http://orcid.org/0000-0002-5737-4498
Figures
Abstract
Aim
To study the transcriptome profiles in the blood of recurrent implantation failure (RIF), recurrent miscarriage (RM) and fertile women during the window of implantation, and further analysis the correlation of transcriptome profiles between blood and endometrium.
Methods
This is an observational prospective study. In total 9 subjects were recruited, 3 RIF, 3 RM, and 3 controls. Paired samples (endometrium and peripheral blood) from the same subjects were precisely timed on the 7th days after luteal hormone surge (LH+7). RNA sequencing was applied to investigate the transcriptome profiles.
Citation: Huang J, Jin N, Qin H, Shi X, Liu Y, Cheung W, et al. (2017) Transcriptomic profiles in peripheral blood between women with unexplained recurrent implantation failure and recurrent miscarriage and the correlation with endometrium: A pilot study. PLoS ONE 12(12): e0189159. https://doi.org/10.1371/journal.pone.0189159
Editor: Kwang-Hyun Baek, CHA University, REPUBLIC OF KOREA
Received: August 14, 2017; Accepted: November 20, 2017; Published: December 7, 2017
Copyright: © 2017 Huang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The raw sequencing data was uploaded to NCBI with BioProject ID: PRJNA379542.
Funding: Direct Grant for Research (Ref No. 2014.1.042) from the Chinese University of Hong Kong. HQ is supported by a General Research Fund (GRF14102014) from the Hong Kong Government to TFC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The endometrium becomes receptive to the embryo only in the mid-luteal phase, but not in other stages of the menstrual cycle, and it has been considered as the initial critical stage of implantation. The specific time when the endometrium becomes receptive to the embryo is often referred to as the window of implantation (WOI). Implantation failure is considered to be an important cause of infertility, whereas defective implantation may lead to miscarriage, either sporadic or recurrent [1, 2]. In clinical practice, endometrial receptivity may be investigated by morphological, immune-histochemical or genomic study of a biopsy specimen [3] but the biopsy procedure itself may disturb the implantation process and cannot be carried out, for ethical reasons, in cycles when conception may take place. Non-invasive approaches such as ultrasonographic measurement of endometrial thickness and blood flow are of limited use [4] and provide very little information on the very rapid biological changes occurring in endometrium at this specific time. To achieve a breakthrough in the understanding of the complex processes involved in human implantation requires the development of a non-invasive method which can be repeated serially but which can accurately reflect the physiological changes in the endometrium.
Peripheral blood, a heated target for liquid biopsy, contains proteins, DNAs and RNAs from various sources of cells, which could provide a special window to measure specific information for organs required monitoring [5, 6]. Transcriptomic study using micro-array or RNA sequencing enables the simultaneous study of most of the genes involved in the implantation process. Several studies have used this approach to examine the endometrium in the peri-implantation period of women with recurrent implantation failure (RIF) [7–9] or recurrent miscarriage (RM) [10, 11]. Our previous study has found different pathway regulations between RIF and RM [12]. Furthermore, whole blood transcriptome studies could reveal other physical changes of human body, like blood pressure [13], pre-term labor [14], or immune response [15]. However, none of the earlier studies applied this technique to compare and correlate the transcriptome profiles of endometrium and peripheral blood and examine the potential usefulness of transcriptome study of peripheral blood to characterize different groups of women with reproductive failure.
In this study, we wish to investigate whether there is a correlation between the transcriptome in peripheral blood and endometrium during the window of implantation, and to know whether transcriptome in peripheral blood seven days after the LH surge (LH+7) could be used to characterize women with RIF and RM.
Materials and methods
Subject
Three groups of subjects were recruited from the Prince of Wales Hospital, Chinese University of Hong Kong. The inclusion criteria of all subjects recruited include: age no more than 40 years, with regular cycles (25–35 day), had not used steroid hormone in the preceding 2 months. The exclusion criteria were, briefly: chromosomal anomaly, positive for anticardiolipin antibody or lupus anticoagulant, abnormal thyroid function test, uncorrected uterine anomalies, intra-uterine device in situ, or serious systematic disease. Women with unexplained RIF was defined as failure to achieve a clinical pregnancy after transfer of at least four good-quality embryos in more than three cycles with her age under 40 years, and no obvious cause had been identified [1]. Unexplained recurrent miscarriage was defined as three or more consecutive miscarriages before 24 weeks of gestation [2]. Fertile control subjects referred to women who had one or more live birth following spontaneous conception, stopped breastfeeding for more than 6 months, and without any history of spontaneous miscarriage. In total, 9 women were recruited, 3 women with unexplained RIF, 3 women with unexplained RM, and 3 fertile controls.
Endometrial and peripheral blood samples
In the cycle of study, all subjects started daily urine LH test from day 9 of the cycle onwards until the LH surge had been identified. An endometrial biopsy was obtained on day LH+7 as an outpatient procedure with the use of a Pipelle® sampler. The samples were immediately snap-frozen and stored in liquid nitrogen for later processing. Simultaneously, 3ml of peripheral blood was collected in one dipotassium ethylenediaminetetraacetic acid (EDTA) Vacuette® tubes. Blood was mixed with TRIzol LS (Invitrogen) in a ratio of 1:3 immediately after collection, and after homogenization, the samples were stored in -80°C for later processing. The whole process blood collection, homogenization, and storage would finish within 30 minutes.
RNA extraction and expression calculation
Total RNA was extracted from endometrium by TRIzol according to manufacturer’s protocols (Invitrogen). Blood total RNA was extracted by RNeasy Mini Kit (QIAGEN). RNA quality control was confirmed by Bioanalyzer 2100 Eukaryote Total RNA Pico (Agilent Tech, Inc). There were 9 pairs of RNA samples (3 with RIF, 3 with RM, 3 controls) with high total RNA quality and integrity for RNA-seq. All total RNA samples were rRNA depleted by Ribozero (Illumina) and the paired-ends strand-specific libraries were prepared by TrueSeq Stranded Total RNA Library Prep Kit (Illumina). All samples were sequenced by Illumina HiSeq2000. After sequencing, low quality reads whose sequencing quality below 20 were trimmed. All reads were mapped to human genome hg38 by Tophat2 [16] with default parameters. The Reads Per Kilobase Per Million Reads (RPKM) of gene expression was calculated based on the GENCODE v23 annotation [17]. All expressions were normalized by quantile normalization method using median [18].
Unsupervised hierarchical clustering and principle component analysis (PCA)
The unsupervised hierarchical clustering and PCA were done as previously described [12]. Briefly, the raw normalized expression of genes were scaled and used for unsupervised hierarchical clustering and PCA by R packages gplots [19] and prcomp. After PCA was done, the vector of each principle component was calculated and support vector machine (SVM) was performed by Python library sklearn 0.17.0 [20]. Genes whose absolute values of the contribution scores were larger or equal to 0.01 were considered to have significant contribution.
Gene ontology and pathway analysis
Genes with significant contribution scores were retrieved for gene ontology (GO) and pathway analysis. Pathway enrichment was analyzed by DAVID 6.7 (the Database for Annotation, Visualization and Integrated Discovery) [18].The pathways whose correlated p-value (q-value) smaller than 0.05 were considered significantly enriched [21].
Blood and endometrium correlation analysis
Coefficients of all the genes among three groups were calculated by Pearson’s correlation. And the association of gene expressions in blood and endometrium in paired samples were also tested by Spearman’s correlation. The difference of coefficients among three groups was compared by t test.
Results
The demographics of the recruited subjects are summarized in S1 Table. There was no significant difference in age, BMI, cycle length and endometrium thickness at the time of biopsy amongst RIF, RM and fertile groups.
Differentially expressed genes in peripheral blood
The reads mapping of all 9 pairs of samples for RNA-Seq were satisfactory. For the blood samples, after trimming the adapters and low quality bases (Q<20 in 4bp sliding window) using Trimmomatic (v0.32), RNA-Seq reads were mapped to the human genome from the Genome Reference Consortium (GRCh38) using Tophat2 and Cufflinks. The average percentage of mapped reads was 87.38%, including 19.44% in intronic region, 37.96% in protein coding region, 0.0064% in rRNA region and 42.59% in other regions (e.g. intergenic, antisense strand etc.). Of the same data size, median expression of genes in endometrium were about 3 times higher than those in blood of each paired samples in Table 1. The raw sequencing data was uploaded to NCBI with reference BioProject ID: PRJNA379542.
All genes were included in the unsupervised clustering, and there were 19,079 genes in endometrium and 18,446 genes in blood. Among the three studied groups, RM and RIF showed distinct different gene expression patterns in unsupervised clustering in endometrium; whereas in peripheral blood, though the 3 RM cases grouped together, the difference from the other two groups was not distinct in Fig 1. To further examine the differences between RIF and RM, principle component analysis (PCA) was performed. In endometrium, RM and RIF samples showed distinct spatial distribution in the three-dimensional space constructed by the first two components; whereas in blood, RM and RIF could not be linearly separated, and only RM and control could barely be separated in components 3 and 4 in Fig 2. Pathway analysis by DAVID 6.7 identified no significant pathways among the gene expression profiles in peripheral blood of the three groups in S2 Table.
RIF: recurrent implantation failure; RM: recurrent miscarriage; C: fertile control.
Correlated gene expression of peripheral blood and endometrium
Of all the genes (18446) which had sequencing reads, there were 1960 genes correlation coefficients more than 0.5, among which 126 genes involved in immune response. The correlation coefficients of the 9 paired samples were all over 0.64 and the R2 for each paired samples were about 0.3 in Fig 3, and it seems that the correlation between blood and endometrium were higher in RIF than those in RM, with p < 0.05 in Fig 4. As the value of correlation coefficients over 0.5 was considered as a moderate positive relationship and 0.7 as a strong positive relationship [22], the correlation of gene expression profiles between peripheral blood and endometrium were moderate.
RIF: recurrent implantation failure; RM: recurrent miscarriage; C: fertile control.
Results shown were mean ± SE. *t test p = 0.027; SE: standard error; RIF: recurrent implantation failure; RM: recurrent miscarriage; C: fertile control.
Discussion
In this study, we have found that the transcriptome profiles in peripheral blood of the 3 groups studied (RM, RIF and fertile controls) may not be as distinctively different from one another as in those of endometrium in Figs 1 and 2. We observed that controls C3 and C42 appeared to have rather different gene profiles. One possible explanation and a limitation of our study is the biological repeat of samples (3 in each group) is relatively small in the study human whole blood, known to contain various cell subtypes and heterogeneous molecules, which may mask the critical signature difference [23]. Therefore, no significant differentially expressed pathways could be drawn from blood samples from the 3 groups.
However, we have found correlated gene expression between blood and endometrium. Over 10.6% (1960/18446) genes in blood showed similar expression trend as in endometrium among three groups could potentially represent the changes in endometrium, in which about 6.4% (126/1960) genes involved in immune response. Tissue transcriptome correlations with peripheral blood are gradually explored in searching of potential biomarkers to investigate the functional changes of the targeted tissue. For example, circulating RNAs provide a potential approach to investigate placental transcriptome, and some RNAs presenting in maternal blood are proved to be produced by placenta [24]. Through investigating the differential transcriptome profiles in peripheral blood, several critical RNAs have been identified as biomarkers in pregnancy complications, such as preeclampsia [25–27], fetal growth restriction [28], and even real-time monitoring fetal hypoxic status [29]. It suggested that circulating RNA could be used dynamically to interrogate changes in fetal and placental health. However, so far there is no report of correlations between peripheral blood and endometrium transcriptome.
Hence, novelty of this study is for the first time trying to investigate the transcriptome in paired samples among three groups of women. Previous studies had either measured individual markers or RNA markers in blood or endometrium [30]. We have chosen in our analysis to use support vector machine to draw a two-class comparison; we acknowledge that other clustering methods could be used instead. We consider that the use of unsupervised analysis strategy helped to reduce the selection bias. Therefore, our study suggests that it is possible that more genes could be identified in blood as non-invasive markers to represent the correlated gene expression levels in the endometrium.
The relative gene expression level of whole blood to endometrium is another valuable contribution of this study for exploration of circulating markers of endometrium receptivity. As suggested by the ratio of medium expression levels of genes in each paired samples is 2–4 in Table 1, for better exploring the difference of blood transcriptome difference, about 3 times sequencing coverage should be necessary to reach the same gene expression level as that of endometrium. This could be of special importance when performing investigations for those low-level expressed correlated genes in both endometrium and blood.
Previous studies had successfully used transcriptomic profiles of the endometrium to characterize different groups of women with reproductive failure (RIF [7–9, 31, 32] or RM [10–12]); our observation suggested that it was not possible to do so with transcriptomic profile of the peripheral blood. It raises questions about the possible clinical usefulness of the peripheral blood to monitor the molecular changes in the endometrium in the window of implantation. Correlations of the transcriptome profiles are quite good between endometrium and peripheral blood given that no gene selection is being made. The difference in the genes expressed among the three groups of women is not consistently observed, but rather a relevant subset related to the condition of interest. To further explore the gene expression features of different groups, more samples analyzed with another selective strategy may be need.
Whilst transcriptomic profiles of the peripheral blood do seem to have a moderate degree of correlation with profiles of the endometrium around the time of implantation, this indirect & non-invasive approach does not seem able to replace the direct but somewhat invasive approach (endometrial biopsy) to characterize different groups of women with reproductive failure. Consequently, it is still not yet possible to use serial (for example daily) transcriptomic profiling of the peripheral blood to examine the dynamic and rapidly changing molecular events in the endometrium at this time. Until an accurate, sensitive and reliable and yet non-invasive test is available (which does not disturb the embryo and the implantation process), it is not possible to overcome the ethical dilemma posed on the study of human implantation.
Supporting information
S1 Table. Demographic characteristics in women with recurrent reproductive failure or fertile controls.
C: fertile subjects, RIF: recurrent implantation failures, RM: recurrent miscarriages, NS: not significant, SD: standard deviation.
https://doi.org/10.1371/journal.pone.0189159.s001
(DOCX)
S2 Table. KEGG pathways and GO terms involved in comparison of blood transcriptome between recurrent implantation failures and recurrent miscarriages.
https://doi.org/10.1371/journal.pone.0189159.s002
(XLSX)
References
- 1. Coughlan C, Ledger W, Wang Q, Liu F, Demirol A, Gurgan T, et al. Recurrent implantation failure: definition and management. Reproductive biomedicine online. 2014;28(1):14–38. pmid:24269084.
- 2. Saravelos SH, Li TC. Unexplained recurrent miscarriage: how can we explain it? Human reproduction. 2012;27(7):1882–6. pmid:22552692.
- 3. Haouzi D, Mahmoud K, Fourar M, Bendhaou K, Dechaud H, De Vos J, et al. Identification of new biomarkers of human endometrial receptivity in the natural cycle. Human reproduction. 2009;24(1):198–205. pmid:18835874.
- 4. Riad ON, Hak AA. Assessment of endometrial receptivity using Doppler ultrasonography in infertile women undergoing intrauterine insemination. Gynecological endocrinology: the official journal of the International Society of Gynecological Endocrinology. 2014;30(1):70–3. pmid:24256372.
- 5. Jalbrzikowski M, Lazaro MT, Gao F, Huang A, Chow C, Geschwind DH, et al. Transcriptome Profiling of Peripheral Blood in 22q11.2 Deletion Syndrome Reveals Functional Pathways Related to Psychosis and Autism Spectrum Disorder. PloS one. 2015;10(7):e0132542. pmid:26201030; PubMed Central PMCID: PMCPMC4511766.
- 6. Petrov PD, Bonet ML, Reynes B, Oliver P, Palou A, Ribot J. Whole Blood RNA as a Source of Transcript-Based Nutrition- and Metabolic Health-Related Biomarkers. PloS one. 2016;11(5):e0155361. pmid:27163124; PubMed Central PMCID: PMCPMC4862680.
- 7. Koot YE, van Hooff SR, Boomsma CM, van Leenen D, Groot Koerkamp MJ, Goddijn M, et al. An endometrial gene expression signature accurately predicts recurrent implantation failure after IVF. Sci Rep. 2016;6:19411. pmid:26797113; PubMed Central PMCID: PMCPMC4726345.
- 8. Ruiz-Alonso M, Blesa D, Diaz-Gimeno P, Gomez E, Fernandez-Sanchez M, Carranza F, et al. The endometrial receptivity array for diagnosis and personalized embryo transfer as a treatment for patients with repeated implantation failure. Fertility and sterility. 2013;100(3):818–24. pmid:23756099.
- 9. Ledee N, Munaut C, Aubert J, Serazin V, Rahmati M, Chaouat G, et al. Specific and extensive endometrial deregulation is present before conception in IVF/ICSI repeated implantation failures (IF) or recurrent miscarriages. The Journal of pathology. 2011;225(4):554–64. pmid:22025212.
- 10. Kosova G, Stephenson MD, Lynch VJ, Ober C. Evolutionary forward genomics reveals novel insights into the genes and pathways dysregulated in recurrent early pregnancy loss. Human reproduction. 2015;30(3):519–29. pmid:25586782; PubMed Central PMCID: PMC4325674.
- 11. Othman R, Omar MH, Shan LP, Shafiee MN, Jamal R, Mokhtar NM. Microarray profiling of secretory-phase endometrium from patients with recurrent miscarriage. Reproductive biology. 2012;12(2):183–99. pmid:22850470
- 12. Huang J, Qin H, Yang Y, Chen X, Zhang J, Laird S, et al. A comparison of transcriptomic profiles in endometrium during window of implantation between women with unexplained recurrent implantation failure and recurrent miscarriage. Reproduction. 2017;153(6):749–58. pmid:28283674.
- 13. Zeller T, Schurmann C, Schramm K, Muller C, Kwon S, Wild PS, et al. Transcriptome-Wide Analysis Identifies Novel Associations With Blood Pressure. Hypertension. 2017;70(4):743–50. pmid:28784648.
- 14. Chim SSC, Wong KKW, Chung CYL, Lam SKW, Kwok JSL, Lai CY, et al. Systematic Selection of Reference Genes for the Normalization of Circulating RNA Transcripts in Pregnant Women Based on RNA-Seq Data. Int J Mol Sci. 2017;18(8). pmid:28777335.
- 15. Song F, Qian Y, Peng X, Li X, Xing P, Ye D, et al. The frontline of immune response in peripheral blood. PloS one. 2017;12(8):e0182294. pmid:28771541; PubMed Central PMCID: PMCPMC5542476.
- 16. Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 2013;14(4):R36. pmid:23618408; PubMed Central PMCID: PMCPMC4053844.
- 17. Harrow J, Frankish A, Gonzalez JM, Tapanari E, Diekhans M, Kokocinski F, et al. GENCODE: the reference human genome annotation for The ENCODE Project. Genome Res. 2012;22(9):1760–74. pmid:22955987; PubMed Central PMCID: PMCPMC3431492.
- 18. Risso D, Ngai J, Speed TP, Dudoit S. Normalization of RNA-seq data using factor analysis of control genes or samples. Nat Biotechnol. 2014;32(9):896–902. pmid:25150836; PubMed Central PMCID: PMCPMC4404308.
- 19.
Warnes GR, Bolker B, Bonebakker L, Gentleman R, Huber W, Liaw A, et al. gplots: Various R programming tools for plotting data. R package version. 2009;2(4).
- 20.
Pedregosa. Scikit-learn: Machine Learning in Python. JMLR 12, pp. 2011:2825–30.
- 21. Alles MC, Gardiner-Garden M, Nott DJ, Wang Y, Foekens JA, Sutherland RL, et al. Meta-analysis and gene set enrichment relative to er status reveal elevated activity of MYC and E2F in the "basal" breast cancer subgroup. PloS one. 2009;4(3):e4710. pmid:19270750; PubMed Central PMCID: PMCPMC2650420.
- 22.
Rumsey DJ, Unger D. U Can: Statistics For Dummies: John Wiley & Sons; 2015.
- 23. Krjutskov K, Koel M, Roost AM, Katayama S, Einarsdottir E, Jouhilahti EM, et al. Globin mRNA reduction for whole-blood transcriptome sequencing. Sci Rep. 2016;6:31584. pmid:27515369; PubMed Central PMCID: PMCPMC4981843.
- 24. Tsui NB, Jiang P, Wong YF, Leung TY, Chan KC, Chiu RW, et al. Maternal plasma RNA sequencing for genome-wide transcriptomic profiling and identification of pregnancy-associated transcripts. Clinical chemistry. 2014;60(7):954–62. pmid:24740362.
- 25. Farina A, Zucchini C, De Sanctis P, Morano D, Sekizawa A, Purwosunu Y, et al. Gene expression in chorionic villous samples at 11 weeks of gestation in women who develop pre‐eclampsia later in pregnancy: implications for screening. Prenatal diagnosis. 2011;31(2):181–5. pmid:21268038
- 26. Founds SA, Conley YP, Lyons-Weiler JF, Jeyabalan A, Hogge WA, Conrad KP. Altered global gene expression in first trimester placentas of women destined to develop preeclampsia. Placenta. 2009;30(1):15–24. pmid:19027158; PubMed Central PMCID: PMCPMC2667803.
- 27. Paiva P, Whitehead C, Saglam B, Palmer K, Tong S. Measurement of mRNA transcripts of very high placental expression in maternal blood as biomarkers of preeclampsia. The Journal of Clinical Endocrinology & Metabolism. 2011;96(11):E1807–E15.
- 28. Whitehead CL, McNamara H, Walker SP, Alexiadis M, Fuller PJ, Vickers DK, et al. Identifying late-onset fetal growth restriction by measuring circulating placental RNA in the maternal blood at 28 weeks' gestation. Am J Obstet Gynecol. 2016;214(4):521 e1–8. pmid:26880734.
- 29. Whitehead C, Teh WT, Walker SP, Leung C, Mendis S, Larmour L, et al. Quantifying circulating hypoxia-induced RNA transcripts in maternal blood to determine in utero fetal hypoxic status. BMC medicine. 2013;11:256. pmid:24314237; PubMed Central PMCID: PMCPMC4029381.
- 30. Achache H, Revel A. Endometrial receptivity markers, the journey to successful embryo implantation. Human reproduction update. 2006;12(6):731–46. pmid:16982667.
- 31. Koler M, Achache H, Tsafrir A, Smith Y, Revel A, Reich R. Disrupted gene pattern in patients with repeated in vitro fertilization (IVF) failure. Human reproduction. 2009;24(10):2541–8. pmid:19542175.
- 32. Altmae S, Martinez-Conejero JA, Salumets A, Simon C, Horcajadas JA, Stavreus-Evers A. Endometrial gene expression analysis at the time of embryo implantation in women with unexplained infertility. Molecular human reproduction. 2010;16(3):178–87. pmid:19933690.