Babesiosis is a tick-borne zoonosis caused by protozoans of the genus Babesia, apicomplexan parasites that replicate within erythrocytes. However, unlike related Plasmodium species, the pathogenesis of Babesia infection remains poorly understood. The primary etiological agent of babesiosis in the United States is B. microti. In healthy individuals, tick-transmitted infection with Babesia causes no specific clinical manifestations, with many having no symptoms at all. However, even in asymptomatic people, a Babesia carriage state can be established that can last up to a year or more. Current blood bank screening methods do not identify infected donors, and Babesia parasites survive blood-banking procedures and storage. Thus, Babesia can also be transmitted by infected blood, and it is currently the number one cause of reportable transfusion-transmitted infection in the United States. Despite a significant impact on human health, B. microti remains understudied. In this study, we evaluated the course of Babesia infection in three strains of mice, C57BL/6J, BALB/cJ, and C3H-HeJ, and examined the contribution of multiple immune parameters, including TLRs, B cells, CD4+ cells, IFN-γ, and NO, on the level of parasitemia and parasite clearance during acute babesiosis. We found that B. microti reaches high parasitemia levels during the first week of infection in all three mice strains before resolving spontaneously. Our results indicate that resolution of babesiosis requires CD4 T cells and a novel mechanism of parasite killing within infected erythrocytes.

Human babesiosis is an emerging tick-borne disease that is endemic in the northeast, upper midwest, and Pacific northwest of the United States. Babesia is a threat to the blood supply and is now the most common transfusion-transmitted infection in the United States that is reported to the U.S. Food and Drug Administration (1, 2). In North America, Babesia microti is the primary etiologic agent for most cases of babesiosis, with B. duncani responsible for cases of babesiosis in the Pacific northwest (1, 3, 4).

Babesia are piroplasms, small pear-shaped protozoan parasites that belong to the phylum Apicomplexa, which includes the other human pathogens Plasmodium, Toxoplasma, and Cryptosporidium. Babesia are intracellular pathogens that replicate exclusively in the cytosol of erythrocytes (5). Natural transmission occurs through the bite of an infected Ixodes tick. Tick transmission in otherwise healthy individuals usually results in a mild illness with general viral-like symptoms or an asymptomatic infection. In either case, a carrier state can be established that can persist in some individuals for up to 2 y (6, 7). Immunocompromised or asplenic individuals, the very young, and those older than 50 y are at risk for severe clinical manifestations, including adult respiratory distress syndrome, pulmonary edema, disseminated intravascular coagulation, congestive heart failure, renal failure, coma, splenic rupture, or prolonged relapsing illness. Mortality of 10% is reported for hospitalized individuals, despite treatment (1, 4). Our blood supply is at risk from asymptomatic Babesia carriers donating infected blood (810). The parasite survives standard blood-banking procedures and storage, is easily transmitted with infected erythrocytes (1113), and there is no U.S. Food and Drug Administration–approved method to detect the parasites in donated blood (1, 10, 14). Transfusion-transmitted babesiosis typically results in more severe disease than tick-transmitted infection, because the number of parasites introduced via transfusion is higher than transmitted by the tick and because transfusion recipients are more likely to be members of a high-risk population. Mortality of 30% is reported for transfusion-transmitted babesiosis. It is now the most common reported cause of death in the United States from transfusion-transmitted infection (1, 2, 14, 15).

Despite the impact of babesiosis on human health, the pathogenesis of babesiosis and the immune response to B. microti and B. duncani is poorly defined. Complicating the ability to study these parasites is the fact that B. microti and B. duncani cannot be cultured in vitro, requiring passage in live hosts. The few studies performed generally support a role for T cells and macrophages, but not B cells, in the control of parasitemia during acute disease in murine infection models (1619). However, these studies are inconsistent in defining the essential function of each of these factors in the resolution of infection.

In this study, we compared the course of B. microti infection in three strains of mice: C57BL/6J, BALB/cJ, and C3H/HeJ, to evaluate the effect of strain diversity on the level of parasitemia and parasite clearance during acute babesiosis. We also evaluated the role of CD4 cells, B cells, TLR signaling, IFN-γ, and inducible NO synthase (iNOS) in acute babesiosis. Our results indicate that resolution of babesiosis occurs through a novel mechanism that involves the death of B. microti parasites within infected erythrocytes. Death of parasites within infected erythrocytes did not require MyD88, iNOS, mature B cells, CD4 cells, or IFN-γ. However, CD4 cells and, to a lesser extent, IFN-γ and B cells are important for complete resolution of parasitemia and in preventing prolonged carriage of parasites.

The following mouse strains were used for experiments: BALB/cJ, C57BL/6J, C3H/HeJ, B6.129P2 (SJL)-Myd88tm.1.1Defr/J, B6.129S7-Ifngr1tm1Agt/J, B6.129P2-Nos2tmaLau/J, B6.129S2-Ighmtm1Cgn/J, B6.129S2-Cd4tm1Mak/J. All mice were females between 6–10 wks of age and were obtained from The Jackson Laboratory (Bar Harbor, ME). All animal studies were approved by the New York Medical College Institutional Animal Care and Use Committee.

The B. microti Gray strain, originally isolated from a human infection on Nantucket Island, Massachusetts in 1970, was obtained from the American Type Culture Collection (ATCC 30221) and was used for all of the studies (20, 21). For experiments, four or five mice per group were infected by i.p. inoculation of 100 μl of infected blood obtained from an infected mouse at the height of parasitemia (5–7 d postinfection; 30–40% parasitemia). Mice were infected at the same time with the same parasite inoculum to ensure that all mice received equal numbers of viable parasites. B. microti parasites were maintained using the same methods by serial passage in BALB/cJ mice and C57BL/6J mice. Blood was collected daily by tail snip, and blood smears were prepared and stained with Giemsa or New Methylene Blue to assess the number of parasites per erythrocyte and the percentage of mature erythrocytes versus reticulocytes.

Blood smears were stained with Giemsa Stain (Richard-Allan Scientific, Kalamazoo, MI), as per the manufacturer’s instructions, to monitor parasitemia or with New Methylene Blue Solution supravital stain (Electron Microscopy Sciences, Hatfield, PA), as per the manufacturer’s instructions, to monitor the number of mature erythrocytes versus reticulocytes. For parasitemia, mature erythrocyte, and reticulocyte analysis, four or five counts of 100 RBCs were performed under 100× brightfield magnification per mouse, and the average number and SD determined.

Plasma cytokine levels were quantified using a Bio-Rad Bio-Plex Pro Mouse Cytokine 23-plex Assay at the CytoPlex-Core Facility at the Yale School of Medicine. Mouse cytokines analyzed in this study were IL-1α, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-9, IL-10, IL-12 (p40), IL-12(p70), IL-13, IL-17, Eotaxin (CCL11), G-CSF, GM-CSF, IFN-γ, KC, MCP-1 (CCL2), MIP-1α (CCL3), MIP-1β (CCL4), RANTES (CCL5), and TNF-α. Blood was collected from uninfected or infected mice on days 5, 7, 11, and 15 postinfection in the presence of 3.0% sodium citrate as an anticoagulant. Blood was centrifuged, and plasma was separated and diluted 1:3 in PBS and stored at −80°C prior to chemokine and cytokine analysis.

A comparative analysis evaluating cytokine levels at each time point for each mouse strain was conducted using the nonparametric Kruskal–Wallis test. The Mann–Whitney U test was then used for pairwise comparison of each time point with uninfected mice and day 7 postinfection, which represented the onset of significant clearance of parasites. Pearson correlations were also estimated to assess any linear relationship between the cytokine and the parasite concentrations. Because the correlation tests are for linear association and because the parasite growth is nonlinear (it increases until day 7 and then decreases), the data were divided into two sets, one containing days 0 and 7 data, during which time parasite growth is increasing, and another set containing days 7 and 15 data, during which time parasite growth is decreasing. Significance was defined at α ≤ 0.05. All analyses were done using IBM-SPSS version 23.0. Statistical analysis of parasitemia, parasite morphology, and the number of reticulocytes was performed using a nonpaired t test comparing two conditions with one another using GraphPad Prism.

We evaluated the susceptibility to B. microti in three commonly used mouse strains, C3H/HeJ, BALB/cJ, and C57BL/6J, comparing peak parasitemia and the rate of parasite clearance. The course of infection was similar in all of the mouse strains. B. microti achieved high levels of parasitemia before it was cleared (Fig. 1). Parasitemia peaked in each of the three strains ∼7 d postinfection. The level of peak parasitemia was highest in C3H/HeJ mice (60%), followed by BALB/cJ mice (50%) and C57BL/6J mice (40%). Despite the increased parasite burden, C3H/HeJ mice cleared the infection at the same rate as BALB/cJ and C57BL/6J mice (day 13, Fig. 1). B. microti levels were monitored until day 30 postinfection, but no recrudescence of parasitemia was observed in any of the three mouse strains. Surprisingly, during acute infection, all three mice strains tolerated high B. microti levels well and only exhibited ruffled fur for a 24-h period on day 7 postinfection, at which time they also showed evidence of hemoglobinuria, which is indicative of erythrocyte hemolysis (4, 22).

FIGURE 1.

Parasitemia levels in C3H/HeJ, BALB/cJ, and C57BL/6J mice during babesiosis. Babesia microti reached high parasitemia levels, followed by resolution of infection in C3H/HeJ, BALB/cJ, and C57BL/6J mice. The mean and SE from five counts of 100 erythrocytes per mouse obtained from Giemsa-stained blood smears are shown. n = 5 mice per group. Experiment shown is representative of two independent experiments. Day 7 post-infection parasite numbers are significantly higher in C3H mice compared to BALB/c or C57BL/6 mice, p < 0.01.

FIGURE 1.

Parasitemia levels in C3H/HeJ, BALB/cJ, and C57BL/6J mice during babesiosis. Babesia microti reached high parasitemia levels, followed by resolution of infection in C3H/HeJ, BALB/cJ, and C57BL/6J mice. The mean and SE from five counts of 100 erythrocytes per mouse obtained from Giemsa-stained blood smears are shown. n = 5 mice per group. Experiment shown is representative of two independent experiments. Day 7 post-infection parasite numbers are significantly higher in C3H mice compared to BALB/c or C57BL/6 mice, p < 0.01.

Close modal

Parasites and erythrocytes were examined by histopathology of blood smears over the course of infection for each of the mouse strains. Parasite burden increased steadily for the first 7 d of infection. During this time, intracellular parasites appeared largely as rings within erythrocytes, with some erythrocytes containing multiple parasites (Fig. 2) (<day 7 postinfection). In contrast, by day 7 postinfection, >75% of intraerythrocytic parasites appeared degenerate within intact erythrocytes; this phenotype is historically referred to as “crisis forms” that have been shown to be characteristic of intraerythrocytic death of parasites (Fig. 2) (2326). Consistent with the predominance of crisis-form parasites within erythrocytes, the overall percentage of erythrocytes containing detectable parasites decreased dramatically by days 9 and 10 postinfection (Fig. 1). Infection was resolved by day 13 postinfection, with no subsequent waves of parasitemia (monitored for 30 d postinfection; data not shown).

FIGURE 2.

Morphology of parasites and RBCs throughout infection. Reticulocytosis and parasite crisis forms inside infected erythrocytes correlates with parasite clearance. (A) Giemsa-stained blood smears from BALB/cJ mice during B. microti infection during parasite replication (days 3 and 5), onset of parasite clearance (day 7), and resolution of infection (days 10 and 11). (B) Morphology of parasites in infected RBC (iRBCs) on day 5 compared with day 7 postinfection. The morphology of intraerythrocyte parasites changes from ring forms on day 5 postinfection to crisis forms by day 7 postinfection. (C) Percentage of infected erythrocytes containing crisis forms of the parasite on day 5 or 7 postinfection of C3H/HeJ, BALB/cJ, and C57BL/6J mice. The mean and SE from four counts of 100 infected erythrocytes per mouse obtained from Giemsa-stained blood smears are shown. n = 5 mice per group. Scale bars, 5 μm.

FIGURE 2.

Morphology of parasites and RBCs throughout infection. Reticulocytosis and parasite crisis forms inside infected erythrocytes correlates with parasite clearance. (A) Giemsa-stained blood smears from BALB/cJ mice during B. microti infection during parasite replication (days 3 and 5), onset of parasite clearance (day 7), and resolution of infection (days 10 and 11). (B) Morphology of parasites in infected RBC (iRBCs) on day 5 compared with day 7 postinfection. The morphology of intraerythrocyte parasites changes from ring forms on day 5 postinfection to crisis forms by day 7 postinfection. (C) Percentage of infected erythrocytes containing crisis forms of the parasite on day 5 or 7 postinfection of C3H/HeJ, BALB/cJ, and C57BL/6J mice. The mean and SE from four counts of 100 infected erythrocytes per mouse obtained from Giemsa-stained blood smears are shown. n = 5 mice per group. Scale bars, 5 μm.

Close modal

Polychromatophilic red cells that stained blue-gray via Giemsa stain due to residual RNA in their cytoplasm accounted for 50% of total RBCs by day 7 postinfection (Figs. 2, 3A). We confirmed that these polychromatophilic red cells were reticulocytes using New Methylene Blue, a cationic vital stain that precipitates rRNA in reticulocytes (27, 28). Intracellular parasites were clearly evident in mature erythrocytes and reticulocytes when stained with this dye (arrowheads and arrows, respectively, Fig. 3D). B. microti displayed a clear tropism for mature erythrocytes, because >70% of these cells were infected by day 7 postinfection, whereas <10% of reticulocytes contained intracellular parasites at any point of infection, even when reticulocytes dominated the bloodstream during the later stages of infection (Fig. 3B, 3C). The tropism of Babesia for mature erythrocytes could be a consequence of preferential parasite invasion of these cells or enhanced parasite survival within these cells compared with reticulocytes. Thus, we determined whether reticulocytes and mature erythrocytes differed in their ability to support intracellular survival of Babesia following parasite invasion. As shown in Fig. 4, parasites that invaded reticulocytes were more likely to be killed (crisis forms) than were those that invaded mature erythrocytes at all time points postinfection. On day 6 postinfection, when reticulocytes made up ∼25% of RBCs, 60% of infected reticulocytes contained crisis-form parasites, whereas the majority of infected mature erythrocytes contained parasites in ring form, and <20% had crisis forms. As infection progressed (after day 6 postinfection), crisis-form parasites predominated in infected mature erythrocytes (60% crisis forms), as well as infected reticulocytes (80% crisis forms).

FIGURE 3.

Mature erythrocytes are replaced with reticulocytes that appear more refractory to parasite growth/replication. (A) Reticulocytes replace mature erythrocytes in the peripheral blood ∼7 d postinfection in all three mouse strains. (B) The percentage of infected mature erythrocytes during infection. (C) The percentage of infected reticulocytes during infection. (D) New Methylene Blue stain of erythrocytes from uninfected mice (upper panel) compared with erythrocytes from B. microti–infected mice on day 7 postinfection (lower panel). Reticulocytes are evident from their deposition of a dark precipitate throughout the RBC, as well as increased size. Arrowheads mark mature erythrocytes containing intracellular parasites, and arrows mark reticulocytes. The mean and SE from four counts of 100 erythrocytes per mouse per time point are shown for a single experiment. However, similar results were obtained in two additional independent experiments using Giemsa stain instead of methylene blue (original magnification ×100, NA 1.3).

FIGURE 3.

Mature erythrocytes are replaced with reticulocytes that appear more refractory to parasite growth/replication. (A) Reticulocytes replace mature erythrocytes in the peripheral blood ∼7 d postinfection in all three mouse strains. (B) The percentage of infected mature erythrocytes during infection. (C) The percentage of infected reticulocytes during infection. (D) New Methylene Blue stain of erythrocytes from uninfected mice (upper panel) compared with erythrocytes from B. microti–infected mice on day 7 postinfection (lower panel). Reticulocytes are evident from their deposition of a dark precipitate throughout the RBC, as well as increased size. Arrowheads mark mature erythrocytes containing intracellular parasites, and arrows mark reticulocytes. The mean and SE from four counts of 100 erythrocytes per mouse per time point are shown for a single experiment. However, similar results were obtained in two additional independent experiments using Giemsa stain instead of methylene blue (original magnification ×100, NA 1.3).

Close modal
FIGURE 4.

Crisis-form parasites in infected mature erythrocytes versus infected reticulocytes during infection in C57BL/6J (A), BALB/cJ (B), and C3H/HeJ (C) mice. Infected reticulocytes, unlike mature RBCs, predominantly contain crisis-form parasites throughout infection in all three mice strains. The percentages of infected mature erythrocytes and infected reticulocytes that contain crisis forms of the parasite increase after day 6 postinfection (p < 0.05). Cells were stained with Giemsa to detect intracellular parasites and to distinguish mature erythrocytes from reticulocytes. The mean and SE from four counts of 100 infected mature erythrocytes or >25 infected reticulocytes per mouse per time point are shown for a single experiment. n = 5 mice for each strain. The experiment was performed once.

FIGURE 4.

Crisis-form parasites in infected mature erythrocytes versus infected reticulocytes during infection in C57BL/6J (A), BALB/cJ (B), and C3H/HeJ (C) mice. Infected reticulocytes, unlike mature RBCs, predominantly contain crisis-form parasites throughout infection in all three mice strains. The percentages of infected mature erythrocytes and infected reticulocytes that contain crisis forms of the parasite increase after day 6 postinfection (p < 0.05). Cells were stained with Giemsa to detect intracellular parasites and to distinguish mature erythrocytes from reticulocytes. The mean and SE from four counts of 100 infected mature erythrocytes or >25 infected reticulocytes per mouse per time point are shown for a single experiment. n = 5 mice for each strain. The experiment was performed once.

Close modal

Parasitemia peaked on day 7 postinfection and then steadily decreased (Fig. 1); this is a time course consistent with the onset of an early adaptive immune response. To clarify a role for lymphocytes in the intraerythrocytic death of parasites, we evaluated B. microti parasitemia in CD4 T cell–deficient (CD4−/−) mice and in μMt mice lacking mature B cells. We assessed the presence of singlets (single parasites), multiples (more than one parasite), and crisis forms (dying parasites) in mature erythrocytes and reticulocytes in CD4−/− and μMt mice and compared the results with wild-type (WT) C57BL/6J mice. Peak B. microti levels were not different in CD4−/− or μMt mice compared with WT C57BL/6J mice (Fig. 5). Intraerythrocytic death of parasites was also associated with a sharp decline in parasitemia in CD4−/− and μMt mice. However, CD4−/− mice were unable to completely eliminate parasites and maintained a persistent parasitemia for >1 mo (Fig. 5B). The persistent stage of parasitemia in CD4−/− mice was associated with a marked alteration in the steady-state population structure of Babesia, with ∼30% of infected RBCs containing singlets, 30% containing multiples, and 30% containing crisis-form parasites. In contrast, WT C57BL/6J mice had very few multiples and cleared the parasites by day 20 (Fig. 5A). μMt mice maintained a low level of parasitemia (<5%) out to 1 mo postinfection, with very few multiples within infected erythrocytes (Fig. 5C). Thus, mature B cells or CD4 cells are not essential for the death of parasites within infected erythrocytes. However, CD4 cells and, to a lesser extent, B cells are important to enable clearance of parasites to prevent their prolonged carriage.

FIGURE 5.

Effect of CD4 cell deletion (CD4−/− mice) and B cell deletion (μMt−/− mice) on parasitemia. The percentage of RBCs containing single parasites, multiple parasites, or crisis forms in WT C57BL/6J mice (A), CD4−/− mice (B), and μMt−/− mice (C). CD4−/− and μMt−/− mice control the first wave of parasitemia and do not prevent parasite death within infected erythrocytes. However, CD4−/− mice are unable to clear parasites, resulting in prolonged parasite carriage. Parasite persistence in CD4−/− mice is characterized by ∼30% parasitemia consisting of nearly equal populations of RBCs containing single parasites, multiple parasites, and crisis-form parasites. The mean and SE from four counts of 100 erythrocytes per mouse per time point are shown for a single experiment. n = 5 mice of each strain. The experiment was performed once. All gene deletions are on the C77BL/6J mouse background. CD4−/−, B6.129S2-Cd4tm1Mak/J; μMt−/−, B6.129S2-Ighmtm1Cgn/J.

FIGURE 5.

Effect of CD4 cell deletion (CD4−/− mice) and B cell deletion (μMt−/− mice) on parasitemia. The percentage of RBCs containing single parasites, multiple parasites, or crisis forms in WT C57BL/6J mice (A), CD4−/− mice (B), and μMt−/− mice (C). CD4−/− and μMt−/− mice control the first wave of parasitemia and do not prevent parasite death within infected erythrocytes. However, CD4−/− mice are unable to clear parasites, resulting in prolonged parasite carriage. Parasite persistence in CD4−/− mice is characterized by ∼30% parasitemia consisting of nearly equal populations of RBCs containing single parasites, multiple parasites, and crisis-form parasites. The mean and SE from four counts of 100 erythrocytes per mouse per time point are shown for a single experiment. n = 5 mice of each strain. The experiment was performed once. All gene deletions are on the C77BL/6J mouse background. CD4−/−, B6.129S2-Cd4tm1Mak/J; μMt−/−, B6.129S2-Ighmtm1Cgn/J.

Close modal

Given the differential ability of mature erythrocytes versus reticulocytes to support survival of intracellular parasites, we assessed whether the absence of CD4 cells or mature B cells altered peripheral reticulocyte numbers during infection. The percentage of peripheral blood cells made up of reticulocytes was similar in WT, CD4−/−, and μMt mice out to day 10 postinfection. Thereafter, the percentage of reticulocytes decreased in WT mice but remained elevated in CD4−/− and μMt mice (Supplemental Fig. 1). This difference in peripheral reticulocytes was clearly evident by 18 d postinfection and suggests that, although μMt mice had largely cleared peripheral parasites by this time point, there was an ongoing need for replenishment of peripheral blood cells or a delay in erythrocyte maturation.

Soluble mediators, including cytokines, have been hypothesized to contribute to B. microti death within infected erythrocytes during infection (2326). We hypothesized that production of inflammatory cytokines analogous to a “cytokine storm” may contribute to death of parasites within infected erythrocytes. To test this hypothesis, we evaluated plasma cytokine and chemokine levels on days 7, 11, and 15 postinfection. These time points were chosen to correlate with the emergence of crisis forms (day 7), decreasing parasitemia (day 11), and parasite clearance (day 15). Cytokine and chemokine levels were evaluated in all three strains of mice to ensure that any observed changes were not specific to a single genetic background. Results are shown in Table I. Although B. microti parasitemia reached 30–60% on day 7 postinfection, depending on the strain of mouse, a robust systemic inflammatory cytokine response was not evident. However, trends in plasma cytokines/chemokines were noted. Inflammatory cytokines TNF-α, IL-6, and IL-1β were elevated in C57BL/6J mice at day 7 but not in BALB/cJ or C3H/HeJ mice. IL-10, IL-12p40 (but not IL-12p70), and G-CSF were elevated at peak parasitemia (day 7) and decreased as the parasites were cleared (days 11 and 15) in all three mice strains. The levels of chemokines CCL2 (MCP-1), CCL4 (MIP-1β), and CCL5 (RANTES) were also elevated in all three mice strains on day 7 and declined by day 15. In C57BL/6J and C3H/HeJ mice, CCL3 (MIP-1α) levels showed similar kinetics. Plasma eotaxin was sharply decreased in infected mice at all time points and in all mouse strains. Table II shows linear trends between cytokines and chemokines in relation to parasite numbers. Linear trends were examined between days 0 and 7 postinfection when parasite numbers were increasing and between days 7 and 15 when parasite numbers were decreasing. IL-10, MCP-1, MIP-1α, MIP-1β, and RANTES all showed a positive correlation with parasite numbers. Eotaxin and IL-12 (p70) were negatively correlated with parasite number.

Table I.
Plasma cytokine and chemokine levels (pg/ml) during B. microti infection at various days postinfection
C57BL/6 Mice
C3H Mice
BALB/c Mice
Day 0Day 7Day 11Day 15p ValueDay 0Day 7Day 11Day 15p ValueDay 0Day 7Day 11Day 15p Value
IL-1α 11.62 14.59 47.11 82.51a 0.022* 5.00 0.00 55.9b 65.63 0.010* 0.00 0.00 17.53 18.97 0.057 
IL-1β 208.80 562.19 103.69 92.44b 0.016* 371.43 308.19 169.53 179.63 0.136 185.17 225.95 88.28 261.44c 0.025* 
IL-2 0.00 0.00 0.00 0.00 NP 0.00 0.00 46.08 56.25 0.333 0.00 10.69 0.00 0.00 NP 
IL-3 11.44 13.42 0.00 9.65 0.357 13.13 0.00 0.00 0.00 NP 0.00 0.00 0.00 29.11 NP 
IL-6 27.30 34.46 24.73 30.86 0.729 19.34 10.65 3.41 51.08 0.091 22.55 10.49 10.89 67.94 0.061 
IL-10 81.39 344.10a 229.39 116.61 0.008* 73.52 394.17 95.45 51.5 0.128 45.22 237.68 45.42b 71.52 0.016* 
IL-12(p40) 458.63 973.57 840.36 848.09 0.200 605.92 786.76 429.98 428.9 0.103 721.13 1246.4 326.2b 471.05 0.012* 
IL-12(p70) 440.68 219.22 41.22a 64.19 0.007* 421.74 181.74 93.34 49.86a 0.017* 232.45 183.98 46.78 84.6 0.032* 
IL-13 718.65 550.63 238.18a 298.83 0.010* 641.79 485.49 451.66 357.98 0.200 428.05 462.94 243.18 365.16 0.172 
IL-17 0.00 77.91 0.00 0.00 NP 40.35 9.57 10.04 0.00 0.062 0.00 31.27 11.36 91.12 0.175 
Eotaxin 1520.85 146.33 0.00a 716.33 0.044* 1535.94 416.41 346.92 651.65 0.205 1989.95 246.02 0.00a 0.00a 0.009* 
G-CSF 68.64 107.84 14.37b 36.62 0.038* 62.74 108.60 14.37 42.76 0.061 18.72 72.17 0.00 121.85 0.082 
GM-CSF 79.65 19.86 0.00 16.74 0.207 22.40 0.00 0.00 0.00 NP 74.19 49.81 0.00a 0.00a 0.014* 
IFN-y 15.82 14.47 0.00 0.00 0.052 32.21 17.57 36.85 45.59 0.461 15.23 13.11 0.00 0.00 0.127 
KC 28.58 123.98 82.69 13.15 0.060 36.91 73.08 45.89 64.01 0.0936 8.56 91.01 11.73 21.53 0.042 
MCP-1 177.09 949.40 501.41 79.23b 0.007* 244.67 388.55 154.99b 141.84b 0.018* 172.36 718.78 84.71b 96.32b 0.011* 
MIP-1α 24.00 48.00a 31.29 33.63 0.029* 25.67 34.34 5.98 19.38 0.115 8.48 41.90 0.00b 21.78 0.015* 
MIP-1β 16.68 239.81a 106.33 52.51 0.004* 41.65 229.66 45.06 61.41 0.419 18.34 381.08a 42.96 95.51 0.007* 
RANTES 44.63 119.24 135.26a 83.46 0.015* 29.71 64.34 46.95 42.04 0.112 28.45 80.47a 55.85 91.56 0.021* 
TNF-α 312.24 624.42 161.81b 165.41 0.011* 415.45 312.24 204.38 207.1 0.062 311.54 327.52 291.41 881.97 0.135 
Parasited 0.00 38.21a 7.70 4.83 0.008* 0.00 60.78a 4.12 9.25 0.009* 0.00 43.13a 4.32 7.92 0.008* 
C57BL/6 Mice
C3H Mice
BALB/c Mice
Day 0Day 7Day 11Day 15p ValueDay 0Day 7Day 11Day 15p ValueDay 0Day 7Day 11Day 15p Value
IL-1α 11.62 14.59 47.11 82.51a 0.022* 5.00 0.00 55.9b 65.63 0.010* 0.00 0.00 17.53 18.97 0.057 
IL-1β 208.80 562.19 103.69 92.44b 0.016* 371.43 308.19 169.53 179.63 0.136 185.17 225.95 88.28 261.44c 0.025* 
IL-2 0.00 0.00 0.00 0.00 NP 0.00 0.00 46.08 56.25 0.333 0.00 10.69 0.00 0.00 NP 
IL-3 11.44 13.42 0.00 9.65 0.357 13.13 0.00 0.00 0.00 NP 0.00 0.00 0.00 29.11 NP 
IL-6 27.30 34.46 24.73 30.86 0.729 19.34 10.65 3.41 51.08 0.091 22.55 10.49 10.89 67.94 0.061 
IL-10 81.39 344.10a 229.39 116.61 0.008* 73.52 394.17 95.45 51.5 0.128 45.22 237.68 45.42b 71.52 0.016* 
IL-12(p40) 458.63 973.57 840.36 848.09 0.200 605.92 786.76 429.98 428.9 0.103 721.13 1246.4 326.2b 471.05 0.012* 
IL-12(p70) 440.68 219.22 41.22a 64.19 0.007* 421.74 181.74 93.34 49.86a 0.017* 232.45 183.98 46.78 84.6 0.032* 
IL-13 718.65 550.63 238.18a 298.83 0.010* 641.79 485.49 451.66 357.98 0.200 428.05 462.94 243.18 365.16 0.172 
IL-17 0.00 77.91 0.00 0.00 NP 40.35 9.57 10.04 0.00 0.062 0.00 31.27 11.36 91.12 0.175 
Eotaxin 1520.85 146.33 0.00a 716.33 0.044* 1535.94 416.41 346.92 651.65 0.205 1989.95 246.02 0.00a 0.00a 0.009* 
G-CSF 68.64 107.84 14.37b 36.62 0.038* 62.74 108.60 14.37 42.76 0.061 18.72 72.17 0.00 121.85 0.082 
GM-CSF 79.65 19.86 0.00 16.74 0.207 22.40 0.00 0.00 0.00 NP 74.19 49.81 0.00a 0.00a 0.014* 
IFN-y 15.82 14.47 0.00 0.00 0.052 32.21 17.57 36.85 45.59 0.461 15.23 13.11 0.00 0.00 0.127 
KC 28.58 123.98 82.69 13.15 0.060 36.91 73.08 45.89 64.01 0.0936 8.56 91.01 11.73 21.53 0.042 
MCP-1 177.09 949.40 501.41 79.23b 0.007* 244.67 388.55 154.99b 141.84b 0.018* 172.36 718.78 84.71b 96.32b 0.011* 
MIP-1α 24.00 48.00a 31.29 33.63 0.029* 25.67 34.34 5.98 19.38 0.115 8.48 41.90 0.00b 21.78 0.015* 
MIP-1β 16.68 239.81a 106.33 52.51 0.004* 41.65 229.66 45.06 61.41 0.419 18.34 381.08a 42.96 95.51 0.007* 
RANTES 44.63 119.24 135.26a 83.46 0.015* 29.71 64.34 46.95 42.04 0.112 28.45 80.47a 55.85 91.56 0.021* 
TNF-α 312.24 624.42 161.81b 165.41 0.011* 415.45 312.24 204.38 207.1 0.062 311.54 327.52 291.41 881.97 0.135 
Parasited 0.00 38.21a 7.70 4.83 0.008* 0.00 60.78a 4.12 9.25 0.009* 0.00 43.13a 4.32 7.92 0.008* 

Mean cytokine values from three or four mice are shown. The p values were calculated using the Kruskal–Wallis test. Statistical tests were not done when all or most of the values were below the level of detection (considered as zero for statistical analysis).

a

Versus day 0.

b

Versus day 7.

c

Versus day 11.

d

Mean parasite number from three or four mice.

*

α = 0.05.

NP, not performed.

Table II.
Pearson correlation test for linear trend between cytokines/chemokines and parasite numbers
C57BL/6
C3H
BALB/c
Parasite Correlation Days 0–7Parasite Correlation Days 7–15Parasite Correlation Days 0–7Parasite Correlation Days 7–15Parasite Correlation Days 0–7Parasite Correlation Days 7–15
MCP-1 0.763* 0.807* 0.56 0.70 0.90** 0.90** 
MIP-1a 0.871* 0.594 0.62 0.27 0.82* 0.62 
MIP-1b 0.905** 0.883** 0.70 0.62 0.96** 0.90** 
RANTES 0.971** 0.804* 0.80* 0.54 0.90** −0.30 
Eotaxin −0.935** −0.512 −0.72 0.42 0.96** 0.23 
IL-10 0.839* 0.815* 0.58 0.60 0.95** 0.87** 
IL-12(p70) −0.898** 0.869** −0.93** 0.83* −0.49 0.47 
C57BL/6
C3H
BALB/c
Parasite Correlation Days 0–7Parasite Correlation Days 7–15Parasite Correlation Days 0–7Parasite Correlation Days 7–15Parasite Correlation Days 0–7Parasite Correlation Days 7–15
MCP-1 0.763* 0.807* 0.56 0.70 0.90** 0.90** 
MIP-1a 0.871* 0.594 0.62 0.27 0.82* 0.62 
MIP-1b 0.905** 0.883** 0.70 0.62 0.96** 0.90** 
RANTES 0.971** 0.804* 0.80* 0.54 0.90** −0.30 
Eotaxin −0.935** −0.512 −0.72 0.42 0.96** 0.23 
IL-10 0.839* 0.815* 0.58 0.60 0.95** 0.87** 
IL-12(p70) −0.898** 0.869** −0.93** 0.83* −0.49 0.47 

In the first correlation test (days 0–7), levels for a particular cytokine were compared between day 0 and day 7 and tested for correlation with parasite growth between days 0 and 7; the results columns show the correlation coefficient. In the second correlation test (days 7–15), levels for a particular cytokine between days 7 and 15 were compared and analyzed for correlation with parasite growth between days 7 and 15.

*

α = 0.05, **α = 0.01.

To evaluate the contribution of TLR signaling to parasite clearance, we infected MyD88−/− mice with B. microti. MyD88 is required for the activation of the NF-κB signaling pathway by most TLRs. However, TLR3 and TLR4 can be activated independently of MyD88. The absence of MyD88 had no apparent effect on peak parasitemia, the rate of B. microti clearance, intraerythrocytic killing of parasites (Fig. 6A, 6B), or reticulocytosis (Supplemental Fig. 1). Thus, TLR signaling via MyD88 does not seem to be essential for control of B. microti infection. These results are also consistent with the lack of a robust proinflammatory serum cytokine response during infection. We also compared cytokine levels in WT and MyD88−/− mice on days 5 and 11 postinfection, corresponding to time points when parasite numbers are increasing and decreasing, respectively. As shown in Supplemental Table I, the absence of MyD88 did not strongly alter serum cytokine levels, with the exception of increased IL-1β on day 11 and increased IL12p70 on day 5, in the absence of MyD88.

FIGURE 6.

The effect of MyD88, IFN-γR, and iNOS gene deletion on parasitemia. The percentage of RBCs containing single parasites, multiple parasites, or crisis-form parasites in WT C57BL/6J mice (A), MyD88−/− mice (B), IFN-γR−/− mice (C), and iNOS−/− mice (D). MyD88−/− mice (MyD88 KO) and iNOS−/− mice controlled parasitemia as well as WT mice. IFN-γ−/− mice (IFN KO) had a higher peak parasitemia level but still controlled the primary wave of parasites. However, IFN-γ−/− mice were impaired for complete parasite clearance. The mean and SE from four counts of 100 erythrocytes per mouse from Giemsa-stained blood smears are shown. The experiment shown is representative of two experiments. n = 4 or 5 mice per group per experiment. All gene deletions are on the C57BL/6J mouse background. IFN-γR−/−, B6.129S7-Ifngr1tm1Agt/J; iNOS−/−, B6.129P2-Nos2tmaLau/J; MyD88−/−, B6.129P2 (SJL)-Myd88tm.1.1Defr/J.

FIGURE 6.

The effect of MyD88, IFN-γR, and iNOS gene deletion on parasitemia. The percentage of RBCs containing single parasites, multiple parasites, or crisis-form parasites in WT C57BL/6J mice (A), MyD88−/− mice (B), IFN-γR−/− mice (C), and iNOS−/− mice (D). MyD88−/− mice (MyD88 KO) and iNOS−/− mice controlled parasitemia as well as WT mice. IFN-γ−/− mice (IFN KO) had a higher peak parasitemia level but still controlled the primary wave of parasites. However, IFN-γ−/− mice were impaired for complete parasite clearance. The mean and SE from four counts of 100 erythrocytes per mouse from Giemsa-stained blood smears are shown. The experiment shown is representative of two experiments. n = 4 or 5 mice per group per experiment. All gene deletions are on the C57BL/6J mouse background. IFN-γR−/−, B6.129S7-Ifngr1tm1Agt/J; iNOS−/−, B6.129P2-Nos2tmaLau/J; MyD88−/−, B6.129P2 (SJL)-Myd88tm.1.1Defr/J.

Close modal

Because IFN-γ has been suggested to play a role in pathogen clearance (17, 18), we assessed the importance of IFN-γ to the control and resolution of B. microti infection. We infected IFN-γR−/− mice and assessed parasitemia and B. microti clearance times. The absence of IFN-γR resulted in an increase in peak parasitemia, but mice were still capable of intraerythrocytic killing of parasites and suppressing the initial wave of parasitemia (Fig. 6A, 6C). A low level of prolonged parasitemia occurred in IFN-γR−/− mice, but they were ultimately able to clear the infection, unlike CD4−/− mice (Figs. 5B, 6C). The absence of IFN-γR did not result in a marked changed in plasma cytokine levels compared with WT mice (Supplemental Table I) or the percentage of reticulocytes in the periphery (Supplemental Fig. 1). iNOS induction and the subsequent formation of NO radicals are a common mechanism of IFN-γ–induced pathogen clearance. To determine whether IFN-γ’s effects were mediated through iNOS, B. microti clearance was compared in WT, IFN-γ−/−, and iNOS−/− mice. As shown in Fig. 6D, peak parasitemia and parasite clearance in iNOS−/− mice were similar to WT mice, indicating that the role of IFN-γ in the resolution of babesiosis is largely independent of its role in inducing NO production via iNOS.

Although B. microti is an emerging pathogen in the United States and an increasing risk for transfusion medicine, the immune response to B. microti is not well defined. To assess the course and immune response to B. microti infection, we used three strains of inbred mice, C57BL/6J, BALB/cJ, and C3H/HeJ, each with a distinct, well-characterized immune response pattern. C57BL/6J mice have strong inflammatory responses, BALB/cJ mice are Th2 biased, and C3H/HeJ mice lack Tlr4 and are susceptible to Gram-negative bacteria but resistant to septic shock. The peak level of parasitemia differed in each of the three strains, but the kinetics of parasite death and clearance were similar. Despite differences in peak parasitemia, parasite crisis forms signifying intraerythrocytic B. microti death, decreased parasitemia, and the development of reticulocytosis occurred at the same time point in all three mice strains. Throughout infection, B. microti conspicuously parasitized mature erythrocytes but not reticulocytes. Thus, reticulocytosis with the replacement of infected mature peripheral erythrocytes with immature erythrocytes is a potential mechanism for controlling B. microti infection.

In humans, the clinical manifestations of malaria and babesiosis are similar. Mild to moderate disease is characterized by fever, sweats, chills, headache, and myalgia. Cell-mediated immunity and IFN-γ have been shown to contribute to control of acute babesiosis, although it is not clear that they are essential to disease resolution (1619). Earlier studies showed that infection of mice with bacillus Calmette-Guérin provided protection against B. microti (King strain) and B. rodhaini (Antwerp strain). Parasites degenerated within circulating erythrocytes, leading the investigators to hypothesize that a soluble, nonspecific mediator was induced that killed intracellular parasites (24, 26). A subsequent comparison of self-limiting B. microti infection and fatal infection with the Babesia WA1 strain showed an association of IL-10 production with B. microti infection and TNF-α with WA1. TNF-α−/− mice were protected against fatal WA1 infection (29). In our model, we show that mice infected with B. microti did not generate a robust serum proinflammatory cytokine response but killed parasites within infected erythrocytes. Thus, intraerythrocytic death of parasites is unlikely to be a consequence of a systemic proinflammatory cytokine response. Instead, intraerythrocyte parasite death was associated with elevated levels of the anti-inflammatory cytokines IL-10 and G-CSF. Other studies of B. microti infection, like the current study, noted serum IL-10 production but also reported modest elevation of serum proinflammatory cytokines, including IFN-γ (30, 31), which we did not observe. IL-12p40 was also elevated; this cytokine can form homodimers that may play an antagonistic role in the development of IL-12p70 responses, although this remains unclear. Similar levels of bacteremia or viremia are typically associated with very strong cytokine responses or cytokine storm. Based on the results of this study and available literature, we hypothesize that the ability to avoid a potentially pathological systemic proinflammatory response in the presence of high parasitemia may be important for survival of the host.

B. microti clearance was not dependent on TLR signaling (via MyD88). Because MyD88 is also the signaling adaptor protein for IL-1R family members, this also suggests that IL-1R and IL-18R signaling is not essential for control of acute infection or B. microti clearance. That MyD88 is not required raises the question of how the innate immune system recognizes B. microti. TLR9 contributes to the generation of an effective cytokine- and cell-mediated immune response to Plasmodium through its recognition of hemozoin (32). However, unlike Plasmodium, Babesia does not produce hemozoin; thus, the role of TLRs during B. microti infection is unknown. We did not observe significant changes in plasma IFN-γ levels during the course of infection. Yet the findings in IFN-γR−/− mice, higher parasitemia and delayed B. microti clearance, indicate that IFN-γ does play a role in this infection. Thus, the fact that a specific cytokine is not increased systemically in babesiosis does not rule out its possible role in other tissues during infection (33).

Crisis-form parasites were evident by 6 d postinfection within infected, but morphologically intact, erythrocytes. Intraerythrocytic crisis forms of B. microti were previously identified by Clark et al. during resolution of babesiosis (23). That study confirmed that crisis-form parasites were noninfective, and electron microscopy confirmed that the abnormal forms were degenerating intraerythrocytic parasites (23, 25). It is unlikely that direct killing by PMNs, macrophages, or cytolytic lymphocytes is involved, although it has been hypothesized that they may deliver reactive oxygen products or other toxic mediators to infected erythrocytes, resulting in parasite death inside the RBC (23, 25). Our results show that parasite death occurs within infected erythrocytes in CD4−/− and μMt−/− mice, although CD4−/− mice were unable to fully eliminate the parasites. This suggests that parasite death inside infected erythrocytes is not due to Abs against the parasite. However, a role for B cells and Abs cannot be entirely ruled out, because μMt−/− mice retain B1 and perhaps other B cell subpopulations (34). We also show that NO production via iNOS is not critical for parasite clearance. However, although iNOS is the primary inducer of NO production during immune responses, NO from other sources, such as endothelial NO synthase, may be important in NO production during babesiosis. A possibility that we have yet to explore is the potential role of acute-phase proteins and platelets induced during babesiosis in the intraerythrocytic death of B. microti. Using Plasmodium infection models of uncomplicated malaria, it was discovered that platelet-deficient mice had increased parasitemia, and platelets were shown to kill Plasmodium in vitro in infected erythrocytes (3537). However, the role of platelets in the clearance of blood-stage Plasmodium remains controversial (38).

Parasite tropism for specific maturation states of erythrocytes is well established in malarial disease. If the preferred erythrocyte population for a Plasmodium species is in limited supply, parasitemia is curtailed (39). Our study strongly indicates that alterations in the erythrocyte population during B. microti infection may play a relatively unexplored role in disease resolution, as well as recrudescence. Our findings clearly show that, in early infection, nearly 100% of the peripheral RBCs are mature erythrocytes that appear permissive for parasite replication. In contrast, erythrocyte death results in replacement of the mature parasite-permissive erythrocytes with less mature reticulocytes, and this repopulation correlates with the appearance of intraerythrocyte crisis forms and parasite clearance. This is consistent with a previous study that showed that B. microti primarily invades mature erythrocytes in mice, and reticulocytes are rarely infected (40). That study concluded that the intensity of reticulocytosis varied with the degree of parasitemia, and that the timing of reticulocytosis appeared to be delayed in susceptible (DBA/2 C.B.10.scid) versus resistant (C.B-17, BALB/cBy, B10.D2.) strains of mice. Although that study concluded that parasites largely failed to invade reticulocytes, our study suggests that parasites invade reticulocytes, because reticulocytes contained degraded parasites in stained blood smears. Compared with mature RBCs, parasites that invaded reticulocytes were more often actively killed or their growth was not supported. It is unclear why reticulocytes are less favorable for B. microti survival compared with mature erythrocytes. We favor the hypothesis that reticulocytes have a greater arsenal of metabolites, proteins, and noncoding RNAs that may be important for constitutive and induced resistance to B. microti, although this has yet to be defined. Thus, reticulocytes may play a critical role in the clearance of the parasites, particularly in cases of high parasitemia, by being less supportive of parasite replication than mature erythrocytes.

Although primarily focused on acute babesiosis, our findings have implications for persistent babesiosis (6, 7, 41). CD4−/− mice were unable to completely eliminate parasites, and a prolonged low level of persistent parasitemia was evident even at 1 mo postinfection. The fact that IFN-γ−/− mice also had a delay in parasite clearance suggests that CD4 T cell production of IFN-γ contributes to, but is not required for, the resolution of B. microti infection. Our results indicate a potential role for B cells in resolution of B. microti infection, because μMt−/− mice had a delay in their ability to completely clear parasites. Parasitemia remained at ∼5% for a month postinfection compared with 30% in CD4−/− mice. This is consistent with the observation that rituximab treatment that targets CD20+ B cells is associated with an increased risk for persistent babesiosis in humans (6, 42). Persistent parasitemia in CD4−/− mice was associated with a change in the population structure of B. microti, with ∼30% of infected erythrocytes containing more than one parasite. For B. divergens, in synchronized erythrocyte culture, multiple parasites within infected RBCs is a consequence of extended cycles of replication prior to egress, rather than independent invasion events (43). Such a dynamic population structure may allow the parasite to adapt to changing environmental conditions, allowing it to choose when to egress and invade other cells or to delay egress and stay sequestered within its current host cell and continue multiplication. Our hypothesis is that the absence of CD4 cells enables the parasite to maintain a dynamic reservoir of infected RBCs, each sequestering multiple parasites that lyse and serve to continuously reseed blood, maintaining relatively stable numbers of infected RBCs containing one parasite, multiple parasites, or degenerating parasites (crisis forms).

Although studies have demonstrated an important role for T cells in controlling parasitemia, studies in SCID mice have suggested that innate immunity may be of primary importance (16, 17, 4446).

Our results, along with available literature, are consistent with a model in which innate immunity is progressively activated during infection to kill parasites within infected erythrocytes. Parasite death within infected erythrocytes may be a consequence of intrinsic changes in the erythrocyte population during infection and/or in factors that act on infected erythrocytes whose production or efficacy may be enhanced by CD4 T cell activation. The mechanism of intraerythrocyte killing has yet to be elucidated, but it is not dependent on CD4 cells, B cells, MyD88, IFN-γ, or iNOS. However, engagement of the adaptive immune response, specifically CD4 T cells, potentiates intraerythrocytic killing of parasites and is required for complete clearance of parasites. In the absence of CD4 cells, and to a lesser extent IFN-γ and B cells, complete resolution of parasitemia is prevented or delayed, resulting in a low-grade persistent parasitemia. Known risk factors for severe babesiosis, including lack of a spleen, a compromised immune system, or age > 50 y are all consistent with deficiencies in elements of the adaptive immune system or clearance of damaged RBCs in the spleen. The current study highlights the importance of investigating the effect of the adaptive immune response in these high-risk individuals at the level of the biology of the infected RBC and the parasite.

We thank the Biostatistics, Epidemiology, and Biomathematics Research Core at Weill Cornell Medicine-Qatar for services provided.

This work was supported by National Institutes of Health Grant 5R41AI108006 (to D.G.M. and R.J.D.; multiple principal investigator grant).

The online version of this article contains supplemental material.

Abbreviations used in this article:

iNOS

inducible NO synthase

WT

wild-type.

1
Lobo
,
C. A.
,
J. R.
Cursino-Santos
,
A.
Alhassan
,
M.
Rodrigues
.
2013
.
Babesia: an emerging infectious threat in transfusion medicine.
PLoS Pathog.
9
:
e1003387
.
2
U.S.Food and Drug Administration
.
2014
.
Fatalities reported to FDA following blood collection and transfusion
. .
3
Conrad
,
P. A.
,
A. M.
Kjemtrup
,
R. A.
Carreno
,
J.
Thomford
,
K.
Wainwright
,
M.
Eberhard
,
R.
Quick
,
S. R.
Telford
III.
,
B. L.
Herwaldt
.
2006
.
Description of Babesia duncani sp. (Apicomplexa: Babesiidae) from humans and its differentiation from other piroplasms.
Int. J. Parasitol.
36
:
779
789
.
4
Vannier
,
E.
,
P. J.
Krause
.
2012
.
Human babesiosis.
N. Engl. J. Med.
366
:
2397
2407
.
5
Rudzinska
,
M. A.
,
W.
Trager
,
S. J.
Lewengrub
,
E.
Gubert
.
1976
.
An electron microscopic study of Babesia microti invading erythrocytes.
Cell Tissue Res.
169
:
323
334
.
6
Krause
,
P. J.
,
B. E.
Gewurz
,
D.
Hill
,
F. M.
Marty
,
E.
Vannier
,
I. M.
Foppa
,
R. R.
Furman
,
E.
Neuhaus
,
G.
Skowron
,
S.
Gupta
, et al
.
2008
.
Persistent and relapsing babesiosis in immunocompromised patients.
Clin. Infect. Dis.
46
:
370
376
.
7
Krause
,
P. J.
,
A.
Spielman
,
S. R.
Telford
III.
,
V. K.
Sikand
,
K.
McKay
,
D.
Christianson
,
R. J.
Pollack
,
P.
Brassard
,
J.
Magera
,
R.
Ryan
,
D. H.
Persing
.
1998
.
Persistent parasitemia after acute babesiosis.
N. Engl. J. Med.
339
:
160
165
.
8
Moritz
,
E. D.
,
C. S.
Winton
,
L.
Tonnetti
,
R. L.
Townsend
,
V. P.
Berardi
,
M. E.
Hewins
,
K. E.
Weeks
,
R. Y.
Dodd
,
S. L.
Stramer
.
2016
.
Screening for Babesia microti in the U.S. blood supply.
N. Engl. J. Med.
375
:
2236
2245
.
9
Leiby
,
D. A.
2011
.
Transfusion-associated babesiosis: shouldn’t we be ticked off?
Ann. Intern. Med.
155
:
556
557
.
10
Leiby
,
D. A.
2011
.
Transfusion-transmitted Babesia spp.: bull’s-eye on Babesia microti.
Clin. Microbiol. Rev.
24
:
14
28
.
11
Cursino-Santos
,
J. R.
,
A.
Alhassan
,
M.
Singh
,
C. A.
Lobo
.
2014
.
Babesia: impact of cold storage on the survival and the viability of parasites in blood bags.
Transfusion
54
:
585
591
.
12
Gubernot
,
D. M.
,
C. T.
Lucey
,
K. C.
Lee
,
G. B.
Conley
,
L. G.
Holness
,
R. P.
Wise
.
2009
.
Babesia infection through blood transfusions: reports received by the US Food and Drug Administration, 1997-2007.
Clin. Infect. Dis.
48
:
25
30
.
13
Eberhard
,
M. L.
,
E. M.
Walker
,
F. J.
Steurer
.
1995
.
Survival and infectivity of Babesia in blood maintained at 25 C and 2-4 C.
J. Parasitol.
81
:
790
792
.
14
U.S. Food and Drug Administration
.
2015
.
Issue summary for FDA Blood Products Advisory Committee
. .
15
Gubernot
,
D. M.
,
H. L.
Nakhasi
,
P. A.
Mied
,
D. M.
Asher
,
J. S.
Epstein
,
S.
Kumar
.
2009
.
Transfusion-transmitted babesiosis in the United States: summary of a workshop.
Transfusion
49
:
2759
2771
.
16
Clark
,
I. A.
,
A. C.
Allison
.
1974
.
Babesia microti and Plasmodium berghei yoelii infections in nude mice.
Nature
252
:
328
329
.
17
Clawson
,
M. L.
,
N.
Paciorkowski
,
T. V.
Rajan
,
C.
La Vake
,
C.
Pope
,
M.
La Vake
,
S. K.
Wikel
,
P. J.
Krause
,
J. D.
Radolf
.
2002
.
Cellular immunity, but not gamma interferon, is essential for resolution of Babesia microti infection in BALB/c mice.
Infect. Immun.
70
:
5304
5306
.
18
Aguilar-Delfin
,
I.
,
P. J.
Wettstein
,
D. H.
Persing
.
2003
.
Resistance to acute babesiosis is associated with interleukin-12– and gamma interferon–mediated responses and requires macrophages and natural killer cells.
Infect. Immun.
71
:
2002
2008
.
19
Igarashi
,
I.
,
R.
Suzuki
,
S.
Waki
,
Y.
Tagawa
,
S.
Seng
,
S.
Tum
,
Y.
Omata
,
A.
Saito
,
H.
Nagasawa
,
Y.
Iwakura
, et al
.
1999
.
Roles of CD4(+) T cells and gamma interferon in protective immunity against Babesia microti infection in mice.
Infect. Immun.
67
:
4143
4148
.
20
Van Peenen
,
P. F.
,
G. R.
Healy
.
1970
.
Infection of Microtus ochrogaster with piroplasms isolated from man.
J. Parasitol.
56
:
1029
1031
.
21
Ruebush
,
M. J.
,
W. L.
Hanson
.
1979
.
Susceptibility of five strains of mice to Babesia microti of human origin.
J. Parasitol.
65
:
430
433
.
22
Homer
,
M. J.
,
I.
Aguilar-Delfin
,
S. R.
Telford
III.
,
P. J.
Krause
,
D. H.
Persing
.
2000
.
Babesiosis.
Clin. Microbiol. Rev.
13
:
451
469
.
23
Clark
,
I. A.
,
J. E.
Richmond
,
E. J.
Wills
,
A. C.
Allison
.
1977
.
Intra-erythrocytic death of the parasite in mice recovering from infection with Babesia microti.
Parasitology
75
:
189
196
.
24
Clark
,
I. A.
,
E. J.
Wills
,
J. E.
Richmond
,
A. C.
Allison
.
1977
.
Suppression of babesiosis in BCG-infected mice and its correlation with tumor inhibition.
Infect. Immun.
17
:
430
438
.
25
Clark
,
I. A.
,
J. E.
Richmond
,
E. J.
Wills
,
A. C.
Allison
.
1975
.
Immunity to intra-erythrocytic protoza.
Lancet
2
:
1128
1129
.
26
Clark
,
I. A.
,
A. C.
Allison
,
F. E.
Cox
.
1976
.
Protection of mice against Babesia and Plasmodium with BCG.
Nature
259
:
309
311
.
27
Brecher
,
G.
1949
.
New methylene blue as a reticulocyte stain.
Am. J. Clin. Pathol.
19
:
895
.
28
Riley
,
R. S.
,
J. M.
Ben-Ezra
,
R.
Goel
,
A.
Tidwell
.
2001
.
Reticulocytes and reticulocyte enumeration.
J. Clin. Lab. Anal.
15
:
267
294
.
29
Hemmer
,
R. M.
,
D. A.
Ferrick
,
P. A.
Conrad
.
2000
.
Role of T cells and cytokines in fatal and resolving experimental babesiosis: protection in TNFRp55−/− mice infected with the human Babesia WA1 parasite.
J. Parasitol.
86
:
736
742
.
30
Jeong
,
Y. I.
,
S. H.
Hong
,
S. H.
Cho
,
W. J.
Lee
,
S. E.
Lee
.
2012
.
Induction of IL-10–producing CD1dhighCD5+ regulatory B cells following Babesia microti-infection.
PLoS One
7
:
e46553
.
31
Terkawi
,
M. A.
,
S.
Cao
,
M. S.
Herbas
,
M.
Nishimura
,
Y.
Li
,
P. F.
Moumouni
,
A. H.
Pyarokhil
,
D.
Kondoh
,
N.
Kitamura
,
Y.
Nishikawa
, et al
.
2015
.
Macrophages are the determinant of resistance to and outcome of nonlethal Babesia microti infection in mice.
Infect. Immun.
83
:
8
16
.
32
Gowda
,
N. M.
,
X.
Wu
,
D. C.
Gowda
.
2012
.
TLR9 and MyD88 are crucial for the development of protective immunity to malaria.
J. Immunol.
188
:
5073
5085
.
33
Jason
,
J.
,
L. K.
Archibald
,
O. C.
Nwanyanwu
,
M. G.
Byrd
,
P. N.
Kazembe
,
H.
Dobbie
,
W. R.
Jarvis
.
2001
.
Comparison of serum and cell-specific cytokines in humans.
Clin. Diagn. Lab. Immunol.
8
:
1097
1103
.
34
Ghosh
,
S.
,
S. A.
Hoselton
,
J. M.
Schuh
.
2012
.
μ-Chain-deficient mice possess B-1 cells and produce IgG and IgE, but not IgA, following systemic sensitization and inhalational challenge in a fungal asthma model.
J. Immunol.
189
:
1322
1329
.
35
McMorran
,
B. J.
,
V. M.
Marshall
,
C.
de Graaf
,
K. E.
Drysdale
,
M.
Shabbar
,
G. K.
Smyth
,
J. E.
Corbin
,
W. S.
Alexander
,
S. J.
Foote
.
2009
.
Platelets kill intraerythrocytic malarial parasites and mediate survival to infection.
Science
323
:
797
800
.
36
Love
,
M. S.
,
M. G.
Millholland
,
S.
Mishra
,
S.
Kulkarni
,
K. B.
Freeman
,
W.
Pan
,
R. W.
Kavash
,
M. J.
Costanzo
,
H.
Jo
,
T. M.
Daly
, et al
.
2012
.
Platelet factor 4 activity against P. falciparum and its translation to nonpeptidic mimics as antimalarials.
Cell Host Microbe
12
:
815
823
.
37
McMorran
,
B. J.
,
L.
Wieczorski
,
K. E.
Drysdale
,
J. A.
Chan
,
H. M.
Huang
,
C.
Smith
,
C.
Mitiku
,
J. G.
Beeson
,
G.
Burgio
,
S. J.
Foote
.
2012
.
Platelet factor 4 and Duffy antigen required for platelet killing of Plasmodium falciparum.
Science
338
:
1348
1351
.
38
Gramaglia
,
I.
,
J.
Velez
,
V.
Combes
,
G. E.
Grau
,
M.
Wree
,
H. C.
van der Heyde
.
2017
.
Platelets activate a pathogenic response to blood-stage Plasmodium infection but not a protective immune response.
Blood
129
:
1669
1679
.
39
Okada
,
H.
,
K.
Suzue
,
T.
Imai
,
T.
Taniguchi
,
C.
Shimokawa
,
R.
Onishi
,
J.
Hirata
,
H.
Hisaeda
.
2015
.
A transient resistance to blood-stage malaria in interferon-gamma-deficient mice through impaired production of the host cells preferred by malaria parasites.
Front. Microbiol.
6
:
600
.
40
Borggraefe
,
I.
,
J.
Yuan
,
S. R.
Telford
III.
,
S.
Menon
,
R.
Hunter
,
S.
Shah
,
A.
Spielman
,
J. A.
Gelfand
,
H. H.
Wortis
,
E.
Vannier
.
2006
.
Babesia microti primarily invades mature erythrocytes in mice.
Infect. Immun.
74
:
3204
3212
.
41
Bloch
,
E. M.
,
A. E.
Levin
,
P. C.
Williamson
,
S.
Cyrus
,
B. H.
Shaz
,
D.
Kessler
,
J.
Gorlin
,
R.
Bruhn
,
T. H.
Lee
,
L.
Montalvo
, et al
.
2016
.
A prospective evaluation of chronic Babesia microti infection in seroreactive blood donors.
Transfusion
56
:
1875
1882
.
42
Raffalli
,
J.
,
G. P.
Wormser
.
2016
.
Persistence of babesiosis for >2 years in a patient on rituximab for rheumatoid arthritis.
Diagn. Microbiol. Infect. Dis.
85
:
231
232
.
43
Cursino-Santos
,
J. R.
,
M.
Singh
,
P.
Pham
,
M.
Rodriguez
,
C. A.
Lobo
.
2016
.
Babesia divergens builds a complex population structure composed of specific ratios of infected cells to ensure a prompt response to changing environmental conditions.
Cell. Microbiol.
18
:
859
874
.
44
Ruebush
,
M. J.
,
W. L.
Hanson
.
1980
.
Transfer of immunity to Babesia microti of human origin using T lymphocytes in mice.
Cell. Immunol.
52
:
255
265
.
45
Ruebush
,
M. J.
,
W. L.
Hanson
.
1980
.
Thymus dependence of resistance to infection with Babesia microti of human origin in mice.
Am. J. Trop. Med. Hyg.
29
:
507
515
.
46
Aguilar-Delfin
,
I.
,
M. J.
Homer
,
P. J.
Wettstein
,
D. H.
Persing
.
2001
.
Innate resistance to Babesia infection is influenced by genetic background and gender.
Infect. Immun.
69
:
7955
7958
.

The authors have no financial conflicts of interest.

Supplementary data