Streptomyces coelicolor DNA homologous with acyltransferase domains of type I polyketide synthase gene complex

FEMS Microbiology Letters 157 (1997) 195^200

Streptomyces coelicolor DNA homologous with acyltransferase domains of type I polyketide synthase gene complex Katarzyna Kuczek *, Krzysztof Pawlik, Magdalena Kotowska, Marian Mordarski Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, Weigla 12, 53-114 Wroclaw, Poland Received 21 July 1997 ; revised 30 September 1997; accepted 13 October 1997

Abstract

An acyltransferase-homologous DNA fragment was amplified in a PCR reaction on a cosmid DNA template from the genomic DNA library of the soil bacterium Streptomyces coelicolor A3(2). The putative amino acid sequence of the fragment resembles acyl-CoA:ACP acyltransferase domains from several bacterial enzymatic complexes of polyketide synthase. There is a high similarity with acyltransferase domains from so-called type I polyketide synthases. Such synthases catalyze production of the aglycone portion of macrolides and polyethers that are important as antibiotics or immunosuppressants. The amplified fragment is considered to be a part of a larger gene complex. Keywords : Acyltransferase; Polyketide synthase; Streptomyces

1. Introduction

Streptomycetes, Gram-positive soil bacteria, are known to be producers of a variety of secondary metabolites. Among these, biologically active compounds of a polyketide structure, most of which are antibiotics, are of great concern [1]. Multifunctional protein complexes called polyketide synthases (PKS) contain enzyme activities catalyzing successive rounds of elongation and reduction of the polyketide product in a manner similar to fatty acid synthesis. Based on enzyme organization and the chemical nature of the polyketide formed, polyketide synthases have been classi¢ed into two distinct types. Type I and II enzymes catalyze formation of complex and * Corresponding author. Tel.: +48 (71) 732274; Fax: +48 (71) 679111; E-mail: [email protected]

aromatic polyketides, respectively. Polyketide biosynthetic genes are clustered and the respective clusters involve genes for the major activities L-ketoacyl synthase (KS), acyltransferase (AT) and acyl carrier protein (ACP) as well as genes for other activities [2]. In the present study we searched for a DNA sequence thought to be a part of a larger fragment comprising a new KS-homologous sequence of Streptomyces coelicolor A3(2) [3].

2. Materials and methods

2.1. Strains cosmid clones and plasmids. Escherichia coli SURE strain (e143 (McrA3 )v (mcr CB- hsdSMR-mrr)171 endA1 supE44 thi-1 gyrA96 relA1 lac recB recJ sbcC umuC: :Tn5 (Kanr ) uvrC

0378-1097 / 97 / $17.00 ß 1997 Federation of European Microbiological Societies. Published by Elsevier Science B.V. PII S 0 3 7 8 - 1 0 9 7 ( 9 7 ) 0 0 4 7 6 - X

FEMSLE 7904 20-11-97

K. Kuczek et al. / FEMS Microbiology Letters 157 (1997) 195^200

196

[FP

proAB

q

vM15

lacI Z

r

Tn10 (Tet )]) was a host for

cosmid clones 1A12, 1G7, 2C4 and 7G5 of

color

S. coeli-

genetic procedures were performed as described earlier [7].

A3(2) genomic DNA in SuperCos vector (Stra-

tagene) which were kindly provided by H.M. Kieser,

2.5. DNA sequencing and sequence analysis

John Innes Institute, Norwich, UK. PCR product was cloned in pBluescript SK

‡

The

phagemid vector

DNA sequence was determined using

the

E. coli XL1 Blue strain (recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [FP proAB lacIq ZvM15 Tn10 (Tetr )]). S. coelicolor A3(2) chromosomal DNA was iso3 lated from strain M110 (hisA1 uraA1 strA1 SCP1 SCP2*) and from strain YU105 (proA1 argA1 redE60 vact (ermE) vwhiE (hyg)) [4], made available

chain-termination method with the Sequenase v. 2.0

by D.A. Hopwood.

and PILEUP programs from GCG (software pack-

(Stratagene)

in

sequencing kit of USB and both double-stranded and single-stranded templates. Single-stranded phagemid DNA was obtained upon infection of host cells with VCSM13 helper phage (Stratagene) [7]. Comparisons of nucleotide and amino acid sequences with the databases were performed with the FASTA

age of the Wisconsin University Biotechnology Cen-

2.2. Isolation of DNA, PCR reaction, labeling and detection Puri¢ed cosmid DNA (Qiagen tip-100 columns,

ter).

3. Results and discussion

Diagen) was used as a template. The primers KS : 5P-TCCACCGGCTGCACCTC-3P

and

AT :

5P-

3.1. Selection of cosmid clones

AGCGAGTGGCCGACCAT-3P were used. The reaction was carried out for 31 ampli¢cation cycles (1

In our previous studies we used the Sc probe con-

L-ketoacyl synthase

min at 94³C, 1 min at 54³C, 1 min at 72³C followed

sidered to be a part of a putative

by 10 min extension at 72³C) with PrimeZyme ther-

gene of

mostable polymerase of Biometra.

Sc probe, which is homologous to the KS domains

For the probe labeling in a course of PCR reaction

Streptomyces coriofaciens

ISP5485 [3]. The

of 6-deoxyerythronolide B synthase (DEBS) of

Sac-

and detection of hybrids ECL-probe amp reagents of

charopolyspora erythraea,

Amersham were used. Chromosomal DNA was iso-

numerous

lated according to the procedure of Hopwood et al.

strains the probe hybridized to chromosomal DNA

[5].

of

2.3. Digestion of DNA and Southern hybridization

polyketide

S. coelicolor

hybridized to the DNA of

producers.

Among

A3(2) known to produce two aro-

matic polyketides : actinorhodin (act gene cluster) and a spore pigment (whiE gene cluster) [2]. A PCR product was obtained on

S. coelicolor

BamHI

A3(2) chromosomal DNA template with the same

cuts all of the inserts

primers which had been used to produce the Sc

DraI

probe (data not shown). The nucleotide sequence

hybridization

of the ampli¢ed fragment was homologous to KS

with the £uorescein-labeled PCR product was done

domains of DEBS and di¡erent from the known

at 73³C according to the ECL-probe amp Amersham

type II PKS genes from the strain (act,

Cosmid DNA was digested with restriction enzymes.

BamHI

DraI

other

and

into fragments of a wide length range while leaves

the

inserts

intact.

Southern

instruction.

Based on hybridization of the

whiE) S. coelicolor

[3]. A3(2)

genomic DNA library with the Sc probe, four cos-

2.4. Cloning in T-vector

mid clones were selected : 1A12, 1G7, 2C4 and 7G5 (Section 2). The cosmids are located in the

PCR reaction product was cloned into T-overhang vector prepared from pBluescript SK

‡

fragment of

S. coelicolor

AseI

B

A3(2) chromosome (D.A.

according to

Hopwood, unpublished). The cosmid clones were

E. coli XL1

screened further for the presence of putative polyke-

the method of Hadjeb and Berkowitz [6].

Blue cells were transformed by electroporation. All

tide synthase genes.

FEMSLE 7904 20-11-97

K. Kuczek et al. / FEMS Microbiology Letters 157 (1997) 195^200

Fig. 1. Alignment of nucleotide and amino acid sequences of active sites of

L-ketoacyl

197

synthase and acyltransferase domains from several

polyketide synthases. Derived sequences of KS and AT primers are shown. DEBS m5 : module 5 of 6-deoxyerythronolide B synthase

Saccharopolyspora erythraea (GenBank M63677) ; Act : actinorhodin synthase, S. coelicolor A3(2) (GenBank X63449) ; Fren : frenoliS. roseofulvus (GenBank L26338) ; PKS curacoi : polyketide synthase S. curacoi (GenBank X62518) ; Rap m12 : module 12 of rapamycin synthase S. hygroscopicus (GenBank X86780). from

cin synthase,

3.2. Ampli¢cation and analysis of a putative AT domain By comparison of sequences of several known

tained

on

the

templates

from

all

cosmid

clones

whereas a product of 550 bp was obtained on

coelicolor

S.

A3(2) M110 chromosomal DNA (Fig. 2).

Chromosomal

DNA

of

this

strain

contains

two

PKS gene clusters a set of primers was designed

known

based on the conservative sequences of

The appearance of a 550-bp product can be ex-

L-ketoacyl

synthase (primer KS) and acyltransferase (primer

PKS

gene

sets

for

aromatic

polyketides.

plained as a result of preferential binding of the pri-

act PKS S. coelicolor act and whiE PKS

AT) active sites (Fig. 1). Considering the common

mers to the respective sequences of the

organization of major PKS genes, the AT sequence

genes. When chromosomal DNA from

is likely to follow the KS domain as observed in any

A3(2) YU105, deleted for both

type of PKS gene organization [2]. The PCR product

genes, was used as a template, only the shorter frag-

was expected to span the region between the active

ment appeared.

sites of the two domains. In gene complexes of poly-

The ampli¢ed 450 bp fragment from 2C4 cosmid

ketide synthases studied so far the respective distan-

was cloned and three independent clones were se-

ces are around 1200 bp for type I PKS and 500 bp

quenced. The obtained sequence, excluding primers

for type II ([2] and references therein, [8^10]).

(413 bp), is referred to as B5 (accession number

A PCR product of approximately 450 bp was ob-

U88833). The putative amino acid sequence of B5

FEMSLE 7904 20-11-97

198

K. Kuczek et al. / FEMS Microbiology Letters 157 (1997) 195^200

[8], FK506 polyketide synthase of sp. [9] and 6-methylsalicylic acid synthase from [10] (Fig. 3) as well as 6-deoxyerythronolide B synthase from S. erythraea [11]. The homology is 31^44% identical amino acids. The amino acid motif GHSxG (x is any amino acid) ^ a consensus sequence found at the active site regions of all acyltransferase domains studied so far [12] ^ is present in B5. Another conserved motif, HAxHS, can be found 97 residues downstream of the active site. The second His in this motif could be involved in catalysis [9]. There is a correlation between amino acid sequence and substrate speci¢city of acyltransferase domains. Twenty amino acids located 11 residues upstream of the active site are predictive. Divergent motifs were identi¢ed in this region depending on whether the AT domain is speci¢c for malonyl- or methylmalonyl-CoA as a substrate [13]. The amino acid sequence of B5 resembles the consensus motif of AT domains from rapamycin synthase modules: 2, 5, hygroscopicus Streptomyces Penicillium patulum

Fig. 2. Products of the PCR reaction with KS and AT primers on the following templates: lane 1, S. coelicolor A3(2) M110 total DNA; 2, S. coelicolor YU105 total DNA; 3, B5 clone (PCR product cloned in pBSK‡ ); 4, cosmid 1A12; 5, cosmid 1G7; 6, cosmid 2C4; 7, cosmid 7G5; 8, control (no DNA); 9, control (B5 clone, no polymerase). Molecular mass markers: M1, HindIII digested V DNA; M2, AluI digested pBR322 DNA.

revealed homology with the AT domains of type I PKS such as rapamycin synthase from Streptomyces

Fig. 3. Alignment of putative amino acid sequence of B5 with di¡erent acyltransferase domains. The acyltransferase active site motif is in the shaded box. The second conserved motif is in the empty box. Upper case letters indicate amino acids identical with the B5 sequence. Bold letters indicate amino acids identical with the consensus motif (Cons A) of acetate incorporating domains [13]. R_2^R14, modules of rapamycin synthase from Streptomyces hygroscopicus (GenBank X86780); FK_7^FK10, modules of FK506 synthase, Streptomyces sp. (GenBank Y10438); MSAS,- 6-methylsalicylic acid synthase, Penicillium patulum (GenBank X55776).

FEMSLE 7904 20-11-97

K. Kuczek et al. / FEMS Microbiology Letters 157 (1997) 195^200 B5

was used, after

BamHI

digested

199

labeling, as a

chromosomal

2).

Southern

and

DraI

from cosmid clones digested with (Section

probe against

DNA

hybridization

and

at

DNA

BamHI

73³C

con-

¢rmed the presence of the putative acyltransferase domain on the fragments of around 3 kb in chromosomal DNA as well as 2C4 and 7G5 cosmid DNA and on the shorter one (around 1.5 kb) in 1G7 cosmid (Fig. 4). This shorter fragment probably lies at the end of the insert in the cosmid. In digested 1A12 DNA a hybridizing band of approximately 3.5 kb was

found

only

under

less

stringent

conditions

(68³C). Additional weaker bands were observed in other

cosmids

and

chromosomal

DNA

at

68³C

(not shown). This could indicate the presence of other AT-homologous sequences in the studied DNA. DNA bands from cosmid clones which gave a signal were selected for further cloning and the studies are under way.

3.3. Final remarks In conclusion, the DNA fragment from

Fig. 4. Southern blot hybridization with the B5 fragment used as a probe against the total

BamHI

with

cloned in the

(lane 1) ;

SmaI

S. coelicolor

pBSK

‡

A3(2) M110 DNA digested

containing the

site, digested with

EcoRI

B5 PCR product and

PstI

(lane 2) ;

DNA of the cosmids 1A12, 1G7, 2C4, 7G5 digested with and

BamHI

DraI

color

S. coeli-

A3(2) was ampli¢ed, whose potential transla-

tion product is highly homologous to acyltransferase domains, speci¢c for malonyl-CoA, found in PKS type I complexes. The presented results correspond

(lanes 3^6).

to our preliminary data on the presence in this strain of

at

least

two distinct

whiE

genes ([3] and studies

Although

they

do

di¡erent from

10

under

[9]

as

well

as

6-methylsalicylic

acid

synthase

way).

act

KS-homologous sequences

and

8, 9, 11, 12, 14 [8,13] FK506 synthase modules : 7, 8,

not

directly

show

[11,13] responsible for incorporation of acetate resi-

the existence of a gene cluster, we consider it possible

dues into the growing polyketide chain. This suggests

that new PKS genes are present in the

similar activity for the studied domain (Fig. 3).

A3(2) chromosome. This suggestion is supported by

S. coelicolor

The presence of an active site motif in the center

the fact that the studied cosmids and the `homologue

of the PCR product and homology exclusively with

of a macrolide PKS gene' communicated by Sher-

AT domains preclude binding of both primers at the

man and cited by Redenbach et al. [14] are located

expected sites. The nucleotide sequence of B5 within

in the same

the putative active site region turned out to be only

color

AseI

restriction fragment of the

S. coeli-

A3(2) chromosome.

partially complementary to primer AT (11 nucleotides out of 17) which would explain its non-speci¢c binding. We suspect a similar situation to be the case

Acknowledgments

for primer KS. More stringent PCR conditions had no in£uence on the length of the product. Neverthe-

We

gratefully

thank

D.A.

Hopwood

and

H.M.

less, we were interested in further investigation of B5

Kieser

and

clones. This work was supported by the Committee

surrounding

regions

since

the

B5

sequence

is

for

the

di¡erent from all known acyltransferase genes from

of

S. coelicolor

P206 009 07.

A3(2) ([2] and references therein).

Scienti¢c

FEMSLE 7904 20-11-97

gift

of

Research

strain

of

YU105

Poland

and

under

cosmid

Project

6

K. Kuczek et al. / FEMS Microbiology Letters 157 (1997) 195^200

200

References

The biosynthetic gene cluster for the immunosuppressant rapamycin. Proc. Natl. Acad. Sci. USA 92, 7839^7843.

[1] Okami, J. and Hotta, K. (1988) Search and discovery of new antibiotics. In : Actinomycetes in Biotechnology (Goodfellow, M., Williams, S.T. and Mordarski, M., Eds.), pp. 33^67. Academic Press, London. [2] Katz, L. and Donadio, S. (1993) Polyketide synthesis : prospects for hybrid antibiotics. Annu. Rev. Microbiol. 47, 875^ 912. [3] Kuczek, K., Mordarski, M. and Goodfellow, M. (1994) Distribution

of

Streptomyces

oxoacyl

synthase

homology

sequences

within

DNA. FEMS Microbiol. Lett. 118, 317^326.

[4] Yu, T.-W. and Hopwood, D.A. (1995) Ectopic expression of the

Streptomyces coelicolor whiE

genes for polyketide spore

pigment synthesis and their interaction with the

act

genes

for actinorhodin biosynthesis. Microbiology 141, 2779^2791. [5] 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. and Schrempf, H. (1985) Genetic Manipulation of

ces.

Streptomy-

A Laboratory Manual. The John Innes Foundation, Nor-

wich. [6] Hadjeb, N. and Berkowitz, G.A. (1996) Preparation of Toverhang vectors with high PCR product cloning e¤ciency. BioTechniques 20, 20^22. [7] Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning : A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. [8] Schwecke, T., Aparicio, J.F., Molnar, I., Konig, A., Khaw, L.E., Haydock, S.F., Oliynyk, M., Ca¡rey, P., Cortes, J., Les-

[9] Motamedi, H., Cai, S.-J., Sha¢ee, A. and Elliston, K.O. (1997) Structural organization of a multifunctional polyketide synthase involved in the biosynthesis of the macrolide immunosuppressant FK506. Eur. J. Biochem. 244, 74^80. [10] Beck, J., Ripka, S., Siegner, A., Schiltz, E., Schweizer, E. (1990) gene of

The

multifunctional

Penicillium patulum.

6-methylsalicylic

acid

synthase

Eur. J. Biochem. 192, 487^498.

[11] Bevitt, D.J., Cortes, J., Haydock, S.F. and Leadley, P.F. (1992) 6-Deoxyerythronolide-B synthase 2 from

spora erythraea.

Saccharopoly-

Cloning of the structural gene, sequence anal-

ysis and inferred domain structure of the multifunctional enzyme. Eur. J. Biochem. 204, 39^49. [12] Revill, W.P., Bibb, M.J. and Hopwood, D.A. (1995) Puri¢cation of a malonyltransferase from

Streptomyces coelicolor

A3(2) and analysis of its genetic determinant. J. Bacteriol. 177, 3946^3952. [13] Haydock,

S.F., Aparicio,

J.F., Molnar,

I., Schwecke, T.,

Khaw, L.E., Konig, A., Marsden, A.F.A., Galloway, I.S., Staunton, J. and Leadlay, P.F. (1995) Divergent sequence motifs correlated with the substrate speci¢city of (methyl)malonyl-CoA : acyl carrier protein transacylase domains in modular polyketide synthases. FEBS Lett. 374, 246^248. [14] Redenbach, M., Kieser, H.M., Denapaite, D., Eichner, A., Cullum, J., Kinashi, H. and Hopwood, D.A. (1996) A set of ordered cosmids and a detailed genetic and physical map for the 8 Mb

Streptomyces coelicolor

Microbiol. 21, 77^96.

ter, J.B., Bohm, G.A., Staunton, J. and Leadlay, P.F. (1995)

FEMSLE 7904 20-11-97

A3(2) chromosome. Mol.