The dynamic changes in the number of uterine natural killer cells are specific to the eutopic but not to the ectopic endometrium in women and in a baboon model of endometriosis

Abstract

Background: Endometriosis is a common condition associated with growth of endometrial-like tissue beyond the uterine cavity. Previous reports have suggested a role for uNK cells in the pathogenesis of endometriosis postulating that survival and accumulation of menstrual endometrial tissue in the peritoneal cavity may relate to a reduction in the cytotoxic activity of peripheral blood NK cells. We aimed to assess the differences in percentage of uNK cells and their phenotypical characterization in eutopic and ectopic endometrial samples from women with and without endometriosis and baboons with induced endometriosis.

Methods: Eutopic and ectopic endometrial samples from 82 women across the menstrual cycle with/without endometriosis and from 8 baboons before and after induction of endometriosis were examined for CD56 and NKp30 expression with immunohistochemistry, quantified using computer assisted image analysis. Curated secretory phase endometrial microarray datasets were interrogated for NK cell receptors and their ligands. In silico data was validated by examining the secretory phase eutopic endometrium of women with and without endometriosis (n = 8/group) for the immuno-expression of BAG6 protein.

Results: The percentage of uNK cells increased progressively from the proliferative phase with the highest levels in the late secretory phase in the eutopic endometrium of women with and without endometriosis. The percentage of uNK cells in ectopic lesions remained significantly low throughout the cycle. In baboons, induction of endometriosis increased the percentage of uNK in the ectopic lesions but not NKp30. Published eutopic endometrial microarray datasets demonstrated significant upregulation of NKp30 and its ligand BAG6 in women with endometriosis compared with controls. Immunohistochemical staining scores for BAG6 was also significantly higher in secretory phase eutopic endometrium from women with endometriosis compared with the endometrium of healthy women (n = 8/group).

Conclusions: The dynamic increase in the percentage of uNK cells in the secretory phase is preserved in the endometrium of women with endometriosis. The low number of uNK cells in human and baboon ectopic lesions may be due to their exaggerated reduction in hormonal responsiveness (progesterone resistance).

Keywords: Baboon; Endometriosis; Humans; Primate; Uterine natural killer cells.

Conflict of interest statement

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures

Fig. 1

Expression of CD56 and NKp30 in human endometrium. Representative micrographs showing CD56 expression by immunohistochemistry (brown DAB staining) in eutopic endometrial stromal cells of fertile control women (a, A-C) and in women with endometriosis (b, A-C) (400× magnification). NKp30 expression in eutopic endometrial stromal cells from fertile control women (a, D-F) and women with endometriosis (b, E-G). Staining in ectopic lesions are shown in (bD) (uNK cells) and (bH) (NKp30). Graphs comparing %CD56 (1c), %NKp30 (d) and ratio of NKp30:CD56 (e) in ectopic lesions (n = 6–9) and at different time points in the menstrual cycle (PP = proliferative phase; MSP = mid-secretory phase; LSP = late-secretory phase; n = 10 for each group in human samples (in both fertile controls ‘normal’ and patients with endometriosis). (f) Graph showing percentage of CD56+ uNK cells in eutopic endometrium and ectopic lesions across the menstrual cycle (n = 9 ectopic lesions with matched eutopic endometrium in 7/36 cases) demonstrating that levels remain low in ectopic lesions. (g) Graph showing percentage of CD56+ uNK cells in matched eutopic and ectopic endometrium (n = 7). P = 0.03, Wilcoxon matched pairs signed rank test

Fig. 2

Co-localisation of CD56 (a) and NKp30 (b) positive cells on serial sections from late secretory endometrium. Examples of cells stained with both markers are shown by black arrows

Fig. 3

Expression of CD56 and NKp30 in a baboon model of induced endometriosis. a Representative micrographs depicting CD56 (A, C, E, G, I), or NKp30 (B, D, F, H, J) expression in eutopic endometrial stroma cells of baboon samples during the three time-points: pre-inoculation (A, B), 3 (C, D) and 15 months (G, H) post-inoculation of the disease and expression in ectopic endometrial lesions 3 months (E, F) and 15 months (I, J) post-inoculation (400× magnification). Graphs comparing percentage of stromal CD56+ (b), NKp30+ (c) and ratio of NKp30+ to CD56 (d) cells prior to inoculation (n = 5), at 3 (n = 7) and 15 months (n = 5) in eutopic endometrium after the induction of endometriosis and in ectopic lesions at 3 (n = 5) and 15 months (n = 4) after induction of endometriosis

Fig. 4

BAG6 expression in mid-secretory phase human endometrium. Representative micrographs depicting BAG6 expression in the functional layer of the endometrium from a fertile control women (400× magnification) and b women with endometriosis (400× magnification). c Quickscore data comparing BAG6 immuno staining in the endometrium of normal control women compared with women with endometriosis during the mid-secretory phase and demonstrating significantly increased BAG6 immunoexpression scores in the endometriosis group (n = 8/group, P = 0.01, Mann Whitney U test). In full thickness endometrium, a gradient in staining intensity was observed from the functionalis to the basalis layer (D, 40× magnification). The basalis/functionalis demarcation is indicated by the dotted line with the basalis to the left of the line and the functionalis to the right