Chemical modification of dipeptidyl peptidase IV: Involvement of an essential tryptophan residue at the substrate binding site

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 234, No. 2, November 1, pp. 622-623,1934

Chemical Modification of Dipeptidyl Peptidase IV: Involvement of an Essential Tryptophan Residue at the Substrate Binding Site MINORU

HARADA,l

Department

B. YUKIHIRO HIRAOKA, KAYOKO AND KATSUHIKO FUKASAWA

of oral

Biochemistry,

Received

April

Md.sumoto

Dental

19, 1934, and in revised

College, form

July

M. FUKASAWA,

Shiqfiri

39%07, Japan

3, 1964

Inactivation of pig kidney dipeptidyl peptidase IV (EC 3.4.14.5) by photosensitization in the presence of methylene blue at pH 7.5 was observed to have pseudo-first-order kinetics. During the process, until over 95% inactivation was achieved, the histidine and tryptophan residues were decreased from 14.0 to 2.7 and 12.6 to 7.1, respectively, per 94,000-Da subunit, without any detectable changes in other photosensitive amino acids. Modification of four histidine residues per subunit using diethylpyrocarbonate resulted in only 30% inactivation of the enzyme, while N-bromosuccinimide almost completely inactivated the enzyme with the modification of only one tryptophan residue per subunit, as determined by absorption spectrophotometry at 280 nm. The protective action of the substrate and inhibitors such as Ala-Pro-Ala and Pro-Pro against the modification of tryptophan residues with N-bromosuccinimide was observed both fluorometrically and by measurement of activity. On the basis of these results it is suggested that one of the tryptophan residues in the enzyme subunit is essential for the functioning of the substrate binding site of pig kidney dipeptidyl peptidase IV. Q1~8.1 Academic PWS. IX

Dipeptidyl peptidase IV (DPP IV) (EC 3.4.14.5) was discovered in rat liver and acylase I preparations from pig kidney by Hopsu-Have and Glenner (l), and was designated as glycylproline+naphtylamidase. In the group of dipeptidyl peptidases, the unique substrate specificityX-Pro(Ala)-Y(X and Y for amino acids)-which occurs at a slightly alkaline pH, has been well documented (2-11). Few studies on the amino acid residues which comprise the active site of the enzyme have been reported, except for those showing serine involvement by inhibitory studies with diisopropylfluorophosphate (5,6). In our previous report we suggested that the active site of the enzyme involves a hydrophobic interaction of the enzyme with the proline at the Pl position in the substrates as determined by an inhii To whom 0003-9861/34 Copyright All rights

correspondence

should

$3.00

Q 1984 by Academic Press, Inc. of reproduction in any form reserved.

bition study using proline-containing peptides (12). In the present study we report the chemical modification of pig kidney DPP2 IV by three different methods, and provide evidence suggesting that tryptophan and histidine residues might be involved in the catalytic action of DPP IV. The interaction of the proline (or alanine) residue of the substrate with a tryptophan residue in the active site of the enzyme is demonstrated by the protective action of substrate or inhibitor against a chemical modification with N-bromosuccinimide. MATERIALS

AND

METHODS

Pig kidney purified dipeptidyl prepared according to the previous

peptidase IV was report (8) without

t Abbreviations used: DPP dipeptidyl peptidase; NBS, N-bromosuccinimide; DTNB, 5,5’-dithiobis(2nitrobenzoic acid); DEPC, diethylpyrocarbonate; DFP, diisopropylfluorophosphate.

he addressed. 622

CHEMICAL

MODIFICATION

OF

the final affinity chromatography step. The specific activity of the enzyme was 52.0 units/mg protein. Gly-Pro-pNA was kindly provided by the Ajinomoto Company. L-Prolyl-L-proline (Pro-Pro) and L-alanyl-L-prolyl-L-alanine (Ala-Pro-Ala) were purchased from Bachem. N-Bromosuccinimide (NBS) and 5,5’-dithiobis(2-nitrobenxoic acid) (DTNB) were purchased from Nakarai Chemical Company. Diethylpyrocarbonate (DEPC) was purchased from Sigma Chemical Company. Mercaptoethanesulfonic acid (3 N) was obtained from Pierce Chemical Company. All other chemicals used were reagent grade. Enzymatic activity was determined by the method of Nagateu d d (13). using Gly-Pro-pNA as substrate, except that 10 mM Tris-HCl buffer (pH 8.0) and a scaled-down volume of the assay mixture to 0.20 ml were used. Protein concentrations were measured as described by Hartree (14) except for a scaled-down mixture volume. Pig kidney purified DPP IV was inactivated by photosensitization (15). The mixture consisted of an accurately determined amount of enzyme (6 to 7 mg) and 6.0 X lo-’ M methylene blue in 10.0 ml 0.05 M potassium phosphate buffer, pH 7.8, and it was exposed with constant stirring to a 300-W tungsten lamp at a distance of 40 cm at 25°C. Samples (2.4 ml) were withdrawn at 0,4,14, and 25 min. then an aliquot (5 ~1) of each was used for the determination of residual activity after adequate dilution for the assay. Most of the sample. volume was dialped against distilled water, lyophilised, and then used for amino acid analysis. Amino acid analyses were performed with a JEOL Model 6AH automatic amino acid analyser after hydrolysis of the lyophilised material (0.5-0.7 mg) for 24 h at 110°C in 3 N mercaptoethanesulfonic acid in a sealed tube ,in ‘u(ccuo (16). Cysteine content was separately determined using 0.5-0.7 mg of the lyophilized sample by a calorimetric method with DTNB (17). As the control experiment, the same reaction was also performed under dark conditions for 25 min. DEPC was diluted to a 100 rnlld concentration with acetonitrile for a stock solution. The modification of the enzyme was performed as follows (18): Two cuvettes containing 1.13 mg of the enzyme in 2.35 ml 0.1 M potassium phosphate buffer, pH 6.5, were placed in the sample and reference compartments of a Hitachi A220 double-beam spectrophotometer at room temperature. Difference spectra were recorded between 209 and 280 nm before and at various time intervals after the addition of 0.15 ml 10 mM DEPC in 95% ethanol to the sample cuvette and an equal volume of 95% ethanol to the reference cuvette. Ten-microliter aliquots of the reaction mixtures were removed immediately after each difference spectrum was recorded, and diluted with 1.0 ml 50 rnlld Tris-HCl buffer, pH 8.0, in dialysis tubing. Then

DIPEPTIDYL

PEPTIDASE

IV

623

they were dialyzed against the same buffer. The relation between activity and the number of histidy1 residues modified was calculated from the absorbance at 240 nm (Ab = 3200 cm-’ I&-‘). Conversion of the indol chromophore of tryptophan to oxyindole, which absorbs much less at 280 nm, was performed as described by Spande and Witkop (19). In a typical experiment, 2.5 ml 0.1 M potassium phosphate buffer, pH 6.4, containing an accurately determined amount of enzyme (0.69 mg), was reacted in a l.O-cm quartz cell. Changes in the absorption spectrum and loss of activity were followed after successive additions of lO-p1 aliquots of a 10 mM aqueous solution of NBS. Changes in the absorption spectrum were recorded 2 min after the addition and adequate mixing of NBS by a Hitachi A220 double-beam spectrophotometer equipped with a digital recorder for absorbance measurement. After 4 min, 10 pl of the reaction mixture was removed and diluted to 1.0 ml with 0.1 M Tris-HCl buffer, pH 8.0, in ice-cold water and then used for measurement of activity. Fluorescence experiments were carried out on a Hitachi fluorescence spectrophotometer, Model 650lOS, equipped with a circulating water bath to maintain the temperature at a constant 25°C. Excitation was at 295 nm (band width, 2 nm) to ensure that the absorption was due to tryptophanyl groups. Emission was recorded in the range 260-506 nm (band width, 3 nm). Solutions of 0.6 ml in 0.1 M potassium phosphate buffer, pH 6.4, containing an exactly determined amount of enzyme (13 fig). were used. Emission spectra of DPP IV modified with NBS in the presence or absence of protective peptides (substrate and inhibitor) and amino acids were recorded. Aliquots (10-26 ~1) of the mixture were diluted 20 times with 0.1 Y Tris-HCl buffer, pH 8.0, and used for the determination of residual activity after the spectral measurements. RESULTS

When pig kidney DPP IV was exposed to light in the presence of 6.0 X lo-’ M methylene blue in 50 mM potassium phosphate buffer, pH 7.8, inactivation of the activity was observed, and the inactivation process closely followed pseudo-first-order kinetics until over 95% inactivation had occurred (Fig. 1). No inactivation was observed in the dark condition. An apparent inactivation rate constant, k, was calculated to be 0.246 min-‘. Complete amino acid analyses of enzymes photooxidized for three different lengths of time demonstrated that the inactivation of the enzyme was mainly associated with the de-

HARADA 100

‘0 10 Photoinactivation

20

30 time(min)

FIG. 1. Photoinactivation of dipeptidyl peptidase IV sensitized by methylene blue. Reaction mixtures containing 6.0 X 10-z M methylene blue, 0.05 M potassium phosphate, and 6.7 mg enzyme in a total volume of 9.9 ml at pH 7.8 and 25°C were exposed, at a distance of 40 cm, to the light emitted by a 369W tungsten lamp. These mixtures were stirred in a water bath at 25°C throughout the period of illumination, and aliquots were removed at various intervals for assay of the residual activity.

struction of both tryptophan and histidine residues (Table I). All these amino acids, except alanine and leucine, are quite susceptible to photooxidation. Other amino acid residues, although not listed in the table, were unchanged by photooxidation. The most photosensitive amino acid was histidine, whose number decreased linearly with time from 14.0 to 2.7 residues. Tryptophan residues were not susceptible to photooxidation within 4. min, though 1.4 to 5.5 residues were modified from 14 to 25 min. The number ‘of methionines in the amino acid analysis was reduced, but there was no linear relationship with activity. When the enzyme activity was arranged as a function of the reduced number of tryptophan and histidine residues, a linear relationship would be expected if these residues were involved in the active site. It would seem from these results that DPP IV contains some tryptophan and histidine residues that might be involved in the catalytic activity of DPP IV. Then, modification of histidine residues with DEPC was carried out. The difference spectra showed the appearance of a positive peak at 240 nm due to the formation

ET AL.

of N-carbethoxyhistidine, and no negative peak at 278 nm due to the formation of O-carbethoxytyrosine (18). Absorbance at 240 nm rapidly increased within 7 min, then gradually increased, reaching a plateau after 25 min, with the addition of 1.5 pmol DEPC to 1.13 mg enzyme in 2.5 ml 0.1 M potassium phosphate buffer, pH 6.5. When the enzyme activity was measured as a function of the extent of histidine modification, as calculated from a previously published equation (18), a good linear relationship was found (Fig. 2). In spite of the blockage of four histidine residues, the activity remained at 70% of its original. It would seem from the results that histidine residues must participate in the enzymatic action, but no specifically reactive histidine residue was observed in the active site of DPP IV under these conditions. Modification of DPP IV with NBS at pH 6.4 resulted in a gradual decrease in the absorption spectrum at 280 nm, and also in the enzyme activity, depending on the concentration of NBS. When the percentage residual activity was plotted against the number of tryptophan residues modified, as calculated from the equation according to Spande and Witkop (19) using TABLE AMINO

I

ACID ANALYSIS OF DIPEPTID~L PEPTIDASE BEFORE AND AFTER PHOTOOXIDATION

IV

Residues/mol enzyme monomer (94,000) Amino acid

0 min

4 min

14 min

25 min

Tryptophan Histidine Alanine Cysteine Methionine Leucine Tyrosine Phenylalanine

12.6 14.0 28.6 0.8 9.9 41.2 41.1 23.2

12.6 9.0 29.4 0.8 8.4 41.4 40.2 22.2

11.2 5.9 28.3 0.9 8.5 40.8 38.8 22.4

7.1 2.7 28.4 1.0 8.2 41.0 39.4 23.0

Note Aliquots of photoinactivated and untreated enzymes were subjected to acid hydrolysis and amino acid analysis, as described under Materials and Methods.

CHEMICAL

Blocked

histidyl (no./%,OOO)

MODIFICATION

OF DIPEPTIDYL

residues

FIG. 2. Changes in the enzymatic activity of carbethoxylated dipeptidyl peptidase IV as a function of blocked histidyl residues. The reaction mixture was composed of 1.13 mg enzyme and 1.5 pmol diethylpyrocarbonate in 0.1 M potassium phosphate buffer, pH 6.5. The reaction started just after the addition of DEPC at 25°C. At each time period (after observation of the spectrum), an aliquot (10 ~1) was withdrawn, dialyzed against 0.1 M Tris-HCl buffer, pH 8.0, and then brought to a constant volume of 1.0 ml. The enzyme activity was then measured using 10 or 20 pl of the dialyzed enzyme. The modified histidine residues were calculated using AC, = 3200 cm-’ M-‘.

1.6 as the absorptivity factor, a linear line was obtained for up to 90% inactivation (Fig. 3). When the line was extrapolated to zero activity, 0.8 tryptophan residue was indicated. This result clearly demonstrates that one tryptophan residue is responsible for the activity. This result is further supported by the fluorometric observation of tryptophan modification with NBS and the decrease in activity (Fig. 4). The greater decrease in activity than in fluorescence intensity would indicate that modification was affecting a tryptophan residue in the active site while, with the addition of Ala-Pro-Ala, the substrate of the enzyme, to the reaction mixture of NBS modification, neither fluorescence intensity nor activity were decreased (Fig. 5) with increasing concentration of the substrate. At the maximum concentration of Ala-Pro-Ala (135 PM) used, the activity remained almost 100% while its fluorescence intensity was slightly reduced, suggesting the oxidation of a tryptophan residue at some site other than the active site.

PEPTIDASE

625

IV

Tryptophans oxidized ( no. I subunit )

FIG. 3. Changes in dipeptidyl peptidase IV activity as a function of tryptophan oxidation. Purified enzyme (0.65 mg) was dissolved in 2.5 ml 0.1 M potassium phosphate buffer, pH 6.4. Modification of the tryptophan residues was repeated five times by the successive addition of 100 nmol N-bromosuccinimide. Each time, enzyme activity and spectral changes at 280 nm were determined.

The protective action of amino acids and peptides against the modification of DPP IV with NBS was further studied (Table II). Dipeptidyl Ala-Ala and ProPro exhibited a more effective protective action than the amino acids tested, indicating that these peptides may interact with the substrate binding site. When tryptophan (20 and 40 PM) was modified with NBS (200 PM), neither a protective -1

100

5 Ptl5-I

Moles

NBS

I Moles

Enzyme

(subunit)

4. Modification of tryptophan residues by Nbromosuccinimide. The reaction mixture was composed of 0.23 NM purified enzyme and 1.7-15.3 PM NBS in 0.6 ml 0.1 M potassium phosphate buffer, pH 6.4. Fluorescence intensities (0) at emission of 340 nm and excitation of 295 nm became constant after 2 min by adding NBS to the mixture. Aliquota (10 al) of the reaction mixture were diluted to 200 pl using 0.1 M Tris-HCl buffer, pH 8.0, and 10 to 25 ~1 was used for the measurement of enzyme activity (0). FIG.

626 1

HARADA

ET AL.

7100

shown to be quite photosensitive (20), though complete inactivation was not demonstrated until 10 to 11 residues had been modified. On the other hand, modification of 1 to 2 tryptophan residues result in instant inactivation. If two kinds of amino acid residues were directly responsible for the catalytic activity, the inactivation kinetics should not be linear

Ala-Pro-Ala

FIG. 5. Protective action of Ala-Pro-Ala against modification of dipeptidyl peptidase IV with Nbromosuccinimide. The reaction mixture was composed of 0.23 FM purified enzyme, 83 PM NBS, and different concentrations of Ala-Pro-Ala in 0.6 ml 0.1 Y potassium phosphate buffer, pH 6.4. The NBS was added after a 5-min preincubation at 3’7°C. Fluorescence intensities were observed at an emission of 340 nm and an excitation of 295 nm (0). Aliquots (10 ~1) of the reaction mixture were diluted to 200 pl using 0.1 M Tris-HCl buffer, pH 8.0, and 10 to 20 pl was used for the measurement of enzyme activity (0).

effect of nor activation with Pro-Pro (200 and 400 PM) was observed fluorometrically. DISCUSSION

TABLE

II

PROTECTIVE ACTION OF AMINO ACIDS AND PEPTIDES AGAINST MODIFICATION OF DIPEFTIDYL PEPTIDASE IV WITH N-BROMOEWXINIMIDE

Amino acids and peptides (PM) None Ala 83 167 Ala-Ala a3 167

Remaining activity (W) with N-bromosuccinimide 16.7 CM 56.2

33.0 pY 8.0

69.9 44.4 89.7 70.8

Pro

One of the amino acid residues in the active site of DPP IV was proved to be serine by inhibition studies with diisopropylfluorophosphate (DFP), although Kenny et al. (6) argued that these were different 50% inhibitory concentrations of DFP for Ala-X and Pro-X bond-hydrolyzing activities, with presumably only one active-site serine in each subunit as reported by Barth et al (5). The time course of inactivation of DPP IV with DFP was considerably slower; with 1 h being required for complete inactivation of the enzyme (2 mg/5 ml buffer) with 1.25 rmol DFP (6) and 50% inactivation occurring with 3 PM DFP (5). The present inactivation study employing photosensitization showed that inactivation of DPP IV took place rapidly and closely followed pseudo-first-order kinetics until over 95% of the activity was gone (Fig. 1). The rate constant, k, depends upon the concentration of methylene blue (data not shown). Histidine has been

83 167

69.8 13.7

Pro-Pro 83 167

91.0

GUY 83 167

57.7

Asp 83 167

69.4

62.0

14.8

27.8

I&l 83

167

69.4

40.8

Nota Reaction mixtures contained 0.05 M potassium phosphate, 6.5 pg enayme, and 83 or 167 fl amino acid or peptide in a total volume of 25 ~1 at pH 6.4. After the mixture was preincubated for 5 min at 37”C, 5 ~10.2 rnM N-bromosuccinimide was added to each tube and incubation was continued for 5 min. The residual activity of each assay mixture was determined after it was diluted to 609 ~1 with 0.1 M ‘I’ria-HCl OH 8.0. _.._ ---~ buffer. ~~~~~~~, _~~ ~, in an ice bath.

CHEMICAL

MODIFICATION

OF

on a semilog plot (20). The participation of a histidine residue in the catalytic site cannot be determined until photooxidation studies have been done. DEPC, although not quite specific for histidyl residues, does react more rapidly with them than with other amino acid residues (18). The N-carbethoxylation of histidine residues was monitored by the absorbance at 240 nm using a quite limited concentration of DEPC (1.5 pmol) against 1.13 mg enzyme protein. The inactivation percentage plotted against the number of N-carbethoxyhistidine residues (Fig. 2) gave results quite similar to those obtained from the photooxidation study. DPP IV was sensitive to the ethanol used as the solvent for DEPC, therefore, the complete reactivation with hydroxylamine was not observed. Modification of a tyrosine residue has not been observed from the decreasing absorption spectrum at 278 nm (18), suggesting no involvement of tyrosine residue for activity of DPP IV (Table I). Reaction of the tryptophan nucleus with DEPC in the present experiment was not demonstrated from the relative fluorescence intensity (excitation at 285 nm; emission at 345 nm) of the modified enzyme (22). The insensitivity of the histidine residue with DEPC in the catalytic site of a-chymotrypsin was demonstrated (23), as stated in the present results for DPP IV. A highly specific requirement for histidine residues for DPP IV activity was not demonstrated by two experimental methods. On the other hand, although the tryptophan residue is less sensitive than histidine to photosensitization (20), destruction of 1 to 2 residues per subunit of DPP IV resulted in an almost complete inactivation. This result was further supported by the study involving NBS modification (Fig. 3). The decrease in DPP IV activity was quite linear with respect to the number of tryptophans modified in the range of 0 to 1 tryptophans oxidized. This result implicates the involvement of tryptophan residues in the hydrolysis of the peptide bond between X-Pro and Y. The involvement of a tryptophan residue in the active site of a hydrolytic enzyme

DIPEPTIDYL

PEPTIDASE

IV

627

is not without precedent, as three tryptophan residues at 62, 63, and 108 on the surface of the molecule have been implicated in the binding site of lysozyme, and modification of them with NBS or iodine can lead to loss of activity, as reviewed by Witkop (24). It is therefore suggested that the tryptophan residue of DPP IV which was modified with NBS indicates a similar function of the catalytic action. The role of the tryptophan residue in the active site of DPP IV must be further elucidated. Fluorescence intensity (excitation at 285 nm; emission at 355 nm) of the tryptophan residue susceptible to modification with NBS was recorded. The decrease in DPP IV activity quite paralleled the lowering of fluorescence intensity (Fig. 4), but the rate of decrease in activity was greater than that in fluorescence intensity. It was further demonstrated that the substrate and inhibitor peptides sterically shield the tryptophan from NBS modification. Ala-Pro-Ala, Ala-Ala, and Pro-Pro effectively protected the modification of the enzyme with NBS, because the fluorescence intensity and activity of the enzyme were maintained even after strong reaction with NBS, whereas the fluorescence intensity of tryptophan itself decreased in the presence of substrate and inhibitor peptides. This protective action of enzyme activity implies a structural relationship between the positioning site and the substrate, and it plays an important part in the recognition of the substrate X-Pro configuration at the microvillous membrane (25). The system responsible for the reabsorption of dipeptides in the proximal tubules of the microvillous membrane of the kidney has been identified (26). It is quite interesting to note that Gly-Pro and Ala-Ala are reabsorbed through the same channel in rat kidney (27); therefore, DPP IV might constitute the channel and play an important part in the absorption of the dipeptides. The catalytic action of this enzyme on X-Pro (or Ala)-Ypeptides might be explained by an acylation-deacylation mechanism, because serine occurs in the active site of the enzyme (28). It remains

628

HARADA

to be proven that a special histidine residue participates in the mechanism. The present results have not proven the presence of a histidine residue by two methods of modification, since the histidine residue in the catalytic site of DPP IV should be involved in hydrogen bonding in a socalled “catalytic triad” with an aspartate and serine. ACKNOWLEDGMENTS We thank Ajinomoto Company, Inc. for their generous gift of the substrate. We are also grateful to Miss. S. Fukushima for typing the manuscript and drawing the figures.

REFERENCES 1. HOPSU-HAVE, V. K., AND GLENNER, G. G. (1966) Hi&o&em+ 7, 197-201. 2. HOPSU-HAVE, V. K., AND SARIMO, S. R. (1967) Hoppe-seyler’s 2. Physiol Chem 348, 15401550. 3. HOPSU-HAVE, V. K., RINTOLA, P., AND GLENNER, G. G. (1968) Acta Chem .%a& 22, 2S9-30Kr 4. OYA, H., NAGATSU, I., AND NAGATSU, T. (1972) Biochim. Biophys. Actu 253,591-599. 5. BARTH, A., SCHULZ, H., AND NEUBERT, K. (1974) Acta Bid Me& Germ 32,157-174. 6. KENNY, A. J., BOOTH, A. G., GEORGE, S. G., INGRAM, J., KERSHAW, D., WOOD, E. J., AND YOUNG, A. R. (1976) Biochem. J. 157, 169-182. 7. YOSHIMO~, T., AND WALTER, R. (1977) Biochim Biqphys Ada 485, 391-401. 8. FUKASAWA, K. M., FUKASAWA, K., AND HARADA, M. (1978) Biochim Biuphys. Acta 535, 161166. 9. SVENSON, B., DANIELSEN, M., STAUN, M., JEPPESEN, L., NOR%N, O., AND SJ~STR~M, H. (1978) Eur. J. Biochem 90,489-498. 10. MACNAIR, R. D. C., AND KENNY, A. J. (1979) Biochem J 179,379-395.

ET

AL.

11. PUSCHEL, G., MENTLEIN, R., AND HEYMANN, E. (1982) Eur. J. Biochem. 126, 359-365. 12. HARADA, M., FIJKASAWA, K. M., FUKASAWA, K., AND NAGATSU, T. (1982) B&him Biophys. Acta 705, 288-290. 13. NAGATSU, T., HINO, M., FUYAMADA, H., HAYAKAWA, T., SAKAKIBARA, S., NAKAGAWA, Y., AND TAKEMOTO, T. (1976) Ad Biochem 74, 4% 476. 14. HARTREE, E. F. (1972) And Biochem 48, 42% 427. 15. FORMAN, H. J., EVANS, H. J., HILL, R. L., AND FRIDOVICH, I. (1973) Biochemistry 12.823-827. 16. PENKE, R., FERENCZI, R., AND KOVACS, K. (1974) Anal Biochem 60,45-54X 17. ELLMAN, G. L. (1959) Arch, B&hem Biophys. 82.70-77. 18. MILES, E. W. (1977) in Methods in Enzymology (Hirs, C. H. W., and Timasheff, S. N., eds.), Vol. 47, pp. 431-442, Academic Press, New York. 19. SPANDE, T. F., AND WITKOP, B. (1967) in Methods in Enzymology (Hirs, C. H. W., ed.), Vol. 11, pp. 498-506, Academic Press, New York. 20. WEIL, L. (1965) Arch Biochem Bhiophys. 110, 57-68. 21. RAY, W. J., JR., AND KOSHLAND, D. E., JR. (1961) J. BioL Chem 236, 1973-1979. 22. ROSEN, C. G., GEJVALL, T., AND ANDERSSON, L. 0. (1970) Biochim Biophy& Actu 221, 207213. 23. MELCHIOR, W. B., JR., AND FAHRNEY, D. (1970) Biochemistry 9,251-258. 24. WITKOP, B. (1968) Science (Washington, D. C.) 162,318-326. 25. FUKASAWA, K. M., FUKASAWA, K., SAHARA, N., HARADA, M., KONDO, Y., AND NAGATSU, I. (1981) J. Histochmn Cytochem 29, X37-343. 26. FONTELES, M. C., GANAPATHY, V., PASHLEY, D. H., AND LEIBACH, F. H. (1983) Lzfe Sti 33, 431-436. 27. WISEMAN, G. (1983) J. Physiol 342.421-435. 28. BARTH, A., MAGER, H., FISCHER, G., NEUBERT, K., AND SCHWARZ, G. (1980) Acta Bid Mtd Germ 39, 1129-1142.