Characterization of cyanobacterial ferredoxin-NADP+ oxidoreductase molecular heterogeneity using chromatofocusing

4NALl’TICAL

BIOCHEMISTRY

154, 441-448

(1986)

Characterization of Cyanobacterial Molecular Heterogeneity

Ferredoxin-NADP+ Oxidoreductase Using Chromatofocusing

AURELIO SERRANO

Received

October

2 I ~ 1085

Chromatofocusing has been used as an analytical tool to check preparations of the enzyme ferredoxin-NADP’ oxidoreductase (EC I. 18. I .2) purified in either the presence or absence of the serint: protease inhibitor phenylmethylsulfonyl fluoride from the cyanobacterium .tr~ahuenu sp. strain 71 19. Only one isoelectric species was found when the crude extract was processed in the presence of the protease inhibitor. Nevertheless, when the inhibitor was omitted, four ionic forms of the enzyme-showing apparent pi’s in the range 4.3-4.6--were separated after chromatofocusing ofthe purified preparation. These forms were found to differ in their specific activities. exhibiting, on the other hand. lower values than the single one obtained in the presence of the protease inhibitor. Analysis hy acrylamide gel clectrophoresis revealed virtually a single main protein band except for the ionic form of pl 4.39, which was clearly resolved into two active components. Except for the more basic form. which seems to be an homodimer of ,\I, 80.000. all the protein components were found to he monomeric species in the range hl, 33,000-38,000. These results indicate that the molecular heterogeneity ofthe ferredoxin-NADP+ oxidoreductase purified from the cyanohacterium .trrahuc~r~u sp. strain 7 I 19 may result from the activity of a protease present in the whole cell homogenates. On thr other hand. these data also point out that chromatofocusing should be considered as an etfcctive technique in the isolation and characterization of the different molecular forms of this enzyme. r 1986 Academw Press, Inc. KEY WORDS: protein chromatography: chromatofocusing: molecular heterogeneity: flavoproteins; cyanohacteria.

The flavoprotein ferredoxin-NADP’ oxidoreductase ‘(FNR)’ (EC 1.18.1.2) is an FADcontaining enzyme which is present in all the organisms which perform oxygenic photosynthesis ( 1). Since the early works of Arnon and co-workers (2.3) it seems clear that the main physiological function of FNR is the reduction of NADP+ using photosystem I-reduced ferredoxin. but the study of the molecular properties of the enzyme has been hindered by the occurrence of multiple molecular forms, in both cell-free extracts and purified prepara’ Abbreviations used: BAPNA. .Vcy-benzoyl-DL-arginine-p-nitroanilrde hydrochloride: FNR. ferredoxinNADP’ oxidoreductase: INT. 2-(p-rodophenyl)-3-nitrophenyl-5-phenyltetrazolium chloride; PBE. Polyhuffer Exchanger: PMSF. phenylmcthylsulfonyl fluoride: SDS. sodium dodecyl sulfate.

tions (4). The various enzyme species isolated either from chloroplasts (5.6) or cyanobacterial cells (7.8) differ only slightly in their molecular weights. pl, and amino acid composition. There are. on the other hand, discrepancies concerning number-from two to nine-and relativ’e amounts of forms found by different authors, even when using the same source of enzyme (9-12). Although recent observations indicate that the molecular heterogeneity observed in the FNR from both spinach ( 13) and the non-Nzfixing cyanobactcrium Spimlinrr plutcnsis ( 14.15) was due to proteolytic digestion of the flavoprotein at the amino terminus. the subtle differences in physicochemical parameters of the multiple molecular forms make it difficult to isolate them using the usual chromato-

graphic techniques ( 13), isoelectric focusing providing, on the other hand. only small amounts of protein ( 13). In this communication the proteolytic origin of the molecular heterogeneity of FNR from the N,-fixing cyanobacterium ..lr~ahuc~?rr sp. strain 7 1 19 is reported. The obtained results suggest. therefore, that a single molecular species of the enzyme-which should be modified during the purification procedureis synthesized by the cyanobacterial cells. Chromatofocusing (16) has been used as the technique of choice for separating the different molecular forms of the FNR. Taking advantage of the high resolution and the semipreparative character of this chromatographic method (17). the several ionic species of this tlavoprotein have been obtained in quantities sufficient to allow further studies on their structure and biological significance. MATERIALS

AND

METHODS

Centrifugations were carried out in a Sorvall RC-2B centrifuge equipped with a rotor SS34. Spectrophotometric determinations were performed using a Pye-Unicam SP8-I 50 uv double-beam recording spectrophotometer. c’l~cwicds. 2’,5’-ADP-Sepharose 4B, Polybuffer Exchanger (PBE) 94, Polybuffer 74. and protein standards for electrophoresis were purchased from Pharmacia Fine Chemicals (Uppsala. Sweden): t-amino-n-caproic acid, bovine serum albumin, N-cu-benzoyl-DL-arginine-I)-nitroanilide hydrochloride (BAPNA), leupeptine. lysozyme. piperazine. phenylmethylsulfonyl fluoride (PMSF), Tris. Triton X-100, and EDTA from Sigma Chemical Company (St. Louis, MO.); NADPH and cytochrome c from Boehringer (Mannheim, West Germany): DEAE-cellulose DE-53 from Whatman, Inc. (Maidstone, England): and 2(I, - iodophenyl) - 3 - nitrophenyl - 5 - phenyltetrazolium (INT) and ingredients for polyacrylamide gel electrophoresis from Serva (Heidelberg. West Germany). All other chemicals were of analytical grade and were acquired from Merck (Darmstadt. West Germany).

Gt’ct~c./lr c!/ lltc ot:qrtti.vtt.

,Yo.slftc~ ttiu.wttattt

strain 7 I 1‘$-reclassified as . Itdwcwu sp. strain 7 I I9 ( I X)-from the Department ofCell Physiology. Berkeley. California (a gift of Dr. D. I. Arnon) was grown photoautotrophically under the conditions previously described ( 19). except that no combined nitrogen source was added to the culture medium. The cells were harvested at the late exponential grown phase (about 4 days after inoculation) with a continuous flow system Szent-Gyorgyi-Blum (Sorvail. Newton, Conn.), washed twice in 50 mM Tris-HCI buffer. pH 7.5, containing 0.1 mM EDTA. and stored at -20°C before use. The yield was 3-3 g (wet wt) per liter of culture medium. Protein drt~rtt?itlntiotl. Protein in crude extracts was determined by a modification ofthe method of Lowry ct al. (10). In latter stages of the purification procedure and in pure preparations, protein was determined according to the method of Bradford (3 I) or by uv absorption (22). Bovine serum albumin and lysozymc were used as standards. Etz~ww CI.V.SC~J~.Y. Ferredoxin-NADP+ oxidoreductase was assayed spectrophotometritally by following the ferredoxin-dependent NADPH-cytochrome c reductase activity according to the procedure described by Shin (13). One unit of enzyme is defined as the amount which catalyzes the reduction of 1 pmol of cytochrome c per minute at 35°C. Ferredoxin from .-ltluhucrzu sp. strain 7 1 I9 was prepared as described by Mitsui and Arnon (24). Protease activity was assayed by spectrophotometrically determining the p-nitroaniline formed upon hydrolysis of the artificial substrate BAPNA (amidase activity) (35). Reaction mixtures contained in a final volume of 1 ml: 100 ymol of Tri-HCl buffer, pH 8.0: 0.35 pmol of EDTA: 1 pmol of BAPNA: and an adequate quantity of cell-free extract. One unit of amidase activity was defined as the amount of enzyme which produces 1 pmol of p-nitroaniline per minute at 25°C. G’c)l c~i~~l~oj~/2or.rsis. Polyacrylamide gel electrophoresis was carried out according to Jovin (V c/l. (76) on running columns of 5

CHROMATOFOCUSING

OF

FERREDOXIN-N.4DP+

X 75 mm at the concentrations specified in each case, using a stacking gel of 3.5% (w/v) acrylamide. Samples containing about 25 pg protein were applied to the gels; electrophoresis was first performed at 2 mA/gel tube for 15 min and then at 4 mA/gel tube for 2 h. Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS) was performed according to Laemmli (37) using stacking and separating gel columns containing 3 and 10% (w/v) acrylamide, respectively. A current of 2 mA/gel tube was applied for 1-4 h. Proteins were stained with 1% (w/v) Coomassie brilliant blue in 7% (v/v) acetic acid for 4-6 h at room temperature. NADPH diaphorase activity in the gels was determined as described by Gozzer et rrl. (6). using INT as electron acceptor and buffered by 0.1 M TrisHCl, pH 8.0. nc,tc~rnzirlLItiOll qf mc~l~~cxlrrrttvi,yht.s. The native protein molecular weights were estimated by running adequate amounts of protein on polyacrylamide gels of 3.75. 5. 6.25, 7.5, and IOr:: (w/v) acrylamide concentrations as described by Hedrick and Smith (28). Molecular weights under denaturing conditions were determined by SDS-polyacrylamide gel electrophoresis as described above. by comparing the relative mobility of the enzyme with those of stanldard proteins. ..lrnino CccYd nnu/~~si.s.The amino acid mixtures obtained by treating samples of the purified FNR preparations (0.5-I mg of protein) with 6 N HCI in vacuum sealed tubes (I 10°C for 24 and 48 h) were processed with a JEOL JLC-6AH analyzer (Japan), using norleucine as an internal standard and ninhydrin as a postcolumn staining reagent (39). Plwificatioti of’ .~lrlulwIcnu sp. slruirr 7119 FNR. The enzyme has been purified according to the method described by Serrano and Rivas (30). The cells were resuspended in 50 mM Tris-HCI buffer. pH 7.5, at the ratio of 10 ml/ g wet wt cells. and disrupted by incubation at 4°C with gentle stirring in the dark for 6-X h in the presence of 2% (v/v) of the nonionic detergent Triton X- 100. Batches of .trzcthactrn sp. strain 7 I 19 cell paste coming from the

OXIDOREDUCTASE

443

same culture were divided into two similar portions. typically about 70 g (wet wt) each, one of which was processed in the presence of 0.1 mhl PMSF and 2 mM EDTA during the enzyme extraction treatment. In both cases, however, the resulting cell-free extract was considered as the crude extract and used as such for the subsequent chromatographic procedures, namely ion exchange chromatography on DEAE-cellulose DE-52 and affinity chromatography on ?,5’-ADP-Sepharose 4B. C’I?r’or)?Llf(!fi)(.lI.si11,4. Column chromatofocusing in the pH range 5 to 4 was carried out at 4°C on a PBE 94 column (1 X 75 cm) equilibrated with 25 mM piperazine-HCI buffer. pH 5.3. according to the indications of the manufacturer (16). Samples of the purified FNR preparations. containing 3-13 mg of protein in a volume of 5 ml. were dialyzed against 10 mM piperazine-HCI buffer. containing 0.1 mM EDTA. and applied to the column which was afterward washed with 5 ml of equilibrating buffer. The enzyme was eluted with the pH gradient created by washing the column with I:! bed volumes of 1O-fold diluted Polybuffer 74-HCl, pH 4.0, containing 0.1 mM EDTA, at a descending flow rate of 11 ml/h. All buffers were completely degassed before use. The pH values of the fractions were measured by using a Beckman SS-2 pH meter with an expanded scale. RESULTS Extraction with the nonionic detergent Triton X-100 is a mild and very efficient treatment which has been previously used to solubilize active FNR (30) and other flavoproteins (3 I ) from .tncihaf~t~r sp. strain 7 1 19 cells. Practically all the enzyme activity was recovered in the soluble protein fraction after centrifugation (40.000~~. 20 min) of the cell-free extract (sp act, 0.36 LJ/mg of protein). A significant proteolytic activity-about I50 mU/ mg of protein-was also determined in the supernatant by using the amidase assay with BAPNA as a substrate. Nevertheless. in the presence of 0.05-0.15 mM PMSF a clear decrease in the amidasc activity, up to 60% of

444

AURELIO

SERRANO

the control without inhibitor, was observed. A similar but less pronounced effect was produced by higher concentrations ( l-2 mM) of other serine protease inhibitors. namely leupeptine and t-amino+-caproic acid. On the other hand, the presence of PMSF during the extraction procedure resulted in a significant increase of the FNR activity-from I .5 to 2.0 times as much as the observed without inhibitor-when assayed in the crude extracts by using the ferredoxin-dependent NADPH-cytochrome c reduction, which is a specific test for this enzyme. The presence of the protease inhibitor PMSF during the extraction of the cyanobacterial FNR did not affect the behavior of the enzyme during the subsequent chromatographic steps ofthe purification. Moreover. the purified enzyme showed, in any case, an absorbance ratio 2,1274/,+t 45h-a characteristic index of purity for flavoproteins-of 8.0, similar to those obtained for the apparent homogeneous enzyme isolated either from chloroplasts ( 13) or other cyanobacterium strains (8). Nevertheless, two differential features. clearly related to the presence or absence of the protease

l

.

inhibitor during the solubilization treatment. were inferred from the comparative study of both purified FNR preparations: (i) when the enzyme was exttactcd in the presence of PMSF a specific activity of I I5 U/mg of protein was determined, which is about twice the value obtained without inhibitor (65 U/mg); and (ii) as previously described (30). a single protein band was observed upon electrophoresis on 7.5% (w/v) polyacrylamide gels when the FNR was purified with the protease inhibitor present: otherwise this technique revealed a cluster of four protein bands, all of them showing NADPH-INT diaphorase activity in si/l/ and NADPH-cytochrome ~‘reductase activity after extraction of the gels. Taking advantage of the high resolution and the semipreparative character of column chromatofocusing ( 17). this method has been used to analyze the molecular heterogeneity of the cyanobacterial FNR and also to isolate the different forms of this enzyme. Figure I shows the chromatofocusing elution profile of protein and enzyme activity of the FNR purified in the presence of PMSF during the solubilization treatment. The only

.

.

ye-*-

_ 20

30

l

B 50

40 FRACTION

60

NUMBER

FIG. I. Column chromatofocusing on PBE 94 of Admm sp. strain 7 I 19 FNR purified in the presence of the protease inhibitor PMSF. A sample containing 3.5 mg of protein was applied to a PBE 94 column (I x 25 cm). and the enzyme was eluted by using a pH gradient (0) generated by 12 bed volumes of Polybuffer 74-HCI. pH 4.0 (7.5 pmol/pH unit/ml). Fractions of 7 ml were collected. Absorbance at 280 nm (0) and enzyme activity CM) were measured for each fraction.

(‘HROMATOFOCUSING

o ~.~n 20

OF

FERREDOXIN-NADP+

I--40

OXIDOREDUCTASE

445

L60

FRACTION

80

NUMBER

FK;. 1. Column chromatofocusing on PRE 94 of .4nnhacwu sp. strain 7 I I9 FNR purified in the absence of the proteasc inhibitor PMSF. A sample containing 13.5 mg of protein, purified from the same batch of cells used in Fig. I. was chromatographed under the conditions described in the legend of Fig. I.

peak, which eluted at a pH value ofabout 4.35 (estimated apparent ~1). shows a close parallelism between both parameters. thus suggesting that a single isoelectric species is present in the purified enzyme preparation. Nevertheless, column chromatofocusing performed under the same conditions clearly resolved the FNR preparation purified in the absence of proteasc inhibitor into four components with pi‘s in the range 4.3-4.6. all of them being catalytically active (Fig. 3). They are namedfollowing the elution order under chromatofocusing-as FNR-I to FNR-IV, going from the most basic to the most acidic form. FNRIV appeared as the predominant species in these preparations. being about 50% ofthe total protein: in contrast, FNR-I was only a minor component. All the forms were flavoproteins. since the ionic species obtained upon chromatofocusing exhibited an intense yellowish appearance. and absorbance spectra of peak fractions showed maxima at 390 and 456 nm, typical of FADcontaining enzymes (32). Spectral ratios A& ,A456were in the range 7.7-8.7 (see Table l), very similar to those observed for the assumed

homogeneous FNR purified from other sources (8.12). Most notably, as the apparent pl of the diKerent forms decreased. the specific activity fell down in a parallel manner. so that the specific activity of the most acidic form, FNR-IV. was only about 30% of that determined for FNR-II (Fig. 2 and Table I). The peak fractions showed in t; ~ :-hromatofocusing elution profile of Fig. 3 corresponding to FNR-I, II. and IV gave one major protein band on polyacrylamide gel electrophoresis, whereas FNR-III clearly exhibited two electrophoretically distinct protein components (Fig. 3). The minor band-less than 5’;; of the total protein applied to the gelobserved upon electrophoresis of FNR-II and III may be due to a slight overlapping with the main tbrm FNR-IV. Protein bands of all the four fractions corresponded with bands of catalytic iactivity (NADPH-INT diaphorase) in duplicate gels. A correlation was evident between apparent pl of FNR forms and electrophoretic mobility of their protein components, suggesting that the resolution of the different forms by electrophoresis is mainly due to actual charge differences among the molecules

446

AURELIO

SERRANO TABLE

I

CHARACTERIZA-I-IONOF MULTIPLE Iolvrc FORMSof: FNR FROM .~&ww sp. strain 71 19 P~WFIED IN Tfw ABSENCE:OF THE PROTEASI:INHIBITOR PMSF Estimated parameter”

FNR-I

FNR-II

Molecular weight Gel multiple concentrations SDS-gel electrophoresis

79.000

3X.500h

38.000

34,000

36,000

35.000h

35,000

33,000

4.56

4.39

x.75

7.87 83

4.63

PI

&74/&6

ND’

Sp act (U/mg protein) K,,, NADPH (/AM)

ND ND

105 8.6

FNR-III

7.8

FNR-IV

4.31 7.76 35 9.1

a Data were obtained from the analysis of peak fractions from column chromatofocusing of Fig. 2. ’ Molecular weights of the two protein components were identical within experimental error of methods. ’ ND: Not determined

rather than the molecular sieving action of the gel. Chromatofocusing has been used successfully in this work to obtain milligram quantities of each major isoelectric species present in the cyanobacterial FNR preparations. Therefore, this technique allowed characterization, both in physical and catalytic properties, ofthese forms (Table 1). The three main components are monomeric proteins exhibiting molecular weights in the range 33,00038,000. These data are, therefore, in agreement

0

FIG. 3. Polyacrylamide gel electrophoresis of the indicated peak fractions from the chromatofocusing of Fig. 2. Samples containing about 25 pg of protein were run in 7.5% polyacrylamide gels which were stained for protein. The arrow indicates the tracking dye position.

with the slight differences observed in apparent p1 values and electrophoretic mobility, being moreover very similar to the estimated molecular mass (38,000 Da) of the single form obtained in the presence of the protease inhibitor. The minor form FNR-I was a noteworthy exception since it was an homodimer of ,Ur 80,000 (Table 1). On the other hand, the presence of two proteins as components of the ionic form FNR-III could be due either to the proteins having charge isomers, or to proteins migrating on chromatofocusing in association with one another as a complex: nevertheless, they can eventually be resolved by using other techniques such as electrophoresis or isoelectric focusing. The amino acid composition of the three main ionic forms of .-lrzuhac~ct sp. strain 7 I 19 FNR revealed no clear distinctive features, as could be expected from the very similar molecular properties of these proteins (data not shown). As shown in Table 1 there are clear differences in the rate of ferredoxin-dependent cytochrome c reduction catalyzed by the multiple ionic forms of the cyanobacterial FNR with only FNR-II reaching a value similar to that presented by the single form obtained with PMSF. Nevertheless, no appreciable differences were observed for both the Michaelis constant (K,) for NADPH and I ’ of the main forms: thus, it would appear that the possible structural changes giving the differences in

CHROMAT’OFOCUSING

OF

FERREDOXIN-NADP’

specific activity observed are due to changes in the affinity for the ferredoxin-cytochrome c acceptor. DISCUSSION

As soon as a crude extract is obtained, the resulting loss of cellular organization allows protein interactions which do not occur in the living cell. Proteolytic degradation is one of such processes: therefore, protease inhibitors are now widely used in enzyme purification (33). The usefulness of the serine protease inhibitor PMSF in the purification of FNR from .~l~h~nrl sp. strain 71 19 is shown in this work. In fact, a proteolytic activity is present in crude extracts of this microorganism. PMSF behaving as an inhibitor of BAPNA hydrolysis. Nevertheless. only a partial inhibition was achieved. a fact that can be explained by the presence of cystcinyl proteases in the extracts. These proteolytic enzymes show amidase activity but are not affected by this inhibitor (25). On the other hand. a protective effect of PMSF on the cyanobacterial FNR-probably avoiding a proteol)tic alteration of the enzyme-is suggested by the higher specific activity value of cell-free extracts obtained in the presence of the protease inhibitor. The presence of PMSF during the first step of the purification procedure was also essential to obtain in the purified FNR preparations both a single protein band on polyacrylamide gel electrophoresis and only one isoelectric species upon chromatofocusing. Otherwise, four ionic forms and at least five protein components, all of them catalytically active, were resolved by using these techniques. Molecular heterogeneity of FNR has been a controversial subject for a long time. and authors have been in disagreement about the physiological significance of this phenomenon (6.1x,14.34). Nevertheless, a limited proteolytic digestion at the amino terminus of the enzyme has been recently demonstrated for both the spinach (I 3.35) and the Spirdinu p/~~fr~n.si.s( 14,151) FNR. These results are. therefore. consistent with the proteolytic origin and the very similar properties showed in this

447

OXIDOREDUCTASE

work for the multiple ionic forms of the FNR from the N2-fixing cyanobacterium .,inc~huenr~ sp. strain 7 119. On the other hand, the slight differences observed for molecular properties of the different forms severely complicated the achievement of good separation by conventional chromatographic techniques, such as ion-exchange chromatography ( 14); moreover, isoeleclric focusing gave only small quantities of protein ( 13). In this work chromatofocusing has been revealed as a useful technique in the resolution of multiple ionic forms of the cyanobacterial FNR. The main conclusion of this work is, therefore, not only demonstration of the proteolytic origin of molecular heterogeneity of the FNR isolated from a N2-fixing cyanobacterium strain, but also illustration of the usefulness of chromatofocusing in the isolation and characterization of multiple forms of this enzyme. The high resolution and the semipreparative character of this method make it suitable to obtain these isoproteins in quantities large enough for further studies about their structure (immunology, X-ray diffraction) and possible biological signiticance. ACKNOWLEDGMENTS This research was supported Asesora de Investigacibn, Spain The author is deeply grateful to his generous encouragement and

by a grant of Cornis& (Project No. 1 I52/8 I I. Professor M. Losada for help.

REFERENCES I

2. 3. 4.

5. 6.

Biiger. P. ( 1979) ipz Encyclopedia of Plant Physiology, hew Series (Gibbs. M., and Latzko. E.. eds.). Vol. 6. pp. 401-404. Springer-Verlag, Berlin. Shin. M., Tagawa. K.. and Arnon. D. I. ( 1963) B/oc~hcvn. L:. 338, 84-96. Shin. M., and Arnon. D. I. (I 965) J. Biol C’hn. 240, 1~405-1411. Carrillo. N.. and Vallejos. R. H. (1986) in Current Topics in Photosynthesis: Photosynthetic Electron Transport (Barber. J.. ed.). Vol. 8, Chapter IO. Elsevier. New York/Amsterdam, in press. Fredricks. W. W.. and Gehl, J. M. (1976) ..lrclr. b”rt~c~11cvn. Bioptin. 174, 666-674. Gozrer. C.. Zanetti. G.. Galliano. M., Sacchi. G. A.. Minchiotti. L.. and Curti. B. (1977) Biochirn. Bioptr)x .-lctu 485, 278-290.

448

AURELIO

7. Rowell. P.. Dicz. J., Apte, S. K., and Stewart. W. D. P. ( I98 I ) Bioc~him Biophyx :lcra 657, 507-5 16. 8. Masaki. R., Wada. K.. and Matsubara. H. ( 1979) J. Biochcm. 86, 95 I-962. 9. Keirns. J. J.. and Wang. J. H. (lY72) J. Biol. C’hw~. 24.7374-7382. IO. Effelson. W. L., and Krogman. D. W. (1979) ;Irc& Biochem. Bioph!x 194, 593-599. 1 1. Dutton, J. E.. and Rogers. L. J. ( 1979) Bio~h~r~. Sot. Trunx 7, I262- 1264. 12. Shin, M., and Oshino. R. (1978)J. Biochenz. 83,357361. 13. Hasumi, H.. Nagata. E.. and Nakamura, S. (1983) Biochern. Biophy~s. Rev. Commun. 110, 280-786. 14. Wada. K., Tamura, T.. Matsubara. H.. and Kodo. K. (1983) J. Biocl~cw. 94, 387-393. IS. Yao. Y.. Tamura, T.. Wada, K.. Matsubara, H.. and Kodo, K. (1984) J. Biochem 95, 15 13-l 5 16. 16. Chromatofocusing ( 1980) Pharmacia Fine Chemicals. Uppsala. Sweden. 17. Sluyterman. L. A. R. (1982) Truzd,s Rio&m Scr 7, 168-170. 18. Rippka. R., Deruelles. J. B.. Waterbury, M.. Herdman, M.. and Stanier, R. Y. (1979) .I. Gm. Microhiol. Ill, I-61. 19. Serrano. A., Rivas, J., and Losada, M. (198 1) Photos,vnrh. Res. 2, 175-184. 20. Yocum, R. R.. Blumberg. P. M., and Strominger. J. L. (1974) J. Bin/. Chem. 249, 4863-487 I. 2 I. Bradford, M. M. (1976) .dnal. Biochenz. 72, 248-254. 22. Layne, E. (I 957) in Methods in Enzymology (Colow-

SERRANO

73.

14. 25.

26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

ick, S. P., and Kaplan. N. 0.. eds.). Vol. 3. pp. 447-454, Academic Press, New York. Shin. M. (1971) in Methods in Enzymology (San Pietro. A.. rd.). Vol. 23. pp. 440-447. Academic Press. New York. Mitsui. A.. and Amon. D. 1. (I97 I) Ptij,c/r~i. Plartt. 25, 135- 140. Arnon. R. (1970) in Methods in Enzymology (Perlmann. G. E., and Lorand. L.. eds.). Vol. 19. pp. 226-244. Academic Press, New York. Jovin. T., Charamback, A.. and Naughton. M. A. ( 1964) Anal. Biochm~. 9, 35 l-364. Laemmli. U. K. (1970) Nufurc (London) 227, 680685. Hedrick. J. L.. and Smith. A. J. ( 1968) .irc,h. Biochm. Biophys. 126. I55- 164. Spackman, D. H.. Stein. W. H., and Moore. S. (1958) .jnal. Chcm 30, I 190-1206. Serrano. A.. and Rivas. J. ( 1982) .4na/. Rio&m 126, 109-l 15. Serrano, A.. Rivas, J.. and Losada. M. ( 1984) J Bacc leririoi. 158, 3 17-374. Morton. R. A. (1975) Biochemical Spectroscopy, Vol. II. pp. 427-435, Adam Hilger. London. Scopes. R. (I 982) Protein Purification. Principles and Methods, pp. l97-lY9, Springer-Verlag. New York. Hutber. G. N., and Rogers. L. J. (198 I) Photoswzih. Rc\. 2, 269-280. Karplus. P. A., Walsh. K. A., and Herriott. J. K. (1984) Biochcrnixlr~~ 23, 6576-6583.