Patients and samples

This case–control study was performed between March 2013 and February 2017 at the Recurrent Pregnancy Loss Outpatient Clinic of the Gemelli Hospital of Rome, Italy. The study population included 70 women with history of idiopathic RPL with three or more spontaneous consecutive pregnancy losses (≤ 12 weeks of gestation) clinically documented by ultrasonography [1] at least 6 months before the study and 30 healthy women with two or more previous uncomplicated term pregnancies (control group).

The inclusion criteria for both groups were as follows: caucasian, age ≤ 39 years, healthy, regular ovulatory cycles (28–32 days), normal serum levels of follicle-stimulating hormone (FSH < 10 mIU/ml), luteinizing hormone (LH < 10 mIU/ml) and anti-mullerian hormone (AMH > 2 ng/ml) on day 3 of the menstrual cycle, absence of abnormal ovarian and endometrial ultrasonographic features, no use of any contraceptive drugs or intrauterine device in the last 6 months. Exclusion criteria were: smoking, alcohol consumption, presence of abnormalities at the screening for RPL (anatomical abnormalities, endometrial cavity anomalies assessed by hysteroscopy, luteal phase deficiency, hyperprolactinemia, hyperinsulinemia or insulin resistance, endocrine disorders, vaginal infections, karyotype anomalies, thrombophilic disorders, autoimmune diseases), obesity, endometriosis, refusal to undergo intestinal permeability test, any pathological conditions potentially interfering with intestinal permeability, like celiac disease or gluten sensitivity, history of viral hepatitis, diarrhea, diverticulosis, irritable bowel syndrome, inflammatory bowel diseases, or bariatric or other abdominal surgery, immunoglobulin A or immunoglobulins deficiency, impaired renal function, use of non-steroidal anti-inflammatory drugs, antibiotics, probiotics, or anti-secretory drugs within the 3 months preceding enrollment.

Samples collection

Peripheral blood samples (5 ml) from all RPL (n = 70) and healthy control women (n = 30) enrolled were collected by vein puncture. Samples were spun at 3000 g at 4 °C for 20 min, then the supernatant were aliquoted and stored at − 80 °C until use.

Thirty-five of seventy RPL women and twenty of thirty controls accepted a diagnostic mini-hysteroscopy during the putative window of implantation (days 19th to 24th), with endometrial biopsies. The biopsies timing was chosen according to the last menstrual period, monitoring the follicle size by transvaginal ultrasound and then confirmed by histologic assessment [15]. All women were advised to avoid sexual intercourses from menses to planned hysteroscopic procedure. Serum progesterone and β-hCG levels were determined just before hysteroscopy. Endometrial biopsy were performed using a 3-mm Novak curette, avoiding any contact between the curette and the vaginal walls, and then immediately washed in normal saline solution and stored at − 80 °C.

Biopsies were also analysed for Chlamydia trachomatis, Mycoplasma, Ureaplasma urealyticum, Neisseria gonorrheae or yeast colonization. Patients with endometrial samples positive for pathogens colonization as well as those with evidence of chronic endometritis at hysteroscopy were not included in the study.

Anxiety and depression state evaluation

Acute or stable anxiety of enrolled women was assessed by administration of the State-Trait Anxiety Inventory (STAI) Y test [16, 17], divided in two parts: the first evaluating the anxiety state by inquiring current emotional state asking twenty questions with responses ranging from ‘‘not at all’’ (score 1) to ‘‘extremely’’ (score 4) on a 1–4 Likert scale. Total scores ranged from 20 to 80, with higher scores indicating higher degrees of state anxiety. The second part of STAI Y test assesses trait anxiety by asking to subjects to describe how they usually feel. Twenty questions were asked to enrolled patients with responses ranging from ‘‘almost never’’ (score 1) to ‘‘almost always’’ (score 4) on a Likert 1–4 scale. Higher scores reflected greater symptomatology and a suggested cutoff of 39 was established for clinically significant anxiety (score: 40–50 mild; 50–60 moderate; > 60 severe [17]. STAI -Y questionnaires were completed in approximately 20 min.

Depressive symptomatology was investigated with the use of the Zung Self-Rating Depression Scale (Z-SDS) [18], a short self-administered survey. The Z-SDS scale consists of 20 items rating the four common characteristics of depression: (1) the pervasive effect, (2) the physiological equivalents, (3) other disturbances and (4) psychomotor activities. Each question is scored on a scale of 1–4 [corresponding to (1) a little of the time, (2) some of the time, (3) good part of the time and (4) most of the time]. The scores range from 25 to 100 to be set on a depression scale with higher scores reflecting greater symptomatology: 25–49 Normal Range; 50–59 Mildly Depressed; 60–69 Moderately Depressed; 70 and above Severely Depressed.

Assessment of circulating lipopolysaccharide levels

Sera from RPL and control women were thaw and analyzed for concentration of the Gram negative bacterial membrane component, LPS by LAL chromogenic endpoint assay (HyCult Biotechnology, Uden, Netherlands) according to the manufacturer’s instructions. Results were analysed by spectrophotometer (Titertek Multiscan plus Mk II plate reader ICN Flow Laboratories, Irvine, CA) measuring the absorbance at wavelengths of 450 nm.

Western blot analysis of endometrial Nalp-3 expression

To investigate inflammasome Nalp-3 endometrial expression by Western blot, total cellular lysates obtained from endometrial biopsies of ten RPL and ten control women were separated by 10% SDS-PAGE electrophoresis under reducing conditions. After gel electrophoresis and transfer of proteins to a nitrocellulose membrane, nitrocellulose sheets were blocked at room temperature for 1 h in 5% non-fat dry milk and incubated overnight at 4 °C with a specific primary mouse monoclonal anti-Nalp-3 antibody (ThermoFisher Scientific, Rockford, IL, USA). The membranes were then washed with PBST and incubated with specific horseradish peroxidase-conjugated IgG diluted 1:2000 in PBST with 5% of non-fat dried milk. Bound secondary antibody was detected by chemiluminescence. Bands were analyzed by a Gel Doc 200 Image Analysis System and quantified with the Quantity One Quantitation Software (both from BioRad). As positive load control cytosolic β-actin band (42-kDa) was detected by a mouse monoclonal anti-human β-actin antibody (Sigma-Aldrich, St Louis, MO, USA).

Endometrial assessment of caspase-1, IL-1β and IL-18

Caspase-1, IL-18 and IL-1β levels were measured in endometrial lysates from 35 RLP and twenty control women by a commercial enzyme-linked immunoassay (ELISA), according to manufacturer’s instructions (USCN Life Science Inc. and Cloud-Clone Corp. Houston, TX, USA). Briefly, 100 µl of samples or standard were added to each well coated with human monoclonal anti-caspase-1 or IL-18 or IL-1β antibodies. After 2 h of incubation at 37 °C, wells were washed and incubated with a specific enzyme-linked polyclonal antibody, conjugated to horseradish peroxidase. Then, tetramethyl-benzidine substrate solution was added to each well, and the color developed in proportion to the amount of the proteins bound in the initial step. The plate was read on a Titertek Multiscan plus Mk II plate reader (ICN Flow Laboratories, Irvine, CA) measuring the absorbance at wavelengths of 450 nm. Results were normalized for protein content of each sample.

Statistical analysis

Results are expressed as mean ± standard deviation (SD). Data were analyzed using one-way analysis of variance (ANOVA) followed by a post–hoc test (Bonferroni test) or Chi square according to type of variables. Statistical significance was determined at P < 0.05. Anxiety and depression state data are expressed as percentages and P value was determined at P < 0.05 (Student’s t test for unpaired data).

Patients

RPL was associated with abnormal intestinal permeability

Intestinal permeability of enrolled patients was assessed by ⁵¹Cr-EDTA absorption test. We observed a higher prevalence of abnormal intestinal permeability in RPL patients compared to the control group (Table 1). At gastrointestinal evaluation, none of women with increased intestinal permeability referred clear symptoms of gastrointestinal diseases. None of the recruited women showed any abnormal levels of immunoglobulin (Ig) A or E, or any positivity for anti-endomysial, anti-gliadin, anti-transglutaminase antibodies, HLA DQ2-DQ8 haplotype, lactose/lactulose/sorbitol breath tests, fecal calprotectin assay, pANCA, cANCA, ASCA or ASMA antibodies.

RPL women showed higher circulating levels of lipopolysaccharides

Assessment of serum LPS levels revealed significantly increased circulating levels of LPS in women with RPL compared to the controls (P < 0.05; Table 1). The cut off for defining abnormal LPS circulating levels was established to be > 2SD of the LPS average levels in a healthy population (0.4 EU/ml).

Women with RPL showed a higher prevalence of depression and anxiety

Results of the psychological tests are reported in Table 1. Overall, RPL patients showed more symptoms of depression and anxiety than control healthy subjects.

Eighty-seven percent of the RPL women, compared with 35% of the control group, had STAI-Y1 (state anxiety) scores > 39 (95% mild and 5% moderate for the RPL group; 90% mild and 10% moderate for the control group).

Sixty percent of the RPL women, compared with 31% of the control group, had STAI-Y2 (trait anxiety) scores > 39 (79% mild and 21% moderate for the RPL group; 89% mild and 11% moderate for the control group).

Forty-eight percent of the RPL women, compared with 24% of the healthy control group, had a score > 50 on the Z-SDS test (40% mildly depressed and 8% moderately depressed; P < 0.05).

Endometrial Nalp-3 inflammasome was over expressed and activated in RPL

Western Blot analysis showed a significant increase of Nalp-3 proteins expression in endometrial tissues of women with idiopathic RPL compared to the controls (P < 0.0001) (Table 1 and Fig. 1a). The cut off for defining Nalp-3 endometrial expression to be abnormal was > 2SD of endometrial Nalp-3 average expression of the control population (3000 O.D.).

We also examined the levels of caspase-1 in endometrial lysates obtained from hysteroscopic biopsies by ELISA. As shown in Fig. 1b, increased levels of caspase-1 were shown in the RPL women compared to the controls (P < 0.05). Finally, through an ELISA assay, we evaluated the levels of IL-1β and IL-18 in endometrial lysates obtained from the controls and RPL women. Significantly increased levels of pro-inflammatory cytokine IL-1β (Fig. 1c), but not of IL-18 (Fig. 1d), were found by ELISA in endometrial tissues of the RPL women compared to the controls (P < 0.001).

Abnormal intestinal permeability and circulating LPS levels were associated with increased expression of endometrial Nalp-3 in women with RPL

Abnormal intestinal permeability and circulating LPS levels are not associated to anxiety and depression state in women with RPL

Abnormal intestinal permeability was not associated to anxiety state or trait, or depression (Table 2). RPL women with higher circulating levels of LPS did not show significant higher prevalence of anxiety and depression state compared to once with normal levels of serum LPS (Table 3). A significant association was found only between abnormal serum levels of LPS and anxiety trait (Table 3). As a whole, abnormal intestinal permeability and serum LPS levels do not seem to be related to psychological disorder in a strong cause-effect relationship.