Changes in the Molecular and Functional Phenotype of Bovine Monocytes during Theileria parva Infection

Abstract

Theileria parva is the causative agent of East Coast fever (ECF), a tick-borne disease that kills over a million cattle each year in sub-Saharan Africa. Immune protection against T. parva involves a CD8⁺ cytotoxic T cell response to parasite-infected cells. However, there is currently a paucity of knowledge regarding the role played by innate immune cells in ECF pathogenesis and T. parva control. Here, we demonstrate an increase in intermediate monocytes (CD14⁺⁺ CD16⁺) with a concomitant decrease in the classical (CD14⁺⁺ CD16^-) and nonclassical (CD14⁺ CD16⁺) subsets at 12 days postinfection (dpi) during lethal infection but not during nonlethal T. parva infection. Ex vivo analyses of monocytes demonstrated upregulation of interleukin-1 beta (IL-1β) and tumor necrosis factor alpha (TNF-α) mRNA and increased nitric oxide production during T. parva lethal infection compared to nonlethal infection at 10 dpi. Interestingly, no significant differences in peripheral blood parasite loads were observed between lethally and nonlethally infected animals at 12 dpi. In vitro stimulation with T. parva schizont-infected cells or Escherichia coli lipopolysaccharide (LPS) resulted in significant upregulation of IL-1β production by monocytes from lethally infected cattle compared to those from nonlethally infected animals. Strikingly, monocytes from lethally infected animals produced significant amounts of IL-10 mRNA after stimulation with T. parva schizont-infected cells. In conclusion, we demonstrate that T. parva infection leads to alterations in the molecular and functional phenotypes of bovine monocytes. Importantly, since these changes primarily occur in lethal infection, they can serve as biomarkers for ECF progression and severity, thereby aiding in the standardization of protection assessment for T. parva candidate vaccines.

Keywords: Theileria parva; bovine; correlate of protection; function; innate immunity; intermediate; monocyte function; monocyte phenotype; monocytes; phenotype.

FIG 1

Gating strategy for the identification of monocyte subsets in cattle and proportions of monocyte populations in T. parva-infected animals. (A) PBMC were gated according to their complexity (side scatter [SSC]) and size (forward scatter [FSC]). Log₁₀ was used on the y axis (SSC) and a linear scale used on the x axis (FCS). (B) Dot plot depicting the three monocyte subsets in total PBMC, as follows: classical CD14⁺⁺ CD16⁻ (blue), intermediate CD14⁺⁺ CD16⁺ (red), and nonclassical CD14⁺ CD16⁺ (green). In addition, NK cells are shown in pink, and the identity of this population was verified by the expression of CD16 and CD335 (data not shown). (C) Dot plot showing classical (blue), intermediate (red), and nonclassical (green) monocytes. (D to F) Dot plots showing the relative proportions of monocyte subsets in representative uninfected, T. parva lethally infected, and T. parva nonlethally infected cattle. For dot plots shown in panels B to F, log₁₀ was used on both the x and y axes, and the numbers indicate the percentage of each monocyte subset. (G to I) Frequency of monocyte subsets in uninfected (uninfect, n = 15), T. parva lethally infected (lethal, n = 6), and T. parva nonlethally infected (non-lethal, n = 4) cattle. Bars indicate the mean percentage of each monocyte population. Monocyte subset proportions were compared using analysis of variance (ANOVA) and Tukey’s post hoc test. *, P < 0.05.

FIG 2

Total white blood cell counts and kinetics of changes in the proportion of classical, intermediate, and nonclassical monocytes during lethal and nonlethal T. parva infection of cattle. (A to C) Absolute numbers of total white blood cells (A), lymphocytes (B), and monocytes (C) in peripheral blood of lethally and nonlethally infected animals. Leukocyte counts were performed using a ProCyte Dx hematology analyzer (Idexx). (D to F) Percentages of classical (D), intermediate (E), and nonclassical (F) monocytes in PBMC from lethally and nonlethally infected cattle throughout acute infection. Averages of subset percentages were compared using Student's t test. *, P < 0.05.

FIG 3

Differential expression of MHC-II by bovine monocyte subsets during T. parva infection. (A to C) Percentages of MHC-II expression on classical (CD14⁺⁺ CD16⁻) (A), intermediate (CD14⁺⁺ CD16⁺) (B), and nonclassical (CD14⁺ CD16⁺) (C) monocyte subsets in uninfected (uninfect, n = 6), T. parva lethally infected (lethal, n = 6), and T. parva nonlethally infected (non-lethal, n = 4) cattle. Percentages of MHC-II-positive monocytes were compared using ANOVA and Tukey’s post hoc test. *, P < 0.05. Bars indicate the mean percentage of each monocyte population.

FIG 4

T. parva infection alters the expression of CD163 on the surface of bovine PBMC and monocyte subsets and the levels of soluble CD163 (sCD163) in sera from infected cattle. (A) Percentages of CD163⁺ cells in PBMC of uninfected (uninfect, n = 6), T. parva lethally infected (lethal, n = 6), and T. parva nonlethally infected (non-lethal, n = 4) cattle. (B) Amount (in nanograms per milliliter) of soluble CD163 (sCD163) in peripheral blood of uninfected (uninfect, n = 6), T. parva lethally infected (lethal, n = 6), and T. parva nonlethally infected (non-lethal, n = 4) cattle. (C to E) Percentages of CD163⁺ cells within classical (C), intermediate (D), and nonclassical (E) subsets in PBMC of uninfected (uninfect, n = 6), T. parva lethally infected (lethal, n = 6), and T. parva nonlethally infected (non-lethal, n = 4) cattle. CD163 surface expression and serum sCD163 levels in uninfected and infected animals were compared using ANOVA and Tukey’s post hoc test. *, P < 0.05. Bars in panels A and C to E indicate the mean percentage of each monocyte population.

FIG 5

T. parva upregulates the production of inflammatory mediators by monocytes during lethal infection. (A) Percentage of CD172a⁺ cells in PBMC of an uninfected, representative steer. (B) Positive selection of CD172a⁺ cells from PBMC of an uninfected, representative steer. (C) CD172a⁺ cells include the classical (CD14⁺⁺ CD16⁻), intermediate (CD14⁺⁺ CD16⁺), and nonclassical (CD14⁺ CD16⁺) monocyte subsets. (D and E) Expression of mRNA for IL-1β (D) and TNF-α (E) in CD172a⁺ cells from T. parva lethally infected (n = 6) and nonlethally infected (n = 4) cattle. Gene expression levels were compared using Student's t test. (F) Production of nitric oxide (NO) in CD172a⁺ cells from T. parva lethally infected (n = 6) and nonlethally infected (n = 4) cattle following 96 h of exposure to exogenous recombinant bovine IFN-γ (50 U/ml; Ciba-Geigy) plus recombinant human TNF-α (2,500 U/ml; R&D Systems). NO production was compared using ANOVA and Tukey’s post hoc test. (G) T. parva peripheral blood parasite load, determined by quantitative PCR, in T. parva lethally (n = 6) and nonlethally (n = 4) infected cattle. Parasite loads were compared using Student's t test. *, P < 0.05.

FIG 6

Differential response of monocytes from T. parva lethally and nonlethally infected cattle to T. parva infected cells and E. coli LPS. (A) Intracellular staining for the T. parva polymorphic immunodominant molecule in a representative schizont-infected cell line. (B) Intracellular staining for IFN-γ in a representative schizont-infected cell line. (C and D) Expression of mRNA for TNF-α (C) and IL-10 (D) in a representative schizont-infected cell line at 4 and 16 h after Histopaque centrifugation. (E) Levels of soluble IL-1β in CD172a⁺ cell supernatant following stimulation of monocytes from uninfected (n = 6), T. parva lethally infected (n = 6), and T. parva nonlethally infected (n = 4) cattle with T. parva schizont-infected cells (using a transwell format) or E. coli LPS (1 μg/ml). Amounts of soluble IL-1β were compared using ANOVA and Tukey’s post hoc test. (F) Expression of IL-10 mRNA in CD172a⁺ cells from T. parva lethally infected (n = 6) and nonlethally infected (n = 4) cattle following stimulation with T. parva schizont-infected cells (using a transwell format) or E. coli LPS (1 μg/ml). IL-10 mRNA expression was compared using ANOVA and Tukey’s post hoc test. *, P < 0.05.