|Home | About | Journals | Submit | Contact Us | Français|
It is known from earlier studies that the heat shock (HS) response in Malpighian tubules (MTs) of Drosophila larvae is different from that in other tissues because instead of the Hsp70 and other common heat shock proteins, Hsp64 and certain other new proteins are induced immediately after HS. In the present study, we examined the kinetics of the synthesis of Hsp70 and Hsp64 immediately after HS and during recovery from HS by 35S-methionine labeling and Western blotting. In addition, we also examined the transcriptional activity of hsp70 genes in larval MT cells at different times after HS by in situ hybridization and Northern blotting. The HS-induced synthesis of Hsp64 ceased by 1 hour of recovery from the HS when synthesis of the Hsp70 commenced. Our results revealed that the induced synthesis of Hsp64 immediately after HS was dependent on new transcription. Although the levels of Hsp70 in MT cells rapidly increased after its synthesis began during recovery, the levels of Hsp64 remained unaltered irrespective of its new synthesis occurring during or after HS. Inhibition of new Hsp64 synthesis by transcriptional or translational inhibitors also did not affect the total amount of this protein in MTs. The Hsp64 polypeptides synthesized in response to HS are degraded rapidly. Apparently, the cells in MTs maintain a balance between new synthesis of Hsp64 and its turnover so that under all conditions a more or less constant level of this protein is maintained. Although the Hsp70 synthesis started only after 1 hour of recovery, the hsp70 genes were transcriptionally activated immediately after HS and they continued to transcribe till at least 4 hours after the HS. The hsp70 transcripts in MT cells that recovered for 2 hours or longer did not contain the 3′ untranslated regions (UTRs), which may allow their longer stability and translatability at normal temperature. Synthesis of Hsp70 during recovery period was dependent on continuing transcription. Assessment of the β-galactosidase activity in 2 transgenic lines carrying the LacZ reporter gene under hsp70 promoter and different lengths of the 5′UTR suggested that the delayed translation of hsp70 transcripts in MTs is probably regulated by some elements in the 5′UTR.
A rapid induction of synthesis of a new set of polypeptides is the hallmark of the response mounted by nearly all cell types of most living organisms when subjected to thermal or several other cellular stresses (Ashburner 1982; Feder and Hoffmann 1999). This rapidly induced synthesis of the heat shock proteins (Hsps) is generally the result of transcriptional activation of heat shock (HS) genes, followed by a quick and preferential translation of the HS messenger ribonucleic acid (mRNA) (Morimoto et al 1994; Wu et al 1994, 1995). However, some exceptions to such a general paradigm for cellular response to thermal and other stresses have been reported earlier (Feder and Hoffmann 1999; Lakhotia 2001a, 2001b). Earlier studies in our laboratory (Lakhotia and Singh 1989, 1996; Singh and Lakhotia 1995) revealed that unlike in other tissues HS fails to induce any of the usual Hsps in Malpighian tubules (MTs) of Drosophila larvae. Instead, synthesis of a different set of proteins, with a 64-kDa protein being the most prominent, is induced immediately after a 30-minute HS. Further studies (Lakhotia and Singh 1996) established that the HS-induced 64-kDa protein in the larval MT of Drosophila was a member of the Hsp60 family. Krebs and Feder (1997) confirmed our observation on noninducibility of the Hsp70 immediately after HS but showed that the Hsp70 did appear in the MT after some hours of recovery from HS and persisted till 21–24 hours. In several other insect species also, the HS response in MT has been reported to be somewhat different from that in most other cell types (Nath and Lakhotia 1989; Tiwari et al 1995, 1997; Singh and Lakhotia 2000). It appears therefore that the HS response in MT of insects is regulated in a manner different from that in other tissues.
A recent study in our laboratory (Lakhotia and Prasanth 2002) on transcription of the individual hsp70 genes in response to HS in different cell types of Drosophila revealed tissue- and developmental stage–specific variations in induction and stability of the hsp70 transcripts.
During the course of this study, we noted that the hsp70 genes were as quickly activated in response to HS in the larval MT cells as in other tissues, notwithstanding the fact that the Hsp70 did not appear till some time after recovery from the stress. In the present study, therefore, we examined the time kinetics of the induction of synthesis of Hsp70 and Hsp64 and their steady-state levels in larval MT cells immediately after HS and during recovery. Our results show that the regulation of synthesis and turnover of Hsp70 and Hsp64 during and after HS in MTs of Drosophila larvae is complex, involving transcriptional, translational, and posttranslational controls.
All flies were reared at 22 ± 1°C on standard food containing agar, maize powder, yeast, and sugar. For cytological preparations, staged larvae were grown in petri plates with food supplemented with additional yeast for healthy growth. In addition to wild-type D melanogaster (Oregon R+) flies, the following 2 transgenic lines carrying the bacterial LacZ reporter gene under hsp70 gene promoter were also used. Bg9/Bg9 stock is an hsp70-LacZ transgenic stock containing the hsp70 promoter (up to −194 bp), 5′ untranslated region (UTR) (259 bp), and 3′UTR (250 bp) of the proximal gene from 87C locus with the bacterial LacZ as the reporter gene in between (Lis et al 1983). The −194a/−194a line is another hsp70-LacZ transgenic stock (Weber and Gilmour 1995), which contains the hsp70 regulatory elements and LacZ reporter gene as in Bg9, except that the promoter (up to −194 bp) is associated with only the first 84 bp of the hsp70 5′UTR.
Salivary glands (SGs) and larval MT along with the gut from wild-type late third instar larvae of D melanogaster were dissected in Poels' salt solution (PSS; Lakhotia and Tapadia 1998) and were heat shocked in cavity slides kept in a moist chamber at 37°C for the desired duration (see Results). In some experiments, whole larvae were given HS in plastic vials with moist filter papers for the desired duration, and the tissues were later quickly dissected out in PSS. For control samples, the tissues were dissected out at 22°C (room temperature [RT]) in PSS directly from unstressed larvae and, if required, incubated in PSS at RT for the desired duration. In experiments requiring recovery from HS, the heat-shocked larvae were allowed to recover at RT for the desired duration before dissection. In case of dissected tissues, the heat-shocked tissues were kept in PSS at 22 ± 1°C (RT) for the desired duration before further processing.
Wild-type larval SG and MT were dissected in PSS and were heat shocked in the presence of actinomycin D (10 μg/mL; Sigma Chemical Co, St Louis, MO, USA) for 40 minutes. For recovery studies, the tissues were heat shocked in the absence of the drug and were allowed to recover at RT in the presence of actinomycin D for various durations.
Wild-type larval SG and MT were either heat shocked in the presence of the translational inhibitor cycloheximide (50 μg/mL; Sigma Chemical Co), for 40 minutes or were heat shocked without the drug and then allowed to recover in the presence of cycloheximide for the desired period.
After the desired treatments, 10 pairs of SGs or MTs from 50 animals were labeled for 30 minutes with 35S-methionine (Activity 400 μCi/mL; Specific Activity ~800 Ci/mM; BRIT, Mumbai, Maharashtra, India) at RT (control and recovery from HS) or at 37°C (HS samples). After repeated washing of the tissues in PSS, protein samples were prepared and electrophoresed under denaturing conditions in vertical sodium dodecyl sulfate (SDS) polyacrylamide slab gels and the 35S-methionine–labeled protein bands detected by fluorography of dried gels, as described earlier (Lakhotia and Singh 1989).
In 1 set, batches of MT were heat shocked for 30 minutes at 37°C and then labeled with 35S-methionine at 37°C as above. One batch (0 minute) was immediately processed for sample preparation, whereas 2 other batches were washed in radioisotope-free PSS and further incubated at 37°C in PSS containing “cold” methionine (100 μg/mL) for 30 minutes and 60 minutes, respectively, before preparation of the sample for polyacrylamide gel electrophoresis (PAGE) and fluorography, as above.
Proteins separated by SDS-PAGE were electroblotted onto Immobilon-P membrane, following the manufacturer's (Millipore Corp, Bedford, MA, USA) instructions. The blots were sequentially or independently challenged with the 7Fb, SPA805, SPA806, or actin (or all) antibodies. The 7Fb, used at 1:2000 dilution, is a rat monoclonal antibody, which detects only the heat-inducible form of Hsp70 in D melanogaster (Velazquez and Lindquist 1984). The SPA805 rabbit polyclonal antibody of Stressgen (Victoria, BC, Canada), used at 1:1000 dilution, identifies the Hsp60 family proteins (Lakhotia and Singh 1996). The SPA806 is a monoclonal mouse antibody of Stressgen that, as reported by the vendor, specifically recognizes the epitope from 383 to 447 amino acid residues of the human Hsp60. This antibody was used at 1:500 dilution for Western detection. To ascertain the relative levels of proteins in different samples on a blot, anti-actin antibody (Sigma Chemical Co) was used at a dilution of 1:200.
After incubation in the primary antibody, horseradish peroxidase–labeled appropriate secondary antibody was used and the signals were detected with the enhanced chemiluminescence detection system, following manufacturer's instructions (Amersham Pharmacia Biotech, Buckinghamshire, UK).
After the desired treatments, the intact MT were immunostained with the 7Fb antibody (1:400 dilution), and the antibody binding was detected chromogenically using horseradish peroxidase–labeled anti-rat antibody, as described earlier (Lakhotia and Prasanth 2002).
Late third instar larvae of the hsp70-LacZ transgenic lines (Bg9 and −194a) were heat shocked at 37 ± 1°C for 1 hour and allowed to recover at 22°C for 1 hour. Gut with the attached MT was dissected out in PSS and fixed in 2.5% glutaraldehyde in 50 mM sodium phosphate buffer, pH 8.0, followed by X-gal staining, as described earlier (Lakhotia and Singh 1989).
The hsp70 transcripts were detected by in situ hybridization in intact or partially squashed MTs and by Northern hybridization with isolated RNA from SGs and MTs. Three different riboprobes were used. The clone pPW18 includes the entire hsp70 complementary deoxyribonucleic acid (cDNA) (2.15 kb) and 1.1 kb of upstream sequences but completely lacks the 3′UTR (Sharma and Lakhotia 1995). The clone pVZ-70-3′#1 contains only the 3′UTR sequence (270 bp) of the proximal hsp70 gene at the 87A7 locus (Dellavalle et al 1994), whereas the clone pVZ-70-3 contains only the 3′UTR sequence (256 bp) of the proximal hsp70 gene at the 87C1 locus (Dellavalle et al 1994). Whereas the pPW18 riboprobe detects all the hsp70 transcripts from any of the 5 hsp70 genes in D melanogaster, the riboprobes from pVZ-70-3′#1 and pVZ-70-3′ clones detect the hsp70 transcripts carrying the 3′UTR sequence from the 87A or 87C locus, respectively (see Lakhotia and Prasanth 2002). For in situ hybridization, the digoxigenin (DIG)-labeled riboprobes were hybridized to RNA in wild-type larval MT cells (control, heat shocked, or recovering from HS; see Results), and the hybridization signal was detected using either anti–DIG-alkaline phosphatase (AP) or anti–DIG-rhodamine antibody, as described earlier (Prasanth et al 2000; Rajendra et al 2002). The anti–DIG-rhodamine–stained tissues were counterstained with diaminophenylindole (1 μg/mL) and mounted in Vectashield (Vector Labs, Burlingame, CA, USA). For anti–DIG-AP antibody, nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate were used as chromogenic substrates, and the slides were mounted in 50% glycerol.
For Northern hybridization, total RNAs (20 μg to 30 μg) from wild-type larval SG and MT (control, heat shocked, and recovering from HS) were electrophoresed on formaldehyde-agarose denaturing gel, as described by Sambrook et al (1989), after which the RNA was transferred to a nylon membrane by capillary transfer. 32P-Labeled pPW18 antisense riboprobe (1 × 108 cpm/μg of RNA) was used for hybridization. After washing, the filter was exposed to X-ray film for 3 days.
It has been reported earlier that instead of the normal set of Hsps, a different set of proteins, with Hsp60 being the most prominent, is induced by HS in MTs of Drosophila larvae (Lakhotia and Singh 1989, 1996). Krebs and Feder (1997) later reported that Hsp70 is induced in the larval MT after 4 hours of recovery from HS. Therefore, we examined the synthesis of these 2 Hsps in larval MT immediately after HS and during recovery by 35S-methionine labeling as well as by Western blotting and immunostaining.
35S-Methionine labeling of proteins followed by SDS-PAGE and fluorography revealed that as reported earlier (Lakhotia and Singh 1989, 1996), and unlike that in SGs (Fig 1A, left panel), none of the typical Hsps were induced immediately after HS in the MT (Fig 1A, right panel). Hsp64 was strongly induced by a 30-minute HS in the larval MT (Fig 1A, right panel, lane 2) but not in SGs (Fig 1A, left panel, lane 2). However, after 1 hour of recovery from the HS, Hsp64 was no longer being synthesized in the MT; instead, the Hsp70 was actively synthesized in this tissue at this time and this continued, at a relatively reduced level, during the second hour of recovery (Fig 1A, right panel, lanes 3 and 4).
Protein samples from control and heat-shocked SG and MT and those recovered from HS for 2 hours and 4 hours, respectively, were separated by SDS-PAGE, Western blotted, and challenged sequentially with the Hsp70, Hsp64 (SPA805), and actin antibodies. Results presented in Fig 1B showed that as expected Hsp70 was rapidly induced in SGs after HS (Fig 1B, left panel, Hsp70 row) and persisted till 4 hours of recovery. However, Hsp70 was not detectable in MT immediately after the HS, but as seen after the 35S-methionine labeling, Hsp70 was detectable after 1 hour (not shown) and continued to be present in high amounts after 2 hours and 4 hours of recovery (Fig 1B, right panel, Hsp70 row).
The delayed synthesis of Hsp70 in polytene cells of MT was further confirmed by in situ immunostaining of control, heat-shocked, and recovering MTs with the 7Fb antibody. As shown in Fig 2 A–D, immediate induction of Hsp70 was seen only in the small stellate cells, which are characteristically interspersed between the large polytene cells (Fig 2B, also see Lakhotia and Singh 1996), whereas the polytene cells of MT showed the presence of Hsp70 only during recovery periods (Fig 2 C,D). Interestingly, unlike the larger polytene cells, which showed a strong presence of Hsp70 till 4 hours after HS, the stellate cells did not show any staining with the 7Fb antibody after 2 hours or 4 hours of recovery from HS.
In agreement with the little induction of Hsp64 synthesis in larval SGs after HS, Western blotting with SPA805 antibody showed little change in the levels of Hsp64 in control and heat-shocked SGs (Fig 1B, left panel, Hsp64 row). Intriguingly, however, although HS caused a very strong induction of new synthesis of Hsp64, as evident from the 35S-methionine labeling (Fig 1A), Western blotting did not reveal any significant detectable increase in the amount of Hsp64 after HS (Fig 1B, right panel, Hsp64 row). Likewise, even when new Hsp64 was not being actively synthesized during recovery from HS (Fig 1A), the level of Hsp64, detected with the SPA805 antibody, remained unchanged (Fig 1B). That all the lanes contained comparable levels of cellular proteins was confirmed by a more or less similar signal obtained for anti-actin antibody in the different samples (Fig 1B, Actin row).
We used another Hsp60 monoclonal antibody, viz, SPA806, for immunoblotting also. As per the specifications provided by the vendor (StressGen), this antibody recognizes a specific epitope of human Hsp60. This region of Hsp64 is conserved in all the 3 known forms of Hsp64 in Drosophila (see Discussion). The SPA806 antibody also did not show any difference in the levels of Hsp64 in control and heat-shocked MT samples (not shown).
The control and heat-shocked MTs were labeled with 35S-methionine, and the label was chased in cold medium for 0 minutes, 30 minutes, or 60 minutes before PAGE and fluorography. In the 0-minute chase sample, the 64-kDa band was heavily labeled (Fig 1C). However, radioactivity in the 64-kDa polypeptides was rapidly lost because by a 30-minute chase at 37°C this band was very feebly labeled and by 60 minutes very little labeling was detectable at the 64-kDa band position. Compared with Hsp64, many other bands continued to show good labeling even after 1 hour of chase (Fig 1C, lane 3).
To check if the induced synthesis of Hsp64 immediately after HS is due to new transcription or is independent of it, MTs were heat shocked in the presence of actinomycin D or cycloheximide and then labeled with 35S-methionine. The proteins were separated by SDS-PAGE and the labeling detected by fluorography. As shown in Fig 3A, inhibition of transcription by actinomycin D inhibited 35S-methionine labeling of Hsp64 in unstressed as well as heat-shocked MT. Cycloheximide treatment also inhibited synthesis of Hsp64 and other proteins in control and heat-shocked MTs. A parallel Western blot with protein samples, as in Fig 3A, was challenged with the Hsp64 antibody (SPA805). It was seen (Fig 3B) that unlike the above results with 35S-methionine labeling, the actinomycin D or cycloheximide treatments did not affect the steady-state levels of Hsp64 in MTs. Comparable signal with the anti-actin antibody in all the lanes indicated more or less similar levels of proteins in the different samples.
RNA in situ hybridization with the hsp70 cDNA as well as the 87A- or 87C-specific 3′UTR riboprobes showed very intense signals in larval MTs soon after HS with the transcripts being mostly restricted in the nucleus (Fig 4 a,e). Thirty minutes after recovery, the MT cells continued to show intense hybridization with all the riboprobes (Fig 4 b,f). The hybridization patterns with the 87A- and 87C-specific 3′UTR riboprobes were similar. After 2 hours of recovery, both the 3′UTR riboprobes showed significantly weaker hybridization (Fig 4g) than did the cDNA (Fig 4c), and after 4 hours of recovery, no hybridization was detectable with any of the 3′UTR riboprobes (Fig 4h). Unlike the 3′UTR riboprobes, the cDNA riboprobe continued to show intense hybridization with transcripts in MT cells after 2 hours and 4 hours of recovery from HS (Fig 4 c,d). Four hours after HS, the cDNA probe showed positive staining only in the cytoplasm of the large polytene cells of the MT (Fig 4d).
Fluorescence in situ hybridization on partially squashed heat-shocked MT cells with the cDNA riboprobe revealed 2 closely placed signals of intense hybridization on polytene chromosomes (Fig 4i). These intensely stained nuclear spots most likely correspond to the sites of hsp70 transcription at the 87A and 87C loci. After 2 hours of recovery from HS also, both the hsp70 loci in the nucleus showed detectable hybridization with the cDNA riboprobe (Fig 4j). These signals were not due to hybridization of the riboprobe to chromosomal DNA in polytene nuclei of MTs because in other experiments it was seen that RNase treatment abolished the signal, whereas DNase treatment did not and also control MT cells did not show any signal under identical conditions of in situ hybridization (not shown). Therefore, the hybridization signals reflect the presence of hsp70 RNA molecules at the 87A and 87C sites because of continued transcription of the hsp70 genes.
Total RNAs were isolated from unstressed or heat-shocked larval MT and SG or from these tissues that were allowed to recover for 4 hours at RT after a 40-minute HS. These were subjected to Northern hybridization with the hsp70 cDNA riboprobe. As expected, no positive signal was observed in lanes in which RNAs from unstressed larval tissues were loaded (Fig 4k, lanes 1 and 2). After HS, strong induction of hsp70 was observed in SG as well as in MT (Fig 4k, lanes 3 and 4) with the transcript size varying between 1.8 kb and 2.4 kb. RNA from SG that were allowed to recover for 4 hours did not show any positive signal (Fig 4k, lane 5), but RNA from MT that recovered for 4 hours hybridized to an RNA species of ~2.2 kb in size (Fig 4k, lane 6).
Our results showed that the hsp70 genes in MT are immediately activated by HS, although Hsp70 is synthesized only after 1 hour of recovery. In view of this and the continued presence of the hsp70 transcripts in MT during recovery period, it was interesting to examine if the transcripts that were induced immediately after HS or that were synthesized during the recovery were translated. For this purpose, MTs were allowed to recover from HS in the absence or presence of actinomycin D for 2 hours or 4 hours after they were heat shocked at 37°C for 40 minutes in the absence of the drug. A Western blot containing samples from control and heat-shocked MTs and MTs recovering for 2 hours or 4 hours from HS in the absence or presence of actinomycin D was sequentially challenged with Hsp70, Hsp64, and actin antibodies (Fig 3C). As expected, Hsp70 was absent in control and heat-shocked MTs (Fig 3C, lanes 1 and 2) but was abundantly present in MTs that recovered from HS for 2 hours in the absence of actinomycin D (Fig 3C, lane 3). However, inhibition of new transcription by actinomycin D during recovery from HS completely prevented the appearance of Hsp70 in Western blots (Fig 3C, lanes 4 and 5). In agreement with the results presented in Fig 4B, the Hsp64 levels in these different samples remained more or less similar (Fig 3C).
To understand the basis for the delayed translation of the hsp70 transcription in heat-shocked MTs, we examined the synthesis of the reporter enzyme (β-galactosidase) in 2 transgenic lines carrying different lengths of the 5′UTR sequence of the hsp70 gene upstream of the LacZ reporter gene (see Materials and Methods). Gut and the associated MTs from heat-shocked Bg9/Bg9 and −194a/−194a larvae were quickly dissected out and processed for X-gal staining to check if the length of the hsp70 5′UTR affected translatability of the HS-induced β-galactosidase transcripts. As reported earlier (Lakhotia and Singh 1989), MTs from Bg9/Bg9 larvae did not show any X-gal staining after HS (Fig 5A), although, as already known (Lis et al 1983; Lakhotia and Singh 1989), the other larval organs showed strong β-galactosidase activity soon after HS (not shown). However, heat-shocked MTs from the −194a/−194a larvae showed strong X-gal staining soon after HS (Fig 5B). Interestingly, the β-galactosidase activity in other tissues of heat-shocked −194a/−194a larvae was significantly poorer when compared with that in heat-shocked Bg9 larvae (not shown).
The basal and more or less universal regulation of the HS response operates through the transcriptional induction of HS genes by the activated HS factor (Morimoto et al 1994; Wu et al 1994, 1995). However, it is becoming increasingly clear that there must exist additional regulatory mechanisms that cause some cells to respond to stress in a subtle or more noticeably different manner (Feder and Hoffmann 1999; Lakhotia 2001a, 2001b; Lakhotia and Prasanth 2002). The HS response in MTs of Drosophila larvae is very unusual because, as reported earlier (Lakhotia and Singh 1989), none of the typical Hsps are induced after HS in MT, but a different set of polypeptides is induced. Among these, a 64-kDa polypeptide belonging to the Hsp60 family was found to be maximally induced (Lakhotia and Singh 1989, 1996). Unlike the other tissues, synthesis of Hsp70, and presumably of the other common Hsps, in larval MTs is induced only later during recovery (Krebs and Feder 1997).
Present results reveal 2 significant aspects of the HS response in MT cells of Drosophila larvae. The first relates to the unchanging levels of Hsp60 in the MT cells, whether the HS-induced new synthesis is allowed or prevented by transcriptional or translational inhibitors. The second relates to immediate activation of the hsp70 genes after HS but a delayed appearance of Hsp70.
As initially reported by Lakhotia and Singh (1989) and reconfirmed in the present study, 35S-methionine incorporation in Hsp64 is increased severalfold immediately after HS to MTs. However, Western blotting did not show a corresponding increase in the level of Hsp64 after HS, although the levels of Hsp70 were found to change in the same blots in parallel with 35S-methionine–labeling data. The constant level of Hsp64 in MT cells under different conditions contrasts with the well-known increase in the net cellular levels of the various Hsps after their synthesis is stimulated by stress. A possible explanation for the apparent contradiction between the 35S-methionine–labeling and Western data could be that the antibodies do not recognize the 35S-methionine–labeled 64-kDa polypeptide in heat-shocked MTs and, therefore, fail to detect its increased level after HS. However, this possibility is ruled out by an earlier observation (Lakhotia and Singh 1996) that the 64-kDa polypeptide, immunoprecipitated from 35S-methionine–labeled heat-shocked MT using the SPA805 antibody, showed as much increased labeling over that from control MT as in total protein samples. Thus, the SPA805 polyclonal antibody recognizes the HS-induced Hsp64. In addition, we have now also used another monoclonal antibody SPA806 from Stressgen, which identifies only a specific epitope of the Hsp64. This particular epitope is fairly well conserved (see Fig 6) in the 3 known Hsp64 forms in D melanogaster (Kozlova et al 1997; Timakov and Zhang 2001; Berkeley Drosophila Genome Project). Thus, this antibody is expected to recognize all forms of the Hsp64 in Drosophila cells. But it is interesting to note that the SPA806 antibody did not reveal any increase in the level of Hsp64 in heat-shocked MT samples. Thus, we conclude that despite the increased synthesis of Hsp64 in MTs immediately after HS, the net amount of the Hsp64 in these cells does not increase.
Interestingly, not only did the level of total Hsp64 in MT cells not increase after HS but also when this synthesis was inhibited by actinomycin D or cycloheximide the Hsp64 level in MT remained unaltered. Put together, these results suggest that the new synthesis and turnover of existing Hsp64 are tightly coupled so that the steady-state level of Hsp64 in MT cells remains unaltered. Our finding that newly synthesized (35S-methionine labeled) Hsp64 disappears within 30 minutes to 60 minutes (Fig 1C) also highlights rapid turnover of this protein. A comparable situation that the increased synthesis of a protein is not accompanied by a corresponding increase in the net amount of that protein has been reported earlier for Hsp56 (Sanchez 1990) and Cyp40 (Mark et al 2001). Reasons and mechanisms for the regulated level of Hsp64 in MT cells are not clear. Nevertheless, it appears that an increase in Hsp64 in cells, even under stress conditions, may have some unwanted effect, and yet the MT cells may need newly synthesized Hsp64 under conditions of HS. Because actinomycin D treatment during HS prevented 35S-methionine labeling of Hsp64 in the MT cells, it is clear that the increased synthesis of this protein after HS in MT cells is dependent on new transcription rather than on increased translatability of preexisting transcripts.
Singh and Lakhotia (1995) demonstrated that the absence of Hsp70 synthesis in larval MTs during or immediately after HS was neither because these tissues already had a basal level of Hsp70 nor because the cognate forms (Hsc70) were heat inducible in these cells. Krebs and Feder (1997) further confirmed the absence of Hsp70 in larval MTs immediately after HS but showed that the Hsp70 is detectable in these cells during recovery. In this context, the present finding that hsp70 genes at 87A as well as 87C sites are readily induced by HS in larval MTs is intriguing. Present results further showed that unlike in other cells where much of the hsp70 transcripts were degraded within 1–2 hours of recovery, the hsp70 genes in MTs continued to actively transcribe even after 2 hours of recovery from HS and the transcripts persisted till at least 4 hours of recovery. The hsp70 transcripts at this time were most likely not carrying 3′UTR because both the 3′UTR riboprobes failed to detect any transcripts and also the size of the hsp70 transcripts at 4 hours of recovery as seen in the Northern blot was smaller.
The absence of 3′UTR in hsp70 transcripts persisting after 4 hours of recovery is significant because normally hsp70 transcripts are degraded within a short time of recovery as a result of the destabilizing role of the AU-rich elements in 3′UTRs of hsp70 transcripts (Simcox et al 1985; Yost et al 1990). Therefore, the absence of 3′UTRs in the hsp70 transcripts present in larval MTs later during recovery may help stabilize the transcripts and allow their translation. Whether the absence of 3′UTR in hsp70 transcripts made during recovery in MT cells is due to a transcriptional or posttranscriptional regulation remains to be examined.
Because the hsp70 genes in larval MTs are quickly induced to transcribe by HS but the protein is not detectable, except in the small stellate cells (Singh and Lakhotia 1995; our observations), till after some recovery period (Krebs and Feder 1997; our observations), it is apparent that a posttranscriptional regulation prevents Hsp70 synthesis in these cells. It was seen that compared with other cell types, the hsp70 transcripts in heat-shocked larval MT cells were mostly nuclear and seemed to move to the cytoplasm only during recovery. This delayed transport may be one of the factors that prevents a rapid translation of these mRNAs. Present results further revealed that the delayed appearance of Hsp70 in MT cells was dependent on continuing transcription because inhibition of transcription by actinomycin D during recovery prevented Hsp70 synthesis. It appears therefore that the hsp70 transcripts synthesized during HS are turned over without translation, and the Hsp70 synthesis during recovery depends on the continued transcription of the hsp70 genes.
The 2 hsp70-LacZ reporter transgenic lines shed some light on the possible regulatory events that affect Hsp70 synthesis in MT cells. The rapid expression of HS-induced reporter LacZ in −194a, but not in Bg9, larval MT is significant. Although in most of the larval tissues in the −194a transgenic line carrying the depleted hsp70 5′UTR the β-galactosidase activity was poorly induced by HS, in the MT cells the reporter activity was easily detectable soon after the HS. Apparently, the difference in the response of the reporter gene to HS in these 2 transgenic lines is related to the deletion of a part (from +85 to +259) of the 5′UTR sequence (see Materials and Methods). Studies by McGarry and Lindquist (1985) on Drosophila cell lines showed that deletion of sequences from 5′UTRs of hsp70 mRNAs prevented their translation at HS temperature, but these transcripts were translated during recovery (Hess and Duncan 1994; Duncan 1996). This may explain the absence of β-galactosidase activity in most tissues of the −194a larvae immediately after the HS. The same region may also be responsible for the opposing pattern of posttranscriptional and translational regulation of the hsp70 mRNA in MT. In MT cells, the presence of this region in the mRNA (as in those transcribed by the resident hsp70 genes or by the transgene in Bg9) seems to prevent translation (and possibly transport to the cytoplasm) during or immediately after HS but in its absence (as in the −194a transgenic larvae) this translational delay does not occur. Present results thus show that the 5′UTR in the hsp70 mRNAs has important regulatory elements, which in addition to being parts of the core promoter (Wu et al 2001) also have additional posttranscriptional regulatory functions. These need to be identified.
The regulatory mechanisms that cause the transcriptional activation of Hsp64 in MT cells and maintain the steady-state levels of Hsp64 in MT cells under all conditions remain to be understood. The significance of this unusual HS response is also not clear. However, because MTs in several other insects also show varying degrees of variations from the basal HS response (Nath and Lakhotia 1989; Tiwari et al 1995, 1997; Singh and Lakhotia 2000), it is likely that the specific cellular physiology of MT cells has some specific requirements.
Another recent study from our laboratory (Lakhotia and Prasanth 2002) has shown that there are cell- and developmental stage–specific variations in the HS-induced transcription of the different hsp70 genes of D melanogaster and turnover of their transcripts. The present studies reveal another example of the fine-tuning of the HS response in relation to specific cellular requirements. Apparently, we still have much to learn about the finer details of an individual cell's response to stress.
We thank Prof. Susan Lindquist for the generous supply of the 7Fb antibody. This work was supported by a research grant from the Department of Biotechnology, Government of India, New Delhi, India, to S.C.L. P.S. and K.V.P. have been supported by research fellowships from the Council of Scientific and Industrial Research, New Delhi, India.