|
|
||||||||
Bodega Marine Laboratory, University of California, PO Box 247, Bodega Bay, California 94923
* To whom correspondence should be addressed. E-mail: eschang{at}ucdavis.edu
| Abstract |
|---|
|
|
|---|
Abbreviations: HSP, heat-shock protein mosM, milliosmolar SDS, sodium dodecyl sulfate SSPE, sodium chloride, sodium phosphate, ethylenediaminetetraacetic acid
| Introduction |
|---|
|
|
|---|
Early studies of lobster environmental physiology were concerned with the determination of optimal and suboptimal conditions for maintaining lobsters in captivity. McLeese (1956) found that the survival of adult Homarus americanus was negatively impacted by various levels of hypo-osmotic stress. By acclimating lobsters over several reduced salinities, he determined that there were interactive effects of low salinity and temperature on survival (McLeese, 1956). Gonzalez and Bradley (1994) observed both up- and down-regulation in the synthesis of specific proteins in the copepod Eurytemora affinis in response to stress from low and high salinities.
To examine the osmotic stress response at the level of gene transcription, we used several homologous molecular probes now available for stress-responsive genes in H. americanus (Shean and Mykles, 1995; Chang et al., 1999; Spees et al., 2002). As indexes of lobster stress in vivo, we examined the levels of the mRNAs encoding the molecular chaperones HSP70 (= 70-kDa heat-shock protein) and HSP90 (=90-kDa heat-shock protein), and polyubiquitin. HSPs are well-characterized and pervasive "chaperones" of nascent peptides during protein synthesis; they also help to re-fold denatured proteins resulting from environmental perturbation of cells (reviewed in Morimoto et al., 1990; Hartl, 1996; Nover and Scharf, 1997; Bukau and Horwich, 1998; Feder and Hofmann, 1999). HSP90 is one of the most abundant proteins in eukaryotic cells, accounting for as much as 1%2% of total cytosolic protein even in the absence of stress (Welch and Feramisco, 1982; Parsell and Lindquist, 1993; Jakob and Buchner, 1994).
Although HSP70 appears to be a "chaperoning generalist" assisting in the folding and re-folding of most proteins, HSP90 chaperones specific targets, many of which require its presence for final maturation (Nathan et al., 1997). Classical examples of HSP90 substrates include a variety of steroid hormone receptors, the aryl hydrocarbon receptor, kinases, the heat-shock transcription factor, and the tumor-suppressor protein p53 (Buchner, 1999). HSP90 plays dual roles in environmental stress responses (Buchner, 1999; Rutherford and Lindquist, 1999) and cellular signal transduction (Pratt, 1997, 1998). HSP90 is likely to regulate the crustacean ecdysteroid (steroid molting hormone) receptor as demonstrated in Drosophila (Arbeitman and Hogness, 2000). In the absence of environmental stress, lobster HSP90 mRNA may be induced by injections of ecdysteroid (20-hydroxyecdysone; Chang et al., 1999) and increases significantly in claw muscle during premolt atrophy (an ecdysteroid-dependent process) (Spees et al., unpubl.).
During cell stress, damaged proteins that cannot be salvaged by chaperones are no longer functional and are potentially cytotoxic (Feder, 1999). The process of destroying proteins and the recycling of useful peptides and amino acids occurs via the ubiquitin/proteasome pathway. Ubiquitin (a 76 amino acid peptide) is bound to targets through a multi-step enzyme cascade. The resulting pattern of moieties signals a degradation complex (26S proteasome) to remove targeted proteins or regulatory subunits from the cell (Varshavsky, 1997). To expedite the production of ubiquitin, eukaryotic genetic coding regions for multiple ubiquitin molecules are linked and transcribed as a single "polyubiquitin" transcript. The resulting polypeptide is post-translationally cleaved into functional ubiquitin units. Because its rapid and robust production is correlated with environmental stress, ubiquitin (ubiquitous in cells) was identified early as one of the first "stress proteins." In mussels (Mytilus trossulus), conjugation of ubiquitin has been used to quantify irreversible protein damage in nature (Hofmann and Somero, 1995). Polyubiquitin mRNA levels increase severalfold in both Manduca sexta (Myer and Schwartz, 1996) and Drosophila (Lee et al., 1988) following thermal stress. Quantification of polyubiquitin mRNA has also been used to examine the degradation process during premolt claw atrophy in the land crab Gecarcinus lateralis (Shean and Mykles, 1995). Recently, increases in ubiquitin concentrations have been observed in grass shrimp following their exposure to heat shock, cadmium, and petroleum fuels (Downs et al., 2001).
Osmotic conditions strongly affect the survival of H. americanus (McLeese, 1956). In response to acute osmotic stress, we hypothesized that stenohaline juvenile lobsters would upregulate mRNA levels of HSP70, HSP90, and polyubiquitinstress-responsive genes often induced during protein synthesis or turnover.
| Materials and Methods |
|---|
|
|
|---|
Osmotic stress
Two-liter jars were fitted with air stones and filled with either 1.0 1 of 100% seawater (controls; 32 ppt), 50% seawater (0.5 l seawater with 0.5 l distilled and 0.2 µm-filtered freshwater; 16 ppt), or 150% seawater (1.0 l total volume with 16 g additional dissolved sea salts; 48 ppt; Sigma). Salinity was determined with a refractometer. The jars were thermally equilibrated by partial submersion in the flow-through system at 15 °C, the rearing temperature of the juvenile lobsters. Lobsters (n = 4 for all treatments) were removed from their compartments and placed directly into the jars, where they remained for 30, 60, or 120 min. During these time points we have previously measured significantly elevated levels of HSP70 and HSP90 mRNAs in lobsters exposed to acute thermal stress (Spees et al., 2002). Controls (100% seawater) were incubated for 30, 60, and 120 min in the same manner. Following all of the treatments (100% survival), animals were quickly removed from the jars and sacrificed. Samples of abdominal muscle and hepatopancreas (midgut gland) were dissected, frozen in liquid N2, and stored at -70 °C.
We ran a parallel experiment to examine the effect of salinity on hemolymph osmolarity. Hemolymph (50 µl) was removed repeatedly from individual lobsters at 0, 30, 60, and 120 min of exposure to 50, 100, or 150% seawater (n = 4 for each time point), and osmolarity was measured with a vapor pressure osmometer (Wescor). The osmolarity of seawater from the flow-through system was 940 mosM. Students t test was used to compare the hemolymph osmolarity of animals incubated in 100% seawater to that of animals incubated in 50% or 150% seawater.
Northern analysis
Total RNA was isolated from the abdominal muscle and hepatopancreas samples (RNAgents kit, Promega), quantified by absorbance at 260 nm with a spectrophotometer, and equally loaded (15 µg abdominal muscle total RNA; 25 µg hepatopancreas total RNA) onto denaturing (formaldehyde) 1% agarose gels. These gels were washed (15 min, diethylpyrocarbonate water); blotted overnight onto nylon membranes (Magnagraph, MSI), UV cross-linked; and prehybridized (2 h) in 5x SSPE buffer (0.75 M NaCl, 50 mM NaH2PO4, 5 mM ethylenediaminetetraacetic acid, pH 7.4), 50% (w/v) formamide, 5x Denhardts, 1% sodium dodecyl sulfate (SDS), and 100 µg/ml denatured sheared salmon sperm DNA at 42 °C. A partial lobster HSP90 clone (350 bp, cloning described in Chang et al., 1999) was 32P-labeled (Prime-It RmT, Stratagene), added directly to the prehybridization solution, and allowed to hybridize overnight at 42 °C. Following hybridization, the blots were washed twice with 2x SSPE at room temperature and placed on film overnight at -70 °C.
Following exposure of the film, the blots were stripped with several washes of buffer (15 mM NaCl, 1.5 mM sodium citrate, pH 7, with 0.1% SDS at 65 °C) until background was minimal; they were then prehybridized, hybridized, and washed as above, except that a partial lobster polyubiquitin cDNA probe was added (690 bp fragment from Acc I digest of lobster polyubiquitin clone; described in Shean and Mykles, 1995). To check for equal loading of RNA, the blots were probed with a partial lobster actin cDNA (700 bp, Harrison and El Haj, 1994). Lastly, following similar washes and prehybridization, a partial lobster HSP70 cDNA probe was hybridized with the blots (500 bp, Spees et al., 2002). Films were scanned on a high-resolution scanner, and densitometry was performed with NIH Image software. The signals from these northern blots were quantified, normalized against the actin signal to control for equal loading of RNA, and expressed as percent control mRNA level. All data were log10-transformed to normalize variance prior to statistical analysis. Students t test was used to identify differences between treatment and control mRNA levels. P values
0.01 were considered significant for the mRNA experiments and
0.05 for the hemolymph osmolarity measurements.
| Results |
|---|
|
|
|---|
|
|
|
|
Polyubiquitin mRNA levels
Hypo- and hyper-osmotic stress both significantly increased polyubiquitin mRNA levels in abdominal muscle (Fig. 5). Significant elevation over control levels was detectable by 30 min of exposure in 50% seawater (P = 0.004). Polyubiquitin mRNA levels were significantly elevated at 60 min of exposure to either salinity treatment (50% seawater, P < 0.001; 150% seawater, P < 0.001).
|
| Discussion |
|---|
|
|
|---|
Tissue-specific patterns of stress-responsive gene expression
We observed tissue-specific patterns of lobster HSP90 expression following osmotic challenge. Whereas HSP90 expression was significantly increased in lobster abdominal muscle by hypo-osmotic treatments, expression in hepatopancreas was unaffected by any duration of hypo-osmotic stress that we tested. Both tissues, however, displayed significant elevations in HSP90 mRNA levels in response to hyper-osmotic stress. As the composite protein pools of abdominal muscle and hepatopancreas may differ considerably, variation in expression patterns may be due to differential stability of proteins arising from ionic or other solute interactions.
In lobster abdominal muscle and hepatopancreas, HSP70 and HSP90 mRNA levels are both raised by thermal stress (Chang et al., 1999; Spees et al., 2002) and by osmotic stress in this study (only HSP90 mRNA was examined in hepatopancreas). Abdominal muscle appears to be a more thermally stable tissue than hepatopancreas (in vivo expression of polyubiquitin does not occur in abdominal muscle from the same lobsters that do express it in the hepatopancreas during and after a 13 °C acute heat-shock; Spees et al., 2002). In the present study, however, abdominal muscle did show an increase in expression of polyubiquitin during both hypo- and hyper-osmotic stress. In fact, it did so to a much greater extent than hepatopancreas, which did not show a significant increase in polyubiquitin mRNA levels in response to any duration of hyper-osmotic stress that we tested. Although less sensitive to thermal stress, abdominal muscle appears to be more sensitive than hepatopancreas to osmotic stress.
Does osmotic stress cause protein instability or denaturation in vivo?
It is thought more likely that low availability of ions perturbs cells by affecting enzyme-ligand interactions rather than by altering enzyme conformations substantially (protein structure) (Somero and Yancey, 1997). The cations Na+, K+, Mg+2, Ca+2, and the anion Cl- are all present in seawater and are known to destabilize proteins at excessive concentrations (>1 M) (Somero and Yancey, 1997). In vitro results from 3T3 and SV-3T3 (rat) cells in culture show that the presence of the osmolyte betaine (N-trimethylglycine) reduces the production of HSP70 during hypertonic (500 mosM) incubation (Petronini et al., 1993). Betaine also inhibits HSP70 protein induction in Madin-Darby canine kidney (MDCK) cells during thermal stress (Sheikh-Hamad et al., 1994), suggesting that, as is commonly observed during thermal stress, some form of protein destabilization may occur during osmotic stress. Interestingly, the heat-shock transcription factor (HSF1) can be activated by either hyper- or hypo-osmotic stress in mammalian (HeLa) cells, but does not induce HSP70 mRNA, indicating that it may play a different role in regulating osmo-sensing pathways or osmotic stress proteins (Caruccio et al., 1997).
Because the hypo- and hyper-osmotic conditions used in this study may not generate ionic conditions in vivo that have been observed to denature proteins in vitro, it is important to consider alternative hypotheses concerning osmotic stress and stress-responsive gene expression. In the rat kidney, both the mRNA and protein levels of HSP72 and HSP25/27 increase steeply along the corticopapillary axis in a pattern that matches tissue solute levels in the distal tubule. It is believed, however, that hypertonicity rather than hyperosmolarity per se is actually responsible for these patterns, because increased HSP synthesis correlates with the addition of relatively membrane-impermeable substances (NaCl) and not with the addition of membrane-permeable substances (urea). Thus, alterations in membrane fluidity or shrinking and swelling may influence HSP expression in some cellular systems without protein denaturation (reviewed in Beck et al., 2000).
Increased molecular chaperone mRNA levels may be required in a particular tissue for signal transduction or osmolyte responses to intracellular ionic changes. Pan et al. (2000) found that, in salmon, hyper-osmotic stress could raise the levels of HSP90 mRNA in the branchial lamellae (both in vitro and in vivo), but not in the kidney. Because cortisol is believed to govern osmoregulatory capacity in salmon by influencing chloride cell differentiation and ATPase activity in the gill, and HSP90 is known to regulate glucocorticoid receptors, Pan et al. (2000) hypothesized that HSP90 might be playing a specific role in signal transduction during osmotic stress. In the lobster, there is evidence that molting hormone (ecdysteroid) titers can influence molecular chaperone gene expression (Chang et al., 1999; Spees et al., unpubl.).
The changes in lobster polyubiquitin gene expression we observed during osmotic stress indicate that protein turnover is occurring. Although the osmotic stress used in this study may not be excessive enough to denature proteins that have achieved their native state, it may have been enough to affect the folding of nascent proteins post-translation. As linear molecules arising from the ribosome, proteins may be more susceptible to ionic fluctuations. Improperly folded proteins in this case would be ubiquitinated and degraded by the 26S proteasome. Alternatively, the synthesis of proteins required for osmotic stress responses could account for the molecular chaperone gene expression we observed, while degradation (ubiquitination) of proteins that are unnecessary or detrimental during osmotic stress could account for increased polyubiquitin mRNA levels. Further study will be required to determine the effects on gene expression of osmotic stresses that are not as extreme as those that we used and that are more like the stresses encountered in nature. In addition, studies will be needed to elucidate the mechanisms responsible for our observations and the extent to which ionic fluctuations can alter protein stability in vivo.
| Acknowledgments |
|---|
| Footnotes |
|---|
1 Current address: Center for Gene Therapy, Tulane Health Sciences Center, New Orleans, LA 70112
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Santos, M. Gaspar, A. Caeiro, C. Branco-Price, A. Teixeira, and R. B. Ferreira Exposure of Lemna minor to Arsenite: Expression Levels of the Components and Intermediates of the Ubiquitin/Proteasome Pathway Plant Cell Physiol., September 1, 2006; 47(9): 1262 - 1273. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. S. Chang Stressed-Out Lobsters: Crustacean Hyperglycemic Hormone and Stress Proteins Integr. Comp. Biol., January 1, 2005; 45(1): 43 - 50. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |