Cloning and expression of a nitroaryl reductase gene from <i>Streptomyces aminophilus </i>strain MCMB411 in <i>E. coli </i>JM109 and <i>Streptomyces lividans </i>TK64

IIndian Journal of Experimental Biology Vol. 37. August 1999. pp. 787-792

Cloning and expression of a nitroaryl reductase gene from Streptomyces aminophilus strain MCMB411 in E. coli 1M 109

and Streptomyces lividans TK64

Sabita Dey

Division of Microbial Sciences, Agharkar Research Institute, G G Agarkar Road,Pune 41 I 004, India

Fax: 91-020-351542; email dey.dipan@axcess.net.in

Received 2 December 1998; revised 28 April 1999

A partial genomic library was prepared in Ecoli JM I 09 using pBR322 as vector and 2.4kb Sau 3A I chromosomal fragment, encoding a nitroaryl reductase (nbr A) gene, from Streptomyces aminophilus strain MCMB 411. From the library.

2.4 kb fragment was recloned in Ecoli JMI09 and Slividans TK64 using pUCI8 and pIJ702 as vectors respectively. The recombinant plasmids pSD 103 and pSD 105 expressed the reductase gene and exported the enzyme in peri plasmic space of Ecoli and in cytoplasm of Slividans TK64. The proteins expressed by Ecoli and Slividans had the same molecular mass (70kD) as that expressed by parent strain, which suggested .that the enzyme was processed similarly by all strains. Activities of the enzymes cloned in Ecoli JMI09 and Slividans TK64 containing recombinant plasmids pSDI03 and pSDI05 respectively were optimum at 30°C and pH 9 and requirement of cofactors was same as that of the parent strain.

Streptomycetes are economically important organisms not only for production of useful secondary metabolites but also for their actIvIty as widespread scavengers in soil and contribution to the fertility and geochemical stability of the biosphere.

As most of the man made organic molecules contain aromatic ring so biodegradation of this group of chemicals whi~h follow /3 keto adipate pathway', was well studied. The enzymes required for aromatic metabolism mainly belong to oxido reductases²,

rubredoxins and hydroxylases³. Cloning and expression of Streptomyces gene have been mostly carried out in Streptomyces⁴,5. However a few examples of expression of cloned Streptomyces genes in E coli have been reported⁶,7 but cloning of biodegradative gene from Streptomyces to Ecoli is rare. Streptomyces aminophilus strain MCMB41I , isolated from polluted soil has the capability of degradation of 92% of m-dinitrobenzene from the wastes⁸ and it was also observed that the nitrogroup of m-dinitrobenzene changed to amino group before ring c1eavage⁹. When compared with a type strain, without biodegradative activity, it was revealed that nitro aryl redu~tase, the key enzyme required for alteration of aromatic molecule, was absent in the type strain'o. For waste treatement it is better to have a microbial consortia capable of producing the key enzyme. Hence, the isolation of nitroaryl reductase

(nbrA) gene from Saminophilus strain SD2 as well as cloning and expression of this gene in other microorganisms become essential. ]n this paper we report the isolation of a nitroaryl reductase (nbr A) gene from Streptomyces aminophilus strain MCMB411 and its cloning and expression in Ecoli and Slividans.

Materials and Methods

Bacterial strains and plasmids--Characteristics of S lividans TK64 and E. coli JM 109 are shown in Table I. Saminophilus strain SD2 was isolated from polluted soil and identified as per the method of Locci" and maintained as spore suspension at -70°C in presence of glycerol (25% v/v). pBR 322 and pUC 18 were used for cloning in E.coli; and p]J702 for cloning in Slividans TK 64. pSD 101, pSD 103 and pSD 105 were constructed by ligating a fragment (2.4 kb) of Sau3A digested chromosomal DNA with BamHI restricted pBR 322, pUC 18 and PST I digested plJ 702 respectively. pSD 105, plJ 702 and plJ 922 were used for cloning in Slividans TK 64.

Culture media and antibiotics-Ecoli JM 109 was grown at 37°C in nutrient broth and nutrient agar supplemented without and with ampicillin (50 Ilg mr'). S aminophilus SD2 and S lividans TK 64 were grown in NCCP at 30°C. To study nbr A activity these media were supplemented with m-dinitro-

788 INDIAN 1. EXP BIOL., AUGUST 1999

Table I-Description of Bacteria and plasmid used in the experiments

Bacteria! Genotypelrelevant References Plasmid

S. lividans Str-6, SLP2', SLP3-- Hopwood et at. ¹² TK64

E. coli ^recAl, endoAI, gyrA96 Miller^l)

JMI09 thi", hsd R 17, sup E44 reI A·' (Iacpro) f^ltraD36, pro AB lac 1-, LacZM 15

pBR 322 Col E I, relaxed, Amp' tet' Sutcliffe^l4

pUCI8. Amp',lacZk Vierra &

Messing^l5 plJ I 02 Ts', Mel+ Katz et al.¹⁶ pSD 101 pBR 322/2.4 kbAmp', tct' Dey unpublished

DNB+hybrid

pSD 103 pUC 18/2.4 kb Amp', Dey unpublished DNB+ hybrid

pSD 105 p1J702/2.4 kbTs', Mcl+, Dey unpublished DNB+ hybrid

benLene (200J.1g m)"') and thiostrepton (SOflg mr') as required.

Isolation of total DNA and plasmid-Total DNA from S aminophilus SD2 was prepared as per Chaterl7 with slight modification. Mycelia harvested from 24 hr old .liquid culture of S aininophilus.SD2 were resuspended in Bimboim buffer (2mM tris.cl pH 7.S, 10mM EDTA with 0.9% wiv glucose, 2mg lysozyme mr'). After addition of SDS (final conc. 1%

wiv), 1ml pronase solution (IOmg mrl) predigested at 37°C for 2 hr; and 4ml TE buffer containing 20J.1g mr' of boiled RNase were added to harvested mycelia and incubated for I hr at 37°C with repeated trituration. After addition of Sml TE, the same was incubated at 37°C for 2 hr. Then S MNaCl (to a final concentration I M) was added, followed by precipitation with chilled 100% ethanol and DNA was spooled with a glass rod, washed with 70 and 100% ethanol, dried and resuspended in Sml TE and purified by CsCIlEtBr density gradient using Beckman LM 60 Ultracentrifuge. P~asmid · isolation was carried out as per Bibb et af. 18 and small scale preparation of plasmid DNA was performed .by alkaline hydrolysis methodl9

.

Cloning and expression of nitroaryi reductase gene in E. coli-All restriction endonuclease diges- tions and ligations were carried out as described by Maniatis et al.²⁰ and agarose gels (0.8% wiv) were used to separate DNA bands. Chromosomal DNA of Saminophilus was digested to completion with Sau 3A and size (ractionated in 2 to 10kb by sucrose

gradient, reprecipitated and ligated to pBR 322 restricted with Bam HI. Partial genomic library was used to transform the competent Ecoli JM 109. All Ampr, tet' colonies obtained were replicated on dinitrobenzene master plates containing ampicillin and tetracycIin using sterile velvet pad.

Restriction endonuclease mapping and sub- cloning-Restriction endonucleases~ EcoRl, Hind III, BamH I, BgI II, BcI I, Sau 3A, Xho I, Sma I, Sal I were used for cloning and mapping. HindIII digested (lambda DNA) was used as size marker. The maps of the plasm ids have been shown in Fig. I. In order to locate nitroaryl reductase (nbr A) gene in the insert pSD 101 (pBR 322+ insert) was digested overnight with BamHI and HindIIl, extracted with phenol: CHCh and then religated (pSD 102). For the same purpose -2.4 kb EcoRIIBamHI fragment was extracted from agarose gel using prep-a-gene (Biorad) as per Girvitz et af. 2' and cloned into pUC 18 restricted with same enzymes (pSD 103). Before ligation the sticky ends were made compatible using CIP22. For cloning and expression of nbr A gene in Slividans TK64, a hybrid plasmid (pSD 105) containing the same 2.4 kb fragment and ligated to a high copy Streptomyces vector pIJ 702 restricted with pst I was cloned to S hvidans TK 64. Recombinant plasm ids (pSD 101 and pSD 103) were isolated from E. coli and pSD lOS was used to transform Slividc.Jns TK 64 using methods as described by Thompson et

af. ^2J. Preparation of S lividans protoplasts and their transformation were performed as described by Hopwood et al. ^24. The inserts from pSD 103 and pSD 105 were treated with Sma I and religated with pUC 18 and plJ 702. Transformation was carried out with the resultant plasmids pSD 104, pSD 106 and 107. Similarly, 2.4 kb fragment was ligated to plJ 922 and transformation experiment was repeated with S.

lividans TK 64.

Southern hybridization-Southern hybridization was carried out as per Southern^25. DNA (5J.1g/lane) from Saminophilus SO 2 digested with Sau 3A, Pst 1 and Eco RI and fractionated on 1 % agarose gel. DNA was transferred to Hybond-N-membrane and probed with nick translated plasmid DNA of clone (1 x 1 07 cprn/ J.1g). Hybridization was carried out at 30%, formam ide; 5X, SSC; 5X, Dendardt's solution; 0.1 %, SDS; 0.1 mg/ml, BSA and 50 mM phosphate buffer (PH 7) at 42° C. The blot was washed with (0.2X) SSC containing (0.1 %) SDS at 55° C for 40 min and autoradiograph of the blot was carried out.

DEY: CLONING & EXPRESSION OF NITROARYL REDUCTASE GENE 789

Preparation of enzyme, protein determination and SDS PAGE-Total intracellular enzymes, (concen- trated 80 fold), was prepared by disrupting cultures of E. coli pSD WI, S. aminophilus SD2 and S. lividans TK64 (pSD 105) by sonication and cell debris was removed by centrifugation (10000 rpm). The soluble protein was precipitated with ammonium sulphate (70% w/v). The precipitate was resuspended in phosphate buffer (PH 7.5), dialysed twice against the same buffer containing 25% v/v glycerol at room temperature overnight. Finally stored at -20°C after further purification upto homogeneity using adsorption chromatography and gel filtration. The proteins were assayed by Lowry:s method²⁶.

Approximately I OO~g of protein was analysed by SDS-PAGE for determination of molecular weighe⁷.

The proteins were denatured by mixing with 50~1 of loading buffer containing 5% v/v, mercaptoethanol;

amp r

(toR,

amp'

PSD 101

PSD 103

3.4% w/v, SDS; 15% v/v, glycerol; and 0.0 I % w/v, bromophenol blue in 10mM tris.c1 (PH 8.4) and boiling at 100°C for 3 min and after electrophoresed for 4 hr using 10% acrylamide gel, and was observed by staining with Coomassie Brilliant Blue R 250 followed by destaining.

Nitroaryl reductase asst:ry--N itroaryl reductase (nbr A) in sonicated and centrifuged supernatant

Table 2-Localization of nbr A expressed by E. coli 1M 109 (pSD 10 I). S. lividans (pSD 105) and S. aminophilus SD2

Fraction nbrA activit~ {%}

E. coli S. lividans S. amino- Alkaline phi/us phosphatase

Supernatant 10±2 5±1 3±2 12±3

Periplasml 75 ± 3 82 ± 3 87± 4 88± 2 cytoplasm

Membrane IS ± 4 12 ± 3 9±3 0±2

i"

<c,v ~$ nbr A <b ftt r

11PSD 101 I I 2.4 kb

,

PBR322

~ ~ ^~ .~ ~

<c,rJ ~: nbrA <b t.t'

IIP5D 102

I I Hkb I

PBR322

~ ~

"- ~ ~ ... ^-

~'$ ~

't f

2'4 kb •

PUC 18

~

-t"

t ' ~i;; ~$ '-,''l'<> <b. nbrA 4" /p

Bell

PS 0 lOS

D~~·-+I-~'~4'~+1-=2~.~4k~b~·~,~+~~m~.~1 IIIPSDIOS

PIJ 702

II

~ "- ~V:- "-

<:-1> ~ ^.. ~ ~.. ^....'

~

"q .

I ' I I 2-Okb I II PSD 106 PU 702

"- .:::-

~.. ,-'

"q

, I~I Pm.1 NPSDI07

2·0 k b

PIJ 702 'v" ~ nb' A

I 2.4kb

~ ~ ~~

q,1> q,o, t '

., I • IIPSDI08 PU 922

Fig. I- Restriction enzyme map ofpSD101 and plasmids obtained from subcloning experiments. [The arrows represent the orientation of nbr A gene in relation to tet' of pBR 322, LacZ promoter in pUC 18, mel promoter of plJ 702. The nbr activity was not there in pSD 102 where orientation was opposite.]

790 INDIAN 1. EXP BIOL., AUGUST 1999

1 2 3 4 5

Fig. 2- Shot gun Floning of nbr A gene from Saminophilus SD2 using pBR322, and plJ702. Gel electrophoresis was carried out using O.S% agarose gel and a current of I .SmA per cm of the gel.

[Lane I-size marker (digested with Hind III), Lane 2-pSDIOI, Lane3-pBR322; Lane 4-pI1702; Lane 5-pSD I 05]

obtained from recombinant E. coli 1M 109 and S livi- dans TK 64 were compared with that of parent strain S aminophilus SD2. Enzyme units were calculated by modified method of Braton and ~arshal28 using the reaction mixture containing NADH (6 flM), FAD (5flM) and 50mM of potassium phosphate buffer (PH 7.5), and mDNB (3mM) incubated for 30min at 30°e. The enzyme activity could be observed in the electrophoresed gel after incubating it in the reaction mixture, followed by addition of a detection mixture containing sodium sulfamale and sodium nitrite (I %) and a-naphthyldiamine dihydrochloride. A pink diazo band appeared in the gel. The activity of the enzyme from recombinant E.coli and Slividans as well as S aminophilus was analysed by thin layer chromatography using petroleum ether (40°-60°C):

anaesthatic ether (75:25) as solvent system and detected by diazotization. Rf values of the substrate and the product '" ere compared with those of authentic compounds. E. coli 1M 109 (pSD 10 I) Slividans TK64 (pSD 105) and S aminophilus, was grown in low salt medium (Neu and Heppel29

) to an OD₅₀₀ of 1.2 and fractionated by the method of Minton et al.^30. nbr A was assayed as per the method described above. The experiments were carried out in triplicate with a variation of 2-4%. Alkaline phos- phatase was used as a periplasmic/cytoplasmic marker and assayed as described by Toriani et a/.'I

2 3 I. 5 6 7

-13QOOO

•

•

Fig. 3- SDS-PAGE of nbrA from £. coli JMI09 (pSDI03) and Slividans (pSDI05). [Lane I-nbr activity of E coli JMI09 (pSDI03); Lane 2-nbr A activity of £. coli JMI09 (pUCIS)]: Lane 3-protein from £. coli JM 109 (pSD I 03); Lane 4-size marker (130kDa, 75 kDa, 50kDa & 30kDa); Lane 5-proteins from S lividans TK64 (pSD I 05); Lane 6-nbr activity of S lividans TK64 (pI1702): Lane 7-nbrA activity of S fividans TK64 (pSDI05)]

Results and Discussion

Transformation of E. coli 1M 109 by Sau 3 A library of S aminophilus SD2 in pBR 322 resulted more than 2x I 0³ Amp" tet^r colonies, only I % of which were showing nbr A positive reaction which were presumed to be recombinants. After replicating on LB plate supplemented with Amp (50flg mrl), tet (50flg mrl) and m-dinitrobenzene (200flg mrl) nbr A positive colonies were found to have a recombinant plasmid pSD 101 containing an insert of 2.4 kb.

Transformation with this hybrid plasmid produced all nbr A positive strains. Control experiment was carried out with E. coli transformed with pBR 322 only. The reductase activity was found to be located in the periplasmic fraction of E. coli ^{and in cytoplasm} 'Of Streptomyces (Table 2). nbr A gene of S aminuphilus SD2 was cloned a~d expressed in E.coli JM 109 and S lividans TK 64 and more than 75% of its activity was present in the periplasmic fraction in E. coli and in cytoplasm of S lividans.

This supports the previous report that Gram positive enzymes are periplasmic when expressed in E. COIP3. It is not clear at this stage whether other proteins, beside nbr are required for the periplasmic location in E. coli. Comparison of the activities of the fractions from S. lividans TK 64 (pSD 105) with those of S aminophilus SD2 revealed cytoplasmic presence of nbr in both the strains.

DEY: CLONING & EXPRESSION OF NITROAR YL REDUCTASE GENE 791

Mapping and sub-cloning-The maps of the plasm ids resulting sub-cloning has been shown in Fig. I. It was found that nbr A gene was carried by 2.4 kb BamH I1Hind III fragment of pSD 101 and started from the side of EcoR I as deduced from the lack of nbr A expression from E. coli harbouring pSD 102. Plasm ids pSD 103 and 104 isolated from E. coli were used to transform S Iividans TK 64 and Ampf (Ampicillin) colonies were picked up. The plasmid constructs were extremely unstable in S lividans.

They were rearranged and most probably lost the nbr A gene. Cloning in S lividans was successful when Hind rrVSph I fragment carrying nbr A gene was inserted into high copy number plasmid pU 702 cut with Sph I after making the s~tes compatible with mungbean (Phaseolus mungo) nuclease³² and Met+

Tsf colonies were selected. The recombinant plasmid was designated as pSD 105. Large nbr A activity zones were produced by S Iividans TK 24 (pSD 105) whereas the control TK 64 (pU 702) did not exhibit any activity. Hind IH/Sph I fragment (2.4 kb) was also subcloned into the low copy number plasmid 922 recombinant plasmid (pSD 108) and introduced in E. coli and S Iividans. E. coli colonies containing this construct were nbr A negative whereas S lividans containing the construct produced moderate nbr A activity. The insert was cut with Sma I and 2kb fragment was religated with pUC 18. The resultant pSD 104 exhibited less nbr A activity. The smallest insert still expressing nbr A activity was 2 kb as could be seen by repeated restriction, ligation and cloning of other constructs as pSD 104, pSD I 06 and pSD 107.

S lividans TK 64 containing these plasmids showed the same activity as with pSD 105, but E. coli with pSD 104 exhibited less activity which may be due to the fact that 2.4 kb insert may contain · the gene required for peri plasmic location of the protein. DNA seqeuncing of this insert and further sub-cloning may reveal the information that whether for secretion of the protein in the peri plasmic space, is also located in the coding sequence of nbr A gene or not. The expression of nbr A gene in E. coli was not observed when the insert was sub-cloned either in the opposite orientation to lacZ promotor or in a vector that lacks this promotor. This barrier to expression of Streptomyces genes in E. coli has been previously observed³⁴•36

. Similarly 2kb fragment was religated with pU 702 and transformed S lividans showed nbr A activity. S lividans TK64, transformed with pSD 106 and 107, containing 2kb insert, exhibited nbr A

activity, but in E. coli with pSD 104 containing same insert the activity was very less. It was also observed that though the lacZ promoter was necessary for expression of nbr A but IPTG induction was not there. This is against the observation of Hoshiko et af. 37 who reported [PTG induction of amylase gene.

Southern hybridization-Streptomyces origin of the protease gene was confirmed by probing a Sau 3A bulk digestion of Saminophilus SD2 total DNA with radioactively labelled pSD 104. A single 2.4 kb fragment of SD2 showed homology with the probes.

SDS-PAGE and charac;terisation of nitroaryl reductase (nbr A}--Concentrated and purified enzyme solution from S aminophilus SD2 and E. coli (pSD 103) and S lividans (pSD 105) were analysed in the same gel (Fig. 2). Coomassie staining of the protein bands in the gel revealed co-migration of nbr from E. coli and S lividans TK64 alongwith S aminophilus SD2 showing that the processing of the signal peptide by E. coli could differ either by a few amino acids or may be same. But this has to be studied in detail in amino acid analyzer. Though it is not a general case that E. coli can process the signal peptide in exactly the same manner as Gram positive parent strain³⁸ but in this experiment E. coli containing pSD 103 retained nearly 75% of nbr in the periplasmic fraction (Table 2).

Both the recombinant strains showed presence of a 70 kD protein which was same as parent enzyme and the activity showed by E. coli and S lividans proteins was also same as that of parent protein. The molecular mass of nbr produced by E. coli (pSD 103) and S lividans TK 64 (pSD 105) was nearly 70 kD which is same as that of parent strain. This similarity may result from the ability of E. coli and S lividans to process the signal peptide correctly as in the original Streptomyces host. This can be proved by sequencing the N-terminus. This suggests that both E. coli and S lividans can process nbr A protein similar to the form found in the parent strain. Control experiment with E. coli (pUC 18) and S lividam TK 64 (pU 702) did not show any activity (Fig 3). The nbr activity of the homogenous protein from E. coli (pSD 103), S lividans TK64 (pSD 105) was compared with that of the parent strain (Table 2) and it was observed that difference in activity was also negligible (6-16%). TLC analysis and Rf values of the products after nbr activity, were compared with that of the substrate and it was seemed to be different from m-DNB and similar to the authentic compound

792 INDIAN J. EXP BIOL., AUGUST 1999

m-aminophenol as observed previously9. Comparison of the other characters viz. optimum pH and temperature and requirement of cofactors revealed that characters were nearly same as those of the parent strain.

So present results strongly suggest that nbr A gene is transcribed in the direction of BamH I to pSt I in 2 kb fragment inserted into pIJ 702. It is not clear whether 2 kb fragment contains the promoters of nbr A gene or whether expression in S. lividans is caused by read through from promoter within the vector.

This enzyme may be used for construction of promotor-probe vectors since the coloured expression zones appeared on the gel and expression in E. coli makes easier isolation of this gene.

AcknowledgeOlent

This work was supported by grants from Department of Biotechnology, Ministry of Science and Technology, Government of India. The auth'or thanks Professor K Dharmalingam for his advice and Dr T Kieser for supplying S.lividans TK64.

References

Stanier R Y, Adelberg E A & Ingraham J L in General microbiology (Macmillan Publishers Ltd. Hongkong) 1985, 170.

2 Gibson D T, Science. 161 (1968~ 1093.

3 Sala Trepat J M & Evans W C, Eur J Biochem. 20 (1971) 400.

4 Bibb M J, Chater K F & Hopwood D A in Experimental manipulation of gene 5 exprefsion. edited by M Inouye, (Academic Press London) 1983; 54.

5 Vara J, Malpatrida F, Hopwood D A & Jimmez A, Gene. 33 (1985) 197.

G Hercomb J Thierbach G, Baumerg S & Parish J H, J Gen Microbiol, 133 (1987) 317.

7 Gil J A & Kieser H M & Hopwood D A, Gene. 38 (1985) 1- 8.

8 Dey S, Kanekar P & Godbole S H, Indian J Environ Health.

29 (1986a) 118.

9 Dey S & Godbole S H, Iridian. J Exp Bioi. 24 (1986b) 29.

10 Dey S & Godbole S H, in Comptemporary themes in biochemistry enzymo!ogical aspects of di-nitrobenezene degradation, edited by 0 L Kon, M C M Chung, P L H Hwang, S Leong. K H Loke, P Thiyagarajah & P T H Wong (Fed. Asian and Oceanic Biochemistry, Oxford Press, London) 1987,34.

II Locci R in Bergey's manual of systematic bacteriology. Vol 4, edited by S T Williams, M E Sharpe, and J G Holt (Williams & Wilkins, London) 1989, 2451.

12 Hopwood D A, Keiser T, Wright M J & Bibb M J, J Gen Microbiol, 129 (1983) 2257.

13 Miller H, Meth Enzymol. 152 (1987) 145.

14 Sutclife J G, Nucleic Acid Res. 5 (1978) 2721.

15 Vierra J & MeS1>ing J, Gene. 19 (1982) 258.

16 Katz F, Thompson C J & Hopwood D A, J Gen Microbiol.

129 (1983) 2703.

17 Chater K F, Hopwood D A, Kieser T & Thompson C J, Curr Topic Microbiol Immunol. 96 (1982) 69.

18 Bibb M J, Freeman R F & Hopwood D A, Mol Gen Genetics, 154 (1977) 155.

19 Birnboim H C & Dolly J, NucleiC; Acid Res. 7 (1979) 1513.

20 Girvitz S C, Bacchilli S & Rainbow J A & Graham L F, Anal Biochem. 106 (1980) 492. .

21 Sambrook J, Fritsch E F & Maniatis T, Molecular cloning- A laboratory manual (Cold Spring Harbor Laboratory Press) 1989, 1.53 & 9.38.

22 Cobianchi F & Wilson S H, Methods in enzymology. edited by S L Berger & A R Kimmel (Academic Press, NY) 1987, 152,94.

23 Thomson C J, Ward J M & Hopwood D A, J Bacterio/' lSI (1982) 668.

24 Hopwood D A, Bibb M J, Chater K F, Kieser T, Bruton. C J, Kieser H M, Lydiate D J, Smith C P & Ward J M in Genetic manipulation of streptomycetes- A laboratory manual (The John Innes Foundation, Norwich 1985, 12.

25 Southern E M, J Mol Bioi, 98 (1975) 503.

26 Lowry O.I;I, Rosenbrough N J, Fair A L & Randel R J, J Bioi Chem, 193 (1951) 265.

27 Laemmli U K, Nature, 227 (1970) 680.

28 Braton A C & Marshall E K, J Bioi Chem. 128 (1939) 538.

29 Neu H C & Heppel L A, J Bial Chem, 240 (1964) 3685. 30 Minton N P, Atkinson T & Sherwood R, J Bacteriol. 156

(1983) 1222.

31 Torriani A in Procedures in nucleic acid recearch. edited by G I Cantom & D R Davies (Harper & Row, NY) (1966) 224.

32 Gubler U & Hoffman B J, Gene, 25 (1983) 263.

33 Pugsley A P in Protein Synthesis and organelle biosynthesis, edited by Dass and Robbins (Academic Press, London)

1988,607.

34 Horinouchi S, Uozumi T & Beppll. T, Agric Bioi Chem, 44 (1980) 367.

35 Bibb M J & Cohen S N, Mol Gen Gener. 187 (1982) 265.

36 Bahri S M & Ward J M, J Gen Microbiol. 136 (1990) 811. 37 Hoshiko S, Maka.tse 0, Nojiri C, Katsllmata K, Satoh E &

Nagaoka K, J Bacteriol, 169 (1987) 1029.

38 Robbins P W, Trimble R, Wirth D F, Hering C, Maley F, Maley G, Das R, Gilsen B W, Royal S & Biemank I, J Bioi Chem.259 (1984) 7577.