Sequential analysis of Anaplasma phagocytophilum msp2 transcription in murine and equine models of human granulocytic anaplasmosis

Abstract

Anaplasma phagocytophilum causes human granulocytic anaplasmosis by inducing immunopathologic responses. Its immunodominant Msp2 protein is encoded by a family of >100 paralogs. Msp2 (msp2) expression modulates in the absence of immune pressure, and prolonged in vitro passage modulates in vivo virulence. Because programmed MSP2 expression occurs in Anaplasma marginale, we hypothesized a similar event in A. phagocytophilum in vivo, with specific Msp2 expression triggering immunopathologic injury or clinical manifestations of disease. We examined msp2 transcripts in 11 B6 mice and 6 horses inoculated with low- or high-passage A. phagocytophilum Webster strain. Blood was sequentially obtained through 3 weeks postinfection for msp2 reverse transcription-PCR. Horses were additionally assessed for clinical manifestations, seroconversion, complete blood count, blood chemistry, and cytokine gene transcription. In both species, there was no consistent emergence of msp2 transcripts, and all 22 msp2 variants were detected in both passage groups. Clinical severity was much higher for high-passage-infected than for low-passage-infected horses, preceded by higher levels of blood gamma interferon transcription on day 7. Antibody was first detected on day 7, and all horses seroconverted by day 22, with a trend toward lower antibody titers in low-passage-infected animals. Leukocyte and platelet counts were similar between experimental groups except on day 13, when low-passage-infected animals had more profound thrombocytopenia. These findings corroborate studies with mice, where msp2 diversity did not explain differences in hepatic histopathology, but differ from the paradigm of low-passage A. phagocytophilum causing more significant clinical illness. Alteration in transcription of msp2 has no bearing on clinical disease in horses, suggesting the existence of a separate proinflammatory component differentially expressed with changing in vitro passage.

FIG. 1.

Cumulative severity scores for both low- and high-passage-infected horses, with three animals represented in each group. Clinical parameters that were examined and included in the severity score were ataxia (grade 1 to 5), lethargy (grade 1 to 4), limb swelling (grade 1 to 4), reluctance to move (grade 1 to 4), and presence of petechiae (number observed on oral mucosa). Aph, A. phagocytophilum.

FIG. 2.

(Top) Curves for low- and high-passage-infected horses, with fold change in IFN-γ and IL-8 transcription plotted with platelet counts to demonstrate that thrombocytopenia is preceded by increasing IFN-γ but not IL-8 transcription. lp Aph and hp Aph, low- and high-passage A. phagocytophilum infection, respectively. (Bottom) Curves for low- and high-passage-infected horses, with fever plotted against accumulated clinical scores. Note the increase in clinical scores and fever at days 7 to 10. A second peak in clinical score severity occurred between days 12 and 16, which was unrelated to fever but coincided with the presence of morulae between days 13 and 16 followed by a peak in IL-8 transcription (see Results).

FIG. 3.

When results were normalized to starting platelet counts, low-passage-infected horses developed a more profound thrombocytopenia than high-passage-infected animals, but only on day 13 (P < 0.05, Student's t test). No significant differences were noted in the white blood cell (WBC) count change between low- and high-passage-infected horses throughout the entire experimental period. Ap, A. phagocytophilum. Error bars indicate standard errors of the means.

FIG. 4.

Diversity of msp2 gene transcription of 22 variants in the peripheral blood of horses infected with high-passage (A) or low-passage (B) A. phagocytophilum. Transcription became evident by day 7 in both groups, peaked between days 13 and 16, and was mostly resolved by day 22. Of the 22 msp2 variants examined, all were detected in both groups of horses, and the patterns of transcription appeared to be similar.

FIG. 5.

Mean quantities of msp2 transcripts in low- and high-passage-infected horses through day 22 after infection. Note the similarity in emergence of transcripts between days 7 and 20, with a greater number of transcripts in high-passage- than in low-passage-infected horses. This reflects the disproportionate emergence of a few msp2 transcripts in high-passage- compared to low-passage-infected animals at day 13. Bars represent standard errors of the means.

FIG. 6.

msp2 transcript diversity found in mice inoculated with passage 5 and passage 26 Webster strain A. phagocytophilum. Note the lack of programmed transcript emergence within and among mice over time. Each mark for “days after infection” represents a different single mouse at that specific time point.