Figure 1.
FGF2 stimulates C. trachomatis L2 binding in an HSPG-dependent manner.
(A, B) HeLa cells were serum starved for 2 hr and then infected with C. trachomatis L2 in SFM, in FBS-containing media, or in SFM supplemented with the indicated growth factors (100 ng/mL) for 1 hr. Binding and vacuole formation was quantified at 1 hpi (A) and 20 hpi (B), respectively. Shown is the mean ± SEM of three independent experiments normalized to SFM. *** p<0.001 (C) HeLa cells were treated with heparinase for 2 hr in SFM and infected for 1 hr with C. trachomatis L2 either in SFM, in FBS-containing media, or in SFM supplemented with FGF2 (100 ng/mL). Shown is the mean binding ± SEM normalized to mock-treated cells infected in SFM from four independent experiments. * p<0.05, *** p<0.001
Figure 2.
FGF2 co-localizes with cell surface bound EBs.
HeLa cells were infected for 1 hr with C. trachomatis L2 in SFM supplemented with FGF1 (left panels; 100 ng/mL) or with FGF2 (right panels; 100 ng/mL). (A) Cells were fixed and stained with FGF1 antibody (left panels) or FGF2 antibody (right panels). Cell surface associated EBs and host cell nuclei were visualized by DAPI (blue) staining. Arrows point to EBs. (B) Co-localization of EBs with FGF1 or FGF2 was quantified from at least 8 different fields. The data are expressed as a mean percentage of FGF-associated bacteria (± SEM) compared to total bacteria. *** p<0.001
Figure 3.
FGF2 binds to renograffin-purified C. trachomatis L2 EBs.
Renograffin-purified EBs were incubated with 100 ng/mL of FGF1 (A) or FGF2 (B) in SFM containing 0.1% BSA for 1 hr at 37°C. The EB-FGF mixture was centrifuged onto coverslips, fixed, and stained with DAPI (to visualize the EBs) and FGF1 antibody (A) or FGF2 antibody (B). Red arrows in the enlarged images show co-localization of FGF2 with EBs. (C) Co-localization of EBs with FGF1 or FGF2 was quantified from at least 8 different fields. The data are expressed as a mean percentage of FGF-associated bacteria (± SEM) compared to total bacteria. *** p<0.001 compared to the absence of growth factor (D) EBs that had been either mock- or heparinase-treated for 2 hrs were incubated with FGF2 as described in 3A. Co-localization of EBs with FGF2 was quantified as described in 3C. The values are expressed as a mean percentage (± SEM) of FGF2-associated bacteria compared to total bacteria from three independent experiments. The data are normalized to the mock treated EBs. The difference in FGF2 binding between mock- and heparinase-treated EBs was not statistically significant (P>0.05) (E) EBs and FGF2 were incubated in the presence of isotype matched Goat IgG or FGF2 neutralizing antibody for 1 hr. The FGF2-antibody mixture was centrifuged onto coverslips, fixed, and stained. The values are expressed as a mean percentage (± SEM) of FGF2-associated bacteria compared to total bacteria. The data are normalized to the no antibody control. *** p<0.001
Figure 4.
Cell associated FGF2 contributes to C. trachomatis L2 binding, activation and recruitment of FGFR/FRS2α, and bacterial entry.
(A) Left panel; HeLa cells were transfected with shRNA to GFP (control) or FGF2. After 72 hr, cell lysates were immunoblotted with anti-FGF2 or anti-GAPDH (control) to assess efficiency of depletion. The molecular weights of the five FGF2 isoforms (upper panel) and GAPDH (lower panel) are marked. Right panel; HeLa cells transfected with GFP or FGF2 shRNA were infected with C. trachomatis L2 in SFM and binding was measured at 1 hpi. Shown is the mean binding ± SEM compared to GFP shRNA treated cells from three independent experiments. **p<0.01 (B) HeLa cells that were serum-starved for 2 hr were infected with C. trachomatis L2 for 45 min. Cells were fixed and stained with antibodies to phospho-FGFR (pFGFR; green) or to phospho-FRS2α (pFRS2α; green). EBs and host cell nuclei were visualized with DAPI (blue). Enlargements of the boxed areas are shown to the right. Red arrows point to EBs. (C) HeLa cells were incubated in SFM with DMSO, FGFR inhibitor PD173074 (200 nM), PDGFR inhibitor AG1296 (10 µM), or both inhibitors for 2 hr. Samples were either stimulated with FGF2 (50 ng/mL) for 5 min (upper panel) or infected with C. trachomatis L2 for 45 min (lower panel) in the presence of the indicated drug. Lysates were immunoblotted with antibodies to pFRS2α or to FRS2α. FRS2α migrates as multiple bands, delineated by the bracket. Immunoblots are representative of three independent experiments. (D) The percentage of pFRS2α compared to total FRS2α was quantified using densitometry analysis of the immunoblots (Fig 4C) and normalized to DMSO (DM) treated uninfected samples. The values represent the mean (± SEM) of three independent experiments. *p<0.05 compared to DMSO treated uninfected cells. (E) The effect of FGF2 depletion by shRNA on co-localization of pFGFR with surface bound EBs was quantified in at least 8 different fields. Shown are the average percent of EBs that co-localized with pFGFR (± SEM). ***p<0.001 (F) FGFR and PDGFR have redundant roles in C. trachomatis L2 entry. HeLa cells pre-incubated in SFM with DMSO, PD173074, AG1296, or both inhibitors for 2 hr were infected with C. trachomatis L2 for 1 hr in the presence of drug in SFM. Cells were fixed and analyzed by inside-out staining to distinguish between total cell associated bacterial and internalized bacteria as described in Materials and Methods. The internalization efficiency (the percentage of internalized EBs/total cell associated EBs) in each case was normalized to DMSO (DM) treated cells. Shown is the mean (± SEM) of three independent experiments. ***p<0.001 compared to DMSO treated cells
Figure 5.
C. trachomatis L2 infection stimulates FGF2 expression, production, and release.
(A) Total RNA was isolated from the HeLa cells infected with C. trachomatis L2 for indicated time. fgf2 mRNA was assessed by qRT-PCR and normalized to gapdh mRNA. (B) Total FGF2 protein (cell associated and secreted) was measured by Quantikine kit and normalized to uninfected cells (UI) and normalized for total cell number by quantifying total LDH activity. (C) Cell lysates (upper panel) and cell supernatants (lower panel) of FGF2 were collected from HeLa cells infected with C. trachomatis L2 for the indicated times. FGF2 was detected by immunoblot (upper panel) or by immunoprecipitation (lower panel). GAPDH serves as a loading control.
Figure 6.
C. trachomatis L2 induces a biphasic activation of Erk1/2 which contributes to the induction of fgf2 mRNA expression.
(A) Upper 2 panels: Erk1/2 activation in HeLa cells infected with C. trachomatis L2 for the indicated times was examined by immunoblotting cell lyates with antibodies to phospho-Erk (pERK) and to total Erk. Arrows indicate Erk1/2 (p42/p44). Two peaks of Erk1/2 activation are noted, at 45 min pi and at 10–12 hpi. Lower 2 panels: The same lysates from upper panel were immunoblotted with antibodies to FGF2 or to GAPDH (loading control). In the FGF2 immunoblot, arrows indicate the 24, 22.5/22, 18, and 16 kDa isoforms of FGF2. An abrupt change in FGF2 isoforms is noted at 10 hpi. (B) Total mRNA was isolated from HeLa cells infected with C. trachomatis L2 for the indicated times and the fold change in fgf2 mRNA relative to gapdh mRNA was measured by qRT-PCR. An increase in fgf2 mRNA is detectable by 6 hpi and increases further at 10–12 hpi. **p<0.01, ***p<0.001, compared to uninfected (UI) samples. (C, D) HeLa cells were infected with C. trachomatis L2 for 12 hrs in the presence of 10 µM U0126 (U) between 0–3 hpi or 8–12 hpi. Lysates were immunoblotted with antibodies to pErk or to total Erk. The fold change in fgf2 mRNA relative to gapdh mRNA was quantified by qRT-PCR using total RNA isolated at 12 hpi. Inhibition of either the early peak of Erk activation or the late peak of Erk activation decreased C. trachomatis L2 induction of fgf2 mRNA but did not affect the distribution of FGF2 isoforms. **p<0.01, ***p<0.001 compared to C. trachomatis L2 infected cells. (E, F) HeLa cells were infected with C. trachomatis L2 for 45 min or 12 hrs in the absence or presence of chlorampenicol (CAM; 100 µg/mL) for the indicated times. (E) Erk activation and FGF2 isoforms were detected by immunoblotting cell lyates with the indicated antibody. MOMP serves as a control for bacterial replication. (F) Total RNA was isolated at 12 hpi and the fold change in fgf2 mRNA relative to gapdh mRNA was monitored by qRT-PCR. Bacterial protein synthesis is required for Erk activation and for the change in FGF2 isoforms. *p<0.05, ***p<0.001 compared to uninfected (UI) cells.
Figure 7.
HMW FGF2 isoforms are degraded by C. trachomatis L2-induced host proteases.
(A) HeLa cells were infected with C. trachomatis L2 for 12 hr. For the indicated samples, cycloheximide (CHX; 100 µM) was present from 8–12 hpi. Lysates were immunoblotted with an antibody to FGF2 or GAPDH (loading control). (B) Hela cell lysates were incubated with recombinant CPAF under the indicated conditions and the distribution of FGF2 isoforms was assessed by immunoblotting with an antibody to FGF2. For comparison, the distribution of FGF2 isoforms in C. trachomatis L2-infected HeLa cells (Inf) at 12 hpi is shown in the right lane. Vimentin serves as a positive control for CPAF activity. GAPDH serves as a loading control. (C) HeLa cells were infected with C. trachomatis L2 for 12 hrs and treated either with DMSO, MG132 (MG; 15 µM) or Lactacystin (LC; 10 µM or 50 µM) from 9–12 hpi. Cell lysates were collected at 9 hpi or 12 hpi as indicated and immunoblotted with antibodies to FGF2, MOMP, (control for bacterial replication), or GAPDH (loading control).
Figure 8.
C. trachomatis L2-induced FGF2 release correlates with cell lysis.
FGF2 or LDH in the cell lysate or in the supernatant was measured by FGF2 Quantikine kit or the Cytotox 96 non-radioactive cytotoxicity assay kit, respectively. Shown is the percentage of FGF2 (black bars) or LDH (grey bars) in the supernatant compared to total.
Figure 9.
C. trachomatis L2-induced FGF2 facilitates secondary rounds of infection.
(A) Schematic of experiment. HeLa cells were mock-infected or infected with C. trachomatis L2 for 20 hr in 2% FBS. Conditioned media from mock-infected cells (mock-CM) or from C. trachomatis L2 infected cells (CT-CM) were collected and filtered to remove residual bacteria and cell debris. (B) Renograffin-purified EBs were resuspended in mock-CM or CT-CM and binding to HeLa cells was measured at 1 hpi. Shown is the mean (± SEM) of three independent experiments. ***p<0.001 compared to Mock-CM. (C) Filtered CT-CM was immunoprecipitated with goat IgG (control), FGF2 antibody, or EGF antibody and immunoblotted with FGF2 antibody. The ∼24 kDa band in the IgG sample likely represents a breakdown product of IgG light chain (LC). (D) Renograffin-purified EBs were resuspended with CT-CM that had been immunodepleted (see Panel C) with the indicated antibody and binding to HeLa cells at 1 hpi was quantified. The values are normalized to EB binding in CT-CM without antibody incubation. Shown is the mean (± SEM) of three independent experiments. ***p<0.001 compared to no antibody depletion.
Figure 10.
C. trachomatis serovar E co-opts the FGF2 pathway.
(A) HeLa cells were treated with heparinase for 2 hr in SFM and infected with C. trachomatis serovar E either in SFM or in SFM supplemented with FGF2 (100 ng/mL) for 1 hr. Shown is the representative mean binding (± SEM) from three independent experiments. **p<0.01. (B) HeLa cells were serum starved for 2 hrs and then infected for 1 hr with C. trachomatis serovar E either in SFM or in SFM supplemented with FGF2 (100 ng/mL) in the presence or absence of heparin (1 mg/mL). Shown is the representative mean binding (± SEM) from three independent experiments. ***p<0.001 (C) Renograffin-purified serovar E EBs were incubated with 100 ng/mL of FGF1 or FGF2 in SFM containing 0.1% BSA for 1 hr at 37°C. The EB-FGF mixture was centrifuged onto coverslips, fixed, and stained with DAPI (to visualize the EBs) and FGF1 or FGF2 antibody. Co-localization of EBs with FGF1 or FGF2 was quantified from at least 8 different fields. The data are expressed as a mean percentage of bacteria associated with FGF (± SEM) compared to total bacteria. ***p<0.001. (D) Upper 2 panels: Erk1/2 activation in HeLa cells infected with C. trachomatis serovar E was examined by immunoblotting cell lyates with antibodies to pERK or to total ERK. The 42 and 44 kDa forms of ERK are indicated by the arrows. Middle panel: Cell lysates were immunoblotted with C. trachomatis MOMP antibody. Lower 2 panels: The change in cell-associated FGF2 isoforms in HeLa cells infected with C. trachomatis serovar E for the indicated times was assessed by immunoblotting cell lysates with antibodies to FGF2. GAPDH serves as a loading control. Arrows indicate the 24, 22.5/22, 18, and 16 kDa isoforms of FGF2. An abrupt change in FGF2 isoforms is noted at 10 hpi. (E) Total mRNA was isolated from HeLa cells infected with C. trachomatis serovar E for the indicated time and the fold change in fgf2 mRNA relative to gapdh mRNA was measured by qRT-PCR. An increase in fgf2 mRNA is detectable by 9 hpi and increases further at 12 hpi. ***p<0.001 compared to mock-infected cells. (F) HeLa cells were mock-infected for 72 hrs or infected with C. trachomatis serovar E in 5% FBS for 24, 48, or 72 hrs. Conditioned media were collected, filtered, immunoprecipitated with FGF2 antibody, and immunoblotted with FGF2 antibody. (G) Renograffin-purified serovar E EBs were resuspended in mock-CM or CM from serovar E infected (CT-CM). Binding to HeLa cells was measured at 1 hpi. Shown is the representative mean binding (± SEM) from three independent experiments. ***p<0.001 compared to mock-CM.
Figure 11.
Working model for how C. trachomatis co-opts the FGF2 signaling pathway to enhance infection.
(A) FGF2 binds to EBs and facilitates attachment and internalization through cell surface HSPG and FGFR. Both early (45 min pi, C. trachomatis replication independent) and late (10 hpi, C. trachomatis replication dependent) Erk1/2 activation induce fgf2 transcription, resulting in an increase of all FGF2 isoforms. Subsequent intracellular replication of C. trachomatis activates host cell proteases that lead to the processing and/or loss of the HMW FGF2 isoforms and increased amounts of the 16 and 18 kDa isoforms. Upon host cell lysis, EBs, along with the 16 and 18 kDa FGF2 isoforms, are released. Released FGF2 binds to EBs and serves as a bridging molecule to enhance subsequent rounds of C. trachomatis binding and entry. This positive feedback loop between C. trachomatis and FGF2 may enhance the efficiency and spread of subsequent rounds of infection. (B) L2-FGF2 binding may involve synergistic interactions with OmcB-HSPG interactions. (C) The HSPG-binding domain of OmcB of serovar E is non-functional.