The Promyelocytic Leukemia Zinc Finger Transcription Factor Is Critical for Human Endometrial Stromal Cell Decidualization

Abstract

Progesterone, via the progesterone receptor (PGR), is essential for endometrial stromal cell decidualization, a cellular transformation event in which stromal fibroblasts differentiate into decidual cells. Uterine decidualization supports embryo implantation and placentation as well as subsequent events, which together ensure a successful pregnancy. Accordingly, impaired decidualization results not only in implantation failure or early fetal miscarriage, but also may lead to potential adverse outcomes in all three pregnancy trimesters. Transcriptional reprogramming on a genome-wide scale underlies progesterone dependent decidualization of the human endometrial stromal cell (hESC). However, identification of the functionally essential signals encoded by these global transcriptional changes remains incomplete. Importantly, this knowledge-gap undercuts future efforts to improve diagnosis and treatment of implantation failure based on a dysfunctional endometrium. By integrating genome-wide datasets derived from decidualization of hESCs in culture, we reveal that the promyelocytic leukemia zinc finger (PLZF) transcription factor is rapidly induced by progesterone and that this induction is indispensable for progesterone-dependent decidualization. Chromatin immunoprecipitation followed by next generation sequencing (ChIP-Seq) identified at least ten progesterone response elements within the PLZF gene, indicating that PLZF may act as a direct target of PGR signaling. The spatiotemporal expression profile for PLZF in both the human and mouse endometrium offers further support for stromal PLZF as a mediator of the progesterone decidual signal. To identify functional targets of PLZF, integration of PLZF ChIP-Seq and RNA Pol II RNA-Seq datasets revealed that the early growth response 1 (EGR1) transcription factor is a PLZF target for which its level of expression must be reduced to enable progesterone dependent hESC decidualization. Apart from furnishing essential insights into the molecular mechanisms by which progesterone drives hESC decidualization, our findings provide a new conceptual framework that could lead to new avenues for diagnosis and/or treatment of adverse reproductive outcomes associated with a dysfunctional uterus.

Fig 1. Progestin induction of PLZF expression in hESCs undergoing decidualization.

A) Relative transcript levels of PLZF, IGFBP-1, and PRL in hESCs cultured with vehicle (black bar (control)) or the EPC cocktail (red bar) at the indicated time points of cell culture. Note: Day 0 is the first day that hESCs were treated with vehicle or the EPC cocktail. B) Immunofluorescent detection of PLZF protein expression in hESCs cultured with vehicle or the EPC cocktail at the indicated time points of culture. Left and right columns show images of the same cell field stained for DAPI and PLZF immunopositivity respectively. After 6 days of EPC treatment, note the prominent nuclear staining for PLZF expression in a decidualized hESC cell (arrowhead). C) Relative transcript levels of PLZF in hESCs cultured for 6 days in vehicle control, E2, cAMP, or MPA alone. Note: For each transcript type (panel A) or reagent (panel C), results are reported relative to day 0 or vehicle treatment group respectively as the mean ± standard error (SE) from triplicate samples from one subject (three subjects were tested in total). *P<0.05, **P<0.01, and ***P<0.001. D-F) Immunohistochemical analysis of PLZF expression in human endometrial tissue biopsied during proliferative phase (D) and secretory phase (E and F) of the menstrual cycle. Black and white arrowheads in panel D and E indicate glandular epithelial and stromal cells respectively; white arrowhead in panel F indicates the nuclear localization of PLZF protein in stromal cells. Scale bar in panel D denotes 100 μm and applies to panel E; scale bar in panel F represents 100 μm.

Fig 2. The progesterone receptor mediates progestin induction of PLZF expression in hESCs.

A) Relative transcript levels of PLZF in hESCs cultured with vehicle (control), MPA, or MPA in the presence of RU486 for three days in stripped media. Note the marked reduction in MPA-induced PLZF transcription when RU486 is included. B) PLZF and PR transcript levels in hESCs transfected with control siRNA or siRNA targeting PGR and cultured in EPC media at the indicated time points. Results are reported as the mean ± SE from triplicate samples from one subject (three subjects were tested in total); *P<0.05 and **P<0.01. C) Distribution of PGR binding intervals (designated 1 to 10) along the PLZF gene locus is shown along with H3K27Ac active sites and DNase hypersensitivity clusters. The sequence logo for progesterone-response element (PRE) is shown along with the sequence of PREs for sites 3, 4, 6, and 8. D) Validation of PGR binding to the PLZF gene using chromatin immunoprecipitation (ChIP). The histogram shows PCR results following PGR ChIP from hESCs on sites 1 to 10 as depicted in panel C. A non-binding untranslated region (Utr12 (blue bar)) and the FKBP5 intronic region (FKBP5i (green bar)) were used as negative and positive controls respectively for PGR binding.

Fig 3. Progestin-dependent decidualization of hESCs requires PLZF.

A) Cultured for six days in EPC media, the comparative hESC morphology following transfection with control siRNA or PLZF siRNA is shown. A typical epithelioid hESC with a polygonal appearance is indicated with a black arrowhead in the presence of control siRNA. A white arrowhead shows a hESC that remains fibroblastic when cultured in the presence of PLZF siRNA. Bottom panels represent higher magnifications of regions demarcated by hatched boxes in the corresponding top panels. B) Transcript levels of IGFBP-1, PRL, and PLZF in hESCs transfected with control siRNA (black bar) or PLZF siRNA (red bar) and cultured in EPC media for the indicated time period. C) Western blot analysis of PLZF protein levels to determine the effectiveness of siRNA against PLZF from hESCs transfected with control siRNA or PLZF siRNA and cultured in EPC media. Lanes 1–3 and 2–4 represent hESCs transfected with control siRNA or PLZF siRNA respectively. Lanes 1–2 and 3–4 represent hESCs cultured in EPC media for 0 days and 6 days respectively; β-actin was used as loading control. D) Transcript levels of the progesterone-responsive genes: FOXOA1, HAND2, and HOXA10 in hESCs transfected with control siRNA or PLZF siRNA and cultured in EPC media. Results are reported as the mean ± SE from triplicate samples from one subject (three subjects were tested in total). *P<0.05, **P<0.01, and ns>0.05.

Fig 4. Identification of the PLZF cistrome by ChIP-Seq.

A) Histogram representing the distance of PLZF binding peaks from the transcriptional start site (TSS) of all genes. B) Representative zf-C2H2 DNA binding domain motif (MC00418) identified by SeqPos motif enrichment analysis. C) Venn diagram comparing genes bound both by PLZF with genes for which their expression changes in hESCs following EPC induced decidualization [8].

Fig 5. PLZF-mediated transcriptional repression of EGR1 is critical for hESC decidualization.

A) Location of PLZF binding sites in the promoter region of EGR1. B) The histogram shows the qPCR result following ChIP using the PLZF antibody on the EGR1 promoter region from panel A. C) Comparative transcript levels of EGR1 in hESCs transfected with control siRNA (black bar) or PLZF siRNA (red bar) and cultured in EPC media. D) Western blot analysis of PLZF protein levels from hESCs transfected with control siRNA (lanes 1–3) or PLZF siRNA (lanes 2–4) cultured in EPC media for 0 days (lanes 1–2) or 3 days (lanes 3–4).; β-actin was used as loading control. E) Relative transcript levels of IGFBP1, PRL, and EGR1 in non-transduced hESCs or hESCs virally transduced with control eGFP or EGR1 cDNAs (green bar) and cultured in EPC media. Results are reported as the mean ± SE from triplicate samples from one subject (a total of three subjects were tested). *P<0.05, **P<0.01, and ***P<0.001. F) Morphology of hESCs transduced with control eGFP or EGR1 cDNAs following six days of culture in the presence of the EPC cocktail; non-transduced hESCs cultured in vehicle were used as control.