| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711
b Department of Anatomy, School of Medicine, Chiba University, Chiba 260-8670, Japan
c Core Research for Evolutional Science and Technology CREST, Kawaguchi City, Saitama 332-0012, Japan
ABSTRACT
Testicular heat shock was used to characterize cellular and molecular mechanisms involved in male fertility. This model is relevant because heat shock proteins (HSPs) are required for spermatogenesis and also protect cells from environmental hazards such as heat, radiation, and chemicals. Cellular and molecular methods were used to characterize effects of testicular heat shock (43°C for 20 min) at different times posttreatment. Mating studies confirmed conclusions, based on histopathology, that spermatocytes are the most susceptible cell type. Apoptosis in spermatocytes was confirmed by TUNEL, and was temporally correlated with the expression of stress-inducible Hsp70-1 and Hsp70-3 proteins in spermatocytes. To further characterize gene expression networks associated with heat shock-induced effects, we used DNA microarrays to interrogate the expression of 2208 genes and thousands more expression sequence tags expressed in mouse testis. Of these genes, 27 were up-regulated and 151 were down-regulated after heat shock. Array data were concordant with the disruption of meiotic spermatogenesis, the heat-induced expression of HSPs, and an increase in apoptotic spermatocytes. Furthermore, array data indicated increased expression of four additional non-HSP stress response genes, and eight cell-adhesion, signaling, and signal-transduction genes. Decreased expression was recorded for 10 DNA repair and recombination genes; 9 protein synthesis, folding, and targeting genes; 9 cell cycle genes; 5 apoptosis genes; and 4 glutathione metabolism genes. Thus, the array data identify numerous candidate genes for further analysis in the heat-shocked testis model, and suggest multiple possible mechanisms for heat shock-induced infertility.
apoptosis, gamete biology, gene regulation, spermatogenesis, testis
INTRODUCTION
It was shown more than 75 years ago that mammalian testes must descend from the abdominal cavity for normal development to occur, and that the elevated testicular temperature caused by cryptorchidism disrupts spermatogenesis and causes infertility [1]. More recently, a number of studies have documented the adverse effects of hyperthermia on the normal adult testis in several species, including mouse [2], rat [3], cow [4], pig [5], sheep [6], and human [7, 8]. The reported effects include a temporary reduction in relative testis weight accompanied by a temporary period of partial or complete infertility [9–11]. Sperm quality has also been shown to suffer, with a reduction in progressive sperm motility and a significantly lower in vitro fertilization rate of oocytes by sperm from heat-shocked males [9]. Precoital testicular heating appears not only to reduce the number of successful matings [9], but also produces a transient retardation in embryo growth [9–11], and has been seen to increase the rate of embryonic degeneration in rats [10] and sheep [6].
Although the physiological and cellular responses to heat treatment of the testes have been well documented, gene expression following increased scrotal temperature has been poorly described, and the molecular mechanisms through which these responses are directed remain largely unknown. One of the best characterized responses by eukaryotic tissues to environmental stresses, such as elevated temperature or drug or chemical exposure, is that of the so-called heat shock protein (HSP) genes [12]. Tissieres et al. [13] found that RNAs encoding HSPs are induced following heat shock treatment of fruit fly larvae. Although a number of HSPs are now known to be induced by such environmental stress, one particular family of stress proteins, the HSP70s, compose the major class of proteins induced by elevated temperatures. The HSP70s are 70-kDa chaperones that assist in the folding, assembly, and disassembly of other proteins [14]. They have been shown to be important both in spermatogenesis [15–17] and in protecting developing embryos from the lethal effects of hyperthermia [18, 19]. Because the HSPs are clearly important in facilitating and maintaining reproductive vigor, the perturbation of either or both of their regulation and normal expression patterns by environmental exposure to heat or chemicals could cause adverse effects on the fertility of humans and other mammalian species.
In order to better characterize the disruption of spermatogenesis by heat shock, DNA arrays were employed to monitor gene expression in testis. DNA arrays [20] are a powerful tool for functional genomics studies because they permit analysis of the expression of thousands of genes simultaneously. Such arrays, with hundreds or thousands of genetic elements on each, often include large numbers of stress-inducible genes and others encoding proteins that regulate cell cycle and cell death. Such genes are of high toxicological interest and can provide useful information on how environmental agents either regulate or disrupt entire gene expression networks [21, 22], thereby facilitating the characterization of mechanisms of action of known and suspected toxicants.
MATERIALS AND METHODS
Animals
Male C57BL/6 mice were obtained from Charles River Laboratories (Raleigh, NC) at 8 wk of age, and maintained in a temperature- and humidity-controlled room on a 12-h light/dark cycle. Animals were housed singly in polycarbonate cages with a bedding of pine shavings, and they had free access to food and water. The animals were treated and housed in accordance with approved guidelines (Guidelines for the Care and Use of Laboratory Animals), and approved project reviews (Animal Care and Use Committees of the U.S. Environmental Protection Agency [EPA] and the National Institute of Environmental Health Sciences).
Heat Shock
At 10–11 wk of age, animals were exposed to a single heat shock. Each was sedated and the lower half of the torso of each animal was submerged in a water bath for 20 min, after which time the animals were dried and returned to their cages. Heat shocks were carried out in a 43°C bath, controls at 33°C, and conditioning at 39°C and 40°C.
Histology
Two to 72 h after heat shock, animals were killed and one testis and epididymis from each animal was immersion-fixed in 4% paraformaldehyde for 16 h at 4°C, paraffin-embedded, sectioned, and stained with hematoxylin and eosin. Periodic acid Schiff (PAS) reagent staining was used to differentiate spermatocytes from spermatids.
RNA Extraction and Reverse Transcription-Polymerase Chain Reaction
Halves of one testis from each animal were flash-frozen in liquid nitrogen and stored at -80°C for protein or RNA extraction. Testis halves were thawed and homogenized in Tri Reagent (Sigma, St. Louis, MO), and RNA was extracted according to the manufacturer's instructions. Duplicate batches of 4-µg pooled testis RNA from three control or three heat-shocked mice were treated for 15 min at room temperature with 4 units of amplification-grade DNase I (Life Technologies, Rockville, MD), then heat-inactivated by adding 1/10 volume of 25 mM EDTA at 65°C for 10 min. Duplicate RNA samples were then split for reverse-transcription (RT) with and without reverse transcriptase. Four micrograms of oligo dT (Research Genetics Inc., Huntsville, AL) was added to 4 µg of total RNA, and the mixture heated to 70°C for 10 min, then cooled on ice. Samples were reverse transcribed in a total volume of 70 µl with 600 units Superscript II (Life Technologies) according to the manufacturer. Two microliters of these cDNAs were used as templates in 25 µl polymerase chain reactions (PCRs) to detect hsp70-1 and hsp70-3 as previously described [23]. PCRs were run for 25–40 cycles to identify linear ranges of amplification. Amplicons from hsp70-1 (285 base pair [bp]) were generated at 25 cycles, hsp70-3 (220 bp) at 30 cycles.
Protein Extraction and Western Blot
Testis halves were thawed and homogenized in 1 ml protein extraction buffer (50 mM Tris pH 7.5; 100 mM NaCl; 1 mM EDTA; 2.5 mM EGTA; 0.1% NP-40; 10 µg/ml each of soybean trypsin inhibitor, leupeptin, and aprotinin; and 50 µg/ml PMSF), centrifuged at 13 000 x g for 20 min at 4°C, and protein supernatants transferred to clean tubes. Protein concentrations were determined by bicinchoninic acid protein assay. Proteins were separated by SDS-PAGE on 7.5% gels and blotted to nitrocellulose. Hsp70-1 and Hsp70-3 were detected using mouse monoclonal antibody C92F3A-5 (StressGen Biotechnologies Corp., Victoria, BC, Canada) at a 1:1000 dilution. Horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G (Boehringer-Mannheim Biochemicals, Indianapolis, IN) was used to detect the primary antibody and visualized by ECL-Plus (Amersham, Arlington Heights, IL).
Immunohistochemisty
Unstained testis sections were immunostained with an Elite ABC-peroxidase kit (Vector Laboratories, Burlingame, CA) and counterstained with hematoxylin. Hsp70-1 and Hsp70-3 were detected using mouse monoclonal antibody C92F3A-5 (StressGen Biotechnologies) at a 1:400 to 1:1000 dilution. The intensity of peroxidase staining was assessed using National Institutes of Health Image software to analyze optical density within individual cells from black and white video frames. A 14-step optical density standard was used to confirm that optical densities from the immunoperoxidase experiments were within the linear range of response of the video microscopy system. For each dose or time group, optical density was measured in 81 cells from nine sections from three different animals. Control values were derived from mice exposed to 33°C and allowed to recover for 16 h. Means were compared by one-way ANOVA and Tukey-Kramer multiple comparisons tests. The percentage of cells expressing Hsp70-1 and 70-3 was derived by counting all spermatocytes and spermatids in three seminiferous tubules from three animals from each time-dose point. Cells were considered positive for Hsp70-1 and Hsp70-3 expression if the optical density was above the 99% confidence interval for the mean of control cells.
TUNEL and Immunofluorescence
In situ TUNEL of apoptotic cells followed by immunofluorescent detection of Hsp70-2 was performed as previously described [15]. Rhodamine B-labeled streptavidin (Molecular Probes, Eugene, OR) was bound to biotin-labeled DNA using a modified TUNEL method. After TUNEL labeling, sections were processed for immunofluorescent detection of Hsp70-2 using rabbit polyclonal antibody 2A [24] to identify spermatocytes and spermatids. The percentage of seminiferous tubules containing three or more TUNEL-positive cells was determined by counts on five different sections per mouse (mean ± standard deviation = 75.4 ± 22.7 tubules per mouse). Nonparametric ANOVA (Kruskal-Wallis) and Dunn multiple comparisons tests were used to compare means and determine significance of variation.
Mating Study
Male mice were assigned randomly into five groups of six and two groups of three. The first group of three males was killed 4 h after heat shock. The second group of three males, identical except for not having received heat shock, was also killed at this time. The right testis of these six animals was fixed in 4% paraformaldehyde and embedded in paraffin. The left testis was snap-frozen in liquid nitrogen and stored at -80°C. Three 10-wk-old virgin females were added to each male of the remaining five groups of six males at predetermined time intervals after heat shock (see Table 1) and allowed to mate for 5 days. Females were checked for postcoital plugs each morning. If a plug was observed, the female was noted as being at Gestation Day (g.d.) 0.5. On Day 5 of the mating period males were removed and killed. Their testes and epididymides were collected as before. Plug-positive females were killed on g.d. 14.5 and litters assessed for number and weights of embryos, resorptions, and terata.
|
DNA Microarrays
Testes of three animals harvested 4 h after heat shock were pooled and total RNA extracted using Tri Reagent according to the manufacturer's instructions. The same procedure was carried out for the testes of the three untreated males. Two DNA array systems were utilized. In the first approach, 1 µg of pooled total RNA from the untreated (control) animals and 1 µg from those killed 4 h after heat shock were submitted to Incyte Genomics Inc. (St. Louis, MO) for hybridization to the mouse Gene Expression Microarray I (GEM 1, http://gem.incyte.com/gem/mousegem1.html). Hybridization of the samples and analysis of the gene expression changes was carried out by Incyte staff and the data were returned electronically. Incyte does not consider a change in balanced differential expression significant unless it is greater than 1.7-fold. In addition, signal over background should be greater than 2.5-fold. Furthermore, if both the treated and control values are very low (below about 700 on their intensity scale), expression may be hard to verify. These figures were used as benchmarks to report genes showing altered expression with the Incyte GEM. In the second approach, 32P-labeled cDNAs were produced from 2 ;gmg of each pooled RNA using the reagents supplied with the Atlas mouse cDNA expression and Stress/Tox array kits (Clontech Laboratories Inc., Palo Alto, CA). Hybridization to these membrane arrays was carried out according to the manufacturer's instructions, and images were developed using a PhosphorImager (Molecular Dynamics Inc., Sunnyvale, CA). The control and heat-shock RNA pools were each hybridized against two replicate membranes. Images were quantitated and analyzed using Atlas Image version 1.2 (Clontech) as follows: Expression data from replicate membranes were averaged and normalized. Genes whose adjusted signal intensity (intensity minus background) were not at least twofold the background value on both membranes were not considered genuine signals and were discarded, as were those that had opposing directional changes in replicate pairs. A twofold change in treated versus control was used as the cutoff point for tabulating expression effects.
RESULTS
Figure 1 shows the relative testis weight (RTW) of experimental males at various time points after a single heat shock of 43°C for 20 min. The initial drop in RTW reaches a minimum 15 days after heat shock, then gradually recovers. However, even 68 days after heat shock, RTW is still not fully back to control levels. A mating study conducted at predetermined intervals after application of the heat shock (Table 1) provides evidence that spermatocytes are most susceptible to this treatment (Fig. 2). Only one small litter was sired from males 23–28 days after heat shock. No significant differences were observed between heat-shocked and control males in terms of fetuses per male in other time periods. By the time heat-shocked spermatagonia have developed into mature sperm, the average number of fetuses sired per male has returned to control levels (matings on Days 34–38; Fig. 2).
|
|
Histological examination of testes harvested shortly after treatment revealed that control shock (33°C) had no effect on spermatogenesis (Fig. 3A). A 20-min exposure at 39°C also had no effect on spermatogenesis (data not shown). However, 8 h after a 43°C heat shock, vacuoles were common in tubules, many germ cell nuclei contained highly condensed chromatin (pyknotic nuclei), and some nuclei appeared to be breaking up into apoptotic bodies (Fig. 3B). Sixteen hours after 43°C heat shock, tubules had degraded even further and many contained giant degenerating cells of primarily spermatocyte origin (Fig. 3C). In part due to abnormal cytoplasmic retention, no spermatid/sperm tails filled tubule lumens 16 h after heat shock.
|
Concordant with disruption of spermatogenesis, a 43°C heat shock of the testis induced the expression of the major stress-inducible Hsp70s, with both hsp70-1 and hsp70-3 mRNAs being detected by RT-PCR only 4 h later (Fig. 4A). Expression of Hsp70-1 and Hsp70-3 proteins followed within 16 h after heat shock as confirmed by Western blot (Fig. 4B). Both the mRNAs and proteins were only detected in extracts from testes exposed to 43°C (e.g., exposure to 39°C or 40°C had no affect on Hsp70-1 and Hsp70-3 expression detectable by Western blot (Fig. 4B).
|
Immunohistochemical detection was used to link the expression of Hsp70-1 and Hsp70-3 to spermatocytes perturbed by heat shock of the testis. The time course of Hsp70-1 and Hsp70-3 expression in the testis was coincident with the disruption of spermatogenesis, peaking at 16 h post-43°C and predominant in spermatocytes (Fig. 5). Optical density of immunoperoxidase staining for Hsp70-1 and Hsp70-3 in the heat-shocked group compared with control spermatocytes indicated that significant expression of Hsp70s peaked 16 h after treatment. Within 48 h of 43°C heat shock, the expression of Hsp70-1 and Hsp70-3 was back to control levels, coincident with apparent seminiferous tubule recovery.
|
To confirm the morphological assessment that spermatocytes became apoptotic following heat shock, TUNEL detection of DNA fragmentation was performed (Fig. 6). An increase in TUNEL-positive cells coincided with the disruption of spermatogenesis and degeneration of seminiferous tubules seen in Figure 3. Increased numbers of apoptotic spermatocytes were evident as early as 8 h after heat shock. Twenty-four hours after a 43°C heat shock, significantly more tubules had TUNEL-positive cells and the majority of these apoptotic cells were still spermatocytes. Germ cells were identified by immunofluorescent detection of Hsp70-2, the Hsp70 expressed developmentally in spermatocytes. Exposure at 39°C had no significant effect detectable by TUNEL.
|
To characterize mechanisms underlying the disruption of spermatogenesis and the induction of apoptosis by heat shock, DNA microarrays were employed. Altered gene expression was assessed using Incyte GEM 1 slide-based arrays and Clontech filter arrays (Fig. 7). These two types of DNA microarrays interrogated the expression of 2208 genes, and thousands more ESTs expressed in the mouse testis. Of these genes, 27 were up-regulated and 151 were down-regulated after heat shock. Selected expression changes are shown in Table 2 (up-regulated) and Table 3 (down-regulated). Complete sets of the gene expression data from the GEM 1 Incyte arrays and Clontech filter arrays will be available via the U.S. EPA MicroArray Consortium (EPAMAC) [21, 22], website (http://www.epa.gov/nheerl/epamac).
|
|
|
One apparent pattern emerging from both array platforms is that the ratio of up-regulated to down-regulated genes is low (Table 2 versus Table 3). However, only 75 named genes were present on both the Incyte GEM and either the Clontech Stress/Tox array or cDNA expression array, and only nine of these were present in all three arrays (BRCA1, DNA Ligase 1, DNA polymerase catalytic subunit, GAPDH, HO1, HPRT, an Hsp60, MAPKAPK2, and Myosin 1). Thus, the results provide limited information to make direct comparisons between the two array formats.
For two stress response genes, heme oxygenase 1 (HO1) and mitochondrial matrix Hsp60, both the Incyte and Clontech arrays did detect substantial increases in expression (Table 2). These two genes were the only effects on gene expression detected on both the Incyte and Clontech arrays. The other 25 genes up-regulated in response to heat shock were interrogated on only one type of array, Clontech or Incyte.
Gene classification by biological process was used to characterize the biological significance of array results. Biological processes are defined as broad biological goals accomplished by ordered assemblies of molecular functions. Classification was according to conventions of the Gene Ontology Consortium (http://www.geneontology.org/). Seven of the 27 up-regulated genes were stress response regulators and effectors (Table 2), including Hsp25, an Hsp40, and the aforementioned Hsp60. Other members of the stress response/HSP family, including Hsp86, which is expressed at a constitutively high level in male germ cells, was down-regulated (Table 3). Expression of MTJ1 (another Hsp40) and heat shock factor 1 (HSF1) also decreased. Unfortunately, the expression of stress-inducible HSP70s, which is regulated by HSF1, was not interrogated by the commercial arrays utilized in this study. Other genes expressing proteins that interact with Hsp86 were also down-regulated, including cyclophilin 40, FK506-binding protein 1A, and the glucocorticoid receptor A. Seven genes encoding T-complex protein subunits of the cytosolic chaperone complex, which also interact with a number of HSPs, were down-regulated in response to hyperthermia as well.
Besides HSP genes, expression of a number of genes related to oxidative stress and DNA repair were affected. Four glutathione metabolism genes and two superoxide dismutases were all down-regulated (Table 3). This is in contrast to the up-regulation of HO1 and another oxidative stress-induced protein. Ten DNA repair and recombination genes were down-regulated, including base and nucleotide excision repair, and mismatch repair genes. The nucleotide excision repair gene, RAD23B, was up-regulated twofold.
Nine cell cycle genes and five apoptosis genes on the Clontech arrays had reduced mRNA levels. For example, cell cycle regulating cyclins A1, B1, B2, and D2, and the M-phase inducing phosphatase, CDC25a, were all down-regulated. Bak1, Bax, and procaspase2, all promoters of cell death, were also down-regulated 4 h after heat shock, as were the antiapoptotic proteins, Dad1 and Bag1.
Expression of additional genes associated with male fertility was also affected by heat shock. Both laminin receptor (the 40S ribosomal protein SA) and laminin gamma 2 were up-regulated. A number of other genes significant to meiotic and postmeiotic spermatogenesis were down-regulated, including c-abl; the murine homolog of ataxia telangiectasia (Atm), translin; and the nonhistone chromosomal protein, HMG-14. Ca2+/Calmodulin-dependent protein kinase IV (CamKIV) catalytic subunit and insulin-like growth factor binding protein 2 precursor expression were also down-regulated.
DISCUSSION
The physiological effect of heating the adult mammalian testis is documented in several species—testicular weight loss and a period of infertility, followed by a gradual return to normality over a period of one to two spermatogenic cycles. The aforementioned observations have been attributed to a reduction in spermatid and spermatozoa numbers, caused by a failure of spermatocytes to complete their maturation cycle. This has been shown by flow cytometry [25, 26] and histopathology [3, 27], and is in agreement with results from our mating studies.
Results from our mating studies agree with earlier, somewhat preliminary findings with mice [9, 11] and rats [10] characterizing a transient subfertility post-heat shock. A dominant lethal breeding assay to determine definitively which spermatogenic cell types are damaged by heat shock had not been conducted and reported before in the hyperthermic mouse model. These results are significant for comparing with recent histopathology results in rat [28], in which the middle stages of rat spermatogenesis were resistant to the adverse effects of heat. Our mating studies, along with the earlier studies in mice [9, 11] and rats [10], indicate that the initial histopathology evident in the early and late stages may not reveal the full story. It appears the middle stages of spermatogenesis were not fully resistant to the effects of heat shock, resulting in a 14-day period of subfertility in mice, from Days 20 to 33 post-heat shock. In additional mating studies we have conducted (as yet unpublished) using wild-type and HSF1 gene knockout mice, we have put individual mice through a 40-day continuous breeding assay following heat shock. Results from these experiments separate the histopathologic lesions in early and late spermatogenic stages (i.e., apoptotic spermatocytes), which are dependent on HSF1 and inducible HSP expression, from longer-term effects on reproductive performance relating to spermatocyte-specific but stage-independent effects. The results from the present study are in agreement with the interpretation that mouse spermatocytes are sensitive to heat, independent of the spermatogenic stage of the seminiferous tubule.
The question then arises as to the mechanism through which the effects of heat on testis weight, spermatogenesis, and reproductive performance are taking place. Hand et al. [2] demonstrated that following heat shock of adult male mice, testicular weight loss is manifest by 1 wk and remains unchanged for at least 3 wk after treatment. It was later suggested that heat may cause denaturation of some of the cytoplasmic bridges in the syncytium, which in turn, initiates the degeneration of neighboring cells [27]. This degeneration appears to be carried out through an apoptotic mechanism. Apoptotic germ cells can be detected in the testis a day after inducing experimental cryptorchidism in mice [29], and 1 day after heat shock of adult rat testes [28]. We have now shown that apoptosis is visibly apparent in germ cells within 8 h of heat shock in adult mice and, in agreement with previous findings, that it is the spermatocytes that are primarily affected.
The molecular mechanisms controlling these cellular events are currently undefined, but we postulate that HSPs may be involved. HSP70s are rapidly induced following heat shock. Concordant with the induction of Hsp70 expression, the DNA arrays detected up-regulation of an Hsp40 (DNA J-like 2), which could partner with Hsp70s and provide some additive role in protective functions. Improperly folded proteins are believed to stimulate HSPs, which in turn, prevent further protein folding while the stress conditions are maintained. In this study, Hsp70-1/-3 was detected primarily in spermatocytes, and the temporal expression was closely associated with that of TUNEL-positive cells. It is likely that a general protective response against heat shock would have shown a broader induction pattern across multiple cell types, which suggests that the expression of Hsp70-1/-3 was specifically related to the apoptosis observed in spermatocytes. HSP70s have been shown to prevent heat-induced apoptosis in human cell lines [30]. It appears that the induction of Hsp70-1/-3 in spermatocytes may also be a protective mechanism. Further evidence for the role of HSPs in germ cell apoptosis comes from a report that constitutive HSF1 expression induces stage-specific apoptosis of pachytene spermatocytes in transgenic mice [31]. The mechanism through which HSF1 is acting in this case is uncertain, but because the expression of Hsp70-1/3 is mediated by HSF1 in testicular cells, this provides further evidence that the expression of Hsp70s is an important part of the apoptotic process.
DNA arrays were employed in this study in order to more broadly link gene expression to mechanisms of heat toxicity, and provide leads for better understanding the role of HSPs in germ cell apoptosis. The overall picture of gene expression was one of cellular shutdown. This was true of data from both the Clontech stress/toxicology and cDNA expression arrays, where only 18 of 728 genes were up-regulated, and from the Incyte mouse GEM 1 where only 9 of 1480 genes were up-regulated. In contrast, 117 of 728 and 34 of 1480 genes were down-regulated on the Clontech and Incyte arrays, respectively. This general response is intuitively understandable, wherein an acute environmental insult causes cellular activity to cease, with the exception of a few protective and regulatory genes that required to initiate a defensive or reparative response. More specifically, the meiotic cell cycle of spermatocytes appears to be shutting down at the G2/M-phase transition. This result is indicated by both the morphology of the heat-shocked testes, spermatocyte apoptosis, and the down-regulation of G2/M-specific cyclinB1, cyclin B2, cyclin A1, and M-phase inducer phosphatase CDC25a genes.
Results from the DNA array experiments also provide some information on the nature of the apoptotic events. It appears that 4 h after heat shock, some proapoptotic genes are down-regulated, perhaps as an initial survival response to the acute stress. Bag-1, an antiapoptotic gene that modulates Hsp70 and heat shock cognate 70 (Hsc70) function [32], was down-regulated 2.7-fold. Bag-1 appears to inhibit Hsp/Hsc70-mediated in vitro refolding of unfolded proteins. Overexpression of Bag-1 has been found to protect certain cell lines from heat-induced cell death [32]. Our hypothesis of events is that heat shock causes a rapid shutdown of the majority of cellular transcription machinery, while concomitantly inducing protective proteins such as HSPs. However, the reduction in expression of Hsc70/Hsp70 cofactors (e.g., Bag-1) may serve to limit this protective function. The modulatory activity of Bag-1 appears related to its molar ratio to HSP/Hsc70 [33], and Hohfeld et al. [34] postulated that its antiapoptotic function may be exerted through modulation of the chaperone activity of HSP/Hsc70 on specific protein folding and maturation pathways. Thus, reduction of Bag-1 following heat shock may permit inappropriate HSP-mediated protein refolding during a time when the cell is still recovering from stress. The abnormally folded proteins may in turn induce apoptosis of the affected cells shortly thereafter.
Further evidence that chaperonin-assisted protein folding was perturbed by testicular heat shock was the down-regulation of seven genes encoding T-complex proteins of the cytosolic chaperonin complex. In contrast, the Hsp60 subunit of the mitochondrial chaperonin complex is up-regulated. Both the cytosolic and mitochondrial chaperonins depend on Hsp70s to provide protein-folding intermediates for additional processing, so further exploration of the significance of these effects on expression of HSPs and chaperonins is warranted.
Although Bax is a proapoptotic gene, it is required for the maturation of spermatocytes and to prevent spermatocyte apoptosis. Bax-deficient male mice are infertile as a result of disordered seminiferous tubules with hyperproliferating spermatogonia, multinucleated giant cells, and apoptotic spermatocytes [35]. Thus, a reduction in Bax expression initiated by heat shock may also contribute, directly or indirectly, to maturation arrest, spermatocyte apoptosis, and a decline in postmeiotic germ cells. Our array results indicate a 7.7-fold reduction in Bax mRNA levels in the whole testis. The next step in elucidating the mechanism of heat-induced germ cell apoptosis will be to characterize the cellular and intracellular distribution of Bax protein in the mouse testis, and potential interactions of HSPs with Bax and other Bcl family members. The redistribution of Bax protein from cytoplasm to perinuclear is an early step in germ cell apoptosis triggered by testicular hyperthermia in rats [36]. However, in contrast to our results, there appeared to be no difference in rat Bax mRNA (or protein) levels six hours after heat shock. The reasons for this discordance of results between the rat and mouse models requires further investigation.
A potentially significant trio of genes up-regulated in response to heat shock are laminin receptor 1, laminin 
2, and Hsp25. Laminin, along with collagen type IV, is a major component of the peritubular basal lamina of the seminiferous epithelium [37]. Interactions between the basal lamina and Sertoli cells are essential for maintaining Sertoli cell barrier functions [38], which in turn are required for maintaining spermatogenesis [39]. Sertoli cells express the laminin receptor and in this way are responsive to extracellular laminin exposure. Disruption of basal lamina-Sertoli cell contact leads to the perturbation of F-actin bundles, which accumulate just under the basal and apical surfaces of Sertoli cells. Thus, these F-actin bundles are adjacent to the basal lamina and spermatid contacts of Sertoli cells. The F-actin bundles in Sertoli cells colocalize with Hsp25 [40], and Hsp25 expression in the testis is also up-regulated in response to heat shock according to our array results. The up-regulated expression of laminin, laminin receptor, and HSP25 mRNAs in the heat-shocked testis may indicate a cellular signaling pathway connecting the interior of the seminiferous tubule to the extratubular environment.
In the heat-shocked testis, the aforementioned perturbation of basal lamina/Sertoli contacts could also disrupt Ca2+-mediated signaling between cells [41] and lead to CaMKIV dysfunction. CaMKIV is particularly abundant in T lymphocytes, granule cells of the cerebellum, and meiotic male germ cells [42]. Evidence for an effect on CaMKIV function comes from our array results, wherein expression of the CaMKIV gene is down-regulated by heat shock. CaMKIV activates transcription factors CREM
and CREB, which bind cAMP response elements (CREs) in the promoters of various genes whose function is critical for spermatogenesis. One potential confounding issue that needs to be resolved is to separate the expression of CaMKIV from calspermin. Calspermin is the C-terminal 169 amino acids of CaMKIV, whose transcript is initiated from a promoter within an intron of the CaMKIV gene [42]. Calspermin is expressed exclusively in postmeiotic male germ cells, is the most abundant calmodulin-binding protein in spermatozoa, and is of unknown function.
Although we are far from understanding how the recovery process is affected by the coincident down-regulation of transcripts for glutathione metabolizing enzymes, DNA repair enzymes, and the other genes represented in Table 3, we have taken some positive steps in identifying candidate pathways that may be related to the cellular observations. However, the utility of commercial arrays to further elucidate gene expression in the testes may be limited. Not only are such arrays expensive, they often do not contain the necessary complement of genes expressed in testis. For this reason, it seems reasonable to pursue the assembly of testis-specific arrays for characterizing gene expression during spermatogenesis and linking it to mechanisms of toxicity.
The timing of gene up- and down-regulation is important. For Hsp70-1 and Hsp70-3 mRNA and protein, we have characterized the time course of mRNA and protein expression relative to disruption of spermatogenesis and appearance of apoptotic spermatocytes. However, our gene expression analyses by microarray were limited to a single, early time point in the present study. First, because of the great expense of the commercial arrays being used. And second, because these commercial arrays were being used to identify genes for further analysis with custom-made arrays tailored for expression analysis in the mouse testis. Use of these custom arrays in future studies will allow fuller determination of the time course of gene expression.
In summary, the combination of histological, immunohistochemical, and molecular biological results following heat shock of the testis has yielded an intuitively understandable combination of observations that have begun to inform our mechanistic understanding of heat's disruption of spermatogenesis. Clearly it will be valuable to further analyze the functions of Hsp70s in spermatogenic cell cycle and cell death. Furthermore, genes and gene networks identified as significant by microarrays provide important leads for pursuing a more complete understanding of male reproductive toxicity.
FOOTNOTES
First decision: 21 November 2000.
1 The information in this document has been funded in part by the U.S. Environmental Protection Agency (EPA). It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of EPA, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. We thank Drs. James Allen and Jeffrey Welch (EPA) for scientific review of the manuscript prior to submission. 
2 Correspondence: David Dix, U.S. Environmental Protection Agency, Reproductive Toxicology Division (MD-72), Research Triangle Park, NC 27711. FAX: 919 541 4017; dix.david{at}epa.gov
Accepted: March 13, 2001.
Received: November 1, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  C. Paul, A. A Murray, N. Spears, and P. T K Saunders A single, mild, transient scrotal heat stress causes DNA damage, subfertility and impairs formation of blastocysts in mice Reproduction, July 1, 2008; 136(1): 73 - 84. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Paul, D. W. Melton, and P. T.K. Saunders Do heat stress and deficits in DNA repair pathways have a negative impact on male fertility? Mol. Hum. Reprod., January 1, 2008; 14(1): 1 - 8. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T. Selsby, S. Rother, S. Tsuda, O. Pracash, J. Quindry, and S. L. Dodd Intermittent hyperthermia enhances skeletal muscle regrowth and attenuates oxidative damage following reloading J Appl Physiol, April 1, 2007; 102(4): 1702 - 1707. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-K. Kim, J. Suarez, Y. Hu, P. M. McDonough, C. Boer, D. J. Dix, and W. H. Dillmann Deletion of the Inducible 70-kDa Heat Shock Protein Genes in Mice Impairs Cardiac Contractile Function and Calcium Handling Associated With Hypertrophy Circulation, June 6, 2006; 113(22): 2589 - 2597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Wistuba, C. M. Luetjens, R. Wesselmann, E. Nieschlag, M. Simoni, and S. Schlatt Meiosis in Autologous Ectopic Transplants of Immature Testicular Tissue Grafted to Callithrix jacchus Biol Reprod, April 1, 2006; 74(4): 706 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Farias, E. Bustos-Obregon, and J. G. Reyes Increase in Testicular Temperature and Vascularization Induced by Hypobaric Hypoxia in Rats J Androl, November 1, 2005; 26(6): 693 - 697. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Honaramooz, E. Behboodi, C. L. Hausler, S. Blash, S. Ayres, C. Azuma, Y. Echelard, and I. Dobrinski Depletion of Endogenous Germ Cells in Male Pigs and Goats in Preparation for Germ Cell Transplantation J Androl, November 1, 2005; 26(6): 698 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T. Selsby and S. L. Dodd Heat treatment reduces oxidative stress and protects muscle mass during immobilization Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2005; 289(1): R134 - R139. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Banks, S. A King, D S. Irvine, and P. T K Saunders Impact of a mild scrotal heat stress on DNA integrity in murine spermatozoa Reproduction, April 1, 2005; 129(4): 505 - 514. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Takano, H. Matusi, and T. Takahashi Granzyme N, a Novel Granzyme, Is Expressed in Spermatocytes and Spermatids of the Mouse Testis Biol Reprod, December 1, 2004; 71(6): 1785 - 1795. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Rohmer, J. R. David, B. Moreteau, and D. Joly Heat induced male sterility in Drosophila melanogaster: adaptive genetic variations among geographic populations and role of the Y chromosome J. Exp. Biol., July 15, 2004; 207(16): 2735 - 2743. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Somwaru, S. Li, L. Doglio, E. Goldberg, and B. R. Zirkin Heat-Induced Apoptosis of Mouse Meiotic Cells Is Suppressed by Ectopic Expression of Testis-Specific Calpastatin J Androl, July 1, 2004; 25(4): 506 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. H. Ing, A. M. Laughlin, D. D. Varner, T. H. Welsh Jr., D. W. Forrest, T. L. Blanchard, and L. Johnson Gene Expression in the Spermatogenically Inactive "Dark" and Maturing "Light" Testicular Tissues of the Prepubertal Colt J Androl, July 1, 2004; 25(4): 535 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. R. Hunt, D. J. Dix, G. G. Sharma, R. K. Pandita, A. Gupta, M. Funk, and T. K. Pandita Genomic Instability and Enhanced Radiosensitivity in Hsp70.1- and Hsp70.3-Deficient Mice Mol. Cell. Biol., January 15, 2004; 24(2): 899 - 911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Said, U. Paasch, H.-J. Glander, and A. Agarwal Role of caspases in male infertility Hum. Reprod. Update, January 1, 2004; 10(1): 39 - 51. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Izu, S. Inouye, M. Fujimoto, K. Shiraishi, K. Naito, and A. Nakai Heat Shock Transcription Factor 1 Is Involved in Quality-Control Mechanisms in Male Germ Cells Biol Reprod, January 1, 2004; 70(1): 18 - 24. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. K. Tarka-Leeds, J. D. Suarez, N. L. Roberts, J. M. Rogers, M. P. Hardy, and G. R. Klinefelter Gestational Exposure to Ethane Dimethanesulfonate Permanently Alters Reproductive Competence in the CD-1 Mouse Biol Reprod, September 1, 2003; 69(3): 959 - 967. [Abstract] [Full |
