Altered autophagic flux enhances inflammatory responses during inflammation-induced preterm labor

Abstract

Cellular organelles and proteins are degraded and recycled through autophagy, a process during which vesicles known as autophagosomes fuse with lysosomes. Altered autophagy occurs in various diseases, but its role in preterm labor (PTL) is unknown. We investigated the role of autophagic flux in two mouse models of PTL compared to controls: 1) inflammation-induced PTL (IPTL), induced by toll-like receptor agonists; and 2) non-inflammation (hormonally)-induced PTL (NIPTL). We demonstrate that the autophagy related genes Atg4c and Atg7 (involved in the lipidation of microtubule-associated protein 1 light chain 3 (LC3) B-I to the autophagosome-associated form, LC3B-II) decrease significantly in uterus and placenta during IPTL but not NIPTL. Autophagic flux is altered in IPTL, as shown by the accumulation of LC3B paralogues and diminishment of lysosome associated membrane protein (LAMP)-1, LAMP-2 and the a2 isoform of V-ATPase (a2V, an enzyme involved in lysosome acidification). These alterations in autophagy are associated with increased activation of NF-κB and proinflammatory cytokines/chemokines in both uterus and placenta. Similar changes are seen in macrophages exposed to TLR ligands and are enhanced with blockade of a2V. These novel findings represent the first evidence of an association between altered autophagic flux and hyper-inflammation and labor in IPTL.

Figure 1. Autophagy-related proteins in various forms of labor.

Panels A and B show mRNA expression of Atg4c and Atg7 in uterus and placenta recovered from preterm labor groups and controls. N = 6–11 each group. Panels C and D show tissue distribution of Atg4c and Atg7 protein, respectively. Corresponding immunostaining index scores are shown in E and F [(ISIS) = Stained area score (SAS) X Immunostaining intensity score (IIS)]. Six tissue sections per animal were analyzed. Original magnification: 400X. PBS, LPS and PGN+poly(I:C): intrauterine injections on day 14.5; MIF-C: subcutaneous DMSO control, MIF: subcutaneous mifepristone in DMSO on day 14.5. Error bars = ±SEM. ^aP ≤ 0.05 Significant difference vs. respective control.

Figure 2. LC3B protein is increased and a2V mRNA is decreased in inflammation-induced preterm labor.

Panels A and C show western blots of LC3B-I, LC3B-II and GAPDH in uterus and placenta recovered from preterm labor groups and controls with corresponding densitometric analysis (LC3B-II/GAPDH) in panels B and D. LC3B-I densitometric analysis shows a similar pattern (data not shown). N = 4–5 each group. Panels E and F show LC3B and panels G and H show a2V mRNA expression in uterus and placenta recovered from preterm labor groups and controls. N = 6–11 each group. PBS, LPS and PGN+poly(I:C): intrauterine injections on day 14.5; MIF-C: subcutaneous DMSO control, MIF: subcutaneous mifepristone in DMSO on day 14.5. Error bars = ±SEM. *P ≤ 0.05, **P ≤ 0.01 Significant difference vs. respective control. Full-length blots are presented in Supplementary Figure 12.

Figure 3. Expression of a2V is decreased and LC3B is increased in inflammation induced preterm labor.

Uterus shown in (A) and placenta in (B). LC3B stained in green; a2V stained in red. Nuclei stained with DAPI (blue) in merged images. N = 4–5 each group. Six sections per animal were analyzed. Original magnification: 200X. PBS, LPS and PGN+poly(I:C): intrauterine injections on day 14.5; MIF-C: subcutaneous DMSO control on day 14.5; MIF: subcutaneous mifepristone in DMSO on day 14.5.

Figure 4. Expression of LAMP-1 and LAMP-2 is decreased in both inflammation-induced and non-inflammation-induced preterm labor.

(A) LAMP-1 and (B) LAMP-2 (green) by immunofluorescence staining in uterus and placenta recovered from preterm labor and control groups. Six sections per animal were analyzed. Original magnification: 400X. Panels C and D show sample western blot and corresponding densitometric analysis of LAMP-1, LAMP-2 and GAPDH in uterus recovered from preterm labor and control groups. N = 4–5 each group. Error bars = ±SEM. *P ≤ 0.05 Significant difference vs. respective control. Full-length blots are presented in Supplementary Figure 12.

Figure 5. Decrease of a2V is associated with reduction of LAMP-2 in the RAW 264.7 mouse macrophage cell line.

RAW 264.7 cells were treated for 30 min with PBS, PGN+poly(I:C) or LPS, followed by up to 2 h incubation with IgG control or anti-a2V antibody. (A) mRNA expression of a2V; (B) protein for LAMP-2 (green) and a2V (red); (C) LAMP-2 in the absence or presence of an a2V neutralizing antibody. Original magnification: 400X. Each experiment was done three times with triplicates. Error bars = ±SEM. **P ≤ 0.01 Significant difference vs. respective control.

Figure 6. NF-κB p65 activation is increased in inflammation-induced preterm labor.

(A) Distribution of LC3B, NF-κB p65 and TNF (brown) in serial sections of uterus recovered from PBS, PGN+poly(I:C) and LPS-treated animals. Six sections per animal were analyzed. Original magnification: 400X. (B) Total and phosphorylated NF-κB p65 by ELISA (control group set to 1). N = 4–5 each group. (C) mRNA expression of IL-1β and TNF in uterus by RT-PCR. N = 6–9 each group. PBS, LPS and PGN+poly(I:C): intrauterine injections on day 14.5; MIF-C: subcutaneous DMSO control, MIF: subcutaneous mifepristone in DMSO on day 14.5. Error bars = ±SEM. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001 Significant difference vs. respective control.

Figure 7. Blockade of a2V enhances baseline and LPS- and PGN+poly(I:C)-induced activation of NF-κB p65 in the RAW 264.7 mouse macrophage cell line.

RAW 264.7 cells were treated for 30 min with PBS, PGN+poly(I:C) or LPS, followed by up to 3 h incubation with IgG control or anti-a2V antibody. Each experiment was done three times with triplicates. Shown is the translocation of NF-κB p65 (green) from the cytoplasmic to the nuclear compartment (blue) in 2 h (A). Original magnification: 400X. (B) Fold change in the concentration of phosphorylated NF-κB p65 by ELISA after 1 h, 2 h and 3 h of antibody incubation (IgG group set to 1). Error bars = ±SEM. **P ≤ 0.01 Significant difference between PGN+poly(I:C) or LPS treated with PBS/IgG vs. treated with anti-a2V.

Figure 8. Model of autophagy in preterm labor.

Red arrows represent observed changes; Blue X represents hypothesized mechanism based on our observations.