Maintained thymic output of conventional and regulatory T cells during human pregnancy

Abstract

During pregnancy, immunological tolerance toward the semiallogeneic fetus needs to be established while at the same time an effective immune defense must be maintained.1Mor G. Aldo P. Alvero A.B. The unique immunological and microbial aspects of pregnancy.Nat Rev Immunol. 2017; 17: 469-482Crossref PubMed Scopus (470) Google Scholar The pregnancy-associated thymic involution reported in rodents2Clarke A.G. Kendall M.D. The thymus in pregnancy: the interplay of neural, endocrine and immune influences.Immunol Today. 1994; 15: 545-551Abstract Full Text PDF PubMed Scopus (130) Google Scholar has been suggested to support maternal immune regulation by reducing the output of potentially harmful TH cells. However, the functional importance of this thymic involution remains unclear and it is not known whether it even occurs in humans.3Swami S. Tong I. Bilodeau C.C. Bourjeily G. Thymic involution in pregnancy: a universal finding?.Obstet Med. 2012; 5: 130-132Crossref PubMed Google Scholar In fact, the role of thymus during human pregnancy and in pregnancy-associated tolerance remains largely unknown,4Wagner M.I. Mai C. Schmitt E. Mahnke K. Meuer S. Eckstein V. et al.The role of recent thymic emigrant-regulatory T-cell (RTE-Treg) differentiation during pregnancy.Immunol Cell Biol. 2015; 93: 858-867Crossref PubMed Scopus (22) Google Scholar albeit a role for thymic-derived regulatory T (Treg) cells in pregnancy complications has been suggested,4Wagner M.I. Mai C. Schmitt E. Mahnke K. Meuer S. Eckstein V. et al.The role of recent thymic emigrant-regulatory T-cell (RTE-Treg) differentiation during pregnancy.Immunol Cell Biol. 2015; 93: 858-867Crossref PubMed Scopus (22) Google Scholar supporting a role for thymus in immune regulation during human pregnancy. The aim of this study was to assess the role of thymus in TH-cell regulation during human pregnancy by analyzing the output of different TH-cell populations. To determine whether pregnancy influences thymic function we used RT-PCR to analyze T-cell receptor excision circle (TREC) content in flow cytometry–sorted TH-cell subpopulations (Fig 1, A and B) in second-trimester pregnant and nonpregnant women (see detailed methods and Table E1 in this article's Online Repository at www.jacionline.org). TREC content is a well-established measure of cells with recent thymic origin,5Kong F.K. Chen C.L. Six A. Hockett R.D. Cooper M.D. T cell receptor gene deletion circles identify recent thymic emigrants in the peripheral T cell pool.Proc Natl Acad Sci U S A. 1999; 96: 1536-1540Crossref PubMed Scopus (148) Google Scholar based on the presence of episomal DNA fragments, generated during the T-cell receptor rearrangement in the thymus, and because TRECs are not duplicated during mitosis they are diluted with each cell division. Consequently, cells with a more recent thymic origin will have a higher TREC content compared with TH cells that have undergone peripheral postthymic proliferation.5Kong F.K. Chen C.L. Six A. Hockett R.D. Cooper M.D. T cell receptor gene deletion circles identify recent thymic emigrants in the peripheral T cell pool.Proc Natl Acad Sci U S A. 1999; 96: 1536-1540Crossref PubMed Scopus (148) Google Scholar As expected, TREC content was significantly higher in cells with a naive (CD45RA+) phenotype, compared with memory (CD45RA−) cells (P < .0001; Fig 1, C), hence validating our PCR assay to measure TREC. We found that second-trimester pregnant, compared with nonpregnant, women had significantly higher TREC levels (P = .043: Fig 1, D) in the naive Treg-cell population (defined as CD25++CD45RA+Foxp3+), indicating an increased thymic output of Treg cells in pregnancy. The TREC content did not differ in the naive T conventional (Tconv; CD25−CD45RA+) cells (Fig 1, E), in line with a maintained output of conventional TH cells in pregnancy. To corroborate and extend the finding of maintained or increased thymic output during pregnancy, we used CD31 expression, assessed by flow cytometry, as an additional marker of recent thymic emigrants (RTEs).6Tanaskovic S. Fernandez S. Price P. Lee S. French M.A. CD31 (PECAM-1) is a marker of recent thymic emigrants among CD4+ T-cells, but not CD8+ T-cells or gammadelta T-cells, in HIV patients responding to ART.Immunol Cell Biol. 2010; 88: 321-327Crossref PubMed Scopus (47) Google Scholar Here, we investigated thymic output in both the second and third trimesters of pregnancy, the period when pregnancy-induced immune modulation is most pronounced (Fig 2, A; see Table E1 and Fig E1 in this article's Online Repository at www.jacionline.org). CD31+, as opposed to CD31−, naive T cells are significantly enriched for TRECs,7Junge S. Kloeckener-Gruissem B. Zufferey R. Keisker A. Salgo B. Fauchere J.C. et al.Correlation between recent thymic emigrants and CD31+ (PECAM-1) CD4+ T cells in normal individuals during aging and in lymphopenic children.Eur J Immunol. 2007; 37: 3270-3280Crossref PubMed Scopus (127) Google Scholar and in agreement, we show that naive T cells, with higher TREC content, also have higher CD31 expression compared with memory T cells (P < .0001; see Fig E2 in this article's Online Repository at www.jacionline.org). CD31 and TRECs are known to correlate, although with a slight difference in dynamics.8Kilpatrick R.D. Rickabaugh T. Hultin L.E. Hultin P. Hausner M.A. Detels R. et al.Homeostasis of the naive CD4+ T cell compartment during aging.J Immunol. 2008; 180: 1499-1507Crossref PubMed Scopus (170) Google Scholar Overall, we found that pregnant and nonpregnant women had similar proportions and absolute numbers of RTE Tconv (CD31+ Foxp3−CD45RA+) cells (Fig 2, B and C), supporting a maintained output of most T cells during pregnancy. The significantly higher RTE Tconv-cell count that was noted in third-trimester compared with second-trimester pregnant women (P = .0036, Fig 2, C) most likely reflects the corresponding increase noted in total CD4+ T-cell counts in the third trimester (compared with the second trimester; P = .0067; data not shown). In contrast to the TREC-based difference in Treg cells, we found no differences between nonpregnant and pregnant women, neither in the second trimester nor in the third trimester, concerning the proportion and absolute numbers of RTE Treg cells (CD31+ FoxplowCD45RA+) (Fig 2, D and E). This was also true for the cellular level of expression of CD31 (median fluorescent intensity; data not shown). The reason why CD31 findings did not corroborate the TREC-based increase in the Treg-cell population might be the difference in dynamics of these markers. Indeed, CD31 expression is maintained on naive T cells for several rounds of homeostatic proliferation while TREC levels decrease,8Kilpatrick R.D. Rickabaugh T. Hultin L.E. Hultin P. Hausner M.A. Detels R. et al.Homeostasis of the naive CD4+ T cell compartment during aging.J Immunol. 2008; 180: 1499-1507Crossref PubMed Scopus (170) Google Scholar supporting the notion that differences in kinetics may well explain the difference in our findings regarding Treg cells when based on CD31 (maintained output) or TREC (increased output). Proliferation of T cells influences TREC levels, although proliferation in naive T cells is assumed to be too low to significantly affect TREC levels. However, indications of a more activated maternal immune system during pregnancy9Steinborn A. Rebmann V. Scharf A. Sohn C. Grosse-Wilde H. Soluble HLA-DR levels in the maternal circulation of normal and pathologic pregnancy.Am J Obstet Gynecol. 2003; 188: 473-479Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar prompted us to make a more detailed analysis of the dynamics of the T-cell pool (for gating strategies see Fig E1). Overall, we found no indications of increased proliferation in naive and memory TH-cell subsets during pregnancy, neither when comparing pregnant and nonpregnant women, nor when comparing the second and third trimesters of pregnancy (see Fig E3 in this article's Online Repository at www.jacionline.org). Instead, we found significantly lower Ki67 expression in RTE Tconv cells in pregnant women compared with the nonpregnant state, which could affect measured TREC levels, albeit Ki67 expression was very low in all groups (see Fig E3, B). Another factor that potentially could affect measured RTE frequencies, and hence also TREC content, is peripheral consumption, leading to a different distribution of subpopulations. However, we found no major shifts in the T-cell distributions because pregnant and nonpregnant women had similar proportions of mature naive (CD31−CD45RA+) and memory Tconv and Treg cells, as well as of nonsuppressive Foxp3+ TH cells (see Fig E4 in this article's Online Repository at www.jacionline.org). Taken together, it is unlikely that peripheral events such as proliferation and consumption of cells have had a large impact on our results. In conclusion, we found that thymic output of both Tconv and Treg cells is maintained during second and third trimesters of human pregnancy. The output of Treg cells may even be increased, which would fit with the demand for fetal tolerance during pregnancy and could also contribute to the pregnancy-associated improvement of autoimmune diseases, like multiple sclerosis, that are associated with defects in thymic function. Our findings challenge the general conception, based on studies in rodents, of an inactive thymus during pregnancy, but rather support a maintained and important function of thymus in human pregnancy. We thank the midwives at the maternity care unit at Vrinnevi Hospital, Norrköping, Sweden, for their invaluable help with recruiting and taking samples from the pregnant women included in the study and the nurses at Clinical Immunology, Linköping University Hospital, for taking blood samples from all the healthy controls. We also thank J. Raffetseder for help with the Foxp3 RT-PCR. A total of 56 healthy pregnant women (n = 30 second-trimester pregnant, n = 26 third-trimester pregnant) with no signs of pregnancy complications at inclusion, visiting the maternity care unit in Norrköping, Sweden, were included in the study. A review of medical records following delivery showed that 2 pregnant women (recruited in the second trimester) had delivered preterm (before gestational week 37) and were therefore not considered as normal pregnancies and were excluded. In addition, 1 woman (second trimester) who received tinzaparin (Innohep) 3 days before blood sampling was also excluded because of the potential immunological effects of low-molecular weight heparin.E1Luley L. Schumacher A. Mulla M.J. Franke D. Lottge M. Fill Malfertheiner S. et al.Low molecular weight heparin modulates maternal immune response in pregnant women and mice with thrombophilia.Am J Reprod Immunol. 2015; 73: 417-427Crossref PubMed Scopus (21) Google Scholar Of the remaining 27 second-trimester pregnant women (see Table E1), it was decided that the following 3 women should remain included: 1 woman who subsequently developed intrahepatic cholestasis but delivered at term (gestational week 42), hence considered healthy at the time of blood sampling; 1 woman on antihistamine (cetirizine) treatment; 1 woman on selective serotonin reuptake inhibitor (SSRI) and inhalation budesonide (Pulmicort) treatment. Of the 26 included third-trimester pregnant women (see Table E1), it was decided that the following women should remain included: 2 women on SSRI; 2 women on Trombyl, and 3 women who used pain medication (Citodon). Thirty nonpregnant women were recruited among students and personnel at Linköping University and Linköping University Hospital (see Table E1). All nonpregnant women were healthy and had normal blood cell counts. One nonpregnant control reported use of inhalation budesonide (Pulmicort) and terbutaline sulfate (Bricanyl Turboinhaler) treatment and 2 controls used SSRIs. There were no significant differences between the groups in terms of age, smoking, or parity. The pregnant women had been pregnant significantly more times compared with the nonpregnant controls (P = .012 for second trimester vs healthy control and P = .004 for third trimester vs healthy control). Because of limitations in cell numbers available from each woman, different numbers of subjects are included in the different analysis. All women gave written informed consent before inclusion and blood sampling. The study was approved by the regional ethical review board in Linköping (M39-08).Table E1Information about the participating womenSubject characteristicsPregnantNonpregnantSecond trimesterThird trimesterNumber of subjects (n)272630Age at inclusion (y)28 (22-34)28 (19-35)27.5 (22-38)Use of hormonal contraceptives (yes/no)NANA13/17Menstrual cycle (luteal/follicular)∗Based on a 28-d menstrual cycle.NANA11/15†Four individuals on hormonal contraceptive reported not having a period.Smoking (yes/no)1/260/260/30Current pregnancyGestational age at inclusion (wk)26 (24-30)36 (35-37)NAParity week (wk)41 (40-43)40.5 (38-43)NASex of baby (male/female)15/1213/13NABirth weight (g)3550 (2735-4465)3408 (2460-4710)NABirth method (PN/VE/CS)25/1/120/1/5NAPregnancy history Previous pregnancies (n)1 (0-4)‡P < .05, compared with healthy controls using Kruskal-Wallis with Mann-Whitney test.1 (0-5)§P < .01 compared with healthy controls using Kruskal-Wallis with Mann-Whitney test.0 (0-2) Previous births (n)1 (0-3)1 (0-4)0 (0-2)Data are shown as median and ranges (in parentheses) or as categorical data.CS, Cesarean section; NA, not applicable; PN, normal delivery; VE, vacuum extraction.∗ Based on a 28-d menstrual cycle.† Four individuals on hormonal contraceptive reported not having a period.‡ P < .05, compared with healthy controls using Kruskal-Wallis with Mann-Whitney test.§ P < .01 compared with healthy controls using Kruskal-Wallis with Mann-Whitney test. Open table in a new tab Data are shown as median and ranges (in parentheses) or as categorical data. CS, Cesarean section; NA, not applicable; PN, normal delivery; VE, vacuum extraction. Blood was collected in an EDTA-tube (BD Vacutainer; BD Biosciences, Franklin Lakes, NJ) for flow cytometry and in sodium-heparin tubes (Greiner Bio-One, Kremsmünster, Austria) for isolation of PBMCs for cell sorting and subsequent analysis of TREC content with PCR. PBMCs were isolated by density centrifugation over a Lymphoprep gradient (Axis-Shield, Dundee, UK) followed by washing in HBSS (Gibco BRL, Invitrogen, Life Technologies, Paisley, UK). The PBMCs were filtered through a preseparation filter (Miltenyi Biotec, Bergisch Gladbach, Germany), and CD4+ T cells were immunomagnetically isolated by negative selection using the CD4+ T-cell isolation kit (Miltenyi Biotec) using MS columns and a MiniMACS separator (Miltenyi Biotec) according to the instructions provided by the manufacturer. The CD4+ T cells were labeled with mouse antihuman CD4-PeCy7 (clone SK3), CD45RA-V450 (clone HI100), and CD25-APC (clone M-A251; all from BD Biosciences) for 30 minutes at 4°C in the dark and washed in PBS (pH 7.4; Medicago, Uppsala, Sweden) supplemented with 0.1% heat-inactivated FBS (HyClone, GE Healthcare, Little Chalfont, UK) by centrifugation at 500g, 4°C for 5 minutes. CD4+ T cells were sorted into different T-cell subpopulations on the basis of their expression of CD25 and CD45RA (Fig 1, A). Naive (CD4+CD25−CD45RA+) and memory (CD4+CD25−CD45RA−) Tconv cells, and naive (CD4+CD25++CD45RA+) and memory (CD4+CD25+++ CD45RA−) Treg cells were gated using a similar gating strategy as described by Miyara et al.E2Miyara M. Yoshioka Y. Kitoh A. Shima T. Wing K. Niwa A. et al.Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor.Immunity. 2009; 30: 899-911Abstract Full Text Full Text PDF PubMed Scopus (1646) Google Scholar The CD25+++ gate for memory Treg cells was based on the lowered expression of CD25 on naive Treg cells (CD25++), which in turn was set to include cells with a lower CD25 expression while still being positive. The CD25− gate for the Tconv cells was set to include a major proportion of the CD25− cells while avoiding potential contamination of CD25dim-expressing cells. Naive (CD45RA+) and memory cells (CD45RA−) were set on the basis of the contour of the positive and negative populations. We set stricter gates for both CD25 and CD45RA to ensure high purity of the sorted populations (Fig 1, A). The cells were sorted using the FACSAria III Cell sorter (BD Biosciences) with a 70-μM nozzle and collected in PBS with 0.5% FBS. After sorting, the cells were centrifuged and lysed in Buffer RLT Plus (Qiagen, Hilden, Germany) and stored at −70°C until RNA and DNA extraction. RNA and DNA were extracted from the sorted cells using the Allprep DNA/RNA Micro Kit (Qiagen, Hilden, Germany) according to the instructions provided by the manufacturer. RNA concentrations were determined spectrophotometically (ND-1000, NanoDrop Technologies Inc, Wilmington, Del). The DNA concentration was determined with a fluorescence-based nucleic acid method using a Quantus Fluorometer (Promega, Madison, Wis) with QuantiFluor Dye (Promega) according to the instructions provided by the manufacturer. The levels of the δRec-ψJα signal joint TREC (sjTREC) were measured by TaqMan real-time quantitative PCR using previously validated primers and probes directed toward the signal joint region of the TRECs.E3Sairafi D. Mattsson J. Uhlin M. Uzunel M. Thymic function after allogeneic stem cell transplantation is dependent on graft source and predictive of long term survival.Clin Immunol. 2012; 142: 343-350Crossref PubMed Scopus (27) Google Scholar, E4Nobile M. Correa R. Borghans J.A. D'Agostino C. Schneider P. De Boer R.J. et al.De novo T-cell generation in patients at different ages and stages of HIV-1 disease.Blood. 2004; 104: 470-477Crossref PubMed Scopus (40) Google Scholar, E5Douek D.C. Vescio R.A. Betts M.R. Brenchley J.M. Hill B.J. Zhang L. et al.Assessment of thymic output in adults after haematopoietic stem-cell transplantation and prediction of T-cell reconstitution.Lancet. 2000; 355: 1875-1881Abstract Full Text Full Text PDF PubMed Scopus (471) Google Scholar, E6Chan K. Puck J.M. Development of population-based newborn screening for severe combined immunodeficiency.J Allergy Clin Immunol. 2005; 115: 391-398Abstract Full Text Full Text PDF PubMed Scopus (297) Google Scholar The PCR reaction was performed by mixing 4 μL of DNA (TREC) or 1 μL of DNA (glyceraldehyde-3-phosphate dehydrogenase [GAPDH]) with 2× Taqman Universal Master Mix (Applied Biosystems, Foster City, Calif), together with primers and probes for sjTREC (forward primer: 5′-CACATCCCTTTCAACCATGCT-3′; reverse primer: 5′-GCCAGCTGCAGGGTTTAGG-3′; both Invitrogen; probe: 6-FAM-5′-ACACCTCTGGTTTTTGTAAAGGTGCCCACT-3′-TAMRA; Applied Biosystems) or GAPDH (forward primer: 5′-GGACTGAGGCTCCCACCTTT-3′; reverse primer: 5′-GCATGGACTGTGGTCTGCAA-3′; both Invitrogen; probe: VIC-5′-CATCCAAGACTGGCTCCTCCCTGC-3′-TAMRA; Applied Biosystems). The final concentrations of the primers and probes in the reaction mixture were 500 and 300 nM for the TREC forward and reverse primer, respectively, 50 and 300 nM for GAPDH, and 200 nM of both probes. The mRNA expression of Foxp3, the lineage-specific transcription factor for Treg cells, was also analyzed in a portion of the sorted T-cell populations (Fig 1, B). RNA was converted to cDNA using the high-capacity cDNA reverse transcription kit (Applied Biosystems) according to the instructions provided by the manufacturer. Thermal cycling was carried out using the Arktik Thermal Cycler (Thermo Scientific, Waltham, Mass). The reaction was carried out by mixing 4 μL (Foxp3) or 1 μL (18S) of cDNA with 2× Taqman Universal Master Mix together with primers and probes (all from Applied Biosystems) for Foxp3 (forward primer: 5′-GTGGCCCGGATGTGAGAA-3′; reverse: 5′-GCTGCTCCAGAGACGTACCATCT-3′; probe: FAM-5′-CCTCAAGCACTGCCAGGC-GGAC-3′-TAMRA) and 18S (forward primer: 5′-CGGCTACCACATCCAAGGAA-3′; reverse: 5′-GCTGGAATTACCGCGGCT-3′; probe: FAM- 5′-TGCTGGCACCAGACTTG-CCCTC-3′-TAMRA) as previously used.E7Mjosberg J. Svensson J. Johansson E. Hellstrom L. Casas R. Jenmalm M.C. et al.Systemic reduction of functionally suppressive CD4dimCD25highFoxp3+ Tregs in human second trimester pregnancy is induced by progesterone and 17beta-estradiol.J Immunol. 2009; 183: 759-769Crossref PubMed Scopus (132) Google Scholar, E8Svensson-Arvelund J. Mehta R.B. Lindau R. Mirrasekhian E. Rodriguez-Martinez H. Berg G. et al.The human fetal placenta promotes tolerance against the semiallogeneic fetus by inducing regulatory T cells and homeostatic M2 macrophages.J Immunol. 2015; 194: 1534-1544Crossref PubMed Scopus (187) Google Scholar The final concentrations of the primers and probes in the reaction mixture were 300 and 900 nM for Foxp3 forward and reverse primer, respectively, 200 and 100 nM for 18S, and 200 nM (Foxp3) and 50 nM (18S) of each probe. All PCR amplifications were performed using the 7500 Fast Real-Time PCR system (Applied Biosystems), with the thermal cycling conditions set to an initial step at 95°C for 20 seconds followed by 40 cycles at 95°C for 3 seconds and a final step at 60°C for 30 seconds. All samples were run in duplicates. Baseline was set using the automatic baseline feature of the 7500 software version 2.3 (Applied Biosystems). Threshold values were adjusted manually and only those samples that had a Ct value of less than 35 were considered detectable. Samples with a value above Ct 35, that is, undetectable, were assigned a value of Ct 35. TREC content in each sample was calculated using the delta Ct-method where the Ct values for sjTREC were subtracted from the Ct values for the housekeeping gene GAPDH (formula 2Ct GAPDH-Ct sjTREC), as previously described.E3Sairafi D. Mattsson J. Uhlin M. Uzunel M. Thymic function after allogeneic stem cell transplantation is dependent on graft source and predictive of long term survival.Clin Immunol. 2012; 142: 343-350Crossref PubMed Scopus (27) Google Scholar For Foxp3, the RNA content was normalized against the housekeeping gene 18S rRNA and quantification was performed using the standard curve method. Whole blood was stained with CD127-PerCP-Cy5.5 (clone HIL-7R-M21), CD45RA-V450 (clone HI100), CD31-PeCy7 (clone WM59), and CD4-APC-Cy7 (clone SK3; all from BD Biosciences) for 30 minutes at 4°C in darkness. Subsequently, the blood samples were incubated for 15 minutes at room temperature in the dark with ammonium chloride (NH4Cl, diluted 1:10 with dH2O; Merck, Kenilworth, NJ), for lysis of the erythrocytes. The cells were washed in PBS supplemented with 0.1% FBS by centrifugation at 500g, 4°C for 5 minutes. The cells were fixed and permeabilized using the Foxp3/Transcription Factor Buffer Staining Set (eBioscience, Thermo Fisher Scientific) according to the instructions provided by the manufacturer. Briefly, the cells were fixed for 30 minutes at 4°C in the dark and washed twice in permeabilization buffer and labeled with antibodies toward Ki67-FITC (clone B56; BD Biosciences) and Foxp3-PE (clone PCH101; eBioscience, Thermo Fisher Scientific, Waltham, Mass) for 30 minutes, 4°C, in the dark. The cells were then washed in permeabilization buffer and resuspended in PBS with 0.1% FBS before flow cytometry analysis. To determine the absolute cell count, a BD TruCount tube was used (BD Biosciences) as described by the manufacturer. The TruCount tubes contain an exact number of beads that, when combined with a known volume of blood, can be used to determine the concentration, that is, the number of cells per liter of blood, of different cell populations. Data were acquired using FACS Canto II (BD Biosciences) and analyzed with Kaluza software version 1.2 (Beckman Coulter, Brea, Calif). Lymphocytes were gated on the basis of size (forward scatter) and granularity (side scatter), and further characterized as CD4+. Treg and Tconv-cell populations were defined on the basis of their expression of Foxp3 and CD45RA; naive (CD4+FoxplowCD45RA+) and memory (CD4+Foxp3highCD45RA−) Treg cells, naive (CD4+Foxp3−CD45RA+) and memory (CD4+ Foxp3−CD45RA−) Tconv cells, and nonsuppressive Foxp3+ TH cells (CD4+ FoxpdimCD45RA−). The Foxp3high gate was set on the basis of lowered expression of Foxp3 on naive Treg cells (Foxp3low), which in turn was gated to include cells that had a lower Foxp3 expression while still being positive. The lack of Foxp3 expression (Foxp3−) was set on the basis of contour of the naive Tconv cells. For gating strategy, see Fig E1. The definition of naive and memory Treg cells was based on the gating strategy described by Miyara et al.E2Miyara M. Yoshioka Y. Kitoh A. Shima T. Wing K. Niwa A. et al.Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor.Immunity. 2009; 30: 899-911Abstract Full Text Full Text PDF PubMed Scopus (1646) Google Scholar To ensure that our gating strategy was correct, we added CD127 in addition to Foxp3 to the definition of Treg cells,E2Miyara M. Yoshioka Y. Kitoh A. Shima T. Wing K. Niwa A. et al.Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor.Immunity. 2009; 30: 899-911Abstract Full Text Full Text PDF PubMed Scopus (1646) Google Scholar which gave the same results, that is, no differences between pregnant and nonpregnant women (data not shown). Furthermore, more than 92% of the naive Treg cells were also CD127low. The naive Treg and Tconv cells were further gated for CD31 expression on the basis of contour of the positive and negative populations. Ki67 expression in TH subpopulations was set on the basis of expression in the whole CD4+ population. All gating was performed in a blinded manner, that is, the evaluator did not know the origin of the samples.Fig E2Correlation between TREC levels and CD31 expression. TREC levels and CD31 expression in (A) naive and memory Tconv cells (n = 9) and (B) naive and memory Treg cells (n = 9). TREC levels were determined by RT-PCR, and CD31 expression was analyzed by flow cytometry.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E3Ki67 expression on CD4+ T-cell subsets. Proportion of Ki67+ (A) mature naive and memory Tconv and Treg cells and nonsuppressive Foxp3+ TH cells, (B) RTE T conv cells, and (C) RTE Treg cells, analyzed by flow cytometry. Mean and SDs are shown. Missing values are due to limited availability in cell numbers required to perform analysis. *P < .05, ***P < .001, ****P < .0001.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E4Proportion of T-cell subsets. Frequency of (A and B) mature naive Tconv and Treg cells, (C and D) memory Tconv and Treg cells, and (E) nonsuppressive Foxp3+ TH cells, analyzed by flow cytometry. Missing values are due to limited availability in cell numbers required to perform analysis. MN, Mature naive. Mean and SDs are shown.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Most flow cytometry data from the phenotyping and absolute counts followed Gaussian distribution, and differences between groups were therefore determined using 1-way ANOVA and Tukey multiple comparisons test. RT-PCR data were analyzed with Mann-Whitney U test, and correlations between CD31 and TREC were performed using Spearman rank correlation. Flow cytometry data are expressed as mean and SD and RT-PCR data as median and interquartile ranges. P values less than or equal to .05 were considered statistically significant. Data were analyzed using GraphPad Prism version 7.03 (GraphPad Software Inc, La Jolla, Calif).