Vitamin K-dependent carboxylase: Influence of the “propeptide” region on enzyme activity

ARCHIVESOFBIOCHEMISTRYAND BIOPHYSICS Vol. 274, No. 2, November 1, pp. 574-581,1989

Vitamin K-Dependent Carboxylase: influence of the “Propeptide” Region on Enzyme Activity’ ALEX CHEUNG,

JEAN A. ENGELKE,

Department

CYNTHIA

SANDERS,

AND J. W. SUTTIE2

of Biochemistry, College of Agricultural and Lzfe Sciences, University of Wi.wonsin, Ma.dism, Wisumsin 55706

Received April 17,1989, and in revised form July lo,1989

The liver microsomal vitamin K-dependent carboxylase catalyzes the post-translational conversion of specific glutamyl to y-carboxyglutamyl (Gla) residues in precursor forms of a limited number of proteins. These proteins contain an amino-terminal extension (propeptide) that is presumed to serve as an enzyme recognition site to assure their normal processing. The free, noncovalently bound propeptide has also been shown to stimulate the in vitro activity of this enzyme. This peptide has now been shown to lower the app K, of a low-molecular-weight Glu site substrate while having no influence on the app K,,, of the other substrates, vitamin KH2, 02, and CO,/HCO;. Propeptide addition was shown to have no influence on the ratio of the two products of the enzyme, Gla and vitamin K-2,3-epoxide. Stimulation of carboxylase activity by the propeptide from human factor X was observed in a number of rat tissues and in the liver of a number of different species. Stability of the enzyme in crude microsomal preparations was greatly enhanced by the presence of propeptide. These observations are consistent with the hypothesis that this region of the protein substrates for the carboxylase not only serves an enzyme recognition or docking function but also modulates the activity of the enzyme by altering the affinity for one of its substrates. 0 1989 AcademicPress,Inc.

The liver microsomal vitamin K-dependent carboxylase catalyzes the post-translational conversion of glutamyl to y-carboxyglutamyl (Gla)3 residues in intracellular precursors of a limited number of proteins (1). These proteins include the classical vitamin K-dependent plasma clotting factors (II, VII, IX, and X), plasma 1Research was supported by the College of Agricultural and Life Sciences of the University of Wisconsin-Madison, and in part by Grants DK-14881, HL29586, and GM-07215 from the National Institutes of Health, Bethesda, MD. ’ To whom correspondence should be addressed. a Abbreviations used: Gla, y-carboxyglutamic acid; vitamin KO, vitamin K-2,&epoxide; vitamin KHz, the reduced form of vitamin K quinone; Boc, t-butyloxycarbonyl; OME, methoxy; Bz, benzoyl; OEt, ethoxy; propeptide, KSLFIRREQANNILARVTRA. 0003-9861/89 $3.00 Copyright All rights

Q 1989 by Academic Press, Inc. of reproduction in any form reserved.

proteins C and S, and proteins of undetermined function in other tissues. The intracellular precursor form of rat prothrombin (factor II) has been shown to be larger and more basic than the secreted plasma form (2) and knowledge of cDNA sequences has revealed that the primary gene product of all vitamin K-dependent plasma proteins contains a basic amino acid rich “propeptide” between the hydrophobic signal peptide region and the amino-terminal end of the mature plasma form of the protein (3, 4). This is an extremely homologous region of these proteins, and a similar amino-terminal extension or internal homologous region has been observed (5,6) in the precursor form of two vitamin K-dependent bone proteins which shows no other sequence homology to the vitamin K-dependent plasma proteins. These observations 574

VITAMIN

K-DEPENDENT

have suggested (2,5, ‘7) that this region of these proteins might be an important recognition site for the enzyme catalyzing the post-translational y-glutamyl carboxylation event. The recognition or “docking” function of this region has now been established by a number of in vitro and cellular studies. The isolated cellular precursor of prothrombin which contains the propeptide region has been shown (8-11) to be a better substrate for the enzyme than the des-y-carboxy forms of plasma prothrombin. Des-y-carboxy recombinant Factor IX containing a propeptide extension (12), recombinant protein C containing a partial propeptide sequence (ll), or a low-molecular-weight peptide substrates with an amino-terminal propeptide extension (13) have all been shown to be good substrates for the enzyme. It has also been shown that effective carboxylation of recombinant constructs of factor IX (14, 15) or protein C (16) expressed in mammalian cell lines is dependent on the presence of a native propeptide region. These observations have been interpreted as evidence for a role of the propeptide region in recognizing the carboxylase during intracellular processing. More recently (17) we have demonstrated that a peptide containing the propeptide sequence of human factor X effectively stimulates the carboxylation of a low-molecular-weight peptide substrate of the carboxylase even if the two peptides are not eovalently attached. A number of aspects of the interaction of this regulatory peptide with the enzyme have now been studied in more detail. MATERIALS

AND

METHODS

Animals. All rats were male, 250-300 g (SpragueDawley, Madison, WI), fasted overnight for 18 h before decapitation. Normal rats were maintained on commercial rat chow. Warfarin-treated rats received commercial chow and an ip injection of warfarin (5 mg/kg) 18 h prior to sacrifice. Vitamin K-deficient rats were maintained in coprophagy-preventing cages (18) and fed a vitamin K-deficient diet (19) for 9 days prior to sacrifice. Some rats of all dietary groups were given 1 mg phylloquinone (aquamephyton, Merck Sharpe & Dohme, Westpoint, PA) ic 15 min

CARBOXYLASE

575

prior to sacrifice. In experiments designed to determine propeptide response in various species, the carboxylase activity of liver microsomes from rats were compared to those obtained from 130-145 g male hamsters (Sprague-Dawley), 400-500 g male guinea pigs (Sprague-Dawley), 30-40 g male mice (U.W. Department of Biochemistry, Madison, WI). 600-800 g male Leghorn chicks (U.W. Poultry Science Department), a mature cow, and a mature female pig (U.W. Animal Sciences Dept., Madison, WI). These animals were fed a commercial ration suitable for each species and were not fasted prior to sacrifice. Microscrm,al preparation. Excised livers were placed in cold (5°C) sucrose (0.25 M) imidazole (0.025 M) buffer, pH 7.3 (SI buffer), and homogenized in SI buffer (2 ml vol/g liver) using a Potter-Elvehjem homogenizer. The homogenate was centrifuged at 10,OOOg(11,500 rpm) for 15 min at 5°C in a Beckman 52-21 centrifuge with a JA-20 rotor and 8-ml aliquots of this supernatant centrifuged at 105,000~ (40,000 rpm) for 60 min at 5°C in a Beckman Model L-2 using a TI 50 rotor. The pellets were surface washed using SI buffer containing 0.5 M KC1 (SIK buffer). Pellets were frozen in liquid nitrogen and stored at -70°C until assayed. Vitamin K-dependent car5oxylase assays. Microsomal pellets were suspended in SIK buffer containing 1.5% Triton X-100 so that 1 ml of suspension was equal to 0.5 g of liver. Incubation mixtures contained 200 ~1 of microsomal suspension, 100 ~1 of propeptide in SIK buffer, 50 ~1 Boc-Glu-Glu-Leu-OMe in SIK buffer and 10 ~1 of NaH’%Oa (1.0 mCi/ml, 55 mCi/ mmol). After incubation for 2 min at 17°C reactions were initiated by the addition of 5 pl(40 pg) of ehemieally reduced (20) vitamin K and quenched at 30 min by the addition of 0.2 ml of incubation mixture to 1.0 ml of cold 10% trichloroacetic acid. After centrifugation, the supernatant was removed and free ‘%O, was removed by bubbling for 5 min with CO,. A 0.2-ml aliquot of the supernatant was mixed with 3.8 ml of BioSafe II (Research Products International, Elk Grove Village, IL), and radioactivity determined in a liquid scintillation spectrometer. Precursor content of microsomes was assayed as carboxylation of endogenous microsomal protein as previously described (21). Kinetic and stoichiometric measurements. Apparent K,,, values for the four earboxylase substrates were obtained from initial rate measurements of enzyme activity at 17°C. Values for the Glu site substrate Boc-Glu-Glu-Leu-OMe and vitamin KH2 were determined as previously described (17, 22). Concentrations of 0, were varied by flushing sealed incubation tubes with Na and adding a known amount of OZ. Variations in CO$HCO, concentrations were achieved by lyophilizing the microsomal suspension followed by the addition of increasing amounts of

576

CHEUNG ET AL.

oc 000

a25

Propeptide

0.75

Concentration Q.&l)

FIG. 1. Effect of propeptide concentration on carboxylase activity. Enzyme activity was assayed with 0.5 m&i Boc-Glu-Glu-Leu-OMe as a substrate and varying concentrations of the peptide KSLFIRREQANNILARVTRA.

HCO, at a constant specific radioactivity. While concentrations of one substrate were varied, the others were held constant at 0.5 mM Boc-Glu-Glu-Leu-OMe, 220 pM vitamin KHz, ambient air (20% O,), and endogenous C02/HCOs (-2.0 mM). Initial rates were obtained for at least five different substrate concentrations, and app Km values were calculated by the nonlinear least-squares technique described by Cleland (23). Stoichiometry of vitamin K epoxide and Gla formation at different propeptide concentrations was determined essentially as described by Wood and Suttie (20). Substrates and propeptides. The carboxylase substrates Boc-Glu-Glu-Leu-OMe and Phe-Leu-GluGlu-Leu were obtained from Bachem (Torrance, CA). Boc-Leu-Glu-Glu-Leu and Ala-Leu-Glu-Glu-ProAla were synthesized as previously described (24). Boc-Glu-OBz was obtained from Sigma (St. Louis, MO) and Bz-Glu-OEt and Bz-Glu-OMe were a gift from Dr. James Slama of the University of Texas, San Antonio. The 20-residue peptide containing the propeptide sequences of human factor X with an additional lysine at the amino terminus and an alanine at the carboxy terminus was synthesized and purified as previously described (17) and is referred to in the text as propeptide. RESULTS

The initial report (17) of the ability of the free propeptide to stimulate the activity of the vitamin K-dependent carboxylase indicated that maximal activation was achieved at a propeptide concentration near 1 PM. The data in Fig. 1 indicate that

half-maximal stimulation of the enzyme by the propeptide utilized in the studies reported here was approximately 0.1 PM. The vitamin K-dependent carboxylase is normally assayed with short peptides containing Glu-Glu sequences as Glu site substrates (1). The propeptide has previously been shown to stimulate the activity of the enzyme toward the substrate, Boc-GluGlu-Leu-OMe, and this observation is extended in Table I. Addition of propeptide caused a greater than lo-fold increase in the rate of carboxylation of four short peptides with Glu-Glu sequences, but had little influence on the rate of carboxylation of three substrates that were simple derivatized Glu residues. Microsomal preparations from vitamin K-deficient rats contain a pool of bound protein precursors which are substrates for the carboxylase (21), and the carboxylation of these proteins was not substantially influenced by the addition of the propeptide. The data in Table II confirm previous observations that the presence of the propeptide has a profound effect on the Km of the Glu site substrate, lowering it nearly lofold. The data also indicate that propeptide addition has no influence on the Km of the other three substrates, 02, vitamin KHz, and C02/HCOa. The products of the carboxylase reaction are Gla and the 2,3-epox-

VITAMIN TABLE

K-DEPENDENT

I

TABLE

EFFECTOFPROPEPTIDEADDITION ONVARIOUS Glu SITE SUBSTRATES Carboxylase activity (dpm) Substrate Boc-Glu-Glu-Leu-OMe Phe-Leu-Glu-Glu-Leu Boc-Leu-Glu-Glu-Leu Ala-Leu-Glu-Glu-Pro-Ala Bz-Glu-OEt Bz-Glu-OME Boc-Glu-OBz Endogenous precursor

577

CARBOXYLASE

-Propeptide 3865 3279 770 364 250 462 1283 1360

+Propeptide 41,501 41,957 14,103 10,590 707 1,225 2,385 1,810

Note. Microsomes were obtained from vitamin Kdeficient rats and prepared as described under Materials and Methods. Assays were conducted for 30 min at 1’7°C in the presence of 0.5 mM peptide substrates and -Cl0 FM propeptide. Values are means of duplicate assays.

III

EFFECT OF PROPEPTIDE CONCENTRATION ON EPOXIDE TO Gla RATIOS

Propeptide concentration (M) 0 1o-8 10 -7

10 +’ 1om5

Vitamin K epoxide ‘*CO2 fixed (nmol/ml) (dpm/ml X 10m4) KO/Gla 0.28 0.56 1.23

3.31 4.96 10.75

1.28

16.41

1.32

13.63

2.5 3.5 3.2 2.4 3.0

Note. Values are means of duplicate incubations at each level of propeptide with a substrate, Boc-GluGlu-Leu-OMe, concentration of 5 mM. Epoxide/Gla ratios (KO/Gla) are molar ratios calculated from the specific activity of the H’%O: used and an endogenous CO,/HCO; concentration of 2.0 mM.

min K-deficient or warfarin-treated rats contain higher vitamin K-dependent caride of vitamin K (KO). The ratio of KO to boxylase activity, and this response has Gla is a function of the COJHCO; concen- been postulated to be due to stabilization tration of the incubation and approaches of the enzyme by bound precursor proteins unity at saturating concentrations of COZ which accumulate under these conditions. (20,25). The data in Table III indicate that These preparations would also contain the stoichiometry of the reaction is not higher concentrations of bound endogesubstantially influenced by propeptide nous propeptide, and it is possible that the concentration. A similar lack of effect of enhanced activity results from its prespropeptide concentration on stoichiometry ence. The data in Table IV show that under was seen at a Glu site substrate concentra- normal assay conditions, carboxylase action of 0.5 mM (data not shown). tivity is higher when rats are vitamin K It has previously been demonstrated (26) deficient or have been given the vitamin K that microsomal preparations from vita- antagonist warfarin, and that increased activity is related to the level of bound precursors. The ability to stimulate microTABLE II somal carboxylase activity by propeptide EFFECTOFPROPEPTIDEADDITION ONKm addition was decreased when there were OF CARBOXYLASE SUBSTRATES higher levels of bound precursor. When bound precursor concentrations were deAPP i-k creased by administration of vitamin K just prior to sacrifice of the animals, stimSubstrate -Propeptide +Propeptide ulation of enzyme activity by added proBoc-Glu-Glu-Leu-OMe 8.8+ 0.7 mM 1.0-c 0.1 rnM peptide was greatly increased. Although '79 +23p~ 102 fl'l@M Vitamin KH, the vitamin K-dependent carboxylase ac1.4 * 0.3% 0, 1.2 r!z 0.3% 02 02 tivity appears highest in liver (27), other 1.9-+ 0.9rnM 2.0 k 0.9rnM COJHCO, tissues do have activity. The data in Table Note. Values were obtained as described under Ma- V demonstrate that the propeptide of human factor X is capable of stimulating low terials and Methods and were obtained k10 pM prolevels of carboxylase activity present in peptide.

578

CHEUNG ET AL. TABLE IV EFFECT OF VITAMIN K STATUS ON PROPEPTIDE STIMULATION OF CARBOX~LASE

Carboxylase activity (dpm) Precursor content

Animal treatment Normal Normal + K K Deficient K Deficient + K

22Ok 93 33+ 8 1231 f 306 50 + 23

Warfarin treated Warfarin treated + K

1064 + 324 299 f 193

-Propeptide

+Propeptide

Fold increase

1919f 925 993k 58 3670 k 1007 3014 k 875 2979 + 802 2352+ 711

16,343f 6538 12,837f 2041 18,952If: 2682 27,836 f 6401

8.5 12.9 5.2 9.2

10,904+ 3691 14,192+ 1508

3.7 6.0

Note. Rats were fed a laboratory chow diet (normal) or a vitamin K-deficient diet (K deficient) or were injected with 5 mg/kg warfarin ip 18 h before they were killed. The +K rats were given 1 mg of phylloquinone ic 15 min before they were killed. “Precursor content” was measured as vitamin K-dependent incorporation of “COz into endogenous microsomal proteins (protein carboxylase activity, dpm/50 ~1). Carboxylase assays were conducted for 30 min at 17°C in the presence of 0.5 mM Boc-Glu-Glu-Leu-OMe as a substrate and &lo fiM propeptide. Values are mean + SE for three animals per group.

lung and kidney microsomes. The ability to stimulate hepatic carboxylase activity was demonstrated in a number of species (Table VI) with the most pronounced response observed in the guinea pig, a species which had very low activity prior to stimulation. The vitamin K-dependent carboxylase activity is relatively unstable in the crude microsomal preparation usually used for studying the activity. The data in Fig. 2 demonstrate that the activity was greatly

TABLE V EFFECT OF TISSUE ON PROPEPTIDE STIMULATION OF CARBOXYLASE

Carhoxylase activity (dpm) Tissue

-Propeptide

+Propeptide

Liver Kidney Lung

7769 f 999 70f 24 lo+ 4.7

45,279+ 3004 459 + 169 199k 76

Note. Tissues were obtained from vitamin K-deficient rats and microsomes prepared as described under Materials and Methods. Assays were conducted for 30 min at 17°C in the presence of 0.5 mM Boc-GluGlu-Leu-OMe as a substrate and k10 PM propeptide. Values are mean f SE for four animals per group.

stabilized by the presence of increasing concentrations of the propeptide. Similar increases in stability were seen at 1’7°C and at room temperature. DISCUSSION

Evidence from a number of different approaches has now clearly demonstrated that the primary gene product of vitamin K-dependent proteins contains a “propeptide” region between the signal peptide and the amino terminus of the mature protein that is involved in the interaction of these proteins with the enzyme that carries out a specific y-glutamyl carboxylation. The available data support the hypothesis that this region serves both a docking function to orient the enzyme and its substrate and also a regulatory role to modulate the activity of the enzyme. The results reported here have attempted to more fully describe the latter function and interpret our previous understanding of the properties of the enzyme in light of new observations. The stimulation of carboxylase activity in the presence of propeptide noncovalently bound to the carboxylase substrate is the result of a 50-100% increase in the

VITAMIN

K-DEPENDENT

enced by the propeptide have high (-10 mM) Km values and the influence of the propeptide on the natural, presumably tight binding, protein substrates is not yet known. The presence of a propeptide covalently attached to a low-molecular-weight peptide substrate has, however, been shown to lower the Km of this substrate to the micromolar range (13). It is of interest that the carboxylation of simple derivatized Glu residues, which have even higher Km values than peptides with Glu-Glu sequences, are not substantially influenced by the presence of the propeptide. It has previously been shown (28) that the enzyme will carboxylate Asp to /3carboxy Asp residues only if they are present as a simple derivatized Asp residue but not when included in a tripeptide substrate. These data suggested that a derivatized Asp or Glu residue had a very weak interaction with the enzyme which would be consistent with the failure of the propeptide to modify this interaction. The nature of the propeptide activation of the enzyme has been clarified by the observation that the Km of the other three

TABLE VI EFFECTOFSPECIESONPROPEPTIDESTIMULATION OF~ARBOXYLATION Carboxylase activity (dpm) Species Rat Mouse Hamster Guinea pig cow

Pig

-Propeptide

+Propeptide

Fold increase

817 f 71

10,367+ 898

12.7

726f26 636+67

10,210+226 3,934 + 289 4,496+288 833 6,651

14.1 6.2 53.5 26.0 12.9

34+19 32 512

Note. All laboratory species were young adult males fed laboratory chow 1 week before liver microsomes were prepared as described for rat microsomes under Materials and Methods. Assays were conducted for 30 min at 17°C in the presence of 0.5 mM Boc-GhGlu-Leu-OMe as a substrate and ?lO pM propeptide. Values for laboratory animals are means + SE for four animals per group and those for cow and pig livers are means of duplicate assays of a single liver.

observed V,,, of the reaction and a nearly lo-fold decrease in the app Km for Glu site substrate. The Glu site substrates influ-

I

I 1

2

579

CARBOXYLASE

4 3

4

1 5

I 6

7

Days

FIG. 2. Effect of propeptide concentration on carboxylase stability. Microsomes in SIK buffer and Triton X-100 as described under Materials and Methods were held at 4°C in the presence of varying propeptide concentrations and aliquots were taken for assay at 17°C at the times indicated. Data are expressed as the percentage of carboxylase activity remaining after storage in the presence of 0 PM (Cl), 1 PM (m), 10 pM (0), and 100 pM (0) of propeptide.

580

CHEUNG

ET AL.

substrates, OZ, vitamin KHz, and COz, is lize this activity should greatly aid these not altered by its presence. Neither does efforts. the presence of propeptide alter the efficiency of coupling of vitamin K epoxidaREFERENCES tion to Gla formation. This is consistent with previous studies (20,25) demonstrat1. SUSIE, J. W. (1985) Annu Rev. Biochem 54,459ing that these functions are closely linked 477. and independent of the rate of the reaction. These observations are consistent with a 2. SWANSON, J. C., AND SUTXE, J. W. (1985) Bie chemistry 24.3890-3897. mechanism whereby interaction of the en3. DAVIE, E. W. (1987) in Hemostasis and Thrombozyme with the propeptide region of its subsis: Basic Principles and Clinical Practice (Colstrate serves not only to restrict the possiman, R. W., Hirsh, J., Mardar, V. J., and Salzble sites of interaction of the enzyme with man, E. W., Eds.), pp. 242-267, Lippincott, the substrate Glu residues but also to inPhiladelphia, PA. crease the affinity of the catalytic site for 4. FURIE, B., AND FURIE, B. C. (1988) Cell 53,505-518. its substrate. The compatability of a dock5. PAN, L. C., AND PRICE, P. A. (1985) Proc. Nat1 Acad Sci. USA 82,6109-6113. ing site for an enzyme that must carboxyl6. PRICE, P. A., AND WILLIAMSON, M. K. (1985) J. ate Glu residues as near as 6 residues from Biol Chem. 260,14,9’71-14,975. the terminus of the propeptide region or as 7. DIUGUID, D. L., RABIET, M. J., FURIE, B. C., LIEBfar as 40 is not clarified by these observaMAN, H. A., AND FURIE, B. (1986) Proc. Nat1 tions. Neither is the relationship between Aced Sci USA 83,5803-5807. the propeptide region and a postulated in8. SOUTE, B. A. M., VERMEER, C., DE METZ, M., tra Gla region consensus sequence (29) for HEMKER, H. C., AND LIJNEN, H. R. (1981) Bie carboxylase substrates clarified by the chim Bicphys. Ada 676.101-107. data presented here. 9. SHAH, D. V., SWANSON, J. C., AND SUSIE, J. W. The recognition of the enzyme by the (1983) Arch B&hem. Biophys. 222.216-221. propeptide appears to be universal. The en- 10. EVANS, M. R., SUNG, M. W. P., AND ESNOUF, M. P. (1984) Biochem. Sot. Trans. 12,1051-1052. zyme was activated by propeptide addition 11. SUTTIE, J. W., HOSKINS, J. A., ENGELKE, J., HOPFin all species studied, and the activation GARTNER, A., EHRLICH, H., BANG, N. U., BELAwas apparent in both nonhepatic and heGAJE, R. M., SCHONER, B., AND LONG, G. L. patic tissues. The data in Table IV have (1987) Proc. Natl. Acad. Sci USA 84.634-637. clarified the basis for previous observa- 12. VERMEER, C., AND SOUTE, B. A. M. (1988) in Curtions (26) relative to the influence of vitarent Advances in Vitamin K Research (Suttie, min K status on carboxylase activity. J. W., Ed.), pp. 25-39, Elsevier Science, New Those preparations which have a high York. level of endogenous protein precursors, 13. ULRICH, M. M. W., FURIE, B., JACOBS, M. R., VERMEER, C., AND FURIE, B. C. (1988) J. BioL Chem and therefore a high concentration of co263,9697-9702. valently attached natural propeptide, have a higher activity and are less capable of 14. JORGENSEN, M. J., CANTOR, A. B., FURIE, B. C., BROWN, C. L., SHOEMAKER, C. B., AND FURIE, stimulation by added propeptide. Previous B. (1987) Cell 48,185-191. observations that vitamin K deficiency in15. RABIET, M.-J., JORGENSEN, M. J., FURIE, B., AND creased carboxylase activity are, thereFURIE, B. C. (1987) J. Biol Chem 262,14,895fore, probably explained by an increase in 14,898. the amount of endogenous propeptide. The 16. FOSTER, D. C., RUDINSKI, M. S., SCHACH, B. G., presence of propeptide also has a substanBERKNER, K. L., KUMAR, A. A., HAGEN, F. S., tial influence on the stability of the enSPRECHER, C., INSLEY, M., AND DAVIE, E. W. zyme, and it is likely that this is another (1987) Biochemistry 26.7003-7011. reason for the higher activity in vitamin 17. KNOBLOCH, J. E., AND SUTTIE, J. W. (1987) J. Bid K-deficient rats. Efforts to purify the carChem. 262,15,334-15,337. boxylase have been of only limited success, 18. METTA, V. C., NASH, L., AND JOHNSON, B. C. (1961) and instability of the activity has been a J. Nutr. 74.473-476. significant problem. The observation that 19. MAMEESH, M. S., AND JOHNSON, B. C. (1959) Proc the free propeptide will effectively stabiSC. Exp. Biol. Med 101,467-468.

VITAMIN

K-DEPENDENT

20. WOOD, G. M., AND SUITIE, J. W. (1988) J. Biol.

Chem. 263,3234-3239. 21. SHAH, D. V., AND SUTTIE, J. W. (1983) Proc. Sot. Exp. Biol. Med 173,148-152. 22. CHEUNG, A., AND SUTTIE, J. W. (1988) BioFactors 1,61-65. 23. CLELAND, W. W. (1979) in Methods in Enzymology (Purich, D. L., Ed.), Vol. 63, pp. 103-138, Academic Press, San Diego. 24. RICH, D. H., LEHRMAN, S. R., KAWAI, M., GOODMAN, H. L., AND SUITE, J. W. (1981) J hfed Chem 24,706-711.

CARBOXYLASE

581

25. LARSON, A. E., FRIEDMAN, P. A., AND SUTTIE, J. W. (1981) J. Bid Ch.em. 256, 11,03211,035. 26. SHAH, D. V., AND SUTTIE, J. W. (1978) Arch. Bic&em. Biophys. 191,571-577. 27. VERMEER, C. (1986) Haewwstasti 16,239-245. 28. MCTIGUE, J. J., AND SUTTIE, J. W. (1983) J. Biol. Chem. 258,12,129-12,131. 29. PRICE, P. A., FRASER, J. D., AND METZ-VIRCA,

G. (1987) Proc. Nat1 Acad Sci USA 84, 78567860.