Innate Lymphoid Cell Diversity in Pregnancy and Ovarian Cancer
Many parallels have been drawn between the mechanisms that support successful pregnancies and those that facilitate tumour expansion. These include an immune profile bearing the hallmarks of immune suppression and reduced cytotoxicity. Innate Lymphoid Cells (ILCs) are a diverse group of lymphocytes which exhibit tissue specific phenotypes and functions. They are found in both the uterine lining during early pregnancy (decidua) and the ascites of ovarian cancer patients. Despite several attempts, there is still no consensus regarding the phenotypic profiles, subset composition and functional roles of ILCs in either tissue. To address this, I have developed a mass cytometry panel and stained resting and stimulated ILCs from peripheral blood, decidua and ascites.
Within the decidua, a number of ILC subsets are identifiable including three main subsets of decidual NK cells (dNK), dNK1-3. The most abundant subset, dNK1, express receptors for HLA class I found on invading extravillous trophoblast (EVT) including Killer-cell Ig-like receptors (KIR), LILRB1 and NKG2A. In contrast to dNK2 and dNK3, dNK1 respond poorly to ‘missing self’ and PMA plus ionomycin treatment. However, upon cross-linking of an activating KIR they can degranulate and produce cytokines/chemokines. dNK1 responsiveness to activating KIR increases with the expression of additional KIR which may be a result of increased granularity. This is the first time mass cytometry has been used to characterise the phenotype and function of decidual ILCs.
KIR variants are associated with infectious disease and pregnancy outcome. KIR2DL1 is an inhibitory KIR present on both KIR A and KIR B haplotypes. KIR2DL1 alleles typically found on KIR A haplotypes (KIR2DL1A) in Caucasians are associated with increased risk of developing pre-eclampsia. Using a novel antibody combination, I am able to stain for NK cells expressing particular KIR2DL1 allotypes. KIR2DL1A and KIR2DL1B are co-dominantly expressed by NK cells from peripheral blood (pbNK) and decidua but KIR2DL1A is preferentially expressed. Allotypes display similar inhibitory capacities but KIR2DL1A+ pbNK degranulate more in response to ‘missing self’ than KIR2DL1B+ pbNK. This approach is particularly powerful because KIR2DL1 allotypes can be compared within the same individual and have therefore experienced the same HLA class I environment.
NK cells are increasingly being targeted in cancer immunotherapies and can exhibit both pro- and anti-tumorigenic properties. They are the major ILC subset within the ascites of ovarian cancer patients but phenotypic and functional descriptions are conflicting. Once again applying the CyTOF methodology, I find that there are three major subsets of ascites derived NK cells (aNK). These include previously described CD56bright and CD56dim aNK subsets and an additional NKG2Ahigh KIRlo subset which expresses tissue resident markers (TR aNK). Upon PMA plus ionomycin treatment, the proportion of TR aNK cells producing XCL1 and IFNg is higher than that of other aNK subsets. TR aNK are also highly responsive to ‘missing self’ and, similar to dNK cells, increased responsiveness is associated with increased granularity. A detailed characterisation of ascites derived ILCs is important to better understand existing cancer treatments and to provide insight for designing new therapeutic strategies.