Patient and public involvement
The patients and/or the public were not involved in the design, conduct, reporting, or dissemination of this study.
Study design and patients
The present study is a proof-of-concept and subsequent implementation of basic research on physiological trials that include samples from two cohorts: cohort of samples from a biobank and newly recruited pregnant women. Ten biobank AF samples (five pregnant women without CD and five pregnant women with confirmed CD) were obtained from the La Paz University Hospital (Spanish IdiPAZ, Madrid, Spain). The newly recruited cohort included 125 pregnant women (124 non-CD pregnant women and one CD-confirmed pregnant woman) from two hospitals, Virgen de Valme University Hospital (Seville, Spain) and Quirónsalud Sagrado Corazón Hospital (Seville, Spain), between November 2015 and October 2020 (Fig. 1).
This study was conducted in accordance with the principles of the Declaration of Helsinki. The study protocol was reviewed and approved by the research ethics committee (REC) of each institution involved (ID/Numbers: Celifluid and PI-2611). Written informed consent was obtained from all the participants. The participants consented to IdiPaz for the use of their data and/or samples for health-related research purposes.
The 10 AF samples collected during pregnancy screening were obtained from the Spanish IdiPAZ biobank. The maternal age ranged from 18 to 44 years. The entire biobank cohort had available primary care and specialty data. The inclusion criteria for the biobank cohort included pregnant women underwent amniocentesis in the second trimester (15–25 weeks): 1) five non-CD pregnant women on a gluten-containing diet (GCD) and 2) five CD pregnant women on a gluten-free diet (GFD) for ≥ 24 months before the time of sampling. Women with CD were diagnosed by the detection of positive anti-endomysium and tTG antibodies in the serum, confirmed by a small intestinal biopsy and cleared of digestive symptoms suggestive of CD.
The inclusion criteria for the newly recruited cohort of pregnant women were as follows: 1) second trimester non-CD pregnant women on a GCD who underwent amniocentesis (16–24 weeks), indicated for clinical reasons and not for the purpose of this study; 2) a cohort of term non-CD pregnant women who gave birth with GCD and were classified according to their mode of delivery (caesarean section or vaginal delivery) and 3) a full-term pregnant woman with GFD who was diagnosed with CD by the detection of positive anti-endomysium and tTG antibodies in the serum and confirmed by a small intestinal biopsy. She had consumed a GFD for ≥ 24 months before the time of inclusion in the study.
The exclusion criteria for all pregnant women included the presence of known severe medical conditions, and the use of prescription medications and antibiotics 2 months prior to inclusion in the study.
We recruited AF samples from two different stages of pregnancy: 1) amniocentesis from weeks 15 to 25 of pregnancy. AF collection before week 14th may lead to complications and miscarriage; 2) at term delivery (after 37 weeks). The newly enrolled pregnant women included 92 at term deliveries, of which 50 had vaginal deliveries (28 eutocic and 22 instrument-assisted) while 42 underwent caesarean sections (32 elective and 10 emergency C-sections). The amniocentesis samples included 33 prenatal diagnostic amniocentesis samples. The amniocentesis cohort included 14 pregnancies with normal foetuses and 19 pregnancies with foetal anomalies (Supplemental material).
Amniotic fluid and urine collection
AF samples were collected in a sterile fashion under careful conditions to avoid blood contamination and stored at -20°C until analysis. The collection of AF samples was conducted as follows: 1) at vaginal delivery, after cervical opening and amniotomy by lancet; 2) during elective (planned, with no ongoing labour) or acute (labour already started) caesarean delivery, at the time of caesarean section by aspirating through intact amniotic membranes using a sterile probe and 20-mL syringe; and 3) during amniocentesis in the second trimester after abnormal findings on foetal ultrasound or biochemical markers, by inserting a needle through the abdomen into the utero to remove a sterile 5-mL AF sample. Additionally, all pregnant participants were instructed to collect 50–100 mL of midstream urine samples in a sealed sterile container. This sample was taken at the time of AF collection and stored at -20°C until the time of processing.
Total protein quantitation
Frozen AF samples were analysed for total protein in duplicate batches using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, Madrid, Spain), as recommended by the manufacturer. Bovine serum albumin, ranging from 25 to 2 mg/mL, was used as a standard. A 96-microplate well was placed in a shaker for 30 s and incubated at 37°C for 30 min. The absorbance of each sample was measured at 562 nm using a plate reader (UVM340; Asys Hitech GmbH, Eugendorf, Austria).
Gluten peptides concentration and conditioning
The AF samples were centrifuged at 4,550 g for 5 min. The supernatant was mixed with three volumes of ice-cold acetone for 2 h, followed by centrifugation at 11,000 g for 10 min. The supernatant was removed, and the protein pellets were resuspended in 1% trifluoroacetic acid (TFA) and concentrated by solid-phase extraction (SPE) with preconditioned cartridges. For urine samples, 3 mL of a mixture of 50% urine in TFA was centrifuged for 10 min at 4,550 × g. The resulting supernatant was concentrated and cleaned using SPE [27].
The SampliQ C18 cartridges (Agilent Technologies, Wilmington, DE, USA) were prepared according to the manufacturer’s recommendations. The resulting supernatants from the AF and urine samples were applied to the cartridge. The target compounds were eluted with 0.5 to 1 mL of phosphate buffered saline for further use in the immunochromatographic assays.
Immunochromatographic test for GIP detection
Lateral flow immunoassays were performed on AF and urine samples to detect GIP. After SPE, 100 µL of the blind concentrate was added to a G12 lateral flow immunochromatographic (LFI) strip for 30 min. If gluten peptides are present, they react with the conjugated antibodies previously fixed on the LFI strip, thus producing a red line in the resulting window. A control Ab-antigen reaction was performed to indicate correct test performance (green line). To quantify the signal, the LFI strip was introduced into the cassette of an LFI reader (iVYCHECK®Reader, Biomedal S.L., Seville, Spain) and irradiated with light; the reflection was subsequently measured. The limit of quantification (LOQ) of the technique was established as 6.25 ng GIP/mL. The limit of detection (LDT) was 2.25 ng GIP/mL.
Statistical analysis
The sample size required for each experiment was calculated. The minimum expected difference was the LOQ (0.062) and the dropout rate was 0.2, with an alpha risk of 5%, a beta risk of 20% (80% statistical power), and a standard deviation of 0.101. The calculation was made using the GRANMO v7.12 April 2012 (Institut Municipal d'Investigació Mèdica, Barcelona, Spain).
All results are expressed as mean standard deviation (SD). Each human specimen was analysed in triplicate. Positive controls and buffer blanks were used for each assay. Statistical analyses were performed using GraphPad Prism 6 for Windows (GraphPad Software Inc., San Diego, CA, USA). An unpaired two-tailed Student’s t-test was used; statistical significance was set at P < 0.05.
We used the non-parametric Fisher’s exact test, Cochran–Armitage test, and Spearman’s correlation coefficient to assess the presence of GIP in pregnant women with and without CD as well as to estimate the level of association between the two variables. SAS version 9.4 (Cary, NC, USA) was used for all statistical calculations. All P-values were two-sided.