- Email: [email protected]
Enolase isozymes in Coho salmon
('Oral,, Biocht'm Physiol., ~,ol 60B, pp. 383 to 388.
0305-049178/0715-0383502,00/0
(0 Pcr~lamon Pre~s Ltd 1978. Printed iJl Great Britain
ENOLASE ISOZYMES IN COHO SALMON SCOTT R. LANDREY, ROGER APPLEGATE* and JANET M. CARDENAS'I" The Department of Biochemistry & Biophysics, Oregon State University, Corvallis, OR 97331, U.S.A.
(Received 13 October 1977) Abstract 1. Analysis of the enolase isozymic distribution has been performed in tissues of the Coho salmon, using electrophoretic separation on cellulose acetate strips followed by localization of enzymatic activity. 2. A total of six electrophoretically distinct forms are seen in Coho salmon in patterns that differ both qualitatively and quantitatively from one tissue to another. 3. The isozymes in skeletal muscle and liver are sufficiently similar to one another that a purification procedure previously developed for trout muscle enolase by Cory & Wold (1966) can be used to partially purify enolase from either of the above-mentioned Coho tissues. The main form of enolase in Coho muscle has an isoelectric point of 7.57. 4. Both liver and skeletal muscle enolases can be reversibly denatured in guanidine HC1 and subsequently renatured. Liver enolase appeared to renature somewhat faster than muscle enolase under the same conditions. 5. While polyploidy among salmonids may contribute to the complexity of enolase patterns in fish, the differences in isozymic patterns seen from one tissue to another indicate the presence of distinct, nonallelic genes, probably arising through gene duplication.
INTRODUCTION
MATERIALS AND METHODS
Multiple forms of the glycolytic enzyme enolase (E.C. 4.2.1.11) have been k n o w n to exist since M a l m s t r o m (1957) observed them in yeast. Subsequent studies have revealed enolase isozymes in the skeletal muscle of several salmonids (Tsuyuki & Wold, 1964) and in various tissues of rats (Rider & Taylor, 1974, 1975) and h u m a n s ( C h e n & Giblet, 1976; Kamel & Schwarzfischer, 1975; Pearce et al., 1976). Rider & Taylor (1974, 1975) found three electrophoretically distinguishable forms of enolase in rats in isozymic distributions that differ from tissue to tissue. Subsequent studies by Chen & Giblet (1975) and Pearce et al. (1976) revealed at least four different electrophoretic forms of the enzyme in h u m a n tissue extracts, including a major form in muscle that apparently does not occur in other tissues. While the skeletal muscle of salmonids is k n o w n to contain enolase isozymes (Tsuzuki & Wold, 1964; Cory & Wold, 1966; Ruth et al., 1970), the isozymic distribution a m o n g the various tissues of salmonids or, as far as we know, of any other n o n m a m m a l i a n , multicellular organism was unknown. The studies reported in this paper reveal several additional electrophoretic forms of enolase in other tissues of the C o h o salmon that do not occur in muscle. Furthermore, while the isozymic patterns in this fish are much more complicated than those seen in the previously studied mammals, considerable parallelisms in tissue distributions exist a m o n g these animals.
Juvenile Coho salmon ( 5 1 2 in.) (Oncorhynchus kisutch, Fall Creek strain) were obtained from the USDA Fish Toxicology Laboratory in Corvallis, Oregon. Substrates and lactate dehydrogenase were obtained from Sigma Chemical Co., and bovine pyruvate kinase was prepared by the method of Cardenas et al. (1973).
* Present address: MS-II, University of Oregon Health Sciences Center. Portland, OR 97201, U.S.A. t To whom to address all correspondence: Chemistry Department, University of North Carolina, Chapel Hill, NC 27514, U.S.A.
Preparation of tissues Dissected tissues were weighed and homogenized in 2-4 ml of 0,02 M Tris, 1 mM EDTA, 1 mM fl-mercaptoethanol, pH 7.0, per g of tissue. The homogenate was centrifuged for 30min at 18,0000 and samples of the supernatants were dialyzed for 4-6 hr in electrophoresis buffer containing 0.5 M sucrose, 0.02 M Tris, I mM MgCl 2, and l0 mM fl-mercaptoethanol, pH 7.0. Assay of enolase activity Enolase activity was measured spectrophotometrically at 240 nm (Warburg & Christian, 1942) using a Beckman double-beam recording spectrophotometer and 1 ml assay volumes containing 1 mM 2-phosphoglycerate, 0.1 M KCI, 0.05 M imidazole, and 1 mM MgC12, pH 7.0. 1 unit of enzyme activity was taken as that amount catalyzing the conversion of 1/~mole of substrate to product in l rain at 25°C. Electrophoresis Electrophoresis was performed on cellulose acetate strips using equipment and supplies from the Gelman Instrument Co. Sepraphore III cellulose acetate strips were presoaked in the electrophoresis buffer described above, to which was added 1 mg of bovine serum albumin per ml. Dialyzed tissue homogenates containing more than l0 units of enolase activity per ml were diluted to 7-10 units per ml. If more dilute than %10 units per ml, multiple applications of the sample were made at the origin in order to apply approximately the same amount of enzyme activity to each strip. Electrophoresis was performed at 250V for %9 hr 383
384
SCOIT R. LANDREY,ROGER APPLEGATEand JANET M. CAROENAS
at 4 C, using the procedures described by Cardenas & Dyson (1973). Detection c~["the isozyme bands Localization of enolase activity after electrophoresis was accomplished by a modification of the procedure of Susor &Rutter (1971). The electrophoresed strips were placed on a 1",, agarose film containing 2 mM 2-phosphoglycerate, 2mM ADP, 1 mM NADH. 10mM MgCl2, 0.1M KC1, 0.05 M imidazole, pH 7.0, to which was added 5 units per ml each of pyruvate kinase and lactate dehydrogenase. Electrophoretic patterns were recorded by contact photography on Kodak Kodabromide F-5 paper. Enzyme purification Enolases were partially purified from both muscle and liver using procedures similar to those of Cory & Wold (1966) for rainbow trout muscle enolase, with the following modifications. After the acetone fractionation, 10 mM /3-mercaptoethanol was added to the Mg2+-imidazole buffer in which the precipitate was dissolved. Also, the purification was carried only through the second ammonium sulfate fractionation, after which the precipitate was suspended in 709'0 ammonium sulfate. The enzyme was stored at 4 C for weeks with no apparent loss of activity. In each case, this procedure resulted in an approximately 8-fold purification with a 15~, yield, giving final specific activities of 61 and 5 units of enolase activity per mg protein for muscle and liver, respectively. lsoelectric focusinq This was performed using equipment and supplies purchased from LKB Instruments. A 110 ml capacity column was used with a sucrose density gradieni containing 1% ampholyte carrier of pH range 7-10. Isoelectric focusing was carried out for 45 hr at 3°C and 500 V using a circulating constant temperature bath, Fractions of 1 ml volume were collected for pH measurements and enolase assays. Reversible denaturation of enolases Denaturation of crude extracts and of partially purified Coho enolase was accomplished in 2.4 M guanidine HC1 in renaturation buffer (0.05 M Tris, pH 7.5, 0.5 M sucrose, 0.01 M MgC12, 1 mM EDTA, 0.1 M KC1, and 0.05 M /~-mercaptoethanol). After incubating this mixture on ice for 5 min, 20 ttl was diluted into 0.5 ml of the renaturation buffer described above. The final mixture was incubated at I 5 C for up to 23 hr and assayed periodically for enolase activity.
RESULTS Listed in Table 1 are the concentrations of enolase in the various tissues of the Coho salmon, based on enzyme assays of undialyzed extracts. Muscle contained much more enolase activity than did any other tissue, although adipose tissue, liver and heart contained relatively high concentrations of the enzyme. Shown in Fig. 1 are zymograms obtained with tissues from Coho salmon. For a given zymogram, the relative intensities of the bands are approximately proportional to their relative amounts of enzymatic activity. However, comparison of relative intensities from one zymogram to another is less reliable because of differences in exposure and development time. Muscle and adipose tissue revealed a heavy cathodally migrating band plus two lighter bands of lower cathodic mobility. Heart contains at least three bands; furthermore, examination of the electrophoretic patterns early in the development of the
Table 1. Enolase concentrations in various tissues of Coho salmon Tissue Gill Kidney Brain Eye Liver Adipose Muscle Gas bladder Spleen Heart
Enolase activity (units/g wet wt~ l 2.4 17.3 17.7 15.4 39.1 34.9 80.3 2.7 3.0 25.9
Extracts were prepared and assayed as described in the text.
zymograms, before closely spaced bands have diffused together, reveals that the middle band is actually a doublet. Thus, it appears likely that at least four electrophoretically distinguishable forms of enolase exist in heart. Eye and brain contain additional electrophoretic bands, some of which migrate toward the anode under these conditions. In order to more closely compare the isozymic patterns such as those seen in eye and muscle, these two samples were applied to the same electrophoretic strip. As seen in Fig. 1, the cathodal bands of eye had exactly the same mobility as those of muscle, while the anodal bands of eye had distinct electrophoretic mobilities. Thus, the cathodal bands are probably the same in both eye and muscle, while the anodal bands of eye appear to represent additional, noninterconvertible forms. Electrophoresis performed in buffer lacking MgCI2 or made 5 m M in MgCI2 produced qualitatively similar patterns to those shown in Fig. 1 but with somewhat different electrophoretic mobilities. However, the bands were less sharp in the absence of MgCI2, lending support to earlier studies showing that Mg 2÷ stabilized the active forms of the enzyme (Winstead & Wold, 1965; Pearce et al., 1976). Examination of the isozyme patterns before and after purification of liver and muscle enolases revealed no change in their isozyme patterns (results not shown), in agreement with observations of Cory & Wold 0966) during purification of enolase isozymes from rainbow trout skeletal muscle. This evidence, plus the results cited above that mixtures of extracts applied to the same electrophoretic strip produce additive patterns, indicate that the isozymes are not artifacts of isolation and are relatively stable. The mere fact that the same purification procedure can apparently be applied to both muscle and liver enolases with no apparent alteration in their electrophoretic patterns attests to their similarity in structure and overall properties. Isoelectric focusing with partially purified skeletal muscle enolase revealed the isoelectric point of the main form to be 7.57 (see Fig. 2). We were unable to resolve the two minor forms by isoelectric focusing, presumably partly because 95~, of the enzyme activity in muscle is known to occur in one form (Tsuyuki & Wold, 1964; Ruth et al., 1970). The shoulder on
Enolases in Coho salmon
(+)
0 ~
385
(-) Brain
Eye
Kidney
Liver
Gill
Heart
Muscle
Adipose
Eye and Muscle
Fig. 1. Electrophoresis of Coho enolase isozymes on cellulose acetate. Electrophoresis was performed for 8 hr, using procedures described in the Methods section. O denotes the origin. Note the presence of at least six bands in eye extracts, including the one at the very left edge of the zymogram.
the acidic side of the major peak may well represent the other two isozymes found in muscle. Shown in Fig. 3 are the renaturation curves for muscle and liver enolases after their denaturation in guanidine HC1. Regain of muscle enolase activity occurred very slowly, with only 31~o of the initial activity recovered after 22.5 hr. Liver enolase renaturation proceeded more rapidly, with 64~o of the initial enzyme activity recovered by 3.5 hr. The recovery of enzymatic activity after denaturation of these enzymes in guanidine HCI indicates that the polypeptide chains probably contain all the information required
for their proper folding and thus may not have undergone extensive post-translational modification.
DISCUSSION
The zymograms of Fig. 1 contain evidence for a total of at least six electrophoretically distinguishable forms of enolase in Coho salmon. That all of these bands are actually enolase and not a phosphatase was shown by omitting either 2-phosphoglycerate or ADP from the agarose films used to detect enolase activity
386
SCOTT R. LANDREY, ROGER APPLEGATE a n d JANET M . CARDENAS
I0,0 ~
°~°~o
pH 757
°~°~o
i 9.5-
~,°__°
\.
9.oi
~°
15
~.
f71
I
"<',~
i
i !
8.5 ! Z d~
°~.~.
I
°~,
E ,°
7.5 i
B 7.0j
I/ 6'5 L 25
_
L 75
50
! IO0
/ 125
150
Fraction Numbe~ Fig. 2. Isoelectric focusing pattern of partially purified skeletal muscle enolase from Coho salmon. Open circles denote enolase activity, while closed circles indicate the pH of the fractions, measured at 3:C. Experimental procedures are given in the Methods section.
.2O !
.15i E
.10 >
,d
.05
50
I00
150
200
Time (rain) Fig. 3. Renaturation of partially purified enolase from Coho skeletal muscle (circles) and liver (triangles) enolases after denaturation in 2.4 M guanidine HC1. Experimental procedures are given in the Methods section. For the particular experiments illustrated here, 0.25 ml of liver enolase, 9.1 units per ml, was mixed with 0.13 ml of 7 M guanidine HCI solution at O°C. After 5 min, 0.25 ml of the enolase in guanidine HCI was diluted with gentle swirling into 0.475 ml renaturation buffer and incubated at 16°C. For the renaturation of muscle enolase, 0.28 ml of enzyme, 153 units per mi, was mixed with 0.10ml of 7 M guanidine HCI and incubated on ice for 5 min, after which time 0.01 ml was mixed with gentle swirling into 0.5 ml of renaturation buffer and incubated at 16°C.
Enolases in Coho salmon after, their electrophoretic separation; no bands appeared under these conditions. Muscle and liver appear to have the same three bands, although these differed in relative intensity, indicating probable differences in their relative quantities of enzyme activity. In muscle, the most cathodal band has the greatest amount of activity, while in liver the band of lowest cathodal mobility contains the greatest amount of enzyme activity. Similarly, in kidney the band having the greatest amount of activity is that with the lowest cathodal migration. As in humans (Pearce et al., 1976), the middle band in Coho salmon heart is the heaviest. By comparison, Rider & Taylor (1974, 1975) observed enolase 1 (the most anodal form seen in rat tissues) in liver, spleen, brain, blood, adipose tissue and kidney, whereas enolase 3 was found in skeletal muscle. Both enolases 1 and 3, as well as a form having intermediate electrophoretic mobility, were found in heart. Subsequent work by Chen & Giblett (1975) and Pearce et al. (1976) revealed at least four different electrophoretic forms of the enzyme in human tissue, which led the latter workers to conclude that three distinct genetic loci give rise to functional homo- and heterotetramers. Coho salmon contain a greater number of forms of enolase than have been reported for mammals. Brains from humans and rats each contain three electrophoretic forms (Kamel & Schwarzfischer, 1975; Chen et al., 1975; Pearce et al., 1976). Similarly, brains from Coho salmon contain three heavy bands of activity. Three additional, lighter bands can generally be seen in the original zymograms but are too faint to photograph well. Certainly the most complex pattern seen in Coho salmon tissue is that of the eye, where at least six bands of enolase activity can be detected. Analysis of isozymic patterns in Coho salmon is somewhat complicated by the fact that salmonids are thought to be tetraploid (Ohno et al., 1968; Bailey et al., 1969). While polyploidy may contribute somewhat to the complexity of the enolase patterns, particularly regarding closely spaced doublets, i.e. in heart, we feel that most of the electrophoretic forms seen here are the result of at least three nonallelic genes giving rise to polypeptide chains that combine randomly to form dimers. The most convincing evidence for a role of nonallelic genes is the differential patterns seen from one tissue to another. For instance, muscle contains mainly, or only, the three cathodal bands; brain contains at least four bands, the three main ones of which are centered around the origin, while the total of six bands seen in eye could be produced by all possible combinations of three subunit types to form dimers. This analysis is made on the assumption that all electrophoretic forms of the Coho salmon, like those of skeletal muscle, are dimers (Ruth et al., 1970). Verification of this assumption must await the isolation and characterization of these additional forms. Thus, isozymic differences from one organ to another are apparent in the Coho salmon as well as in both rats and humans, strengthening the possibility of a physiological significance for multiple forms of this enzyme. It is interesting to note that the largest number of electrophoretic forms of enolase occurs in
387
the eye, a tissue known to contain cells of varying developmental levels. In our future studies, we hope to determine whether salmon, like rats (Rider & Taylor, 1975), undergo alterations in the enolase isozyme patterns during development. The limited comparative data now available on enolase isozyme distributions suggests some parallelisms from one animal to another, an observation that would lend support to a regulatory or functional role of the multiple forms of this enzyme. An interesting possible reason for the occurrence of enolase isozymes is given by the work of Oh & Brewer (1973), who found substrate inhibition of swine kidney enolase at much lower concentrations of 2-phosphoglycerate than is found for mammalian muscle enolases: this inhibition could serve to increase 2- and 3-phosphoglycerate concentrations, which in turn would enhance glucose synthesis in kidney, an organ known to be gluconeogenic. Another recent study has shown that enolases, along with a number of other glycolytic enzymes, bind specifically to muscle proteins (Clarke & Masters, 1975). Enolase isozymes, then, could be important for differential binding to particular tissue components. Clearly, further characterization of enolase isozymes is needed to define their metabolic roles. Acknowledyements--Appreciation is extended to Dr Helga Guderley for valuable discussions during the course of this work, and to the U.S. Public Health Service for financial support through their grant No. RR07079. REFERENCES
BAILEY G. S., COCKS G. T. & WILSON A. C. (1969) Gene duplication in fishes: malate dehydrogenases in salmon and trout. Biochem. biophys. Res. Commun. 34, 605 612. CARDENASJ. M. & DVSON R. D. (1973) Bovine pyruvate kinases II. Purification of the liver isozyme and its hybridization with skeletal muscle pyruvate kinase. J. biol. Chem. 248, 6938-6944. CARDENASJ. M., DYSON R. D. & STRANDHOLMJ. J. (1973) Bovine pyruvate kinases I. Purification and characterization of the skeletal muscle isozyme. J. biol. Chem. 248, 6931 6937. CHEN S. H. & GmLETT E. R. (1976) Enolase: human tissue distribution and evidence for three different loci. Ann. hum. Genet. 39, 277-280. CLARKE F. M. • MASTERSC. J. (1975) On the association of glycolytic enzymes with structural proteins of skeletal muscle. Biochim. biophys. Acta 381, 37 46. CORY R. P. & WOLD F. (1966) Isolation and characterization of enolase from rainbow trout (Salmo ,qairdnerii 9airdnerii). Biochemistry 5, 3131-3137. KAMEL R. & SCHWARZFISCHERF. (1975) Multiple forms of enolase (E.C. 4.2.1.11): Their distribution in human tissues. Humangenetik 28, 259-261. MALMSTROMB. G. (1957) The purification of yeast enolase by zone electrophoresis and ion-exchange chromatography, and the existence of several forms of the enzyme. Archs Biochem. Biophys. 70, 58 69. OH S. K. & BREWERJ. M. (1973) Purification and properties of enolase from swine kidney. Archs Biochem. BiDphys. 157, 491~499. OHNO S., WOLF U. & ATKIN N. B. 0968) Evolution from fish to mammals by gene duplication. Hereditas 59, 169-187. PEARCEJ. M., EDWARDSY. H. & HARRISH. (1976) Human enolase isozymes: electrophoretic and biochemical evidence for three loci. Ann hum. Genet. 39, 263-276. RIDER C. C. & TAYLOR C. B. (1974) Enolase isoenzymes
388
SCOTT R. LANDREY. ROGER APPLEGATEand JANET M. CARDENAS
in rat tissues. Electrophoretic, chromatographic, immunological, and kinetic properties. Biochim. biophys. Acta 365, 285 300. RrDER C. C. & TAYLOR C. B. (1975) Enolase isoenzymes. II. Hybridization studies, developmental and phylogenetic aspects. Biochim. biophys. Acta 405, 175~-187. RUTH R. C., SOJA D. M. & WOLD F. (1970) Purification and characterization of enolases from coho (Oncorhynchus kisutch) and chum (Oncorhynchus keta) salmon. Archs. Biochem. Biophys. 140, 1-~10. SUSOR W. A. & RUTTER W. T. (1971) A method for the
detection of pyruvate kinase, aldolase, and other pyridine nucleotide linked enzyme activities after electrophoresis. Analyt. Biochem. 43, 147-155. TSUYUKZH. & WOLD F. (1964) Enolase: multiple molecular forms in fish muscle. Science 146, 535-537. WAR,URn:; O. & CHRISTIAN W. (1942) Isolierung und kristallization des garungsferments enolase. Biochem. Z. 310, 384~425. WINSTEAD J. A. & WOLD F. (1965) Dissociation and reassociation of rabbit muscle enolase. Biochemistry 4~ 2145.2158.
0305-049178/0715-0383502,00/0
(0 Pcr~lamon Pre~s Ltd 1978. Printed iJl Great Britain
ENOLASE ISOZYMES IN COHO SALMON SCOTT R. LANDREY, ROGER APPLEGATE* and JANET M. CARDENAS'I" The Department of Biochemistry & Biophysics, Oregon State University, Corvallis, OR 97331, U.S.A.
(Received 13 October 1977) Abstract 1. Analysis of the enolase isozymic distribution has been performed in tissues of the Coho salmon, using electrophoretic separation on cellulose acetate strips followed by localization of enzymatic activity. 2. A total of six electrophoretically distinct forms are seen in Coho salmon in patterns that differ both qualitatively and quantitatively from one tissue to another. 3. The isozymes in skeletal muscle and liver are sufficiently similar to one another that a purification procedure previously developed for trout muscle enolase by Cory & Wold (1966) can be used to partially purify enolase from either of the above-mentioned Coho tissues. The main form of enolase in Coho muscle has an isoelectric point of 7.57. 4. Both liver and skeletal muscle enolases can be reversibly denatured in guanidine HC1 and subsequently renatured. Liver enolase appeared to renature somewhat faster than muscle enolase under the same conditions. 5. While polyploidy among salmonids may contribute to the complexity of enolase patterns in fish, the differences in isozymic patterns seen from one tissue to another indicate the presence of distinct, nonallelic genes, probably arising through gene duplication.
INTRODUCTION
MATERIALS AND METHODS
Multiple forms of the glycolytic enzyme enolase (E.C. 4.2.1.11) have been k n o w n to exist since M a l m s t r o m (1957) observed them in yeast. Subsequent studies have revealed enolase isozymes in the skeletal muscle of several salmonids (Tsuyuki & Wold, 1964) and in various tissues of rats (Rider & Taylor, 1974, 1975) and h u m a n s ( C h e n & Giblet, 1976; Kamel & Schwarzfischer, 1975; Pearce et al., 1976). Rider & Taylor (1974, 1975) found three electrophoretically distinguishable forms of enolase in rats in isozymic distributions that differ from tissue to tissue. Subsequent studies by Chen & Giblet (1975) and Pearce et al. (1976) revealed at least four different electrophoretic forms of the enzyme in h u m a n tissue extracts, including a major form in muscle that apparently does not occur in other tissues. While the skeletal muscle of salmonids is k n o w n to contain enolase isozymes (Tsuzuki & Wold, 1964; Cory & Wold, 1966; Ruth et al., 1970), the isozymic distribution a m o n g the various tissues of salmonids or, as far as we know, of any other n o n m a m m a l i a n , multicellular organism was unknown. The studies reported in this paper reveal several additional electrophoretic forms of enolase in other tissues of the C o h o salmon that do not occur in muscle. Furthermore, while the isozymic patterns in this fish are much more complicated than those seen in the previously studied mammals, considerable parallelisms in tissue distributions exist a m o n g these animals.
Juvenile Coho salmon ( 5 1 2 in.) (Oncorhynchus kisutch, Fall Creek strain) were obtained from the USDA Fish Toxicology Laboratory in Corvallis, Oregon. Substrates and lactate dehydrogenase were obtained from Sigma Chemical Co., and bovine pyruvate kinase was prepared by the method of Cardenas et al. (1973).
* Present address: MS-II, University of Oregon Health Sciences Center. Portland, OR 97201, U.S.A. t To whom to address all correspondence: Chemistry Department, University of North Carolina, Chapel Hill, NC 27514, U.S.A.
Preparation of tissues Dissected tissues were weighed and homogenized in 2-4 ml of 0,02 M Tris, 1 mM EDTA, 1 mM fl-mercaptoethanol, pH 7.0, per g of tissue. The homogenate was centrifuged for 30min at 18,0000 and samples of the supernatants were dialyzed for 4-6 hr in electrophoresis buffer containing 0.5 M sucrose, 0.02 M Tris, I mM MgCl 2, and l0 mM fl-mercaptoethanol, pH 7.0. Assay of enolase activity Enolase activity was measured spectrophotometrically at 240 nm (Warburg & Christian, 1942) using a Beckman double-beam recording spectrophotometer and 1 ml assay volumes containing 1 mM 2-phosphoglycerate, 0.1 M KCI, 0.05 M imidazole, and 1 mM MgC12, pH 7.0. 1 unit of enzyme activity was taken as that amount catalyzing the conversion of 1/~mole of substrate to product in l rain at 25°C. Electrophoresis Electrophoresis was performed on cellulose acetate strips using equipment and supplies from the Gelman Instrument Co. Sepraphore III cellulose acetate strips were presoaked in the electrophoresis buffer described above, to which was added 1 mg of bovine serum albumin per ml. Dialyzed tissue homogenates containing more than l0 units of enolase activity per ml were diluted to 7-10 units per ml. If more dilute than %10 units per ml, multiple applications of the sample were made at the origin in order to apply approximately the same amount of enzyme activity to each strip. Electrophoresis was performed at 250V for %9 hr 383
384
SCOIT R. LANDREY,ROGER APPLEGATEand JANET M. CAROENAS
at 4 C, using the procedures described by Cardenas & Dyson (1973). Detection c~["the isozyme bands Localization of enolase activity after electrophoresis was accomplished by a modification of the procedure of Susor &Rutter (1971). The electrophoresed strips were placed on a 1",, agarose film containing 2 mM 2-phosphoglycerate, 2mM ADP, 1 mM NADH. 10mM MgCl2, 0.1M KC1, 0.05 M imidazole, pH 7.0, to which was added 5 units per ml each of pyruvate kinase and lactate dehydrogenase. Electrophoretic patterns were recorded by contact photography on Kodak Kodabromide F-5 paper. Enzyme purification Enolases were partially purified from both muscle and liver using procedures similar to those of Cory & Wold (1966) for rainbow trout muscle enolase, with the following modifications. After the acetone fractionation, 10 mM /3-mercaptoethanol was added to the Mg2+-imidazole buffer in which the precipitate was dissolved. Also, the purification was carried only through the second ammonium sulfate fractionation, after which the precipitate was suspended in 709'0 ammonium sulfate. The enzyme was stored at 4 C for weeks with no apparent loss of activity. In each case, this procedure resulted in an approximately 8-fold purification with a 15~, yield, giving final specific activities of 61 and 5 units of enolase activity per mg protein for muscle and liver, respectively. lsoelectric focusinq This was performed using equipment and supplies purchased from LKB Instruments. A 110 ml capacity column was used with a sucrose density gradieni containing 1% ampholyte carrier of pH range 7-10. Isoelectric focusing was carried out for 45 hr at 3°C and 500 V using a circulating constant temperature bath, Fractions of 1 ml volume were collected for pH measurements and enolase assays. Reversible denaturation of enolases Denaturation of crude extracts and of partially purified Coho enolase was accomplished in 2.4 M guanidine HC1 in renaturation buffer (0.05 M Tris, pH 7.5, 0.5 M sucrose, 0.01 M MgC12, 1 mM EDTA, 0.1 M KC1, and 0.05 M /~-mercaptoethanol). After incubating this mixture on ice for 5 min, 20 ttl was diluted into 0.5 ml of the renaturation buffer described above. The final mixture was incubated at I 5 C for up to 23 hr and assayed periodically for enolase activity.
RESULTS Listed in Table 1 are the concentrations of enolase in the various tissues of the Coho salmon, based on enzyme assays of undialyzed extracts. Muscle contained much more enolase activity than did any other tissue, although adipose tissue, liver and heart contained relatively high concentrations of the enzyme. Shown in Fig. 1 are zymograms obtained with tissues from Coho salmon. For a given zymogram, the relative intensities of the bands are approximately proportional to their relative amounts of enzymatic activity. However, comparison of relative intensities from one zymogram to another is less reliable because of differences in exposure and development time. Muscle and adipose tissue revealed a heavy cathodally migrating band plus two lighter bands of lower cathodic mobility. Heart contains at least three bands; furthermore, examination of the electrophoretic patterns early in the development of the
Table 1. Enolase concentrations in various tissues of Coho salmon Tissue Gill Kidney Brain Eye Liver Adipose Muscle Gas bladder Spleen Heart
Enolase activity (units/g wet wt~ l 2.4 17.3 17.7 15.4 39.1 34.9 80.3 2.7 3.0 25.9
Extracts were prepared and assayed as described in the text.
zymograms, before closely spaced bands have diffused together, reveals that the middle band is actually a doublet. Thus, it appears likely that at least four electrophoretically distinguishable forms of enolase exist in heart. Eye and brain contain additional electrophoretic bands, some of which migrate toward the anode under these conditions. In order to more closely compare the isozymic patterns such as those seen in eye and muscle, these two samples were applied to the same electrophoretic strip. As seen in Fig. 1, the cathodal bands of eye had exactly the same mobility as those of muscle, while the anodal bands of eye had distinct electrophoretic mobilities. Thus, the cathodal bands are probably the same in both eye and muscle, while the anodal bands of eye appear to represent additional, noninterconvertible forms. Electrophoresis performed in buffer lacking MgCI2 or made 5 m M in MgCI2 produced qualitatively similar patterns to those shown in Fig. 1 but with somewhat different electrophoretic mobilities. However, the bands were less sharp in the absence of MgCI2, lending support to earlier studies showing that Mg 2÷ stabilized the active forms of the enzyme (Winstead & Wold, 1965; Pearce et al., 1976). Examination of the isozyme patterns before and after purification of liver and muscle enolases revealed no change in their isozyme patterns (results not shown), in agreement with observations of Cory & Wold 0966) during purification of enolase isozymes from rainbow trout skeletal muscle. This evidence, plus the results cited above that mixtures of extracts applied to the same electrophoretic strip produce additive patterns, indicate that the isozymes are not artifacts of isolation and are relatively stable. The mere fact that the same purification procedure can apparently be applied to both muscle and liver enolases with no apparent alteration in their electrophoretic patterns attests to their similarity in structure and overall properties. Isoelectric focusing with partially purified skeletal muscle enolase revealed the isoelectric point of the main form to be 7.57 (see Fig. 2). We were unable to resolve the two minor forms by isoelectric focusing, presumably partly because 95~, of the enzyme activity in muscle is known to occur in one form (Tsuyuki & Wold, 1964; Ruth et al., 1970). The shoulder on
Enolases in Coho salmon
(+)
0 ~
385
(-) Brain
Eye
Kidney
Liver
Gill
Heart
Muscle
Adipose
Eye and Muscle
Fig. 1. Electrophoresis of Coho enolase isozymes on cellulose acetate. Electrophoresis was performed for 8 hr, using procedures described in the Methods section. O denotes the origin. Note the presence of at least six bands in eye extracts, including the one at the very left edge of the zymogram.
the acidic side of the major peak may well represent the other two isozymes found in muscle. Shown in Fig. 3 are the renaturation curves for muscle and liver enolases after their denaturation in guanidine HC1. Regain of muscle enolase activity occurred very slowly, with only 31~o of the initial activity recovered after 22.5 hr. Liver enolase renaturation proceeded more rapidly, with 64~o of the initial enzyme activity recovered by 3.5 hr. The recovery of enzymatic activity after denaturation of these enzymes in guanidine HCI indicates that the polypeptide chains probably contain all the information required
for their proper folding and thus may not have undergone extensive post-translational modification.
DISCUSSION
The zymograms of Fig. 1 contain evidence for a total of at least six electrophoretically distinguishable forms of enolase in Coho salmon. That all of these bands are actually enolase and not a phosphatase was shown by omitting either 2-phosphoglycerate or ADP from the agarose films used to detect enolase activity
386
SCOTT R. LANDREY, ROGER APPLEGATE a n d JANET M . CARDENAS
I0,0 ~
°~°~o
pH 757
°~°~o
i 9.5-
~,°__°
\.
9.oi
~°
15
~.
f71
I
"<',~
i
i !
8.5 ! Z d~
°~.~.
I
°~,
E ,°
7.5 i
B 7.0j
I/ 6'5 L 25
_
L 75
50
! IO0
/ 125
150
Fraction Numbe~ Fig. 2. Isoelectric focusing pattern of partially purified skeletal muscle enolase from Coho salmon. Open circles denote enolase activity, while closed circles indicate the pH of the fractions, measured at 3:C. Experimental procedures are given in the Methods section.
.2O !
.15i E
.10 >
,d
.05
50
I00
150
200
Time (rain) Fig. 3. Renaturation of partially purified enolase from Coho skeletal muscle (circles) and liver (triangles) enolases after denaturation in 2.4 M guanidine HC1. Experimental procedures are given in the Methods section. For the particular experiments illustrated here, 0.25 ml of liver enolase, 9.1 units per ml, was mixed with 0.13 ml of 7 M guanidine HCI solution at O°C. After 5 min, 0.25 ml of the enolase in guanidine HCI was diluted with gentle swirling into 0.475 ml renaturation buffer and incubated at 16°C. For the renaturation of muscle enolase, 0.28 ml of enzyme, 153 units per mi, was mixed with 0.10ml of 7 M guanidine HCI and incubated on ice for 5 min, after which time 0.01 ml was mixed with gentle swirling into 0.5 ml of renaturation buffer and incubated at 16°C.
Enolases in Coho salmon after, their electrophoretic separation; no bands appeared under these conditions. Muscle and liver appear to have the same three bands, although these differed in relative intensity, indicating probable differences in their relative quantities of enzyme activity. In muscle, the most cathodal band has the greatest amount of activity, while in liver the band of lowest cathodal mobility contains the greatest amount of enzyme activity. Similarly, in kidney the band having the greatest amount of activity is that with the lowest cathodal migration. As in humans (Pearce et al., 1976), the middle band in Coho salmon heart is the heaviest. By comparison, Rider & Taylor (1974, 1975) observed enolase 1 (the most anodal form seen in rat tissues) in liver, spleen, brain, blood, adipose tissue and kidney, whereas enolase 3 was found in skeletal muscle. Both enolases 1 and 3, as well as a form having intermediate electrophoretic mobility, were found in heart. Subsequent work by Chen & Giblett (1975) and Pearce et al. (1976) revealed at least four different electrophoretic forms of the enzyme in human tissue, which led the latter workers to conclude that three distinct genetic loci give rise to functional homo- and heterotetramers. Coho salmon contain a greater number of forms of enolase than have been reported for mammals. Brains from humans and rats each contain three electrophoretic forms (Kamel & Schwarzfischer, 1975; Chen et al., 1975; Pearce et al., 1976). Similarly, brains from Coho salmon contain three heavy bands of activity. Three additional, lighter bands can generally be seen in the original zymograms but are too faint to photograph well. Certainly the most complex pattern seen in Coho salmon tissue is that of the eye, where at least six bands of enolase activity can be detected. Analysis of isozymic patterns in Coho salmon is somewhat complicated by the fact that salmonids are thought to be tetraploid (Ohno et al., 1968; Bailey et al., 1969). While polyploidy may contribute somewhat to the complexity of the enolase patterns, particularly regarding closely spaced doublets, i.e. in heart, we feel that most of the electrophoretic forms seen here are the result of at least three nonallelic genes giving rise to polypeptide chains that combine randomly to form dimers. The most convincing evidence for a role of nonallelic genes is the differential patterns seen from one tissue to another. For instance, muscle contains mainly, or only, the three cathodal bands; brain contains at least four bands, the three main ones of which are centered around the origin, while the total of six bands seen in eye could be produced by all possible combinations of three subunit types to form dimers. This analysis is made on the assumption that all electrophoretic forms of the Coho salmon, like those of skeletal muscle, are dimers (Ruth et al., 1970). Verification of this assumption must await the isolation and characterization of these additional forms. Thus, isozymic differences from one organ to another are apparent in the Coho salmon as well as in both rats and humans, strengthening the possibility of a physiological significance for multiple forms of this enzyme. It is interesting to note that the largest number of electrophoretic forms of enolase occurs in
387
the eye, a tissue known to contain cells of varying developmental levels. In our future studies, we hope to determine whether salmon, like rats (Rider & Taylor, 1975), undergo alterations in the enolase isozyme patterns during development. The limited comparative data now available on enolase isozyme distributions suggests some parallelisms from one animal to another, an observation that would lend support to a regulatory or functional role of the multiple forms of this enzyme. An interesting possible reason for the occurrence of enolase isozymes is given by the work of Oh & Brewer (1973), who found substrate inhibition of swine kidney enolase at much lower concentrations of 2-phosphoglycerate than is found for mammalian muscle enolases: this inhibition could serve to increase 2- and 3-phosphoglycerate concentrations, which in turn would enhance glucose synthesis in kidney, an organ known to be gluconeogenic. Another recent study has shown that enolases, along with a number of other glycolytic enzymes, bind specifically to muscle proteins (Clarke & Masters, 1975). Enolase isozymes, then, could be important for differential binding to particular tissue components. Clearly, further characterization of enolase isozymes is needed to define their metabolic roles. Acknowledyements--Appreciation is extended to Dr Helga Guderley for valuable discussions during the course of this work, and to the U.S. Public Health Service for financial support through their grant No. RR07079. REFERENCES
BAILEY G. S., COCKS G. T. & WILSON A. C. (1969) Gene duplication in fishes: malate dehydrogenases in salmon and trout. Biochem. biophys. Res. Commun. 34, 605 612. CARDENASJ. M. & DVSON R. D. (1973) Bovine pyruvate kinases II. Purification of the liver isozyme and its hybridization with skeletal muscle pyruvate kinase. J. biol. Chem. 248, 6938-6944. CARDENASJ. M., DYSON R. D. & STRANDHOLMJ. J. (1973) Bovine pyruvate kinases I. Purification and characterization of the skeletal muscle isozyme. J. biol. Chem. 248, 6931 6937. CHEN S. H. & GmLETT E. R. (1976) Enolase: human tissue distribution and evidence for three different loci. Ann. hum. Genet. 39, 277-280. CLARKE F. M. • MASTERSC. J. (1975) On the association of glycolytic enzymes with structural proteins of skeletal muscle. Biochim. biophys. Acta 381, 37 46. CORY R. P. & WOLD F. (1966) Isolation and characterization of enolase from rainbow trout (Salmo ,qairdnerii 9airdnerii). Biochemistry 5, 3131-3137. KAMEL R. & SCHWARZFISCHERF. (1975) Multiple forms of enolase (E.C. 4.2.1.11): Their distribution in human tissues. Humangenetik 28, 259-261. MALMSTROMB. G. (1957) The purification of yeast enolase by zone electrophoresis and ion-exchange chromatography, and the existence of several forms of the enzyme. Archs Biochem. Biophys. 70, 58 69. OH S. K. & BREWERJ. M. (1973) Purification and properties of enolase from swine kidney. Archs Biochem. BiDphys. 157, 491~499. OHNO S., WOLF U. & ATKIN N. B. 0968) Evolution from fish to mammals by gene duplication. Hereditas 59, 169-187. PEARCEJ. M., EDWARDSY. H. & HARRISH. (1976) Human enolase isozymes: electrophoretic and biochemical evidence for three loci. Ann hum. Genet. 39, 263-276. RIDER C. C. & TAYLOR C. B. (1974) Enolase isoenzymes
388
SCOTT R. LANDREY. ROGER APPLEGATEand JANET M. CARDENAS
in rat tissues. Electrophoretic, chromatographic, immunological, and kinetic properties. Biochim. biophys. Acta 365, 285 300. RrDER C. C. & TAYLOR C. B. (1975) Enolase isoenzymes. II. Hybridization studies, developmental and phylogenetic aspects. Biochim. biophys. Acta 405, 175~-187. RUTH R. C., SOJA D. M. & WOLD F. (1970) Purification and characterization of enolases from coho (Oncorhynchus kisutch) and chum (Oncorhynchus keta) salmon. Archs. Biochem. Biophys. 140, 1-~10. SUSOR W. A. & RUTTER W. T. (1971) A method for the
detection of pyruvate kinase, aldolase, and other pyridine nucleotide linked enzyme activities after electrophoresis. Analyt. Biochem. 43, 147-155. TSUYUKZH. & WOLD F. (1964) Enolase: multiple molecular forms in fish muscle. Science 146, 535-537. WAR,URn:; O. & CHRISTIAN W. (1942) Isolierung und kristallization des garungsferments enolase. Biochem. Z. 310, 384~425. WINSTEAD J. A. & WOLD F. (1965) Dissociation and reassociation of rabbit muscle enolase. Biochemistry 4~ 2145.2158.