341

Metabolism of Arachidonate in Rat Testis: Characterization of 26-30 Carbon Polyenoic Acids W. McLEAN GROGAN, Department of Biochemistry. Box 614 MCV Station. Medical College of Virginia, Virginia Commonwealth University, Richmond, VA 23298 ABSTRACT

Fatty acid methyl esters of long-chain polyenoic fatty acids (LCPA) from rat testis injected with [1-14C] araehidonate were analyzed and separated by reversed-phase high performance liquid chromatography (RP-HPLC). Earlier, all previously identified LCPA were prepared in high purity along with 4 previously unidentified fatty acids, which were further characterized by capillary gas chromatography (GC), catalytic hydrogenation and alkaline isomerization. Unidentified fatty acids proved to be 26:4, 26:5, 28:5 and 30:5. All of these LCPA incorporated 14C from arachidonate (20:4) to specific activities that were comparable to that of 20:4 and previously identified metabolites of 20:4 and much greater than specific activities of 18:1n-9 or 22:6n-3. LCPA were analyzed on a capillary GC system capable of resolving known cis-trans and positional isomers of the n-3, n-6, n-7 and n-9 families of unsaturated fatty acids. Log plots of isothermal retention times and normal plots of temperature programmed retention times were linear (r = 0.999) in carbon number when values for known and previously unidentified LCPA of 4 or 5 double bonds, respectively, were coplotted. Thus, the newly identified fatty acids belong to the n-6 family of fatty acids synthesized from arachidonic acid. Lipids 19: 341-346, 1984. INTRODUCTION

Long-chain polyenoic acids (LCPA) with chain lengths in excess of 22 carbons have been reported in the testes of several species (1-4). The 24 carbon tetraene (24:4n-6) (by this convention, 24 carbons, 4 double bonds and 6 carbons from the methyl end to the highest numbered double bond (n-6), i.e., all cis-9, 12, 15, 18-tetracosatetraenoic acid) and pentaene (24:5n-6) are biosynthesized in the testis from arachidonic acid (20:4n-6) by a series of elongation and desaturation reactions. The presence of 26 carbon fatty acids has been suggested by gas cttromatographic (GC) analysis of catalytically hydrogenated methyl esters of fatty acids extracted from rat testes (2). Incubation of mouse testes or intratesticular injection of mice with [ 14C] arachidonate resulted in incorporation of 14C into fatty acids with GC retention times greater than those of k n o w n 22 or 24 carbon polyenoic acids (5). Thus, circumstantial evidence that the testis biosynthesizes unsaturated fatty acids with chain lengths in excess of 24 carbons exists, but the specific fatty acids have not been isolated or identified. Because the number of potential isomers of LCPA having similar retention times becomes quite large as chain lengths increase, GC alone can provide only a tentative identification for a previously u n k n o w n LCPA (6). In the present study, we have used analytical and preparative reversed-phase high performance liquid chromatography (RP-HPLC) to confirm tentative identifications by capillary GC of 26, 28 and 30

carbon LCPA. We also provide evidence that these fatty acids are members of the n-6 family of LCPA, derived from arachidonic acid. METHODS AND MATERIALS Preparation of Fatty Acid Methyl Esters (FAME)

Total lipid was extracted with CHCla/MeOH (1:1, v/v) from decapsulated testes of adult male Sprague-Dawley rats (150-200 g; Flow Laboratories, Rockville, MD) by the procedure of Bridges and Coniglio (7). Extracts were reduced to dryness under vacuum on a rotary evaporator, redissolved in CHCla, dried under a stream of N2 and transmethylated with NaOCH3 by the procedure of Grogan et al. (4) to yield FAME. For studies ofarachidonate metabolism, rats were anesthetized with ether and injected intratesticularly with 50 ~Ci/testis [1-14C] arachidonate (New England Nuclear, Boston, MA, 56 mCi/mmol) as the albumin complex. After 48 hr, the rats were sacrificed by cervical dislocation and testes were removed for preparation of FAME. High Performance Liquid Chromatography (HPLC)

FAME (50-500 ~g) were dissolved in 10-100 /al acetonitrile (AcCN) and separated on an HPLC system (Waters Assoc., Framingham, MA) consisting of 6000A pumps, U6K injector, 660 gradient programmer and a 7.8 mm x 30 cm reversed-phase column (/~Bondapak Cla). Effluent was monitored on a Schoeffel SF770 variable wavelength absorbance detector and LIPIDS, VOL. 19, NO. 5 (1984)

342

W.M. GROGAN

collector. FAME were eluted using a nonlinear g r a d i e n t ( n o . 5 on the Waters 660) of 60-100% AcCN (solvent B) in 15% M e O H - H 2 0 (solvent A) for 1 hr at 2 ml/min. The gradient was begun 10 min after injection. A second protocol used to separate 18:2 from 22:5 consisted of a 66-100% gradient of MeOH in H 2 0 for 1 hr at 2ml/min. For measurement of radioactivity in eluent, fractions were collected at 0.5 min intervals into scintillation vials that were subsequently filled with scintillation fluid (Budget-Solve, Research Products International Corp.) for liquid scintillation counting. Counts were checked for quenching by the channels ratio method. For GC or chemical characterization, fractions were collected into 50 ml glass stoppered tubes, diluted 3-fold with H 2 0 and extracted 3• with 0.5 vol petroleum ether to recover FAME. FAME that had been previously characterized were identified b y GC retention times. All solvents were HPLC grade. H 2 0 was 0.5 Mohm distilled H2 O redistilled in glass. Gas Chromatography

Capillary GC of FAME was carried out on a Packard 427 GC equipped with flame ionization detector (FID), a splitless injector and a 0.75 m m x 60 m fused silica capillary column coated with 0.2 /am SP2340 (Supelco, Inc., Bellefont, PA). Argon carrier gas was used at a linear flow rate Of 20 cm/sec. Column efficiency was in excess of 1,200 plates/m, sufficient for baseline resolution of the n-3 and n-6 isomers of 22:5. For temperature programmed analysis, a linear gradient from 120-260 C at 2 C/min was used with an initial time of 10 min. FAME were identified by comparing of retention times with those of commercial standards and biological samples of previously determined composition (5). Chemica(Characterization of LCPA

Aliquots of unsaturated FAME purified to 99% by preparative HPLC were catalytically hydrogenated b y the method of Farquhar et al. (8) to yield corresponding saturated methyl esters. These were subsequently indentified b y isothermal (180 C) GC as described above. Double-bond content of purified LCPA was determined by spectrophotometric analysis of alkaline isomerized aliquots. Petroleum ether extracts were dried under N2, dissolved in 10% KOH-glycerol and heated to 180 C for 45 min as described in detail by Holman and Hays (9). Isomerized samples were analyzed for maximum wavelength of absorbance (Xmax) on a Cary 210 recording spectrophotometer (Varian Assoc., Palo Alto, CA) scanning 220-400 nm. LIPIDS, VOL. 19, NO. 5 (1984)

Unknowns were assigned double-bond values on the basis of previously reported Xmax values (9,10) and by comparing them with known LCPA simultaneously subjected to the same analysis. R ESU LTS

Total lipid extract from rat testis was transesterified to yield methyl esters of component fatty acids (FAME) and analyzed by RP-HPLC. Figure 1A shows the elution profile obtained by monitoring effluent for absorbance at 215 nm. All major unsaturated fatty acids were resolved by a single pass on this system with the exception of 18:2 and 22:5, which overlapped completely in retention time. Rechromatography, using the methanol-water gradient, completely resolved these fatty acids (relative retention times were 1.00 and 1.03, respectively), yielding 22:5 with no detectable impurities by HPLC or GC (data not shown). Only unsaturated FAME were detected by this method because detector response was roughly proportional to degree of unsaturation, although not in a linear mode (Table 1). This is best illustrated by a comparison of the percentage of area under the absorbance curve obtained for 18:1 by HPLC with the actual percentage of mass obtained by GC. For each fatty acid, detector response was proportional to concentration up to 5 /ag of fatty acid injected. Figure 2 shows a chromatogram obtained by capillary GC of the same FAME sample. This system resolved the n-7 and n-9 isomers of 16:1, the cis-trans isomers of 18:1n-9 and the n-3 and n-6 isomers of 22:5, as well as achieving baseline resolution of all commonly occurring fatty acids. The differing modes of separation of HPLC and GC are apparent in the fact that retention times are inversely proportional to the degree of unsaturation on the hydrophobic HPLC column and proportional to the degree of unsaturation on the polar GC column, given the same number of carbons in the respective FAME (Table I and Fig. 3). Retention times are proportional to carbon number with b o t h methods, although the principle of separation is based on hydrophobicity in the case of HPLC and boiling point (molecular weight) and polarity in the case of GC. With the use of programmed decreases in solvent polarity (HPLC) or increases in column temperature (GC), the relationship between carbon number and retention time was linear (Correlation coefficient, r>0.999, in every case) for each homologous series (4 or 5 double bonds, respectively) allowing tentative identification of unknown FAME.

343

VERY LONG CHAIN POLYENOIC ACIDS ~n eu

A

w (.)

z tln

mE

_.J W

eJ oJ

I

,6 a

I

I

1

'PEAK 9 e3,000 t ~ r ~o

t4

o

J 'O

19

16

x

i

i

24

32

i

40

48

i

i

52

64

RETENTION TIME (mln)

E

a. t o N

FIG. 2. Capillary GC of FAME from rat-testis lipids. FAME were prepared by transesterification of total lipid extracts and eluted by a linear temperature program from 120-260 C. Detector response is proportional to mass.

>-

F--pO

~6 N 4.

2

i

0

tO

2'0

3'0

40

50

60

70

BO

RETENTION TIME (ram) FIG. 1. RP-HPLC elution profile of unsaturated

fatty acid methyl esters from rat-testis lipids. FAME were prepared by transesterification of total lipid extracts from rats injected intratesticularly with [1,1'C] arachidonate. FAME were eluted by a 1 hr nonlinear gradient of AcCN in 15% McOII-H20 as described in Methods. (A) Absorbance of eluent monitored at 215 nm. (B) Radioactivity of fractions collected at 0.5 min intervals. Identification of Unknown Fatty Acids

Unsaturated F A M E coeluting with each major HPLC peak were collected and analyzed by capillary GC. Their relative r e t e n t i o n times are given in Table 1. Most F A M E could be

identified by comparing r e t e n t i o n times with those of standards or previously identified F A M E from rat or mouse testes. Standards were unavailable for > 2 4 carbon unsaturated f a t t y acids. Aliquots of u n k n o w n F A M E were hydrogenated to corresponding 26, 28 and 30 carbon F A M E and analyzed by GC. Aliquots of F A M E were also isomerized by alkali to determine s p e c t r o p h o t o m e t r i c a l l y the n u m b e r of double bonds present. As shown in Table 2, all 4 unk n o w n F A M E were isomerized to c o m p o u n d s absorbing with k m a x of 317 or 348 nm, indicating 4 or 5 double bonds, respectively. Characteristic k m a x were also obtained for o t h e r F A M E prepared by HPLC, consistent with results reported by previous investigators using different preparative techniques and preparations of lesser purity (10,11 ). Purified F A M E were analyzed under isot h e r m a l conditions by a capillary GC system capable of resolving n-6 f r o m n-3 and n-9 fatty acids. As shown in Figure 3, log plots of retenLIPIDS, VOL. 19, NO. 5 (1984)

W.M. GROGAN

344 1.2

TABLE 2

9 5 double bonds 1 4 double bonds AhydrocJenoted

/

1.0

~.8

//"

Characteristics of Previously Identified and Unknown LCPA Isolated by HPLC a

/" Fatty acid ....."

.2

20

22

24

26

28

30

32

i 34

20:3 22:4 22:5 22:6 24:4 24:5 26:4 26:5 28:5 30:5

Previously identifiedb x x x x x x

H 2 FA

hmax (rim)

20:0 22:0 22:0 22:0 24:0 24:0 26:0 26:0 28:0 30:0

279 317 348 374 317 348 317 348 348 348

CARBON NUMBER FIG. 3. Linear plots of isothermal GC log relative retention times vs carbon number for methyl esters of 20-30 carbon LCPA isolated by preparative HPLC. Retention times are relative to 20:4 for 4 and 5 double-bond LCPA and relative to 20:0 for hydrogenated samples. Points plotted beyond 30 carbons represent peaks detected by GC, but not further characterized. Correlation coefficient, r ~ 0.999 in each case.

TABLE 1

aH 2 FA = fatty acid identified by GC after catalytic hydrogenation; hmax = maximum wavelength of absorbance following alkaline isomerization of double bonds. bSee references 1, 10, 11. m a r k a b l y linear in c a r b o n n u m b e r . Peaks were also seen having r e t e n t i o n t i m e s a p p r o p r i a t e f o r 32 c a r b o n LCPA o f 4 and 5 d o u b l e b o n d s , a l t h o u g h t h e s e were n o t f u r t h e r c h a r a c t e r i z e d . Biosynthesis of LCPA from Arachidonic Acid

Figure 1 B s h o w s t h e d i s t r i b u t i o n o f radioactivity in f r a c t i o n s c o l l e c t e d during HPLC s e p a r a t i o n o f F A M E p r e p a r e d f r o m rat testis llpids 48 h r a f t e r i n t r a t e s t i c u l a r i n j e c t i o n o f GCa HPLC b [ 1-14 C] a r a c h i d o n a t e . R a d i o a c t i v i t y was incorRel. rot. Percentage Rel. ret. Percentage p o r a t e d i n t o 2 2 : 4 , 2 2 : 5 , 24:4 and 2 4 : 5 , w h i c h Fatty acid time of mass time of mass are k n o w n t o be s y n t h e s i z e d f r o m 2 0 : 4 in t h e rat (2), and i n t o 2 6 : 4 , 2 6 : 5 , 28:5 and 3 0 : 5 , 18:1n-9 0.70 22 1.23 1.2 t h e n e w l y i d e n t i f i e d LCPA. With t h e e x c e p t i o n s 18:2n-6 0.76 7.8 1.07 o f 22:5 and 3 0 : 5 , specific radioactivities o f 20:3n-6 0.97 1.6 1.11 1.5 20:4n-6 1.00 29 1.00 27 LCPA, w h i c h i n c o r p o r a t e d 14C, were quite 22:4n-6 1.16 2.9 1.16 4.0 similar to t h e specific r a d i o a c t i v i t y o f 20:4 22:5n-6 1.19 30 1.07 50 c f r o m w h i c h t h e label was d e r i v e d (Table 3). 22:6n-3 1.26 1.4 0.96 1.1 24:4n-6 1.29 1.7 1.33 2.6 Specific radioactivities o f all LCPA derived 24:5n-6 1.33 2.1 1.21 3.6 f r o m 2 0 : 4 were 2- t o 15-fold higher t h a n t h a t 26:4n-6 1.42 0.2 1.52 0.2 o f 18 : 1, w h i c h was labeled during de n o v o 26:5n-6 1.46 0.3 1.37 0.6 s y n t h e s i s f r o m a c e t a t e derived f r o m fl-oxidation 28:5n-6 1.58 0.3 1.54 0.8 30:5n-6 1.70 0.4 1.71 0.3 o f 20:4 or a m e t a b o l l t e . The 2 2 : 6 , w i c h is t h e p r e d o m i n a n t m e m b e r o f t h e n-3 family o f aTemperature programmed 120-260 C at 2 C/min. L C P A , c o n t a i n e d n o d e t e c t a b l e 14C. The 2 2 : 5 , Retention times are relative to 20:4. FID response is linear with respect to mass. Percentage calculation w h i c h c o n s t i t u t e d 30% o f t h e u n s a t u r a t e d f a t t y acid (Table 1), c o n t a i n e d o n l y 11% o f t o t a l includes only unsaturated fatty acids. bSolvent programmed 60-100% AcCN in 15% 14C, b u t a c c o u n t e d f o r a c o r r e s p o n d i n g 28% MeOH-HzO. Retention times are relative to 20:4. Absorbance at 215 nm is dependent on degree of o f 14C in f a t t y acid m e t a b o l i t e s o f 2 0 : 4 (Table 3). In c o n t r a s t , 2 2 : 4 c o n s t i t u t e d o n l y 3% o f unsaturation. CIncludes 18:2. u n s a t u r a t e d f a t t y acid b u t c o n t a i n e d 10% o f t o t a l t4C and 25% o f 14C in m e t a b o l i t e s o f 20:4. Thus, t h e specific r a d i o a c t i v i t y o f 2 2 : 4 , t i o n t i m e s f o r F A M E o f k n o w n n-6 f a t t y acids w h i c h was t y p i c a l o f specific radioactivities o f of 4 and 5 double bonds and those newly iden2 0 : 4 and t h e 24-28 c a r b o n m e t a b o l l t e s , was tified as 2 6 : 4 , 26:5, 28:5 and 30:5 were re7-fold higher t h a n t h a t o f 22:5. A Comparison of Capillary GC and HPLC Analyses of LCPA

LIPIDS, VOL. 19, NO. 5 (1984)

VERY LONG CHAIN POLYENOIC ACIDS

345

DISCUSSION

TABLE 3

The presence of 26 carbon fatty acids in testis has been suggested by GC analysis of hydrogenated fatty acid methyl esters prepared from rat-testis lipids (2). Presence of 14C in FAME having GC retention times greater than those of known LCPA following intratesticular injection of mice or incubation of mouse testicular cells with [ 14C] arachidonate suggested the biosynthesis of metabolites of arachidonate with chain lengths greater than 24 (5). In the present work, we have isolated and characterized 4 of these metabolites from rat testis, along with previously identified LCPA, and presented evidence that they are also derived from arachidonate. LCPA of the 4 double-bond series have maximum wavelengths of absorbance near 317 nm following alkaline isomerization, characteristic of 4 double bonds (9-11), and yield saturated fatty acids of 20-26 carbons following catalytic hydrogenation. LCPA of the 5 doublebond series have maximum wavelengths of absorbance near 348 nm following alkaline isomerization, characteristic of 5 double bonds, and yield saturated fatty acids of 22-30 carbons folowing hydrogenation. The newly identified fatty acids are 26:4, 26:5, 28:5 and 30:5. Identifications of these LCPA are further corroborated by linear plots of GC (isothermal and programmed modes) and HPLC retention times vs carbon number in each homologous series. Because the capillary GC column used was able to resolve completely the n-6 and n-3 isomers of 22:5, the newly identified LCPA are almost certainly of the n-6 family derived from linoleate by way of arachidonate. This assignment is further supported by the incorporation of 14C from arachidonate into each of the newly identified LCPA at specific radioactivities similar to those of 20:4, 22:4, 24:4 and 24:5 n-6 and much higher than the specific radioactivities of 18:1 or 22:6, which would only become labeled by de n o v o synthesis from acetate resulting from degradation of 20:4 or its metabolites. The major n-3 LCPA, 22:6, incorporated no detectable radioactivity. RP-HPLC proved to be a highly efficient method for both analysis and preparation of each of the newly identified LCPA as well as those identified by previous investigators. Whereas conventional chemical techniques were used to support the identification of previously u n k n o w n LCPA, information derived from preparative HPLC and capillary GC is sufficient for unambiguous identification of the u n k n o w n components. A single pass on the HPLC was sufficient to purify most LCPA to 99% homo-

Relative Specific Radioactivities of Metabolites of [ 1-14C] Arachidonate 48 hr After Intratesticular Injection Fatty acid 18:1 20:3 20:4 22:4 22:5 22:6 24:4 24:5 26:4 26:5 28:5 30:5

Percentage of total radioactivity

Relativespecific radioactivity

6

0.1

1 63 10 11 0 S 4 0.4 1 1 0.3

0.3 1.0 1.5 0.2 0 1.3 0.9 0.9 1.5 1.5 0.4

aRats were injected intratesticularly with [1-14C]arachidonate. After 48 hr, lipids were extracted and transesterified to methanol. Methyl esters were isolated by HPLC and assayed for radioactivity by scintillation counter. Relative specific radioactivity -- percentage radioactivity + percentage mass, relative to 20:4.

geneity. In cases where a partial or complete overlap of LCPA with another component took space (e.g., 22:5 and 18:2), rechromatography of the collected fraction with the alternate solvent system was sufficient to achieve complete homogeneity as judged by GC analysis. The range of elongation of n-6 LCPA by rat testis is now at least 30 carbons and minor mass peaks from GC suggest the presence of LCPA with even greater chain lengths. These fatty acids of greater than 24 carbons comprise 1% or more of total fatty acid in the testis and are synthesized from arachidonate at rates that are sufficient to result in 48 hr equilibration of 14C label between most of the precursor-intermediate-product LCPA pools involved. The specific radioactivity of 22:5 suggests a pool slowly equilibrating, probably because of the size and stability of the 22:5 pool. Accumulation of 14C in 22:4, 24:4 and 26:4 suggests that the A4 desaturase, which catalyzes the reaction 22:4 --> 22:5, is relatively slower than the elongation activities. The 24-30 carbon 5 doublebond homologs are probably also derived by sequential elongations from 22:5, since the A6, AI0 and A12 desaturases required for biosynthesis from the corresponding 4 double-bond analogs are nowhere else in evidence. A A8 desaturase has been shown to participate in an alternate pathway for biosynthesis of arachidonate in rat testis (12). However, this desaturase probably would not act only on 20 and 26 carbon fatty acids. LIPIDS, VOL. 19, NO. 5 (1984)

W.M. GROGAN

346

Lack o f a c c u m u l a t i o n of t h e 24-30 c a r b o n L C P A in t h e testis d e s p i t e r a p i d e q u i l i b r a t i o n o f 14C a m o n g t h e 2 0 - 3 0 c a r b o n L C P A suggests a relatively h i g h r a t e o f t u r n o v e r o f t h e s e LCPA, caused e i t h e r b y o x i d a t i o n or r e t r o c o n v e r s i o n t o s h o r t e r - c h a i n analogs, as has b e e n d e m o n s t r a t e d f o r 22 c a r b o n L C P A b y several investig a t o r s (13-16). U n l i k e 2 2 : 5 , t h e 2 4 - 3 0 c a r b o n LCPA are p r o b a b l y n o t p r e s e n t in q u a n t i t i e s s u f f i c i e n t t o m a k e t h e m i m p o r t a n t storage f o r m s o f 2 0 : 4 . H o w e v e r , as h a s b e e n r e c e n t l y r e p o r t e d for 2 2 : 6 n - 3 (17), t h e s e L C P A m a y m o d u l a t e b i o s y n t h e s i s o f biologically active m e t a b o l i t e s o f t h e 20 c a r b o n f a t t y acids o r even give rise to biologically active m e t a b o l i t e s themselves.

3. 4. 5. 6. 7. 8. 9. 10. 11.

ACKNOWLEDGMENTS

12.

This work was supported by a grant from the National Institutes of Health, HD 13019. Gratitude is expressed to Ellen G. Huth for excellent technical assistanee.

13.

REFERENCES 1. Bridges, R.B., and Coniglio, J.G. (1970) J. Biol. Chem. 245, 46-49. 2. Coniglio, J.G., Whorton, A.R., and Beckman, J.K. (1977) in Function and Biosynthesis of

LIPIDS, VOL. 19, NO. 5 (1984)

14. 15. 16. 17.

Lipids (Bazan, N.G., Brenner, R.R., and Giusto, N.M., eds,) pp. 575-589, Plenum Publishing Co., New York. Holman, R.T., and Hofstetter, H.H. (1965) J. Am. Oil Chem. Soc. 42,540-544. Grogan, W.M., Farnham, W.F., and Szopiak, B.A. (1981) Lipids 16,401-410. Grogan, W.M., and Huth, E.G. (1983) Lipids 18, 275-284. Hofstetter, H.H., Sen, N., and Holman, R.T. (1965) J. Am. Oil Chem. Soc. 42, 537-540. Bridges, R.B., and Coniglio, J.G. (1970) Lipids 5,628-635. Farquhar, J.W., Insull, W., Jr., Rosen, P., Stoffel, W., and Ahrens, E.H., Jr. (1959) Nutr. Rev. 17 (suppl.), 1-29. Holman, R.T., and Hayes, H. (1957) Methods Biochem. Anal. 4, 99-138. Grogan, W.M., Coniglio, J.G., and Rhamy, R.K. (1973) Lipids 8,480-482. Davis, J.T., Bridges, R.B., and Coniglio, J.G. (1966) Biochem. J. 98, 342-346. Albert, D.H., and Coniglio, J.G. (1977) Bioehim. Biophys. Acta 489,390-396. Verdino, B., Blank, M.L., Privett, 0.S., and Lundberg, W.O. (1964) J. Nutr. 83,234-238. Schlenk, H., Sand, D.M., and Gellerman, J.L. (1969) Biochim. Biophys. Acta 187,201-207. Sprecher, H. (1967) Biochim. Biophys. Acta 144, 296-304. Bridges, R.B., and Coniglio, J.G. (1970) Biochim. Biophys. Acta 218, 29-25. Aveldano, M.I., and Spreeher, H. (1983) J. Biol. Chem. 258, 9339-9343. [ R e c e i v e d O c t o b e r 14, 1 9 8 3 ]