stage. While hrdB is transcribed from both promoters in all developmental stages, both tandem .... approximately 5 h prior to the appearance of coloured spores.
Nucleic Acids Research, 1993, Vol. 21, No. 16 3647-3652
Differential expression of principal sigma factor homologues of Streptomyces aureofaciens correlates with the developmental stage Jan Kormanec and Marian Farkagovskq Institute of Molecular Biology, Slovak Academy of Sciences, Dubravska cesta 21, 842 51 Bratislava, Slovak Republic Received May 28, 1993; Revised and Accepted July 12, 1993
ABSTRACT In previous experiments, Streptomyces aureofaciens has been shown to contain four genes hrdA, hrdB, hrdD, and hrdE, encoding polypeptides very similar to principal sigma factors of RNA polymerase. Two apparent tandem promoters were identified for each of the hrdA, hrdB and hrdD genes by SI nuclease mapping using RNA prepared of S.aureofaciens in various developmental stages. Under all the conditions studied, tandem promoters of each gene differed significantly in their respective strengths. Transcription from the hrd promoters depended on developmental stage. While hrdB is transcribed from both promoters in all developmental stages, both tandem promoters of the hrdD gene are active only in vegetative stage and transcription of the hrdA tandem promoters temporally correlates with the aerial mycelium formation. In addition to a promoter, hrdB-P2, which lies upstream of the open reading frame, the hrdB gene, proposed to encode functional principal factor, appeared to contain at least one intemal promoter, hrdB-Pl. Activity of all promoters was consistent with Si mapping experiments after insertion of promoter-bearing DNA fragments to promoter-probe vectors plJ486 and pARC1. The results implicate temporally different expression of the hrd genes during the differentiation of S.aureofaciens. INTRODUCTION Streptomycetes are mycelial, Gram-positive bacteria which produce a variety of biologically active secondary metabolites, including the majority of known antibiotics (1). Streptomycetes undergo a complex cycle of morphological differentiation. The initial vegetative stage is characteristic by a network of branching, multinucleate hyphae (the substrate mycelium). Differentiation occurs on solid surfaces. The older parts of the substrate mycelium give rise to aerial mycelium consisting of upwardly protruding hyphae that impart a characteristically fuizzy, white appearance on the colonies. The aerial hyphae undergo septation to form chains of uninucleate cells, which metamorphose into
GenBank accession nos M90410, M90411, M90413
pigmented spores that impart a species-specific colour on the colonies (2). Among several levels of regulation of the differentiation, transcriptional regulation by heterogeneity of a factors has been shown to play an important role (for review see ref. 1). The a factor encoded by whiG has been shown to be obligatory for sporulation in S. coelicolor A3(2) (3). In addition, S. coelicolor contains four other genes (hrd4, hrdB, hrdC and hrdD) that appear to encode a factors highly homologous to a70 of E. coli (4, 5, 6). Only one of them, hrdB, was shown to be essential for growth (4). It encodes d66 that is suggested to direct transcription of 'house-keeping' genes after association with the core of RNA polymerase (7). The function of hrdA, hrdC and hrdD is unclear at present (8). In S.aureofaciens CCM3239 (ATCC 10762), four genes, hrdA, hrdB, hrdD and hrdE, coding for homologues of principal a factor of RNA polymerase were identified (9). Based on the close sequence similarity and on initial transcriptional analysis, three of them, hrdA, hrdB and hrdD, appear to represent analogues of S. coelicolor hrd genes. The hrdE gene probably represents a new rpoD homologue (9). Northern blot hybridization experiments revealed that, similar to S. coelicolor (4), only hrdB and hrdD are efficiently expressed in cells growing in liquid medium (9). To elucidate possible function of the hrd genes in S.aureofaciens, we have analyzed their transcription in various stages of development of the microorganism. The obtained data show temporally different expression of the hrd gene products during the S.aureofaciens differentiation. The observed presence of a possible internal promoter in hrdB coding region, indicating a complex transcription pattern, is also discussed. MATERIAL AND METHODS Strains, plasmids and culture conditions E.coli and Streptomyces strains, plasmids are described in Table 1. Conditions for cultivation of E. coli and Streptomyces strains were as described (10, 1 1). For RNA isolation, S.aureofaciens was cultured to late exponential phase (24 h) in liquid medium NMP (11) containing 0.5% glucose as a carbon source. For RNA
3648 Nucleic Acids Research, 1993, Vol. 21, No. 16 Table 1. Strains and plasmid used
Strain, plasmid
Genotype
Ref. or source
recB recJ sbcC201 uvrC umuC::TnS (kanR) mnrA mcrB mrr lac hsdRMS endAl gyrA96
Stratagene, LaJolla, California
Su-ais E.coli SURE
thi relA1 supE44 F'(proAB lacIqZAM15
TnJO) S.aureofaciens
Wild type
Czechoslovak Collection of Microorganisms, Bruno
S.lividans TK24
str-6 SLP2- SLP3-
(24)
plJ486
High copy promoter-probe Streptomyces vector; thio' for selection; promoterless neo gene for promoter screening
(18)
pARCI
Low copy promoter-probe Streptomyces vector; thioR for selection; promoterless part of whiE operon for promoter screening
(19)
pIJ-hrdBl
pLJ486 containing 360 bp XmnI-Aval[ hrdB-Pl promoter bearing fragment
This work
pU-hrdB2
pU486 containing 360 bp Sall-XmnI hrdB-P2 promoter bearing fragment
This work
pU-hrdDl
pU486 containing 300 bp Bgil-SacI hrdD-P1 promoter bearing fragment
This work
pU-hrdD2
pIJ486 containing 560 bp Sall-Bgfl hrdD-P2 promoter bearing fragment
This work
pU-hrdAl2
pLJ486 containing 520 bp Pvull-BamHI hrd4-P1 and hrd4-P2 promoters bearing fragment
This work
pARC-hrdA
pARCl containing 460 bp PvuI-BamHI hrd4-P1 and hrd4-P2 promoters bearing fragment
This work
pARC-hrdB
pARCl containing 720 bp Sal-AvaIl hrdB-PI and hrdB-P2 promoters bearing fragment
This work
pARC-hrdD
pARCI containing 880 bp Sail-Sac hrdLD-Pl and hrdD-P2 promoters bearing fragment
This work
Plasmids
isolation from surface culture, 108 spores were spread on sterile cellophane membranes placed on Bennet medium (12) and grown to the appropriate phase of development. 13 h growth yielded substrate mycelium, 19 h growth led to the beginning of aerial mycelium formation and 36 h growth late aerial mycelium stage, approximately 5 h prior to the appearance of coloured spores. Level of Neomycin (Neo) resistance of S. lividans TK24 with recombinant p11486 plasmids bearing promoter fragments was assayed on Bennet medium with 50 Ag/ml Thiostreptone (Thio) and with increasing Neo concentration. DNA manipulations and transformations E.coli and Streptomyces DNA manipulations, transformations and plasmid isolation were done as described (10, 11). DNA fragments for DNA manipulations were isolated from agarose gel by GeneClean technique (BIO 101, LaJolla). RNA isolation RNA was isolated from surface cultures scrapped from cellophane membranes and harvested liquid cultures of S.aureofaciens as previously described (13). Integrity of RNA was checked
according to the intactness of internal rRNA after electrophoresis in agarose containing 2.2 M formaldehyde (10).
Si nuclease mapping The high-resolution SI mapping described previously (13) was used with some modifications. DNA fragments for Si nuclease mapping (Fig. 1) were dephosphorylated with calf intestinal alkaline phosphatase (BioLabs) and 5' end-labelled with T4 polynucleotide kinase (BioLabs) and [y-32P]ATP (Amersham, 3000 Ci/mmol) as described (10). After further digestion with appropriate restriction endonuclease, uniquely labelled probes were purified from polyacrylamide gel as described (14). 40 jig of RNA was co-precipitated with 10 ng of the labelled fragment (approximately 200 000 cpm). The pellet was dissolved in 20 ,gl of water and lyophilized. The resulting pellet was resuspended in 20 ,l of 3M NaTCA buffer (15). After 5 min incubation at 65°C, the temperature was slowly decreased to 450C and hybridization continued for 4 h. The samples were further treated as described (15). The protected DNA fragments were analyzed on DNA sequencing gels together with G+A and T+C sequencing ladders derived from the end-labelled fragments (16).
Nucleic Acids Research, 1993, Vol. 21, No. 16 3649 A,
1 kb
.jrd BstEII BamHI BamHI PvuII
P2 |_P1
BamHI
SphI
NotI BamHI
PvuII
II
A
BstEII
P1
-
T
1
hrdD 2
3
4
5
P2
-
T
1
2
3
4
5
BstEII xt.44
hrdA
1IA
probe 1 *
G G
BamHI
A G C
NarI
AvaII
SmaI
A
3 5' T C C e _ A T C G
i.
BstEII
.
i
BamNI
SXmnI ll
T A
.1 .5'3'
KpnI
P2
hrdB
C G A T
a
C G
SalI
G C T A C G C G
1CC GG
U.
Is
f
53'3
13,
probe 3 *
AvaII
hzd1B -i, I
.i,,hdR - P2
probe 2 *
T
A
AvaII
1
2
A
T
1
2
_q Al.,
StyI A T
:l II Ir
f wI
Sal I
j;
OarP21
hrdD
~~~~b~~~CG
17.F
*
A T T A T A ~~~~CG ]T A C G G A T
iC
C G
II.
probe 4
,f t
xIr.Sp
5'3'
StyI
Figure
Figure 1. Restriction maps of the S.aureofaciens hrdA4, hrdB and hrdD genes and strategy for transcript mapping. Filled boxes indicate ORFs of the hrd genes. Bent arrows represent direction of the apparent transcripts from P1 and P2 promoters. The lines bellow the maps represent DNA fragments labelled at the end marked with an asterisk that were used as probes in S1 nuclease mapping. Relevant restriction sites in S1 mapping and cloning are indicated.
RESULTS Expression of the hrd genes in liquid cultures Transcription analysis of the S.aureofaciens hrd
genes by Northern blot hybridization indicated that only hrdB and hrdD are efficiently expressed in cells growing in liquid medium. No transcripts of hrdE could be identified under these conditions and only a very weak signal was identified in the case of hrdA (9). The estimated size of messages corresponded to the deduced open reading frames (ORF), suggesting monocistronic organization of the hrdA, hrdB and hrdD genes (9). To characterize the expression of the hrd genes in more detail, S1 nuclease mapping of apparent promoters corresponding to the identified messages was performed. Similarly to the initial Northern blot experiments, the RNA used was isolated from S.aureofaciens grown to the late exponential phase in minimal medium NMP (11). The probes employed are detailed in Fig. 1. Initially, titration of S1 nuclease with the probe for hrdD was performed to established optimum digestion of the protected heteroduplex (Fig. 2A). 100 units of S1 nuclease were found optimum and this amount of the enzyme was used in all subsequent experiments under the condition described in Material and Methods. Multiple weaker bands present in several of the Si-protection experiments might arise from digestion of SI nuclease into heteroduplex. In the S1
2. High-resolution SI nuclease mapping of the hrdB and hrdD apparent
promoter regions. The RNA was from liquid culture of S.aureofaciens. The details
given in Material and Methods. A, Titration of SI nuclease using probe 4 for the hrdD gene (Fig. 1). Tracks A and T are G+A and T+C Maxam and Gilbert sequencing ladders. 40 jig RNA was hybridized with the hrdD probe and treated with 100 U (lane 1), 200 U (ane 2), 400 U (lane 3) and 600 U (ane 4) of S1 nuclease (Sigma). 40 ;tg tRNA was used as control and treated with 100 U of SI nuclease (lane 5). B, SI nuclease mapping of the hrdB apparent promoter region using probes 2 and 3 (Fig. 1) and 100 U of SI nuclease. The end-labelled DNA fragments were hybridized with 40 jig RNA isolated from liquid culture of S.aureofwiens (ane 1) and with 40 jg of tRNA (ane 2). Thin horizontal arrows indicate the position of RNA-protected fragment and thick bent vertical arrows indicate the nucleotide corresponding to apparent tsp after substraction of 1.5 nt. are
mapping, two RNA protected fragments were identified. The shorter RNA-protected fragment corresponded to a more downstream apparent promoter termed hrdD-Pl with transcription start point (tsp) at T, 72 base pairs (bp) upstream from the most likely translation initiation codon. The larger fragment corresponded to a more upstream apparent promoter termed hrdD-P2 with KlKtsp at G, 332 bp upstream from the most likely translation initiation codon (Fig. 4). For hrdA, no RNA-protected fragment was identified under the conditions used (data not shown). Initial S1 mapping experiments for hrdB, using the probe 5'-labelled at the first NarI site in the coding region (Fig. 1), revealed two RNA-protected fragments. They were located too far from the labelled end to enable precise localization of tsp. Therefore, separated probes for each of the apparent promoter were employed (Fig. 1). Using these probes, two apparent tsps were identified. The one located more upstream, started at A, 161 bp upsteam from the translation initiation codon of hrdB ORF (Fig. 4) which has the N-terminal sequence identical to that of o66 from S. coelicolor (7). The corresponding apparent
3650 Nucleic Acids Research, 1993, Vol. 21, No. 16 hrdA A
1.
I:
A
4
p
A
C' G CG G r G C
C AC G
;>t3
4&.r
181 241
I-4
TAACCAICCGGTCACGCACCGGACTCAGAAAGTGCTCGCACGCCCCTCGTGGCGGCGATG
BstEII
_
1- - -P U
;;1w
.1
_.
Pl r-
301
TGTCGTACCCCACACTGAGGTGCGGTGCTTGAGTCCTCGCCAGCCCTGGAGGTCGCCCCC
1 361
V Q T Q T L T V N V S R P V S D T A G A GTGCAGACCCAGACCCTGACCGTGAACGTGAGCCGGCCCGTCAGCGACACCGCCGGAGCG
120
GATTCGGGCCACGAAGATTGGGCGTAACGCTCCGCGACGTCACGCGATGACGTAAGAGGT
_
"
-
hrdB
V..
II4
0
dP
-
GACATGATGTCGATCATCTCGGACATGCGTCCGATCCGCATCGGGGTTCCCGGCGGGGGG
.w!w
*dP
400
;!
4F
t
S '
1
P2
A r
0
T
P2
180
GATGGCCGACGGAGGGAAGACAGCAGCCCTTGGGGGTGCTGTGCAGCTCCCCGGCCCCTG
240
CCCGCGCCCCCGGCCCATCCCCAGTCGGTGGTCGTCGGCTCCGGTCAGCACCAGGCCGG
1 300
GGTCGGAAGCCGTTCCCATCGTTCCGAGAGGTTGTTCGTGTCGGCCAGCACATCCCGTAC
9 360
XmnI L P P E I A D S E S V M A L I E R G K A GCTCCCGCCGGAGATCGCCGATTCCGAGTCTGTGATGGCGCTCATCGAGCGGGGCAAGGC
29 420
E G Q I A G D D V R R A F E A D Q I P A CGAGGGGCAGATCGCCGGCGATGACGTGCGTCGGGCCTTCGAGGCTGACCAGATTCCTGC
49 480
T Q W K N V LRS L N Q I L E E E G V T GACCCAGTGGAAGAATGTTCTGCGCAGCCTCAACCAGATCCTTGAGGAAGAGGGTGTGAC FspI
69 540
ACTGATGGTCAGTGCCGCGGAGTCGCCCAGGCGCACCCCCCGCAAGAGCGTCGCGGCGAA
240
GGGCGAAACATCCCACTTGTGACGGATTCGGGCATTCGATCGCCAGTCCAGGCAACCGGG
300
GGCCGGTGCGGACCGTCTTCAGGGGTGGAAGTGCGGCCGGGAGCGGTGCCGCGTCTGGGT
_
-b *
(.;C A T C'G T A C G
C (; A T C G C,c C,G
.NW
..._
or T
I
;2
3
S
AvaII A S
T
S
R
T
P1 -
S 3'
A
V
A
4
T
2
L
M
V
S
A
A
E
S
P
R
R
T
P
R
K
S
V
A
A
K
:1 4
hrdD
Wo A 'IT
P2
J
1.
*,~~ G
_.11
~~
S.
A C G
360
GAAAAGCGCCCCACGGATGTGACGGCGGTCTCACGGGGGCGGTGGCCGGGCCGGCCGGCG
C
420
AGGGTGTGCGCGATGGCCTTCGTCTCCGCTCTGACCTGCGCATCCACCGGCTCTTGCGAG
Cc~~A
480
GGTTCTGTCAGCTCCGTGTAAAGAGTGGCGGGGCAGGATACTGCAAGATCCGAAGAACA
._:
;*
Ge A IT
A r
b
BglII P1 r
c G
.,;. aF *.
C: G 1 h
I-'
1C G | ~~~~CG
540
GGGACCGACTGGGAATTTGTCCCGATTCCAGTCGTTGTTTCCTTCGGAAGCGGGCACT EcoRI
1
Flgure 3. High-resolution SI nuclease mapping of apparent tsp for the S.aureofaiens hrdA, hrdB and hrdD) genes with RNA isolated from surface culture of S.aureofaciens. Details are described in Material and Methods. Tracks A and T are G+A and T+C Maxam and Gilbert sequencing ladders. 5'-labelled DNA fragments (Fig. 1) hybridized with 40 sg RNA from 12 h (lane 1), 19 h (lane 2) and 36 h (lane 3) from surface culture of S.aureofaciens and with 40 isg tRNA Oane 4) and treated with 100 U of Si nuclease. Thin horizontal arrows indicate the position of RNA-protected fragment and thick bent vertical arrows indicate the nucleotide corresponding to apparent tsp after substraction of 1.5 nt. All of the SI nuclease mapping experiments were performed three times with independent sets of RNA samples. In every experiment, the same RNA preparations were hybridized parallelly with all of the hrd probes.
promoter was termed hrdB-P2. The more downstream apparent tsp at G, corresponding to apparent promoter hrdB-Pl, was
located withiii the hrdB ORF, 165 bp downstream from the translation initiation codon. Another possible translation initiation codon ATG, preceded by potential ribosome binding site (RBS), is located 43 bp downstream of this apparent promoter (Fig. 4). This part of sequence is almost identical with hrdB of S. coelicolor and it was initially assumed to be the first translation initiation site of S. coelicolor hrdB ORF (6).
Expression of the hrd genes in surface cultures at different developmental stages Since the principal sigma factors are essential for the expression of 'house-keeping' genes, the hrd gene products may direct expression of the genes under various physiological states during development. To determine whether the expression of the hrd genes is developmentally regulated, high-resolution SI nuclease mapping was performed using RNA isolated from surface-grown cells both before and during the aerial mycelium formation.
M
600
CGGGAGAGTGTCCGAGGACGCACCGCGACCGACTCACCGCATCGCAAGGGrAGCACGCATG
2 660
A T R A V A R R Q S T S S A R A V G G E GCAACCCGCGCCGTCGCCCGTCGTCAGTCCACGAGCAGCGCCCGCGCTGTGGGCGGGGAG
Figure 4. Nucleotide sequence around the region of tsp of S.aureofaciens hrdA, hrdB and hrdD genes. Bent arrows indicate the likely position of tsp and the direction of transcription. The nucleotides in tsp and translation initiation codons are bolded. Presumptive RBS are indicate by dots. The one-letter amino acids code is used in the second position of each codon. Relevant restriction sites are underlined. The numbers correspond to the nt positions which refer to the published sequences of hrd genes (9). The GenBank accession No. is M90410 for hrd4, M90411 for hrdB, and M90413 for hrdD.
Surface-grown cells of S.aureofaciens were harvested 13, 19 and 36 h after inoculation of spores. The intervals correspond to the substrate mycelium, the beginning of aerial mycelium formation and to the aerial mycelium about 5 h before pigmentation characteristic for spore maturation appeared, respectively. At 36 h, the aerial mycelium, together with the rest of substrate mycelium, was scraped and used for RNA isolation. For hrdE, no RNA-protected fragment was identified (data not shown). 5'-labelled probes used for hrdA, hrdB and hrdD are indicated in Fig. 1. In the case of hrdB, RNA-protected fragments were identified with all three RNAs at the same position as in the RNA isolated from liquid culture (Fig. 3, lanes 1,2,3). This data indicate a fairly constant transcription initiated in the same tsps of both hrdB-P1 and hrdB-P2 apparent promoters throughout the developmental cycle of S.aureofaciens. For hrdD, RNA-protected fragments were also identified at the same position as with the RNA from liquid culture corresponding to apparent promoters hrdD-Pl and hrdD-P2, but only in the vegetative stage -substrate mycelium (Fig. 3, lane
Nucleic Acids Research, 1993, Vol. 21, No. 16 3651 1). Using RNA isolated from the early (19 h) or from late (36 h) aerial mycelium, no RNA-protected fragment was identified even after prolonged exposition of autoradiogram (Fig. 3, lanes
2,3). For hrdA4, similar to the results with the RNA isolated from liquid culture, no RNA-protected fragment was identified with the RNA isolated from substrate mycelium (13 h) (Fig. 3, lane 1). However, with the RNA from the beginning of aerial mycelium, an RNA-protected fragment was identified (Fig. 3, lane 2). The fragment corresponded to an apparent promoter termed hrdA-P2 with tsp at A, 107 bp upstream from the most likely translation initiation codon (Fig. 4). The activity of the apparent promoter increased more then 10 times with the RNA isolated from late aerial mycelium (Fig. 3, lane 3). With this RNA, second shorter RNA-protected fragment also appeared. The fragment corresponded to a more downstream apparent promoter termed hrd4-Pl with KiKtsp at T or G, 40 or 39 bp upstream from the most likely translation initiation codon (Fig. 4). In all of the experiments, no RNA-protected fragment was identified at these positions with tRNA as control (Fig. 3, lanes 4).
Characterization of the hrdA, hrdB and hrdD promoters of
S.aureofaciens The sequences of the apparent hrdA, hrdB and hrdD promoters are shown in Fig. 4. Based on the sequence around -10 and -35 positions upstream of apparent tsp, all identified apparent promoters belong to the group of streptomyces promoters that do not display typical -10 and -35 regions characteristic of promoters recognized by the major form of RNA polymerase (17). Sequence comparison of the apparent S.aureofaciens hrdB-
P2, hrdD-PL, and hrdD-P2 promoters with previously identified S. coelicolor hrdB-P, hrdD-P1, and hrdD-P2 promoters (4) revealed extensive similarities. However, alignment of the whole upstream regions of the S.aureofaciens and S. coelicolor hrdB and hrdD genes showed that the similarities persist over regions longer than promoters occupy. Comparison of the DNA sequences upstream of the apparent tsp with 139 compiled promoter regions from Streptomyces spp. (17) revealed no significant similarity to any of the apparent hrd promoters.
Activity of the hrd promoters in vivo To support the results of the transcriptional analysis, in vivo promoter activity was examined, using promoter-probe vectors pIJ486 and pARCI (18, 19). DNA fragments bearing apparent promoters were inserted to the vectors as indicated in Material and Methods. The resulting plasmid, containing fragment with the hrdB-Pl apparent promoter (pUJ-hrdBPl) conferred Neo resistance on Streptonyces lividans TK24 to 15 gig/ml, pUhrdBP2 with the hrdB-P2 apparent promoter to 100 ,ug/ml Neo, pUJ-hrdDPl with the hrdD-Pl to 20 jig/ml Neo, and pLU-hrdDP2 with the hrdD-P2 to 100 ,tg/ml Neo (Tab. 1). Plasmid pUhrdAP12, bearing DNA fragment with the both hrd4-P1 and hrd4-P2 apparent promoters caused resistance to less then 5 ,g/ml Neo. S. lividans TK24 with pUJ486 was unable to grow on the medium with 5 ltg/ml Neo. The inability of the hrd4 promoters fragment to confer Neo-resistance on S. lividans TK24 spores when cloned into pIJ486 is not surprising given that the promoters are not expressed until aerial hyphae are formed. The hrd4, hrdB and hrdD promoters-bearing fragments were cloned to promoter-probe vector pARCl (Tab. 1), where it is
possible to see the expression in various developmental stages. Expression of brown marker pigment occurred in substrate mycelium when introduced plasmids pARC-hrdB and pARChrdD (Tab. 1), containing hrdB and hrdD apparent promoters, into S.lividans TK24. The cells with pARC-hrdA (Tab. 1), containing DNA fragment with the hrd4 promoters, started to produce the marker brown pigment after the aerial mycelium had appeared. These results are in line with the results of transcriptional experiments. They suggest that the in vivo activity of all of the identified hrd apparent promoters is in the same correlation with the development as in S1 mapping experiments.
DISCUSSION The experiments described above show the presence of at least two tandem promoters directing expression of the S.aureofaciens
hrdA, hrdB and hrdD genes. In each case, the more downstream promoter is much weaker then the one more upstream. Si nuclease mapping and in vivo promoter-probing show that expression of the hrdA, hrdB and hrdD genes correlates differently with a developmental stage. While hrdB has been showvn to be expressed at the transcriptional level in all of the developmental stages tested, the hrdD appears to be expressed only in vegetative stage-substrate mycelium. On the other hand, expression of hrd4 has been proven to be initiated at the beginming of aerial mycelium formation. This expression pattern suggests a distinct function of protein products of the hrd genes in respective developmental stages. The hrdB gene has been suggested to encode principal sigma factor in S. coelicolor and S.aureofaciens (4, 7, 9). Therefore, it is not surprising that its expression continues also in aerial mycelium to direct the expression of 'house-keeping' genes. The function of hrdD appears to be restricted to substrate mycelium and its protein product appears to be not required in other developmental stage. The function of hrdD is unclear at present. Gene disruption experiments of hrdD in S. coelicolor did not reveal any significant phenotypic changes (4, 8). Based on the data presented here, a likely sigma factor encoded by the hrdA gene is probably involved in some process occurring during aerial mycelium formation or sporogenesis. However, disruption of hrdA in S. coelicolor and S.aureofaciens did not reveal any significant phenotypic alterations and the disrupted strains differentiate comparably with the wild type strains (4, 8; J.Kormanec-unpublished results). The important point will be to prove, whether the temporal expression of hrdA correlates with the spatial location of the gene product in aerial mycelium. SI-protection experiments with RNA isolated from scraped aerial mycelium from 36 h culture support this assumption, but we can not rule out possible expression in a contaminating substrate mycelium. Verification of the spatial location using antibodies against the overexpressed part of the hrdA gene product in E. coli is in progress. If the hrdA gene product is expressed in aerial mycelium, the question will arise why it is not necessary for development. Possible explanation of this obscurity may be that the hrdA gene is expressed in the substrate mycelium in the time of aerial mycelium formation and may have some function in this part of colony. Identification of the genes recognized by this hrd4-encoded sigma factor could help to answer the question. A situation similar to the expression of the S. aureofaciens hrd4 gene has been recently observed inAnabaena. Two new putative sigma factor genes, sigB and sigC, exhibiting high degrees of similarity to principal sigma factors, were identified. Expression
3652 Nucleic Acids Research, 1993, Vol. 21, No. 16 of the genes was detected only under nitrogen-limiting conditions that lead to the Anabaena differentiation. However, disruption of the genes has no obvious effect on differentiation (20). Presence of internal apparent promoter hrdB-Pi in the S.aureofaciens hrdB gene is also of interest. This apparent promoter was not identified in hrdB of S. coelicolor, since the S1 mapping probe used was labelled at a site upstream of the putative tsp (4). The faint band identified by Sl-mapping does not seem to be an artefact caused by a secondary structure, since it was not present in the control experiment with the same fragment and tRNA. Likewise, no secondary structure is present in the region. However, we can not rule out the possibility that the apparent internal tsp results from RNA processing. The level of Neo resistance is rather low (15 ,tg/ml) and might reflect the presence of sequence that can function fortuitously as promoter. However, some features of its location suggest its possible role as a promoter. A possible ATG initiation codon in frame with the hrdB ORF, preceded by potential RBS, is located downstream of the tsp (Fig. 4). This ATG codon was initially suggested as a start codon of the hrdB ORF (6). Similarly, the analysis of the third position G + C of the hrdB ORF by FRAME program (21) preferred this ATG as a start codon. However, N-terminal amino acids sequencing of a56 located translation initiation codon more upstream in the hrdB ORF (7). The major promoter for the hrdB gene appears to be hrdB-P2 apparent promoter located upstream of the translation initiation codon GTG followed by the sequence encoding amino acids identical with those revealed in the N-terminal part of a56 by amino acids sequencing (7, 9). Internal apparent weak promoter hrdB-Pl may play a specialized role in response to particular physiological conditions. This would suggest the presence of two in-frame N-terminally different HrdB proteins that may have specialized functions. In Streptomyces, there are known two examples of two or more polypeptides encoded by the same coding sequence. Two polypeptides with different N-termini encoded by sporulation gene of Streptomyces griseus are generated by temporal control of transcript accumulation (22). Two internal promoters were identified in c repressor gene of Streptomyces phage 4C31 generating two additional N-terminally differing proteins that may have different function in repression of phage development (23). Investigation of conditions for induction of internal hrdB-Pl promoter and further experiments to prove this hrd expression at protein level are in progress. We have detected no induction of the promoter upon heat-shock induction (J.Kormanec-unpublished results). The results in this report may be the basis for further analysis of the biological role of hrd genes.
ACKNOWLEDGEMENTS We would like to thank Mrs Renata Knirschova' for excellent technical assistance, and Dr Stefan Kulela for critical reading of the manuscript. We are grateful to Prof. D.A.Hopwood for
providing us with S.lividans TK24 and plasmid pIJ486 and to Prof. S.Horinouchi for plasmid pARCl. This work was supported, in part, by Grant agency of Slovak Academy of Sciences (grant No. 999135).
REFERENCES 1. Chater, K.F. (1989) Trends Genet. 5, 372-377. 2. Schauer, A., Ranes, M., Santamaria, R., Guijarro, J., Lawlor, E., Mendez, C., Chater, K. and Losick, R. (1988) Science 240, 768-772.
3. Chater, K.F., Bruton, C.J., Plaskitt, K.A., Buttner, M.J., Mendez, C. and Helmann, J.D. (1989) Cell 59, 133-143. 4. Buttner, M.J., Chater, K.F. and Bibb, M.J. (1990) J. Bacteriol. 172,
3367-3378.
5. Tanaka, K., Shiina,, T. and Takahashi, H. (1992) Mol. Gen. Genet. 229, 334-340. 6. Shiina, T., Tanaka, K. and Takahashi, H. (1991) Gene 107, 145-148. 7. Brown, K.L., Wood, S. and Buttner, M.J. (1992) Mol. Microbiol. 6,
1133-1139.
8. Buttner, M.J. and Lewis, C.G. (1992) J. Bacteriol. 174, 5165-5167. 9. Kormanec, J., Farkasovsky, M. and Potd&kovd, L. (1992) Gene 122, 63-70. 10. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.O., Seidman, J.S., Smith, J.A. and Struhl, K. (1987) Current Protocols in Molecular Biology. Wiley, New York, 11. Hopwood, D.A., Bibb, M.J., Chater, K.F., Kieser, T., Bruton, C.J., Kieser, H.M., Lydiate, D.L. Smith, C.P., Ward, J.M. and Schrempf, H. (1985) Genetic Manipulaion of Streptomyces. A Laboratory Manual, Norwich: The John Innes Foundation. 12. Horinouchi, S., Hara, 0. and Beppu, T. (1983) J. Bacteriol. 155, 1238- 1248. 13. Farkagovsk9, M., Kormanec, J. and KollArva, M. (1991) Biochem. Biophys. Acta 1088, 119-126. 14. Kormanec, J., Schaaff-Gerstenschlager, I., Zimmermann, F.K., Perecko, D. and Kiintzel, H. (1991) Mol. Gen. Genet. 230, 277-287. 15. Murray, M.G. (1986) Anal. Biochem. 158, 165-170. 16. Maxam, A. and Gilbert, W. (1980) Methods Enzymol. 65, 499-560. 17. Strohl, W.R. (1992) Nucleic Acids Res. 20, 961-974. 18. Ward, J.M., Janssen, G.R., Kieser, T., Bibb, M.J., Buttner, M.J. and Bibb, M.J. (1986) Mol. Gen. Genet. 203, 468-475. 19. Horinouchi, S. and Beppu, T. (1985) J. Bacteriol. 162, 406-412. 20. Brahamsha, B. and Haselkom, R. (1992) J. Bacteriol. 174, 7273-7282. 21. Bibb, M.J., Findlay, P.R. and Johnson, M.W. (1984) Gene 30, 157-166. 22. Babcock, M.J. and Kendrick, K.E. (1990) Gene 95, 57-63. 23. Smith, M.C.M. and Owen, C.E. (1991) Molec. Microbiol. 5,2833 -2844. 24. Hopwood, D.A., Kieser, T., Wright, H.M. and Bibb, M.J. (1983) J. Gen. Microbiol. 129, 2257-2269.
