The immune cell landscape and response of Marek’s disease resistant and susceptible chickens infected with Marek’s disease virus

Abstract

Genetically resistant or susceptible chickens to Marek’s disease (MD) have been widely used models to identify the molecular determinants of these phenotypes. However, these prior studies lacked the basic identification and understanding of immune cell types that could be translated toward improved MD control. To gain insights into specific immune cell types and their responses to Marek’s disease virus (MDV) infection, we used single-cell RNA sequencing (scRNAseq) on splenic cells from MD resistant and susceptible birds. In total, 14,378 cells formed clusters that identified various immune cell types. Lymphocytes, specifically T cell subtypes, were the most abundant with significant proportional changes in some subtypes upon infection. The largest number of differentially expressed genes (DEG) response was seen in granulocytes, while macrophage DEGs differed in directionality by subtype and line. Among the most DEG in almost all immune cell types were granzyme and granulysin, both associated with cell-perforating processes. Protein interactive network analyses revealed multiple overlapping canonical pathways within both lymphoid and myeloid cell lineages. This initial estimation of the chicken immune cell type landscape and its accompanying response will greatly aid efforts in identifying specific cell types and improving our knowledge of host response to viral infection.

Single-cell RNA-seq mapping of chicken peripheral blood leukocytes

“…This latter study states limited resolution as a reason for not being able to identify different cell types, particularly T-cells, further. In comparison, the latter study reports that approximately 50% less genes were detected per cell [ 18 ] compared to the present results, which possibly explain their poor detection of genes with low expression levels e.g., CD4. Thus, the sequencing depth, numbers of analysed cells and bioinformatic approach of the current study generated considerably more information on leukocyte subtypes compared to these previous reports on chicken leukocytes and can probably be considered a “minimal requirement” for useful leukocyte subtype phenotyping by single-cell RNA-seq.…”

“…The identification of different T-cell populations in the current data set proved more challenging than that of B-cells and monocytes. Such problems have also been reported in other studies of single-cell RNA transcriptomics of T-cell subpopulations, for mouse and human T-cells [ 22 ], as well as for horse PBMC [ 15 ] and chicken spleen cells [ 18 ]. This can be due to limited resolution due to insufficient sequencing depth or to differences between mRNA expression levels and protein expression of markers by which T-cell populations are currently defined, e.g., CD4, CD8α and CD8β, or more probably a combination of both.…”

“…We found that the expression of CD4, CD8α and CD8β never comprised all cells in a cluster. Similarly, very low expression of CD4 and CD8α was reported for chicken spleen cells [ 18 ]. Nonetheless, we were able to further identify the clusters comprising T-cells into: TCRγ/δ + T-cells, CD4 + cells, CD8αβ + cells and “cytolytic cells”.…”

“…We therefore believe that this difference in normalisation method could have impacted the interpretation of results between these two studies. Moreover, a study of single-cell RNA-seq of chicken spleen cells reports 12 clusters that were identified as three general T-cell clusters, two TCRγ/δ T-cell enriched clusters, two B-cell clusters, two macrophage clusters, two granulocyte clusters and one cluster of antigen presenting myeloid cells [ 18 ]. This latter study states limited resolution as a reason for not being able to identify different cell types, particularly T-cells, further.…”

“…Single-cell RNA-seq has fruitfully been used to study mixed immune cell populations from several “non-mainstream” species such as pigs [ 14 ], horses [ 15 ], and teleost fish [ 16 ]. For chicken immune cells, purified “lymphocytes” [ 17 ], spleen cells [ 18 ] and bursal cells [ 19 ] have been used for single-cell RNA-seq but the focus of these studies have been on different viral infections and limited immune cell identification was reported. Peripheral blood is an easily accessible source of leukocytes but mRNA analysis of these in the chicken is complicated by presence of both nucleated red blood cells and thrombocytes.…”