THE JOURNAL OF BIOLOGICAL CHEMISTRY (e 1987 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 262. No. 33. Issue of November 25, pp. 16011-16019,1987 Printed in U.S.A.

Enzyme Termini of a Phosphocreatine Shuttle PURIFICATION AND CHARACTERIZATION OF TWO CREATINE KINASE ISOZYMES FROM SEA URCHIN SPERM*

(Received for publication, December 31, 1986)

Robert M. Tombesz and Bennett M. ShapiroQ From the Department of Bwchemistv, University of Washington, Seattle, Washington 98195

Two isozymes of creatine kinase have been purified from sperm of the sea urchin, Strongylocentrotuspur- puratus. One isozyme was purified from the sperm flagellum, and the other from the head. Both require nonionic detergent for extraction from sperm. The fla- gellar isozyme is a monomeric species with an M, of 145,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and 126,000 from sucrose density gra- dient and gel filtration analyses. Creatine kinase from sperm heads was localized to the mitochondrion by an antibody raised against mouse muscle creatine kinase. This purified mitochondrial isozyme is multimeric, with an M, of 47,000 by sodium dodecyl sulfate-poly- acrylamide gel electrophoresis, but 240,000 for the native enzyme. Peptide mapping indicates that the two isozymes are not related.

The following kinetic characteristics were observed for the purified flagellar and mitochondrial isozymes, respectively. In the direction of ATP formation, at pH 6.6 and 25 “C, specific activities were 235 and 180 unitslmg; pH optima were 6.7 and 6.9 and Michaelis constants were 0.13 and 0.055 mM for ADP and 5.8 and 2.7 mM for phosphocreatine. In the direction of phosphocreatine formation, at pH 7.5 and 25 “C, spe- cific activities were 29 and 47 unitstmg; pH optima were 7.5 and 7.7 and Michaelis constants were 0.89 and 0.31 mM for ATP and 39 and 62 mM for creatine. These unique isozymes constitute the termini of the phosphocreatine shuttle of sea urchin sperm that is responsible for energy transport from the mitochon- drion to the distal flagellum (Tombes, R. M., and Shap- iro, B. M. (1985) Cell 41, 325-334; Tombes, R. M., Brokaw, C. J., and Shapiro, B. M. (1987) Biophys. J. , 52, 75-86).

The reversible transfer of high energy phosphate (-P) between ATP and phosphocreatine by creatine kinase not only maintains -P levels throughout the cell but also may direct energy to specific cellular locations through a phospho- creatine shuttle (reviewed in Bessman and Carpenter, 1985). The discovery and characterization of both mitochondrial (Blum et al., 1983; Grace et al., 1983; Hall et al., 1979; Jacobs et al., 1964; Jacobus and Lehninger, 1973) and myofibrillar or cytoplasmic isozymes of creatine kinase (Turner et al., 1973;

* This work was supported by National Institutes of Health Grants GM23910 and HDO 7183-06. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Wisconsin, Madison, W1 53706. $ Present address: Laboratory of Molecular Biology, University of

8 To whom correspondence should be addressed.

Walliman et al., 1977) suggested the existence of a phospho- creatine shuttle. Indirect evidence supporting this energy transport mechanism came from studies of creatine kinase inhibition (Carpenter et al., 1983; Gerken and Schlette, 1968; Infante and Davies, 1965; Yang and Dubick, 1971), measure- ments of creatine kinase kinetics, and the intracellular com- partmentation of -P (Bessman et al., 1980; Bessman and Carpenter, 1985; Erickson-Viitanen et al., 1982a, 1982b; Gel- lerich and Saks, 1982; Hall et d., 1979; Jacobus and Lehnin- ger, 1973; Moreadith and Jacobus, 1982; Nunnally and Hollis, 1979; Saks et al., 1975, 1976, 1980; Savabi et al., 1983; Walli- man et al., 1984; Yang et al., 1977).

Direct evidence supporting creatine kinase-dependent en- ergy transport has recently come from studies on sea urchin sperm, functionally and energetically polarized cells in which substantial amounts of phosphocreatine (Christen et at, 198313; Winkler et al., 1982; Yanagisawa, 1959a, 1967) and creatine kinase (Tombes and Shapiro, 1985) are present. The motile sperm consumes energy primarily for flagellar motion, and ATP is produced by mitochondrial respiration (reviewed in Shapiro and Tombes, 1985). Sperm have distinct popula- tions of creatine kinase in heads and tails. When creatine kinase activity is inhibited in live sperm with 1-fluoro-2,4- dinitrobenzene (FDNB),’ respiration is impaired and motility patterns are altered, exactly as predicted if a phosphocreatine shuttle mediating energy transport between mitochondria and tail had been inhibited (Tombes and Shapiro, 1985; Tombes et al., 1987). This suggests that there may be two creatine kinase isozymes in sperm, as in muscle, one at the mitochon- drion to catalyze the synthesis of phosphocreatine and the other along the flagellum to catalyze the synthesis of ATP from diffusing phosphocreatine.

This paper describes the purification of creatine kinase isozymes from sea urchin sperm heads and tails, the localiza- tion of the head isozyme to the mitochondrion, and some physical and kinetic characteristics of these isozymes. We propose that they function as the two termini of a phospho- creatine shuttle system to ensure optimal mitochondrial ATP generation and -P transport from mitochondrion to flagel- lum.

The abbreviations used are: FDNB, l-fluoro-2,4-dinitrobenzene; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid Hepes, 4- (2-hydroxyethy1)-1-piperazineethanesulfonic acid; MES, 2-(N-rnor- pho1ino)ethanesulfonic acid MFSW, Millipore-filtered sea water; PAGE, polyacrylamide gel electrophoresis; Pi, inorganic phosphate; SDS, sodium dodecyl sulfate; DTT, dithiothreitol; CrK, creatine kinase, in Miniprint; CrK-MM, mouse muscle creatine kinase; NP- 40, Nonidet P-40; FPLC fast protein liquid chromatography; AMPPNP, adenosine 5’-(@-y-imino)triphosphate.

16011

16012 Sea Urchin Sperm Creatine Kinase Isozymes

1 2 3 4 5 6 7 8 9H)t(12131415

+200

+ 116 + 94 + 68

e? + 45

+ 31

+ 21 + 14

+ FRONT

FIG. 2. SDS-PAGE of sperm creatine kinase fractions. All lanes are from the same gel. Lanes I , 8, and 14 are fluorograms of whole sperm, heads and tails, respectively, labeled with 10 PM ['HI FDNB. The remaining lanes are stained for protein with Coomassie Blue and represent: lane 2, whole sperm; lane 3, heads; lane 4, heads, 100,000 X g supernatant of Nonidet P-40 extract; lane 5, heads, Procion Red peak; lane 6, heads, KCl/glycerol-free pellet; lane 7, heads, DEAE peak (-2 pg); lane 9, mouse muscle creatine kinase (1.5 pg); lane 10, tails; lane 1 1 , tails, 100,000 X g supernatant of Nonidet P-40 extract; lane 12, tails, Procion Red peak; lane 13, tails, DEAE peak (2.1 pg); and lune 15, molecular mass standards, as indicated in kDa. All creatine kinase samples contained 0.5 units of activity, except for lanes 2 and 3 which contained 0.2 units.

EXPERIMENTAL PROCEDURES AND RESULTS*

Purification of Sperm Creatine Kinase Isozymes-Fig. 2 shows the SDS-PAGE pattern of fractions from the purifi- cation of head and tail creatine kinase preparations (lanes 2- 7, and 9-13). Sperm head creatine kinase was purified 30-fold and is a polypeptide of M , = 47,000 (lane 7). Flagellar creatine kinase, purified some 20-fold to homogeneity, is an unusually large form of creatine kinase, with an M, of 145,000 on SDS gels (lane 13). Neither purified isozyme comigrates with pu- rified mouse muscle creatine kinase (lane 9 ) .

As previously shown (Tombes and Shapiro, 1985), a specific in uivo inhibitor of creatine kinase, FDNB, labels four poly- peptides in intact sperm (Fig. 2, lane I ) . The upper band, which is found primarily in sperm tails (Fig. 2, lane 14) , comigrates with the purified flagellar isozyme (Fig. 2, compare lanes 13 and 14). The purified head isozyme comigrates with the middle band of the FDNB labeled triplet associated with sperm head fractions (Fig. 2, compare lanes 7 and 8).

Identification and Localization of Sperm Mitochondrial Creatine Kinase-Affinity purified rabbit polyclonal antibod- ies directed against mouse muscle creatine kinase (Chamber- lain et al., 1985) were found to be cross-reactive with the creatine kinase isozymes purified from sperm heads and tails (Fig. 3, lane 4 ) . The anti-mouse muscle creatine kinase reacted with four proteins of whole sperm (Fig. 3, lane 3 ) , heads (Fig. 3, lane 2), and tails (Fig. 3, lane 1 ) that were labeled in uivo with ['HJFDNB (Fig. 2, lane I ; Tombes and Shapiro, 1985). In heavily loaded sperm tail preparations, there was cross-

* Portions of this paper (including "Experimental Procedures," part of "Results," Tables I and 11, and Fig. 1) are presented in miniprint a t the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 86M-4498, cite the authors, and include a check or money order for $4.80 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

Protein

1 2 3 4 5 PpI - 7 - T -

1 2 3 4 5 Mr

Ix10'31

FIG. 3. Immunoblot of sperm proteins. Protein staining was with 0.1% Amido Black; immunoblotting used affinity purified anti- mouse muscle creatine kinase at 3.5 pg/ml as described under "Ex- perimental Procedures." Lanes contained the following amounts of protein: lane I , sperm tails (15 pg); lane 2, sperm heads (8 pg); lane 3, whole sperm (8 pg); lane 4, a mixture of ta i l creatine kinase (1.2 pg) and sperm head creatine kinase (0.5 pg); and lane 5, mouse muscle creatine kinase, (0.3 pg). a-OK,,,,,,, anti-mouse muscle creatine kinase.

reaction with another protein of 42,000 M, (Fig. 3, lane I ) that was not labeled by ['HIFDNB. The reaction of anti- mouse muscle creatine kinase with purified mouse muscle creatine kinase is shown for comparison (Fig. 3, lane 5 ) .

The sperm mitochondrion is a component of the isolated head fraction (Tombes and Shapiro, 1985). Since one of the proteins of the sperm head preparation that reacts with anti- mouse muscle creatine kinase was the purified head isozyme (Fig. 3, compare lanes 2 and 4 ) , we examined the localization of the anti-creatine kinase binding activity by immunofluo- rescence of whole sperm. When whole sperm were fixed with either methanol (Fig. 4) or 3% paraformaldehyde plus 0.05% Nonidet P-40 (data not shown) and stained with anti-mouse muscle creatine kinase, a bright ring of staining was observed at the region of the mitochondrion at the base of the head. Since the mitochondrion is the only sperm organelle at this site and all of the antigens of the head region were at this locus, it is reasonable to refer to the purified head isozyme of creatine kinase as mitochondrial. At higher levels of the antibody flagellar staining could be observed (data not shown), but no other region of the head reacted with the anti- mouse muscle creatine kinase antibody. We do not know why the antibody reacted equally well with head and tail proteins on immunoblots, but reacted more strongly with the mito- chondrial proteins by immunofluorescence microscopy.

Peptide Analysis of Purified Sperm Isozymes-In order to explore the relationships between the sperm proteins labeled by FDNB and anti-mouse muscle creatine kinase and the purified creatine kinase isozymes, isolated mitochondrial and flagellar creatine kinase were stoichiometrically labeled with ['HIFDNB and then compared with in uiuo ['HIFDNB-la- beled bands by Staphylococcus aureus V8 protease-peptide mapping and fluorography (according to Cleveland et al., 1977). The labeling of intact sperm in uiuo was carried out at

Sea Urchin Sperm Creatine Kinase Isozymes 16013

FIG. 4. Immunofluorescent localization of sperm head crea- t ine kinase. Sperm were fixed in methanol and stained with affinity puritied anti-mouse muscle creatine kinase as described under “Meth- ods.” Several sperm are illustrated with the mitochondria appearing as dark dots a t the base of the head in bright field ( A ) . Immunoflu- orescent exposure was for 60 s ( B ) . Bar indicates 10 pm.

the minimum FDNB concentration needed to completely inhibit creatine kinase, which is a 2- to 3-fold molar excess (see “Experimental Procedures”). The one tail and three head polypeptides accounted for 37, 29, 23, and 10% of the [“HI FDNB incorporation in whole sperm, respectively. Sperm from three different animals, spawned a t different times of the season and labeled with FDNB, contained similar propor- tions of the tail and head polypeptides.

The purified creatine kinase isozymes were labeled with stoichiometric amounts of FDNB for comparison with the labeled proteins in sperm. The purified flagellar isozyme (Fig. 5A, lune 1 ’) and the 145,000 M , band labeled in vivo (data not shown) gave [‘HIFDNB peptides of the same size. Purified mitochondrial creatine kinase (lane 2’) and the 47,000 M, polypeptide labeled in uiuo (lune 5 ‘ ) had identical [‘HIFDNB peptide maps. This similarity suggests that the 50-kDa poly- peptide is closely related to the 47-kDa head isozyme of creatine kinase. Moreover, the two have labeled peptides in common with the 44-kDa polypeptide that reacted only min- imally with FDNB (Fig. 5A, lune 6’). The same three poly- peptides were seen with anti-mouse muscle creatine kinase antibodies (Fig. 3) as well as with [“HIFDNB labeling (Figs. 2 and 5A) of sperm head preparations or intact sperm; thus, they are unlikely to be related by adventitious proteolytic cleavages during preparative procedures, for example in pu- rifying the 47-kDa creatine kinase isozyme. It is not clear whether all are creatine kinase isozymes that may be related by limited proteolytic cleavages during spermatogenesis, whether they have different roles in sperm head phosphocrea- tine metabolism, or, indeed, whether the 50- and 44-kDa proteins even have creatine kinase activity.

The mitochondrial and flagellar creatine kinase isozymes were not a t all similar by peptide mapping, as shown in Fig. 5B. The purified mitochondrial creatine kinase isozyme pro- duced three principal cleavage products of 37,33, and 11 kDa after extensive digestion of purified mitochondrial creatine kinase in solution with the V8 protease (Fig. 5B, lanes 9 and 10). With a greater amount of protease there was more 11- kDa peptide formed, with a reduction in the 37-kDa peptide, suggesting that the 11-kDa peptide arose from the 37-kDa peptide (Fig. 5B, lunes 9 and 10) . Flagellar creatine kinase digested with S. uureus V8 protease yielded different peptide bands. Differences between the head and tail creatine kinase isozymes were also seen after chymotrypsin digestion (Fig. 5B, lanes 3 and 8) . Thus, the head and tail isozymes are distinct molecular species.

145-

50 47 3 44

- - 27 ‘p

--.I -13

B 1 2 3 4 5 6 7 8 9 1 0 I x 10-31 I x 10-31 Mr Mr

47 +

26 +

+37 *33

+11

drial and flagellar creatine kinase and [“HIFDNB labeled sperm FIG. 5. A, Fluorograrn of a proteolytic digest of purified mitochon-

proteins. The following samples were separated by SDS-PAGE, ex- cised, and then separated by a second SDS-polyacrylamide gel with (lanes I “ 6 ’ ) or without (lanes 1-6) S. aureus V8 protease as described under “Experimental Procedures.” This gel represents exposure times of between 7 and 33 days. Lanes represent the following samples: lane I, purified and labeled flagellar creatine kinase; lane 2, purified and labeled mitochondrial creatine kinase; lane 3, 50-kDa band from whole sperm; lane 4,47-kDa band from whole sperm; and lane 5,44- kDa band from whole sperm. Lanes 5 and 5’ are photographs of the same gel, exposed for longer time periods than the other lanes due to the decreased label associated with these bands. Creatine kinase (from rabbit muscle), labeled and digested under the same conditions, exhibited a single band a t 21 kDa. B, Coomassie-stained SDS-PAGE of solution digest of purified mitochondrial and flagellar creatine kinase. 5 g of mitochondrial (lanes 6-10) and flagellar (lanes 1-5) creatine kinase were digested with 0.25 (lanes 2 and 7) or 0.125 (lanes 3 and 8) pg of chymotrypsin or 0.25 (lanes 4 and 9 ) or 0.125 (lanes 5 and 10) pg of S. aureus V8 protease in solution as described under “Experimental Procedures.” M, of the major products of the peptides generated by V8 digestion which most closely comigrate with bands from the “Cleveland” digest are included in kDa with molecular mass markers of the purified flagellar (145-kDa) and head (47-kDa) pro- teins.

16014 Sea Urchin Sperm Creatine Kinase Isozymes

Quaternary Structure of the Creatine Kinase Isozymes- The hydrodynamic properties of the purified isozymes were characterized by gel filtration and sucrose density gradients. The results are summarized in Table 111. Bound detergent increases the partial specific volume of proteins; if detergent is bound to a protein, the protein will have a higher apparent sedimentation velocity in H,O than in D,O (Clarke, 1975). In sucrose gradients, flagellar creatine kinase comigrated with lactate dehydrogenase, and mitochondrial creatine kinase mi- grated slightly more slowly than catalase. There was no shift in the sedimentation velocity with respect to the protein standards when the experiment was performed in DaO (data not shown). From this analysis, the partial specific volume estimates were 0.75 cm3/g for the flagellar isozyme and 0.73 cm3/g for the mitochondrial isozyme. These results were in the range reported for soluble, detergent-free proteins (Sober, 1970) and for muscle creatine kinase (Noda et al., 1954; Yue et al., 1967). Superose-12 gel filtration chromatography of both purified isozymes yielded single sharp peaks, from which a Stokes radius was obtained by linear interpolation with standards. From these results, the M , was calculated using the Svedberg equation. The M, of the flagellar isozyme (126,000) was within 15% of that estimated from SDS-PAGE (145,000), while the mitochondrial isozyme possessed a M , (240,000) five times the size determined by SDS-PAGE. The f/fo values determined for the mitochondrial (1.02) and fla- gellar (1.14) isozymes suggested much less asymmetric mole- cules than suggested by the values (1.21-1.28) for cytoplasmic brain and muscle creatine kinase isozymes (Noda et al., 1954; Yue et al., 1967).

Test for Cell Surface Exposure of Creatine Kinase Isozymes and Presence of Carbohydrate-When sperm were labeled with I3H]FDNB in vivo and then treated with chymotrypsin, Pronase, or trypsin, no change in migration behavior of the labeled proteins was observed on SDS-PAGE. The pattern of ['HIFDNB-labeled polypeptides was exactly as observed in untreated, [3H]FDNB-labeled sperm (Fig. 2, lane 1) .

The purified flagellar isozyme does not seem to possess carbohydrate, as indicated by the absence of staining by the periodate-Schiff or periodate-silver procedures. Ovalbumin, a glycoprotein, and sperm membrane glycoproteins (Podell et al., 1984) were used as positive controls. The mitochondrial isozyme was not tested.

pH Profile and Kinetics of Purified Sperm Creatine Kinase Isozymes-Using the coupled spectrophotometric assays (see "Experimental Procedures"), at 25 "C and pH 6.5 in the reverse direction (ATP formation) and pH 7.5 in the forward direction (phosphocreatine formation), both purified isozymes behaved in normal Michaelis-Menten fashion, exhibiting lin- ear Lineweaver-Burk plots for both substrates in both reac- tion directions. From these double reciprocal plots, the data in Table I1 were obtained. In general, the mitochondrial isozyme had slightly lower K,,, values for its substrates than the flagellar isozyme, and at all pH values the mitochondrial isozyme had at least a 2-fold higher V,,, in the forward direction than the flagellar isozyme, even though both iso-

zymes had similar pH-V,,, profiles in the reverse direction (Table IV). Nonetheless, apparent equilibrium constants com- puted according to

Vt. &cr. KAOP V r . Kc, . KATP KeLlbpp) =

in which K P C ~ , KADP, KC,, and KATP are the Michaelis constants for phosphocreatine, MgADP, creatine, and MgATP, respec- tively, were similar for both sperm isozymes. Those values (2.7 x lob3 for the mitochondrial isozyme and 4.6 X for the flagellar isozyme) were slightly lower than the value (10 x observed for purified mouse muscle creatine kinase at pH 7.4 and 30 "C (Nihei et al., 1961).

The effect of temperature on creatine kinase activity was similar for both sperm isozymes. In both cases, activity dou- bled from 10.5 to 25.0 "C. Above 25 "C, activity increased by only 25% before decreasing above 37 "C, presumably due to denaturation (data not shown).

Inorganic phosphate (Pi), a principal component of active sperm (Christen, 1983b) and presumably distributed in a gradient along the flagellum, is a known competitive inhibitor of phosphocreatine in the reverse direction (Watts, 1973). The K, of Pi in the reverse direction at pH 6.5 was 5 and 7 mM for the flagellar and mitochondrial isozymes respectively. Neither creatine in the reverse direction nor phosphocreatine nor Pi in the forward direction were capable of significantly inhibiting either isozyme at concentrations below 50 mM. Neither ATP nor ADP were tested as product inhibitors due to the nature of the coupled assay.

Amino Acid Composition and Ultraviolet (UV) Spectra- The amino acid compositions of the mitochondrial and fla- gellar creatine kinase isozymes are shown in Table V. For comparison, the amino acid composition of mouse muscle creatine kinase isozyme as determined from the cDNA se- quence (Buskin et al., 1985) is also shown. Protein sequence analysis revealed the amino terminus of the flagellar isozyme to be blocked, while the mitochondrial isozyme gave a partial sequence of Ala-Ala-Asn-Gly. Protein extinction coefficients (in A/cm for 10 mg/ml) were calculated from spectra taken in 150 mM KC1, 10% glycerol, and 10 mM Tris, pH 7.5, and analyzed for protein by the Bradford (1976) assay. Extinction coefficients for the flagellar (7.6) and mitochondrial (8.0) isozymes were similar to the value of 8.9 determined for rabbit muscle creatine kinase (Sober, 1970). UV spectra were similar for the two isozymes.

DISCUSSION

We have previously provided some of the first direct evi- dence for phosphocreatine shuttles in any cell by determining the effects of creatine kinase inhibition on sperm swimming behavior (Tombes and Shapiro, 1985; Tombes et al., 1987). A phosphocreatine shuttle would employ separate isozymes at the mitochondrion and along the flagellum. We have now purified both isozymes to homogeneity and localized them to those organelles. The purified isozymes have been compared by peptide mapping, hydrodynamic analysis, kinetic and chro-

TABLE 111 Physical parameters of purified sperm creatine kinase isozymes

Sperm creatine Isoelectric kinase isozyme point

::::: specific Partial Molecular

s20.w f / f O weight from volume SDS gels

<?so" weight

A c m ' k Alcm (1 %) Mitochondrial 6.2 45 0.73 12.4 1.02 235,000 47,000 8.0 Flaeellar 6.45 41 0.75 6.8 1.14 126,000 145.000 7.6

a Protein determined by Bradford assay (see "Experimental Procedures").

16015 Sea Urchin Sperm Creatine Kinase Isozymes TABLE IV

Kinetics of purified sperm creatine kinase isozymes K,,,, pH optima, and Ki were all measured as described under “Experimental Procedures.” K , values were

averaged from at least three individual experiments with standard deviations as indicated. Cr, creatine; PCr, phosphocreatine.

forward

reverse Apparent MgATP + Cr e-’ MgADP + PCr + H’

Reverse reaction direction Forward reaction direction K , at

Sperm creatine pH 7.5 kinase isozyme vm.= at p~ (X lo3)

pH ‘muat pH ir7at K, for pi K,,,(ATP) K,,,(Cr) optimum PH optimum L ( A D P ) Km(PCr) optimum optimum (7.5)

mM mM unitslmg unitslmg mM rnM mM unitslmg Mitochondrial 0.055 f 0.015 2.7 f 0.5 6.9 182 139 7 0.31 f 0.11 62 f 3 7.7

246 49

Flagellar 0.130 f 0.005 5.8 f 0.3 6.7 136 5 0.89 k 0.23 39 ? 18 7.5 29 4.6 2.7

TABLE V Amino acid composition of sperm creatine kinases

Amino acid composition was determined on 0.04 mg of protein in duplicate.

Amino acid creatine kinase creatine kinase creatine kinase Flagellar Mitochondrial Mouse muscle

Asx 12.6 Glx 9.1 Ser 4.9 GlY 8.8 His 3.3

5.1 Thr 5.6 Ala 5.4 Pro 5.0 Tyr 2.8 Val 6.3 Met 2.3 Ile 4.2 Leu 9.7 P he 4.0 LY s 7.1 CYS ND” Trp 2.8

ND, not determined.

mol % 10.3 4.7 8.9 9.6 3.7 7.6 6.3 6.7 5.2 2.7 8.1 0.9 5.0 8.7 3.6

ND” 5.1

1.8

8.1 7.5 5.7 9.5 4.3 5.2 5.2 4.0 5.7 2.8 8.1 2.8 4.0

10.7 4.6 9.5 1.2 1.2

matographic behavior, extractability, localization, and amino acid composition. They are distinct proteins.

The initial difficulty encountered in extracting soluble crea- tine kinase activity from sperm implied that sperm isozymes differ from those isolated from vertebrate sources. The solu- bilization of cytoplasmic isozymes of creatine kinase from mammalian muscle and brain tissue merely requires homog- enization in physiological buffers (Dawson and Eppenberger, 1970; Hershenson et al., 1986; Keutel et al., 1972; Walliman et al., 1984; Wang and Cushman, 1980; Watts, 1973), while the mitochondrial isozyme has been extracted by swelling mitochondria with 50-100 mM phosphate (Blum et al., 1983; Farrell et al., 1972; Grace et al., 1983; Hall et al., 1979; Scholte et al., 1973). Neither of these conditions successfully extracted more than 15% of the creatine kinase from whole sperm, sperm heads, or sperm tails which had been homogenized, sonicated, or made semipermeable with low concentrations of digitonin (Table I).

High concentrations of nonionic detergents (0.5-5.0%) in the presence of 150 mM KCl, on the other hand, extracted 98-99s of the creatine kinase from sperm heads or tails. Detergents have generally not been used for extraction of any of the other creatine kinase isozymes, although deoxycholate treatment was used to extract mitochondrial creatine kinase (Wevers et al., 1981). The detergent requirement is not nec- essarily due to the impermeant nature of the spermatozoa,

since some sea urchin sperm proteins have been extracted by sonication or homogenization in physiological buffers (Lee and Iverson, 1976, Swarup and Garbers, 1982).

Although creatine kinase isozymes were extracted with Nonidet P-40, hydrodynamic studies suggested that they did not bind significant amounts of detergent (Table 111). Once extracted, neither isozyme required detergent for solubility at 100,000 X g, although the head isozyme required 75 mM KC1 and 10% glycerol (data not shown). Both creatine kinase isozymes may possess some hydrophobic domains since they both bind to either phenyl-Sepharose or polystyrene beads (data not shown); however, the enzymes were not purified to any extent by phenyl-Sepharose chromatography. During ex- traction, 125 mM KC1 facilitated the solubilization of both isozymes and, upon purification, 250 mM KC1 was required to maintain the solubility of the purified mitochondrial iso- zyme. We subsequently used this insolubility at low salt concentrations in the purification of the mitochondrial iso- zyme (Table 11).

The flagellar isozyme has been observed in membrane preparations prepared by certain procedures (Kazazoglou et al., 1985), but not others (Podell et al., 1984) and it is not glycosylated (see “Results”). Neither the 140- nor 150-kDa polypeptides found in sea urchin sperm plasma membranes (prepared according to Podell et al., 1984) possessed similar- ities to purified flagellar creatine kinase when compared by cyanogen bromide-peptide mapping.3 Neither mitochondrial nor flagellar isozymes are exposed on the cell surface to proteases (see “Results”). The mitochondrial isozyme tends to aggregate in the absence of detergent, salt, and glycerol. Mammalian mitochondrial isozymes are associated with the inner mitochondrial membrane, and perhaps this is the loca- tion of sperm mitochondrial creatine kinase. The association of the head creatine kinase with mitochondria may be based upon hydrophobic interactions; upon purification and removal of detergent and salt, those hydrophobic domains may serve as sites for aggregation. Flagellar creatine kinase could be associated either with the inner surface of the flagellar mem- brane or the axoneme.

Sperm creatine kinase isozymes appear to be major cellular components as assessed by their high specific activity in homogenates, the relatively small increase in specific activity that accompanies their purification (Table II), and their prominence in protein-stained gels of whole sperm (Fig. 2, lane 2 ) . Purified sperm flagellar and mitochondrial creatine kinase isozymes possess comparable specific activity to clas- sical preparations of mammalian isozymes (Dawson and Ep- penberger, 1970; Keutel et al., 1972) and comprise approxi-

V. Vacquier, personal communication.

16016 Sea Urchin Sperm Creatine Kinase Isozymes

mately 5 and 3% of the proteins in tails and heads, respec- tively.

The purified, soluble mitochondrial isozyme behaved as an oligomer, four to six times the polypeptide M , of 47,000. The large M , of the native mitochondrial isozyme (235,000) is not due to disulfide cross-linkage. This is of particular interest, since there has been controversy over whether mitochondrial creatine kinase has a native M , higher than the typical 82,000- 84,000 reported for almost ail mammalian mitochondrial and cytoplasmic isozymes (Hall et al., 1979; Farrell et at., 1972; Grace et al., 1983; Blum et al., 1983). Higher native M, forms of the mammalian mitochondrial isozyme (200,000 Da) were observed unless urea was included in the preparations (Grace et al., 1983), a procedure that we have not employed. On the other hand, flagellar creatine kinase is a clearly unique form of this enzyme. Its behavior on SDS gels, as confirmed by gel filtration and sucrose density gradient analysis, indicates that it is a monomer that is approximately three times the size of other known creatine kinase monomers. I t is slightly asym- metric, but not as asymmetric as the prolate ellipsoidal mam- malian creatine kinase monomers (Watts, 1973).

Despite its unusual M,, the flagellar creatine kinase is immunologically related to other mammalian creatine kinase isozymes (Fig. 3). The polyclonal antibody used in these experiments was prepared against partially denatured mouse muscle creatine kinase and then was affinity purified it reacts with all creatine kinase isozymes tested (Chamberlain et al., 1985). This antibody also reacts with all three FDNB-reactive polypeptides found in the sea urchin sperm head and localizes cross-reactive material to the sperm mitochondrial region (Fig. 4). The persistence of several creatine kinase-like poly- peptides in sperm heads is puzzling. If not representing post- translational modifications or an inefficient mitochondrial creatine kinase precursor processing system, the three head creatine kinase polypeptides could conceivably represent ex- perimentally induced proteolytic artifacts of one creatine ki- nase polypeptide. However, evidence presented here suggests that is not the case. Although flagellar creatine kinase has an M, which is roughly the sum of the three mitochondrial creatine kinase polypeptides, peptide mapping indicates that none of the three head creatine kinase polypeptides arose from, or has any similarity to, the flagellar creatine kinase. No high M , precursors that were reactive with FDNB or the antibody were ever observed in sperm heads or tails. Peptide mapping also indicated a high degree of similarity between the three mitochondrial polypeptides. If the three head crea- tine kinase polypeptides arose as proteolytic artifacts from a larger precursor, the necessary cleavages would have always occurred to the same extent and would have arisen very quickly upon cell lysis because sperm labeled with FDNB or analyzed by immunoblotting were dissolved immediately in sample buffer, as described uilder “Experimental Procedures.” Multiple proteins have also been found in the heads of other sea urchin sperm, both immunologically and by FDNB reac- t i ~ i t y . ~ Only an unlikely constellation of proteolytic artifacts would have consistently generated the three mitochondrial polypeptides in the same relative amounts.

The different, related head polypeptides might have arisen from a processing event during spermatogenesis. A slightly larger precursor polypeptide to the rat heart mitochondrial creatine kinase has been detected immunologically from i n vitro translated rat heart mRNA, but the precursor protein was not seen directly (Perryman et al., 1983). The 44-kDa polypeptide which shows diminished labeling by FDNB i n vivo relative to the other two labeled bands in sperm head

‘ R. M. Tombes and B. M. Shapiro, manuscript in preparation.

preparations (Fig. 2, lane 1 ), accounting for -15% of the head radioactivity, possesses some similar proteolytic peptides and reacts immunologically very strongly (Fig. 4, lanes 2 and 3 ) , but it is not clear how closely it is related to the purified mitochondrial creatine kinase. None of the other mitochon- drial creatine kinase polypeptides has been purified, but the low final recovery (7%) of the 47-kDa creatine kinase activity from heads (Table 11) might be accounted for by the removal of other head creatine kinase isozymes.

All K, values observed by others for purified cytoplasmic (Eppenberger et al., 1967; Morrison and James, 1965; Nihei et al., 1961; Saks et al., 1976; Wang and Cushman, 1980) and mitochondrial (Blum et al., 1983; Hall et al., 1979; Saks et al., 1975,1980) creatine kinase isozymes are similar to each other. The K , values determined here for ATP, phosphocreatine, and ADP, but not creatine, fall into the same ranges. For example, the K, of the flagellar isozyme for ATP (0.89 mM), phosphocreatine (5.8 mM), and ADP (0.130 mM) are near the values of 0.45-1.5 mM (ATP), 0.8-4.8 mM (phosphocreatine), and 0.05-0.83 mM (ADP) reported for the cytoplasmic iso- zymes (Eppenberger et al., 1967; Morrison and James, 1965; Nihei et al., 1961). Likewise, the K, values of the sperm mitochondrial isozyme for ATP (0.31 mM), phosphocreatine (2.7 mM), and ADP (0.055 mM) are similar to those reported for the mitochondrial isozyme ranges (Blum et al., 1983; Hall et al., 1979; Saks et al., 1975, 1980): 0.056-1.7 mM (ATP), 0.31-3.0 mM (phosphocreatine), and 0.0015-0.15 (ADP). The exception in both cases is for creatine, where the K,,, for the flagellar (39 mM) and the mitochondrial (62 mM) isozymes are well above values reported for cytoplasmic (4.5-19 mM) or mitochondrial (1.7-13 mM) isozymes (see references above). Creatine kinase from dogfish muscle, however, also has an unusually high K, (53 mM) for creatine (Simonarson and Watts, 1972).

Although the mitochondrial isozyme has slightly lower K, values for all substrates except creatine and higher VmaX values in the forward direction, the inherent kinetic preference of purified sperm isozymes for one reaction or the other is probably insignificant. It has been suggested that the localiza- tion of creatine kinase isozymes in microcompartments pro- vides the enzyme with special access to its substrates, even though the cellular or even organellar substrate concentra- tions might not favor a certain reaction direction at equilib- rium (Bessman and Carpenter, 1985). For example, the bound mammalian mitochondrial isozyme has preferential access to mitochondrially synthesized ATP over cytoplasmic ATP (Booth and Clark, 1978; Erickson-Viitanen et al., 1982a, 1982b; Saks et al., 1975, 1980). Sarcomeric creatine kinase may also be situated such that i t has preferential access to ADP released by the hydrolysis of ATP by myosin (Savabi et al., 1983; Walliman et al., 1984).

The sea urchin sperm system refuted the conventional wisdom that arginine kinase is the invertebrate phosphagen- kinase while creatine kinase is restricted to vertebrates (Bess- man and Carpenter, 1985; Dawson and Eppenberger, 1970). It also has proven to be an ideal system in which to examine the existence and mechanism of phosphocreatine shuttles directly. From the special characteristics of sperm motility, the role of the shuttle system was clearly established (Tombes and Shapiro, 1985; Tombes et al., 1987). This is the first report of the characterization of the enzyme termini of an established phosphocreatine shuttle. In this regard, it is in- teresting that the sperm isozymes that exist at opposite ends of this shuttle do not exhibit inherent reaction preferences, nor differential sensitivity to inhibition by relevant sperm metabolites; they are capable of catalyzing either reaction

Sea Urchin Sperm Creatine Kinase Isozymes 16017

direction with equal facility. Their catalysis of opposite direc- tions of the same reaction in vivo is more likely dependent upon the substrate microenvironments of the two isozymes and the net steady state flux of -P from mitochondrion to tail. The physical differences between the isozymes may direct them to their respective intracellular environments during spermatogenesis. The shuttle is driven by fatty acid oxidation by motile sperm (Shapiro and Tombes, 1985), but whether there are additional controls on this system remains to be determined.

Acknowledgments-We would like to acknowledge the assistance and advice of Lowell Ericsson, Lisa Hager, Nancy McHenry, Lien Pham, Vic Vacquier, Roger Wade, and Don Wothe. We are also grateful to Jeff Chamberlain and Steve Hauschka for the creatine kinase antibodies, Dave Battaglia for constructive comments on the manuscript, and Mary Patella for expert typing assistance.

REFERENCES

Bergmeyer, H. U. (ed) (1974) in Methods of Enzymatic Analysis, Vol. 2, pp.

Bessman, S. P. & Carpenter, C. L. (1985) Annu. Reu. Biochem. 54,831-862 Bessman, S. P., Yang, W. C. T., Geiger, P. J. & Erickson-Viitanen, S. (1980)

Blum. H. E., Deus, B. & Gerok, W. (1983) J. Biochem. (Tokyo) 94, 1247-1257 Booth, R. F. G. & Clark, J. B. (1978) Bmhem. J. 170, 145-151 Bradford, M. M. (1976) Anal. Biochem. 72, 248-254 Buskin, J. N., Jaynes, J. B., Chamberlain, J. S. & Hauschka, S. D. (1985) J.

Carpenter, C. L., Mohan, C. & Bessman, S. P. (1983) Biochem. Biophys. Res.

Chamberlain, J. S.. Jaynes, J. B. & Hauschka, S. D. (1985) Mol. Cell Biol. 5,

Christen, R., Schackmann, R. W. & Shapiro, B. M. (1983a) J. Bid. Chem. 258,

Christen, R., Schackmann, R. W., Dablquist, F. W. & Shapiro, B. M. (1983b)

Clarke, S. (1975) J. Biol. Chem. 250 , 5459-5469 Cleveland, D. W., Fischer, S. G., Kirschner, M. W. & Laemmli, U. K. (1977) J.

Biol. Chem. 252 , 1102-1106 Dawson, D. M. & Eppenherger, H. M. (1970) Methods Enzymol. XVIIA, 995-

1002 Deits, T., Farrance, M., Kay, E: S., Medill, L., Turner, E. E., Weidman, P. J.

& Shapiro, B. M. (1984) J. Bzol. Chem. 259, 13525-13533 Dubray, G. & Bezard, G. (1982) Anal. Biochem. 119,325-329 Eppenberger, H. M., Dawson, D. M. & Kaplan, N. 0. (1967) J. Biol. Chem.

Erickson-Viitanen, S., Viitanen, P., Geiger, P. J., Yang, W. C. T. & S. P.

Erickson-Viitanen, S., Geiger, P. J., Viitanen, P. & Bessman, S. P. (1982b) J .

Fairbanks, G., Steck, T. L. & Wallach, D. F. H. (1971) Biochemistry 10,2606-

Farrell, E. C., Baba, N., Brierly, G. P. & Grumer, H. D. (1972) Lab. Inuest. 27,

Font. B.. Vial. C.. Goldschmidt. D.. Eichenbereer. D. & Gautheron. D. C. (1981)

574-579, Academic Press, New York

Biochim. Biophys. Res. Commun. 96, 1414-1420

Mol. Euol. 22,334-341

Commun. 11 1,884-889

4%-492

5392-5399

Exp. Cell Res. 149, 289-294

242, 204-209

Bessman (1982a) J. Biol. Chem. 257 , 14395-14404

Biol. Chem. 257 , 14405-14411

261 7

209-213

Arch. Biochim.'Biophys. 212, 195-203 Y l ~~~ ~ ,~ ~

Fujimaki, H. & Yanagisawa, T. (1978) Deu. Growth & Differ. 20, 125-131 Gellerlch, F. & Saks, V. A. (1982) Biochem. Biophys. Res. Commun. 1 0 5 , 1473-

1481 Gerken, G. & Schlette, U. (1968) Experientia 24, 17-19 Gibbons, B. H. & Gibbons, I. R. (1972) J . Cell Biol. 54. 75-97

Grace, A. M., Perryman, M. B. & Roberts, R. (1983) J. Bid. Chem. 258,15346-

Gundersen, G. G. & Shapiro, B. M. (1984) Biochim. Biophys. Acta 799 , 68-79 Hall, N., Addis, P. & DeLuca, M. (1979) Biochemistry 18,1745-1751 Hershenson, S., Helmers, N., Desmueles, P. & Stroud, R. (1986) J . Biol. Chem.

Jacobs, H., Heldt, H. W. & Klingenberg, M. (1964) Biochem. Biophys. Res. Infante, A. A. & Davies, R. E. (1965) J. Biol. Chem. 240,3996-4001

Jacobus, W. E. & Lehninger, A. L. (1973) J. Biol. Chem. 248,4803-4810 Kazazoglou, T., Schackmann, R. W., Fosset, M. & Shapiro, B. M. (1985) Proc.

Keutel, H. J., Okabe, K., Jacobs, H. K., Ziter, F., Maland, L. & Kuby, S. A.

Laemmll, U. K. (1970) Nature 227,680-685

Markwell, M. A. K., Haas, S. M., Bieber, L. L. & 'folgert, N. E. (1978) Anal. Lee, M. Y. M. & Iverson, R. M. (1976) Biochim. Bio h s Acta 429, 123-136

Moreadith, R. W. & Jacobus, W. E. (1982) J. Biol. Chem. 257,899-905 Morrison, J. F. & James, E. (1965) Biochem. J. 97, 37-52 Nihei, T., Noda, L. & Morales, M. F. (1961) J. Biol. Chem. 236,3203-3209

Nunnally, R. L.'& Hollis, D. 6: (1979) Biochemistry 18, 3642-3646 Noda, L., Kuby S. A. & Lard H A (1954) J. Bid. Chem. 209,203-210

O'Farrell, P. H. (1975) J. B i d . Chem. 250, 4007-4021 O'Sullivan, W. J. & Morrison, J. F. (1963) Biochim. Biophys. Acta 77, 142-144 Perryman, M. B., Strauss, A. W., Olson, J. & Roberts, R. (1983) Biochem.

Podell, S. B., Moy, G. W. & Vacquier, V. D. (1984) Biochim. Biophys. Acta

15354

261,3732-3736

Commun. 16,516-521

Natl. Acad. Sci. U. S. A. 8 2 , 1460-1464

(1972),Arch. Biochem. Biophys. 150,648-678

Biochem. 87,206-210

Biophys. Res. Commun. 110,967-972

778.2.5-27 Saks V . A. Chernousova G. B., Gukovsky, D. E., Smirnov, V. N. & Chazov,

Saks, V. A., Lipina, N. V., Smirnov, V. N. & Chazov, E. I. (1976)Arch. Biochem.

Saks, V. A., Kupriyanov, V. V., Elizarova, G. V. & Jacobus, W. E. (1980) J.

Savabi, F., Geiger, P. J. & Bessman, S. P. (1983) Biochem. Biophys. Res.

Scholte, H. R., Weijers, P. J. & Wit-Peeters, E. M. (1973) Biochim. Biophys.

Shapiro, B. M. & Tombes, R. M. (1985),Bioessays 3,100-103 Sober, H. A. (ed). (1970) Handbook of Btochemistry, 2nd Ed., Chemical Rubber

,-- - .

E. i. (196) Eur. J. Bioc'hem. 57,273-290

Biophys. 173, 34-41

Biol. Chem. 255,755-763

Commun. 114, 785-790

Acta 291 , 764-773

Cn Studier, F. W. (1973) J. Mol. Biol. 79, 237-248

Tombes, R. M. & Shapiro, B. M. (1985) Ce1!41;3251334- Swamp, G. & Garbers, D. L. (1982) Biol. Re rod 26 953 960

Tombes, R. M., Brokaw, C. J., and Shapiro, B. M. (1987) Biophys. J. 52, 75- 86

Towbin, H., Staehelin, T. & Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 76.43.511-43.54

Turner, D. C., Walliman, T. & Eppenberger, H. M. (1973) Proc. Natl. Acad.

Wallimann, T., Turner, D. C. & Eppenberger, H. M. (1977) J. Cell Bid. 76,

. . , . . . . - - - - Set. U. S. A. 70, 702-705

797-21 7 Wallimann, T., Schlosser, T. & Eppenberger, H. M. (1984) J. Bid. Chem. 259,

".

572R-53Afi F. L. & Cushman, D. W. (1980) Clin. Chim. Acta 106,339-345

Watts, D. C. (1973) in The Enzymes (Boyer, P. D., ed) 3rd Ed, Vol. 8, pp. 383-

"" "_" 4.55. AcadPmir Press NPW Ynrk

Wetlaufer, D. B. (1962) Adu. Protein Chem. 17,303-390 Wevers, R. A., Reutelingsperger, C. P. M., Dam, B. & Soons, J. B. J. (1981)

Winkler, M. W., Matson, G. B., Hershey, J. W. B. & Bradbury, E. M. (1982)

Yanagisawa, T. (1959a) J. Fac. Sci. Uniu. Tokyo Sect. IVZool. 8,473-479 Yanagisawa, T. (1959b) J. Fac. Sci. Uniu. Tokyo Sect. IV Zool. 8, 482-486 Yanagisawa, T. (1959~) J. Fac. Sci. Uniu. Tokyo Sect. IV Zwl. 8,487-498 Yanagisawa, T. (1967) Exp. Cell Res. 46, 348-354 Yang, W. C. T., Geiger, P. J., Bessman, S. P. & Borrebaek, B. (1977) Biochim.

Yue, R. H., Palmieri, R. H., Olson, 0. E. & Kuby, S. A. (1967) Biochemistry 6, Yang, W. C. T. & Dubick, M. (1971) LifeSci. 21, 1171-1178

~~. , . . __", - " .. - "_.

Clin. Chim. Acta 119, 209-223

Exp. Cell Res. 139, 217-222

Biophys. Res. Commun. 76,882-887

3204-3227

Continued on next page.

16018 Sea Urchin Sperm Creatine Kinase Isozymes Supplementary material to: "Enzyme Termini of a Phosphocreatine Shuttle:

Urchm Sperm." Robert M. Tombes and Bennett PI. Shaplro. Purification and Characterization of m o Creatine Kinase ~sozymes from sea

EXPERIMENTAL PROCEDURES

purpuratus, were obtained by intracoelomic inlection Of 0.SM KC1 e 10T. All Materials - Spermatozoa from the purple sea urchin, Strangylocentrotus

water was prepared as previously described Ichristen et e.. 198311). Nonidet CrK purifications were performed at 2-4OC or on ice. Millipore-filtered sea

nucleotides were Obtained as the Sodium salts from Sigma 198-99% pure). Thelr P-40 INP-40) was Obtained from BDH Chemicals. Poole, England. Adenine

concentration wan calibrated using their ~ 2 6 ~ llmnl = 1 5 . 4 [Sober, 19701. The enzymes used In the Coupled assay, including pyruvate kinase, lactate dehy- drogenase, hexokinase and glucose 6-phosphate dehydrogenase were Obtained from Sigma in the lyophilized, salt-free form, to elminate inhibitory effects of inorganlc anions On CTK activity IWattS. 1973). Creatine, NADH NADP. phosphoenol pyruvate, diadenasine pentaphoaphate and N-acetyl cyste;ne were obtained from Sigma. Porcine gamma-globulin and bovine serum albumin IFraction VI were from Pentex. Spectrophotometric assays were carried Out on

Miles, chymotrypsin was from Calbiachem and p r o n a G from Horthington. a Beckman Du-7 spectrophotometer. StaPhVloCoccUs allreus V8 protease was from

heads or tails With the no"-ionic detergent. NP-40 and partially purified Purificatlon of CrK I sozmes - CrK actinty was extracted from sperm

wlth Proclon Red chromatography as previously described ITombes and Shapiro, 19851. except that 10% instead of 5% glycerol Vas used in all buffers and a 0-1.OM KC1 gradient was used to elute CrK from the Procion Red column. CrK

purification of the tail isozyme was carried Out by application to a column from tails eluted at 0 .3M KC1 and from heads a t 0 .6M KC1. Further

11.5cm x 2Ocml Of DEAE-BIOGEL IBIO-RAD1 ~n "DEAE Buffer", Consisting Of 10% glycerol, 15mM Hepes. 15mM TriS. 0.5mM EGTA, 0.2mM U P . 0.2mM DTT, pH 8.0 and elution With a linear 0-0.5M KC1 gradient. Tail CrK eluted at 0.10M KCI.

Into DEAE buffer free of glycerol and contalnlng l O m M KC1. Thls caused the selective precipitation of head CrK, Whlch was collected by Centrlfugation at 10.000 x g for 30 minutes. Fu~ther purification Was carried out by DEAE chromatography as above. and elution at 0.18M KCI. Purified CrK isozymes were stored in 50% glycerol, 10mM imidazole acetate. pH 7.0. 1mM Mg acetate, 0.5mM ADP, 0.5mM DTT, lmM EGTA e -loPC.

CrK from heads was further purifled by dialysis of the Procion Red peak

Lowry method of Markwell s.119781, using bovine serum albwln as standard, or by the method of Bradford 119161, using purifled porcine gamma- globulln as standard.

Protein Determlnatlon - Protein was determined according to the modlfled

examined qualitatively on SDS gels by both the periodate-Schiff's reagent stain of Failrbanks 119711 and the mare sensitive silver Stamina Drocedure of

Carbohydrate determination - The carbohydrate content of praterns was

Dubray and Bezard 119821.

palyacrylamlde slab gel5 were used to assess sample purlty according to

dllutlon into an equal Y ~ l m e of double Strength SDS sample buffer cons1stlng Laemmli (19101 and Studler 119731. Samples were routinely prepared by

of 4% SDS, 0.lOM Trla-C1. pH 6.8. 20% glycerol. 0.0088 bromophenol blue and O.2M DTT far reducing gels . However. if the protein concentration was too law, or if KC1 or NP-40 was present, samples were precipitated on ice with 10% trichloroacetic acid. washed twice with ethanol at -20°C and dissolved in

POlvaCrvlamLde oel electrophoresis - Di~continuous SDS, 5-15% gradient

mg/ml were bmled ~n 0.5% SDS for 2 mmutes, then digested with 0.0175 mg/ml Peptide Analvsls - A1 Proteolyslr in solution: Purlfled proteins at 0.35

StaphYloCoccuS v8 protease for 3 h at 31'C. at vhlch t m e the reactLon was terminated with 2% SDS, 10% P-mercaproethanal and bailed fox 2 mi" accordlng to Cleveland et &., 1977. Peprides were separated on an SDS 10.20% polyacrylamide gel. B l Prore01 sia In gels: Purified CIK isozymes at 0.25mg/ml were labelled wlth 8 W ?H-FDNB [New England Nuclear, 20.4 Cilmmall in 0 . 0 4 M sodium bicarbonate, pH 8.5 for 8 h at 2OC. Intact sperm Were labelled With 10 UM

proteins separated on SDS 5.15% polyacrylamide gels. Radioactive bands 'H-FDNB as preYious1y described ITOmhes and Shapiro, 1 9 8 5 1 . and labelled

weze visualized by soaking gels in the fluoroqraphlc enhancing agent AMPLIFY

-?Doc for various tunes. Labelled bands on SDS gels correspondmg to CrK IAmershaml before drying. then eXpOJlnq KODAK X-OMAT liR-2 pre-flashed film e polypeptldes from pure CrK and whale sperm samples were cut Out of the drled gel , rehydrated in 0.12511 TrIS-Cl. pH 6.8. placed on top Of a second gel, overlaid wlth 0.01 mg/rnl 5 . aweus v8 protease and electrophoretically separ- ated as previouslv described Icleveland et a1.,19771. Gels were dried and visvalizid as =bo&.

equimolar FDNB labe114ng USJng the specifLC activity Of pure sperm CrK reported here: 4 x 10 sperm/ml contain approximately 80Ulml CrK Ireverse1

"

The approximate COncentLLtion of CIK in whole sperm was Calculated for

activity ITombes and Shapira, 19881. If purified CrK~has a specific activity Of 25OU/mg, then CrK 1s present in this particular s u ~ p e n ~ i ~ n at 0.32 mglml. If mlrochondrlal CrK IM - 47 000) comp~~ses 60% Of whole sperm dctlvity and fail CrK IMr = 126,0001r~~rnpr~ses 40% of the activity ITombes and Shaplro, 1985 and Unoubllshed datal. then the maximum concentration of CrK m thls 10- fold dllufei sperm suspens~on is 4 !LM 1I.e.. It ls about 40UM in undlluted sperm]. In such sperm suspensions, FDNB at Concentrations up to 20 W reacts predomlnanrly wlth the CrK polypeptldes of sperm and Inhibits CrK activity half-maximally at 3 UM ITombes and Shapiro, 1 9 8 5 1 .

protease sensltivit at s erm CTK ln v Y O - sperm were labelled at 10 UM 3H-FDNB as above and We=: diluFed to 6 x 10; cellrlml and incubated with O.lmg/ml protease in MFSH IPOdell et e., 1984). After one hour, 0.2 ml protease-treated sperm Were diluted with 1.0 ml MFSW and centrifuged 10.000 x g for 10 mi". The sperm were resuspended in SDS gel buffer and processed by SDS gel electrophoresis and fluorography as described above.

Molecular Yeiaht Determinatlan under Non-denaturing Conditions - Stokes radil ( S 1 were determined by gel filtration of 100 ul samples in 2 0 W potassld phosphate, pH 1 . 0 . 150mH NaCl and 5% glycerol at 25'C on a Pharmacla superose-12 FPLC column I1 cm x 30 cml. Standards were Obtained from 810-R*D and included thyroglobulin 186 AI, gma-glohulin 14461. ovalbumin (30111, myoglobin 119Ai and vitamin 8-12, S estimates Were Obtalned by linear interpolation of lelution volumelvold Y o l d e l VS. log Sr.

Sedimentation velocifv analvsis was performed e 5'C in D70 and ~n H>O by applying a 0 .1 m l sampie containing IO yg of each standad and C ~ K oiita a 3 ml 5.20% SUCrose gradient Overlaid on a 0.5 ml 409 sucrose cushion. Mito- chondrial CrK was centrifuged at 32,000 and tall CTK at 42.000 RPM for 16

was determined by linear interpolation of their refractive index VS. the hours in a Beckman Type SW65 rotor at 5-C. The density of collected fractions

density Of known standards @ 5-C. Buffers con5~sted of 0.5% NP-40, lOmM

POtaSSim phoS hate, pH 7.0, 15OmM NaCl and standards i cluded catalase 111.35. 0.73cmq/gl, lactate dehydrogenase 17.35. 0.74cmq/gl and lacto- peroxidase (5.?3S. 0.164 cm3/g1, (Sober, 1970). all Obtalned from Sigma. Catalase was analyzed by the consumption of l8m H 0 at 240m in l O m M sodium phosphate, pH 7.0 IBergmeyer, 1974, pp. 673iSi41. lactate dehydro- genase by the 1 . S m M Pyruvate-dependent consumption of 0.2m NADH at 340m. l O m M P. pH 1 . 0 IBermeyeI, 1974. pp. 574-5791 and lactoperoxidase by the Produc?ion of I at 353nm from 0.3mM Hz02 and 1OmM KI IDeits g . . hydrodynamic MI from the apparent sedimentation velocity. the &age 19841. The Subdquent estimation Of partial specific volume sz0 and

viscosity and average density experienced by the sample through the gradient was performed as described (Clarke. 19751.

and analyzed by Pica-tag Compobitionll analysis. Tryptophan vas estimated from the UV h m c t r m of soerm isozvmeb In 0.1N NaOH. 8M urea and their

Amino acid composition: Purlfied proteins were hydrolyzed for 2 4 hours

IMTJNOLCGICAL METHODS All imunologica1 analyses were carried out with afflnity purlfied

antibody obtained against muscle CrK-m, kindly Provided by s . Hauschka, lchamherlain e t . . 1 9 8 5 1 . Imunoblotting was performed as previously described IGundersen 6 Shapira, 1984, accordlng to Tavbin et al., 19791. in which proteins separated by SDS-polyacrylamide gel electra&Gris were

Transphor apparatus. Incubations were performed in Trir buffered saline transferred to nitrocellulose sheets at 0 . 4 Amp for 1 hour in a Hoefer

(TBSI, which conslsted of 1 5 0 mM NaC1, 5 mhl E m A and 5 0 mM Trir, pH 1 . 4 , supplemented with 3 % fetal goat serum, 0.5% gelatin and 0.05% NP-IO. Anti ~ ~ ~ ~ ~ " ~ ~ ~ = ~ ~ ~ ~ ~ ~ ~ ~ , w ~ ~ ~ ~ ~ ~ ~ ~ b ~ ~ ~ :~.3;:5':p;, ,~":"8~~d~~~j:$:~d:,

horseradish peroxidase ICaPPel labs: Cooper Blamedicall. war Used at 1 sglml 11:10001. Peroxidase substrate was 0.8 mslml 4-C1-1-na~thol and the reaction was Initiated with 7.4 mM H202.

CrK-MX was used at 9 ug/ml and all incubations were ~n TBS supplemented with lysine-coated coverslips and fixlng in 100% methanol at -20° for 7 mln. Antl

goat anti-rabbit IqG IZYMEDI. CoYerslips were mounted i n TBS with 20% 5 % fetal goat serum. The secondary antibody VIE rhodamine ITRITcl-conjugated

glycerol and sealed With fingernail polish. Cells were examined u s ~ n g a Leitz epiflvorescence mlcroscape and photographed on Tri-X film, ASA 400 IKadakl.

Immunolocaliration was carried out by adhering sperm to 0.1% poly-L-

RESULTS

WLTRACTBILITY AND SOLUBILIn OF SPERn CrK ISOZYMES

and tail, respectlvely ITombes and Shapiro. 19851. The sperm head CrK is

organelle in the sperm t a l l . Consequently the laozrmes are Called located at the mitachondrlon, as Shown below; the flagellum is the only

mitochondrial and flagellar forms Of CrK. respectively. Table I summarizes

made the heads or tails permeable, so that a maximum amount of the CrK attempts to extract CrK from sperm heads and tails. Several conditions which

activity could be measured, did not release the activity into a 100,000 x g supernatant. The conditions vhlch allowed Sperm CrK activity ea be assayed maximally (buffers contaming 0.05% NP-401 work Similarly for sperm dyneln ATPase actlvlty IGibbons and GlbDonS, 19721. When sperm heads or tails were

detergent-accessed dctiviliy was detected. and it was not soluble. Even if Sonicated or homogenized, ~n the absence of detergent, only SO% of the

extracted. but maximal activity is detected. and then incubated at elevated sperm were permeabrllred with digitonin under Condltmns where CrK is not

lonic strengths. only 20% of the CrK was released (Table 11. Attempts to

Sea urchin sperm have two populatmns Of CrK, located ~n the sperm head

extract sperm head CrK by fechnlques chat were successful for mamnalia" mitochondrial CrK, by Swelling mitochondria ln 50-1OOmM phosphate (Font et - a l . . 1981: Blum et e.. 1983: Farrell et e.. 1972). were unsuccessful f F extracting CTK from Sperm heads, even when all of the activ~ty was made accessible to assay after digitonin treatment.

were when hlgh Concentraelons of the "an-ionic detergents NP-40 or octylglucoside were used. The tall actlvlty required less NP-40 ( 0 . 5 % ) than the head activity 15%) for complete extraction and 150 mM KC1 was required to ensure 100% extractmn of CrK from heads ldata not Shown]. Once extracted,

detergent 01 glycerol, whlle the mitochondrial isozyme required 75mM KC1/10% the flagellar isozyme remalned soluble at 100.000 x g in the absence of salt,

glycerol for complete solubility when detergent was removed. Acetone and KC1 have previously been reported to extract CrK from sea urchin sperm IFujimaki

only 5% Of the maxmal activity that we abtalned by our exfrdcrlon and Ydnagisawa, 1918: Yanagisawa. 19598: 1959C1, but Such conditmns released

condlclons. Detergents have been pre~musIy used to extract mltochondrlal CTK from mammal~an sources lscholte g . , 1973; Wevers et g . , 19811.

The only condrtions under Which CrK could be extracted from the sperm

Table I

Heads and tails I@ 10 mg sperm proteinlmll prepared as described in SEA URCHIN SPERM CREATINE KINASE EXTRACTABILITY AND ACCESSIBILITY

Experimental Procedures were homogenized for t en strokes wlth a W m c E homogenizer with the Indicated agents. The 8 BCtIVlty in the cell extracts was compared to the maximal actlvlty obtalned after treatment wlrh 0.05% NP- 40. Soluble actlvlty was assessed as the % Of actlvlty which remained in the supernatant solution after centrlfugatlon at 100.000 x g for 1 hour.

INCUBATION CONDITION

SONICATED OR HOMOGENIZED +I- 0.5M KC1

THEN: l O O m M Pi, pH 9.0 0.28% DIGITONIN.

500mM KC1 2 M KC1

0 . 0 2 % Occylglucaaide/ 0 . 5 M KC1

NP-40 : 0.1% 0.5% 1.0% 5.0%

% CRX SOLUBILIZED

3 4

3 4 8

14 8 20 17

7 7

6 10 45 95

25 85

95

PURIFICATION OF CrK ISOZYMES FROM SPERM. Both isozymes are lablle once extracted from the cell. but thelr

activity 1s preserved by low concentrations of any of severa l adenine nucleotides. including ATP, AMP and AMPPNP. ADP was the most effective

direction used for estimating purify. Glycerol IlOIl was also included in the stabilizer and did not interfere in the typical CrK assay in the reverse

buffer because it Stabilized activity and/or prevented "on-specific adherence to chromatographic resins such as Sepharoce and agarose. Flagellar CrK was inactivated irreverriblv bv 0.5mM divalent cations lZn2+. cu2+ or Cd2+l. but

extracis, but in partially 01 completely purified preparations Of-CrK, no combination Of phenylmethyl sulfonyl fluorrde. rayhean trypsin inhibitor. leupeptin or Pepstatin prevented activlcy losses. Thus, the EpeCifIC activities reported far purified isozymes could be even higher, considering that recoveries for the flagellar and mitochondrial CrK Were 21% and 7%.

As previously reported. Procion Red-agarose was chosen as the first chromatographic step because the lrorymes could be Separated by either KC1 step elutions ITombeS and Shapiro, 19851 or by linear KC1 gradients Idata not Shown), and because the NP-40 could be removed from the protein by extenszve (10-column volumel washes. Using KC1 gradients at pH 1 . 0 , the flagellar isozvme eluted from Procion Red at O.3M KC1. while the head isorvme was elut;d at 0.6M KC1. The flagellar isozyme w i s subsequently purifled to

with 1 linear 0 to 0.5M KC1 gradlent (FLg. lAl. apparent homogeneity by chromatography on DEAE-agarose a t pH 8.0 and elutlan

Sea Urchin Sperm Creatine Kinase Isozymes 16019 " Table I1

WLUyf 1-1

Figlire 1. Elution Profile of CrK isozymes from DEAE-agarose. A: Flagellar CrK, B: sperm head CrK. CrX activlty is represented as (-1 and protein concentration is represented as 0 1 . KC1 concentratmn is shwn as a solid line.

Peak into Pmcion Red buffer containing l O m M KC1 and 0% glycerol, Whereupon The sperm head isozyme vas purified by dialysls of the procion ~~d

centrifugation at 10,000 x 4 for 10 minutes. Conraminatlng flagellar crK It became insoluble but remained fully active and could be collected by

activity was not precipitated, as observed by SDS-PAGE. Other COntamlnating

eluted at sliqhtly higher concentrations Of K C 1 10.1SMI than that requlred Proteins were removed by DEAE chromatography at pH 8.0. Sperm head CTK

to elute the flagellar isozyme lO.llM1, 1114. lBl. This behavlor is as expected since mltochandrial CrK has a Slightly mare acidlc PI 16.21 than flagellar CrK 16.45) as determlned by two-dirnensmnal PAGE (data not Ehwni. ~elther of the other two proteins which were labelled with FDNB in

CIK shown nele ( s e e below]. whole heads ITombeS and Shaplro, 19851 copurifled with the mitachondrlal

PURIFICATION OF CREATINE KINASE FROM SPERM nsms ~VID TAILS and CrK was pur~fied from each fraction as descrlbed In Experlmental Procedures.

Heads and tails were prepared from 50 ml dry sperm 12.5 x l0l2 sperm)

Ims) IUI (%I (-FOLD1 FRRCTION PROTEIN ACTIVITY RECOVERY SP.ICT. PURIFICATION

TAItS 0.5% NP-40 951 10,500 1 0 0 11.0 HOmO4e"?.te

105x4 Super- 2 4 2 10,350 99 42.8 3 . 9

1.0

rate

Pmcion Red 28 3,300 31 118 10.7 Peak

D%E Peak 8.9 2 , 0 9 0 21 2 3 5 21 ~~ ~

HEADS ~

5% NP-40 3970 22,500 100 5.1 1 HOm04e"dtF

1 0 5 ~ ~ super- 1180 22,000 98 nate

18.7 3.3

Procion Red 385 1 2 . 7 5 0 51 3 3 Peak

Glycerol and 52 7,700 3 4 KCi-Free Pellet

1 4 9 26.1

DEAE Peak 8.8 1 .580 1 180 3 1 . 4

5.8