Inactivation of NAD-dependent dehydrogenases from shallow- and deep-living fishes by hydrostatic pressure and proteolysis

Biochimica et Biophysica Acta 913 (1987) 285-291

285

Elsevier BBA32868

I n a c t i v a t i o n of N A D - d e p e n d e n t d e h y d r o g e n a s e s f r o m shallowa n d deep-living fishes by h y d r o s t a t i c p r e s s u r e a n d p r o t e o l y s i s J o h n P. H e n n e s s e y , Jr. a,. a n d J o s e p h F. S i e b e n a l l e r b a College of Oceanography, Oregon State University, M.O. Hatfield Marine Science Center, Newport, OR and b Department of Zoology and Physiology, Louisiana State University, Baton Rouge, 1.,,4 (U.S.A.)

(Received 5 January 1987)

Key words: Malate dehydrogenase; Glyceraldehyde-3-phosphatedehydrogenase; Hydrostatic pressure; Proteolytic inactivation; Pressure inactivation

Cytoplasmic malate dehydrogenase ((L)-malate: NAD + oxidoreductase, EC 1.1.1.37) and glyceraldehyde-3phosphate dehydrogenase (D-glyceraldehyde-3-phosphate:NAD + oxidoreductase, EC 1.2.1.12) homologues from two shallow-living and three deep-living fishes were examined for the effects of hydrostatic pressure on enzyme activity and susceptibility to inactivation by proteinases. These studies were done to determine whether malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues show similar patterns of adaptation to hydrostatic pressure as seen in lactate dehydrogenase (L-lactate:NAD + oxidoreductase, EC 1.1.1.27) homologues from the same species (Hennessey, J.P., Jr. and Siebenaller, J.F. (1987) J. Exp. Zool. 241, 9-15). Fish malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues are much less susceptible to inactivation by hydrostatic pressure than are lactate dehydrogenase homologues from the same species. This difference in susceptibility to inactivation by hydrostatic pressure may be due to the decreased number of intersubunit contacts in malate dehydrogenase and glyceraldehyde3-phosphate dehydrogenase homologues relative to lactate dehydrogenase homologues. Inactivation of malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues by proteinases, both at atmospheric pressure and at elevated hydrostatic pressure, is less than for lactate dehydrogenase homologues from the same species. This suggests that the structural characteristics and conformational perturbations that are responsible for the susceptibility of lactate dehydrogenase to proteolytic inactivation are not found in malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues of the same species.

Introduction Hydrostatic pressures encountered in the deep sea may be sufficient to affect protein structure and function. This may be an important influence

*

Present address: Department of Pharmacology, University of North Carolina at Chapel Hill, Chapel Hill, NC, U.S.A.

Correspondence: J.F. SiebenaUer, Department of Zoology and Physiology, Louisiana State University, Baton Rouge, LA 70803, U.S.A.

on the biology of deep-living organism [1]. Adaptations to hydrostatic pressure have been documented for enzymes of deep-sea fishes, particularly for tetrameric muscle-type (M4) lactate dehydrogenase (L-lactate : NAD + oxidoreductase, EC 1.1.1.27) [2]. Extension of these studies to structural and functional homologues of M4-1actate dehydrogenase has proven to be a useful approach in identifying focal points of adaptation to hydrostatic pressure. Studies of the kinetics of coenzyme binding for M4-1actate dehydrogenase, malate dehydrogenase (L-malate : NAD + oxidoreductase,

0167-4838/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)

286 EC 1.1.1.37) and glyceraldehyde-3-phosphate dehydrogenase (D-glyceraldehyde-3-phosphate : N A D ÷ oxidoreductase, EC 1.2.1.12) homologues from shallow- and deep-living Sebastolobus congeners provide evidence that reduced sensitivity to hydrostatic pressure of coenzyme binding of NAD-dependent dehydrogenases may be an important and ubiquitous adaptation to life at increased hydrostatic pressure [3-5]. Thus, exploitation of structure-function homologies may be an effective means to identify focal points for biochemical adaptations. Hydrostatic pressure, by affecting protein aggregation state and conformation, may influence protein turnover rates by increasing their susceptibility to proteolysis. This is well demonstrated by previous studies of M4-1actate dehydrogenase homologues of marine fishes. In general, lactate dehydrogenase homologues of shallow-living fishes are more susceptible to inactivation by hydrostatic pressure (via dissociation of subunits) and to inactivation by proteases - both at atmospheric pressure and at elevated pressures - than are homologues of deep-living fishes [6,7]. Given the structural and functional homology of malate dehydrogenase and glyceraldehyde-3-phosphate de-

hydrogenase to M4-1actate dehydrogenase, might the pattern of inactivation by pressure and susceptibility to proteolysis seen in M4-1actate dehydrogenase homologues of marine fishes also be found in malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues of these same species? To determine whether malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues of shallow and deep-living species display patterns of adaptation to pressure similar to Ma-lactate dehydrogenase, we have examined the effects of hydrostatic pressure on enzyme activity and susceptibility to proteolytic inactivation of malate dehydrogenase and glyceraldehyde-3phosphate dehydrogenase homologues from two shallow-living and three deep-living marine teleost fishes. These fishes occur from the surface to 4693 m (Table I). Hydrostatic pressure increases 1 atm for every 10 m depth increase; thus, these fishes cover a pressure range of 470 atm. To examine the effects of pressure on proteolytic inactivation of these enzymes we have used two proteinases, trypsin (EC 3.4.21.4) and subtilisin Carlsberg (subtilopeptidase A, EC 3.4.21.14; (formerly EC 3.4.4.16)). Materials and Methods

TABLE I DEPTHS OF ABUNDANCE OF THE MARINE FISHES STUDIED (TAKEN FROM REFS. 26 AND 27) AND RATES OF PROTEOLYTICINACTIVATIONOF THEIR MALATE DEHYDROGENASE HOMOLOGUES. Malate dehydrogenasehomologues were incubated at 10°C and atmosphericpressure with either 1.0 mg/ml trypsin or 0.5 mg/ml subtilisin. + S.E. is given. Species

Depth of abundance (m) Shallow-livingspecies Sebastes melanops O- 100 Parophrysvetulus 20- 330

Inactivation rate(h- 1) trypsin subtilisin 0.377+0.007 0.933+0.075 0.253+0.009 0.202+0.011

Deep-livingspecies Coryphaenoides rupestris 550-1960 0.251+0.012 1.169+0.044 Coryphaenoides acrolepis 475-2825 0.167+0.013 0.262+0.006 Coryphaenoides leptolepis 2288-4693 0.198+0.014 1.422+0.056

Specimens Specimens were taken by otter trawl at their typical depths of abundance on cruises of the R / V Oceanus, R / V Wecoma and R / V Sacajawea. Coryphaenoides acrolepis, Sebastes melanops and Parophrys vetulus were taken off the coast of Oregon, U.S.A. Coryphaenoides rupestris and Coryphaenoides leptolepis were taken in an area south of New England, U.S.A. Specimens were frozen on solid CO 2 at sea and transported to the laboratory where they were maintained at - 85 ° C. Specimens of S. melanops and P. vetulus were maintained at - 20 o C. Chemicals Chemicals, biochemicals and resins for chromatography were purchased from Sigma Chemical. Subtilisin Carlsberg, L-l-tosylamide-2-phenylethylchloromethyl ketone-treated trypsin from bovine pancreas and chicken-egg albumin were

287 obtained from Sigma. Water was processed through a Milli-Q purification system (Millipore).

Enzyme purification Frozen white muscle was homogenized in a blender in 50 mM potassium phosphate (pH 6.8) 5 mM 2-mercaptoethanol and 1 mM EDTA. The homogenate was stirred for 1 h at 4 ° C and centrifuged at 6 0 0 0 × g (0-4°C) for 30 rain. Lactate dehydrogenase was removed from the supernatant by affinity chromatography as described previously [6]. Lactate dehydrogenase-free eluants were brought to 50% saturation with solid ammonium sulfate (300 g/l), allowed to stand for 30 rain and centrifuged at 15 000 × g for 30 min. The pellet was discarded and the supernatant was brought to 90% saturation with solid ammonium sulfate (300 g/l), allowed to stand for 1 h and centrifuged at 15000 x g for 30 min. The pellet was redissolved in a minimal volume of 20 mM Tris-HC1 (pH 8.2 at 5°C), 5 mM 2-mercaptoethanol and 1 mM EDTA (buffer A), and dialyzed exhaustively against buffer A. Malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase were isolated from the dialyzed solution by pseudo-affinity chromatography on a 2.5 × 8.5 cm reactive Blue 2-agarose column equilibrated with buffer A. The column was washed with 2 1 buffer A (0.5 ml/min) to remove non-adsorbing proteins. Glyceraldehyde-3-phosphate dehydrogenase was eluted by washing the column with 1.0 mM NAD in buffer A. Fractions containing glyceraldehyde3-phosphate dehydrogenase activity were retained for electrophoretic analysis of purity. Malate dehydrogenase was eluted by washing the column with 0.3 mM N A D H is buffer A. S. melanops and P. vetulus had multiple isozymes of malate dehydrogenase. The malate dehydrogenase-2 isozyme was separated by ion-exchange chromatography [5]. Enzymes were concentrated by ultrafiltration under N 2 with a PM-10 filter (Amicon), dialyzed against 90% ammonium sulfate, 5 mM 2mercaptoethanol and 1 mM EDTA and stored at 4 o C. EDTA and 2-mercaptoethanol increased the stability of the purified malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase preparations. The purity and subunit relative molecular mass

( M r) of enzyme homologues were determined by electrophoresis in the presence of sodium dodecyl sulfate (SDS) in a 1.5 mm thick, 12.5% polyacrylamide slab gel [8]. The gels were stained in 25% (v/v) 2-propanol, 10% (v/v) acetic acid and 0.25% (w/v) Serva Blue R (Serva Fine Biochemicals). All enzyme preparations produced a single stained band in the gel. The isozyme composition of the enzyme preparations was determined by starch-gel electrophoresis followed by activity staining [9]. All glyceraldehyde-3-phosphate dehydrogenase homologues and the malate dehydrogenase homologues from the three species of Coryphaenoides produced a single stained band. Malate dehydrogenase homologues of S. melanops and P. vetulus exhibited a second minor band, the malate dehydrogenase-1/2 heterodimer. The staining intensity of the malate dehydrogenase heterodimer band was less than 25% of the intensity of the malate dehydrogenase-2 homodimer band. The M r of native malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues were determined by gel filtration on a calibrated 1.5 × 40 cm Sephacryl S-3000 (Superfine) column equilibrated with 50 mM potassium phosphate (pH 6.8) 5 mM 2-mercaptoethanol, 1 mM EDTA and 100 mM NaC1 [6]. Fractions from the column were assayed for protein content and malate dehydrogenase or glyceraldehyde-3-phosphate dehydrogenase activity. Protein was detected by a dye-binding assay [10] using Serva Blue G (Serva Fine Biochemicals) as described by Peterson (11). Malate dehydrogenase activity was determined in a medium containing 100 mM TrisHC1, (pH 8.1 at the assay temperature of 10°C), 20 mM MgC12, 0.4 mM oxaioacetate and 150/~M NADH. The glyceraldehyde-3-phosphate dehydrogenase assay medium contained 80 mM TrisHC1 (pH 8.5 at 10 ° C), 38 mM sodium arsenate, 4 mM L-cysteine, 1.7 mM o-glyceraldehyde 3-phosphate and 1.5 mM NAD. Oxaioacetate and Dglyceraldehyde 3-phosphate solutions were prepared every 3-4 h. Activity, measured by the change in absorbance at 340 nm, was monitored in a Varian Tectron 634S spectrophotometer equipped with a Soltec model 1241 chart recorder. The assay temperature was maintained with a circulating refrigerated water-bath.

288 Pressure inactivation

The inactivation of dehydrogenases by hydrostatic pressure was determined using methods described previously [6]. Enzyme homologues were dialyzed overnight against 1000 vol. of 50 mM Tris-HC1 (pH 7.5 at 10 ° C), 100 mM KCI, 5 mM 2-mercaptoethanol and 1 mM EDTA, then centrifuge-filtered through a 0.45 /~m nylon-66 filter (Rainin Instrument) and diluted to 5/~g per ml. Protein concentration was determined by the Bradford dye-binding assay [10,11], using bovine serum albumin as the standard. Tris-HC1 was used as the buffer because of the low sensitivity of its pK a to hydrostatic pressure ( - 0.019 pH units per 1000 atm [12,13]). Malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues were incubated for 0.5-24 h at 10 °C and pressures up to 1750 atm. Pressures were generated with an Enerpac model P-228 hydraulic pump filled with high grade mineral oil. Pressure was monitored with a Marsh Mastergauge (model E43968). Upon release of pressure, enzyme solutions were assayed for activity at 10 ° C. Assays were begun within 90 s after release of pressure and terminated within 2-5 min after release of pressure. Proteolytic inactivation

Dehydrogenases were dialyzed overnight at 4 ° C against 1000 vol. of 50 mM Tris-HC1 (pH 7.5 at 10°C) 100 mM KC1, 5 mM 2-mercaptoethanol and 1 mM EDTA, and were centrifuge-filtered. Enzymes were diluted to 5/~g protein per ml with dialysis buffer. Trypsin and subtilisin stock solutions were prepared every 30 min with dialysis buffer. Solutions were maintained at their incubation temperatures + 0.2 °C throughout the experiment with a circulating refrigerated water-bath or heating block. Aliquots of 50-100 /~1 were removed periodically during the 2 h incubation to -~ determine the enzyme activity remaining. Proteinase was replaced with an equal concentration of albumin in control incubations. Rates of inactivation were calculated from the slope of the regression of log percent activity remaining versus incubation time. Proteolysis at elevated hydrostatic pressure was performed in 0.2 ml Microflex vials (Kontes Glass) placed in a pressure vessel [6]. Four samples were

incubated for each enzyme homologue: (1)enzyme plus proteinase at 1750 atm pressure; (2) enzyme plus proteinase at atmospheric pressure; (3) enzyme plus albumin at 1750 atm pressure; and (4) enzyme plus albumin at atmospheric pressure. Samples were assayed for enzymatic activity upon release of pressure. The paired samples incubated at atmospheric pressure were assayed immediately afterward. R ~ Subunit structure

In gel-filtration experiments, catalytically active fish glyceraldehyde-3-phosphate dehydrogenase and cytoplasmic malate dehydrogenase-2 homologues eluted from the column in the same volume as rabbit glyceraldehyde-3-phosphate dehydrogenase and pig mitochondrial malate dehydrogenase, respectively. Each enzyme homologue eluted as a single peak of coincident protein and enzyme activity. In SDS-polyacrylamide gels, fish glyceraldehyde-3-phosphate dehydrogenase and cytoplasmic malate dehydrogenase-2 homologues comigrated with rabbit glyceraldehyde-3-phosphate dehydrogenase and pig mitochondrial malate dehydrogenase, respectively. Rabbit glyceraldehyde-3-phosphate dehydrogenase, a tetramer, has a native molecular mass of 146 kDa and a subunit molecular mass of 36 kDa [14,15]. Pig mitochondrial malate dehydrogenase, a dimer, has a native molecular mass of 68 kDa and a subunit molecular mass of 34 kDa [16]. Pressure-inactivation

Incubation of malate dehydrogenase homologues for 30 min at 1750 atm and 10°C resulted in 6-14% loss of activity (Fig. 1). There was no detectable change in the activity of enzyme solutions within the first 30 rain after release of pressure. Malate dehydrogenase homologues of shallow- and deep-living fishes were inactivated to the same extent under these conditions (two-tailed Mann-Whitney test, P > 0.4). Incubations at 1500 atm pressure for 30 min resulted in no detectable loss of malate dehydrogenase activity. Glyceraldehyde-3-phosphate dehydrogenase homologues showed no loss of activity after a 24 h incubation at 1750 atm pressure and temperature

289

100 80 ~ 6o

40 20" 1

2

3

4

2

3

4

5

100] B ..~ 80t

1

~ 20[ 0

i 1

Fig. 1. Inactivation of malate dehydrogenase homologues of marine fishes by hydrostatic pressure and proteolysis.(A), 1.0 mg/ml trypsin; (B) 0.5 mg/ml subtilisin. Malate dehydrogenase was incubated for 30 rain at 10 o C: II, with proteinase at atmosphericpressure; B, with albumin at 1750 atm; n, with proteinase at 1750 atm. The difference in height between the bar on the right and the bar on the left represents pressure-enhancement of proteolyticinactivationof malate dehydrogenase homologues. 1. S. melanops. 2. P. vetulus. 3. C. rupestris. 4. C. acrolepis. 5. C. leptolepis.

less than 25 ° C. Incubation of glyceraldehyde-3phosphate dehydrogenase homologues for 30 min at 30 ° C and 1750 atm resulted in a 9-85% loss of activity (data not shown). At 35 ° C (but not 30 o C), significant inactivation of homologues of shallowliving species occurred at atmospheric pressure (data not shown). Proteolytic inactivation Malate dehydrogenase homologues. T h e rates of inactivation of malate dehydrogenase homologues at 1 0 ° C and atmospheric pressure by 1 m g / m l trypsin were 0.167-0.377 h -1 (Table I). Malate dehydrogenase homologues from shallow-living fishes are more susceptible to tryptic inactivation than homologues from deep-living species (twotailed Mann-Whitney test, P < 0.01). The mean

rate of inactivation for malate dehydrogenase homologues of the two shallow-living fishes was 50% greater than the mean rate for homologues of the three deep-living fishes. The rate of inactivation of malate dehydrogenase homologues by 0.5 m g / m l subtilisin ranged from 0.202 to 1.422 h -1 (Table I). Rates of inactivation for homologues of the two shaUow-occurring species were not significantly different from those for homologues of deep-occurring species (two-tailed Mann-Whitney test, P > 0.4). Loss of malate dehydrogenase activity due to 1750 atm pressure, proteolytic inactivation at atmospheric pressure and proteolytic inactivation at 1750 atm pressure are shown in Fig. 1. Tryptic inactivation of malate dehydrogenase homologues at 1750 atm resulted in a loss of 1-21% more activity than expected from the combined effects of proteolysis at atmospheric pressure and inactivation at 1750 atm. Inactivation of malate dehydrogenase homologues by subtilisin at 1750 atm resulted in a 16-40% greater loss than expected. Proteolytic inactivation at 1750 atm pressure of malate dehydrogenase homologues did not differ between shallow- and deep-living species (two-tailed Mann-Whitney test, P > 0.4, trypsin; P > 0.4, subtilisin; Fig. 1). Glyceraldehyde-3-phosphate dehydrogenase homologues. Glyceraldehyde-3-phosphate dehydrogenase homologues showed no loss of activity after a 24 h incubation with 1 m g / m l trypsin or 1 m g / m l subtilisin at 10 ° C and atmospheric pressure. Incubation at 30 ° C with 1 m g / m l trypsin resulted in activity losses ranging from 0.002-0.040 h -1 (data not shown). At 3 0 ° C and 1 m g / m l subtilisin, glyceraldehyde-3-phosphate dehydrogenase homologues were inactivated at rates ranging from 0.006-0.054 h-1 (data not shown). Glyceraldehyde-3-phosphate dehydrogenase homologues incubated with a 1 m g / m l trypsin or 1 m g / m l subtilisin for 24 h at 10 ° C and 1750 atm pressure showed no loss of activity. Discussion Subunit structure The malate dehydrogenase and glyceraldehyde3-phosphate dehydrogenase homologues of the five fishes studied have subunit and native M r identi-

290

cal to those of porcine mitochondrial malate dehydrogenase and rabbit glyceraldehyde-3-phosphate dehydrogenase, respectively. The result of this and other studies [3,5-7,17,18] indicates that several enzymes of deep-living fishes maintain the multimeric structure typical of their respective homologues in other vertebrates.

Inactivation by hydrostatic pressure Under the conditions used in this study, malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues of marine fishes were far less susceptible to inactivation by hydrostatic pressure than M4-1actate dehydrogenase homologues from the same species (cf. Ref. 7). After 30 min at 1750 atm pressure and 10 o C, fish malate dehydrogenase homologues lost 6-14% activity; fish glyceraldehyde-3-phosphate dehydrogenase homologues lost no activity. Lactate dehydrogenase homologues from these same species lost 65-100% activity under identical conditions [6,7]. Because of the small losses in malate dehydrogenase activity observed following incubation at pressure, it is difficult to tell whether significant reactivation occurs between release of pressure and activity assessment. However, in other cases where reactivation of pressure-dissociated dimeric enzymes was followed [19,20], reactivation between release of pressure and activity assessment was significant only when the enzyme was more than 50% dissociated. Therefore, it is unlikely that the comparisons of these data are significantly affected by rapid reactivation of the enzyme. The decreased susceptibility of malate dehydrogenase to pressure-inactivation, relative to M nlactate dehydrogenase, may be due to the smaller number of contacts between the subunits (32 per subunit for malate dehydrogenase vs. 58 for lactate dehydrogenase [21]) and the smaller number of subunits in the native molecule (i.e., a dimer vs. a tetramer). Though malate dehydrogenase and lactate dehydrogenase have similar secondary and tertiary structures [21-23], the greater number of intersubunit contacts in lactate dehydrogenase resuits in a greater free volume at the subunit interfaces. Greater free volume results in greater overall compressibility of the molecule, which appears to increase susceptibility to dissociation by hydrostatic pressure [24].

The decreased susceptibility of glyceraldehyde3-phosphate dehydrogenase to inactivation by pressure, relative to lactate dehydrogenase, may also be due to the decreased number of subunit interactions in glyceraldehyde-3-phosphate dehydrogenase (38 vs. 58 for lactate dehydrogenase [21,25]). However, the secondary and tertiary structural differences between glyceraldehyde-3phosphate dehydrogenase and lactate dehydrogenase, particularly at the subunit interfaces [25], are likely to contribute significantly to the different responses of the enzymes to hydrostatic pressure.

Proteolysis Proteolytic inactivation rates of fish malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues are less than those of lactate dehydrogenase homologues from the same species. This suggests that the structural features in lactate dehydrogenase homologues that are responsible for their susceptibility to proteolysis are not found in malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues. Additionally, the effects of hydrostatic pressure on proteolytic inactivation of malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues are much less dramatic than those seen with lactate dehydrogenase homologues from the same species [7]. This could well be a reflection of the susceptibility of the various enzymes to inactivation by hydrostatic pressure. There is a 4-fold difference in the rates of proteolytic inactivation between lactate dehydrogenase homologues of shallow- and deep-living fishes (cf. Table I in Ref. 7). No such difference is seen for the malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues used in this study. Additionally, the effects of hydrostatic pressure on tryptic inactivation of malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues are not related to species depth of abundance, and are much less dramatic than those seen with lactate dehydrogenase homologues from the same species [7]. This suggests that the structural characteristics responsible for increased susceptibility to proteolysis and the conformational perturbations responsible for increased susceptibility to inactivation by trypsin

291

at pressure that are found in lactate dehydrogenase homologues of shallow-living species are not found in rnalate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase homologues of the same species. Thus, it appears that hydrostatic pressure is not as strong an influence on susceptibility to proteolytic inactivation of malate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase as it appears to be for M 4lactate dehydrogenase.

Acknowledgements This research was supported by NSF grant DCB-8201251 and DCB-8596001 to JFS. We thank Drs. G.N. Somero and R.L. Haedrich for help in obtaining specimens of C. rupestris and C. leptolepsis.

References 1 Somero, G.N., Siebenaller, J.F. and Hochachka, P.W. (1983) in the Sea (Rowe, G.T., ed.), Vol. 8, pp. 261-330, Wiley, New York 2 Somero, G.N., Siebenaller, J.F. and Swezey, R.R. (1987) Fed. Proc., in the press 3 Siebenaller, J.F. and Somero, G.N. (1978) Science 201, 255-257 4 Siebenaller, J.F. and Somero, G.N. (1979) J. Comp. Physiol. 129, 295-300 5 Siebenaller, J.F. (1984) J. Comp. Physiol. B154, 443-448 6 Hennessey, J.P., Jr. and Siebenaller, J.F. (1985) J. Comp. Physiol. B 155, 647-652 7 Hennessey, J.P., Jr. and Siebenaller, J.F. (1987) J. Exp. Zool. 241, 9-15

8 Laemmli, U.K. (1970) Nature 227, 680-685 9 Harris, H. and Hopkinson, D.A. (1976) Handbook of Enzyme Electrophoresis in Human Genetics, North-Holland, Amsterdam 10 Bradford, M.M. (1976) Anal. Biochem. 72, 248-254 11 Peterson, G.L. (1983) Methods Enzymol. 91, 95-119 12 Neuman, R.C., Kauzmann, W. and Zipp, A. (1973) J. Phys. Chem. 77, 2687-2691 13 Morild, E. (1977) Biophys. Chem. 6, 351-362 14 Harris, J.I. and Perham, R.N. (1965) J. Mol. Biol. 13, 876-884 15 Harrington, W.F. and Karr, G.M. (1965) J. Mol. Biol. 13, 885-893 16 Banaszak, L.J. and Bradshaw, R.A. (1975) The Enzymes 11, 369-398 17 Wilson, R.R., Jr. and Waples, R.S. (1983) Deep-Sea Res. 30, 1127-1145 18 Wilson, R.R., Jr. and Waples, R.S. (1984) Mar. Biol. 80, 227-237 19 Kornblatt, J., Kornblatt, J. and Hui Bon Hoa, G. (1982) Eur. J. Biochem. 128, 577-581 20 Silva, J.L., Miles, E.W. and Weber, G. (1986) Biochemistry 25, 5780-5786 21 Birktoft, J.J. and Banaszak, L.J. (1983) J. Biol. Chem. 258, 472-482 22 Rossmann, M.G., Liljas, A., Br~lnd6n, C.-I. and Banaszak, L.J. (1975) The Enzymes 11, 61-102 23 Rossmarm, M.G. (1983) in Biological Oxidations (Sund, H. and Ullrieh, V., eds.), pp. 33-54, Springer-Verlag, New York 24 Paladini, A.A. and Weber, G. (1981) Biochemistry 20, 2587-2593 25 Moras, D., Olsen, K.W., Sabesan, M.N., Buehner, M., Ford, G.C. and Rossmann, M.G. (1975) J. Biol. Chem. 250, 9137-9162 26 Siebenaller, J.F., Somero, G.N. and Haedrieh, R.U (1982) Biol. Bull. 163, 240-249 27 Miller, D.J. and Lea, R.N. (1972) Calif. Dept. Fish Game, Fish. Bull. No. 157, 235 pp. (1976) addendum, pp. 236-249