 Vom2r1 /// Vom2r3 /// 1.59215 1.11372 -1.06919 1.00416 Vom2r4 /// LOC689435

10759497 ENSRNOT00000061907 Vom2r-ps89 1.09549 1.5073 1.34233 1.06895

10889213 NM_012686 Vsnl1 1.72745 2.5095 -1.75913 1.02618

10736875 NM_001003706 WDNM1 homolog 1.27662 1.75285 -1.78447 1.18102

10808348 NM_133581 Wfdc1 -1.68986 -1.87937 -1.21163 -2.17954

10882933 NM_017154 Xdh 1.0426 -1.13764 1.20022 1.51821

10935289 NM_057155 Xpnpep2 -1.46571 -1.84588 1.48533 -1.07277

10917116 NM_001013181 Zbtb16 -1.35503 -1.27967 1.82377 1.92774

10939570 NM_001039016 Zdhhc9 1.14812 1.51041 -1.14428 -1.10587

10783753 NM_001098803 Zfhx2 1.03393 -1.10515 -1.45469 -1.74366

10935882 NM_001106343 Zfp275 1.25086 1.6133 1.06365 1.24687

10793495 NM_001012051 Zfp367 -1.54725 -1.29467 -1.49766 -1.72459

10813949 ENSRNOT00000014423 Zfp622 -1.44242 -1.26122 -1.5456 1.17364

Table 2: List of genes in Cluster # 1

Gene symbol Gene full name

Spetex-2F Spetex-2F protein rno-mir-297 rno-mir-297

Alcam

Bex 4

Bmp10

Cacna1g

Ccrl1

Cgnl1

152

Dnah11

Dspg3

Prim1 DNA primase, p49 subunit

Fbln1

Igha Igha

IgG-2a IgG-2a

Spetex-2G Spetex-2G protein

Ggh

Gpr126

Igfbp3

Table 3: List of genes in Cluster # 2

Gene symbol Gene full name

Nnat neuronatin

Erbb4 v-erb-a erythroblastic leukemia viral oncogene homolog 4 (avian)

Fam43b Fam43b

Plin perilipin

Pck1 phosphoenolpyruvate carboxykinase 1 (soluble) (EC:4.1.1.32)

Slc30a3 solute carrier family 30 (zinc transporter), member 3

Myoc myocilin

Herc6 potential ubiquitin ligase 4 rno-mir-328 rno-mir-328

Slc24a4 solute carrier family 24 (sodium/potassium/calcium exchanger), member 4

Gpc3 glypican 3

153

Gprasp2 G protein-coupled receptor associated sorting protein 2

Stmn2 stathmin-like 2

Lamc3 laminin gamma 3

Olr1463 olfactory receptor 1463

Id2 inhibitor of DNA binding 2

Sbsn suprabasin

Nkx6-2 NK6 homeobox 2

Tbx5 T-box 5

Uchl1 ubiquitin carboxyl-terminal esterase L1 (ubiquitin thiolesterase)

Neb nebulin

Aebp1 AE binding protein 1

Bex4 brain expressed gene 4

Creb3l2 cAMP responsive element binding protein 3-like 2

Gabra3 gamma-aminobutyric acid (GABA) A receptor, alpha 3

Msln mesothelin

Bex1 brain expressed gene 1 5

Lmod1 leiomodin 1 (smooth muscle)

Vof16 ischemia related factor vof-16

Agt angiotensinogen (serpin peptidase inhibitor, clade A, member 8)

Vom2r1 vomeronasal 2 receptor, 1

Cfh complement factor H

Antxr1 anthrax toxin receptor 1

Smarca1 SWI/SNF related, matrix associated, actin dependent regulator of

154

rno-mir-224 rno-mir-224

Trove2 TROVE domain family, member 2

Sgol2 shugoshin-like 2 (S. pombe)

Slc7a8 solute carrier family 7 (cationic amino acid transporter, y+

Cxcl13 chemokine (C-X-C motif) ligand 13

Tagln transgelin

Atp6v1g2 ATPase, H+ transporting, lysosomal V1 subunit G2

Syp synaptophysin

Adamts8 ADAM metallopeptidase with thrombospondin type 1 motif, 8

Cacna2d3 calcium channel, voltage-dependent, alpha2/delta subunit 3

Adamts6 ADAM metallopeptidase with thrombospondin type 1 motif, 6

Pde7a phosphodiesterase 7A (EC:3.1.4.17)

Gap43 growth associated protein 43

Kcnj5 potassium inwardly-rectifying channel, subfamily J, member 5

Eno3 enolase 3, beta, muscle (EC:4.2.1.11)

Lancl3 LanC lantibiotic synthetase component C-like 3 (bacterial)

Cdc7 cell division cycle 7 homolog (S. cerevisiae)

Tspan2 tetraspanin 2

Ndn necdin homolog (mouse)

Kcnj3 potassium inwardly-rectifying channel, subfamily J, member 3

Col6a6 Col6a6

Ccrl1 Ccrl1

Nefl neurofilament, light polypeptide

155

Upk1b uroplakin 1B

Lgr7 Lgr7

Tcf21 transcription factor 21

Cdca1 Cdca1

Ca3 Ca3

Smad6 SMAD family member 6

Cdo1 cysteine dioxygenase, type I (EC:1.13.11.20)

Fbn1 fibrillin 1

Rbm46 RNA binding motif protein 46

Pygo2 pygopus 2

Sytl2 synaptotagmin-like 2

Hamp hepcidin antimicrobial peptide

Mpz myelin protein zero

Gabre gamma-aminobutyric acid (GABA) A receptor, epsilon

Sall1 sal-like 1 (Drosophila)

Eda ectodysplasin-A

Cldn1 claudin 1

Dnah11 dynein, axonemal, heavy chain 11

Adamtsl2 ADAMTS-like 2

Thy1 Thy-1 cell surface antigen

Cnn1 calponin 1, basic, smooth muscle

Oas1i 2 '' -5 '' oligoadenylate synthetase 1I

Trdn triadin

156

Rgs2 regulator of G-protein signaling 2

Npr3 natriuretic peptide receptor C/guanylate cyclase C (atrionatriuretic

Cd82 Cd82 molecule rno-mir-212 rno-mir-212

Anks1b ankyrin repeat and sterile alpha motif domain containing 1B

Sln sarcolipin

Oact2 Oact2

Slc36a2 solute carrier family 36 (proton/amino acid symporter), member 2

Tm4sf4 transmembrane 4 L six family member 4

Nov nephroblastoma overexpressed gene

Col5a1 collagen, type V, alpha 1

Cp ceruloplasmin (EC:1.16.3.1)

Dhrs9 dehydrogenase/reductase (SDR family) member 9

Tm4sf1 transmembrane 4 L six family member 1

Ltbp1 latent transforming growth factor beta binding protein 1

Gpr23 Gpr23

Spag5 sperm associated antigen 5

Tpm2 tropomyosin 2

Vcam1 vascular cell adhesion molecule 1

Etv1 ets variant 1

Slc5a7 solute carrier family 5 (choline transporter), member 7

Clcn1 chloride channel 1

Nppa natriuretic peptide precursor A

157

Mapk10 mitogen activated protein kinase 10 (EC:2.7.11.24)

Postn periostin, osteoblast specific factor

Tacc3 transforming, acidic coiled-coil containing protein 3

Tubb6 tubulin, beta 6

Nap1l5 nucleosome assembly protein 1-like 5

Myh11 myosin, heavy chain 11, smooth muscle

Epha4 Eph receptor A4

Sncg synuclein, gamma (breast cancer-specific protein 1)

Igfbp5 insulin-like growth factor binding protein 5

Sgne1 Sgne1

Tubb3 tubulin, beta 3

Igfbp3 insulin-like growth factor binding protein 3

Rab15 RAB15, member RAS oncogene family

Fbxo5 F-box protein 5

Snca synuclein, alpha (non A4 component of amyloid precursor)

Atp1b4 ATPase, (Na+)/K+ transporting, beta 4 polypeptide

Ltb4dh Ltb4dh

Kif1a kinesin family member 1A

Tgfb1i1 transforming growth factor beta 1 induced transcript 1

C4-2 complement component 4, gene 2

Actg2 actin, gamma 2, smooth muscle, enteric

Dspg3 Dspg3

Ndst4 N-deacetylase/N-sulfotransferase (heparan glucosaminyl) 4

158

Barx2 BARX homeobox 2

Barx1 BARX homeobox 1

Vnn1 vanin 1 (EC:3.5.1.-)

Fkbp14 FK506 binding protein 14 (EC:5.2.1.8)

Tubb2b tubulin, beta 2b

Dpysl3 dihydropyrimidinase-like 3

Nef3 Nef3

Ggh gamma-glutamyl hydrolase (EC:3.4.22.12)

Gpr126 G protein-coupled receptor 126

Bmp5 bone morphogenetic protein 5

Thrsp thyroid hormone responsive

Efemp1 EGF-containing fibulin-like extracellular matrix protein 1

Aqp5 aquaporin 5

Fstl1 follistatin-like 1

G0s2 G0/G1switch 2

Fam83d family with sequence similarity 83, member D

Kcnn2 potassium intermediate/small conductance calcium-activated channel, subfamily N, member 2

Susd1 sushi domain containing 1

Apoe apolipoprotein E

H19 H19

Gpx3 glutathione peroxidase 3 (EC:1.11.1.9)

Dlgap5 discs, large (Drosophila) homolog-associated protein 5

Col3a1 collagen, type III, alpha 1

159

Dlgh2 Dlgh2

Olfml2b olfactomedin-like 2B

Sqle squalene epoxidase (EC:1.14.99.7)

Sfrp2 secreted frizzled-related protein 2

Sema3a sema domain, immunoglobulin domain (Ig), short basic domain, secreted, (semaphorin) 3A

Sfrp1 secreted frizzled-related protein 1

Nrcam neuronal cell adhesion molecule

Mbp myelin basic protein

Agpat4 1-acylglycerol-3-phosphate O-acyltransferase 4 (lysophosphatidic

Hist2h3c2 histone cluster 2, H3c2

Mybphl myosin binding protein H-like

Plp Plp

C7 complement component 7

Ppp2r2b protein phosphatase 2 (formerly 2A), regulatory subunit B (PR 52), beta isoform

Prph1 Prph1

Depdc1 DEP domain containing 1

Magee2 melanoma antigen, family E, 2

C6 complement component 6

Dkk3 dickkopf homolog 3 (Xenopus laevis)

Hist2h2bb histone cluster 2, H2bb

Syt1 synaptotagmin I

Vsnl1 visinin-like 1

Kcna4 potassium voltage-gated channel, shaker-related subfamily, member 4

160

Spock2 sparc/osteonectin, cwcv and kazal-like domains proteoglycan 2

Ntrk3 neurotrophic tyrosine kinase, receptor, type 3 (EC:2.7.10.1)

Zfp275 zinc finger protein 275

Kcna2 potassium voltage-gated channel, shaker-related subfamily, member 2

Gpr63 G protein-coupled receptor 63

Shox2 short stature homeobox 2

Sbk2 Sbk2

Myl9 myosin, light chain 9, regulatory

Fbxl22 F-box and leucine-rich repeat protein 22

Adipoq adiponectin, C1Q and collagen domain containing

Tdrkh tudor and KH domain containing

Myl7 myosin, light chain 7, regulatory

Spint2 serine peptidase inhibitor, Kunitz type, 2

Scd2 Scd2

Cacna1g calcium channel, voltage-dependent, T type, alpha 1G subunit

Fn1 fibronectin 1

Scd1 stearoyl-Coenzyme A desaturase 1 (EC:1.14.19.1)

WDNM1 WDNM1

Zdhhc9 zinc finger, DHHC-type containing 9

Lrrc8e leucine rich repeat containing 8 family, member E rno-mir-490 rno-mir-490

Col1a2 collagen, type I, alpha 2

Col1a1 collagen, type I, alpha 1

161

Snap25 synaptosomal-associated protein 25

Nsg1 neuron specific gene family member 1

Cgnl1 cingulin-like 1

Mt1a metallothionein 1a

Lrrn4 leucine rich repeat neuronal 4

Mt2A metallothionein 2A

Corin corin, serine peptidase

C4bpa complement component 4 binding protein, alpha

Ptprn2 protein tyrosine phosphatase, receptor type, N polypeptide 2

Alox15 arachidonate 15-lipoxygenase (EC:1.13.11.31)

Cited1 Cbp/p300-interacting transactivator with Glu/Asp-rich

Ucp1 uncoupling protein 1 (mitochondrial, proton carrier)

Smoc1 SPARC related modular calcium binding 1

Rab3c RAB3C, member RAS oncogene family

Loxl1 lysyl oxidase-like 1

Pi15 peptidase inhibitor 15

Bmp10 bone morphogenetic protein 10

Cidec cell death-inducing DFFA-like effector c

Eln elastin

Upk3b uroplakin 3B

Dcdc5 doublecortin domain containing 5

Ica1 islet cell autoantigen 1

Alcam activated leukocyte cell adhesion molecule

162

Ka28 Ka28

Ttk Ttk protein kinase (EC:2.7.12.1)

Sult1c2 sulfotransferase family, cytosolic, 1C, member 2 (EC:2.8.2.-)

Htr2b 5-hydroxytryptamine (serotonin) receptor 2B

Tigit Tigit

Stard10 StAR-related lipid transfer (START) domain containing 10

Olr1567 olfactory receptor 1567

Hist1h2bl histone cluster 1, H2bl

Krt1-19 Krt1-19

Fras1 Fraser syndrome 1 homolog (human)

Fbln1 fibulin 1

Kif20a kinesin family member 20A

Fgf7 fibroblast growth factor 7

Adamts17 ADAM metallopeptidase with thrombospondin type 1 motif, 17

Table 4: List of genes in Cluster #3

Gene symbol Gene full name

Ret ret proto-oncogene

Nrk Nik related kinase

Bok BCL2-related ovarian killer

Hist1h2bc histone cluster 1, H2bc

Cenpt centromere protein T

Casc5 cancer susceptibility candidate 5

163

Rrm2 ribonucleotide reductase M2 (EC:1.17.4.1)

Vom2r-ps89 vomeronasal 2 receptor, pseudogene 89

Lrp3 low density lipoprotein receptor-related protein 3

Cdc2a Cdc2a

Nol4_predicted Nol4_predicted

Zfp367 zinc finger protein 367

Rad51 RAD51 homolog (RecA homolog, E. coli) (S. cerevisiae)

Snx10 sorting nexin 10

Ccnb2 cyclin B2

Ccnb1 cyclin B1

Hist1h2ai_predicted Hist1h2ai_predicted

Col15a1 collagen, type XV, alpha 1

Aspm_predicted Aspm_predicted

Iqgap3 IQ motif containing GTPase activating protein 3

Aldh1l2_predicted Aldh1l2_predicted

Cacna1h calcium channel, voltage-dependent, T type, alpha 1H subunit

Agtrl1 Agtrl1

Plk1 polo-like kinase 1 (Drosophila) (EC:2.7.11.21)

Nudt4 nudix (nucleoside diphosphate linked moiety X)-type motif 4

Ckap2 cytoskeleton associated protein 2

Cenpf centromere protein F

Diaph3 diaphanous homolog 3 (Drosophila)

Calhm3 calcium homeostasis modulator 3

164

Cenpe centromere protein E

Tpx2_predicted Tpx2_predicted

XTroap_predicted XTroap_predicted

Hist1h2bn_predicted Hist1h2bn_predicted

Hist1h1a_predicted Hist1h1a_predicted

Cntnap1 contactin associated protein 1

Ldlr low density lipoprotein receptor

Ces6 carboxylesterase 6

Dtl denticleless homolog (Drosophila)

Ccnf cyclin F

Hist1h2ak Hist1h2ak

Gipc3 GIPC PDZ domain containing family, member 3

Pla2g5 phospholipase A2, group V (EC:3.1.1.4)

Loxl2_predicted Loxl2_predicted

Ucp2 uncoupling protein 2 (mitochondrial, proton carrier)

Nrep neuronal regeneration related protein

Prnd prion protein 2 (dublet)

Etv3l ets variant 3-like

E2f7_predicted E2f7_predicted rno-mir-208 rno-mir-208

Rgs6 regulator of G-protein signaling 6

Slc18a3 solute carrier family 18 (vesicular acetylcholine), member 3

Mki67 antigen identified by monoclonal antibody Ki-67

165

Neu4_predicted Neu4_predicted

Mcam melanoma cell adhesion molecule

Lmnb1 lamin B1

Ndc80 NDC80 homolog, kinetochore complex component (S. cerevisiae)

Stmn4 stathmin-like 4

Aldoc aldolase C, fructose-bisphosphate (EC:4.1.2.13)

Ect2_predicted Ect2_predicted

Cdkn3_predicted Cdkn3_predicted

Pttg1 pituitary tumor-transforming 1

Kif2c kinesin family member 2C

Spink8 serine peptidase inhibitor, Kazal type 8

RT1-S2 RT1 class Ib, locus H2-TL-like (S2)

Tfrc transferrin receptor

Hr hairless

Kif11 kinesin family member 11

Cuzd1 CUB and zona pellucida-like domains 1

Top2a topoisomerase (DNA) II alpha (EC:5.99.1.3)

Ankrd41_predicted Ankrd41_predicted rno-mir-375 rno-mir-375

C1qtnf6 C1q and tumor necrosis factor related protein 6

Brip1 BRCA1 interacting protein C-terminal helicase 1

Hist2h2ab Hist2h2ab

Cd248_predicted Cd248_predicted

166

Psat1 phosphoserine aminotransferase 1 (EC:2.6.1.52)

Fam148b family with sequence similarity 148, member B

Aurkb aurora kinase B (EC:2.7.11.1)

Kntc1_predicted Kntc1_predicted

Sult1c2a sulfotransferase family, cytosolic, 1C, member 2a (EC:2.8.2.-)

Pcdh17_predicted Pcdh17_predicted

Prc1_predicted Prc1_predicted

Col4a1 collagen, type IV, alpha 1

Nid2 nidogen 2

Kif23_predicted Kif23_predicted

Nid1 nidogen 1

Agpat3_predicted Agpat3_predicted

Pbk_predicted Pbk_predicted

Gria1 glutamate receptor, ionotropic, AMPA 1

Adra1d adrenergic, alpha-1D-, receptor

Slc38a8 solute carrier family 38, member 8

Table 5: List of genes in Cluster #4

Gene Symbol Gene full name

Slc22a3 solute carrier family 22 (extraneuronal monoamine transporter), member 3

Kcnd2 potassium voltage gated channel, Shal-related family, member 2

Spcs3 Spcs3

167

Klf15 Kruppel-like factor 15

Slc46a3 solute carrier family 46, member 3

Aldob aldolase B, fructose-bisphosphate (EC:4.1.2.13)

Cml5 camello-like 5

Adh1 alcohol dehydrogenase 1 (class I) (EC:1.1.1.1)

Mgst2_predicted Mgst2_predicted

Slfn3 schlafen 3

Ifit3 interferon-induced protein with tetratricopeptide repeats 3

Cdkn1a cyclin-dependent kinase inhibitor 1A (p21, Cip1)

Cd28 Cd28 molecule

Tcea3 transcription elongation factor A (SII), 3

Sult1a1 sulfotransferase family, cytosolic, 1A, phenol-preferring, member 1

MtRNA-Asp MtRNA-Asp

Rps27 ribosomal protein S27

MtRNA-Asn MtRNA-Asn

Mc4r melanocortin 4 receptor

Cpne2_predicted Cpne2_predicted

Oas1k 2 '' -5 '' oligoadenylate synthetase 1K

Gnrh1 gonadotropin-releasing hormone 1 (luteinizing-releasing hormone)

RT1-CE12 RT1 class I, CE12

Aim2_predicted Aim2_predicted

Nfkbia nuclear factor of kappa light polypeptide gene enhancer in B-cells

MtRNA-Thr MtRNA-Thr

168

Kif18a kinesin family member 18A

Tfb2m transcription factor B2, mitochondrial

Mir29b-2 Mir29b-2

Stmn1 stathmin 1

Kcnip2 Kv channel-interacting protein 2

Gpnmb glycoprotein (transmembrane) nmb

Mfap3l microfibrillar-associated protein 3-like

Oas1b 2-5 oligoadenylate synthetase 1B

Verge Verge

Enc1 ectodermal-neural cortex 1

Zbtb16 zinc finger and BTB domain containing 16

Hmgcs2 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 2 (mitochondrial)

Kcnj11 potassium inwardly rectifying channel, subfamily J, member 11

Gcnt1 glucosaminyl (N-acetyl) transferase 1, core 2 (EC:2.4.1.102)

Rhobtb1_predicted Rhobtb1_predicted

Efnb3 ephrin B3

GalNAc4S6ST GalNAc4S6ST

Ankrd23_predicted Ankrd23_predicted

C3 complement component 3 (EC:3.4.21.43)

Gbp1 guanylate binding protein 1, interferon-inducible, 67kDa

Kbtbd5_predicted Kbtbd5_predicted

Pbef1 Pbef1

Rbm3 RNA binding motif (RNP1, RRM) protein 3

169

Pdk4 pyruvate dehydrogenase kinase, isozyme 4 (EC:2.7.11.2)

Gmpr guanosine monophosphate reductase (EC:1.6.6.8)

Pla2g7 phospholipase A2, group VII (platelet-activating factor

Cyr61 cysteine-rich, angiogenic inducer, 61

Bcat2 branched chain aminotransferase 2, mitochondrial (EC:2.6.1.42)

Clecsf6 C-type (calcium dependent, carbohydrate recognition domain) lectin, superfamily member 6

Ivns1abp_predicted Ivns1abp_predicted

Mx1 myxovirus (influenza virus) resistance 1

Rab30 RAB30, member RAS oncogene family

Per1 period homolog 1 (Drosophila)

Rgs18 regulator of G-protein signaling 18

Irf7 interferon regulatory factor 7

Gp49b Gp49b

Ampd1 adenosine monophosphate deaminase 1 (isoform M) (EC:3.5.4.6)

MtRNA-Ala MtRNA-Ala

Egflam EGF-like, fibronectin type III and laminin G domains

Pex7 peroxisomal biogenesis factor 7

Sts steroid sulfatase (EC:3.1.6.2)

Cyp2e1 cytochrome P450, family 2, subfamily e, polypeptide 1 (EC:1.14.14.1)

Trim7 tripartite motif-containing 7

Dbp D site of albumin promoter (albumin D-box) binding protein

MtRNA-Arg MtRNA-Arg

Atf3 activating transcription factor 3

170

Cdca3 cell division cycle associated 3

Rab1b RAB1B, member RAS oncogene family

Art1_predicted Art1_predicted

Cdh20 cadherin 20

Gsta3 glutathione S-transferase A3 (EC:2.5.1.18)

Timp4 tissue inhibitor of metalloproteinase 4

Ldhd lactate dehydrogenase D

Pop5_predicted Pop5_predicted

Gzmc granzyme C (EC:3.4.21.-)

Dusp14_predicted Dusp14_predicted

Cyp4f5 cytochrome P450 4F5 (EC:1.14.14.1)

Clec4a1 C-type lectin domain family 4, member a1

Cxcl11 chemokine (C-X-C motif) ligand 11

Otud1 OTU domain containing 1

MtRNA-Cys MtRNA-Cys

Mtmr14 myotubularin related protein 14

Lrrc2 leucine rich repeat containing 2

Ptprc protein tyrosine phosphatase, receptor type, C (EC:3.1.3.48)

Epha3 Eph receptor A3 (EC:2.7.10.1)

RT1-S3 RT1-S3

Trim54 tripartite motif-containing 54

Adcy1_predicted Adcy1_predicted

Lig4_predicted Lig4_predicted

171

Mapk12 mitogen-activated protein kinase 12 (EC:2.7.11.24)

Hcn2 hyperpolarization activated cyclic nucleotide-gated potassium

Asb18_predicted Asb18_predicted

Lrrc49 leucine rich repeat containing 49

Actr3b ARP3 actin-related protein 3 homolog B (yeast)

Cd69 Cd69 molecule

Figf c-fos induced growth factor

Slamf9_predicted Slamf9_predicted

Usp18 ubiquitin specific peptidase 18 (EC:3.1.2.-)

Retnlg resistin-like gamma

Cnksr1 connector enhancer of kinase suppressor of Ras 1

MtRNA-Tyr MtRNA-Tyr

Coro1a coronin, actin binding protein 1A

Xdh xanthine dehydrogenase (EC:1.17.1.4 1.17.3.2)

Slc43a3_predicted Slc43a3_predicted

H2afz H2A histone family, member Z

Ppara peroxisome proliferator activated receptor alpha

Slpi secretory leukocyte peptidase inhibitor

Creg_predicted Creg_predicted

Fkbp5 FK506 binding protein 5 (EC:5.2.1.8)

MtRNA-Ser MtRNA-Ser

Dgat2 diacylglycerol O-acyltransferase homolog 2 (mouse) (EC:2.3.1.20)

Ctss cathepsin S (EC:3.4.22.27)

172

RT1-Da RT1-Da

Myot_predicted Myot_predicted

P2ry1 purinergic receptor P2Y, G-protein coupled 1

Tcp11l2 t-complex 11 (mouse) like 2

Phlda1 pleckstrin homology-like domain, family A, member 1

Hmox1 heme oxygenase (decycling) 1 (EC:1.14.99.3)

Nrxn1 neurexin 1

Cited2 Cbp/p300-interacting transactivator, with Glu/Asp-rich

Tst thiosulfate sulfurtransferase, mitochondrial (EC:2.8.1.1)

Tcta T-cell leukemia translocation altered gene

Ly49si2 immunoreceptor Ly49si2

Ly49si1 immunoreceptor Ly49si1

Prp2l1 Prp2l1

Fgf9 fibroblast growth factor 9

Stom stomatin

Chi3l1 chitinase 3-like 1

Slc25a34 solute carrier family 25, member 34

Dpep1 dipeptidase 1 (renal) (EC:3.4.13.19)

Lgals3 lectin, galactoside-binding, soluble, 3

Lyz Lyz

Cd163_predicted Cd163_predicted

Abca1 ATP-binding cassette, sub-family A (ABC1), member 1

Aadat aminoadipate aminotransferase (EC:2.6.1.7 2.6.1.39)

173

Cd53 Cd53 molecule

Slc27a1 solute carrier family 27 (fatty acid transporter), member 1

Dusp15_predicted Dusp15_predicted

Serping1 serine (or cysteine) peptidase inhibitor, clade G, member 1

Ca4 Ca4

Mphosph1_predicted Mphosph1_predicted

Hpd 4-hydroxyphenylpyruvate dioxygenase (EC:1.13.11.27)

Cir1 Cir1

Nr0b2 nuclear receptor subfamily 0, group B, member 2

Maoa monoamine oxidase A

Dcir3 Dcir3

Agpat2_predicted Agpat2_predicted

Ift57_predicted Ift57_predicted

Cd3g CD3 molecule, gamma polypeptide

Triap1 TP53 regulated inhibitor of apoptosis 1

Fmod fibromodulin

Pfkfb2 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 2 (EC:2.7.1.105

Emr1 EGF-like module containing, mucin-like, hormone receptor-like 1

Herpud1 homocysteine-inducible, endoplasmic reticulum stress-inducible, ubiquitin-like domain member 1

Sqrdl sulfide quinone reductase-like (yeast)

Sctr secretin receptor

Fap fibroblast activation protein, alpha

Mpeg1 macrophage expressed gene 1

174

Ggps1 geranylgeranyl diphosphate synthase 1 (EC:2.5.1.1 2.5.1.10 2.5.1.29)

Unc93b1 unc-93 homolog B1 (C. elegans)

Mawbp Mawbp

MtRNA-Phe MtRNA-Phe

Tmem140 transmembrane protein 140

Ddx60 DEAD (Asp-Glu-Ala-Asp) box polypeptide 60

F3 coagulation factor III (thromboplastin, tissue factor)

C1qa complement component 1, q subcomponent, alpha polypeptide

Bckdha branched chain ketoacid dehydrogenase E1, alpha polypeptide

Grid1 glutamate receptor, ionotropic, delta 1

Csdc2 cold shock domain containing C2, RNA binding

Nusap1_predicted Nusap1_predicted

Sema3d sema domain, immunoglobulin domain (Ig), short basic domain, secreted, (semaphorin) 3D

Ephx2 epoxide hydrolase 2, cytoplasmic (EC:3.3.2.10)

Amd1 adenosylmethionine decarboxylase 1 (EC:4.1.1.50)

Ephx1 epoxide hydrolase 1, microsomal (EC:3.3.2.9)

Ankrd6 ankyrin repeat domain 6

Mte1 Mte1

Cyp1a1 cytochrome P450, family 1, subfamily a, polypeptide 1 (EC:1.14.14.1)

Glipr1 GLI pathogenesis-related 1

Pno1 partner of NOB1 homolog (S. cerevisiae)

Dhrs7c dehydrogenase/reductase (SDR family) member 7C

Cenpb centromere protein B

175

Ccl6 chemokine (C-C motif) ligand 6

Oas1 Oas1

Spata13 spermatogenesis associated 13

Ccl5 chemokine (C-C motif) ligand 5

Myo3b Myo3b

Klra17 Klra17

Penk1 proenkephalin 1

Nol3 nucleolar protein 3 (apoptosis repressor with CARD domain)

Cybb cytochrome b-245, beta polypeptide

Sesn1_predicted Sesn1_predicted

Aox3 aldehyde oxidase 3

Fcer1g Fc fragment of IgE, high affinity I, receptor for; gamma polypeptide

C1s complement component 1, s subcomponent (EC:3.4.21.42)

Zfp622 Zfp622

Tlr7_predicted Tlr7_predicted

Apobec1 apolipoprotein B mRNA editing enzyme, catalytic polypeptide 1

Tmem199 transmembrane protein 199

Ccr5 chemokine (C-C motif) receptor 5

Gramd3 GRAM domain containing 3

Acadsb acyl-Coenzyme A dehydrogenase, short/branched chain (EC:1.3.99.-)

MtRNA-Glu MtRNA-Glu

Myh7 myosin, heavy chain 7, cardiac muscle, beta

Ccr2 chemokine (C-C motif) receptor 2

176

isg12(b) putative ISG12(b) protein

Fdft1 farnesyl diphosphate farnesyl transferase 1 (EC:2.5.1.21)

Fmo3 flavin containing monooxygenase 3 (EC:1.14.13.8)

Fundc1 FUN14 domain containing 1

Slain2 SLAIN motif family, member 2

Fmo2 flavin containing monooxygenase 2 (EC:1.14.13.8)

Scn4b sodium channel, type IV, beta

Fmo1 flavin containing monooxygenase 1 (EC:1.14.13.8)

Armc2 armadillo repeat containing 2

MtRNA-Gln MtRNA-Gln

Klra7 Klra7

Apol11a apolipoprotein L 11a

Ms4a6b membrane-spanning 4-domains, subfamily A, member 6B

Angpt1 angiopoietin 1

Ms4a6a membrane-spanning 4-domains, subfamily A, member 6A

Serpine1 serine (or cysteine) peptidase inhibitor, clade E, member 1

RT1-CE5 RT1 class I, CE5

Cytl1 cytokine like 1

Omd osteomodulin

Lmod3 leiomodin 3 (fetal)

RT1-CE1 RT1 class I, CE1

Ces3 carboxylesterase 3 (EC:3.1.1.1 3.1.1.67)

Gabarapl1 GABA(A) receptor-associated protein like 1

177

Ddn dendrin

Arrdc3 arrestin domain containing 3

Fyb FYN binding protein (FYB-120/130)

Emp2 epithelial membrane protein 2

Colq collagen-like tail subunit (single strand of homotrimer) of

Mettl7a methyltransferase like 7A

Gsn gelsolin

Dip2c Dip2c

Hspa4l heat shock protein 4 like

Table 6: List of genes in Cluster #5

Gene Symbol Gene full name

H2a Histone H2a

Bdh1 3-hydroxybutyrate dehydrogenase, type 1 (EC:1.1.1.30)

Inha inhibin alpha

Snurf SNRPN upstream reading frame

Ttr transthyretin

Ckb creatine kinase, brain (EC:2.7.3.2)

B3galt2 UDP-Gal:betaGlcNAc beta 1,3-galactosyltransferase, polypeptide 2

Ngdn neuroguidin, EIF4E binding protein

Hcn4 hyperpolarization activated cyclic nucleotide-gated potassium

Kif22 kinesin family member 22

Lamc2 laminin, gamma 2

Wfdc1 WAP four-disulfide core domain 1

178

Pdia5 protein disulfide isomerase family A, member 5 (EC:5.3.4.1)

Pnpla3 patatin-like phospholipase domain containing 3

Adamts12 ADAM metallopeptidase with thrombospondin type 1 motif, 12

Susd2 sushi domain containing 2

Arhgap11a Rho GTPase activating protein 11A

Bub1b budding uninhibited by benzimidazoles 1 homolog, beta (S.

Gpd1 glycerol-3-phosphate dehydrogenase 1 (soluble) (EC:1.1.1.8)

Mpp7 membrane protein, palmitoylated 7 (MAGUK p55 subfamily member 7)

Atp1a3 ATPase, Na+/K+ transporting, alpha 3 polypeptide (EC:3.6.3.9)

Xpnpep2 X-prolyl aminopeptidase (aminopeptidase P) 2, membrane-bound rno-mir-22 rno-mir-22

Kif4 kinesin family member 4

Hist1h1b histone cluster 1, H1b

Hpse heparanase

Bloc1s2 biogenesis of lysosomal organelles complex-1, subunit 2

Dyrk3 dual-specificity tyrosine-(Y)-phosphorylation regulated kinase 3

Spetex-2D Spetex-2D protein

Prr11 proline rich 11

Ccna2 cyclin A2

Dpep2 dipeptidase 2 (EC:3.4.13.19)

Egln3 EGL nine homolog 3 (C. elegans)

Fam70a family with sequence similarity 70, member A

Sgms2 sphingomyelin synthase 2 (EC:2.7.8.27)

179

Racgap1 Rac GTPase-activating protein 1

Snrpn small nuclear ribonucleoprotein polypeptide N

Acsl6 acyl-CoA synthetase long-chain family member 6 (EC:6.2.1.3)

Acta1 actin, alpha 1, skeletal muscle

Igsf7 immunoglobulin superfamily, member 7

Ucp3 uncoupling protein 3 (mitochondrial, proton carrier)

Cpxm2 carboxypeptidase X (M14 family), member 2

Pfkfb1 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 1 (EC:2.7.1.105

Myo16 myosin XVI

Rfx4 regulatory factor X, 4 (influences HLA class II expression)

Fbxo44 F-box protein 44

Slc4a1 solute carrier family 4 (anion exchanger), member 1

Bub1 budding uninhibited by benzimidazoles 1 homolog (S. cerevisiae)

Spna1 Spna1

Fcer2a Fcer2a

Cdc20 cell division cycle 20 homolog (S. cerevisiae)

Ard1b ARD1 homolog B (S. cerevisiae) (EC:2.3.1.88)

180

The article #1:

Physiological genomics 34: 78-87, 2008.

Physiological and molecular evidence of heat acclimation memory: a lesson from thermal responses and ischemic cross-tolerance in the heart.

Tetievsky A, Cohen O, Eli-Berchoer L, Gerstenblith G, Stern MD, Wapinski I, Friedman N, and Horowitz M.

The article #2:

Journal of applied physiology 109: 1552-1561, 2010.

Posttranslational modifications in histones underlie heat acclimation-mediated cytoprotective memory

Tetievsky A, and Horowitz M

181

182