Complement Activation in Mycoplasma fermentans-Induced Mycoplasma Clearance from Infected Cells: Probing of the Organism with Monoclonal Antibodies against M161Ag

Abstract

Mycoplasma fermentans, a cell wall-less prokaryote, is capable of infecting humans and has been suggested to serve as a cofactor in AIDS development. Recently, we discovered a novel lipoprotein with a molecular mass of 43 kDa originating from M. fermentans. This protein, named M161Ag, activated human complement via the alternative pathway and efficiently induced the proinflammatory cytokines interleukin 1β (IL-1β), tumor necrosis factor alpha, IL-6, IL-10, and IL-12 in human peripheral blood monocytes. It is likely that M161Ag of M. fermentans affects the host immune system upon mycoplasma infection. In this study, we developed monoclonal antibodies (MAbs) against M161Ag and examined the direct role of complement in M. fermentans infection using these MAbs as probes. M. fermentans was rapidly cleared from the surfaces of infected cells by human complement, but a low-grade infection persisted in human tumor cell lines. Mycoplasma particles remaining alive in host cells may cause recurrent infection, and liberated M161Ag may serve as a biological response modifier affecting both innate and acquired immunity.

The complement system is central to innate immunity. It can directly recognize invading substances via the alternative, or lectin, pathway and facilitate removal of infectious organisms by phagocytes (9, 11). Deposition of the third component of complement (C3) is a critical factor for host defense. Several molecules of viral or microbial origin have been identified as activators of human complement (2, 32, 35).

Recently, we discovered a membrane-associated novel C3-activating protein in human tumor cell lines (18–20). Based on the genomic analysis, it was found to originate from Mycoplasma fermentans (21). This protein, designated M161Ag, is a palmitoylated protein with a molecular mass of 43 kDa (21). It activates human complement via the alternative pathway, allowing the deposition of C3b and C3bi on human cells infected by M. fermentans and thus overcoming the functions of the complement regulatory proteins, CD46 and CD55, expressed on these cells (1, 18, 19). Interestingly, M161Ag efficiently promotes the production of interleukin 1β (IL-1β), tumor necrosis factor alpha (TNF-α), IL-6, IL-10, and IL-12 in human peripheral blood monocytes (21). Thus, M161Ag is a bifunctional protein which elicits the innate immune responses via complement activation and stimulation of monocytes.

M. fermentans is a mycoplasma species capable of infecting humans and has been suggested to serve as a cofactor during the development of AIDS (3, 17). M. fermentans DNA has been detected in the peripheral blood mononuclear cells of patients with AIDS by PCR (8, 12). In addition, the products of M. fermentans affect the host immune system via B- or T-cell activation, monocyte/macrophage stimulation, and cytocidal ability (6, 7, 25, 26, 28). However, its role as a cofactor in human immunodeficiency virus disease is still unknown. Recent studies suggest that AIDS-associated mycoplasma species, including M. fermentans, can invade host cells (30, 38), but direct evidence for the latent infection of human cells by M. fermentans has not been found. Furthermore, the role of complement in defense against M. fermentans infection has not been elucidated. In this study, we established monoclonal antibodies (MAbs) against M161Ag and demonstrated a rapid targeting of M. fermentans by human complement using MAbs as probes.

MATERIALS AND METHODS

Antibodies, cells and reagents.

MAbs against M161Ag (M161) and CD46 (M177) were produced and purified in our laboratory as described previously (19, 34). Anti-human C3b MAb (C5G) and anti-CD55 MAb (IA10) were gifts from K. Iida (Takeda Chemical Industries) and T. Kinoshita (Osaka University), respectively (10, 13). Mouse immunoglobulin G (IgG) was purchased from Sigma Chemical Co. (St. Louis, Mo.). Fluorescein isothiocyanate (FITC)-labeled goat F(ab′)₂ anti-mouse IgG was from Cappel (West Chester, Pa.), and horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG and HRP-labeled anti-rabbit IgG were from Bio-Rad Laboratories (Hercules, Calif.).

Gelatin veronal-buffered (GVB) saline containing 2 mM MgCl₂ and 10 mM EGTA (Mg²⁺-EGTA-GVB) or 10 mM EDTA (EDTA-GVB) was used in the C3 deposition assay. Normal human serum (NHS) was collected from 20 healthy donors and stored in aliquots at −70°C. Antibody to M. fermentans was less than the detection limit (1 ng/ml) by enzyme-linked immunosorbent assay in the pooled NHS (data not shown). A 1/20 volume of 40 mM Mg²⁺–200 mM EGTA (pH 7.4) or 200 mM EDTA (pH 7.4) was added to NHS in the preparation of either Mg²⁺-EGTA-NHS or EDTA-NHS.

Human leukemia cell lines, P39 and CEM, were provided by the Japanese Cancer Research Resources Bank. K562 (a chronic myelogenous leukemia cell line) and Jurkat (a T-cell leukemia cell line) were gifts from J. P. Atkinson (Washington University) and S. Nagasawa (Hokkaido University), respectively. The cells were maintained in RPMI 1640 supplemented with 10% fetal calf serum (FCS) (CSL Ltd., Victoria, Australia) in the presence of antibiotics. M. fermentans-infected and M161Ag-expressed leukemia cell lines were denoted (+).

Preparation of MAbs against M161Ag.

MAbs were produced by the method of Köhler and Milstein (15). M161Ag, partially purified from P39(+) cell lysates using mouse IgG-Sepharose, Q-Sepharose, and chromatofocusing columns as described previously (19), was mixed with TiterMax (CytRx Co., Norcross, Ga.) and injected subcutaneously into female BALB/c mice once every week for a total of three times. After 1 week, P39(+) cells (8 × 10⁶) were administered intraperitoneally as a final booster. Three days later, the spleens were extracted and the cells were fused with the mouse myeloma cell line NS-1. The supernatants of hybridomas were screened by Western blotting and protein A rosette assay using P39(+) and -(−) cells (19). Clones producing a MAb that reacted with a 43-kDa protein in P39(+) cells but not in P39(−) cells were established by limiting dilution. Three MAbs, MK53, MK5, and MK36, were purified from mouse ascites fluid by ammonium sulfate precipitation followed by protein G-Sepharose (Amersham Pharmacia Biotech).

Flow cytometry.

Cells (10⁶) suspended in 50 μl of Dulbecco's phosphate-buffered saline (DPBS) containing 0.5% bovine serum albumin (BSA) and 0.1% NaN₃ (BSA-NaN₃-DPBS) were mixed with 50 μl of EDTA-plasma and 5 μg of mouse IgG or MAb and incubated for 30 min at 4°C. After being washed with BSA-NaN₃-DPBS, the cells were suspended in 90 μl of BSA-NaN₃-DPBS and incubated with FITC-labeled goat F(ab′)₂ of anti-mouse IgG at 4°C. After 30 min, the cells were washed twice with DPBS and fixed with paraformaldehyde. The samples were analyzed on an EPICS Profile II (Coulter Corp., Hialeah, Fla.). C3 deposition was assessed as described previously (18). Briefly, 10⁶ cells were incubated with 25% Mg²⁺–EGTA–NHS or EDTA-NHS for 30 min at 37°C. After the cells were washed, C3 fragments bound to the cells were detected with anti-human C3b or C3bi MAb followed by a FITC-labeled secondary antibody.

Immunoblotting.

Cells (10⁷) were lysed in 200 μl of lysis buffer (1% NP-40, 10 mM EDTA, 25 mM iodoacetoamide, 2 mM phenylmethylsulfonyl fluoride, DPBS) for 20 min at room temperature. After centrifugation at 100 × g for 10 min, the supernatant was centrifuged again at 200,000 × g for 1 h at 4°C. Aliquots of 50 μl of the supernatant were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (10% gel) under nonreducing or reducing conditions. After electrophoresis, the resolved proteins were transferred onto nitrocellulose sheets. The sheets were then blocked with 10% skim milk for 1 h at 37°C and then overnight at 4°C and sequentially incubated with MAb and HRP-conjugated goat anti-mouse IgG, followed by staining with an ECL kit (Amersham Pharmacia Biotech). Mycoplasmas grown in the growth medium were centrifuged at 16,000 × g for 30 min, and the cell pellets were washed twice with PBS and resuspended in 500 μl of PBS. The cell suspension was sonicated at 20 kHz for 3 min and used as the mycoplasma cell lysate (31).

Immunoprecipitation.

Cell lysates (50 μl) were precleared with protein G-Sepharose at 4°C for 1 h and incubated with 5 μg of mouse IgG or MAb at 4°C. After 1 h, protein G-Sepharose was added to the mixtures and reacted overnight with rotation. The Sepharose beads were washed three times with 0.2% NP-40, 25 mM iodoacetoamide, 2 mM phenylmethylsulfonyl fluoride, and PBS (pH 7.4) and incubated in 1% SDS, 0.2% NP-40, and PBS (pH 7.4) for 2 min at 100°C. Immunoprecipitates were analyzed by SDS-PAGE followed by immunoblotting.

Immunostaining.

P39(+) cells (10⁶) were labeled with 1 μg of mouse IgG or MAb against M161Ag (MK53) in 100 μl of RPMI supplemented with 10% FCS for 45 min at 4°C. After being washed twice with medium, the cells were incubated with FITC-conjugated goat anti-mouse IgG for 45 min at 4°C. The cells were washed three times with medium and then immediately observed under a fluorescence microscope. For double-staining of M161Ag and C3 fragments, P39(+) cells were pretreated with 25% EDTA-NHS or Mg²⁺-EGTA-NHS for 20 min at 37°C and then sequentially labeled with MAb against human C3b (C5G), rhodamine-conjugated secondary antibody (TAGO Inc., Burlingame, Calif.), and MK53-FITC-conjugated secondary antibody. For staining of Mycoplasma DNA, cells were fixed with 1% glutaraldehyde for 30 min at room temperature. After being washed with DPBS, the cells were stained with 0.17 mM Hoechst 33258 (Molecular Probes Inc., Eugene, Oreg.) and subjected to fluorescence microscopy.

M. fermentans PG18, M. fermentans incognitus, and Mycoplasma pulmonis were grown in a broth medium consisting of 2.1% PPLO broth base (Difco), 10% horse serum, 0.002% phenol red, and 0.25% glucose (31). The propagated mycoplasmas (10⁷ CFU/ml) were washed three times with DPBS and stained with MAbs against M161Ag followed by rhodamine-conjugated anti-mouse IgG. For the C3 deposition assay, the mycoplasmas were pretreated with 25% EDTA–NHS or Mg²⁺-EGTA-NHS for 20 min at 37°C and then sequentially labeled with MAb against human C3b (C5G) and rhodamine-conjugated secondary antibody.

Immuno-electron microscopy.

The method for preparing electron microscopic specimens (for fixation, dehydration, and embedding with Lowicryl K4M resin) was described previously (37). The ultrathin sections of P39(+) cells cultured in RPMI supplemented with 10% FCS or 25% fresh human serum (FS) for 5 days on nickel grids were incubated with MK53 or mouse ascites fluid as a control. Next, they were reacted with colloidal gold (15-nm diameter)-conjugated secondary antibody (Amersham Pharmacia Biotech) and observed in a JEOL 100CX electron microscope operated at 80 kV.

PCR.

P39(+), CEM(+), and Jurkat(+) cells (10⁷) were washed with RPMI and suspended in 10 ml of RPMI supplemented with 10% FCS, 25% heat-inactivated human serum (HIS), or 25% FS. The cells were cultured at 37°C in 5% CO₂, and the medium was changed every other day. After 5 days, the cells were washed twice with RPMI, suspended in 10% FCS–RPMI, and maintained in culture. Genomic DNAs were isolated from the cells using an Iso Quick nucleic acid extraction kit (ORCA Research Inc., Chatsworth, Calif.). M161Ag was amplified using forward (5′-TTGAGTCCTATTGCTGCTATT-3′) and reverse (5′-CACCAAATGCAACAACTCT-3′) primers with Taq-gold (Takara) and 1 μg of DNA template. PCR conditions were 95°C for 10 min followed by 35 cycles of 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s for denaturation, annealing, and extension, respectively. M. fermentans was detected by PCR using Mycoplasma genus-specific rRNA primers as described previously (21). The PCR products were electrophoresed on 1 or 3% agarose gels and stained with ethidium bromide.

RESULTS

Development of MAbs against M161Ag.

We raised three MAbs against M161Ag, MK53, MK5, and MK36, all of which were of the IgG1(κ) subclass. When used in Western blotting, these MAbs recognized a 43-kDa molecule in M. fermentans and P39(+) (M. fermentans-infected cell line) cell lysates but not in P39(−) (noninfected cell line) cell lysates in a fashion similar to that of M161, a previously obtained MAb against M161Ag (Fig. 1A and B). Unlike M161, MK53, MK5, and MK36 immunoprecipitated the antigen in P39(+) cell lysates, indicating that these MAbs efficiently reacted with native M161Ag (data not shown). The immunoreactivity of each MAb to cell surface M161Ag was then measured by flow cytometry using P39(+) and -(−) cells. All MAbs reacted with cell surface M161Ag on P39(+) cells, while none of the MAbs reacted with P39(−) cells (Fig. 1C).

FIG. 1.

Establishment of MAbs against M161Ag. (A) Western blot of M. fermentans PG18 with MAbs against M161Ag. Ten microliters of mycoplasma cell lysate (1.0 mg/ml) was applied to each lane. SDS-PAGE was performed under reducing conditions. Lane 1, Coomassie blue staining of M. fermentans PG18 proteins. Samples were transblotted onto nitrocellulose sheets and detected with the indicated MAbs (lanes 2 to 5). Mouse IgG was used as a control (lane 6). Molecular mass markers are shown to the left. (B) MAbs recognized the 43-kDa protein in P39(+) cells but not P39(−) cells. P39(+) and -(−) cell lysates were subjected to SDS-PAGE (10% gel) under reducing or nonreducing conditions and transferred onto nitrocellulose sheets. Each sheet was blotted with M161, MK53, MK5, or MK36 and then with HRP-conjugated goat anti-mouse IgG. NR, nonreducing conditions; R, reducing conditions. The positions of molecular mass markers are shown on the right. (C) Flow cytometric analysis of M161Ag on the P39 sublines. Cell surface M161Ag was assessed using flow cytometry with each MAb followed by a FITC-labeled anti-mouse IgG using P39(+) and -(−) cells.

C3 deposition on M. fermentans-infected cells.

In human leukemia cell lines infected by M. fermentans, expression of M161Ag was examined by Western blotting and flow cytometry using M161 and MK53. In the T-cell lines Jurkat(+) and CEM(+), M161Ag was detected with MK53 but not with M161, while in the myeloid cell lines, P39(+) and K562(+), M161Ag expression was detected with both antibodies (Fig. 1C and 2; blotting data not shown). An isoform-typing study of M161Ag suggested that P39(+) and K562(+) cells had M161Ag-1 (His¹³⁹), Jurkat(+) cells had M161Ag-2 (Tyr¹³⁹), and CEM(+) cells had M161Ag-3 (Tyr¹³⁹ with Ala²⁸⁵ insertion) (20). Since M161 MAb reacts with both recombinant M161Ag-1 and M161Ag-2 (M. Matsumoto and T. Seya, unpublished data), M161 epitope may be hidden in Jurkat(+) and CEM(+) cells. In contrast, MK53 efficiently detected M161Ag in any infected cells, a reaction profile shared with the other MAbs, MK5 and MK36 (data not shown).

FIG. 2.

C3 deposition induced on M. fermentans-infected cell lines. Jurkat(+), CEM(+), K562(+), P39(+), and P39(−) cells were analyzed for M161Ag expression and C3 deposition by flow cytometry. C3 deposition was assessed after the cells were treated with 25% Mg²⁺-EGTA-NHS or EDTA-NHS at 37°C for 20 min, using anti-human C3b MAb and FITC-labeled secondary antibody. The levels of complement regulatory proteins, CD46 and CD55, in each cell line are also shown. Of note, CEM(+) cells lack CD55 expression. The P39(−) cells (M. fermentans-uninfected cell line) do not express M161Ag and do not induce C3 deposition.

Homologous C3 deposition was induced on these cells but not on P39(−) cells after treatment with Mg²⁺-EGTA-NHS, indicating that the cells infected by M. fermentans are targets for human complement via the alternative pathway regardless of their M161Ag isoforms. However, the extent of C3 deposition varied among cells with equivalent levels of M161Ag, and there was no relation between the levels of complement inhibitors (CD46 and CD55) and C3 deposition (Fig. 2). It is possible that C3-binding molecules expressed on the affected cells and/or mycoplasma may participate in C3 deposition. The C3-bearing cells were not lysed because of the presence of CD59, which protects cells from homologous complement-mediated lysis (16).

We next analyzed the relationship between M161Ag expression on M. fermentans and complement activation via the alternative pathway in the absence of host cells. M161Ag was shown to be expressed on the cell surface of M. fermentans PG18 (Fig. 3B) and incognitus (data not shown) by immunostaining using MK53. After treatment with 25% Mg²⁺–EGTA–NHS, C3 fragments bound directly to the mycoplasmas (Fig. 3D).

FIG. 3.

Immunostaining of M161Ag and C3 fragments deposited on M. fermentans. M. fermentans PG18 was reacted with mouse IgG1 (a and A) or MK53 (b and B) and then with rhodamine-conjugated secondary antibody. For the C3 deposition assay, the organisms were incubated with 25% EGTA–NHS (c, C, d, and D) or 25% EDTA–NHS (e and E). After the cells were washed, C3 fragments deposited on the organisms were stained with anti-human C3b MAb and rhodamine-labeled secondary antibody (D and E). Mouse IgG1 was used as a control antibody (C). The mycoplasmas were observed under a phase-contrast microscope (a to e) or a light fluorescence microscope (A to E; the same field as in panels a to e). Magnification, ×1,000.

Clearance of M. fermentans by human complement.

To understand the role of complement in defense against M. fermentans infection in vivo, infected cells [P39(+) and Jurkat(+)] were cultured in RPMI in the presence of either 25% FS, 25% HIS, or 10% FCS. After 1 day, M161Ag was completely depleted from the cells cultured in FS (Fig. 4A). HIS did not affect M161Ag expression in P39(+) cells. In Jurkat(+) cells, the surface M161Ag level was considerably decreased in HIS culture, but the total protein expression levels were unaffected (Fig. 4B). Immunostaining with MK53 confirmed the absence of M161Ag on P39(+) cells cultured in FS (Fig. 5G). Interestingly, the decrease in M161Ag expression was associated with destruction of M. fermentans by complement in culture. Membrane-attached organisms could not be stained with Hoechst 33258 (Fig. 5E), which was further demonstrated by immuno-electron microscopy (Fig. 6B).

FIG. 4.

Disappearance of M161Ag from M. fermentans-infected cells after treatment with FS. (A) Flow cytometric analysis of M161Ag expression. P39(+) and Jurkat(+) cells were cultured in RPMI supplemented with 10% FCS, 25% HIS, or 25% FS for 5 days, and the medium was changed every other day. After 5 days, the cells were washed, resuspended in 10% FCS–RPMI, and maintained in culture. The cells were harvested at each indicated time point, and M161Ag expression was assessed by flow cytometry using MK53. As a control, nonimmune mouse IgG was used. (B) Immunoblotting analysis of M161Ag. The cells were cultured in 10% FCS–RPMI (lanes 1), 25% HIS–RPMI (lanes 2), or 25% FS–RPMI (lanes 3) for 1 day and lysed in the lysis buffer as described in Materials and Methods. The cell lysates were analyzed by SDS-PAGE followed by immunoblotting with MK53. Cells in which M161Ag was diminished were transferred to 10% FCS–RPMI and maintained in culture for 8 days (lanes 4) or 15 days (lanes 5). M161Ag expression was assessed as described above. The arrowhead indicates M161Ag.

FIG. 5.

Double staining of M161Ag and C3 fragments on P39(+) cells. After treatment of P39(+) cells with Mg²⁺-EGTA-NHS, M161Ag and C3 fragments on the same cells were sequentially stained with MK53-FITC-conjugated anti-mouse IgG and anti-C3b MAb-rhodamine-conjugated anti-mouse IgG. M. fermentans cells attached to the membranes of P39(+) cells were stained with Hoechst 33258. (Left) P39(+) cells cultured in 10% FCS–RPMI. (Center) P39(+) cells cultured in 25% FS–RPMI for 1 day. (Right) P39(−) cells cultured in 10% FCS–RPMI. (a to j) Phase-contrast photomicrographs of P39(+) and -(−) cells. (A, E, and I) Hoechst staining. (B and F) Immunostaining with control mouse IgG. (C, G, and J) Immunostaining of M161Ag with MK53. (D and H) Immunostaining of C3 fragments.

FIG. 6.

Clearance of M. fermentans by complement in P39(+) cell culture. P39(+) cells cultured in 10% FCS–RPMI or 25% FS–RPMI for 5 days were subjected to immuno-electron microscopy. (A) Electron micrograph of P39(+) cells cultured in 10% FCS–RPMI. (B) Electron micrograph of P39(+) cells cultured in 25% FS–RPMI. (C) Postembedding immuno-electron micrograph of P39(+) cells cultured in 10% FCS–RPMI. (D) Magnification (10-fold) of the region indicated with an arrow in panel C. The mycoplasma particles are decorated with immunogold reacting with M161Ag.

These mycoplasma-cleared cells were not targets for human complement (Fig. 5H). On the other hand, cells infected by M. fermentans carrying M161Ag were tagged with C3. Double staining of M161Ag and C3 fragments (Fig. 5C and D) and a previous study using blocking antibodies to complement inhibitors (18) showed that C3 fragments were deposited on the organism itself and on further host cell membrane around the organism.

Persistent infection of M. fermentans in human cells.

Next, we studied whether M. fermentans persistently infected human cells by using an in vitro model system. P39(+) or Jurkat(+) cells cultured in FS for 5 days were washed, resuspended in RPMI supplemented with 10% FCS, and maintained in culture. At timed intervals, M161Ag expression was analyzed by flow cytometry, Western blotting, and PCR using genomic DNA isolated from cells. At the same time, in order to detect the organisms, genomic sequences of M. fermentans rRNA were amplified by PCR. As shown in Fig. 4, M161Ag was not expressed on day 8 but appeared on day 15. Interestingly, both M161Ag and M. fermentans were detected by PCR in the P39(+) and Jurkat(+) cells with diminished M161Ag cultured in FS for 5 days (Fig. 7, lane 3), indicating the presence of the organism at low titer in these cells. Both were amplified to a greater degree after the cells were transferred into 10% FCS–RPMI and cultured for 8 days (Fig. 7, lane 6), whereas the M161Ag protein could not be detected by either flow cytometry or immunoblotting (Fig. 4). On day 15, M161Ag protein expression was detectable by flow cytometry and Western blotting, suggesting that M. fermentans escaped from complement attack, regrew, and expressed detectable M161Ag after the removal of complement.

FIG. 7.

PCR analysis of M161Ag. Genomic DNA was isolated from P39(+) cells (left) or Jurkat(+) cells (right). M161Ag (top), genomic sequence of M. fermentans rRNA (center), and β-actin (bottom) were amplified using specific primers. The solid arrowheads indicate M161Ag, and the open arrowheads indicate rRNA of M. fermentans. To confirm the mycoplasma species amplified by PCR, nested PCR was performed (data not shown). Jurkat(+) cells have a correct-size PCR fragment in addition to a smaller PCR fragment, which overlap (center, lanes 6). M, DNA marker; N, negative control. Lanes 1, 2, and 3, cells cultured for 5 days in FCS, HIS, and FS, respectively; lanes 4, 5, and 6, cells cultured for 5 days in FCS, HIS, or FS were transferred to FCS and cultured for 8 days; lanes 7, 8, and 9, cells cultured for 5 days in FCS, HIS, or FS were transferred to FCS and cultured for 15 days.

We then performed immuno-electron microscopy using MAb against M161Ag (MK53) to detect intracellular organisms. M. fermentans was detected in close proximity to P39(+) cells and also intracellularly at low frequency. Both extra- and intracellular mycoplasma particles were nonuniformly decorated with immunogold, which recognized M161Ag (Fig. 6 and 8).

FIG. 8.

Immuno-electron micrograph of M. fermentans in cultured P39(+) cells. Immunogold staining shows extra- and intracellular M. fermentans. An intracellular mycoplasma particle decorated with 15-nm-diameter immunogold is indicated by the arrow. The arrowhead indicates a representative extracellular mycoplasma. Bar = 500 nm.

DISCUSSION

In this study, we focused on the role of complement in defense against M. fermentans infection in vivo by developing MAbs against the mycoplasma lipoprotein M161Ag. The findings from this study can be summarized as follows. (i) MAbs against M161Ag are useful tools to detect M. fermentans. (ii) The cells infected by M. fermentans were targeted by human complement via the alternative pathway, and homologous C3 deposition was rapidly induced on the cells. (iii) M. fermentans was cleared from the surface of infected cells by human complement, but low-grade infection persisted in human tumor cell lines.

As previously reported, M161Ag purified from P39(+) cell lysates activates human complement via the alternative pathway and allows the deposition of C3 fragments on itself (19), which has been proven with the use of a recombinant protein (M. Nishiguchi, M. Matsumoto, and T. Seya, unpublished data). Our data revealed that M161Ag could induce C3 deposition not only on mycoplasma particles but also on mycoplasma-infected cells. Mycoplasma lipoproteins are known to undergo rapid phase and size variation (39). Although the possibility that the level of M161Ag reflects its expression states could not be excluded, most mycoplasma particles and cell-attached organisms expressed M161Ag (Fig. 3 and 6). Recently, Calcutt et al. (4) reported the differential posttranslational processing and intraspecies variation for a major surface lipoprotein of M. fermentans (MALP-404) which is identical to our M161Ag (21). They also demonstrated that most M. fermentans strains express a 41-kDa protein (4).

A previous report indicated that several species of mycoplasmas could activate complement and be rapidly removed from contaminated cells by the addition of fresh animal serum (42). Unlike the results presented here, they showed complete removal of mycoplasmas by the addition of fresh serum and activation of the classical complement pathway. In our study, low-grade infection persisted in the cells even after treatment with FS (Fig. 4 and 7). Immuno-electron microscopic analysis suggested the presence of intracellular M. fermentans, which could explain the persistent infection by the organism. However, the possibility still remains that complement treatment selects organisms that do not express M161Ag and that a few adapted organisms may still be replicating outside the human tumor cells. This possibility is presently under investigation.

Among the mycoplasma strains, M. fermentans may be a unique strain which possesses the alternative complement activator, invades human cells (38), and stimulates monocytes/macrophages. Although our results showed the intracellular persistence of this organism in human tumor cell lines, whether this also occurs in normal peripheral blood cells remains to be examined. Preferential target cells for M. fermentans invasion may exist in the presence of human serum, since cell-surface M161Ag expression was largely diminished in Jurkat(+) cells by HIS but not in P39(+) cells (Fig. 4A).

Our findings, together with previous observations, demonstrated that in the bloodstream, both M. fermentans-infected cells and the organism itself are rapidly tagged with C3 as nonself cells and release the C5a chemotactic factor (18). The C3b and C3bi molecules on the cells enhance the phagocytic activity of complement receptor (CR1 and CR3)-bearing cells and facilitate the elimination of nonself cells by phagocytes (5, 36). In addition to the destruction of cell surface M. fermentans, infected cells may be cleared from the host (27, 41). However, all of the M. fermentans cells (PG18) were not killed by complement during 1 or 2 h of incubation in FS at 37°C (data not shown), suggesting that the organisms may escape complement attack by their rapid invasion of the cells and tissues and persistently infect host cells.

As M161Ag and related products elicit innate immune responses, such as inflammatory cytokine production and nitric oxide synthesis, by monocytes/macrophages at low concentrations similar to lipopolysaccharide (21, 22, 26), liberated M161Ag serves locally as a potent modulator of the host immune system. Recently, Toll-like receptors (TLRs), which provoke innate immune responses, were identified in humans (23, 29). TLR type 2 was identified as a receptor for lipopolysaccharide of gram-negative bacteria (14, 40) and peptidoglycan and lipoteichoic acid of gram-positive bacteria (33). These receptors are thought to recognize pathogen-associated molecular pattern (24). Interestingly, a characteristic motif in M161Ag structure is shared with specific lipoproteins in a variety of bacterial genera, including Borrelia, Listeria, Mycoplasma, and Treponema, all of which possess immune modulatory activity (22). Thus, these lipoproteins may be akin to pathogen-associated molecular pattern recognized by TLRs. The most important structural feature of M161Ag for complement or monocyte activation, as well as the identification of its receptor, are under investigation in our laboratory using deletion mutants of M161Ag and MAbs as probes.