# The fate of human SUSD2+ endometrial mesenchymal stem cells during decidualization

- **DOI:** [10.1016/j.scr.2022.102671](https://doi.org/10.1016/j.scr.2022.102671)
- **Rating:** Limited (2.5 out of 5)
- **Verdict:** Cultured SUSD2-positive endometrial cells show a decidual secretory response, but assay and identification problems undermine claims about preserved stemness and cell transition.
- **Framework version:** v5.13
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-11-13
- **Evaluation page:** https://www.nabu.science/paper/10.1016%2Fj.scr.2022.102671

## Key claims

1. Our results suggest that SUSD2+ eMSC undergo decidualization in vitro, while maintaining MSC plasma membrane phenotype.
   - Support: Tentative
   - Type: Primary finding
   - Basis: decidual secretion supports a response, but small surface-marker comparisons and 20.3% CD45 positivity undermine maintenance of the stated MSC phenotype
   - Location: Abstract; Section 3.4, Figures 4–5
2. Progesterone treatment induced changes in the decidual gene expression profile in non-pregnant SUSD2+ eMSC.
   - Support: Tentative
   - Type: Secondary
   - Basis: combined progestin–cAMP treatment prevents attribution of the gene-expression response to progesterone alone
   - Location: Abstract; Section 3.4, Figure 5A
3. Histological analysis revealed a significantly lower abundance of SUSD2+ eMSC in 1st trimester and full term samples compared to non-pregnant samples, p = 0.0296 and 0.005, respectively.
   - Support: Tentative
   - Type: Secondary
   - Basis: cross-sectional stained-area fractions cannot distinguish cell abundance from tissue-composition effects
   - Location: Abstract; Section 3.1, Figure 1B
4. Pregnancy reduced the abundance of SUSD2+ eMSC, however eMSC function remains intact.
   - Support: Tentative
   - Type: Secondary
   - Basis: cross-sectional sampling and small functional comparisons after culture expansion cannot establish pregnancy-induced loss or preserved in vivo function
   - Location: Abstract; Sections 3.2–3.3; Discussion
5. Data analysis of a publicly available single cell RNA-seq data set revealed differential expression of several mesenchymal and epithelial signature genes between the SUSD2+ eMSC and the decidual stromal cells, suggesting mesenchymal-to-epithelial transition occurs during decidualization.
   - Support: Tentative
   - Type: Secondary
   - Basis: a static comparison of distinct cell populations with conflicting marker patterns cannot identify a lineage transition
   - Location: Abstract; Section 3.6, Figure 6D; Discussion

## Strengths

- **Decidual secretion measured in donor cultures** PRL and IGFBP1 secretion increased after combined progestin–cAMP treatment in both SUSD2-positive and SUSD2-negative cultures. Measurements were normalized to cell number.
  - Dimension: methodological rigour
  - Location: Sections 2.9, 2.12 and 3.5; Figure 5B–C
- **Term decidua extends the descriptive comparison** The study compares clonogenicity and differentiation in cells isolated from term decidua parietalis and non-pregnant endometrium, extending the tissue-stage characterization of SUSD2-positive cells.
  - Dimension: contribution
  - Location: Sections 3.1–3.3; Figures 1–2
- **Conflicting prior findings receive explicit discussion** The Discussion identifies the failure to reproduce an earlier difference in decidual secretion and acknowledges epithelial-marker results that oppose the proposed transition.
  - Dimension: positioning
  - Location: Discussion, Murakami comparison and MET-marker interpretation

## Limitations

- **Population purity and differentiation controls remain unresolved** The gated population averaged 20.3% CD45-positive cells despite the paper's CD45-negative MSC definition. Untreated controls also showed myocyte differentiation, limiting that assay's specificity.
  - Dimension: methodological rigour
  - Location: Introduction; Sections 3.2 and 3.4; Figures 2 and 4
- **Experimental contrasts cannot identify broader biological claims** Static cell-cluster comparisons do not establish a lineage transition, combined MPA–cAMP treatment cannot isolate progesterone effects, and stained-area fractions do not directly identify changes in cell abundance.
  - Dimension: methodological rigour
  - Location: Sections 2.4, 2.9, 2.13 and 3.6; Figures 1 and 6
- **Preserved function exceeds the acknowledged culture limitations** The Abstract's intact-function conclusion rests on small, non-equivalence comparisons after culture expansion. The Discussion itself explains that expansion and dedifferentiation could make term-derived cells resemble non-pregnant cells.
  - Dimension: positioning
  - Location: Abstract; Sections 3.2–3.3; Discussion, culture limitations

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - Our results suggest that SUSD2+ eMSC undergo decidualization in vitro, while maintaining MSC plasma membrane phenotype: "whereas on average, 20.3% cells were CD45 positive (Fig. 4)." (Section 3.4, paragraph 2) vs "the absence CD45 and CD34 expression (Dominici et al., 2006)" (Section 1, paragraph 1). Compare the reported 20.3% CD45-positive gated cells with the paper's stated absence-of-CD45 criterion for an MSC phenotype. → Not supported by the paper's own data
  - Progesterone treatment induced changes in the decidual gene expression profile in non-pregnant SUSD2+ eMSC: "IL1B1, BDNF, BMP2, KDR, GDF7, PTK2, FZD9 were upregulated" (Section 3.4 (qPCR results)) vs "also upregulated the expression of VEGF-A fourfold in progesterone treated compared with control" (Discussion, paragraph on VEGFR-2/KDR). Compare the significant-gene list, which includes KDR but not VEGFA, with the Discussion's fourfold VEGF-A assertion. → Effect on result unclear
- **Reporting and transparency:** Noted. Ethics approval, consent and competing interests are declared. Statistical-test names are misreported, and assay-specific sample sizes for Figure 2A–B and the underlying SUSD2-negative qPCR results are not supplied.
  - Conflict of interest: declared · Ethics approval: declared
- **Reference integrity:** Clean
  - 67 references in manuscript
  - 66 of 66 checkable references found in an index
  - 1 could not be checked — not the same as no match found
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-11-13
