Viral Infection Sensitizes Human Fetal Membranes to Bacterial Lipopolysaccharide by MERTK Inhibition and Inflammasome Activation

Abstract

Chorioamnionitis, premature rupture of fetal membranes (FMs), and subsequent preterm birth are associated with local infection and inflammation, particularly IL-1β production. Although bacterial infections are commonly identified, other microorganisms may play a role in the pathogenesis. Because viral pandemics, such as influenza, Ebola, and Zika, are becoming more common, and pregnant women are at increased risk for associated complications, this study evaluated the impact that viral infection had on human FM innate immune responses. This study shows that a herpes viral infection of FMs sensitizes the tissue to low levels of bacterial LPS, giving rise to an exaggerated IL-1β response. Using an ex vivo human FM explant system and an in vivo mouse model of pregnancy, we report that the mechanism by which this aggravated inflammation arises is through the inhibition of the TAM receptor, MERTK, and activation of the inflammasome. The TAM receptor ligand, growth arrest specific 6, re-establishes the normal FM response to LPS by restoring and augmenting TAM receptor and ligand expression, as well as by preventing the exacerbated IL-1β processing and secretion. These findings indicate a novel mechanism by which viruses alter normal FM immune responses to bacteria, potentially giving rise to adverse pregnancy outcomes.

Figure 1. Viral infection augments LPS-induced FM IL-1β

Human FM explants were treated with (A) No treatment (NT), LPS (100ng/ml), MHV-68 (1.5×10⁴/ml PFU) or both MHV-68 and LPS (n=7); (B) NT, LPS (100ng/ml), HSV-2 (6.4×10²/ml PFU) or both HSV-2 and LPS (n=3); or (C) NT, LPS (1ng/ml), Poly(I:C) (20µg/ml) or both Poly(I:C) and LPS (n=6). (A–C) Supernatants were measured for IL-1β by ELISA. (D) Pregnant wildtype mice were injected on E8.5 with either PBS or MHV-68 (1×10⁶ PFU), and on E15.5 with either PBS or LPS (20µg/kg). After 6hrs mice were sacrificed. The FMs were harvested and homogenized for RNA and IL-1B mRNA levels measured by qRT-PCR. *p<0.05 relative to the NT or PBS control unless otherwise indicated. Data are expressed as mean±SEM.

Figure 2. Viral infection augments LPS-induced FM IL-1β processing and secretion through activation of the NLRP3 inflammasome

(A) Human FM explants were treated with (A) No treatment (NT), LPS (100ng/ml), MHV-68 (1.5×10⁴/ml PFU) or both MHV-68 and LPS (n=7). Lysates were evaluated by Western blot for expression of pro-IL-1β (31kDa) and the 17kDa active form. Blots are from one representative experiment. Bar charts show pro- and active-IL-1β expression as determined by densitometry and normalized to β-actin (n=4–5). (B & C) Human FM explants were treated with NT, LPS, MHV-68 or both MHV-68 and LPS in the presence of either media or (B) a caspase-1 inhibitor (1µM) (n=6); or (C) the NLRP3 inhibitor, MNS (10µM) (n=5). Supernatants were measured for IL-1β by ELISA. *p<0.05 relative to the NT control unless otherwise indicated. Data are expressed as mean±SEM.

Figure 3. MHV-68 infection differentially modulates FM cytokine/chemokine responses to LPS

Human FM explants were treated with no treatment (NT), LPS (100ng/ml), MHV-68 (1.5×10⁴/ml PFU) or both MHV-68 and LPS (n=4); Supernatants were measured by multiplex analysis. *p<0.05 relative to the NT control unless otherwise indicated. Data are expressed as mean±SEM.

Figure 4. HSV-2 infection differentially modulates FM cytokine/chemokine responses to LPS

Human FM explants were treated with no treatment (NT), LPS (100ng/ml), HSV-2 (6.4×10²/ml PFU) or both HSV-2 and LPS (n=3). Supernatants were measured by multiplex analysis. *p<0.05 relative to the NT control unless otherwise indicated. Data are expressed as mean±SEM.

Figure 5. Poly(I:C) differentially modulates FM cytokine/chemokine responses to LPS

Human FM explants were treated with no treatment (NT), LPS (1ng/ml), Poly(I:C) (20µg/ml) or both Poly(I:C) and LPS (n=4). Supernatants were measured by multiplex analysis. *p<0.05 relative to the NT control unless otherwise indicated. Data are expressed as mean±SEM.

Figure 6. Effect of combined viral infection and LPS on FM TAM receptor and ligand expression

(A) Untreated FM explants (n=3) were homogenized and analyzed for expression of TYRO3, AXL, MERTK, GAS6 and PROS1 mRNA by qRT-PCR. (B–C) Human FM explants were treated with no treatment (NT), LPS (100ng/ml), MHV-68 (1.5×10⁴/ml PFU) or both MHV-68 and LPS in the presence of media or rGas6 (50ng/ml) (n=5). Tissues were homogenized for protein and Western blot performed for (B) AXL (~140kDa) and (C) MERTK (~180kDa). Blots are from one representative experiment. Bar charts show AXL and MERTK expression as determined by densitometry and normalized to b-actin. (D) FM explants were treated with NT, LPS (1ng/ml), Poly(I:C) (20µg/ml) or both Poly(I:C) and LPS (n=4). Tissues were homogenized for protein and Western blot performed for AXL and MERTK. Blots are from one representative experiment. Bar charts show AXL and MERTK expression as determined by densitometry and normalized to β-actin. (E–F) Human FM explants were treated with no treatment (NT), LPS, MHV-68 or both MHV-68 and LPS in either the presence of media or rGAS6. Tissues were homogenized for protein and ELISA performed for (E) sMERTK (n=7); (F) GAS6 (n=5), and (G) PROS1 (n=8). *p<0.05 relative to the NT control unless otherwise indicated. Data are expressed as mean±SEM.

Figure 7. Viral infection augments LPS-induced FM IL-1β through inhibition of the TAM receptor pathway

(A) Human FM explants were treated with no treatment (NT), LPS (100ng/ml), MHV-68 (1.5×10⁴/ml PFU) or both MHV-68 and LPS in the presence of media or rGas6 (50ng/ml). (i) IL-1β secretion in supernatants was measured by ELISA (n=6). Lysates were evaluated by Western blot for expression of (ii) pro-IL-1β and (iii) active IL-1β. Bar charts show pro- and active-IL-1β expression as determined by densitometry and normalized to β-actin (n=4–5) *p<0.05 relative to the NT control unless otherwise indicated. (B) Human FM explants were treated with or without LPS (1ng/ml) in the presence of blocking anti-TAM Abs or isotype control Abs. IL-1β secretion was measured and levels expressed as LPS-induced fold change (FC) relative to the NT control (n=4; *p<0.05 relative to the isotype control). (C–D) Pregnant wildtype or AXL^−/−MERTK^−/− mice were injected on E15.5 with either PBS or LPS (20µg/kg). After 6hrs mice were sacrificed. The FMs were harvested and homogenized for RNA. (D) TYRO3, AXL, MERTK, GAS6 and PROS1 mRNA expression levels in FMs from PBS treated wildtype mice (n=3) were measured by qRT-PCR. (E) IL-1B mRNA expression levels in FMs from PBS and LPS treated wildtype or AXL^−/−MERTK^−/− mice were measured by qRT-PCR. *p<0.05 relative to the control groups unless otherwise indicated. Data are expressed as mean±SEM.