|Home | About | Journals | Submit | Contact Us | Français|
We have previously shown that pathogenic leptospiral strains are able to bind C4b binding protein (C4BP). Surface-bound C4BP retains its cofactor activity, indicating that acquisition of this complement regulator may contribute to leptospiral serum resistance. In the present study, the abilities of seven recombinant putative leptospiral outer membrane proteins to interact with C4BP were evaluated. The protein encoded by LIC11947 interacted with this human complement regulator in a dose-dependent manner. The cofactor activity of C4BP bound to immobilized recombinant LIC11947 (rLIC11947) was confirmed by detecting factor I-mediated cleavage of C4b. rLIC11947 was therefore named LcpA (for leptospiral complement regulator-acquiring protein A). LcpA was shown to be an outer membrane protein by using immunoelectron microscopy, cell surface proteolysis, and Triton X-114 fractionation. The gene coding for LcpA is conserved among pathogenic leptospiral strains. This is the first characterization of a Leptospira surface protein that binds to the human complement regulator C4BP in a manner that allows this important regulator to control complement system activation mediated either by the classical pathway or by the lectin pathway. This newly identified protein may play a role in immune evasion by Leptospira spp. and may therefore represent a target for the development of a human vaccine against leptospirosis.
Leptospirosis, an emerging global infectious disease, is caused by spirochetes belonging to different pathogenic species of the genus Leptospira. In recent decades, large outbreaks of leptospirosis have occurred in many countries, particularly in Southeast Asia and in Central and South America (56). In Brazil, more than 10,000 cases of severe leptospirosis are reported each year due to outbreaks in urban centers (46). Clinical manifestations range from a mild febrile illness to severe disease forms, such as Weil's syndrome, characterized by jaundice, renal failure, and hemorrhage. Leptospirosis-associated severe pulmonary hemorrhagic syndrome has increasingly become recognized as an important manifestation of leptospiral infection (5, 17, 29, 39, 54, 59) and may be considered a relevant prognostic factor associated with fatal outcomes in severe leptospirosis (51).
Pathogenic Leptospira spp. have the ability to disseminate and to trigger a specific immune response after penetrating the host. Like other pathogens, they have evolved strategies to evade innate immune defense systems, thereby causing severe disease. The complement system plays a crucial role in innate immunity, strongly contributing to the elimination of invading microorganisms. During evolution, pathogens have developed several strategies to control the host complement response and evade complement attack, including multiphasic antigenic variation, cleavage of complement activation products by proteases, acquisition of fluid-phase complement regulators, and production of human complement-regulatory analogs (for a review, see reference 60). Moreover, some pathogens are known to exploit multiple mechanisms to inactivate complement attack.
The acquisition of fluid-phase regulators on the surface of a given pathogen normally results in the downregulation of complement activation. Binding of the negative complement regulators factor H (FH) and factor H-related protein 1 (FHR-1) has been demonstrated for serum-resistant and serum-intermediate Leptospira strains (31). Recently, we have shown that pathogenic leptospiral strains also acquire C4b binding protein (C4BP) from the host. Surface-bound C4BP retains its cofactor activity, indicating that acquisition of this complement regulator may contribute to leptospiral serum resistance (3). C4BP is a 570-kDa plasma glycoprotein that displays a spiderlike quaternary structure composed of seven identical α chains and a unique β chain linked together by a central core (11, 22, 48). The α chains each comprise eight complement control protein (CCP) domains (also termed short consensus repeats), while the β chain has three CCP domains (8). C4BP is involved in the downregulation of the classical and/or the lectin pathway of the complement system by interfering with the assembly and decay of the C3 convertase C4bC2a and acting as a cofactor for factor I in the proteolytic inactivation of C4b (15, 48).
In the present study, we investigated whether putative surface proteins encoded by pathogenic Leptospira spp. could bind to C4BP. A novel 20-kDa outer membrane protein (OMP), encoded by the LIC11947 gene, was shown to interact with this complement regulator. To our knowledge, this is the first report describing a Leptospira surface protein that binds the human complement regulator C4BP.
Leptospira biflexa serovar Patoc strain Patoc I, Leptospira interrogans serovar Copenhageni strain 10A, Leptospira interrogans serovar Pomona strain Pomona, Leptospira kirschneri serovar Grippotyphosa strain 8, Leptospira kirschneri serovar Cynopteri strain 3522C, Leptospira noguchii serovar Panama strain CZ 214K, Leptospira borgpetersenii serovar Javanica strain Veldrat Batavia 46, Leptospira borgpetersenii serovar Tarassovi strain 17, and L. interrogans serovar Pomona strain Fromm were used in the assays. All strains are from the Laboratório de Zoonoses, Faculdade de Medicina Veterinária e Zootecnia, Universidade de São Paulo, São Paulo, Brazil. Except for strain Fromm, whose virulence is maintained by iterative passages in hamsters, all pathogenic strains were culture attenuated by successive passages in artificial medium. Leptospires were cultured at 29°C under aerobic conditions in liquid EMJH medium (Difco) with 10% rabbit serum, enriched with l-asparagine (0.015%), sodium pyruvate (0.001% [wt/vol]), calcium chloride (0.001% [wt/vol]), magnesium chloride (0.001% [wt/vol]), peptone (0.03% [wt/vol]), and meat extract (0.02% [wt/vol]) (2).
Normal human sera (NHS) were obtained from healthy human donors. The sera were pooled, divided into aliquots, and stored at −80°C until use. C4BP, purified from normal human plasma, was purchased from Complement Technology, Inc. (Tyler, TX).
Predicted lipoproteins encoded by the LIC10009, LIC10301, LIC10507, LIC10704, LIC11030, LIC11087, and LIC11947 open reading frames (ORFs) were identified from the L. interrogans serovar Copenhageni lipoprotein database (http://mic.sgmjournals.org/cgi/content/full/152/1/113/DC1/7) (50) on the basis of outer membrane localization as predicted by the PSORT program (http://psort.nibb.ac.jp) (36). Predicted structural and functional domains within the amino acid sequences of the selected proteins were identified using the PFAM (http://pfam.sanger.ac.uk/search) (13) and SMART (http://smart.embl-heidelberg.de/) (26, 49) databases. The BLAST program (http://blast.ncbi.nlm.nih.gov/Blast.cgi) was used to identify orthologous genes in the L. bifexa serovar Patoc genome.
ORFs LIC10301, LIC10704, LIC11087, LIC11030, and LIC11947 were amplified without signal peptide tags by PCR from total L. interrogans genomic DNA (strain 10A) using the primer pairs listed in Table Table1.1. PCR fragments were cloned into the pGEM-T Easy vector (Promega Corp., Madison, WI) and transformed into Escherichia coli DH5α. Following digestion with appropriate restriction enzymes, fragments were subcloned into the pAE vector for the expression of recombinant proteins with an N-terminal 6×His tag (45). All constructs were verified by DNA sequencing with appropriate vector-specific primers. The cloning of ORFs LIC10009 and LIC10507 into the pAE vector has been described previously (2, 16). Expression of recombinant proteins in mid-log-phase cultures of E. coli C43 (33) was induced with 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) at 37°C for 3 h. The cells were harvested by centrifugation, resuspended in binding buffer at pH 8.0 (20 mM Tris-HCl and 500 mM NaCl), and lysed by a French pressure cell (Spectronic Instruments Inc., Rochester, NY). The soluble and insoluble fractions were separated by centrifugation at 26,000 × g for 15 min. The His-tagged recombinant proteins rLIC10507 (recombinant LIC10507) and rLIC10704 were purified from the supernatant, and rLIC10009, rLIC10301, rLIC11030, rLIC11087, and rLIC11947 were purified from the insoluble pellet, by nickel affinity chromatography. The soluble fractions were diluted (10-fold) in binding buffer containing 5 mM imidazole. Inclusion bodies were washed with binding buffer (containing 2 mM β-mercaptoethanol, 1 M urea, and 1% Triton X-100) and were then solubilized with binding buffer (containing 8 M urea and 5 mM β-mercaptoethanol) for 16 h at 25°C. For refolding of proteins solubilized in the presence of urea, the suspensions were diluted (50-fold) with binding buffer containing 5 mM imidazole. Protein solutions were loaded onto Ni2+-charged chelating Sepharose columns (GE Healthcare, United Kingdom). After adsorption of proteins, columns were washed sequentially with the same buffer containing 5, 20, 40, and 60 mM imidazole. His-tagged proteins were eluted from the column with 1 M imidazole. Purified proteins were dialyzed extensively against phosphate-buffered saline (PBS), and samples were analyzed by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Protein concentrations were estimated using the Bradford method. The public database accession numbers for each protein sequence analyzed in this work are listed in Table Table2.2. Proteins can also be accessed by their genome nomenclature for the gene locus and LIC (Leptospira interrogans serovar Copenhageni) numbers.
Ten female BALB/c mice (4 to 6 weeks old) were immunized subcutaneously with 10 μg of recombinant protein. Aluminum hydroxide was used as an adjuvant. Two subsequent booster injections of the same protein preparation were given at 2-week intervals. Negative-control mice were injected with PBS and an adjuvant. One week after each immunization, the mice were bled from the retro-orbital plexus, and the pooled sera were analyzed by an enzyme-linked immunosorbent assay (ELISA) for the determination of antibody titers.
Purified recombinant proteins were first subjected to SDS-15% PAGE under reducing conditions and then transferred to a nitrocellulose membrane. Nonspecific binding sites were blocked by using 10% (wt/vol) dry milk in PBS-0.05% Tween (pH 7.4) (PBST) overnight at 4°C. Subsequently, the membrane was rinsed three times in PBST and was then incubated for 90 min at 25°C with 10% NHS diluted in PBST. After five washes with PBST, the membrane was incubated with a mouse monoclonal antibody (MAb) recognizing human C4BP (Quidel, San Diego, CA) at a 1:1,000 dilution. After three washes with PBST, the membrane was incubated with a peroxidase-conjugated secondary antibody against mouse immunoglobulin G (IgG) at a 1:5,000 dilution for 60 min at 25°C. The positive signal was detected by enhanced chemiluminescence (West Pico; Pierce, Thermo Fisher Scientific Inc., Rockford, IL). For the ELISA using nondenatured recombinant proteins, microtiter plates (MaxiSorp; Nunc, Thermo Fisher Scientific Inc., Rochester, NY) were coated with recombinant proteins (100 μl; 10 μg/ml) or bovine serum albumin (BSA) (100 μl; 10 μg/ml) overnight at 4°C. The wells were washed with PBS, blocked with PBS-3% BSA for 2 h at 37°C, and incubated with normal human sera (100 μl; 0 to 5%) or purified C4BP (100 μl; 0 to 50 μg/ml) for 1 h at 25°C. After three washes with PBS, bound proteins were detected with mouse anti-human C4BP (Quidel, San Diego, CA) at a 1:1,000 dilution followed by peroxidase-conjugated anti-mouse IgG (Sigma-Aldrich Co., St. Louis, MO) at a 1:5,000 dilution. The substrate reaction was performed with o-phenylenediamine dihydrochloride (Pierce, Thermo Fisher Scientific Inc., Rockford, IL), and absorbance was measured at 492 nm. For statistical analyses, the binding of rLIC11947 (LcpA) to C4BP was compared to the binding of the negative-control protein BSA or rLIC10301 to C4BP by Student's two-tailed t test. A P value less than 0.05 was considered statistically significant.
The cofactor activity of C4BP bound to intact L. interrogans Fromm cells or to rLIC11947 (LcpA) and rLIC10301 was analyzed by measuring factor I-mediated cleavage of C4b (3). Freshly harvested leptospires (1 × 108) were first washed with a binding buffer (100 mM NaCl, 50 mM Tris-HCl [pH 7.4]) and then incubated with 2 μg purified C4BP for 60 min at 25°C with gentle agitation. Bacteria were washed three times with washing buffer (100 mM NaCl, 50 mM Tris-HCl, 0.05% Tween 20 [pH 7.4]) and were incubated with purified C4b at 500 ng/assay and factor I at 250 ng/reaction (both from Calbiochem-Novachem Corp., San Diego, CA) for as long as 120 min at 37°C. The cells were collected by centrifugation, and the supernatants were subjected to SDS-12% PAGE under reducing conditions and then transferred to nitrocellulose membranes. The blocking treatment and incubations with specific antibodies were performed as described above. C4d, a factor I-dependent cleavage product of C4b, was detected using a mouse monoclonal antibody against human C4d (Quidel, San Diego, CA) at a 1:1,000 dilution. The cofactor activity of C4BP bound to immobilized rLIC11947 (LcpA) was assessed essentially as described previously (18).
OMPs were extracted according to a previously described method (19). Briefly, leptospires cultured as outlined above were washed in PBS containing 5 mM MgCl2 and were then extracted in the presence of 2% Triton X-114 (TX-114) (Sigma-Aldrich Co., St. Louis, MO), 150 mM NaCl, 10 mM Tris-HCl (pH 8.0), and 1 mM EDTA at 4°C for 2 h. The insoluble material was removed by centrifugation at 15,000 × g for 10 min. After centrifugation, 20 mM CaCl2 was added to the supernatant. Phase separation was performed by warming the supernatant at 37°C and then centrifuging for 10 min at 12,000 × g. Three distinct fractions became apparent: the aqueous phase, the TX-114 phase, and the insoluble pellet. The detergent phase was precipitated with four times the sample volume of cold acetone. All fractions were subjected to 12% SDS-PAGE and were transferred to nitrocellulose membranes for immunological analysis with an antiserum against rLIC11947 (LcpA), LipL32, or GroEL.
L. interrogans serovar Pomona strain Fromm cells were washed twice with PBS and were then fixed with 2% paraformaldehyde in PBS for 60 min at 25°C. After two washes in PBS, bacteria were applied to electron microscopy grids and were then incubated for 60 min with anti-rLIC11947 (LcpA), anti-LipL32 (positive control), or a preimmune serum (negative control) diluted 1:100 in PBS containing 1.5% BSA and 0.05% Tween 20. After four washes with PBS plus 1% BSA, cells were incubated for 60 min with a colloidal-gold (particle diameter, 10 nm)-conjugated goat anti-mouse antibody (Sigma-Aldrich Co., St. Louis, MO). After successive washes in PBS plus 1% BSA, 0.85% NaCl, and double-distilled water, bacteria were stained with 2% uranyl acetate and were examined by electron microscopy (Zeiss EM-109 electron microscope; Carl Zeiss, Inc., Thornwood, NY) at an accelerating voltage of 80 kV.
L. interrogans serovar Pomona strain Fromm bacteria (1 × 109) were harvested by centrifugation at 2,000 × g for 10 min, gently washed with PBS containing 5 mM MgCl2, and collected by centrifugation at 2,000 × g for 10 min. Washed spirochetes were then gently resuspended in PBS-5 mM MgCl2, and proteinase K (Sigma-Aldrich) diluted in proteolysis buffer (10 mM Tris-HCl [pH 8.0], 5 mM CaCl2) was added to a final concentration of 12.5 to 100 μg/ml. Proteolysis buffer without proteinase K was added to the negative control. Cell suspensions were incubated for 1 h at 37°C before the reaction was quenched by the addition of 2.5 μl of 100 mM phenylmethylsulfonyl fluoride (Sigma-Aldrich). Bacteria were subsequently collected by centrifugation at 10,000 × g for 5 min and were washed twice with PBS containing 5 mM MgCl2, and the cells were resuspended in sample buffer for SDS-PAGE. Immunoblot analysis was performed using an antibody against rLIC11947 (LcpA) or GroEL.
Leptospira cultures were harvested by centrifugation at 6,000 × g for 25 min and were gently washed twice with sterile PBS. Genomic DNA was isolated from the pellets with a guanidine-detergent lysing solution (DNAzol reagent; Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. DNA fragments were amplified using the oligonucleotides listed in Table Table1.1. PCR was performed in a reaction volume of 25 μl containing 100 ng of genomic DNA, 1× PCR buffer (20 mM Tris-HCl [pH 8.4], 50 mM KCl), 2 mM MgCl2, 10 pmol of each specific primer, 200 μM each deoxynucleoside triphosphate (dNTP), and 2.5 U Taq DNA polymerase (Invitrogen, Carlsbad, CA). Cycling conditions were as follows: 94°C for 5 min, followed by 35 cycles at 94°C for 50 s, 60°C for 50 s, and 72°C for 2 min, with a final extension cycle of 7 min at 72°C. Amplicons were loaded on a 2% agarose gel for electrophoresis and visualization with ethidium bromide. Gel-purified bands (PureLink; Invitrogen, Carlsbad, CA) were cloned into the pGEM-T Easy vector (Promega, Madison, WI), and the products were sequenced with primers M13F (5′-GTTTTCCCAGTCACGA) and M13R (5′-CAGGAAACAGCTATGAC) on an ABI Prism 3730xl sequencer (SeqWright, Houston, TX). Multiple sequence alignment of the deduced amino acid sequences was performed using the Clustal W program (http://www.ebi.ac.uk/clustalw) (21).
Leptospira extracts were fractionated by 15% SDS-PAGE and were transferred to nitrocellulose membranes. Nonspecific binding sites were blocked with 10% (wt/vol) dry milk in PBST overnight at 4°C. After three washes with PBS-T, the membranes were incubated with a mouse anti-rLIC11947 (anti-LcpA) serum (diluted 1:100) in 5% nonfat dry milk-PBST for 60 min. Following three washes with PBST, the membranes were incubated with peroxidase-conjugated anti-mouse IgG as a secondary antibody for 60 min at room temperature at a 1:5,000 dilution, washed, and revealed with ECL reagent (West Pico; Pierce, Thermo Fisher Scientific Inc., Rockford, IL).
The SpLip algorithm (50) allowed identification of 136 putative lipoproteins encoded by the L. interrogans genome. We then used the PSORT program (36) to predict the topological localizations of these proteins. Forty-eight probable outer membrane proteins (PSORT score, ≥70%) were identified. Of these proteins, 7 were given priority because their functions are unknown and they could be expressed and purified in our laboratory in concentrations and purities suitable for the performance of the assays. The signal peptide sequences, GenBank accession numbers, and sizes of these proteins are presented in Table Table2.2. Outer membrane localization was experimentally confirmed for the protein encoded by LIC10507 in a previous study (16) and for the protein encoded by LIC11947 in this work (see below). In addition, LIC10507 and LIC11947 encode hypothetical proteins with no known domains, as assessed by SMART and PFAM analysis (13, 49). In a previous work, rLIC10507 was shown to promote the upregulation of intercellular adhesion molecule 1 (ICAM-1) and E-selectin on monolayers of human umbilical vein endothelial cells (16). The proteins encoded by LIC10704 and LIC11030 present domains of unknown function, DUF400 and DUF1565, respectively. None of these seven proteins have homologs encoded by the genome of the L. biflexa saprophytic strain.
The open reading frames (ORFs) encoding the seven potential outer membrane lipoproteins without signal peptide sequences were amplified by PCR from L. interrogans genomic DNA with specific primers for each gene (Table (Table1).1). The DNA inserts were cloned and expressed as full-length mature proteins in E. coli with a 6×His tag at the N terminus and were purified by nickel affinity chromatography from soluble (rLIC10507 and rLIC10704) or insoluble (rLIC10009, rLIC10301, rLIC11030, rLIC11087, and rLIC11947) fractions as described in Materials and Methods.
To assess the binding of recombinant putative leptospiral outer membrane proteins to C4BP, ligand affinity blotting and ELISA were employed. Proteins were subjected to SDS-PAGE, transferred to nitrocellulose membranes, and then examined for the ability to bind soluble C4BP from normal human sera. C4BP bound only to rLIC11947 protein (Fig. (Fig.11 A). In the absence of NHS or a primary antibody, no bands were detected (Fig. 1B and C). The recombinant proteins encoded by LIC10009, LIC10301, LIC10507, LIC10704, LIC11030, and LIC11087 did not bind to C4BP (data not shown). Binding of rLIC11947 to this complement regulator was further examined by an ELISA using rLIC11947 immobilized in microtiter wells. Dose-dependent binding to both purified and soluble C4BP from human sera was observed (Fig. (Fig.22 A and B). No significant binding to C4BP was detected with BSA or rLIC10301, which were included as negative controls (Fig. 2A and B). Based on these properties, rLIC11947 was named LcpA (for leptospiral complement regulator-acquiring protein A).
C4BP acts as a cofactor for factor I, promoting the cleavage and inactivation of C4b, with the subsequent formation of C4c (145-kDa) and C4d (44.5-kDa) fragments. A monoclonal anti-C4d antibody, which recognizes both the intact α′ chain of C4b and its cleaved product C4d, was used to assess the cofactor activity of C4BP bound to LcpA. Whole bacteria (L. interrogans serovar Pomona strain Fromm) were included as a positive control, since it has been shown previously that C4BP surface bound to this serum-resistant strain efficiently mediates the cleavage of C4b (3). Immobilized recombinant proteins or whole bacteria were first incubated with purified C4BP, and after intensive washes to remove unbound C4BP, C4b and factor I were added. Incubation proceeded for the periods indicated in Fig. Fig.3,3, and the production of C4d in the supernatant was detected by Western blotting using an anti-C4d MAb. The presence of a 45-kDa species corresponding to C4d in the supernatants of both LcpA and whole bacteria indicates that acquired C4BP was able to promote factor I-mediated cleavage of C4b (Fig. (Fig.3).3). A smaller fragment (~40 kDa) from the subsequent cleavage of the C4b α′ chain was also observed when C4b was incubated in the presence of L. interrogans, C4BP, and factor I for 120 min. A similar result was previously observed by our group (3). The 45-kDa fragment was not detected when we used the negative-control protein encoded by LIC10301. We performed additional controls in which we omitted LcpA or C4BP. As expected, the 45-kDa fragment corresponding to C4d was not detected in any control reaction (Fig. (Fig.33).
LcpA was assessed for surface localization by three complementary methods. Cellular fractionation of L. interrogans with the nonionic detergent Triton X-114 yielded three fractions after incubation at 37°C. They were separated by SDS-PAGE, transferred to nitrocellulose membranes, and probed with an anti-LcpA antiserum. LcpA was found to be completely absent from the detergent-insoluble pellet fraction, consisting of inner-membrane-associated components, cytoplasmic contents, and undisrupted cells, but could readily be detected in the TX-114 detergent phase, indicating that this protein is a component of the leptospiral outer membrane (Fig. (Fig.44 A). Control analyses including LipL32, a well-characterized leptospiral outer membrane protein, and GroEL, an abundant cytoplasmic protein, were also performed (Fig. (Fig.4A).4A). LipL32 was detected mostly in the detergent phase, and GroEL could be detected predominantly in the insoluble pellet fraction but was also found in the aqueous fraction, consisting of periplasmic contents. LcpA was also assessed for surface localization by immunoelectron microscopy. Representative micrographs are shown in Fig. Fig.4B.4B. Organisms incubated with preimmune serum did not present bound colloidal-gold particles, while those incubated with anti-LcpA antiserum exhibited particles localized on the surfaces. The number of particles bound to leptospires incubated with anti-LipL32 antiserum (used as a positive control) was much higher, as expected, since LipL32 accounts for as much as 75% of the total outer membrane proteins of pathogenic leptospires (35). Finally, LcpA was shown to be sensitive to proteinase K-mediated degradation of L. interrogans surface proteins, while the cytoplasmic control protein GroEL was not, providing additional evidence that LcpA may be surface exposed (Fig. (Fig.4C).4C). Unexpectedly, LipL32 was not susceptible to proteolysis using the proteinase K concentrations tested (data not shown).
The presence of the gene encoding LcpA among eight serovars from five different leptospiral species was evaluated by PCR. The gene was detected in L. interrogans (serovars Copenhageni and Pomona), L. kirschneri (serovars Grippotyphosa and Cynopteri), L. borgpetersenii (serovars Javanica and Tarassovi), and L. noguchii (serovar Panama) (Fig. (Fig.5A).5A). Although the gene coding for LcpA could not be amplified by PCR in the saprophytic species L. biflexa (serovar Patoc strain Patoc I) by using primers derived from the L. interrogans sequence (Fig. (Fig.5A),5A), a BLASTp search revealed that this strain has a putative protein (YP_001839093) that shares a certain degree of sequence similarity with LcpA (55% similarity and 37% identity over 109 amino acids from a total of 191 amino acids). However, this protein was not detected in crude extracts of L. biflexa by use of an anti-LcpA antiserum (Fig. (Fig.5D).5D). Since L. biflexa serovar Patoc strain Patoc I is not capable of acquiring C4BP efficiently on its surface (3), this protein may be nonfunctional or may carry out functions different from those of LcpA. Alignment of the deduced LcpA amino acid sequences from four pathogenic strains in our collection and from all the pathogenic strains whose genomes have been sequenced (L. interrogans serovar Copenhageni L1-130, L. interrogans serovar Lai strain 56601, L. borgpetersenii serovar Hardjo-bovis strain L550, and L. borgpetersenii serovar Hardjo-bovis strain JB197) (9, 37, 47) indicated that this protein is highly conserved (Fig. (Fig.5B).5B). Since temperature is a key environmental factor known to affect leptospiral protein expression, we performed immunoblot assays with cultures grown at 20°C, 29°C, or 37°C, representing ambient temperatures in the environment, growth under laboratory conditions, and the temperature in mammalian hosts. The same concentration of lysates was applied in all wells. No significant differences in the levels of expression were observed, indicating that the protein seems to be expressed at temperatures ranging from 20°C to 37°C (Fig. (Fig.5C).5C). Our group previously assessed the sensitivity to complement-mediated killing of six strains belonging to five different Leptospira species (3). L. interrogans serovar Pomona strain Fromm and L. kirshneri serovar Cynopteri strain 3522C were classified as resistant, while L. biflexa serovar Patoc strain Patoc I was classified as sensitive. L. interrogans serovar Pomona strain Pomona, L. borgpetersenii serovar Javanica strain Veldrat Batavia 46, and L. noguchii serovar Panama strain CZ 214K were classified as intermediate, since viable leptospires could still be detected after 60 min of incubation but not after 120 min (3). We then tested whether these particular strains produce LcpA by performing Western blot analysis of whole bacterial protein extracts. Our results indicated that the serum-resistant and serum-intermediate strains produce LcpA, while the serum-sensitive strain Patoc I does not (Fig. (Fig.5D5D).
Acquisition of fluid-phase host complement regulators on the surfaces of pathogens is a common complement evasion mechanism. To date, pathogenic Leptospira strains have been shown to bind FH, FHR-1, and C4BP (3, 31). However, only two outer membrane proteins belonging to the Len (leptospiral endostatin-like protein) family, named LenA and LenB, have been reported to interact with a soluble complement regulator, in this case with FH (52, 55). Considering that leptospires are highly invasive microorganisms, these bacteria may be expected to produce many different surface receptors for these and other host molecules.
In this study we have characterized a 20-kDa (calculated molecular mass, 20.1 kDa) outer membrane protein of Leptospira spp., encoded by LIC11947, which interacts with the human complement regulator C4BP. Previous immunoblot studies using sera from leptospirosis patients revealed that this protein (previously referred to as LipL22 based on its recombinant molecular mass and on a predicted lipoprotein cleavage site [http://www.cbs.dtu.dk/services/SignalP]) is expressed during the course of human infection (14). Its usefulness as an antigen for the diagnosis of the disease or as a vaccine candidate remains to be evaluated. In the present study, LipL22 was renamed LcpA, based on its function as a complement regulator-acquiring protein.
In a previous study by our group, we showed that recombinant LcpA (rLIC11947) does not bind to extracellular matrix (ECM) components (2). Interestingly, proteomic and microarray studies analyzing the effects of temperature on gene expression patterns in Leptospira have shown that the gene coding for LcpA (LIC11947; also referred to as LA1957) was upregulated upon an overnight temperature shift from 30°C to 37°C (27, 28). Thus, LcpA gene expression may be upregulated during the early stages of infection. However, no significant differences in protein levels were observed in cultures grown over relatively long periods (at least a week) at 29°C or at 37°C (Fig. (Fig.5C5C).
The present results demonstrate that LcpA is a surface protein that binds both purified and soluble C4BP from human sera in a dose-dependent manner. Moreover, C4BP retains cofactor activity when bound to LcpA, as indicated by the production of the C4d cleavage fragment after incubation with C4b and factor I (Fig. (Fig.3).3). Cellular localization assays indicated that the protein is a component of the leptospiral outer membrane, which is compatible with its role in establishing interactions with a host fluid-phase complement regulator. Sequence analyses of pathogenic Leptospira species indicate that LcpA is highly conserved. Moreover, this protein is expressed by leptospiral strains that are at least partially able to resist complement-mediated killing. These observations suggest that the binding of LcpA to C4BP may contribute to leptospiral serum resistance to host complement.
Sequestration of host C4BP as a strategy for complement evasion has been described for a number of pathogens, including Gram-negative and Gram-positive bacteria (Neisseria gonorrhoeae, Neisseria meningitidis, Streptococcus pyogenes, Streptococcus pneumoniae, Moraxella catarrhalis, Escherichia coli, Yersinia enterocolitica, Bordetella pertussis, Borrelia recurrentis, Borrelia duttonii, Borrelia garinii, Borrelia burgdorferi sensu stricto, Leptospira interrogans, L. noguchii, L. borgpetersenii, L. kirschneri, and Porphyromonas gingivalis) (3, 4, 8, 12, 23, 25, 32, 38, 40, 41, 42, 43, 53), fungi (Candida albicans , Aspergillus fumigatus, and Aspergillus terreus ), and even the nematode Loa loa (20). To date, a few bacterial receptors have been shown to interact with C4BP. M proteins, important virulence factors of Streptococcus pyogenes, bind with high affinity to the complement control protein 1 (CCP1) and CCP2 domains of the C4BP α chain (1, 6) through their hypervariable N-terminal regions (24, 34). Interestingly, the acquisition of C4BP by these bacteria has been correlated with evasion of phagocytosis (10). C4BP also binds N. gonorrhoeae porins 1A and 1B, as well as type IV pili (7, 44), further enabling these pathogens to evade complement. In Escherichia coli K1, the leading cause of neonatal meningitis, serum resistance is mediated by outer membrane protein A (OmpA) (58), which interacts mainly with the CCP3 domain of the C4BP α chain (42). Moraxella catarrhalis ubiquitous surface proteins 1 and 2 (Usp1 and Usp2) also interact with this complement inhibitor, partially contributing to bacterial resistance to complement lysis (38). In spirochetes, a single receptor for C4BP has been described to date. Of unknown identity, this 43-kDa (P43) outer membrane protein was shown to be shared by Borrelia garinii and Borrelia burgdorferi sensu stricto strains (40).
The present report is the first description of a Leptospira surface protein that binds to this human complement regulator. We believe that the acquisition of C4BP by pathogenic Leptospira spp. could, in particular, control activation mediated by the classical pathway of complement, since the lectin pathway does not seem to play an important role in complement-mediated lysis in Leptospira spp. (3). As already mentioned, pathogenic Leptospira spp. also bind FH and FHR-1 (31), indicating that multiple complement evasion strategies may be used by these bacteria to circumvent the host's immune defense systems. After entry into the host, acquisition of FH enables complement subversion in the absence of specific antibodies. Further evasion of antibody-mediated clearance as a consequence of downregulation of the classical pathway may occur upon acquisition of C4BP. Considering that direct binding of C1q to bacterial surfaces or to natural IgM may also occur, we cannot rule out the possibility that the classical pathway would also be activated in the absence of specific antibodies during early stages of infection. Therefore, besides control of the activation of the alternative pathway, downregulation of the classical pathway certainly contributes to the ability of pathogenic Leptospira spp. to survive and disseminate to multiple organs after penetrating the host. Within the past few years, considerable research focusing on pathogen receptors for human complement regulators has been conducted. The identification and characterization of these molecules is of great relevance, since they may represent interesting targets for immune interference.
We thank Shaker Chuck Farah (Instituto de Química, Universidade de São Paulo, São Paulo, Brazil) for critical reading of the manuscript.
This study has benefited from grants provided by FAPESP (09/01778-0 and 06/54701-6) and CNPq (478081/2007-3 and 564618/2008-0).
Editor: A. J. Bäumler
Published ahead of print on 19 April 2010.