The ram-Dependence of Streptomyces lividans ... ram-Dependence of Streptomyces lividans Differentiation is Bypassed ... The complex process of aerial hyphae and spore formation is

S. lividans Differentiation 565

The ram-Dependence of Streptomyces lividansDifferentiation is Bypassed by Copper

Received March 10, 2000; revised June 2, 2000; accepted June 5, 2000.*For correspondence. Email [email protected]; Tel. (+31) 715274278; Fax. (+31) 71 5274349.

J. Mol. Microbiol. Biotechnol. (2000) 2(4): 565-574.

© 2000 Horizon Scientific Press

JMMB Research Article

Bart J.F. Keijser1, Gilles P. van Wezel1, Gerard W.Canters1, Tobias Kieser2 and Erik Vijgenboom1*

1Leiden Institute of Chemistry, Gorlaeus Laboratories,Leiden University, P.O. box 9502, 2300 RA Leiden, TheNetherlands2John Innes Centre, Norwich Research Park, Colney,Norwich NR7 4UH, UK

Abstract

The onset of morphological differentiation inStreptomyces lividans is intrinsically delayed incomparison to Streptomyces coelicolor, but can beadvanced by adding extra copper to the medium.Copper-specific chelators block aerial hyphaeformation in both strains illustrating the crucial roleof copper in morphogenesis. The S. coelicolor ramcluster was isolated as a clone that complements thecopper-dependent differentiation of S. lividans. The S.lividans ram cluster was cloned and shown to be 99.6%identical to the S. coelicolor clone. The difference indevelopment between S. lividans and S. coelicolorcould neither be related to functional differencesbetween the two ram clusters nor to differences in thetranscription level. In both strains the low level oframAB transcription correlated with aerial myceliumformation and was coupled to the upstream ORF ramS.An increased ramAB expression level in S. lividans bythe introduction of an extra copy of ram stimulatedthe development. In S. lividans disruption of ramABRresulted in the inability to produce aerial hyphae.Conversely, the identical mutant of S. coelicolorretained its developmental capacities, indicating thepresence of a ram-independent developmental routethat is not present or not activated in S. lividans. Aerialhyphae and spore formation in the S. lividans ramABRmutant was restored when grown near wild-typestrains, suggesting that the ram gene products areinvolved in transport of a factor essential for normaldevelopment. In addition, an elevated copperconcentration in the medium also relieved thedevelopmental block of these mutants. These findingssuggest that higher copper concentrations render thisram-associated factor obsolete.

Introduction

Streptomycetes undergo a complex process ofmorphological differentiation that normally results insporulation. The complete lifecycle has three characteristicstages: 1) the formation of a branched vegetative mycelium,

2) the formation of aerial hyphae followed by 3) theproduction of spores. On solid medium the morphologicaldifferentiation of Streptomycetes can easily be followedsince upon aerial hyphae formation colonies get a whiteand fuzzy appearance. Once spores are being formed thecolonies turn grey, due to the biosynthesis of the WhiEspore pigment. The complex process of aerial hyphae andspore formation is the result of the interplay between themetabolic status of the cells, stress responses andextracellular signalling (Chater, 1998). Insight in thisprocess has been gained by studying two classes ofmutants that are disturbed in one or more facets of this lifecycle, the so-called bld mutants, which fail to erect aerialhyphae (Merrick, 1976; Champness, 1988; Willey et al.1993; Nodwell et al., 1996; Nodwell et al., 1999) and thewhi mutants that fail to form spores or the grey sporepigment (Chater, 1972; Hopwood et al. 1970; Ryding etal., 1999).

Other factors involved in morphogenesis have beenidentified by their ability to stimulate the differentiation ofwild-type strains or to restore the development of mutantstrains when present in multiple copies. This approach led,among others, to the identification of the amf gene clusterand amfC in S. griseus (Ueda et al., 1993; Kudo et al.,1995). Ma and Kendall (1994) showed that the S. coelicolorram cluster stimulated aerial hyphae development whenintroduced into S. lividans and that the disruption of ramBin S. lividans severely affects the development. In S.lividans, Kieser and Hopwood (1991) have reported astimulation of spore production as the result of increasingthe concentration of copper. Similar effects of copper havebeen reported for Streptomyces tendae, and Streptomycestanasiensis (Dionigi et al., 1996, Ueda et al., 1997).

Copper is an important element in biology amongothers as the co-factor in a variety of enzymes that playcrucial roles in metabolism. Copper also catalyses theformation of reactive oxygen species via the so-calledFenton chemistry (Chevion, 1988) which can cause severedamage in the cell. Thus, intracellular copper levels needto be tightly controlled. Copper tolerance systems that takecare of the efflux of excess copper have been describedfor numerous bacteria (Silver, 1996). Contrary, systemsinvolved in active uptake of copper and the mechanismsto regulate this process are not well understood. One ofthe few exceptions is the mechanism involved inmaintaining copper homeostasis in the Gram-positivebacterium Enterococcus hirae. Here intracellular copperlevels are maintained by two P-type ATPases, CopA andCopB which are involved in copper efflux or influx whenthe extracellular copper concentration is either too high ortoo low for optimal growth (Odermatt et al., 1993). Amechanism involving the copper chaperone CopZ and thecopper-inducible repressor, CopY, is responsible for thecopper-dependent regulation of the cop operon (Wunderli-Ye et al., 1999).

In this paper we present the results of complementationexperiments which led to the identification of the S.

• MALDI-TOF Mass Spectrometry in Microbiology

Edited by: M Kostrzewa, S Schubert (2016) www.caister.com/malditof

• Aspergillus and Penicillium in the Post-genomic Era

Edited by: RP Vries, IB Gelber, MR Andersen (2016) www.caister.com/aspergillus2

• The Bacteriocins: Current Knowledge and Future Prospects

Edited by: RL Dorit, SM Roy, MA Riley (2016) www.caister.com/bacteriocins

• Omics in Plant Disease Resistance

Edited by: V Bhadauria (2016) www.caister.com/opdr

• Acidophiles: Life in Extremely Acidic Environments

Edited by: R Quatrini, DB Johnson (2016) www.caister.com/acidophiles

• Climate Change and Microbial Ecology: Current Research and Future Trends

Edited by: J Marxsen (2016) www.caister.com/climate

• Biofilms in Bioremediation: Current Research and Emerging Technologies

Edited by: G Lear (2016) www.caister.com/biorem

• Microalgae: Current Research and Applications

Edited by: MN Tsaloglou (2016) www.caister.com/microalgae

• Gas Plasma Sterilization in Microbiology: Theory, Applications, Pitfalls and New Perspectives

Edited by: H Shintani, A Sakudo (2016) www.caister.com/gasplasma

• Virus Evolution: Current Research and Future Directions

Edited by: SC Weaver, M Denison, M Roossinck, et al. (2016) www.caister.com/virusevol

• Arboviruses: Molecular Biology, Evolution and Control

Edited by: N Vasilakis, DJ Gubler (2016) www.caister.com/arbo

• Shigella: Molecular and Cellular Biology

Edited by: WD Picking, WL Picking (2016) www.caister.com/shigella

• Aquatic Biofilms: Ecology, Water Quality and Wastewater Treatment

Edited by: AM Romaní, H Guasch, MD Balaguer (2016) www.caister.com/aquaticbiofilms

• Alphaviruses: Current Biology

Edited by: S Mahalingam, L Herrero, B Herring (2016) www.caister.com/alpha

• Thermophilic Microorganisms

Edited by: F Li (2015) www.caister.com/thermophile

• Flow Cytometry in Microbiology: Technology and Applications

Edited by: MG Wilkinson (2015) www.caister.com/flow

• Probiotics and Prebiotics: Current Research and Future Trends

Edited by: K Venema, AP Carmo (2015) www.caister.com/probiotics

• Epigenetics: Current Research and Emerging Trends

Edited by: BP Chadwick (2015) www.caister.com/epigenetics2015

• Corynebacterium glutamicum: From Systems Biology to Biotechnological Applications

Edited by: A Burkovski (2015) www.caister.com/cory2

• Advanced Vaccine Research Methods for the Decade of Vaccines

Edited by: F Bagnoli, R Rappuoli (2015) www.caister.com/vaccines

• Antifungals: From Genomics to Resistance and the Development of Novel Agents

Edited by: AT Coste, P Vandeputte (2015) www.caister.com/antifungals

• Bacteria-Plant Interactions: Advanced Research and Future Trends

Edited by: J Murillo, BA Vinatzer, RW Jackson, et al. (2015) www.caister.com/bacteria-plant

• Aeromonas

Edited by: J Graf (2015) www.caister.com/aeromonas

• Antibiotics: Current Innovations and Future Trends

Edited by: S Sánchez, AL Demain (2015) www.caister.com/antibiotics

• Leishmania: Current Biology and Control

Edited by: S Adak, R Datta (2015) www.caister.com/leish2

• Acanthamoeba: Biology and Pathogenesis (2nd edition)

Author: NA Khan (2015) www.caister.com/acanthamoeba2

• Microarrays: Current Technology, Innovations and Applications

Edited by: Z He (2014) www.caister.com/microarrays2

• Metagenomics of the Microbial Nitrogen Cycle: Theory, Methods and Applications

Edited by: D Marco (2014) www.caister.com/n2

Caister Academic Press is a leading academic publisher of advanced texts in microbiology, molecular biology and medical research. Full details of all our publications at caister.com

Further Reading

Order from caister.com/order

566 Keijser et al.

coelicolor ram cluster as a clone complementing thecopper-dependent phenotype of S. lividans. Wedemonstrate that fundamental differences exist in theimportance of this cluster of genes during the developmentbetween S. lividans and S. coelicolor, and that the needfor the ram gene products strictly depends on theconcentration of copper in the media.

Results

Copper Specific Chelators Specifically InhibitDevelopment of S. lividans and S. coelicolorThe relatively clear differences between the various stagesof morphological differentiation of surface-grownStreptomyces has allowed easy monitoring of this processin wild-type strains and developmental mutants. Thisfeature was also the key for the finding that copper stronglyinduces aerial hyphae formation of S. lividans. Raising thecopper concentration, supplied as chloride or sulphatesalts, in the standard rich medium (R2YE) from 0.2 to 5µM stimulated aerial hyphae formation in S. lividans andadvanced the switch from vegetative mycelium to aerialhyphae by 24 to 36 hours. No stimulation was observedwith the addition of other metal salts such as FeCl3, ZnCl2,NiCl2, CoCl2 or MgCl2 (Figure 1a). Aerial hyphae formationof the closely related strain S. coelicolor was not affectedby the elevation of the copper concentration in the media.

To study the behaviour under conditions of anextremely low copper concentration in the media, we testedthe effect of copper-chelating agents on the growth anddifferentiation of S. lividans and S. coelicolor. Interestingly,we found that the addition of 50 µM of a Cu(I)-specificchelator, bathocuproin disulfonic acid (BCDA), specificallyinhibited aerial hyphae formation of both S. lividans and S.coelicolor leaving vegetative growth and secondarymetabolism unaffected (Figure 1b). Normal developmentwas allowed to proceed upon the addition of extra copperto the BCDA plates.

The strong effect of copper on the development of S.lividans and the inhibition of development of both S. lividansand S. coelicolor by the copper-specific chelator seems tobe restricted to rich media. Neither the stimulatory effectof copper on the development of S. lividans nor the blockingeffect of BCDA on the development of S. coelicolor and S.lividans were observed on mannitol containing minimalmedium (data not shown).

A S. coelicolor Genomic Fragment Containing the ramCluster Complements the Copper-DependentPhenotype of S. lividansOn the rich medium, R2YE, the onset of aerial hyphaeformation of S. coelicolor occurs 24-36 hours earliercompared to S. lividans. Such an advanced onset ofdifferentiation and the abundance of aerial hyphae andspore production are reached by S. lividans only at a ‘high’copper concentration. Thus, as far as the differentiation isconcerned, S. coelicolor seems to be less dependent onthe copper concentration than S. lividans.

In an attempt to reveal the genetic background of thecopper-dependent development of S. lividans twoindependent libraries of S. coelicolor genomic DNA werescreened for clones that could render S. lividansdevelopment independent of extra copper. In total, threecomplementing clones, designated 1/1/11, 2/2/1, and 2/2/

10, were identified. These clones where subsequentlysubjected to restriction analysis, partial DNA sequencingand Southern hybridisation, and were all found to containthe ram (rapid aerial mycelium) gene cluster that wasidentified several years ago (Ma and Kendall, 1994). Theram cluster contains the genes encoding proteins belongingto the large family of ABC (ATP binding cassette)transporters (RamA, RamB) and a two-componentresponse regulator (RamR). The ram cluster consists oftwo additional open reading frames, which were notobserved before by Ma and Kendall (1994). A small ORFupstream of the ramA (ramS) with an unknown functionand a large open reading frame upstream of ramSputatively encoding a membrane bound serine/threonine

Figure 1. (A): Aerial hyphae formation of a surface-grown S. lividans cultureafter three days of growth was stimulated in a zone surrounding the cottonfilter disk via which 20µl of a 4mM copper(I) (Cu1) or copper(II)-chloride(Cu2) solution was added. The addition of other metal-salt solutions did notstimulate morphogenesis. (B): Phenotypes of S. lividans (top) and S.coelicolor (bottom) after five days of growth on standard R2YE (left) andafter the addition of 50 µM BCDA (right). In both strains, the developmentwas blocked at the onset of aerial hyphae formation by the presence of thecopper chelator BCDA.

A

B

S. lividans Differentiation 567

kinase (ramC). The latter one is dispensable forcomplementation since of this ORF only the C-terminalpart is present on clones 2/2/1 and 2/2/10 (see Figure 2A).The over-all organisation of the ram gene cluster is similarto that of the S griseus amf cluster (Ueda et al. 1993). Theintroduction of any of the complementing clones in S.coelicolor did not affect the growth and development inthis strain. Complementation experiments with deletionderivatives of plasmid pGC151 indicated that the minimumrequirement for complementation is a clone containing

ramSAB. This was in agreement with earlier observationsby Ma and Kendall (1994).

The S. lividans ram Cluster Also Fully Complementsthe Copper-Dependent DevelopmentIn order to adopt a S. coelicolor like differentiation pattern,S. lividans requires either an elevated copper concentrationor an extra copy of ramSAB. This difference between S.coelicolor and S. lividans may be due to the absence of a(functional) ram cluster in S. lividans. However, Ma and

Figure 2. (A): The genetic organisation of the ram cluster, present on thethree clones that complemented the copper-dependent differentiation of S.lividans. The clones that were used to construct the ram gene disruptionmutants were constructed by placing a hygR cassette in the BclI site presentin ramR (pBK154) or by displacing a 3.5 Kb BclI-BclI fragment by thiscassette (pB130/132). (B): The S. lividans ram cluster (GenBank accessionnumber AF139177) has been sequenced (part analysed represented bythe thick line in figure) and was shown to be over-all 99.6% identical to theS. coelicolor ram cluster. The C-terminus of ramA of S. lividans was foundto contain two inserts of 6 and 12 nt in respect to the S. coelicolor sequence.The C-terminal half of the S. lividans ramB ORF, was found to contain a 12nt. deletion in a glutamate/proline repeat of the putative protein. In addition,several point mutations were found. (C): Chimeric ram clusters wereconstructed by combining part of Slram with the complementary part ofScram.

C

B

A

568 Keijser et al.

Kendall (1994) demonstrated the presence in S. lividansof a DNA fragment homologous to ram by Southern blotting.More detailed Southern blotting experiments showed thepresence of an apparently complete ram cluster on the S.lividans genome, designated Slram. We therefore assumedthat the Slram cluster was impaired compared to the fullyactive S. coelicolor ram cluster (Scram) and cloned andsequenced the entire Slram cluster to reveal the apparentdefect (GenBank accession number AF139177).

By Southern hybridisation using fragments of the S.coelicolor ram cluster as probes, the S. lividans ram clusterwas localised on two DNA fragments. ramC, ramS, ramAand part of ramB were cloned on a 5.9 kb EcoRI - BamHIfragment. The remaining part of ramB and ramR werecloned as a 2.9 kb BamHI - SphI fragment. Both fragmentswere sub-cloned and sequenced on both strands.

The sequence of Slram and Scram gene clusters are99.6% identical (Figure 2B). A few significant differenceswere observed. The S. lividans ramA gene harbours twosmall inserts of 12 and 6 bp in respect to S. coelicolorramA. Furthermore, the S. lividans ramB ORF contained a12 bp deletion in a region encoding a glutamate/prolinerepeat at the C-terminal part of the putative protein. Theintergenic regions, putatively involved in the regulation oframAB ramS and ramR transcription were found to beidentical in both strains.

DNA fragments containing either Slram or Scram werecloned in the integrating vector pSET152 (pBK125,pGC151) and subsequently introduced into the S. lividansgenome. Morphological differentiation of the transformantscarrying an extra copy of ramSABR was compared to wild-type strains in the presence and absence of extra copper.These experiments showed that the Slram cluster fullycomplements the copper-dependent development of S.lividans and that the transformants were indistinguishablefrom the integrant containing the Scram cluster. In addition,the complete functionality of Slram was also demonstratedby introducing chimeric ram clusters constructed bycombining part of Slram with the complementary part ofScram. Again, the S. lividans transformants carrying thechimeric ram clusters were indistinguishable fromtransformants carrying either the S. coelicolor or the S.lividans ram cluster as an extra copy.

Construction and Analysis of the ramABR and ramRDisruption MutantsTo study the role of ramAB and ramR during morphologicaldifferentiation and to explore a possible relation with copper,two types of gene-disruption mutants were constructed inS. coelicolor and S. lividans. The disruption of ramR wasachieved by inserting a hygromycin B resistance cassettein the single BclI site of ramR (pBK154). A ramABRdisruption mutant was constructed by replacing a 3.2 kbBclI fragment containing most of ramA, the entire ramBand a large portion of ramR by a hygromycin B resistancecassette (pBK130/132) (Figure 2A and Experimentalprocedures).

The disruption of ramR in S. coelicolor or S. lividansresulted in strains that were 12 to 24 hours delayed in theonset of differentiation compared to the wild-type strains.The disruption of ramABR in S. lividans 1326 resulted in acomplete loss of the ability to form aerial hyphae andspores. Contrary, S. coelicolor retained most of itsdevelopmental capacities upon the disruption of ramABR.

Although delayed, aerial hyphae and spores were producedto near wild type levels (Figure 3). The effects of thedisruption of ramB in S. lividans TK64 (HM21) were alreadyknown to some extend (Ma and Kendall 1994). However,the effects we observed upon disruption of ramABR in S.lividans 1326 differ somewhat with the observationsdescribed earlier. Contrary to mutant HM21, the disruptionof ramABR in S. lividans 1326 results in a completely bldphenotype. An increase in the production of a red pigmentas was observed for HM21 was not observed for S. lividans1326 ∆ramABR in which vegetative growth and antibioticproduction were not affected by the disruption. Nor did weobserve any problems with the viability of the ram mutants.Finally, contrary to HM21, all ram disruption mutantsshowed a carbon source dependent behaviour andproduced normal aerial hyphae and spores when grownon either R2YE or minimal media containing mannitol ascarbon source. Under these conditions, no difference couldbe detected between wild type and mutant strains.

The difference between HM21 (a TK64 derivative) andthe S. lividans 1326 ramABR disruption mutant must becaused by the presence of intact ramA and ramR in HM21.Disruption of ramR does have a negative effect ondevelopment (see above) but also partial activity of RamAin HM21 can not be ruled out as suggested by Ma & Kendall(1994). Differences in pigmentation could be the result ofthe genetic background of the two strains, 1326 is a wildtype strain while HM21 is pro-2, str-6 SLP2- SLP3-

(Hopwood et al., 1985).

Figure 3. The phenotypes of the S. coelicolor (wt) and S. lividans (wt) andthe ramABR mutants after five days of growth on R2YE plates. The disruptionof ramABR in S. lividans and S. coelicolor affected normal development.The S. lividans ramABR mutant (top right) was unable to form aerial hyphaeand spores. The development of the S. coelicolor ramABR mutant (bottomright) was delayed compared to the S. coelicolor wild-type strain (bottomleft).

S. lividans Differentiation 569

Aerial hyphae and spore production in all ramdisruption mutants was stimulated when grown near wild-type strains. Even the bald phenotype of the S. lividansramABR mutant was fully complemented when grown nearwild-type S. lividans. Aerial hyphae and spores were formedin a zone of roughly 1 cm from the wild-type strain (Figure4a). This complementation could also be accomplished byusing spent agar media from wild-type strains. Since theseeffects were not observed using spent media from theramABR mutant itself, the observed effects must be dueto secretion of a stimulating compound by the wild typestrain. Extracellular complementation by wild type strainshas also been observed with some of the bld mutants and

proposed to be due to the supplementation a small peptide(SapB), essential for the onset of aerial hyphae formation(Willey et al. 1991). Extracellular complementation was alsoobserved with certain combinations of bld mutants. Thisphenomenon has led to the proposal of a hierarchicalcascade of intercellular signals leading to the onset of aerialhyphae formation (Willey et al. 1993). To test whether theram gene knock-out mutants could be placed within thissignalling cascade, we tested the extracellularcomplementation of the various bld mutants with the S.lividans ramABR disruption mutant. None of the bldA, bldB,bldC, bldD, bldF, bldG, bldH mutants were able to induceaerial hyphae formation in the S. lividans ramABR mutant.With the exception of bldA, none of the bld mutants wasstimulated to form aerial hyphae when grown near to theS. lividans ramABR disruption mutant.

Interestingly, raising the concentration of copper in themedia from 0.2 to 2 µM was sufficient to fully restore aerialhyphae and spore formation in all S. coelicolor and S.lividans ramABR and ramR mutants to wild type properties(Figure 4b). Similar amounts of iron-, magnesium-, zinc-and nickel chloride did not stimulate aerial hyphaeformation. Elevating the copper concentration in the mediahad no effect on the growth and development of the testedbld mutants.

Complementation of the ram Disruption MutantsThe ramABR disruption mutants of both S. coelicolor andS. lividans were transformed with the integrating vectorpSET152 harbouring the complete Slram cluster (pBK125),the complete Scram cluster (pGC151), or ScramR(pBK117). Both the Slram and the Scram clustercomplemented the S. lividans mutant to almost wild-typelevels, whereas ramR alone did not restore normaldevelopment. The delay in the development of the S.coelicolor mutant was fully restored by both ram clusters,and by ramR alone. The results obtained in thecomplementation experiments with the complete ramclusters were confirmed by similar experiments usingchimeric ram clusters (Figure 2c).

Transcription Analysis of ramABThe complementation experiments show that the delay intiming of S. lividans differentiation in comparison to that ofS. coelicolor, cannot be explained by a significant functionaldifference between their respective ram gene products.This observation prompted transcriptional analysis of thechromosomal region encompassing ramAB and upstreamsequences.

For this purpose total RNA was isolated from S.coelicolor, S. lividans and S. lividans containing the S.lividans cluster integrated as an extra copy in the genome(S. lividans::pBK125) grown for 1-5 days on polycarbonatemembranes on R2YE plates at 30°C.

Nuclease S1 mapping was performed with a 379 bpDNA probe harbouring the region encompassing the endof ramS and the start of ramA, and 64 bp of vectorsequence, the latter allowing discrimination between probereannealing and full length protection by RNA (Figure 5).Approximately 30 µg RNA was hybridised with 0.02 pmol(5x104 Cerenkov counts min-1) probe, processed andanalysed on a 6% (w/v) acrylamide gel under denaturingconditions (experimental procedures).

Interestingly, the appearance of the ramAB transcripts

Figure 4. (A): Aerial hyphae and spore formation of the S. lividans ramABRdisruption mutant (bottom) is restored when grown near a S. lividans wildtype strain (top). (B): Normal development of the S. livdans ramABR mutantwas restored when the concentration of copper in solid R2YE was raisedfrom 0.01 (left) to 2 µM (right).

A

B

S. lividans 1326 ∆ramA/B/R0.01 µM Cu2+ 2 µM Cu2+

570 Keijser et al.

corresponded to the onset of aerial mycelium formation;during the vegetative phase and spore stadium, the ramABexpression level was much lower. In all samples, bandscould be detected corresponding to full length protectionof the probe, indicating that the transcription of ramAB andramS are coupled.

In RNA isolated from S. coelicolor and S. lividans wild-type cultures, few weak bands could be detected only afterseveral days exposure of the X-ray film (Figure 5b and5c). In contrast, in RNA derived from cultures of S. lividanscontaining an extra copy of S. lividans ram, ramABtranscripts could readily be detected after a few hoursexposure of the X-ray film (Figure 5a). Apparently, additionof an extra copy of ram dramatically increases ramABtranscription.

The larger band (FLP) of approximately 315 ntcorresponds to full-length protection of the probe, stronglysuggesting that ramAB transcription arises at least partiallyfrom a promoter upstream of ramS. A shorter band (X) ofapproximately 190 nt was observed, corresponding to aposition 20 nt downstream of ramS. Since no promoter-like consensus sequences could be identified upstream ofthis position, we suggest that band X is the result ofprocessing of a larger transcript.

Discussion

Copper Plays a Crucial Role During the Onset of AerialHyphae FormationCopper ions play an important role in the development ofStreptomyces lividans and Streptomyces coelicolor. Thiswas clearly demonstrated by the stimulatory effect ofcopper ions on the differentiation of S. lividans and thedevastating effect of the Cu(I) specific chelator BCDA onthe development of S. lividans and S. coelicolor (Figure1). The developmental block that occurred after the additionof the BCDA could be relieved by adding extra Cu2+,indicating that the effects observed are not due to any toxiceffects of BCDA but specifically to the chelation copper.The specific effect of BCDA on developmental pathwaysmay be due to the inactivation of an extracellular copper-dependent protein, essential for the onset of aerialmycelium formation. Alternatively the effect may also beexplained by a change in copper metabolism during theswitch from vegetative to aerial growth resulting in a BCDA-sensitive phenotype. The apparent insusceptibility ofvegetative growth and secondary metabolism to thepresence of the copper chelator indicates that the observedeffects are not due to a copper deficiency. One of the manyconsequences of a copper deficiency is a malfunctioningin the respiratory pathways due to a lower activity of the

Figure 5. Nuclease S1 mapping of theramAB transcripts of total RNA of S. lividanscarrying an extra copy of the S. lividans ramcluster (a) S. lividans (b) and S. coelicolor(c), isolated at various time points duringgrowth on solid media. Samples ofvegetative mycelium (V) were taken earlyduring growth and at the appearance of thepigment undecylprodigiosin (Red). Whenonly one vegetative sample is shown, thatof the latter growth phase is used. Aerialhyphae samples (A) were taken at theappearance of aerial hyphae and just beforesporulation started. Samples taken duringsporulation (S) were taken at theappearance of the grey spore pigment. FLPmarks the signal for full-length protection.P indicates the 379 bp probe and M the end-labelled ΦX174 HinfI digest. Band X isprobably the consequence of processing ofa larger transcript (see text). The drawingon the top represents the 379 bp probe usedduring the experiments with the asterisksrepresenting the 32P labelled end of theprobe.

S. lividans Differentiation 571

copper containing cytochrome c oxidase, which would havebeen reflected on vegetative growth as well. The possibilityof the copper-BCDA complex being taken up can be ruledout since this is known to lead to severe DNA damage andinhibition of DNA polymerase I (Pope et al., 1982).

The ram Cluster is Crucial for Morphogenesis of S.lividans But Not of S. coelicolorThe S. coelicolor ram cluster was identified by its ability tocomplement the copper-dependent development of S.lividans. Based of the difference in copper dependenceduring the development of S. coelicolor and S. lividans weinitially assumed the Slram cluster to be either functionally‘impaired’ or expressed at a lower level compared to theScram cluster. However, results obtained fromtranscriptional analysis and additional complementationstudies of the wild-type strains and the ram-disruptionmutants indicated that this assumption is not valid.

Significant functional differences between Scram andSlram could be excluded since an additional copy of eitherthe S. lividans or the S. coelicolor ram cluster fullycomplemented the copper-dependent differentiation of S.lividans 1326. Furthermore, no differences could bedetected between the S. lividans ramABR disruption mutantcomplemented by either Slram or Scram. Although aerialmycelium formation by S. coelicolor is much more abundantthan by S. lividans, transcription analysis failed to revealsignificant differences in the ramAB transcript levels in thetwo wild-type strains. Interestingly, the timing of ramABtranscription corresponds well to the development of aerialmycelium, again linking these genes to morphologicaldifferentiation. Introduction of one extra copy of the S.lividans ram cluster into S. lividans 1326 resulted inabundant aerial mycelium formation, similar to that

observed for S. coelicolor. It is likely that this is caused bythe strongly increased expression of ramAB, as reflectedby the abundant transcript levels observed (see Figure 5).While we cannot exclude a contribution of the increasedexpression of ramS to S. lividans development, introductionof a truncated construct harbouring the complete ramS andramA genes, but not ramB (pBK122), was not sufficient tocomplement the copper-dependent differentiation of S.lividans 1326. Apparently, the development of S. coelicoloris less dependent on the ramAB expression level than S.lividans, which may indicate the presence of a ramAB-independent developmental pathway in S. coelicolor(Figure 6).

This hypothesis was supported by the phenotypes ofthe various S. lividans and S. coelicolor ram gene disruptionmutants. The S. coelicolor ramABR disruption mutant was,contrary to the S. lividans ramABR disruption mutant,capable of producing aerial hyphae and spores.

The delay that was observed in the onset of aerialmycelium formation of the S. coelicolor ramABR mutant ismore likely to be due to the disruption of ramR than to anyeffect on ramAB since this mutant could be fullycomplemented by the introduction of ramR. In S. coelicolorramAB seem to be dispensable for normal developmentand thus indicating the presence of an alternativedevelopment route which is apparently missing ofmalfunctioning in S. lividans. This S. coelicolor-specificalternative development route may either by-pass the needfor Ram activity or may represent an alternative route withthe same activity. The latter option seems to be the mostlikely since aerial hyphae and spore formation in the S.lividans ramABR disruption mutants is stimulated whengrown near the S. coelicolor ramABR mutant (data notshown).

The absence of this ram-independent system in S.lividans may be the cause of the intrinsic difference in S.lividans copper-dependent development compared to thatof S. coelicolor. Since the components involved in thissystem were not identified during the complementationexperiments with the two independent genomic libraries,the genes encoding this system may either be distributedover multiple parts of the genome or span a size largerthan 25 kb.

RamAB Appears to be Involved in the Transport of aFactor Essential for Aerial Hyphae FormationThe bald phenotype of the S. lividans ramABR disruptionmutant can be relieved by either placing the mutant near aS. lividans wild-type strain or by increasing the copperconcentration in the media (Figure 4). The firstphenomenon can be explained by the requirement of anextracellular factor that under normal conditions is exportedby the ABC-transporters RamA and RamB and is absentin the S. lividans ramABR disruption mutant. The secondobservation indicates that at high copper levels this ram-associated factor (RAF) becomes obsolete (Figure 6).

The question that remains is whether or not the effectsof copper and the role of the ram geneproducts duringdifferentiation are linked at a functional level. If thisfunctional relation does not exist, then an elevated copperconcentration in the media may make alternative, ramindependent, differentiation routes accessible. Alternatively,if a functional relation between copper and ram does exists,then the extracellular ram-associated factor may for

Figure 6. A schematic overview of the observations made for the copper-dependent onset of aerial hyphae formation and the role of the ram genes.The middle part of the figure depicts the ram-dependent aerial hyphaeformation on R2YE (0.2 µM copper). The dotted line indicates a ram-independent route for aerial hyphae formation that seems to be present inS. coelicolor. On the right site, the ram-independent growth of aerial hyphaein the presence of a high copper concentration (>2µM) is depicted. The leftsite shows the specific block of aerial hyphae formation in the presence ofa copper-specific chelator (no copper).

ram

572 Keijser et al.

instance be involved in copper trafficking, which includesits uptake and transport, as well as the incorporation ofcopper into apo-proteins. Higher copper concentrationswould render a high affinity trafficking system redundant,possibly accounting for the observed complementation.However, analysis of the aminoacid sequence of the ramgene products does not provide indications for a linkbetween ram and copper.The nature of the Ram-associated factor and its role duringcopper-dependent differentiation of S. lividans is subjectof further research.

Experimental Procedures

Strains and MediaAll E. coli and Streptomyces strains used in this study are listed in Table 1.E. coli strains were grown at 37°C in Luria broth or on Luria broth agarplates as described by Sambrook et al. (1989).

Streptomyces strains were grown on the rich agar medium R2YEcontaining 0.2 µM copper chloride (Hopwood et al., 1985) or the minimalagar medium SMM (Strauch et al., 1991). The concentration of copper inthe media was elevated by the addition of copper salt solutions(CuSO4,CuCl2). For liquid cultures we used YEME (Hopwood et al., 1985)or TSBS (3% (w/v) Tryptone Soya Broth with 10% (w/v) sucrose). Theantibiotics (ampicillin, apramycin and hygromycin B) were purchased fromDucheva (Haarlem, The Netherlands). Thiostrepton was a kind gift of Bristol-Meyers Squibb.

General Cloning Procedures and DNA TechniquesStandard genetic techniques, agarose gel electrophoresis, plasmidpreparations and in vitro DNA manipulations were performed according to

published protocols (Sambrook et al., 1989). All plasmids used andconstructed during this study are listed in Table 2. Restriction enzymes andT4 DNA ligase were purchased from Pharmacia (Sweden). The handlingand manipulation of Streptomyces strains was performed according toprotocols described by Hopwood et al. (1985). Streptomyces genomic DNAwas isolated using a genomic DNA isolation kit (Puregene) in accordancewith the manufacturers instructions for Gram-positive bacteria with themodification that prior to lysis, the cells were incubated in 4 mg/ml lysozymefor 30 minutes at 30°C. Southern blotting and hybridisation was performedas described by Sambrook et al. (1989). DNA probes for Southern blotanalysis were labelled by random prime incorporation of digoxigenin-dUTP(DIG-labelled DNA) using the DIG DNA labelling kit (Boehringer Mannheim).The hybridisation was performed at 45°C in Easy Hyb (BoehringerMannheim) and stringent washes (0.1x SSC, 0.1% (w/v) SDS) at 65°C.For the detection of the alkaline phosphatase-antibody conjugate whichrecognises the DIG-moiety, CSPD (Boehringer Mannheim) was used as asubstrate.

Complementation With the S. coelicolor Genomic LibraryDuring the complementation experiments, two S. coelicolor genomiclibraries, a kind gift of J. Ryding and K. Chater, have been used. The firstlibrary consisted of fragments S. coelicolor M145 genomic DNA partiallydigested with Sau3AI with an average size of 10 kb, ligated in BglII-digestedpIJ698. The library was introduced in S. coelicolor J1501 by protoplasttransformation (Hopwood et al., 1985). A total of 2500 transformants wasmaintained as sporulated patches on R2YE plates. This number oftransformants should cover close to 100% of the S. coelicolor genome.The library was introduced in S. lividans by replica plating and mating onMM mannitol plates containing 50 µg/ml histidine and 7.5 µg/ml uracil. Thesubsequent counter selection for the donor, J1501, was carried out by replicaplating the ‘mating plates’ on MM plates containing mannitol without uraciland histidine.

The second library was derived from S. coelicolor 1147 genomic DNA,partially digested with Sau3AI. Fragments with an average length of 20-25

Table 2. Plasmids

Plasmid Description Reference/source

pUC18 E.coli cloning vector Veira and Messing, 1982pTZ19R/18R pUC18/19 derivative containing f1 ori allowing the generation of single stranded DNA Pharmacia, SwedenpIJ698 Low copy number Streptomyces cloning vector (carrying thiostrepton resistance marker) Kieser and Melton, 1988pSET152 E. coli / Streptomyces shuttle vector (carrying apramycin resistance Bierman et al., 1992

marker) that allows integration in the ΦC31 attachement site on the Streptomyces chromosomepGC151 S. coelicolor ram cluster (BglII insert of 1/1/11) in BamHI site of pSET152 This workpGC153 S. coelicolor ram cluster (BglII insert of 1/1/11) cloned into BamHI site of pUC18. This workpBK121 The reconstructed S. lividans ram cluster in pUC18 (5.9 Kb EcoRI/BamHI fragment ligated to the 2.9 Kb This work

BamHI/SphI ram fragment)pBK122 pGC153 ∆BamHI This workpBK125 The insert of pBK121 cloned in pSET152 This workpBK 130 pBK121 ∆BclI :: hyg, apra This workpBK 132 pGC153 ∆BclI :: hyg apra This workpBK154 2.3 Kb BamHI/HindIII fragment of pGC153 containing ramR hyg (BclI) apra (HindIII) This work

Table 1. Strains

Strain Genotype/phenotype Reference/source

E. coliJM109 F' traD36 lacIq ∆(lacZ)M15 proA+B+l e14-(McrA-) ∆(lac-proAB) thi gyrA96 Yanish-Perron et al., 1985

(Nalr) endA1 hsdrR17 (rk-mk+) relA1 supE44 recA1ET12567 F- dam-13::Tn9 dcm-6 hsdM hsdR recF143 zjj-202::tn10 galK2 galT22 MacNeil et al., 1992

ara14 lacY1 xyl-5 leuB6 thi-1 tonA31 rspL136 hisG4 tsx78 mtl-1 glnV44StreptomycesS. lividans 1326 wild type Hopwood et al., 1985S. coelicolor M145 Prototrophic SCP1- SCP2- Hopwood et al., 1985J1700 bldA39 hisA1 ura1 strA1 SCP1-SCP2- Pgl- Lawlor et al., 1987J669 bldB43 mthB2 cysD18 agaA7 NF Merrick et al., 1976J660 bldC18 mthB2 cysD18 agaA7 NF Merrick et al., 1976J774 bldD53 cysA15 pheA1 mthB2 strA1NF Merrick et al., 1976WC103 bldG103 hisA1 uraA1 strA1 SCP1-SCP2-Pgl- Champness, 1988WC109 bldH109 hisA1 uraA1 strA1 SCP1-SCP2-Pgl- Champness, 1988

S. lividans Differentiation 573

kb were ligated in BglII-digested pIJ698 and introduced directly in S. lividans1326 by protoplast transformation.

Cloning and Sequencing of the S. lividans ram ClusterThe S. lividans ram cluster was localised on two DNA fragments by Southernhybridisation using the S. coelicolor ram genes as probes. The first, a 5.9kb EcoRI - BamHI fragment containing ramC, ramS, ramA and a part oframB, the second a 2.9 kb BamHI - SphI fragment containing the remainingpart of ramB and ramR. DNA fragments of the corresponding size wereisolated from 1% (w/v) agarose TAE gels and cloned in pTZ18R. For thesequence determination a total 42 overlapping subclones of the twofragments were constructed in the phagemid vectors pTZ18R or pTZ19R.Single stranded DNA templates were sequenced at Baseclear (Leiden, TheNetherlands) using a LiCor Automated DNA sequencer. Regions containingcompressions were sequenced using the Thermo Sequenase dye terminatorkit (Amersham Life Science) and analysed a an ABI Prism 373 XL DNAsequencer (Perkin Elmer). The complete cluster was sequenced on bothstrands. To rule out the possibility of an additional BamHI site close to thesite used for cloning, a fragment covering this region of the ram cluster wasamplified by PCR and sequenced. GenBank accession number of the S.lividans ram cluster: AF139177

Transcription AnalysisMycelium grown on R2YE plates overlaid with polycarbonate track-etchmembranes (Osmotics, Livermore CA, USA) was harvested at various pointsduring growth. Total RNA was isolated using the SV RNA isolation kit(Promega) according to the procedures described by Venendaal and Wösten(1999).

For each nuclease S1 protection assay, approximately 0.02 pmol(5x104 Cerenkov counts min-1) of labelled probe was hybridised to 30 µg ofRNA in Na-TCA buffer (Murray, 1986). After denaturation at 72°C for 15minutes, samples were allowed to cool down O/N to a final temperature of45°C. Subsequently, samples were diluted in 300 µl ice-cold S1 buffer(Favaloro et al., 1980) containing 100 units of S1 nuclease and chilled onice. Digestions were done at 37°C for 45 minutes after which the sampleswere precipitated with 0.1 volume 3M sodium acetate and 2 volumes ethanol.After centrifugation, the pellets were dissolved in standard sequencing gel-loading dye, denatured by heat and electrophoresed on denaturing 6% (w/v) polyacrylamide sequencing gels. The ramAB S1 probe that was used todetect the ramAB transcripts was generated by PCR using pramAB (5'-CGG CCG CCG CGA CCG AGC AC-3'), which was designed based on thesequence, approximately 80 bp downstream of the putative translationstartsite of RamA, and the M13 reverse sequencing primer (New EnglandBiolabs) (5'-AAC AGC TAT GAC CAT G-3'). As template we used a PstI-BclI subclone containing parts of ramS and ramA. The polymerase chainreaction was performed in a minicycler (MJ research, Watertown MA, USA)using Pfu polymerase (Stratagene, La Jolla CA, USA) and the bufferprovided by the supplier in the presence of 5% (v/v) DMSO, 25 pmol ofeach primer and 0.2 mM of each dNTP and 10 ng template (usuallypredigested genomic DNA). The amplification was performed in 30 cyclesconsisting of 60s at 94°C, 60s at 54°C and 60s at 72°C. The 379 nt probecontaining a 64 nt non- homologous extension was labelled at the 5’ endwith [γ-32P]-ATP and T4 polynucleotide kinase.

Construction of ram Gene Disruption MutantsIn all plasmid constructs for insertional inactivation of ram genes on thegenome, the hygromycin B resistance gene was used as the selectablemarker (Blondelet-Rouault et al., 1997). In addition a 2.3 Kb fragmentcontaining the apramycin resistance gene was cloned into the HindIII siteof the multiple cloning site of the vector, allowing discrimination betweensingle and double homologous cross-over events. Subsequently, theplasmids were transferred to E. coli ET12567 to circumvent the restrictionbarrier of S. coelicolor. Prior to the protoplast transformation the DNA wasdenatured by heating the sample for 5 minutes at 95°C in an ethyleneglycol/glycerol/EDTA buffer followed by quick cooling on ice (Oh and Chater1997). After transformation, the S. coelicolor and S. lividans protoplastswere regenerated on R2YE plates. Selective antibiotic pressure was appliedafter 18 hrs by a hygromycin B overlay resulting in a final concentration of100 µg/ml. After 7 days incubation at 30°C, the hygromycin B resistantmutants were checked for the absence of apramycin resistance.

The ramABR genes were disrupted by replacing the 3.5 kb BclIfragment containing most of ramA, the entire ramB gene and a large portionof ramR by the 1.7 Kb fragment containing the hygromycin B resistancegene, in pBK121 and pGC153. The ramR disruption mutants wereconstructed by inserting the hygromycin B resistance gene in the BclI sitein the C-terminal half of the ramR ORF. For the isolation of the ramR knock-out mutants in S. lividans a total of 16 hygromycin B resistant (hygR) colonieswas checked and of these four were found to be apramycin sensitiveindicating a correct double crossover event. In S. coelicolor 2 of the 18hygR mutants were found to be apramycin sensitive. In the case of the

ramABR knock-out mutant constructions a total of 12 hygR mutants werechecked in S. lividans of which four were apramycin sensitive. In S. coelicolorthree mutants out of the selected 18 hygR mutants were found to beapramycin sensitive. The integrity of all the hygromycin B resistant andapramycin sensitive mutants was checked by Southern blotting using threeprobes: a 1.6 Kb BclI-SphI fragment of the S. coelicolor ram cluster, theentire 1.7 Kb hygromycin B resistance gene and the 2.3 Kb apramycinresistance gene.

Acknowledgements

This work was supported by a NWO/British Council grant (to E.V.). Wethank J. Ryding and K. Chater for providing the S. coelicolor genomiclibraries.

References

Bierman, M., Logan, R., O’Brien, K., Seno, E.T., Roa, R.N., Schoner, B.E.1992. Plasmid cloning vectors for the conjugal transfer of DNA fromEscherichia coli to Streptomyces spp. Gene 116: 43-49.

Blondelet-Rouault, M.H., Weiser, J., Lebrihi, A., Branny, P. and Pernodet,J.L. 1997. Antibiotic resistance gene cassettes derived from the omegainterposon for use in E. coli and Streptomyces. Gene 190: 315-317.

Champness, W.C. 1988. New loci required for Streptomyces coelicolormorphological and physiological differentiation. J. Bacteriol. 170: 1168-74.

Chater, K.F., 1998. Taking a genetic scalpel to the Streptomyces colony.Micrbiology 144; 1465-1478.

Chater, K.F. 1972. A morphological and genetic mapping study of whitecolony mutants of Streptomyces coelicolor. J. Gen. Microbiol. 72: 9-28.

Chevion, M. 1988. A site-specific mechanism for free radical inducedbiological damage: the essential role of redox-active transition metals.Free Radic. Biol. Med. 5: 27-37.

Dionigi, C.P., Ahten, T.S. and Wartelle, L.H. 1996. Effects of several metalson spore biomass and geosmin productionby Streptomyces tedae andPenicillum expansum. J. Ind. Microbiol. 17: 84-88.

Favaloro, J., Treisman, R. and Kamen, R. 1980. Transcription maps ofpolyoma virus-specific RNA: analysis by two-dimensional nuclease S1gel mapping. Methods Enzymol. 65: 718-749.

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 streptomyces: A laboratory manual. Norwich: TheJohn Innes Foundation.

Hopwood, D.A.,Wildermuth, H. and Palmer, H.M. 1970. Mutants ofStreptomyces coelicolor defective in sporulation. J. Gen. Microbiol. 61:397-408.

Horinouchi, S. and Beppu, T. 1994. A-factor as a microbial hormone thatcontrols cellular differentiation and secondary metabolism in Streptomycesgriseus. Mol. Microbiol. 12: 859-864.

Kieser, T. and Hopwood, D.A. 1991. Genetic manipulation of Streptomyces:integrating vectors and gene replacement. Methods Enzymol. 204, 430-458.

Kieser T., Melton R.E. 1988. Plasmid pIJ699, a multi-copy positive-selectionvector for Streptomyces. Gene, 65: 83-91.

Kudo, N., Kimura, M., Beppu, T. and Horinouchi, S. 1995. Cloning andcharacterization of a gene involved in aerial mycelium formation inStreptomyces griseus. J. Bacteriol. 177: 6401-6410.

Lawlor, E.J., Baylis, H.A. and Chater, K.F. 1987. Pleitropic morphologicaland antibiotic deficiencies result from mutations in a gene encoding atransfer RNA-like product in Streptomyces coelicolor A3(2). Genes Dev.1: 1305-1310.

Ma, H. and Kendall, K. 1994. Cloning and analysis of a gene cluster fromStreptomyces coelicolor that causes accelerated aerial myceliumformation in Streptomyces lividans. J. Bacteriol. 176: 3800-3811.

MacNeil, D.J., Gewain, K.M., Ruby, C.L., Dezeny, G., Gibbons, P.H. andMacNeil, T. 1992. Analysis of Streptomyces avermitilis genes requiredfor avermectin biosynthesis utilizing a novel integration vector. Gene 111:61-68.

Merrick, M.J. 1976. A morphological and genetic mapping study of baldcolony mutants of Streptomyces coelicolor. J. Gen. Microbiol. 96: 299-315.

Murray, M.G. 1986. Use of sodium trichloroacetate and mung bean nucleaseto increase sensitivity and precision during transcript mapping. Anal.Biochem. 158: 165-170.

Nodwell, J.R. and Losick, R. 1998. Purification of an extracellular signalingmolecule involved in production of aerial mycelium by Streptomycescoelicolor. J. Bacteriol. 180: 1334-1337.

Nodwell, J.R., McGovern, K. and Losick, R. 1996. An oligopeptide permeaseresponsible for the import of an extracellular signal governing aerialmycelium formation in Streptomyces coelicolor. Mol. Microbiol. 22: 881-

574 Keijser et al.

893.Nodwell, J.R.,Yang, M.,Kuo, D. and Losick, R. 1999. Extracellular

complementation and the identification of additional genes involved inaerial mycelium formation in Streptomyces coelicolor. Genetics 151: 569-84.

Odermatt, A., Suter, H., Krapf, R. and Solioz, M. 1993. Primary structure oftwo P-type ATPases involved in copper homeostasis in Enterococcushirae. J. Biol. Chem. 268: 12775-12779.

Oh, S.H. and Chater, K.F. 1997. Denaturation of circular or linear DNAfacilitates targeted integrative transformation of Streptomyces coelicolorA3(2)., possible relevance to other organisms. J. Bacteriol. 179: 122-127.

Pope, L.M., Reich, K. A., Graham, D. R. and Sigman, D. S. 1982. Productsof DNA cleavage by the 1,10-phenanthroline-copper complex. Inhibitorsof Escherichia coli DNA polymerase I. J. Biol. Chem. 257: 12121-12128.

Ryding, N.J., Bibb, M.J., Molle, V., Findlay, K.C., Chater, K.F. and Buttner,M.J. 1999. New sporulation loci in Streptomyces coelicolor A3(2). J.Bacteriol. 181: 5419-5425

Sambrook, J., Fritsch, E.F. and Maniatis, T. 1989 Molecular cloning: alaboratory manual. Cold Spring Harbor, New York: Cold Spring HarborLaboratory Press.

Silver, S. 1996. Bacterial resistances to toxic metal ions—a review. Gene179: 9-19.

Strauch, E., Takano, E., Baylis, H.A. and Bibb, M.J. 1991. The stringentresponse in Streptomyces coelicolor A3(2). Mol. Microbiol. 5: 289-298.

Ueda, K., Miyake, K., Horinouchi, S. and Beppu, T. 1993. A gene clusterinvolved in aerial mycelium formation in Streptomyces griseus encodesproteins similar to the response regulators of two-component regulatorysystems and membrane translocators. J. Bacteriol. 175: 2006-2016.

Ueda, K., Tomaru, Y., Endoh, K. and Beppu, T. 1997. Stimulatory effect ofcopper on antibiotic production and morphological differentiation inStreptomyces tanashiensis. J. Antibiot. 50: 693-695.

Veenendaal, A.K.J. and Wösten, H.A.B. 1998. SV Total RNA IsolationSystem in Action. Total RNA from the Gram-positive bacteriumStreptomyces lividans. Promega Benelux News 17: 7.

Vieira, J. and Messing, J. 1982. The pUC plasmids, an M13mp7-derivedsystem for insertion mutagenesis and sequencing with synthetic universalprimers. Gene 19: 259-268.

Willey, J., Santamaria, R., Guijarro, J., Geistlich, M. and Losick, R. 1991.Extracellular complementation of a developmental mutation implicates asmall sporulation protein in aerial mycelium formation by S. coelicolor.Cell 65: 641-650.

Willey, J., Schwedock, J. and Losick, R. 1993. Multiple extracellular signalsgovern the production of a morphogenetic protein involved in aerialmycelium formation by Streptomyces coelicolor. Genes Dev. 7: 895-903.

Wunderli-Ye, H. and Solioz, M. 1999. Copper homeostasis in Enterococcushirae. Adv. Exp. Med. Biol. 448: 255-264.

Yanish-Perron, C., Viera, J. and Messing, J. 1985. Improved M13 phagecloning vectors and host strains: nucleotide sequences of the M13mp18and pUC19 vectors. Gene 33: 103-119.