THE JOURNAL OF Bmmorca~ CHE~STRY Vol. 241, No. 5, Issue of March 10, 1966

Printed in U.S.A.

Purification and Properties of D-Se&e Dehydrase

from Escherichia coZi*

(Received for publication, August 6, 1965)

DANIEL DUPOURQUE, W. AUSTIN NEWTON, AND ESMOND E. SNELL

From the Department of Biochemistry, University of California, Berkeley, California 9@l?O

SUMMARY

A procedure is described for obtaining crystalline o-serine dehydrase from a mutant of Escherichia coli which produces this enzyme constitutively. The enzyme appears pure by ultracentrifugal and electrophoretic criteria. In sucrose gradients, serine dehydrase sediments to the same position as horseradish peroxidase, indicating a molecular weight of about 40,000. This corresponds closely to its combining weight of 42,800 per pyridoxal phosphate. Its absorption maximum at 410 rnp due to bound pyridoxal phosphate is not materially changed by interaction with substrate or by variation in pH between 6.0 and 9.0. n-Serine and D- threonine are the only substrates found; 0-methylserine is a potent competitive inhibitor. In the presence of suitable protective agents, the enzyme can be resolved completely by dialysis against D- or L-cysteine, and reactivated by addi- tion of pyridoxal phosphate. Tris buffer inactivates the enzyme; this inactivation is prevented by the presence of sufficient K+ or NHa, and less effectively by Na+ or pyri- doxal phosphate.

The enzymatic, nonoxidative deamination of serine to pyru- vate and ammonia has been known since 1938 (1). Many differ- ent enzymes carry out this reaction with L-serine. In addition to the relatively specific n-serine dehydrases of Neurospora crassa (2)) Escherichia coli (3, 4), Clostridium acidi-urici (5), and sheep liver (6), this reaction is also catalyzed by apparently pure or highly purified preparations of tryptophanase (7, 8), the B protein of tryptophan synthetase (9), cystathionine synthetase (lo), and threonine dehydrase (11). Most of these enzymes have been shown to require pyridoxal phosphate as a coenzyme.

Cell-free pyridoxal phosphate enzymes that carry out the corresponding deamination of n-serine to pyruvate and ammonia but do not attack n-se&e were first obtained fr0m.E. coli by Metzler and Snell (4) and from N. crassa by Yanofsky (12). We report here the preparation of crystalline n-serine dehydrase (n-serine hydrolyase (deaminating), EC 4.2.1.14) from E. coli, and some of its properties. Labow and Robinson (13) report similar findings in a parallel article.

* Supported in part by Grants A-1448 and E-1575 from the Na- tional Institutes of Health, United States Public Health Service.

MATERIALS AND METHODS

Assay of D-Serine Dehydrase-Unless specified otherwise, assays were performed in 12-ml conical centrifuge tubes which contained the enzyme preparation, 10 c(g of pyridoxal-P, and 100 pmoles of potassium phosphate, pH 7.8, in a volume of 0.8 ml. After 5 min at 37” in a rotary shaker, the reaction was started by adding 0.2 ml of solution (warmed to 37”) which contained 10 mg of n-serine at pH 7.8. Shaking was continued for 10 min; then the reaction was stopped by adding 0.5 ml of 20% trichloroacetic acid. After centrifugation, pyruvate was determined by a slight modification of the procedure of Friede- mann and Haugen (14). A l.O-ml portion was transferred to 2 ml of absolute ethanol, and 1.0 ml of a 0.1 y0 solution of 2,4- dinitrophenylhydrazine in 2 N HCl was added. After 5 min, the color was developed by addition of 5 ml of 2.5 N NaOH. The absorbance at 520 rnp was determined, and the amount of pyru- vate formed was calculated by reference to a standard curve prepared with crystalline sodium pyruvate. One unit of ac- tivity is that amount of enzyme catalyzing the formation of 1.0 pmole of pyruvate in 1.0 min under these conditions; the specific activity represents the number of units per mg of protein.

Other Methods-Protein was determined by the method of Lowry et al. (15) with serum albumin as standard. Pyridoxal-P was determined by the phenylhydrazine procedure of Wada and Snell (16). Sucrose gradient centrifugation and the determina- tion of alcohol dehydrogenase were carried out as described by Martin and Ames (17).

RESULTS

Source of Akrine Dehydrase

The n-serine dehydrase of wild-type E. coli is an inducible enzyme (4, 18). The enzyme content of strains B and W of E. coli grown under conditions of maximum induction, however, proved to be only about one-third that of two mutants of E. coli W which were originally described by McFall (19), and which are constitutive for this enzyme (Table I). Strain Cg was selected for purification of the enzyme.

PuriJication of o-Serine Dehydrase

Mutant strain Cg of E. coli W was grown from 10% inoculum in carboys containing 18 liters of minimal medium (20) supple- mented with 0.16 $ZO glycerol and 1 .O To of acid-hydrolyzed casein. After incubation for 12 to 18 hours at 37” with vigorous aeration, cells were harvested in the Sharples centrifuge. The purifica-

1233

by guest on April 23, 2020

http://ww

w.jbc.org/

Dow

nloaded from

1234 o-Serine Dehydrase Vol. 241, No. 5

TABLE I

Activities of wild-type and mutant strains of E. coli as source of o-serine dehydrase

I Specific activity of cell-free extracts

Additions to growth medium* Wild-type strains Constitutive mutantst of

E. coli W 3828

E. coli B E. coli W 3828 CP C6

None. 0 0 1.4 1.3 n-Serine 0.50 0.40 1.5 1.7

* o-Serine (2 mg per ml) was added to the growth medium where indicated. The growth conditions are those described in the text except that acid-hydrolyzed casein was omitted from the growth medium.

i We are indebted to Dr. Elizabeth McFall for these constitu- tive mutants.

TABLE II

PuriJication of D-serine dehydrase

Purification step*

1. Crude extract., 2. Protamine filtrate.. 3. Treatment at 55”. 4. Hydroxylapatite 1. 5. DEAE-cellulose.. 6. Hydroxylapatite 2.

- Volume

ml

800 1,100

300

3

-

Protein

%

56,000 15,500

5,400 280

60 15

Specific activity

0.85 3.0 7.0

90 200 300

-.

- * The numerals correspond to the description in the text.

Yield

Cl& 96 68 45 21

8

A

.6 .a 1.6

B

Protein, curve 1 Enzyme, curve 2

a; & ZE

‘32 VI

4p r1

5

22 a

1 0 .l .2 .3 .4 .5 .06 .lO .14

EFFLUENT, LITERS

FIG. 1. Purification of o-serine dehydrase by successive ab- sorption and elution from hydroxylapatite (A), diethylamino- ethyl cellulose (B), and hydroxylapatite (C). See the text for details.

tion procedure used is described below; a protocol is given in

Table II.

Step 1: Preparation of Cell Extracts-Wet cells (364 g) har- vested from four carboys were suspended in 900 ml of distilled water and then treated in 50-ml portions for 20 min in a Raytheon ultrasonic disintegrator (10 Kc) at maximum power. The sus- pension was centrifuged for 30 min at 27,000 x g and the pellet was discarded.

Step 9: Protamine Precipitation-The combined supernatant solutions from Step 1 were adjusted to pH 6.0 with 1 N acetic acid and brought to 14”. A 2% solution of protamine sulfate, pH 5.0, was added slowly with stirring to provide 0.3 mg of protamine sulfate per mg of protein in the extract. After stirring for 5 min, the mixture was cooled to between 0” and 5”, and the inactive precipitate was removed by centrifugation and dis- carded. The supernatant fraction was adjusted to pH 7.8 with 1 N KOH. Finely divided ammonium sulfate was then added to 65% of saturation, and the active precipitate was collected by centrifugation.

Step 3: Heat Treatment-The active protein from Step 2 was dissolved in sufficient Buffer A (0.1 M potassium phosphate, 0.01 M m-serine, 5 mM mercaptoethylamine, and 0.04 mM pyridoxal-P, pH 8.0) to give a solution containing 40 mg of protein per ml. This solution was warmed in loo-ml portions in a 65” bath to 55” and kept at 55” for 5 min; then the solution was cooled in ice and the denatured protein was removed by centrifugation.

Step 4: Chromatography on Hydroxylapatite-The supernatant solution from Step 3 was dialyzed for 2 days against five changes (4 liters each) of 0.001 M potassium phosphate, pH 6.8, precipi- tated protein was removed by centrifugation, and the super- natant solution was filtered over a column (6 x 45 cm) of hy- droxylapatite prepared as described by Jenkins (21). The column was developed with 0.005 M potassium phosphate buffer, pH 7.8. The greenish yellow, active fractions between 600 and 800 ml of effluent (Fig. 1A) were pooled, and the enzyme was precipitated by addition of ammonium sulfate to 65 y0 saturation, as described in Step 2.

Step 5: Chromatography on Diethylaminoethyl Cellulose-The precipitated protein from Step 4 was dissolved in 0.01 M potas- sium phosphate, pH 7.8, at a concentration of 40 mg per ml. After dialysis against an excess of this buffer, the protein solu- tion was applied to a column (2 x 30 cm) of diethylaminoethyl cellulose (Bio-Rad, Cellex-D) which had been equilibrated against this same buffer. The column was eluted in a linear gradient between 0.01 and 0.1 M potassium phosphate, pH 7.8. The enzyme appears in the first protein peak to be eluted (Fig. 1B).

Step 6: Rechromatography on Hydroxylapatite-The active fractions obtained between 150 and 300 ml of effluent (Fig. 1B) from the column described in Step 5 were pooled; the protein was concentrated by ammonium sulfate precipitation, dialyzed, and reapplied to a hydroxylapatite column (0.5 x 20 cm) as described in Step 4. The first fractions to be eluted (Fig. 1C) contained essentially pure enzyme.

Step 7: Crystallization-The enzyme was precipitated from the active fractions obtained in Step 6 with ammonium sulfate (see Step 2) and dissolved at a concentration of 20 mg of protein per ml of 0.1 M potassium phosphate buffer containing ammonium sulfate at 25% saturation (4”), pH 7.8. On standing overnight at 3-5”, the crystalline enzyme usually precipitates. If crystals fail to appear, a few drops of saturated ammonium sulfate are added with stirring until a slight silkiness appears. By increas- ing the ammonium sulfate concentration slightly, a second crop of crystalline enzyme can be obtained.

The purified enzyme is bright yellow, and crystallizes in re- markably thin, transparent, elongated platelets (Fig. 2) of the same specific activity as the fraction from Step 6.

by guest on April 23, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Issue of iUarch 10, 1966 Dupouqw et al. 1235

FIG. 2. Crystalline n-serine dehydrase (X 2400)

FIG. 3. Migration of purified n-serine dehydrase in starch gel. Each sample contained 400 ,tg of protein and was subjected to electrophoresis at about pH 8.0 as described by Barrett, Frieson, and Astwood (22). A, crystalline bovine serum albumin; B, crystalline n-serine dehydrase.

Electrophoretic and Sedimentation Properties

Crystalline n-serine dehydrase migrates as a single band in starch gel electrophoresis at pH 8.3 (Fig. 3), and sediments as a single component (s~0,~ = 3.343) in the analytical ultracentrifuge (Fig. 4).’ On centrifugation in a sucrose gradient, good separa- tion from added yeast alcohol dehydrogenase was achieved; horseradish peroxidase and serine dehydrase sedimented at identical rates (Fig. 5). From the SZO,,,, values reported (23) for alcohol dehydrogenase of 7.6 to 6.72, ~~0,~ values for serine dehydrase of 3.8 to 3.36 may be calculated (17). These values are similar to those found by use of the analytical ultracentrifuge and also with those reported for peroxidase, which vary from 3.85 to 3.48 (24). The approximate molecular weight calculated (25) for serine dehydrase varies with the standard chosen. Be- cause serine dehydrase and peroxidase sediment identically, the value of 40,000 found for the latter enzyme (24) appears most

nounced absorption maximum between 405 and 415 mN, indi- cating that the formyl group of bound, pyridoxal-P forms an azomethine link to an amino group of the protein, as in other pyridoxal-P enzymes so far studied. In contrast to glutamate decarboxylase (26) and to aspartate-glutamate transaminase (27), this spectral band is not affected by variations in pH be- tween 6.0 and 9.0 (Fig. 6), but changes are observed between 300 and 350 rnp. On addition of substrate, an intense absorption maximum near 317 rnp appears, a result of pyruvate (cf. Refer- ence 28) formation by enzymatic action; no obvious change in spectrum of the enzyme itself was observed. At pH values be- low 5.5, denatured protein begins to precipitate and the spectrum of free pyridoxal-P appears in the supernatant solution. Quanti- tative measurement of pyridoxal-P released by heating the crystalline enzyme for 20 min at 60” with phenylhydrazine reagent (16), or by heating with cysteine and measuring the amount of pyridoxal-P thiazolidine released by its extinction at 330 rnp (29), both showed the presence of 0.0233 pmole of pyri- doxal-P per mg of protein, corresponding to the presence of 1 mole of pyridoxal-P per 42,800 g of protein. This value cor- responds closely with the approximate molecular weight of 40,000 obtained by comparing the sedimentation of serine dehydrase with that of peroxidase in a sucrose gradient. Like the recently described cystathionine synthetase (lo), n-serine dehydrase is unusual among pyridoxal-P enzymes in possessing only a single binding site for pyridoxal-P per molecule. Pyridoxal-P is not fully released from the enzyme by precipitation by trichloro- acetic acid or treatment with the phenylhydrazine reagent at room temperature.

Substrate SpecQicity

1 We are indebted to Dr. I-I. K. Schachman for this ultracentrifu- n-Serine is the preferred substrate for the enzyme, but D-

gal run. threonine is attacked at a lower rate (Table III). Through its

reasonable; this differs substantially from the value of 58,000 calculated on the assumption (23) of an ~20,~ value of 7.4 and a molecular weight of 150,000 for yeast alcohol dehydrogenase.

Spectrophotometric Behavior and Pyridoxal Phosphate Content The spectrum of n-serine dehydrase (Fig. 6) shows a pro-

by guest on April 23, 2020

http://ww

w.jbc.org/

Dow

nloaded from

1236 D-Serine Dehvdrase Vol. 241, No. 5

FIG. 4. Sedimentation patterns of n-serine dehydrase after (from 1 to 4) 16, 26, 56, and 80 min at 59,780 rpm in the Spinco model E analytical centrifuge. Photographs were taken at bar angles of 75”, 75”, 70”, and 70”, respectively. The sample cell contained 1.0% protein in 0.1 M potassium phosphate buffer, pH 7.8. The blank cell contained buffer only.

0.10, I /Al.2 I I I II’1

E

5 0.08 - E

2 9 4

g 0.06 -

% z z - ‘z, it pH 9.0

8 0.04- I II I II I ,,I,,,,I

300 350 400 450

9 WAVELENGTH, mp

FIG. 6. Variation in spectrum of n-serine dehydrase with pH. 0 0 Cells contained 3.5 mg of protein (specific activity, 230) per ml of

0 4 8 12 16 20 40 0.1 M potassium phosphate buffer at the indicated pH. The

TUBE NUMBER curves have been offset slightly to show the identity in shape of the absorption curve above 400 rnp. The absorbance at 410 rnp was

FIG. 5. Separation of yeast alcohol dehydrogenase (ADH, 0.08 identical in the three Cakes. mg), horseradish peroxidase (Worthington, 0.08 mg), and n-serine dehydrase (SDH, 0.08 mg) by centrifugation in a sucrose gradient. same active site. Centrifugation was performed for 11 hours at 38,000 rpm and 0” in

n-Cysteine, D-tryptophan, and several other

the SW-39 rotor of the Beckman model L centrifuge. The sub- n-amino acids are inactive both as substrates and as inhibitors

strate for peroxidase was o-dianisidine. For alcohol dehydro- under standard assay conditions in which pyridoxal-P is added. genase and peroxidase, the plotted values represent the change in o-Serine dehydrase is thus a much more specific enzyme than absorbance at 340 rnp and a;t 460 mE.r, respectively, per 10 ~1 of sam- ple per 30 set; for n-serine dehydrase they represent micromoles

the tryptophanase of ,%‘. coli (7, 8), although both are inducible

of pyruvate produced per ~1 of sample in 10 min. enzymes and catalyze formally similar (Y ,p elimination reactions. O-Methylserine is a potent competitive inhibitor of the reaction.

action as a competitive substrate, n-threonine also inhibits the rate of keto acid formation from n-serine, and its K; value in

v “1 I

this capacity is equal to its K, value when it serves as sole sub- The optimum pH for action of o-serine dehydrase is 7.8 to

strate,2 indicating that both substrates are acted upon at the 8.0 (Fig. 7A). As reported previously (4), Tris buffer inhibits the enzyme (Fig. 7A), but simultaneous additions of sufficient

2 When both n-serine and n-threonine were added as competitive substrates, pyruvate was determined with lactate dehydrogenase.

K+ or NHd+ were found to prevent this inhibition (Fig. 7B)

Since n-serine is the nreferred substrate for n-serine dehvdrase, for lactate dehydrogenase, the error in determination of the Ki and pyruvate is much superior to a-ketobutyrate as a substrate value for n-threonine is small (cf. Reference 30).

by guest on April 23, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Issue of March 10, 1966 Dupourque et al. 1237

almost completely. Na+ is much less effective. The nature of this protective effect of monovalent cations is not understood; tryptophanase (7, 31), the B protein of tryptophan synthetase (9), and the n-serine and n-threonine dehydrases of mammalian livers (ll)-all of which are pyridoxal-P enzymes and catalyze (Y,P elimination or addition reactions-all are activated to a greater or lesser extent by K+ or NH4+. With n-serine dehy- drase the effect may be related in some way to the binding of pyridoxal-P, for immediate addition of pyridoxal-P to a Tris- enzyme preparation partially protects against inhibition, al- though less effectively than K+. If the enzyme is incubated in

TABLE III

Substrate specificity of o-se&e dehydrase*

Substrates Ki value of competitive inbibitorst

* The following compounds were found to be inactive: L-serine, L-threonine, n-cysteine, IX-homoserine, n-tryptophan, n-alanine, and n-aspartate. These compounds were neither substrates nor inhibitors of n-serine degradation when assayed under standard conditions at concentrations up to 13.2 mrvr. At higher concen- trations (0.1 M), several amino acids (glycine, DL-allothreonine, p-alanine, cu.n-alanine, cu-methyl-DL-serine, DL-homoserine) in- hibit; such inhibitors are more effective when n-threonine re- places n-serine as substrate.

t Tested with n-serine (1.32 mM) as substrate.

6 7 a 9 1 10

PH CONCENTRATIOt$mM

FIG. 7. A, pH optimum of n-serine dehydrase in 0.1 M potassium phosphate (Curve 1), 0.1 M Tris-chloride plus 0.1 M KC1 (Curve Z), 0.1 M Tris-chloride plus 0.01 M KC1 (Cuurve 8), and 0.1 M Tris- chloride alone (Curve 4). The enzyme preparation was freed of alkali metal ions by passage through Sephadex G-50 immediately before the determinations. B, stabilization of n-serine dehydrase in 0.1 M Tris buffer, pH 7.8, by increasing amounts of NaCl, KCl, or NH&I. The enzyme preparation was the same as that used in A.

1 2 3 4 5 6

PYRIDOXAL PHOSPHATE,pM

FIG. 8. Reactivation of n-serine apodehydrase by pyridoxal-P (PLP). The enzyme was dialyzed for 54 hours against 0.04 M cysteine plus 0.02 M mercaptoethanol, and then assayed. Stand- ard assay conditions were followed except for addition of graded amount of pyridoxal-P.

Tris buffer alone for more than 3 min at 37”, no reactivation by either K+ or pyridoxal-P is obtained.

Resolution and Reconstitution of dkrine Dehydrase

In the absence of added pyridoxal-P, both D- and n-cysteine at moderate concentrations (0.005 to 0.02 M) and pH 7.8 rapidly inactivate n-serine dehydrase, even though potassium phosphate is present. This inactivation is prevented by added pyridoxal-P. After dialysis against potassium phosphate buffer, such inac- tivated preparations no longer exhibit the spectrum of a pyri- doxal-P protein, showing that the prosthetic group has been removed. However, no reactivation by added pyridoxal-P is obtained. This inactivity appears to result from sensitivity of the resolved enzyme to denaturation and oxidation, for if small amounts (2 mg per ml) of bovine serum albumin are added, the enzyme can be completely resolved by dialysis for several days against 0.04 M nn-cysteine plus 0.02 M mercaptoethanol, and activity can be fully restored by addition of pyridoxal-P (Fig. 8). Half-maximum activity under these conditions is restored by 0.0014 lllM pyridoxal-P, but this figure probably underesti- mates the true afhnity of apoenzyme for its coenzyme because of the presence of cysteine.

1. 2.

3.

4.

5.

6.

7.

8.

REFERENCES

GALE, E. F., AND STEPHENSON, M., Biochem. J., 32,392 (1938). YANOFSKY, C., AND REISSIG, J. L., J. Biol. Chem., 202, 567

(1953). WOOD, W. A., AND GUNSALUS, I. C., J. Biol. Chem., 181, 171

(1949). METZLER, D. E., AND SNELL, E. E., J. Biol. Chem., 198, 363

(1952). BENZIMAN, M., SAGERS, R. D., AND GUNSALUS, I. C., J. Bac-

teriol., 79, 474 (1960). SAYRE, F. W., AND GREENBERG, D. M., J. Biol. Chem., 220,

787 (1956). NEWTON, W. A., AND SNELL, E. E., Proc. Natl. Acad. Sci.

77. S., 61, 382 (1964). NEWTON, W. A., MORINO, Y., AND SNELL, E. E., J. Biol.

Chem., 240, 1211 (1965).

by guest on April 23, 2020

http://ww

w.jbc.org/

Dow

nloaded from

1238 D-Serine Dehydrase Vol. 241, No. 5

9. CRAWFORD, I. P., AND ITO, J., Proc. Natl. Acad. Sci. U. S., 51, 390 (2964).

10. NAGABHUSHANUM, A., AND GREENBERG, D. M., J. Biol. Chem., 240, 3002 (1965).

11. NISHIMURA, J. S., AND GREENBERG, D. M., J. Bio.?. Chem., 236, 2684 (1961).

12. YANOFSKY, C., J. Biol. Chem., 198, 343 (1952). 13. LABOW, R. S., AND ROBINSON, W. G., J. Biol. Chem., 241, 1239

(1966). 14. FRIEDEMANN, T. E., AND HAUGEN, G. E., J. Biol. Chem., 147,

415 (1943). 15. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDFLL,

R. J., J. Biol. Chem., 193, 265 (1961). 16. WADA, H., AND SNELL, E. E., J. Biol. Chem., 236, 2089 (1961). 17. MARTIN, R. G., AND AMES, B. N., J. Biol. Chem., 236, 1372

(1961). 18. PARDEE, A. B., AND PRESTIDGE, L. S., J. Bacterial., 70, 667

(1955). 19. MCFALL, E., J. Mol. Biol., 9, 746 (1964).

20. RICKENBERG, H., YANOFSKY, C., AND BONNER, D. M., J. Bacterial., 66, 683 (1953).

21. JENKINS, W. T., Biochem. Prep., 9, 83 (1962). 22. BARRETT, R. J., FRIESEN, H., AND ASTWOOD, E. B., J. Biol.

Chem., 237, 432 (1962). 23. SUND, H., AND THEORELL, H., Enzymes, 7, 59 (1963). 24. MAEHLY, A. C., Methods Enzymol., 2, 808 (1955). 25. SCHACHMAN, H. K., Ultracentrifugation in biochemistry, Aca-

demic Press, Inc., New York, 1959. 26. SHUKUYA, R., AND SCHWERT, G. W., J. Biol. Chem., 235, 1653

(1960). 27. JENKINS, W. T., AND SIZER, I. W., J. Biol. Chem., 234, 1179

(1959). 28. ERRERA, M., AND GREENSTEIN, J. P., Arch. Biochem., 15, 477

(1947). 29. SCHIRCH, L., AND MASON, M., J. Biol. Chem. 238, 1032 (1963). 30. FLAVIN, M., AND SLAUGHTER, C., J. Biol. Chem., 239, 2212

(1964). 31. HAPPOLD, F. C., AND STRUYVENBERG, A., Biochem. J., 68,

379 (1954).

by guest on April 23, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Daniel Dupourque, W. Austin Newton and Esmond E. SnellEscherichia coliPurification and Properties of d-Serine Dehydrase from

1966, 241:1233-1238.J. Biol. Chem. 

  http://www.jbc.org/content/241/5/1233Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/241/5/1233.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on April 23, 2020

http://ww

w.jbc.org/

Dow

nloaded from