- Email: [email protected]
regulatory region
Gene, 129 (1993) 113-117 0 1993 Elsevier Science Publishers B.V. All rights reserved. 0378-I 119/93/%06.00
113
GENE 07147
Pseudomonas aeruginosa 1asA gene: determination
point and analysis of the promoter/regulatory
of the transcription start region
(Elastase; LasR; cystic fibrosis; virulence; pseudoknots; protease)
Lynn C. Freck-O’Donnell
and Aldis Darzins
Department of Microbiology, The Ohio State University, Columbus, OH 43210, USA
Received by M. Bagdasarian: 15 January 1993; Accepted: 19 February 1993; Received at publishers: 11 March 1993
SUMMARY
The elastolytic activity of the opportunistic pathogen Pseudomonas aeruginosa is due to the combined activities of at least three secreted proteins: elastase, LasA and alkaline protease. Transcription of both the lasA gene and the elastase structural gene, la@ requires the transcriptional activator LasR. In order to localize the promoter elements involved in lasA expression, the transcription start point (tsp) for lasA was localized by Sl protection and primer extension analysis. The DNA sequence of the region upstream from the tsp was determined, and a putative 070 promoter was identified. Sequence comparison with the 1asB promoter region revealed two areas of considerable homology which could act as potential binding sites for LasR or other, as yet unidentified, regulatory proteins.
INTRODUCTION
Pseudomonas aeruginosa is an opportunistic pathogen that is able to cause severe and often fatal infections in compromised hosts, particularly burn and cystic fibrosis patients. The pathogenesis of P. aeruginosa infections is due in part to the secretion of a large number of virulence factors including exotoxin A and elastase. Elastase has been shown to inactivate several biologically important molecules such as IgG (Holder and Wheeler, 1984), IgA (Diiring et al., 1981), human cl,-proteinase inhibitor Correspondence to: Dr. A. Darzins, Department of Microbiology, The Ohio State University, 484 West 12th Avenue, Columbus, OH 43210, USA. Tel. (614) 292-0147; Fax (614) 292-1538.
Abbreviations: aa, amino acid(s); bp, base pair(s); CF, cystic fibrosis; cpm, counts per minute; E., Escherichia; Ig, immunoglobulin; LasA, protein involved in the degradation of elastin; lasA, gene encoding LasA, lasB, gene encoding elastase; LasR, transcriptional activator of IasA and lasB, lasR, gene encoding Las&, LB, Luria-Bertani (medium); nt, nucleotide(s); oligo, oligodeoxyribonucleotide; p, plasmid; P., Pseudomonas; PAGE, polyacrylamide gel electrophoresis; PI, isoelectric point; R, C or T, RBS, ribosome-binding site(s); tsp, transcription start point(s); u, unit(s); V., Vibrio.
(Morihara et al., 1979), and complement components (Schultz and Miller, 1974) and to degrade such structural proteins as collagen (Heck et al., 1986) and elastin (Morihara and Homma, 1985). The elastolytic activity of P. aeruginosa elastase is enhanced by the presence of a second secreted protein, LasA (Peters and Galloway, 1990). It has been shown that LasA degrades elastin in the absence of elastase (Peters et al., 1992; Toder et al., 1991), thereby classifying LasA as a second elastase produced by P. aeruginosa. LasA-enhanced elastolysis is due to separate activities of LasA and elastase on elastin and is not due to the activation of elastase by LasA (Peters et al., 1992). LasA is emerging as an important virulence factor due to its ability to enhance the elastolytic activities of both P. aeruginosa elastase and human neutrophil elastase (Peters et al., 1992) and may be a contributing factor to the destruction of connective tissue seen in the CF lung by Bruce et al. (1985). Thus, studies on the regulation of lasA will provide a better understanding of the environmental conditions that promote LasA-enhanced elastolysis. A positive regulator, designated LasR, has been identified and shown to have significant homology to
114
LuxR of Vibrio jscheri (Gambello and Iglewski, 1991). An intact 1asR gene is required for transcription of both 1asB (Gambello and Iglewski, 1991) and lasA (Toder et al., 1991). All three Zas genes have been mapped to distinct regions of the P. aeruginosa PA0 chromosome (Howe et al., 1983; Holloway and Zhang, 1990; Gambello and Iglewski, 1991). It is likely, therefore, that LasR alone, a LasR-protein complex, or another regulatory protein that is controlled at the transcriptional level by LasR activates transcription in trans by binding upstream from the lasA and 1asB promoters. The aim of the present study was the localization of the tsp for EasA and the analysis of the DNA sequence upstream from this site. This sequence was examined for putative promoter elements and used in comparison studies with the ZasB promoter region.
EXPERIMENTAL
AND DISCUSSION
(a) Identification of the tsp of 1asA Determining the tsp of a particular gene is often simpli-
fied by knowing the N-terminal aa of the gene product. The N terminus of the precursor form of LasA, however, has not been determined experimentally. Two different start codons separated by ca. 100 bp have been proposed based on independent determinations of sequence, size and pI of purified, mature LasA and on the size of the precursor form produced in E. coli. For this reason the tsp was first localized by low resolution Sl analysis (Freck et al., 1992) to a position ca. 110 nt upstream from the proposed ATG start of Schad and Iglewski (1988) and ca. 250 nt upstream from the proposed TTG start codon proposed by Darzins et al. (1990). This region is ca. 35 bp downstream from the unique SmaI site located near the beginning of the published lasA sequences. To avoid ambiguity, tsp and other reaction products have been localized with respect to this site rather than to potential start codons. From this information, an oligo primer (LA390, 5’-CGAAGGGAAACCTTGAAAGC) complementary to the sense strand ca. 75 bp from the 5’ end of the message was synthesized for use in high resolution Sl analysis. A plasmid containing the initiation region served as the template for extension from LA390 to produce a continuously labelled probe for Sl studies. The results from Sl analysis of PA01 RNA with the lasAspecific probe are shown in Fig. 1A. The two largest Slprotected fragments indicated by arrowheads (lanes 1 and 2) correspond to two 5’ ends of the message that are located 29 bp and 31 bp downstream from the SmaI site. The bracket in Fig. 1A denotes a ladder of smaller Slprotected fragments with ends ca. 47 bp downstream from the SmaI site. Multiple attempts at primer extension analysis using
LA390 were unsuccessful. Analysis of the region within 100 nt of the largest Sl-protected fragment revealed the presence of possible secondary and tertiary structures that could have interfered with primer extension studies. Tertiary structures due to interactions between loops and single-stranded regions have been described (Studnicka et al., 1978). These structures, known as pseudoknots, are hairpin loops with a single strand folded back to pair with the nt in the loop and generally involve 4 or fewer nt such that the single strand does not pass completely through the loop. This structure is stabilized by stacking of the two double helical segments (Pleij et al., 1985). The secondary structure (stem-loop) shown in Fig. 2 has a favorable free energy (- 10 kcal). This alone, however, is probably not enough to prevent reverse transcriptase from extending from LA390 through the region. Analysis of single-stranded regions for the ability to form a pseudoknot with the loop revealed 4 nt, 5’-CUGG (+16 to + 19), able to base pair with the sequence 5’-CCAG (+ 41 to +44) found in the loop. The 5’ ends of the smaller Slprotected fragments (Fig. IA) are located in the + 16 to + 19 region; therefore, it is possible that pseudoknot formation leaves the Sl probe susceptible to digestion in that region. The presence of this pseudoknot has not, however, been confirmed experimentally. Primer-extension analysis using a second oligo primer (LA359,5’-CGACATCGCCTCCGACGGG) that should destabilize these structures on the RNA confirmed the predicted tsp. The two major extended products indicated by arrows (Fig. lB, lane 1) correspond exactly to the 5’ ends of the two larger protected fragments determined by Sl mapping (Fig. 1A). No additional products were seen that mapped to the same region as the smaller Slprotected fragments. Thus, it was determined that transcription of lasA is initiated at two different sites separated by only 1 nt. (h) The nt sequence of the fasA promoter region and identification of a 0” promoter
Since the SmaI site is located at ca. -30 with respect to the tsp of lasA, a 0” promoter, if present, should span the SmaI site. In addition, transcriptional activators such as LasR would be expected to bind upstream from the promoter between -200 and -20 (Collado-Vides et al., 1991). Therefore, the nt sequence upstream from the lasA tsp (ca. 300 bp) was determined and is shown in Fig. 3A. The two previously reported sequences of 1asA (Schad and Iglewski, 1988; Darzins et al., 1990) overlap with the sequence presented here as described in the legend to Fig. 3A. A comparison of the sequence immediately upstream from the SmaI site with the sequence of Schad and Iglewski (1988), however, revealed 3 additional bp not present in the original 1asA sequence.
115
A
A
C
G
B
123
T
G C A
C
G
T
1
2
T C C \iT G
G
E G T\
A
i
=
A C C C G
1
G C C
I
:
E G T G C A
iA
L G G
Fig. 1. Determination of the tsp of 1asA. (A) Sl nuclease analysis. RNA was prepared from 50-ml cultures of P. aeruginosa PA01 grown 12 h in LB by a guanidine isothiocyanate-hot phenol method (Rothmel et al., 1991). Sequencing of the antisense strand was performed using 10 uCi [a-“P]dCTP and SequenaseTM (US Biochemical, Cleveland, OH, USA) by the dideoxy chain-termination method (Sanger et al., 1977) as recommended by the manufacturer. The Sl probe was synthesized by modifying the sequencing reaction to allow polymerization of a high-molecular-weight probe from a plasmid template (pLO593, lusA SmaI-XhoI subclone in pUC18) and an oligo primer (LA390, see Fig. 3A). RNA (50 ug, 100 ug, and 0 ug; lanes 1, 2, and 3, respectively) was hybridized with 10’ cpm of probe overnight at 65°C and digested with 100 u of Sl nuclease (Bethesda Research Laboratories, Bethesda, MD, USA) as described previously (Dixon, 1984). Digestion products were analyzed by 8M urea-containing 6% PAGE adjacent to a reference sequence produced from pL0593 and LA390 (lanes A, C, G, T). Arrowheads and underlined nt indicate the two 5’ ends of the transcript (tsp),and the bracket and underlined nt mark an area protected from Sl nuclease due to secondary structure (see section a). (B) Primer extension analysis. RNA (50 ug and 0 ug; lanes 1 and 2, respectively) was analyzed using a primer extension system from Promega (Madison, WI, USA) and the oligo LA359 (see Fig. 3A). Products were analyzed as in A except that the sequencing primer was LA359 (lanes A, C, G, T). Arrowheads and underlined nt indicate two tsp (as in A).
The region upstream from the tsp of 1asA was examined for potential o”-type promoter sequences by comparison to other Pseudomonas promoters. Deretic et al. (1989) A&!Z G G
i C: :G G: :C
A::U A u I G::C C G::C C::G GoU G::C C::G C::G AGGCGAUGUCGCCGGGCUGCUGGC... GGCGUGCAACCCGGCEYEEj;CGAAAUA 3' 5' pneudoknot
intar8ctions:
5'CUGG 3' 3’GACC 5'
Fig. 2. Possible secondary structure of 1asA mRNA. The sequence corresponding to the first 100 nt of the 1asA transcript was analyzed for secondary structure by the method of Jaeger et al. (1989). Watson-Crick base pairing is denoted by ::, while GoU indicates a non-Watson-Crick base pair. The doubly underlined nt are complementary to the singly underlined nt in the loop structure. The possible interactions between these two regions to form a pseudoknot are shown below the secondary structure.
classified twelve Pseudomonas promoters as 0” (rpoD) promoters. From these promoters we derived a consensus sequence which is as follows: TTGACR (- 35 region) and TATRRT (-10 region). The-most invariant nt are underlined. The region upstream from the tsp of lasA was found to contain sequences homologous to the consensus promoter (Fig. 3A). We also compared the lasA sequence to a recent compilation of 29 cr” promoters from streptomycetes (Strohl, 1992), which, like pseudomonads, contain DNA with a high G+C content. The results of this analysis also identified the -10 and -35 regions shown in Fig. 3A.
(c) Comparison of fusA and IasB promoter regions The lasB promoter has been classified as a P. aeruginow virulence promoter by Deretic et al. (1989). This designation was based on the lack of homology to other known promoters and the role of this gene in pathogenesis. The lasB promoter region also displays limited homology to some members of the os4 class of promoters (Deretic et al., 1989). However, it has been shown that a os4 (rpoN) mutant of P. aeruginosa strain PAK produces
116
A . . . HinfI GAATCGCCCGCGGACAAGCGCCTGAGGGTCGTGCATGGCCGAGATCGGAGCCGGCCAC . . .lOO GGCCTCGACGAGGCCGAGCTGGCGCGTCGTGATCCGCTCGACGGTTCCTCCGACGAGG~ . . HhlfI' XmJlI CGCTGAGCGCGCAGACTGTATCGAAGTATTTTCCGACGCCTGGCGTTCTGTGATCGATTCG . Sau3AI. 9200 GCTCGGTTGGAGTGCGCTGCCGCACGCTTTACAGCCGGACGGAGCGGGGCACCGGATCTA . . +1 0300 SmaI CTGCAMAGCTGATAG~TCCCGGGTGACGGCGTTG~CCTGCfiG~GTGCAA . n * -35 -10 . . . CCCGGCCTGGGCGMATACCGGCGGAGCGGACCAGGCTAT CCCGTCC.C.AC.GCGATGTCGC LA359 . . '400 CGGGCTGCTGGCTTTCAAC.GTTTCCCTTCGATGACCBGG LA390 RBS Sau3AI GATC
t&t
B .
. 1asA
:::: :: ::: : ::::
CTGTGATCGATTCGGCTCGGTTGGAGTGC---““‘:
.. .. .. .. ..
:
:
::
CTGAGCGCGTCCCGGAGC--TGGGGGCAACCTAd .
.
.
.
+GCTGATA$&ZT CGGACGGAGCGGGGCAC----::::::::::: TA&TATTC~A&&,ACAT
.
m--TCAAGGCTACCTG .
::
::
::
.
-10 +1 SmaI . CCCGGG-TGACGGCGTTG!XUi!iTCCTGCGGCGTGCAACCCG----GCCTGGG :: :: :: ::: :::: ::: :: ACACGAAAAGCA&T CCAGTTCTGGCAGGTTT-GCGGG;T;TTTT;;GT . . -10 t1 -35
Fig. 3. Sequence of the promoter region of 1asA and comparison to lasB. (A) Sequence of the 1asA promoter region. Both strands of DNA upstream from the tsp of lasA were sequenced as described in Fig. 1 with the exception that [c+%]dCTP was used for labelling. In order to present the overlapping sequence and to keep the promoter intact, portions of this sequence have been published previously [bp 247-424 (Schad and Iglewski, 1988);bp 264424 (Darzins et al., 1990)]. Additional nt not present in the previously published sequence (Schad and Iglewski, 1988) are marked with carets. The sequence is marked with a dot every 20 nt, starting with the first one of the sequence. Restriction sites used for the generation of overlapping subclones are labelled. Important features in the sequence are underlined. The sequences labelled LA359 and LA390 are complementary to the two primers used for transcript mapping. The tsp are shown in bold type and are marked + 1.Putative promoter sequences are labelled - 10 and - 35. One of the potential start codons is marked (Met) along with the putative RBS (Schad and Iglewski, 1988). The sequence data have been deposited with the GenBank data library under accession No. L0849. (B) Comparison with the lush regulatory region. The sequences upstream from the tsp of 1asE (Bever and Iglewski, 1988; Mohr et al., 1990) and lasA were aligned to produce the best homology within 150 bp from +l. The sequence is marked with a dot every 20 nt back from +I. Gaps introduced to obtain the best fit are shown as dashes. Matches between the two sequences are denoted by colons. Regions of considerable homology are doubly underlined and labelled I and II. The tsp are underlined and labelled +l. Putative promoters are underlined and labelled -10 and -35. The SmaI site of 1asA is also shown.
parental-like levels of elastase (Totten et al., 1990). Therefore, the transcription of 1asB is probably not initiated by a RpoN-containing RNA polymerase. Since a putative 07’-type promoter was found for la&, we examined the 1asB promoter region and also found putative -10 and -35 sequences (Fig. 3B). Sequences upstream from the LasR-regulated IasA and 1asB promoters may contain homologous regions to which regulatory protein complexes could bind. The promoter region of 1asA was aligned with the sequence upstream from the 1asB tsp. Two regions of considerable homology between the aligned sequences were found (Fig. 3B). Region I displays 75% homology and is centered around -86 of 1asA and -111 of 1asB. Region II is 69% homologous and is centered around - 51 of 1asA and -71 of 1asB. Since the C-terminal domain of LasR is homologous to the DNA-binding domain of LuxR (Gambello and Iglewski, 1991), the region surrounding the palindromic operator of the lux regulon of I/. Jischeri (Devine et al., 1988; 1989) was also aligned with the 1asA and 1asB promoter sequences. The most significant homology to the lux operator was seen just upstream from and including region I. When gaps are introduced near the center of the 20-bp palindrome to allow maximal alignment, the LuxR target site is 35% homologous to 1asB sequence and 45% homologous to 1asA sequence (data not shown). These values increase to 42% and 54%, respectively, when 4 bp located 3’ to the lux operator are included in the comparison. These sequences are currently the focus of further investigations to determine if they serve as binding sites for LasR. (d) Conclusions (1) Transcription of 1asA is initiated at two different sites separated by 1 nt. (2) The nt sequence upstream from the 1asA tsp was determined, and a putative 07’ promoter was identified. A putative a7’ promoter was also found upstream from the IasB tsp. (3) Comparison of the sequences upstream from the tsp of 1asA and 1asB revealed two regions of significant homology which may serve as potential binding sites of the transcriptional activator LasR or other unidentified regulatory proteins.
REFERENCES Bever, R.A. and Iglewski, B.H.: Molecular characterization and nucleotide sequence of the Pseudomonns aeruginosa elastase structural gene. J. Bacterial. 170 (1988) 4309-4314. Bruce, MC., Poncz, L., Klinger, J.D., Stem, R.C., Tomashefski Jr., J.F. and Dearborn, D.G.: Biochemical and pathologic evidence for pro-
117 teolytic destruction of lung connective tissue in cystic fibrosis. Am. Rev. Respir. Dis. 132 (1985) 529-535. Collado-Vides, J., Magasanik, B. and Gralla, J.D.: Control site location and transcriptional regulation in Escherichia coli. Microbial. Rev. 55 (1991) 371-394. Darzins, A., Peters, J.E. and Galloway, D.R.: Revised nucleotide sequence of the 1asA gene from Pseudomonas aeruginosa PAOl. Nucleic Acids Res. 18 (1990) 6444. Deretic, V., Konyecsni, W.M., Mohr, CD., Martin, D.W. and Hibler, N.S.: Common denominators of promoter control in Pseudomonas and other bacteria. Bio/Technology 7 (1989) 1249-1254. Devine, J.H., Countryman, C. and Baldwin, T.O.: Nucleotide sequence of the 1uxR and 1~x1 genes and structure of the primary regulatory region of the lux regulon of VibrioJischeri ATCC 7744. Biochemistry 27 (1988) 837-842. Devine, J.H., Shadel, G.S. and Baldwin, T.O.: Identification of the operator of the lux regulon from the Vibrio jischeri strain ATCC 7744. Proc. Natl. Acad. Sci. USA 86 (1989) 5688-5692. Dixon, R.A.: Tandem promoters determine regulation of the KZebsieUa pneumoniae glutamine synthetase (gInA) gene. Nucleic Acids Res. 12 (1984) 7811-7830. Doring, G., Obernesser, J. and Botzenhart, K.: Extracellular toxins of Pseudomonas aeruginosa. II. Effect of two proteases on human immunoglobulins IgG, IgA and secretory IgA. Zentralb. Bakterial. Mikrobiol. Hyg. 1 Abt. Orig. A 249 (1981) 89-98. Freck, L.C., Galloway, D.R. and Darzins, A.: Analysis of Pseudomonas aeruginosa LasA production. Abstracts of the General Meeting of the American Society for Microbiology, 1992, p. 36. Gambello, M.J. and Iglewski, B.H.: Cloning and characterization of the Pseudomonas aeruginosa 1asR gene, a transcriptional activator of elastase expression. J. Bacterial. 173 (1991) 3000-3009. Heck, L.D., Morihara, K., McRae, W.B. and Miller, E.J.: Specific cleavage of human type III and IV collagens by Pseudomonas aeruginosa elastase. Infect. Immun. 51 (1986) 115-118. Holder, LA. and Wheeler, R.: Experimental studies of the pathogenesis of infections due to Pseudomonas aeruginosa: elastase, an IgG protease. Can. J. Microbial. 30 (1984) 1118-I 124. Holloway, B.W. and Zhang, C.: Pseudomonas aeruginosa PAO. In: O’Brien, J. (Ed.), Genetic Maps. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1990, pp. 2.71-2.78. Howe, T.R., Wretlind, B. and Iglewski, B.H.: Comparison of two methods of genetic exchange in determination of the genetic locus of the structural gene for Pseudomonas aeruginosa elastase. J. Bacterial. 156 (1983) 58-61. Jaeger, J.A., Turner, D.H. and Zuker, M.: Predicting optimal and suboptimal secondary structure for RNA. Methods Enzymol. 183 (1989) 281-306.
Mohr, CD., Rust, L., Albus, A.M., Iglewski, B.H. and Deretic, V.: Expression patterns of genes encoding elastase and controlling mucoidy: co-ordinate regulation of two virulence factors in Pseudomonas aeruginosa isolates from cystic fibrosis. Mol. Microbial. 4 (1990) 2103-2110. Morihara, K. and Homma, J.Y.: Pseudomonas proteases. In: Holder, I.A. (Ed.), Bacterial Enzymes and Virulence. CRC Press, Boca Raton, FL, 1985, pp. 41-79. Morihara, K., Tsuzuki, H. and Oda, K.: Protease and elastase of Pseudomonas aeruginosa; inactivation of human plasma alphaiproteinase inhibitor. Infect. Immun. 24 (1979) 188-193. Peters, J.E. and Galloway, D.R.: Purification and characterization of an active fragment of the LasA protein from Pseudomonas aeruginosa; enhancement of elastase activity. J. Bacterial. 172 (1990) 2236-2240. Peters, J.E., Park, S.J., Darzins, A., Freck, L.C., Saulnier, J.M., Wallach, J.M. and Galloway, D.R.: Further studies on Pseudomonas aeruginosa LasA: analysis of specificity. Mol. Microbial. 6 (1992) 1155-1162. Pleij, C.W.A., Rietveld, K. and Bosch, L.: A new principle of RNA folding based on pseudoknotting. Nucleic Acids Res. 13 (1985) 1717-1731. Rothmel, R.K., Chakrabarty, A.M., Berry, A. and Darzins, A.: Genetic systems in Pseudomonas. Methods Enzymol. 204 (1991) 485-514. Sanger, F., Nicklen, S. and Coulson, A.R.: DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467. Schad, P.A. and Iglewski, B.H.: Nucleotide sequence and expression in Escherichia coli of the Pseudomonas aeruginosa 1asA gene. J. Bacterial. 170 (1988) 2784-2789. Schultz, D.R. and Miller, K.D.: Elastase of Pseudomonas aeruginosa: inactivation of complement components and complement-derived chemotactic and phagocytic factors. Infect. Immun. 10 (1974) 128-135. Strohl, W.R.: Compilation and analysis of DNA sequences associated with apparent streptomycete promoters. Nucleic Acids Res. 20 (1992) 961-974. Studnicka, G.M., Rahn, G.M., Cummings, I.W. and Salser, W.A.: Computer method for predicting the secondary structure of singlestranded RNA. Nucleic Acids Res. 9 (1978) 3365-3387. Toder, D.S., Gambello, M.J. and Iglewski, B.H.: Pseudomonas aeruginosa LasA: a second elastase under the transcriptional control of 1asR. Mol. Microbial. 5 (1991) 2003-2010. Totten, P.A., Lara, J.C. and Lory, S.: The rpoN gene product of Pseudomonas aeruginosa is required for expression of diverse genes, including the flagellin gene. J. Bacterial. 172 (1990) 389-396.
113
GENE 07147
Pseudomonas aeruginosa 1asA gene: determination
point and analysis of the promoter/regulatory
of the transcription start region
(Elastase; LasR; cystic fibrosis; virulence; pseudoknots; protease)
Lynn C. Freck-O’Donnell
and Aldis Darzins
Department of Microbiology, The Ohio State University, Columbus, OH 43210, USA
Received by M. Bagdasarian: 15 January 1993; Accepted: 19 February 1993; Received at publishers: 11 March 1993
SUMMARY
The elastolytic activity of the opportunistic pathogen Pseudomonas aeruginosa is due to the combined activities of at least three secreted proteins: elastase, LasA and alkaline protease. Transcription of both the lasA gene and the elastase structural gene, la@ requires the transcriptional activator LasR. In order to localize the promoter elements involved in lasA expression, the transcription start point (tsp) for lasA was localized by Sl protection and primer extension analysis. The DNA sequence of the region upstream from the tsp was determined, and a putative 070 promoter was identified. Sequence comparison with the 1asB promoter region revealed two areas of considerable homology which could act as potential binding sites for LasR or other, as yet unidentified, regulatory proteins.
INTRODUCTION
Pseudomonas aeruginosa is an opportunistic pathogen that is able to cause severe and often fatal infections in compromised hosts, particularly burn and cystic fibrosis patients. The pathogenesis of P. aeruginosa infections is due in part to the secretion of a large number of virulence factors including exotoxin A and elastase. Elastase has been shown to inactivate several biologically important molecules such as IgG (Holder and Wheeler, 1984), IgA (Diiring et al., 1981), human cl,-proteinase inhibitor Correspondence to: Dr. A. Darzins, Department of Microbiology, The Ohio State University, 484 West 12th Avenue, Columbus, OH 43210, USA. Tel. (614) 292-0147; Fax (614) 292-1538.
Abbreviations: aa, amino acid(s); bp, base pair(s); CF, cystic fibrosis; cpm, counts per minute; E., Escherichia; Ig, immunoglobulin; LasA, protein involved in the degradation of elastin; lasA, gene encoding LasA, lasB, gene encoding elastase; LasR, transcriptional activator of IasA and lasB, lasR, gene encoding Las&, LB, Luria-Bertani (medium); nt, nucleotide(s); oligo, oligodeoxyribonucleotide; p, plasmid; P., Pseudomonas; PAGE, polyacrylamide gel electrophoresis; PI, isoelectric point; R, C or T, RBS, ribosome-binding site(s); tsp, transcription start point(s); u, unit(s); V., Vibrio.
(Morihara et al., 1979), and complement components (Schultz and Miller, 1974) and to degrade such structural proteins as collagen (Heck et al., 1986) and elastin (Morihara and Homma, 1985). The elastolytic activity of P. aeruginosa elastase is enhanced by the presence of a second secreted protein, LasA (Peters and Galloway, 1990). It has been shown that LasA degrades elastin in the absence of elastase (Peters et al., 1992; Toder et al., 1991), thereby classifying LasA as a second elastase produced by P. aeruginosa. LasA-enhanced elastolysis is due to separate activities of LasA and elastase on elastin and is not due to the activation of elastase by LasA (Peters et al., 1992). LasA is emerging as an important virulence factor due to its ability to enhance the elastolytic activities of both P. aeruginosa elastase and human neutrophil elastase (Peters et al., 1992) and may be a contributing factor to the destruction of connective tissue seen in the CF lung by Bruce et al. (1985). Thus, studies on the regulation of lasA will provide a better understanding of the environmental conditions that promote LasA-enhanced elastolysis. A positive regulator, designated LasR, has been identified and shown to have significant homology to
114
LuxR of Vibrio jscheri (Gambello and Iglewski, 1991). An intact 1asR gene is required for transcription of both 1asB (Gambello and Iglewski, 1991) and lasA (Toder et al., 1991). All three Zas genes have been mapped to distinct regions of the P. aeruginosa PA0 chromosome (Howe et al., 1983; Holloway and Zhang, 1990; Gambello and Iglewski, 1991). It is likely, therefore, that LasR alone, a LasR-protein complex, or another regulatory protein that is controlled at the transcriptional level by LasR activates transcription in trans by binding upstream from the lasA and 1asB promoters. The aim of the present study was the localization of the tsp for EasA and the analysis of the DNA sequence upstream from this site. This sequence was examined for putative promoter elements and used in comparison studies with the ZasB promoter region.
EXPERIMENTAL
AND DISCUSSION
(a) Identification of the tsp of 1asA Determining the tsp of a particular gene is often simpli-
fied by knowing the N-terminal aa of the gene product. The N terminus of the precursor form of LasA, however, has not been determined experimentally. Two different start codons separated by ca. 100 bp have been proposed based on independent determinations of sequence, size and pI of purified, mature LasA and on the size of the precursor form produced in E. coli. For this reason the tsp was first localized by low resolution Sl analysis (Freck et al., 1992) to a position ca. 110 nt upstream from the proposed ATG start of Schad and Iglewski (1988) and ca. 250 nt upstream from the proposed TTG start codon proposed by Darzins et al. (1990). This region is ca. 35 bp downstream from the unique SmaI site located near the beginning of the published lasA sequences. To avoid ambiguity, tsp and other reaction products have been localized with respect to this site rather than to potential start codons. From this information, an oligo primer (LA390, 5’-CGAAGGGAAACCTTGAAAGC) complementary to the sense strand ca. 75 bp from the 5’ end of the message was synthesized for use in high resolution Sl analysis. A plasmid containing the initiation region served as the template for extension from LA390 to produce a continuously labelled probe for Sl studies. The results from Sl analysis of PA01 RNA with the lasAspecific probe are shown in Fig. 1A. The two largest Slprotected fragments indicated by arrowheads (lanes 1 and 2) correspond to two 5’ ends of the message that are located 29 bp and 31 bp downstream from the SmaI site. The bracket in Fig. 1A denotes a ladder of smaller Slprotected fragments with ends ca. 47 bp downstream from the SmaI site. Multiple attempts at primer extension analysis using
LA390 were unsuccessful. Analysis of the region within 100 nt of the largest Sl-protected fragment revealed the presence of possible secondary and tertiary structures that could have interfered with primer extension studies. Tertiary structures due to interactions between loops and single-stranded regions have been described (Studnicka et al., 1978). These structures, known as pseudoknots, are hairpin loops with a single strand folded back to pair with the nt in the loop and generally involve 4 or fewer nt such that the single strand does not pass completely through the loop. This structure is stabilized by stacking of the two double helical segments (Pleij et al., 1985). The secondary structure (stem-loop) shown in Fig. 2 has a favorable free energy (- 10 kcal). This alone, however, is probably not enough to prevent reverse transcriptase from extending from LA390 through the region. Analysis of single-stranded regions for the ability to form a pseudoknot with the loop revealed 4 nt, 5’-CUGG (+16 to + 19), able to base pair with the sequence 5’-CCAG (+ 41 to +44) found in the loop. The 5’ ends of the smaller Slprotected fragments (Fig. IA) are located in the + 16 to + 19 region; therefore, it is possible that pseudoknot formation leaves the Sl probe susceptible to digestion in that region. The presence of this pseudoknot has not, however, been confirmed experimentally. Primer-extension analysis using a second oligo primer (LA359,5’-CGACATCGCCTCCGACGGG) that should destabilize these structures on the RNA confirmed the predicted tsp. The two major extended products indicated by arrows (Fig. lB, lane 1) correspond exactly to the 5’ ends of the two larger protected fragments determined by Sl mapping (Fig. 1A). No additional products were seen that mapped to the same region as the smaller Slprotected fragments. Thus, it was determined that transcription of lasA is initiated at two different sites separated by only 1 nt. (h) The nt sequence of the fasA promoter region and identification of a 0” promoter
Since the SmaI site is located at ca. -30 with respect to the tsp of lasA, a 0” promoter, if present, should span the SmaI site. In addition, transcriptional activators such as LasR would be expected to bind upstream from the promoter between -200 and -20 (Collado-Vides et al., 1991). Therefore, the nt sequence upstream from the lasA tsp (ca. 300 bp) was determined and is shown in Fig. 3A. The two previously reported sequences of 1asA (Schad and Iglewski, 1988; Darzins et al., 1990) overlap with the sequence presented here as described in the legend to Fig. 3A. A comparison of the sequence immediately upstream from the SmaI site with the sequence of Schad and Iglewski (1988), however, revealed 3 additional bp not present in the original 1asA sequence.
115
A
A
C
G
B
123
T
G C A
C
G
T
1
2
T C C \iT G
G
E G T\
A
i
=
A C C C G
1
G C C
I
:
E G T G C A
iA
L G G
Fig. 1. Determination of the tsp of 1asA. (A) Sl nuclease analysis. RNA was prepared from 50-ml cultures of P. aeruginosa PA01 grown 12 h in LB by a guanidine isothiocyanate-hot phenol method (Rothmel et al., 1991). Sequencing of the antisense strand was performed using 10 uCi [a-“P]dCTP and SequenaseTM (US Biochemical, Cleveland, OH, USA) by the dideoxy chain-termination method (Sanger et al., 1977) as recommended by the manufacturer. The Sl probe was synthesized by modifying the sequencing reaction to allow polymerization of a high-molecular-weight probe from a plasmid template (pLO593, lusA SmaI-XhoI subclone in pUC18) and an oligo primer (LA390, see Fig. 3A). RNA (50 ug, 100 ug, and 0 ug; lanes 1, 2, and 3, respectively) was hybridized with 10’ cpm of probe overnight at 65°C and digested with 100 u of Sl nuclease (Bethesda Research Laboratories, Bethesda, MD, USA) as described previously (Dixon, 1984). Digestion products were analyzed by 8M urea-containing 6% PAGE adjacent to a reference sequence produced from pL0593 and LA390 (lanes A, C, G, T). Arrowheads and underlined nt indicate the two 5’ ends of the transcript (tsp),and the bracket and underlined nt mark an area protected from Sl nuclease due to secondary structure (see section a). (B) Primer extension analysis. RNA (50 ug and 0 ug; lanes 1 and 2, respectively) was analyzed using a primer extension system from Promega (Madison, WI, USA) and the oligo LA359 (see Fig. 3A). Products were analyzed as in A except that the sequencing primer was LA359 (lanes A, C, G, T). Arrowheads and underlined nt indicate two tsp (as in A).
The region upstream from the tsp of 1asA was examined for potential o”-type promoter sequences by comparison to other Pseudomonas promoters. Deretic et al. (1989) A&!Z G G
i C: :G G: :C
A::U A u I G::C C G::C C::G GoU G::C C::G C::G AGGCGAUGUCGCCGGGCUGCUGGC... GGCGUGCAACCCGGCEYEEj;CGAAAUA 3' 5' pneudoknot
intar8ctions:
5'CUGG 3' 3’GACC 5'
Fig. 2. Possible secondary structure of 1asA mRNA. The sequence corresponding to the first 100 nt of the 1asA transcript was analyzed for secondary structure by the method of Jaeger et al. (1989). Watson-Crick base pairing is denoted by ::, while GoU indicates a non-Watson-Crick base pair. The doubly underlined nt are complementary to the singly underlined nt in the loop structure. The possible interactions between these two regions to form a pseudoknot are shown below the secondary structure.
classified twelve Pseudomonas promoters as 0” (rpoD) promoters. From these promoters we derived a consensus sequence which is as follows: TTGACR (- 35 region) and TATRRT (-10 region). The-most invariant nt are underlined. The region upstream from the tsp of lasA was found to contain sequences homologous to the consensus promoter (Fig. 3A). We also compared the lasA sequence to a recent compilation of 29 cr” promoters from streptomycetes (Strohl, 1992), which, like pseudomonads, contain DNA with a high G+C content. The results of this analysis also identified the -10 and -35 regions shown in Fig. 3A.
(c) Comparison of fusA and IasB promoter regions The lasB promoter has been classified as a P. aeruginow virulence promoter by Deretic et al. (1989). This designation was based on the lack of homology to other known promoters and the role of this gene in pathogenesis. The lasB promoter region also displays limited homology to some members of the os4 class of promoters (Deretic et al., 1989). However, it has been shown that a os4 (rpoN) mutant of P. aeruginosa strain PAK produces
116
A . . . HinfI GAATCGCCCGCGGACAAGCGCCTGAGGGTCGTGCATGGCCGAGATCGGAGCCGGCCAC . . .lOO GGCCTCGACGAGGCCGAGCTGGCGCGTCGTGATCCGCTCGACGGTTCCTCCGACGAGG~ . . HhlfI' XmJlI CGCTGAGCGCGCAGACTGTATCGAAGTATTTTCCGACGCCTGGCGTTCTGTGATCGATTCG . Sau3AI. 9200 GCTCGGTTGGAGTGCGCTGCCGCACGCTTTACAGCCGGACGGAGCGGGGCACCGGATCTA . . +1 0300 SmaI CTGCAMAGCTGATAG~TCCCGGGTGACGGCGTTG~CCTGCfiG~GTGCAA . n * -35 -10 . . . CCCGGCCTGGGCGMATACCGGCGGAGCGGACCAGGCTAT CCCGTCC.C.AC.GCGATGTCGC LA359 . . '400 CGGGCTGCTGGCTTTCAAC.GTTTCCCTTCGATGACCBGG LA390 RBS Sau3AI GATC
t&t
B .
. 1asA
:::: :: ::: : ::::
CTGTGATCGATTCGGCTCGGTTGGAGTGC---““‘:
.. .. .. .. ..
:
:
::
CTGAGCGCGTCCCGGAGC--TGGGGGCAACCTAd .
.
.
.
+GCTGATA$&ZT CGGACGGAGCGGGGCAC----::::::::::: TA&TATTC~A&&,ACAT
.
m--TCAAGGCTACCTG .
::
::
::
.
-10 +1 SmaI . CCCGGG-TGACGGCGTTG!XUi!iTCCTGCGGCGTGCAACCCG----GCCTGGG :: :: :: ::: :::: ::: :: ACACGAAAAGCA&T CCAGTTCTGGCAGGTTT-GCGGG;T;TTTT;;GT . . -10 t1 -35
Fig. 3. Sequence of the promoter region of 1asA and comparison to lasB. (A) Sequence of the 1asA promoter region. Both strands of DNA upstream from the tsp of lasA were sequenced as described in Fig. 1 with the exception that [c+%]dCTP was used for labelling. In order to present the overlapping sequence and to keep the promoter intact, portions of this sequence have been published previously [bp 247-424 (Schad and Iglewski, 1988);bp 264424 (Darzins et al., 1990)]. Additional nt not present in the previously published sequence (Schad and Iglewski, 1988) are marked with carets. The sequence is marked with a dot every 20 nt, starting with the first one of the sequence. Restriction sites used for the generation of overlapping subclones are labelled. Important features in the sequence are underlined. The sequences labelled LA359 and LA390 are complementary to the two primers used for transcript mapping. The tsp are shown in bold type and are marked + 1.Putative promoter sequences are labelled - 10 and - 35. One of the potential start codons is marked (Met) along with the putative RBS (Schad and Iglewski, 1988). The sequence data have been deposited with the GenBank data library under accession No. L0849. (B) Comparison with the lush regulatory region. The sequences upstream from the tsp of 1asE (Bever and Iglewski, 1988; Mohr et al., 1990) and lasA were aligned to produce the best homology within 150 bp from +l. The sequence is marked with a dot every 20 nt back from +I. Gaps introduced to obtain the best fit are shown as dashes. Matches between the two sequences are denoted by colons. Regions of considerable homology are doubly underlined and labelled I and II. The tsp are underlined and labelled +l. Putative promoters are underlined and labelled -10 and -35. The SmaI site of 1asA is also shown.
parental-like levels of elastase (Totten et al., 1990). Therefore, the transcription of 1asB is probably not initiated by a RpoN-containing RNA polymerase. Since a putative 07’-type promoter was found for la&, we examined the 1asB promoter region and also found putative -10 and -35 sequences (Fig. 3B). Sequences upstream from the LasR-regulated IasA and 1asB promoters may contain homologous regions to which regulatory protein complexes could bind. The promoter region of 1asA was aligned with the sequence upstream from the 1asB tsp. Two regions of considerable homology between the aligned sequences were found (Fig. 3B). Region I displays 75% homology and is centered around -86 of 1asA and -111 of 1asB. Region II is 69% homologous and is centered around - 51 of 1asA and -71 of 1asB. Since the C-terminal domain of LasR is homologous to the DNA-binding domain of LuxR (Gambello and Iglewski, 1991), the region surrounding the palindromic operator of the lux regulon of I/. Jischeri (Devine et al., 1988; 1989) was also aligned with the 1asA and 1asB promoter sequences. The most significant homology to the lux operator was seen just upstream from and including region I. When gaps are introduced near the center of the 20-bp palindrome to allow maximal alignment, the LuxR target site is 35% homologous to 1asB sequence and 45% homologous to 1asA sequence (data not shown). These values increase to 42% and 54%, respectively, when 4 bp located 3’ to the lux operator are included in the comparison. These sequences are currently the focus of further investigations to determine if they serve as binding sites for LasR. (d) Conclusions (1) Transcription of 1asA is initiated at two different sites separated by 1 nt. (2) The nt sequence upstream from the 1asA tsp was determined, and a putative 07’ promoter was identified. A putative a7’ promoter was also found upstream from the IasB tsp. (3) Comparison of the sequences upstream from the tsp of 1asA and 1asB revealed two regions of significant homology which may serve as potential binding sites of the transcriptional activator LasR or other unidentified regulatory proteins.
REFERENCES Bever, R.A. and Iglewski, B.H.: Molecular characterization and nucleotide sequence of the Pseudomonns aeruginosa elastase structural gene. J. Bacterial. 170 (1988) 4309-4314. Bruce, MC., Poncz, L., Klinger, J.D., Stem, R.C., Tomashefski Jr., J.F. and Dearborn, D.G.: Biochemical and pathologic evidence for pro-
117 teolytic destruction of lung connective tissue in cystic fibrosis. Am. Rev. Respir. Dis. 132 (1985) 529-535. Collado-Vides, J., Magasanik, B. and Gralla, J.D.: Control site location and transcriptional regulation in Escherichia coli. Microbial. Rev. 55 (1991) 371-394. Darzins, A., Peters, J.E. and Galloway, D.R.: Revised nucleotide sequence of the 1asA gene from Pseudomonas aeruginosa PAOl. Nucleic Acids Res. 18 (1990) 6444. Deretic, V., Konyecsni, W.M., Mohr, CD., Martin, D.W. and Hibler, N.S.: Common denominators of promoter control in Pseudomonas and other bacteria. Bio/Technology 7 (1989) 1249-1254. Devine, J.H., Countryman, C. and Baldwin, T.O.: Nucleotide sequence of the 1uxR and 1~x1 genes and structure of the primary regulatory region of the lux regulon of VibrioJischeri ATCC 7744. Biochemistry 27 (1988) 837-842. Devine, J.H., Shadel, G.S. and Baldwin, T.O.: Identification of the operator of the lux regulon from the Vibrio jischeri strain ATCC 7744. Proc. Natl. Acad. Sci. USA 86 (1989) 5688-5692. Dixon, R.A.: Tandem promoters determine regulation of the KZebsieUa pneumoniae glutamine synthetase (gInA) gene. Nucleic Acids Res. 12 (1984) 7811-7830. Doring, G., Obernesser, J. and Botzenhart, K.: Extracellular toxins of Pseudomonas aeruginosa. II. Effect of two proteases on human immunoglobulins IgG, IgA and secretory IgA. Zentralb. Bakterial. Mikrobiol. Hyg. 1 Abt. Orig. A 249 (1981) 89-98. Freck, L.C., Galloway, D.R. and Darzins, A.: Analysis of Pseudomonas aeruginosa LasA production. Abstracts of the General Meeting of the American Society for Microbiology, 1992, p. 36. Gambello, M.J. and Iglewski, B.H.: Cloning and characterization of the Pseudomonas aeruginosa 1asR gene, a transcriptional activator of elastase expression. J. Bacterial. 173 (1991) 3000-3009. Heck, L.D., Morihara, K., McRae, W.B. and Miller, E.J.: Specific cleavage of human type III and IV collagens by Pseudomonas aeruginosa elastase. Infect. Immun. 51 (1986) 115-118. Holder, LA. and Wheeler, R.: Experimental studies of the pathogenesis of infections due to Pseudomonas aeruginosa: elastase, an IgG protease. Can. J. Microbial. 30 (1984) 1118-I 124. Holloway, B.W. and Zhang, C.: Pseudomonas aeruginosa PAO. In: O’Brien, J. (Ed.), Genetic Maps. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1990, pp. 2.71-2.78. Howe, T.R., Wretlind, B. and Iglewski, B.H.: Comparison of two methods of genetic exchange in determination of the genetic locus of the structural gene for Pseudomonas aeruginosa elastase. J. Bacterial. 156 (1983) 58-61. Jaeger, J.A., Turner, D.H. and Zuker, M.: Predicting optimal and suboptimal secondary structure for RNA. Methods Enzymol. 183 (1989) 281-306.
Mohr, CD., Rust, L., Albus, A.M., Iglewski, B.H. and Deretic, V.: Expression patterns of genes encoding elastase and controlling mucoidy: co-ordinate regulation of two virulence factors in Pseudomonas aeruginosa isolates from cystic fibrosis. Mol. Microbial. 4 (1990) 2103-2110. Morihara, K. and Homma, J.Y.: Pseudomonas proteases. In: Holder, I.A. (Ed.), Bacterial Enzymes and Virulence. CRC Press, Boca Raton, FL, 1985, pp. 41-79. Morihara, K., Tsuzuki, H. and Oda, K.: Protease and elastase of Pseudomonas aeruginosa; inactivation of human plasma alphaiproteinase inhibitor. Infect. Immun. 24 (1979) 188-193. Peters, J.E. and Galloway, D.R.: Purification and characterization of an active fragment of the LasA protein from Pseudomonas aeruginosa; enhancement of elastase activity. J. Bacterial. 172 (1990) 2236-2240. Peters, J.E., Park, S.J., Darzins, A., Freck, L.C., Saulnier, J.M., Wallach, J.M. and Galloway, D.R.: Further studies on Pseudomonas aeruginosa LasA: analysis of specificity. Mol. Microbial. 6 (1992) 1155-1162. Pleij, C.W.A., Rietveld, K. and Bosch, L.: A new principle of RNA folding based on pseudoknotting. Nucleic Acids Res. 13 (1985) 1717-1731. Rothmel, R.K., Chakrabarty, A.M., Berry, A. and Darzins, A.: Genetic systems in Pseudomonas. Methods Enzymol. 204 (1991) 485-514. Sanger, F., Nicklen, S. and Coulson, A.R.: DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467. Schad, P.A. and Iglewski, B.H.: Nucleotide sequence and expression in Escherichia coli of the Pseudomonas aeruginosa 1asA gene. J. Bacterial. 170 (1988) 2784-2789. Schultz, D.R. and Miller, K.D.: Elastase of Pseudomonas aeruginosa: inactivation of complement components and complement-derived chemotactic and phagocytic factors. Infect. Immun. 10 (1974) 128-135. Strohl, W.R.: Compilation and analysis of DNA sequences associated with apparent streptomycete promoters. Nucleic Acids Res. 20 (1992) 961-974. Studnicka, G.M., Rahn, G.M., Cummings, I.W. and Salser, W.A.: Computer method for predicting the secondary structure of singlestranded RNA. Nucleic Acids Res. 9 (1978) 3365-3387. Toder, D.S., Gambello, M.J. and Iglewski, B.H.: Pseudomonas aeruginosa LasA: a second elastase under the transcriptional control of 1asR. Mol. Microbial. 5 (1991) 2003-2010. Totten, P.A., Lara, J.C. and Lory, S.: The rpoN gene product of Pseudomonas aeruginosa is required for expression of diverse genes, including the flagellin gene. J. Bacterial. 172 (1990) 389-396.