THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 260, No. 1, Issue of January 10, pp. 237-241,1985 0 1985 by The American Society of Biological Chemists, Inc. Printed in U.S.A.

Correlation of Chloroplast and Bacterial Ribosomal Proteins by Cross-reactions of Antibodies Specific to Purified Escherichia coli Ribosomal Proteins*

(Received for publication, April 23,1984)

Marius BartschS From the Max-Planck-Znstitut fkr Molekulare Genetik, Abteilung Wittmann, Ihnestrasse 63-73, 0-1000 Berlin 33, Federal Republic of Germany

Immunological homology between chloroplast ribo- somal proteins (r-proteins) from a higher plant (Spi- nacia) and bacterial r-proteins was examined using antibodies prepared against 35 purified Escherichia coli r-proteins. Cross-reactions were determined on cellulose acetate gels and on nitrocellulose paper, after electrophoretic transfer of r-proteins from one- and two dimensional polyacrylamide gels, using peroxidase and fluorescein-conjugated second antibodies for de- tection (immunoblotting). The specificity of positive cross-reactions was confirmed by absorption experi- ments using purified E. coli r-proteins.

Antisera against five proteins of the small subunit and six proteins of the large subunit of E. coli ribosome (i.e. anti-S7, -S9, 4311, -512, and 4319; anti-L1, -L2, -L3, -L6, -L13, and -L17) gave cross-reactions. As an inference from this work, and a recent study on the synthesis of certain chloroplast r-proteins in iso- lated chloroplasts (Eneas-Filho, J., Hartley, M. R., and Mache, R. (1981) Mol. Gen. Genet. 184,484-488), we suggest that chloroplast r-proteins S7 and L2 are en- coded in the organelle DNA.

Higher plant chloroplast ribosomes contain between 50 and 60 proteins, as do eubacterial ribosomes. The system of chlo- roplast r-proteins’ is interesting because of its evolution; chloroplast r-protein genes are distributed on both the orga- nelle and nuclear DNA (1) and, therefore, the synthesis and assembly of chloroplast ribosomes requires a process of inter- genomic regulation not encountered in bacteria.

The primary structure of one chloroplast r-protein (2) and the deduced amino acid sequences of four others are currently available (3-5). Each of these five amino acid sequences can be correlated with a corresponding Escherichia coli r-protein, namely proteins L12, S4, S7, S12, and S19 (2-5). Identities in the corresponding sequences range from 40 to 68% of the total number of residues and, therefore, these higher plant

* This work was performed in the laboratories of Dr. A. R. Subra- manian and Professor G. Stoffler. Please address all correspondence on this paper to Dr. A. R. Subramanian, Max-Planck-Institut fiir Molekulare Genetik, Abteilung Wittmann, Ihnestrasse 63-73, D-1000 Berlin 33, Federal Republic of Germany. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ Present address: Kinderklinik und -Poliklinik (WE16), Freie Universitat Berlin, Heubnerweg 6, D-1000 Berlin 19.

The abbreviations used are: r-protein, ribosomal protein; BSA, bovine serum albumin; SDS, sodium dodecyl sulfate.

and bacterial r-proteins are apparent homologues of each other.

Early immunological studies using antisera specific to total chloroplast and E. coli r-proteins and the immunoprecipitin technique could not discover homologies between r-proteins from higher plants and eubacteria (6). I have reinvestigated this problem using antibodies raised against purified E. coli r-proteins and the immunoblotting technique (7) which is very sensitive and allows the simultaneous identification of the positions of the antigens on two-dimensional acrylamide gels. The results of this study are presented in this paper.

EXPERIMENTAL PROCEDURES~

RESULTS

General Strategy-As a first screening procedure, r-proteins from 30 and 50 S subunits of E. coli and chloroplast ribosomes were separated on one-dimensional polyacrylamide gels and were transferred onto nitrocellulose sheets which were cut into 5-mm wide strips and incubated with one of the antisera as described under “Experimental Procedures.’’

E. coli r-proteins always gave a strong immune reaction. Chloroplast r-proteins caused weaker reactions. A cross-re- action with a noncognate subunit protein was never observed. In addition to the band due to the main reaction, several fainter bands were frequently observed with both homologous and heterologous r-proteins. These were either due to known contaminating antibodies (e.g. S14 in anti-Sl9, see Fig. 3) or due to minor contaminants and nonspecific reactions. Two examples of the observed data from the first screening are given in the Miniprint (Figs. 1 and 2).

Cross-reactivities of the sera agianst protein L1, L2, and L13 were also determined from blots of cellulose acetate gels. Cellulose acetate gel electrophoresis has no molecular sieving effect, and, therefore, it provides more accurate information about the net charge of the cross-reacting protein. An example is shown in Fig. 2.

The final identifications of the cross-reacting chloroplast r-proteins were made on nitrocellulose blots of two-dimen- sional polyacrylamide gels. Definite proof of the specificity of the cross-reaction was obtained by absorption experiments

* Portions of this paper (including “Experimental Procedures,” part of “Results,” Figs. 1 and 2, Table 1, and “Acknowledgments”) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document 84M-1213, cite the authors, and include a check or money order for $2.40 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

237

238 Immurwcorrelatwn of Chloroplast and E. coli Ribosomal Proteins

1

E.co1i 30s

0 0 (.....I:. Q [,j.. ..:

0 '"~"

.....

0

.... .... 0 2 7

s9

€? &2., O Q

. .

0 s 19

Chloroplast 30s

J ?

a n t i 4 9

. 4 J -6

anti-SI1

anti - SI9

anti -SI2

FIG. 3. Schematic drawing of the two-dimensional gel patterns of E. coli and chloroplast 30 S ribosomal proteins and the immune reactions obtained with antisera against E. coli 57, S9, Sll, 512, and S19. Immune reactions were obtained with peroxidase-conjugated second antibodies, except for anti419 for which fluorescein-conjugated antibodies were used. The left picture of each pair is the homologous reaction with E. coli r-protein, and the right picture is the heterologous reaction with cross-reacting chloroplast r-protein. The serum against ribosomal protein S19 contained antibodies against E. coli r-protein S14 as a known contamination. The cross-reaction is due to antibodies against protein S19, since an antiserum against pure protein S14 did not cross-react with chloroplast r-proteins (electrophoretic properties of the chloroplast protein that cross-reacted with anti419 also more resemble those of E. coli S19 than of E. coli S14).

Immunocorrelation of Chloroplast and E. coli Ribosomal Proteins

QSt Dim. -

E.co1i 50s

anti- L1

0 L12

L17 0 0

a 0

Chloroplast 50s

anti - L2

I

anti- L3 anti-L6

anti - L13 anti- L17 FIG. 4. Schematic drawing of the two-dimensional gel patterns of E. coli and chloroplast 50 S

ribosomal proteins and the immune reactions obtained with antisera against E. coli L1, L2. L3, L6, L13, and L17. Immune reactions were obtained with peroxidase-conjugated second antibodies. The left picture of each pair is the homologous reaction with E. coli r-protein, and the right picture is the heterologous reaction with cross-reacting chloroplast r-protein.

239

ZndDDim.

\

(details given under “Experimental Procedures”) in which all experiments in which positive results were obtained. Anti- preincubation of the antiserum with the cognate antigen sera against E. coli proteins L1, L2, L3, L6, L13, and L17 and (purified E. coli r-protein) eliminated the cross-reaction. S7, S9, S11, S12, and S19 cross-reacted with chloroplast r-

Summary of the Results-Table I (Miniprint) summarizes proteins. Figs. 3 and 4 show the immunoblots of two-dimen-

240 ~ m m u ~ c o r r e ~ t ~ n of ~ h ~ r o p ~ TABLE I1

Molecular weight and charge c ~ r a c t e r ~ t i c s of cross-reacting E. coli and chloroplast r-proteins

E. coli Chloroplast us. E. coli

M. X Charge M. Charge Protein

S7B 17.1 vb“ Similar >Basic s 9 14.6 vb s11

Similar <Basic 13.7 vb

S l 2 Similar Similar

s19 13.6 wb Similar Similar 10.3

L1 wb Similar Similar

24.6 b L2

-30,000 <Basic 29.4 wb Similar Similar

L3 22.3 b Similar Similar L6 18.8 b Similar Similar L12* 12.2 a L13

Similar Similar 16.0

L17 vb -20,000 Similar

14.4 vb Similar >Basic ‘b, basic; vh, very basic; vvb, very very basic; a, acidic (designation

based on the net positive groups from sequence data and electropho- retic migration). ’ Cross-reaction detected on Ouchterlony plates (see Ref. 2).

sionally separated chloroplast and E. coli r-proteins and the observed homologous and heterologous antigen-antibody re- actions. Schematic pictures of the two-dimensional gel pat- terns of 30 and 50 S subunit r-proteins from E. coli (14, 19) and Spinacia chloroplast (20)3 and the positions of the cross- reacting proteins in these patterns are given (Figs. 3 and 4).

Altogether, 11 of the 35 tested antisera gave cross-reactions with chloroplast r-proteins. Negative or inconclusive results were obtained with sera against E. coli r-proteins S1, S3, S4, S5, S6, S8, S10, S14, S16, S17, S20/L26, L4, L5, L7/L12, L9, L10, L11, L15, L18, L19, L22, L24, L25, and L29. In the case of S4 and L12, four anti-S4 and three anti-112 sera and a serum against the C-terminal fragment residues 74-120 of L12 were tested, but specific cross-reactions were not obtained (see “Discussion”).

DISCUSSION

The experimental results presented here show that it is possible to detect and identify homologous r-proteins from very distantly related organisms such as E. coli and higher plants with immunological methods. As expected for homol- ogous ribosomal proteins, the cross-reactions occurred only between proteins of the same ribosomal subunit. Most of the cross-reacting chloroplast and E. coli r-protein pairs have similar electrophoretic properties, which reflect similarities in their net charge and molecular weight. Estimated molecular weights and charge properties are listed in Table 11.

Additional cross-reacting proteins can probably be found if a battery of different antibody preparations against E. coli r- proteins were tested. For example, in immunoblotting exper- iments three E. coli r-proteins, Biz. S5, S11, and L24, reacted rather strongly with an t ibo~es directed against spinach chlo- roplast ribosomes (20). The proof that the cross-reactions obtained were specific was, however, not achieved in these cases.

I could not show a cross-reaction for chloroplast L12 by immunoblotting, even though E. coli L12 and spinach chlo- roplast L12 share several long peptide sequences, and one of the antisera used precipitates chloroplast L12 in the less sensitive immuno-double-diffusion procedure (2). The latter method requires at least two, if not three, immunological epitopes whereas one epitope suffices for immunoblotting. A

M. Bartsch and A. R. Subramanian, unpublished results.

at and E. coli R ~ s o ~ u ~ protei^

probable explanation is that in this particular case the im- munological epitopes are derived from the tertiary or quater- nary structure of L12 (21) which is not maintained on the nitrocellulose paper in the presence of detergents. A cross- reaction was also not detected for spinach chloroplast S4 although the deduced amino acid sequence of the previously isolated putative gene for maize chloroplast S4 has 39% homology to E. coli S4 (3). S7, S12, and S19 are putative chloroplast coded r-proteins (4, 5) which have cross-reacted with E. coli r-protein antisera (Fig. 3). Sequences of the predicted proteins S7 and S12 (Euglena chloroplast) and S19 (tobacco) show homologies 38,68, and 55%, respectively, with the c o r ~ s p o n ~ n g E. coli r-proteins.

Eneas-Filho et ai. (22) have determined by incubation of intact pea chloroplasts with radioactive amino acids that certain r-proteins are synthesized in situ. Because of the characteristic mobility, the spinach chloroplast r-proteins that cross-react with E. coli S7 and L2 are apparently identical to the pea chloroplast r-proteins designated S9 and L4 by these authors (22). This means that r-proteins homologous to E. coli S7 and L12 are apparently encoded by the chloroplast DNA. Recently other workers (23, 24) have independently concluded that chloroplast ribosomal protein homologous to E. coli L2 must be synthesized within the chloroplast.

REFE~NCES 1. Bogorad, L., Davidson, J. N., and Hanson, M. R. (1977) in Nucleic

Acids and Protein Synthesis in Plants (Bogorad, L., and Weil, J. H., e&) pp. 135-154, Plenum Publishing Corp., New York

2. Bartsch, M., Kimura, M., and Subramanian, A. R. (1982) Proc. Nati. Acad. Sei. U. S. A. 79,6871-6875

3. Subramanian, A. R., Steinmetz, A., and Bogorad, L. (1983) Nu- cleic Acids Res. l l , 5277-5286

4. Sugita, M., and Sugiura, M. (1983) Nucleic Acids Res. 11,1913- 1918

5. Montandon. P.-E.. and Stutz. E. (1984) Nucleic Acids Res. 12. 2851-2859

Bwl. Chem. 249.3347-3355

. . .

6. Gualerzi, C., Janda, H. G., Passow, H., and Stoffler, G. (1974) J.

7. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad.

8. S~anarayana, T., and Subramanian, A. R. (1983) B ~ ~ ~ ~ ~ r y

9. Spitnik-Elson, P. (1965) B ~ h e m . Biophys. Res. Commun. 18,

10. Hardy, S. J. S., Kurland, C. G., Voynow, P., and Mora, G. (1969)

11. Stiiffler, G. (1967) Mol. Gen. Genet. 100,374-377 12. Laemmli, U. K. (1970) Nature (Lond.) 227 ,680485 13. Mets, L. J., and Bogorad, L. (1974) Anal. Biochem. 57,200-210 14. Subramanian, A. R. (1974) Eur. J. Bwchem. 45,541-546 15. Towbin, H., RamjouB, H.-P., Kuster, H., Liverani, D., and Gor-

16. Howe, J. G., and Hershey, J. W. B. (1981) J. Bwl. Chem. 256,

17, Stijffler, G., and Wittmann, H. G. (1971) Proc. Nutl. Acad. Sci.

18. Stiiffler, G., and Wittmann, H. G. (1971) J. Mol. Biol. 62,407-

19. Kyriakopoulos, A., and Subramanian, A. R. (1977) Biochim.

20. Bartsch, M. (1984) Ph.D. thesis, Free University of Berlin. Berlin 21, Leiionmarck. M.. Lilias. A.. and Subramanian. A. R. (1984)

Sci. U. S. A. 76,4350-4354

22,2715-2719

557-562

Biochemistry 8,2897-2905

don, J. (1982) J. Bwl. Chem. 257,12709-12715

12836-12839

U. S. A. 68,2283-2287

409

Biophys. Acta 474,308-311

&ochem. Znt. 8,69176 ‘

. .

22. Eneas-Filho. J.. Hartlev. M. R., and Mache. R. (1981) Mol. Gen. Genet. 184,484-4“.

(1984) Mol. Gen. Genet. 193,129-134

J. E. (1983) J. Cell BWL 9 6 , 1451-1463

. . .

23. Dorne, A. M., Eneas-Filho, J., Heizmann, P., and Mache, R.

24. Schmidt, R. J., Richardson, C. B., Gillham, N. W., and Boynton,

25. Geyl, D., Bock, A., and Isono, K. (1981) Mol. Gen. Genet. 1 8 1 , 309-312

Immunocorrelution of Chloroplast and E. coli Ribosomal Proteins 241

Correlatlon Of Chloroplast and Bacterlal Ribosomal Protelns by Cr08~-Rea~tlons SuPP1eme"t to

of Antlbodle~ Speclflc to Purlfled Escherlchla coll Rlbosomal PrOtelns

m r l ~ s Bartsch by

EXPERIMENTAL. PROCEDURES

Lrlfuqatlon In a llnear qridlent wlth 10 - 30% sucrose I b I b nW poiasalum phos- phate buffer. PI1 7.5, 1 nW IqC12, 100 KC1 18eck:nan Type 15 Rotor, 23000 rpm. 17 hrs, 4-C). The aubunlte yere concentrated by YltraCEnLrlfUqatlon IBeckmaan 45 TI Rotor, 36000 rpm, 17 hrs, 4'Cl after UqCl concentratlon had been raised to 10 nW. The subunits were dISIlOlved and .Corea In 10 nW Tlls-HC1, pH 7.4? 50 nW KC1, 10 nW IlqC12, I O 8 glycerol, 1 IM dlthlothreitol. E. col1 and chloroplast r-proteln were extracted from rlbos~mes or subunits a t 2 n Liclil n urea a8 descrlbed by Spltnlk-Elson 191, or at 671 acetic aCId / 0.03 M ns acetate as descrlbed bv Hardv et al. 1101, wlth the same results. The pror6lns *I(. dlalyzed aqalnit 58 bcetic acid. lyophlllred, and dlasolved

Celluloae acetate gel electrophoresis was perfo-d as descrlbed by StOfflsr In the approprlate sample buffer for electrophoresis (see below).

In qradlenr qels wlth I2 - 209 acrylamlde 112). no-dlmenolonal polyacrylamide 1 1 1 1 . One-dImens10nal SDS polyacrylamide gel eleCtrophoreslS Was carried Out

gel electrophoresls was done accordmq to the nets and Boqorad procedure 1131 * e descrlbed preYIously ( 1 0 . Bdelc fuchsin lwhlch Is transferable to nltro- sellulosel was used to lndlcate the buffer front durlnq both one and two-dlmn- slonal gel electrophoresls. Transfar Of rotelns to nltrocsllul0.e mdpr. FOlloWlnq electrophoresis. Poly- acrylamlde gzls -re aoaked I n transfer buffer 125 mtl h l s , 190 !M qlycine,

paper IschleLcher L SshU11 BA 831 was performed In a Hoefer TE 42 Transfer 0.1% SDS. 20% methanol1 for 30 nln. The transfsr Onto shesta of nltrocellul0.s

Chamber 16v/cm. 4 hrsl. Transfer Of proteins sfparated on celluloDe acetate gals was also accomplished In the same buffer 16V/cm, 1 hrl. Protelns whlch r r c separated by tyo-dlmenslonal polyacrylamide qel electrophoresls rere

method lwlth heparlne/toluldlne blue1 described by Towbin et al. (15). After stadafG?r transfer onto the nItrOCelluloSe paper by the reveralble stalning

tho prOtcIn spots were marked wlth a pencll. they were destained In 90% metha- nol/l\ acetic acld. I m n o l q 1 c a l reactions on nltroceI1~lo.e sh-ts 17.16). Follorlnq proteln transfer and marklnq, the nltrocelluloee sheets wero aaturated wlth 18 BSA In the IncubstIOn Buffer 170 mtl potas8lum phosphate, pH 7.4, 0.9t NaC1, 0.5% hl-

ed sera. dlluted 1:lO - 1:lOO In 1 % BSA/lncuhntlon Fwffrr for 17 hr- n < t r r ton x-100, 0.28 SDSI for 30 mln. The sheets yere then Incubated wlth the deslr-

washlnq wlth Incubation Buffer lflvs changes In 30 mlnl the nltrocelluloee sheets yere Incubated wlth the second antibody and washed as before. F1UoresCCln-cOnuqated antllmdles were used at a concentratlon of 0.5 - 1 rqhl

bodles, as obtained from SI-, were used at a dllutlon Of 1:200 In 18 BSA/ In 18 BSA/Incubatlon Buffer Ilncubatlon: IO m l n ) , Peroxidase-conjugated mtl-

sarco(l1ne IlncubarIon: 3 hre). Fluorescein-conjugated antlbodles were detected 10 mtl pJtassIum phosphate, pH 7.4, 0.9\ NaCl. 0.5t TrltOn X-100, 0 . 2 8 N-lauroyl-

under 330 nm UV-llqht: peroxldase-con]uqated aintlbodles were detected by a red dye produced by the oxldatlon Of 3-amlno-9-ethyl-carbarol. For thln reaction thc nltrocellulose blots were Incubated 'under watch- In a freshly prepared 601utIon Of 10 mtl potasslum acetate-acetlc acld, pH 5.0. 0 . 0 4 8 3-aminO-9-ethyl

where from 1 - 30 mlnl by washlnq wIth water. carbsrol l S l ~ - ? , l , and 0 . 0 3 % H202. The color reaction w a ~ stopped (after any-

~ ~~ .. ~-~~~ """ ." ..

Sources of antlbodles. The antlaera against lndlvldusl E. coli r-proteins were

Fluoreacein-conjuqated ant lbod les were prepared by lncubatlnq antl-rabblt and raised In sheep or rabbits as deocrilmd by StOffler and-WlGnii 117,18).

cyanate lf4arckl In 0.2 I sodium borate buffer, pH 8 . 7 , for 1 hr at 20-C. The antl-sheep IqG at a ConCentratIon Of 2 mq/ml wlth 1 mtl fluoreaceln Isothlo-

reaction was stopped by addlnq excess glycine, and the conjugated IqC was purl- fled by qel-filtration on a Sephadex 6-25 c o l m (Pha-cla). 2 - 5 moles Of fluorerceln lsothlocyanate per mole of I S were coupled a8 determined by

ed antlbodles were purchased from slqma. spectrophotometry at 4 9 5 m Ifluorcscolnl and 280 nm (IqGl. F-eroxldase conjugat-

Absorption erperlnsnts. Indlvldual purlfled E. coli r-proteins 10.3 mg protelnl ml serum) were added to the respectlve sera b preclpltatf specific aintibdles. After lncubatlon 0 2 37.C for 2 hrs and 4'c for 6 hrs the sera were then ccntrl- fuqed and the supernatants taken for control e ~ p e r l w n t s .

RESULTS CrOss-reactlOn~ Of L1 and L2. In Figure 1 the results obtained for proteln L1 arc sumnrarlred. Flqure 1A left, shows blots of One-diPenslonal SDS-polyaoryl- dnldc qels wlth separated E. coll and chloroplast r-protelns. These blots were Incubated wlth 0. serum a q a l n s E coli ribosomal phteln L1 (antl-L1I 01 a second axperlmnt , rlth anCl-Ll -1Ch E. coll Ll had been addrd fa&::- tlm experiment). Wlth E. coll r-protelns, S n t i T i gave on=-&&-bs&-&dr several weak bands. The-slacTlii the high molecular welqht reqlan doe. mot cor- respond to a 505 r-Protein. In the absorptlon experlment the band which cor- responds to protein L1 Is absent, while a l l Other bands re-in unchanged. In ChlorOplast r-proteins, antl-L1 gave 0. band which has a sllghtly hlqher nr than E. Coll Llj a q a h Bome addltlonal falnter bands were seen The reaction

wlth L1 was Used. The molecular welqhts of the Imnoreactlnq protelns ( 3 0 K IS S % c I ~ b e C a u s e the major band 1s absent when an antl-Ll sLrum prOlnCube.ted

chloroplast: 24.6 K E. colll were esrlmated from a one-dlmenslanal SDS p l y - arylarnlde gel, In w h i c h T c o l l and chloroplast protelns yere separated In ad- lacent blots IFlq. lA r1qht)IBlot wtterns In cellulose a r e t n t ~ n-1. .knurl

baalc r-proteins, the chloroplast proteln b e l n q allqhtly less basic (data not that 5. cOlI L1 and the cross-reacting chloroplast protein are teci-&iki;"

was made on a blot of two-dImePnslonslly separated r-protelns and Is-shown In ahom): Tho flnal Idontlflcatlon Of the CroBs-reacClnq chloroplast r-proteln

F l q . IB. Slmllar set of experiments almlnq at the Identlficatlon Of the Chloroplast I‘-protein imUnOloglCally slmilar to E. Foil L2 Is shown In Fig. 2. The left Fide Of Flqure ZA shows the purlty, Cr0ss-rea~tIons~ and the speclflty of the ant1-L2 serum. The rlqht slde show separacmns of E . a and chloroplast r-proteins In adjacent slots In a gel for the molecular rlqht conparleon. E. coll L2 and the CI088-reaCtlnq chloroplast r-proteln have very slmllar mole-

molecular valqht aqqreqates, whlch are detected by the second antlbody.1 & l ~ l q h t s l i 3 O K I . I I n thls gel blot, proteh L2 has formed several hlgher

~ ~ ~ ~ ~ ~ . . ."""

The lieparatlon of Chloroplast and E. coll r-proteins on cellulose acetate gels

41 both belong to the very basic. rlbOboma1 r-oroteln m o m IFlo. 211. Thn showed that peoteln L2 and the cross-reactinq chloroplast proteln 11ames 3 and

A

57 s9 511 s12 S19 L1 L2 L3 L6

Ll 7 L13

FITC, F1uore.celn conjugated second antibody. Perox, peroxldase conluqated second antlbody. Co-elect., co-elactrophoresls experiment.

-S

I thank Prof. H.G. wlttmann for his helpful dlscusslons and support. I am much lndebted LO R. Hasenbat& and also to K.-H. Rak for expert technical a861stanCe. I am most grateful to Dr. A.R. Subramanlan, In whose laboratory rhls york has

slons throughout the course Of rhle A r k .

A n'x- L2 E

crbss-reic'ted ulth antl-L1.

was separated by one-dlmenslonal DolvacrvlamIde oel elecrronhorrsls.