Molecular and Functional Analysis of the lepB Gene, Encoding a Type I Signal Peptidase from Rickettsia rickettsii and Rickettsia typhi

Abstract

The type I signal peptidase lepB genes from Rickettsia rickettsii and Rickettsia typhi, the etiologic agents of Rocky Mountain spotted fever and murine typhus, respectively, were cloned and characterized. Sequence analysis of the cloned lepB genes from R. rickettsii and R. typhi shows open reading frames of 801 and 795 nucleotides, respectively. Alignment analysis of the deduced amino acid sequences reveals the presence of highly conserved motifs that are important for the catalytic activity of bacterial type I signal peptidase. Reverse transcription-PCR and Northern blot analysis demonstrated that the lepB gene of R. rickettsii is cotranscribed in a polycistronic message with the putative nuoF (encoding NADH dehydrogenase I chain F), secF (encoding protein export membrane protein), and rnc (encoding RNase III) genes in a secF-nuoF-lepB-rnc cluster. The cloned lepB genes from R. rickettsii and R. typhi have been demonstrated to possess signal peptidase I activity in Escherichia coli preprotein processing in vivo by complementation assay.

The genus Rickettsia comprises several intracellular pathogens, some of which are responsible for the most severe bacterial diseases of humans. These include R. prowazekii, the causative agent of epidemic typhus, R. typhi, the agent of murine typhus, and R. rickettsii, the agent of Rocky Mountain spotted fever (5, 16). These gram-negative, obligate, intracellular bacteria are transmitted to their mammalian hosts by arthropod vectors such as ticks, fleas, and lice and grow within the cytoplasm of eukaryotic cells (3). Although systematic approaches have revealed substantial information about the biology of rickettsial growth in host cells, the lack of a genetic manipulation system has hampered our ability to characterize the genes involved in the pathogenesis of rickettsiae (16, 30). The molecular basis of the protein secretion mechanism involved in the growth and pathogenesis of these intracellular bacteria remains an important subject of research.

In bacteria, the majority of proteins that are translocated across membranes are synthesized as preprotein with an amino-terminal extension known as the signal or leader peptide. The signal peptide is involved in targeting preproteins for translocation via the Sec system (25). The Sec machinery consists of multiple proteins and provides a channel for translocation of newly synthesized preproteins from the cytosol across the cytoplasmic membrane in bacteria (13). The homotetramer SecB, a chaperone protein, interacts with the newly synthesized preprotein in the cytoplasm and targets the preproteins to the SecAYEG-translocase at the membrane interface. Finally, type I signal peptidase, a membrane-bound endopeptidase presumably located in proximity to SecYEG, cleaves the leader peptide from the preprotein, which results in the release of the mature protein from the membrane (22). It has been demonstrated that inhibition of bacterial type I signal peptidase leads to the accumulation of preproteins and eventual cell death (11, 14, 20, 21). Due to its essential role in bacterial cell growth and relative accessibility of its active site on the outer leaflet of the cytoplasmic membrane, type I signal peptidases have been considered as a potential target for the development of novel antibacterial agents (24).

New opportunities arising from the recent publication of the genome sequences of R. prowazekii (2), R. conorii (23), and R. sibirica (GenBank accession number AABW01000001) now enable us to select and characterize rickettsial genes of interest. Our interest is in characterizing the genes involved in protein secretion pathways of rickettsiae in order to assess their potential roles in the invasion, growth, and pathogenesis of these obligate intracellular bacteria. In this communication, we report the cloning and sequence analysis of the putative lepB gene that encodes type I signal peptidase from R. rickettsii and R. typhi. In addition, we provide the first detailed molecular and functional characterization of the lepB gene of R. rickettsii and R. typhi.

MATERIALS AND METHODS

Bacterial strains.

R. rickettsii strain Sheila Smith (7) and R. typhi strain AZ322 Ethiopian isolate (4) were used in this study. The temperature-sensitive E. coli strain IT41 (17) was used for complementation assay.

Genomic DNA extraction.

Vero cells (African green monkey kidney cells, ATCC number CRL-1573) were cultured in Dulbecco's modified Eagle medium (DMEM) with 4.5 g of glucose per liter with glutamine (Biofluids, Inc., Rockville, Md.) supplemented with 10% fetal bovine serum (Gemini, Calabasas, Calif.). R. rickettsii and R. typhi were propagated in Vero cells as previously described (15, 26). Rickettsiae were partially purified from rickettsiae-infected (>90%) Vero cells as follows. Infected cells were harvested and mechanically ruptured by forcing them through a 27-gauge needle attached to a 10-ml syringe to release the intracellular bacteria. To enrich for rickettsiae, large cell fragments and intact host cells were removed by low-speed centrifugation (275 × g for 10 min). The supernatants were centrifuged at 14,000 × g for 20 min at 4°C to pellet the partially purified rickettsiae. Genomic DNA of R. rickettsii and R. typhi was extracted by using the Wizard genomic DNA purification kit (Promega, Madison, Wis.).

Cloning of the R. rickettsii and R. typhi lepB operon.

The R. rickettsii lepB operon was amplified by PCR in two different fragments. The primers used in PCRs are shown in Table 1. The primers AZ971 (forward) and AZ974 (reverse) were used to amplify the lepB gene of R. rickettsii (first fragment). For 100 μl of PCR, 200 ng of R. rickettsii genomic DNA was used. Thermal cycling conditions consisted of initial denaturation at 94°C for 2 min followed by 30 cycles at 94°C for 1 min, 45°C for 1 min, and 72°C for 3 min, and a final extension step at 72°C for 10 min was performed by using Pfu DNA polymerase (Stratagene, La Jolla, Calif.). The PCR product (1,640 bp) was purified by Strataprep PCR purification kit (Stratagene). The purified PCR product was cloned into pPCR-Script Amp SK(+) vector (Stratagene) and was transformed into Escherichia coli TOP10 cells (Invitrogen Life Technologies, Carlsbad, Calif.). The cloned lepB region of R. rickettsii was sequenced by the dye termination method by using a model 373 automated fluorescent sequencing system (Applied Biosystems, Foster City, Calif.). The second fragment containing the upstream region of the lepB gene was PCR amplified from R. rickettsii DNA by using the forward primer AZ1501 and the reverse primer AZ1375. PCR amplification, cloning, and sequencing of the second fragment (2,978 bp) were performed by following the same conditions as mentioned for the first fragment. The sequences of both fragments of the R. rickettsii lepB region were combined and aligned by using MacVector 6.5.3 software (Genetics Computer Group, Inc., Madison, Wis.).

TABLE 1.

Primers used in PCR reactions

Primer	Sequencea	Nucleotide position
AZ971	GGGTCTGGACTTGGTACAGGTGG	138,001-138,023b
AZ974	CACTTCTTCGCCATGAGTCA	139,601-139,620b
AZ1039	CAGTTAAACAGGAGTTTGCTTC	533-554e
AZ1040	GATTCTTGAATATTCGACTTAATC	1,253-1,276e
AZ1055	CGCCATGAGTCAgAATTcTATAATC	139,588-139,612b
AZ1056	GGTACAGGaGcTcTTATTGTTATGG	138,013-138,037b
AZ1262	CTTCGCCATGAGTCAgAATTcTATA	139,591-139,615b
AZ1263	GGTAgAGctcGTATTATAGTTATGG	2,323-2,347c
AZ1287	CAGCTAAGCAAGAGTGGGGGTC	2,867-2,888c
AZ1286	CGATTTAACCTCACAGATTCAACCC	3,572-3,596c
AZ1372	GAGCCGTTTACCGTTCCAAC	2,938-2,957c
AZ1374	TGTTCTACCGTTTGGCAGTG	3,290-3,309c
AZ1375	TTGGAACGGTAAACGGCTCC	2,937-2,956c
AZ1501	CCTCAAATCCCTAAAGTATCTC	4,415-4,436d
AZ1514	GAGAggATcCAAACAGATAATAC	2,824-2,846c
AZ1515	CAAATGAaTtCATTACGCATCCGTG	3,618-3,642c
AZ1528	GCTTGGAATAGGTGAGGTGG	514-533c
AZ1529	GCAAGAAGCGAGGCGATTGG	1,448-1,467c
AZ1531	GCTTCATCTAAGGCACGCTG	1,786-1,805c
AZ1533	CCTTATTCTTTTTGGCGGTG	1,063-1,082c
AZ1534	CGTGCGTTCTATTTTTTTGTCG	3,213-3,234c
AZ1535	CCCTCTATTGCTTTCGTAAC	2,518-2,537c
AZ1556	ACTTCTTGAGATAAATCAGC	126-145c
AZ1566	TTAGCTCCAACCATGCATATTG	3,883-3,904c
AZ1568	GCTTTGTCATACTCATAAACCCAAG	833-857c

^aThe nucleotides modified to generate restriction sites are indicated in lower case.

^bNucleotide sequence position numbering corresponding to the R. prowazekii genome sequence (GenBank accession number AJ235270).

^cNucleotide sequence position numbering taken from R. rickettsii lepB operon (GenBank accession number AY134668).

^dNucleotide sequence position numbering corresponding to the R. conorii genome sequence under GenBank accession number AE008582.

^eNucleotide sequence positions are taken from R. typhi lepB operon under GenBank accession number AF503336.

The R. typhi lepB gene was amplified by using the primers AZ971 (forward) and AZ974 (reverse). The PCR fragment (1,639 bp) was cloned and sequenced as described above.

The sequence of the lepB operon and deduced amino acid sequence of R. rickettsii and R. typhi were analyzed with MacVector 6.5.3 software. Sequence comparisons to those available in GenBank were performed using BLAST analysis (http://www.ncbi.nlm.nih.gov).

Isolation of RNA and RT-PCR.

R. rickettsii and R. typhi were purified from Vero cells (>90% infection) as described above. Total RNA from the partially purified rickettsiae was isolated by the use of Trizol reagent (Invitrogen Life Technologies) and treated with RQ1 RNase-free DNase (Promega) by following manufacturers' recommendations. Reverse transcription-PCR (RT-PCR) was performed with 300 ng of total RNA in 50-μl reaction volumes by using SuperScript One-Step RT-PCR with Platinum Taq (Invitrogen Life Technologies). The thermal cycling conditions consisted of one cycle of 45°C for 30 min and 94°C for 2 min, followed by 35 cycles of 94°C for 30 s, 48°C for 30 s, 72°C for 2 min, and a final extension step of 72°C for 10 min.

Northern analysis.

The total RNA (6 μg) from R. rickettsii was subjected to Northern analysis by using the NorthernMax kit (Ambion, Austin, Tex.). The [α-³²P]dATP (Amersham Pharmacia Biotech, Piscataway, N.J.)-labeled 297-bp probe specific to the lepB coding sequence corresponding to primers AZ1372 and AZ1534 was prepared by use of the Strip-EZ PCR probe synthesis kit (Ambion). The hybridized membrane (positively charged nylon) was exposed to Kodak Biomax MS film for autoradiography.

Complementation and expression analysis of the rickettsial lepB gene.

The lepB gene of R. rickettsii or R. typhi was cloned into the SacI and EcoRI sites of pPCR-Script Amp SK(+) vector (Stratagene) under a lac promoter by incorporation of the restriction sites into the primers used to amplify the insert sequences. The primers AZ1055 (EcoRI) and AZ1056 (SacI) were used for the cloning of the 1,588-bp fragment of the R. typhi lepB gene to generate the pRTlepB23 plasmid. For the R. rickettsii lepB gene, the 1,592-bp fragment was amplified by AZ1262 (EcoRI) and AZ1263 (SacI) primers in order to clone and generate the pRRlepB569 plasmid. The constructed plasmids pRRlepB569 and pRTlepB23 were checked by sequencing. For the complementation assay, plasmids were transformed into E. coli strain IT41 cells and selected on a Luria broth (LB)-ampicillin (100 μg ml⁻¹) plate incubated at 30°C for 48 h. For controls, plasmid pUC18 or pESL4 [carrying a 2,229-bp fragment of the groESL gene of R. typhi into pPCR-Script Amp SK(+) vector; reference 26] was also transfected into E. coli strain IT41 cells. For the complementation assay by growth curve, the transformed cells were grown in an LB-ampicillin mixture overnight at 30°C. The cultures were diluted 100-fold into a fresh LB-ampicillin mixture and incubated with shaking at nonpermissive temperature of 42°C. The optical density at 600 nm was recorded at 30-min intervals. For the complementation assay by CFU assay, the transformed cells were grown to mid-log phase in LB-ampicillin mixture at 30°C and plated onto two sets of LB-ampicillin plates. One set of plates was incubated at 30°C and the other at 42°C. The colonies were counted after 48 h of incubation to determine the percentage of growth at 42°C with respect to 30°C. All experiments were performed at least three times and the standard deviation was calculated (shown as ±) by using Microsoft Excel software.

To analyze the synthesis of rickettsial signal peptidase I in E. coli strain IT41, the lepB open reading frame (ORF) of R. rickettsii was amplified by primers AZ1514 (BamHI) and AZ1515 (EcoRI) and cloned into the pTrcHisC vector (Invitrogen Life Technologies) at the BamHI and EcoRI sites. The constructed plasmid pTrcHisRR4 that contained the 804-bp ORF of lepB from R. rickettsii was confirmed by sequencing. The constructed plasmids pTrcHisRR4 and vector pTrcHisC were transfected into E. coli strain IT41 cells as mentioned above.

The transformed cells were grown to mid-log phase in LB-ampicillin (100 μg/ml) medium at 30°C and then induced for protein expression by the addition of 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG). Cells were harvested at 4 h postinduction and resuspended in (1/10 vol) 1× phosphate-buffered saline. Cell suspensions were mixed with equal volumes of 2× Tris-glycine-sodium dodecyl sulfate (SDS) sample buffer (Invitrogen Life Technologies) and boiled at 100°C for 5 min. Total cell proteins were separated on 4 to 12% Tris-glycine precast gel (Invitrogen Life Technologies) by using 1× Tris-glycine-SDS running buffer (Bio-Rad, Hercules, Calif.). The proteins were transferred to a polyvinylidene difluoride membrane (Invitrogen Life Technologies). The membrane was blotted with His-Tag monoclonal antibody (Novex, Madison, Wis.) by using the WesternBreeze chromogenic immunodetection system (Invitrogen Life Technologies).

Nucleotide sequence accession numbers.

The GenBank accession numbers for the lepB operons reported in this communication are AY134668 for R. rickettsii and AF503336 for R. typhi.

RESULTS

Cloning and sequence analysis of the rickettsial lepB gene.

The lepB operon of R. rickettsii and R. typhi was cloned and sequenced as described in Materials and Methods. The DNA sequence analysis of the lepB operon revealed a putative ORF of 801 nucleotides for R. rickettsii and 795 nucleotides for R. typhi. Alignment analysis showed that the lepB DNA sequences were 89 to 98% identical among R. rickettsii, R. typhi, R. prowazekii (2), R. conorii (23), and R. sibirica (GenBank accession number AABW01000001). The deduced amino acid sequences of the lepB gene among the rickettsiae species showed a very high degree of identity (ranging from 89 to 98%); however, that with the lepB of E. coli was very low (around 26%) (Fig. 1). Nevertheless, the amino acid sequence alignment shown in Fig. 1 revealed highly conserved amino acid domains (boxes B, C, D, and E) that are considered important for the catalytic activity of bacterial type I signal peptidase (10, 24). Box B (residues 88 to 95; E. coli signal peptidase I numbering) contains the nucleophilic Ser90 (shown in blue) and a conserved Met91 (shown in red). Box C contains residues 127 to 134, and box D (residues 142 to 153) contains the general base Lys145 (shown in blue) and a conserved Arg146 (shown in red). Box E (residues 272 to 282) contains the highly conserved Gly272, Asp273, Asn274, Asp280, and Arg282 (shown in red). These sequence analyses, performed with the web-based HMMTOP program, suggested the presence of a single amino-terminal transmembrane domain for the rickettsiae (Fig. 1) compared to the two transmembrane domains in E. coli signal peptidase I (10).

FIG. 1.

Alignment of amino acid sequences deduced from the putative lepB gene of R. rickettsii (Rr, this work), R. typhi (Rt, this work), R. conorii (Rc, accession number AE008582), R. prowazekii (Rp, accession number AJ235270), R. sibirica (Rs, accession number AABW01000001), and the signal peptidase I of E. coli (Ec, accession number BAA10915). Number of amino acids (a.a. #) is mentioned after the sequence of each species. The molecular weight (MW) and isoelectric point (pI) were computed by using the prediction server available at http://us.expasy.org/cgi-bin/pi_tool.html. The transmembrane domains shown in green were predicted by the HMMTOP program available at http://www.enzim.hu/hmmtop/index.html. The conserved amino acids regions (boxes B through E) are shown in boldface (black, blue, and red). Overall identity with R. rickettsii (Rr) signal peptidase I was calculated with MacVector 6.5.3 software.

Transcriptional analysis of the rickettsial lepB gene.

RT-PCR was performed to analyze rickettsial expression of the lepB gene. Amplification products (predicted from the sequence) of 744 bp (lane 2) and 730 bp (lane 3) as shown in Fig. 2 were obtained for R. typhi and R. rickettsii, respectively, thereby confirming the expression of lepB mRNA for these rickettsia species.

FIG. 2.

Transcription analysis of the R. rickettsii and R. typhi lepB genes. Ethidium bromide-stained 1% agarose gel in 1× TAE (Tris-acetate-EDTA) buffer. Total RNAs isolated from R. rickettsii or R. typhi cultured in Vero cells were used for RT-PCR. Lanes 1 and 2 represent PCR and RT-PCR analysis, respectively, on the total RNA isolated from R. typhi (performed by using forward primer AZ1039 and reverse primer AZ1040 specific to the lepB coding region). Lanes 3 and 4 represent RT-PCR and PCR analysis on the total RNA isolated from R. rickettsii (performed by using forward primer AZ1287 and reverse primer AZ1286 specific to the lepB coding region). The control lanes 1 and 4 demonstrate the absence of DNA in the RNA samples. GeneRuler 100-bp DNA ladder plus (MBI-Fermentas, Hanover, Md.) was used as a DNA size marker (lane M).

Total RNA isolated from R. rickettsii was analyzed by Northern hybridization to assess lepB transcript size (Fig. 3). The hybridization probe specific to the lepB coding sequence detected three bands (4.5, 2.5, and 1.5 kb), thereby indicating the polycistronic transcription of the R. rickettsii lepB operon. The lower hybridization intensity of the 4.5-kb band compared to that of the 2.5- and 1.5-kb bands could be explained by lower stability and posttranscriptional cleavage of the polycistronically transcribed message of the R. rickettsii lepB operon. However, the presence of an additional promoter(s) to generate multiple transcripts cannot be ruled out.

FIG. 3.

Northern blot analysis of total RNA isolated from R. rickettsii cells cultured in Vero cells. Total RNA (6 μg) was separated on a 1% agarose gel and transferred to a positively charged nylon membrane. The membrane was hybridized with a radiolabeled 297-bp probe specific to R. rickettsii lepB. Relative size of the hybridized bands was determined by using a 0.24- to 9.5-kb RNA ladder (Invitrogen-Life Technologies).

For further characterization of the polycistronic transcription of the R. rickettsii lepB operon, a series of RT-PCR analyses were performed on the total RNA isolated from R. rickettsii, which examined the putative secF (protein export membrane protein) and nuoF (NADH dehydrogenase I chain F) genes upstream of the lepB gene and the putative rnc gene (RNase III) downstream of lepB gene as shown in Fig. 4A. The expected RT-PCR products utilizing various forward and reverse primers on the secF-nuoF-lepB-rnc gene cluster (Fig. 4A) of R. rickettsii are shown in Fig. 4B. The RT-PCR products shown in lanes 2 to 5 and 7 (Fig. 4B) suggested the polycistronic transcription of the secF-nuoF-lepB-rnc gene cluster. The lower intensity of the RT-PCR product (1,894 bp) of the primer pair AZ1533 and AZ1375 spanning the secF to lepB genes (Fig. 4B, lane 4) supported the explanation of posttranscriptional cleavage and instability of the polycistronically transcribed single mRNA of the R. rickettsii secF-nuoF-lepB gene cluster. The RT-PCR product shown in lane 6 (Fig. 4B), produced by using forward primer AZ1556 (165 nucleotides upstream from the secF start codon) and reverse primer AZ1568 (from the coding region of secF), indicated that the transcription start site could be located further upstream of the secF-nuoF-lepB-rnc gene cluster of R. rickettsii.

FIG. 4.

Schematic map and RT-PCR analyses of clustered secF-nuoF-lepB-rnc′ genes for R. rickettsii. (A) Schematic map and scale of the secF-nuoF-lepB-rnc′ gene cluster of 3,930 bp for R. rickettsii illustrates three putative ORFs of secF (encoding protein export membrane protein; nucleotide position, 290 to 1,216; green), nuoF (encoding NADH dehydrogenase I chain F; nucleotide position, 1,381 to 2,646; blue), lepB (encoding type I signal peptidase; nucleotide position, 2,830 to 3,630; red) and partial sequence of rnc′ (encoding RNase III; partial 5′ sequence, 3,630 to 3,930; gray). Primers used in RT-PCR analysis are shown by forward and reverse arrows. (B) RT-PCR analyses of the total RNA isolated from R. rickettsii cells cultured in Vero cells. Ethidium bromide-stained 1% agarose gel in 1× TAE (Tris-acetate-EDTA) buffer is shown. RT-PCR products are shown: lane 1, 372 bp using forward primer AZ1372 (lepB) and reverse primer AZ1374 (lepB); lane 2, 1,509 bp using forward primer AZ1529 (nuoF) and reverse primer AZ1375 (lepB); lane 3, 1,475 bp using forward primer AZ1533 (secF) and reverse primer AZ1535 (nuoF); lane 4, 1,894 bp using forward primer AZ1533 (secF) and reverse primer AZ1375 (lepB); lane 5, 1,292 bp using forward primer AZ1528 (secF) and reverse primer AZ1531 (nuoF); lane 6,732 bp using forward primer AZ1556 (165 nucleotides upstream from secF start codon) and AZ1568 (secF); and lane 7, 967 bp using forward primer AZ1372 (lepB) and reverse primer AZ1566 (rnc′). The control PCR using the same primer sets (used for RT-PCR analysis) on the total RNA of R. rickettsii produced no detectable product (data not shown), indicating no DNA contamination in the total RNA used in this analysis. The specificity of each primer pair (used for RT-PCR) to amplify the target sequence was checked (data not shown) by PCR on template DNA. GeneRuler 100-bp DNA ladder plus (MBI-Fermentas) was used as a DNA size marker (lanes M).

Expression and functional analysis of the rickettsial lepB gene in E. coli.

The type I signal peptidase activity of the rickettsial lepB gene was assayed by genetically complementing the temperature-sensitive E. coli strain IT41. The E. coli strain IT41, which has a nonsense mutation in the lepB gene, shows normal growth at 30°C, but the preprotein processing and cell growth are severely affected at 42°C (9). The strain IT41 has been used to demonstrate the complementation ability of many gram-negative and gram-positive bacterial type I signal peptidase genes (24).

The E. coli strain IT41 was transfected with pRRlepB569 and pRTlepB23 and with the control plasmids pESL4 and pUC18. The temperature-sensitive growth was assayed as described in Materials and Methods. It is clearly observed from the growth curves shown in Fig. 5 that the E. coli strain IT41 carrying the plasmid pRRlepB569 (carrying the 1,592-bp fragment of the R. rickettsii lepB gene) or pRTlepB23 (carrying the 1,588-bp fragment of the R. typhi lepB gene) grew much faster than the E. coli strain IT41 with or without control plasmids pESL4 and pUC18 at the nonpermissive temperature of 42°C, thereby indicating functional complementation of the lepB gene from R. rickettsii or R. typhi in E. coli strain IT41. For quantitative comparison of the growth at the nonpermissive temperature of the transformed E. coli strain IT41, survival was also determined by CFU assay. Survival as measured by CFU of E. coli strain IT41 at 42°C was 0.081 ± 0.017% of that grown at 30°C. The control plasmids pESL4 and pUC18 were unable to improve the growth of the strain IT41 at 42°C. However, the temperature-sensitive E. coli strain IT41 carrying the plasmid pRRlepB569 or pRTlepB23 showed a substantial increase in the growth at 42°C by 73.37 ± 22.64% or 78.68 ± 12.82% (compared to that at 30°C), respectively, indicating the functional expression of the lepB gene from R. rickettsii or R. typhi in E. coli.

FIG. 5.

Growth curves showing the complementation in E. coli strain IT41 transfected with appropriate plasmids (as mentioned in Materials and Methods). Cultures pregrown at 30°C were diluted 100-fold into LB-ampicillin broth (IT41 cells without plasmid were grown in absence of ampicillin) and incubated with shaking at 42°C. The growth of cells in culture was monitored by optical density (OD) at 600 nm.

The expression of rickettsial signal peptidase I (expected size, 31 kDa) in E. coli was too low to detect by Coomassie brilliant blue staining of SDS-polyacrylamide gel electrophoresis-separated proteins. Therefore, the expression of rickettsial signal peptidase I in E. coli was investigated by cloning the coding sequence (804-bp ORF) of the R. rickettsii lepB gene at the BamHI and EcoRI sites of pTrcHisC vector containing an N-terminal His₆ tag under the trc (trp-lac) promoter. The expression of the recombinant protein was confirmed by Western blot analysis by using a monoclonal antibody to the N-terminal His₆ tag. A band of approximately 35 kDa was recognized for the E. coli strain IT41 carrying pTrcHisRR4 plasmid (Fig. 6, lane 1) that was not detected for control expression (Fig. 6, lanes 2 and 3). Two minor bands (one around 32 kDa and another below 32 kDa) in lane 1 of Fig. 6 may have resulted from the nonspecific binding in the total proteins or an autocatalytic cleavage, which was previously reported for the E. coli leader peptidase and Bacillus subtilis SipS (29). Complementation analysis performed by using the constructed plasmid pTrcHisRR4 showed a significant increase in the growth of the temperature-sensitive E. coli strain IT41 at 42°C (Fig. 5) and that also assayed by CFU restored the survival by 94.95 ± 4.45%. However, the control plasmid pTrcHisC was unable to restore the growth of the temperature-sensitive E. coli strain IT41 at the nonpermissive temperature.

FIG. 6.

Western blot analysis of the expression of the R. rickettsii signal peptidase I in E. coli strain IT41. Total proteins from the E. coli cells carrying pTrcHisRR4 or pTrcHisC plasmids, separated on 4 to 12% Tris-glycine precast gel, 1× Tris-glycine-SDS running buffer, transferred to polyvinylidene difluoride membrane was probed with His-Tag monoclonal antibodies by using a WesternBreeze chromogenic immunodetection kit. Lane 1, total proteins from E. coli IT41/pTrcHisRR4; lane 2, total proteins from E. coli IT41/pTrcHisC; and lane 3, total proteins from E. coli IT41. Lane M, Bio-Rad Kaleidoscope prestained markers (carbonic anhydrase, 39.7 kDa; soybean trypsin inhibitor, 32.1 kDa).

DISCUSSION

In order to elucidate the mechanisms of protein secretion of rickettsiae, we focused on the characterization of the type I signal peptidases of R. rickettsii and R. typhi. In this communication, we describe the cloning, sequence analysis, transcription, and functional expression of the putative lepB gene that encodes type I signal peptidase of R. rickettsii and R. typhi. The residues of amino acids serine 90 and lysine 145 (E. coli signal peptidase I numbering), which are considered critical for catalytic activity of signal peptidase I in gram-negative and gram-positive bacteria and which are thought to form a catalytic dyad (10), are found to be conserved in the putative signal peptidase I of R. rickettsii and R. typhi. The catalytic domains (boxes B through E) found in bacterial signal peptidase I (10, 24) are also shown to be conserved for the rickettsial signal peptidase I.

Type I signal peptidases from many gram-negative bacteria—including E. coli and Salmonella enterica serovar Typhimurium—have two transmembrane domains at the N terminus for assembly of the enzyme into the membrane and a carboxy-terminal catalytic domain (19, 24). However, type I signal peptidases from gram-positive bacteria (e.g., B. subtilis, Staphylococcus aureus, and Streptococcus pneumoniae) and some gram-negative bacteria (e.g., Bradyrhizobium japonicum and Rhodobacter capsulatus) (6, 19, 24), including rickettsiae, are smaller in size and have only a single transmembrane segment at the N terminus for its assembly into the membrane. The carboxy terminus carrying the conserved catalytic domains (boxes B through E) is also smaller in size compared with that of gram-negative E. coli (10, 24).

The analysis of the recently published genome sequences of R. prowazekii (2), R. conorii (23), Rickettsia sibirica, and the lepB sequence of R. rickettsii reported in this communication reveal that the putative genes secF (encoding protein export membrane protein SecF) and nuoF (encoding NADH dehydrogenase I chain F) are located upstream of the putative lepB (encoding type I signal peptidase) gene. We also show that the putative gene rnc (encoding RNase III, partial sequence of the 5′ end) (Fig. 4A) is located downstream of lepB, such that the termination codon of the signal peptidase I (lepB) gene overlaps with the initiation codon of the RNase III (rnc) gene. The RT-PCR data presented here demonstrate that the putative genes secF, nuoF, lepB, and rnc are transcribed polycistronically from the same promoter in R. rickettsii and that the transcription start site is located further upstream of the polycistronic message of the secF-nuoF-lepB-rnc gene cluster. In Northern analysis, the presence of the 4.5-kb band further supports our explanation that the secF-nuoF-lepB-rnc gene cluster (approximate transcript size, 4.0 kb) (23) cotranscribes in a single polycistronic message in R. rickettsii.

Although polycistronic transcripts usually encode products involved in a common pathway (e.g., the trp and lac operon in E. coli), there are reports that the polycistronically transcribed lep operon (lepA and lepB genes) and lsp locus (lsp and ileS genes) in E. coli have unrelated functions (12, 18). The putative secF and lepB gene products are considered to be involved in the same protein secretion pathway (22); however, the cotranscription of the nuoF and rnc genes of secF-nuoF-lepB-rnc clustered in R. rickettsii could not be explained in terms of related functions. Genome analysis of rickettsiae (1, 2, 23) revealed that genome reduction is an ongoing process for obligate intracellular parasites. It was suggested that this reduction is due to the redundancy of the parasite genes for enzymatic activities supplied by the host cell. Therefore, intracellular parasites typically have fewer genes that code for biosynthetic functions than do free-living bacteria. Thus, one possible explanation for the coordinated expression of seemingly unrelated genes is reduction of transcriptional control. However, it is also possible that the functions may be related by an as-yet- unknown manner for obligate intracellular parasites.

The expression of the putative lepB gene of R. rickettsii and R. typhi from a plasmid in E. coli produced active type I signal peptidase, as demonstrated by complementation assay in this study in an E. coli strain IT41 that was temperature sensitive for preprotein processing at the nonpermissive temperature (42°C). The positive correlation between E. coli strain IT41 growth and the catalytic activity of plasmid-borne signal peptidase I at the nonpermissive temperature has been used to demonstrate the enzymatic activity of the putative type I signal peptidase gene from other gram-negative and gram-positive bacteria (8, 9, 24, 27, 28). Our complementation data presented here indicate that proteins that are processed by E. coli signal peptidase I and are essential for E. coli are also processed by the putative type I signal peptidase of R. rickettsii and R. typhi.