In this study, two approaches were taken to characterise, enumerate and compare the levels of white cell populations in the blood of four genetically distinct macaque populations. Although there are limited reports comparing the immune cells of humans with those in certain macaque populations 10,11 to our knowledge, this is the first time a direct comparison of four genetically distinct macaque populations has been performed using the same assays.
Evaluation of the cellular composition of anti-coagulated blood using a haematology analyser enables an unbiased analysis of cell population frequency and number per ml of blood for all types of cells present. Conversely, data originating from PBMC samples relates to mononuclear cell populations only and is proportional, but the main advantages of flow cytometric immunophenotyping is that it can generate more detailed data and can be applied to archived samples enabling retrospective interrogation of materials.
The haematology analyser derived data set demonstrated that lymphocytes and monocyte counts were different between rhesus and cynomolgus, whereas there was little difference between the genetically distinct cynomolgus populations. Separating the populations into TB disease susceptible (RM and MCM) and less susceptible (ICM and CCM), it was only lymphocytes levels that were different between these groupings. Eosinophil and neutrophils counts were different between groups, but the differences were not common between the susceptible and less susceptible groups and so, it is perhaps unlikely that basal eosinophils and neutrophil numbers have a significant bearing on susceptibility to TB disease.
Immunophenotyping studies of PBMC were used to evaluate the subtypes of the lymphocytes and monocytes present. When the least susceptible populations (ICM and CCM) were grouped and compared to the susceptible populations (RM and MCM), key differences were identified in the CD4:CD8 ratio and mDC proportions.
CD4+ T-cells were identified using PCA to discriminate between the macaque populations and CD4+ T-cells also correlated with TB-induced disease burden measured using a pathology-based scoring system. Although it is known that CD4+ T-cells are necessary for control of TB infection, as CD4+ cell depletion due to HIV is a risk factor for TB infection 12; HIV is a disease state with an abnormally low number of CD4+ T-cells whereas these are healthy animals so in normal conditions it appears that high numbers of CD4+ T-cells may impact on TB disease progression. There are also many subtypes of CD4+ T-cells to be considered and this analysis may be too simplistic, and subtypes and activation status also need to be taken into account in future analysis.
CD8+ T-cells and CD16+ NK cells contributed to the PCA second dimension, highlighting a potential difference in cells that have cytolytic roles between primate populations. We saw a non-significant correlation between higher numbers of CD8+ cells before infection that correlated with lower pathology scores. In CD8+ depletion studies in primates, CD8+ T-cells have been shown to be important in the control of infection13. The CD4:CD8 ratios defined in ICM and CCM showed a balance in CD4+ and CD8+ T-cells, whereas in MCM a skew towards the CD8 population was found, in line with the previous report from Zitsman et al14, whereas the populations in RM were more biased towards CD4. A low CD4:CD8 ratio has been found to be a predictor TB in HIV patients 15 so as MCM do have the lowest CD4:CD8 ratio, this could be a contributing risk factor in their susceptibility to TB.
NK cell transcripts were found to be lower in CMV + infants, that went on to develop TB 7 suggesting a link between NK cells and TB susceptibility. A study comparing NK subsets between persons from a TB endemic country with TB naïve persons showed that there was little difference in the frequency of cytolytic NK cells, but that those NK cells had different reactivity and functional capacity16. Therefore, in this study, the NK subtype proportions were most similar between MCM and CCM which have very different susceptibilities to TB so further investigation into the functionality of the NK cells is required to determine whether there are differences in their capacity to react to TB and influence disease progression.
Monocytes contributed to the variance in the first dimension in the PCA, and there was a difference in the proportion of CD14+ classical monocytes between rhesus and cynomolgus macaques, but they did not correlate with pathology. Dijkman et al saw differences in monocyte subtypes and cytokine production post-infection with TB between rhesus and cynomolgus macaques 17 and so looking post-infection at differences whether there are differences between populations in how they respond to infection that relates to their basal subtypes would be something to examine further in future work.
Antigen presenting cells (APCs) such as mDCs are a key component of T-cell activation, and this population was present at significantly higher frequencies in the macaque populations that are less susceptible to TB. Efficient priming of T-cells is considered to be key in protecting against TB and TB modulates DC activity by delaying their ability to migrate to the lymph nodes, hampering the formation of an effective immune response and giving the TB infection time to establish 18. Furthermore DCs have been found to be present at lower levels in patients with TB 19. Having a higher number of mDCs has the potential to confer an advantage by increasing the likelihood of migration to the lymph nodes and increasing the interactions with T-cells to promote an early immune response to infection. Neutrophils have an important role to shuttle bacteria to mDCs and enhance antigen presentation, and TB has been shown to affect neutrophil apoptosis, leading to less efficient DC priming 20 and so the higher numbers of neutrophils in the CCM, coupled with higher numbers of mDCs may be contributing factors to their control of TB.
Overall, these studies have revealed differences in the cellular composition of peripheral blood in four genetically distinct macaque populations, and particularly between rhesus and cynomolgus macaques in terms of lymphocyte populations. The concordance of findings from haematology analyser-based and flow cytometry-based measurements, supports the concept that there are fundamental differences in the makeup of the immune systems of these species. Others have noted that macaques vary genetically substantially between geographical locations 21, and recommend caution when comparing data from different models for the same diseases as contributing factors could obscure risk factor-disease associations, or lead to artificial associations. Therefore, it is important to understand the genetic background of the animals used in studies, together with the potential implications that any consequent constitutive differences between populations may have on the experimental outcome. Characterisation of macaque populations provides the opportunity to select populations with desirable characteristics for specific studies so differences can be exploited to further understand the factors required to promote a successful immune system.