Experiments were carried out according to the relevant guidelines for animal use as approved by the Institutional Bioethics Committee of Escuela Nacional de Ciencias Biológicas-IPN (Ciudad de México, Mexico). These guidelines are based on the Guide for the Care and Use of Laboratory Animals (National Institutes of Health, Bethesda, MD, USA).

Healthy, virgin, and sexually mature ICR female mice (8 weeks) were obtained from the Animal Center Facilities of Escuela Superior de Medicina-IPN (Mexico City, Mexico). Mice were housed in a room at constant temperature (22°C) with a fixed 12-h light–dark cycle and had access to food and water ad libitum.

Stage of the estrous cycle was determined by cytologic evaluation of vaginal smears. From mice with two continuous estrous cycles, uteri were collected at diestrus, proestrus, estrus, and metestrus (n = 4 mice per stage).

Healthy female mice at the estrus stage were mated with healthy C57BL/6 male mice overnight. The following morning, spermatozoids in females were verified with a vaginal smear and, if spermatozoids were present, the mice were considered to be 0.5 days post coitum (dpc).

Four uteri from pregnant mice were collected 4.5, 5.5, 7.5, and 10.5 dpc (“peri-implantation” period); 12.5, 16.5, 18.5 dpc (“fetal development”); 19.5 dpc (“labor”); 2 days post-labor; 5 days post-labor. Uteri were extracted to measure the levels of mRNA and protein for IL-36 subfamily members. Samples were also embedded in paraffin for immunofluorescent analyses.

A suspension of infective L. monocytogenes was prepared in sterile phosphate-buffered saline (PBS) at 10⁹ CFU/mL. Female BALB/c mice were inoculated (i.v.) in the tail with 100 μL of the L. monocytogenes suspension (10⁸ CFU/mouse) during estrus.

After infection, mice showed typical piloerection after day 3. Then, mice were again cycled and mated overnight with a healthy C57BL/6 male to obtain semiallogenic offspring. A procedure identical to that described above was carried out for non-infected mice to determine pregnancy. Implantation sites from pregnant non-infected and infected BALB/c mice were obtained at diestrus and at estrus, as well as at 4.5, 5.5, 7.5, and 10.5 dpc. To confirm infection with L. monocytogenes in harvested uteri, CFU/mL was determined, and a polymerase chain reaction (PCR) coupled to the DNA sequence of the 16S ribosome was carried out (data not shown).

Uteri were homogenized individually, and RNA was isolated using TRIzol^® reagent (Invitrogen, Carlsbad, CA, USA). Total RNA concentration was evaluated using a NanoDrop spectrophotometer (Thermo Scientific, Wilmington, DE, USA). Total RNA (2 μg) was reacted with DNAse I (Affymetrix, Santa Clara, CA, USA). cDNA was synthesized using MLV reverse transcriptase (Invitrogen). PCR reactions (20 μL) were carried out using 1 μL of the cDNA reaction, 10 μL of AmpliTaq Gold^® Fast PCR Master Mix (Life Technologies, Gaithersburg, MD, USA), and 0.2 μM-each of IL-36R, IL-36α, IL-36β, IL-36γ, and IL-36Ra primers or the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) primer (housekeeping gene) (Table 1). Optimized PCR conditions were 35 cycles at 96°C for 5 s, 60°C for 5 s, and 68°C for 5 s. Amplified DNA (250 bp) was analyzed on a 2% agarose gel and stained using RedGel (Biotium Inc., CA, USA). Gel images were acquired in a ChemiDoc-It™ transilluminator (UVP, Upland, CA, USA), and the integrated pixel density (PD) of each band was calculated using AlphaImager^® (ProteinSimple, San Jose, CA, USA). The PD of each gene band was normalized by dividing the PD of the sample by that of the corresponding housekeeping gene (GAPDH) band. The change in the expression of each gene was calculated by dividing the expression of the normalized gene at a specified point in pregnancy by that of the expression of the normalized gene at diestrus.

A total protein fraction (coming from 3 implantation sites or 1/2 uterus from non-pregnant mice) was extracted from uteri using 400 μL of RIPA buffer [Tris-HCl (pH 7.6), NaCl (150 mM), EDTA (2 mM), glycerol (10%), Triton-X100 (1%), sodium desoxicolate (0.5%), sodium lauryl sulfate (0.2%)], 1 mM of phenylmethylsulfonyl fluoride (Sigma–Aldrich, Saint Louis, MO, USA), and 1× complete protease inhibitor cocktail (Roche Diagnostics, Mannheim, Germany). Proteins were quantified using the Lowry method (Bio-Rad Laboratories, Hercules, CA, USA); 25 μg of protein extract were diluted 1:5 in Laemmli sample buffer (Bio-Rad Laboratories) containing 2-mercaptoethanol (Bio-Rad) and denatured by boiling. Samples were loaded on SDS–PAGE 15% gels and transferred to cellulose sheets (Amersham Biosciences, Amersham, UK). Non-specific binding was blocked with 4% gelatin in PBS 1× containing 0.5% Tween-20 for 1 h at 37°C. Specific primary antibodies to polyclonal goat anti-IL-36α, anti-IL-36β, anti-IL-36γ, anti-IL-36R, and anti-IL-36Ra (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were diluted 1:750 in blocking buffer. The secondary antibody was horseradish peroxidase–IgG mouse anti-goat (1:5000 dilution). Chemiluminescence was developed using an electrochemiluminescence western blotting substrate (Pierce Biotechnology, Waltham, MA, USA) and detected on a ChemiDoc™ Touch imaging system (Bio-Rad).

Uteri were fixed in 4% formalin and embedded in paraffin. Tissue slices (thickness, 6 μm) were obtained using a microtome (RM2125RT; Leica, Wetzlar, Germany) and placed on slides pre-coated with poly-l-lysine. Antigen retrieval was achieved by heating in a pressure cooker at 95°C for 15 min followed by immersion in antigen retrieval buffer (Life Technologies, Carlsbad, CA, USA). Samples were blocked with 0.1% albumin solution and incubated with primary antibody (anti-IL-36α, anti-IL-36β, anti-IL-36γ, anti-IL-36R, or anti-IL-36Ra) diluted in PBS for 50 min at 37°C. After two washes, the secondary antibody (rabbit fluorescein isothiocyanate–IgG anti-goat IgG) was added to samples at 1:1500 dilution and incubated for 1 h at 37°C. Preparations were observed under a confocal microscope (LSM 700; Zeiss, Oberkochen, Germany) for image capture. Captures were analyzed using LSM v5 Image Examiner software (Zeiss).

Results are the mean ± SEM. Statistical analyses were carried out using GraphPad Prism (GraphPad, San Diego, CA, USA). Differences between samples were evaluated using Newman–Keuls multiple comparison tests, after one-way ANOVA. p ≤ 0.05 was considered significant.

The mRNA expression of IL-36α, β, and γ (Figures 1A–C, respectively) and the corresponding production of proteins (Figures 2A–C, respectively) were similar to those of other pro-inflammatory cytokines in the estrous cycle. That is, expression occurred at diestrus, increased slightly at proestrus, reached maximal expression at estrus, decreased at metestrus, was again lowest at diestrus, and then the cycle started again (Figures 1A–C). It was found that mRNA (Figure 1D) and protein (Figure 2E) levels of IL-36R, which behaved as agonists during the estrous cycle, showed maximal expression at estrus, declined at metestrus, and then was reduced to the baseline level at diestrus. IL-36Ra showed a different expression behavior than the other subfamily members: mRNA (Figure 1E) and protein (Figure 2F) were present mainly at proestrus rather than at estrus; and at estrus and metestrus, expression was even lower than that at proestrus.

Embryo implantation is a crucial event for mammals, and inflammatory cytokines are involved. Here, we investigated whether IL-36 subfamily members are involved in implantation.

At the start of implantation, IL-36α expression remained at similar levels at metestrus and decreased at the end of implantation (Figures 1A and 2B). IL-36β expression increased at the beginning of implantation (4.5 dpc) and decreased progressively until 10.5 dpc, and even at 7.5 and 10.5 dpc, the protein could not be detected (Figure 2C). IL-36γ expression was barely detected at the beginning of implantation but had increased by 11-fold at the end of implantation (10.5 dpc) (Figure 1C); IL-36γ protein showed identical behavior (Figure 2D). Expression of IL-36R mRNA remained similar during implantation but increased slightly at 5.5 dpc (Figure 1D). However, expression of IL-36R protein increased at 5.5 and 7.5 dpc during implantation (Figure 2E). Expression of IL-36Ra increased noticeably after metestrus, reaching optimal expression at 5.5 dpc for mRNA and protein (Figures 1E and 2F).

At 12.5 dpc, protein expression of IL-36α, β, and γ was not detectable and increased progressively until labor. Expression of IL-36β mRNA increased considerably, but expression of the protein of IL-36α and γ increased even more (Figures 1E and 2F).

For IL-36R, mRNA expression remained at the same level as during implantation, but the protein increased considerably during fetal development until labor (Figures 1E and 2F). Expression of IL-36Ra increased after 10.5 dpc, reaching optimal mRNA expression at 18.5 dpc, and decreased at labor (Figure 1E).

After a considerable increase in expression at labor, expression of pro-inflammatory cytokines declined after childbirth to hormonal-cycle levels. We also analyzed the behavior of IL-36 subfamily members after labor. Expression of all IL-36 agonists decreased significantly after labor. IL-36R was detected constantly during late pregnancy, even in labor, where production of IL-36 agonists was increased (Figure 1D), but that of their proteins decreased at 5 dpc (Figure 2E). Expression of IL-36Ra was also downregulated after labor (Figures 1E and 2F).

Immunofluorescence detection of IL-36 subfamily members was undertaken. Diestrus showed a reduced presence of agonists distributed randomly in the luminal epithelium, glandular epithelium, and perimetrium. However, IL-36R and IL-36Ra showed a more concentrated presence in glandular and luminal epithelia (Figure 3). Western blotting revealed the presence of the three agonists to be increased during proestrus and estrus and, in these stages, IL-36α showed a characteristic distribution in the glandular epithelium, perimetrium, myometrium, and stratum vasculare. IL-36β and γ were distributed more homogenously, but spots in the glandular epithelium were not seen. IL-36R and IL-36Ra showed a similar pattern of distribution in diestrus, and they were detected in glandular and luminal epithelia, with increased detection in the perimetrium (Figure 3). At the end of the estrous cycle, during metestrus, detection of agonists and antagonists of IL-36 was diminished drastically compared with that at estrus, whereas IL-36R was maintained in the uterus (Figure 3).

Expression of the agonists, receptors, and antagonists of IL-36 was localized by immunofluorescence in uterine tissues at early implantation (4.5 dpc), during the final phase of implantation (10.5 dpc), at culmination of pregnancy (labor), and at 2 days after labor (Figure 4). Similar to previous results, expression of IL-36α and IL-36β was higher than IL-36γ expression in the implantation period (4.5 dpc), and this expression was observed in the decidua. IL-36R showed low expression in the decidua and myometrium, as did IL-36Ra (Figure 4).

IL-36 subfamily members at 4.5 and 10.5 dpc showed a characteristic distribution as small spots close to blood vessels. At labor, expression of the five members of the IL-36 subfamily was upregulated (in accordance with western blotting results) and distributed alongside the interface between the luminal epithelium and placenta (Figure 4). Two days after labor, cells from luminal and glandular epithelia showed a large decrease in the production of IL-36 family members, except for moderate expression of IL-36Ra (Figure 4).

We measured the expression of IL-36 subfamily members in a murine model of L. monocytogenes infection because infection with this pathogen induces an inflammatory response and can eventually cause fetal loss (11). The infection was established before pregnancy in order to ascertain whether the inflammatory response that occurs normally during implantation in healthy pregnancies would be modified due to L. monocytogenes.

Infection by L. monocytogenes reduced the number of implanted embryos (data not shown) and induced severe overexpression of IL-36 subfamily members compared with non-infected mice (Figures 5 and 6). Notably, in the uteri of infected mice, the five members of the IL-36 subfamily showed increases in mRNA expression between 25- and 150-fold higher as compared with expression in the implantation sites of non-infected mice at 4.5, 5.5, and 7.5 dpc (Figure 5). Protein analyses revealed a clear increase in expression of IL-36 subfamily members (Figure 6), suggesting that they have a crucial role in the response against L. monocytogenes, and that the increased expression could be the cause of miscarriage or infertility (Figure 6).