Immature NK cells, capable of producing IL-22, are present in human uterine mucosa

Abstract

NK cells are the dominant population of immune cells in the endometrium in the secretory phase of the menstrual cycle and in the decidua in early pregnancy. The possibility that this is a site of NK cell development is of particular interest because of the cyclical death and regeneration of the NK population during the menstrual cycle. To investigate this, we searched for NK developmental stages 1-4, based on expression of CD34, CD117, and CD94. In this study, we report that a heterogeneous population of stage 3 NK precursor (CD34(-)CD117(+)CD94(-)) and mature stage 4 NK (CD34(-)CD117(-/+)CD94(+)) cells, but not multipotent stages 1 and 2 (CD34(+)), are present in the uterine mucosa. Cells within the uterine stage 3 population are able to give rise to mature stage 4-like cells in vitro but also produce IL-22 and express RORC and LTA. We also found stage 3 cells with NK progenitor potential in peripheral blood. We propose that stage 3 cells are recruited from the blood to the uterus and mature in the uterine microenvironment to become distinctive uterine NK cells. IL-22 producers in this population might have a physiological role in this specialist mucosa dedicated to reproduction.

Figure 1. The uterine mucosa contains committed NK cell precursors

A: Stages 1 – 4 of NK development in SLT [8]. B: Tonsillar leukocytes were mechanically dissociated and magnetically depleted for CD3+ and CD19+ cells. The remaining cells were stained with CD34, CD117 and CD94 and examined by flow cytometry. Cells were gated on scatter, and then on CD34+CD94− or on CD34− cells. Stages 1 - 4 are highlighted. Plots are representative of at least three independent samples. C and D: Decidual and endometrial leukocytes were obtained by collagenase digestion and depleted of stromal cells by a two-hour incubation on plastic. The non-adherent cells were then magnetically enriched for CD34+ cells, stained with CD34, CD117 and CD94 and examined by flow cytometry. Cells were gated by scatter and then on CD34+CD94− or on CD34− cells, demonstrating a single population of CD34+ cells, a population of CD117+CD94− corresponding to stage 3, and a CD94+ population corresponding to stage 4. CD117 and CD94 staining on CD34-enriched and non-enriched cells was similar (Supplementary Figure 1). Each panel is representative of at least three independent samples.

Figure 2. CD34+ cells in the uterine mucosa are endothelial cells

A and B: Serial sections of decidua stained for isotype control (A) and CD34 (B) showing positive staining on endothelial cells. C – E: Gating by scatter and on CD34+ endometrial cells, C, D and E show CD45, ICAM-2 and PECAM-1 staining, respectively. F – H: Phenotype of CD34+ decidual cells, as above. Antibody staining is shown by the bold trace, isotype control staining by a dotted line. Each dot plot is representative of at least three independent samples.

Figure 3. Uterine stage 3 cells are phenotypically similar to SLT stage 3

Phenotype of CD117+CD94−CD3− cells in the decidua and endometrium. Antibody staining is shown by the bold trace, isotype control staining by a dotted line. For CD56 and NKp46, the level of staining on mature CD94+ decidual NK cells in the same sample is represented by a broken line. For decidua, each histogram represents one of at least three independent samples, examined for a single marker. For endometrium, each histogram represents a single sample, examined for a single marker. Endometrial samples were both proliferative and secretory phase. NT: not tested.

Figure 4. NK cell potential of stage 3 cells isolated from uterine mucosa and peripheral blood

A - D: CD117+CD94−CD3− stage 3 cells were sorted from CD117 enriched decidual (A) or peripheral blood (C) leukocytes. The sorted populations were cultured for two weeks in 1nM recombinant human IL-15 and re-examined for CD117 and CD94 expression by flow cytometry (B and D). Numbers indicate the percentage of cells falling into the gate. Cells recovered at the end of decidual cultures were examined for CD56 expression. The dot plots show CD56 staining against side scatter, for cells in the indicated gates, and mean fluorescence intensity is indicated (B). Each culture shown is representative of at least 3 independent experiments. E – F: Decidual CD117+CD94−CD3− cells were cultured as above, either alone or with the introduction of a 0.1% or 1% “spike” of sorted CD117−CD94+CD3− cells. The percentage of cells falling into the CD94+CD117^low gate at the end of culture was determined by flow cytometry. Representative of 2 independent experiments.

Figure 5. Uterine stage 3 cells produce IL-22

A: RT-PCR for IL22, LTA, RORC and GAPDH in three matched pairs of decidual stage 3 and 4 cells. pDNRdual-IL22, pDNRdual-LTA and pDNRdual-RORC (Invitrogen) were positive controls. The identities of the products were confirmed by sequencing. B and C: Intracellular staining for IL-22 on decidual stage 3 (B) and stage 4 (C) cells. Staining on stages 3 and 4 are represented by the bold trace, staining on CD3+ cells in the same sample is represented by a dotted line. Each histogram is representative of five independent experiments