Cish knockout mice exhibit similar outcomes to malaria infection despite altered hematopoietic responses

The Cytokine-inducible Src homology 2 domain-containing (CISH) protein is a negative feedback regulator induced by cytokines that play key roles in immunity and erythropoiesis. Single nucleotide polymorphisms (SNPs) in the human CISH gene have been associated with increased susceptibility to severe malaria disease. To directly assess how CISH might influence outcomes in the BALB/c model of malaria anemia, CISH knockout (Cish−/−) mice on this background were infected with Plasmodium berghei and their hematopoietic responses, cytokine production and ability to succumb to severe malaria disease evaluated. Despite basal erythrocytic disruption, upon P. berghei infection, the Cish −/− mice were better able to maintain peripheral blood cell counts, hemoglobin levels and a steady-state pattern of erythroid differentiation compared to wild-type (Cish+/+) mice. Ablation of CISH, however, did not influence the outcome of acute malaria infections in either the BALB/c model or the alternative C57BL/6 model of experimental cerebral malaria, with the kinetics of infection, parasite load, weight loss and cytokine responses being similar between Cish+/+ and Cish−/− mice, and both genotypes succumbed to experimental cerebral malaria within a comparable timeframe.


Introduction
Malaria is a globally important disease caused by parasites from the genus Plasmodium.Approximately 247 million new cases of malaria occur each year, with the majority of the estimated 619,000 deaths in 2021 occurring in sub-Saharan Africa and from infection with P. falciparum (World Health Organization, 2022).The most common syndromes associated with severe or complicated malaria include severe malaria anemia, cerebral malaria (CM) and respiratory distress, either singly or in combination (Moxon et al., 2020).The causes of severe malaria anemia are multi-factorial and include the destruction of parasitized red blood cells (pRBCs), clearance of uninfected RBC, suppression of erythropoiesis and dyserythropoiesis [reviewed in Lamikanra et al. (2007) and Moxon et al. (2020)].CM is a severe neurological complication, leading to coma and death, and arises from several pathophysiological mechanisms, including sequestration and microvascular obstruction of pRBC (MacPherson et al., 1985;Pongponratn et al., 2003), liberation of pro-inflammatory mediators, dysregulation of coagulation pathways (van der Heyde et al., 2006;Francischetti et al., 2008;Moxon et al., 2009) and disruption of the blood brain barrier (Beare et al., 2009;Dorovini-Zis et al., 2011).Respiratory distress stems from metabolic acidosis leading to respiratory compensation and tissue hypoxia but can also result from damage to the lung due to the infiltration of leukocytes, microhemorrhages and pulmonary edema (Taylor et al., 2012).
The risk of succumbing to severe disease upon Plasmodium infection is dependent on the age and immune status of the infected individual.Most malaria deaths in endemic regions of Africa are seen in young children but in regions such as Asia, where transmission is lower, severe disease occurs in older children and adults.Environmental, socioeconomic and host genetic factors also contribute to the varied responses to infection (Weatherall and Clegg, 2002;Marquet, 2018).For example, mutations and polymorphisms in a variety of human genes, such as those mediating glucose-6-phosphate dehydrogenase deficiency, sickle cell trait and other hemoglobinopathies, have been associated with malaria resistance through linkage analysis and association studies (Driss et al., 2011;Marquet, 2018).However, dissecting human genetic susceptibility to malaria has been challenging because of the heterogeneity in the clinical presentations of malaria.Moreover, selection pressures on human genes might differ between populations and rather than common variants in a small number of genes, rare mutations in multiple genes may be involved.As a consequence, inconsistent findings between studies are frequently observed (Marquet, 2018).
In more recent times, genome-wide association studies (GWAS) and genotyping tools have identified genetic polymorphisms associated with altered susceptibility to P. falciparum malaria, which may pave the way to fully understanding the pathogenesis of severe disease (Driss et al., 2011;Marquet, 2018).In one such study, five single nucleotide polymorphisms (SNPs) in the gene encoding the human cytokineinducible Src homology 2 domain-containing (CISH) protein were associated in African and Asian cohorts with increased susceptibility to severe manifestations of malaria infections, as well as to bacteremia and tuberculosis (Khor et al., 2010).The overall risk of contracting one of these infectious diseases was increased by >18% for people harboring the variant CISH alleles.However, SNP -292 accounted for most of the association and was shown to correlate with reduced induction of CISH expression by IL-2 ex vivo (Khor et al., 2010).Subsequent studies have also revealed associations between CISH polymorphisms and increased susceptibility to tuberculosis or chronic hepatitis B across Asian and African populations (Wang and Wang, 2010;Ji et al., 2014;Sun et al., 2014;Zhao et al., 2014;Naderi et al., 2016).
CISH plays a critical role in immunity and has been shown to regulate the development and homeostasis of T cells, NK cells, dendritic cells and myeloid cells (Yang et al., 2013;Palmer et al., 2015;Delconte et al., 2016;Louis et al., 2020;Naser et al., 2023), and also plays a role in the regulation of erythropoiesis (Maymand et al., 2023).We hypothesized that ablation of CISH could increase susceptibility to severe malaria anemia and cerebral malaria through its impacts on erythropoiesis and/or immunity, possibly via altered inflammatory cytokine responses.To examine this hypothesis, wildtype (Cish +/+ ) and CISH knockout (Cish −/− ) mice were compared in mouse models of malaria anemia (Craig et al., 2012) or experimental cerebral malaria (ECM) (de Oca et al., 2013) resulting from elevated pro-inflammatory immune responses (Hanum et al., 2003).This allowed examination of the impact of CISH ablation on hematopoiesis, inflammatory cytokine levels and susceptibility of mice to severe disease.

Mice and ethics approval
Cish +/+ and knockout Cish −/− mice (Naser et al., 2022) on either a BALB/c or C57BL/6 background were backcrossed 8 times followed by an intercross to generate Cish +/+ and Cish −/− founders that were maintained as separate lines.Mice were fed a standard rodent diet (chow) and housed under controlled conditions at 21ϒC with a 12:12 h light:dark cycle.Female mice between 6-10 weeks of age were used for the experiments unless otherwise stated.All experiments were approved by the Deakin University Animal Welfare Committee (Projects G37/2013 and G09/2017) and performed in accordance with National Health and Medical Research Council recommendations in the ' Australian code for the care and use of animals for scientific purposes' .

Plasmodium berghei infection studies
Mice were infected with wildtype P. berghei ANKA (10 6 parasitized RBCs unless otherwise stated) by intraperitoneal injection and relative parasitemia determined by visualization of Giemsa-stained blood smears and counting a minimum of 1,000 RBC.For the malaria anemia studies, BALB/c mice were used.For the experimental cerebral malaria studies, C57BL/6 mice, which are susceptible to cerebral malaria, were used.Mice were monitored for cerebral malaria symptoms, including ataxia, limb paralysis, presence of seizures and inability to self-right and mice were humanely culled when at they displayed at least three of these symptoms.

Mouse blood analysis
Blood was collected from the tail of mice with minivets (Sarstedt™) and analyzed with a hematology analyzer (SCIL) according to the manufacturer's instructions.

Harvesting of mouse bone marrow and spleen cells
Bone marrow was extracted from the tibias and femurs using a 26-gauge needle and 1 mL RPMI 1640 media (Life Technologies).The entire spleen was placed in 5 mL media and passed through a 40 μm nylon mesh cell strainer (Interpath) to isolate single splenocytes.The cells were then centrifuged at 1,000 × g prior to resuspension in the appropriate buffer.

Statistical analysis
Statistical analysis was performed using Graph Pad Prism v8 (GraphPad Software, La Jolla California, USA).For mouse survival experiments, a Wilcoxon log-rank test was used, and for analysis of parasitemia and change in body weight a two-tailed student's t-test was used.For other experiments, a one-way ANOVA with Šidák correction for multiple testing was used to compare between groups.A p value <0.05 was considered statistically significant.

Cish ablation leads to alteration in blood cell parameters during infection
To assess the impact of Cish ablation on malaria infection, Cish +/+ and Cish −/− mice on a BALB/c background were inoculated with 1×10 6 P. berghei ANKA and at 0, 5 and 7 days post-infection (dpi) peripheral blood was collected.In agreement with previous work (Lakkavaram et al., 2020), the hemoglobin levels, hematocrit and RBC numbers of Cish +/+ mice significantly decreased as the infection progressed to less than half those of uninfected mice, consistent with the development of severe anemia (Figures 1A-C).In contrast, while hemoglobin levels, hematocrit and RBC numbers in uninfected Cish −/− mice were significantly lower than in wildtype mice initially, they remained stable throughout the course of infection such that by 7 dpi they were significantly higher than Cish +/+ mice.No significant change was observed in the mean corpuscular volume of Cish +/+ mice over the course of infection, although it was elevated at 7 dpi in Cish −/− mice (Figure 1D).Platelet counts were comparable in uninfected mice of both genotypes and decreased to a similar extent by 5 dpi, but platelet count recovery was significantly stronger in Cish −/− mice (Figure 1E).The number of WBCs in Cish +/+ mice increased during the course of infection as described (Lakkavaram et al., 2020), with similar results obtained with Cish −/− mice (Figure 1F).

Impact of Cish ablation on hematopoiesis during infection
The impact of Cish ablation on bone marrow and spleen hematopoiesis during a P. berghei infection was assessed.No difference in bone marrow cellularity was observed between BALB/c Cish +/+ and Cish −/− mice before or after infection, with the cellularity decreasing to a similar extent in each case (Figure 2A; Supplementary Figure S1).Infection resulted in a significant decline in the frequency and number of Ter119 + erythroid cells in Cish +/+ mice (Figure 2B; Supplementary Figure S2).This was not the case for Cish −/− mice, so although they had a significantly lower frequency and number of Ter119 + erythroid cells compared to Cish +/+ mice prior to infection and their numbers declined during infection, at 8 dpi the difference in cell frequency and number between genotypes was no longer significant (Figure 2B; Supplementary Figure S2).Ter119/CD44 double staining (Chen et al., 2009) was used to quantify the frequency and number of specific stages of erythropoietic development (Supplementary Figures S1, S2; Figure 2C).Following infection, a statistically significant increase in the frequency of all erythroblast populations and reticulocytes was seen in Cish +/+ mice, while the RBC frequency decreased, although this did not reach statistical significance (Figure 2C).As a consequence, the ratio of pro-:basophilic:polychromatic:orthochromatic erythroblasts did not follow the usual doubling of cell number that occurs as a result of each successive mitosis in infected Cish +/+ mice (Figure 3A).Uninfected BALB/c Cish −/− mice possessed a significantly increased frequency of basophilic, polychromatic and orthochromatic erythroblasts compared to Cish +/+ mice, while RBC frequency was decreased.In response to infection, only the frequency of proerythroblasts and orthochromatic erythroblasts increased significantly in Cish −/− mice, while in contrast the polychromatic erythroblasts and RBC significantly decreased.As such, the frequency of basophilic erythroblasts became significantly lower and orthochromatic erythroblasts significantly elevated compared to infected Cish +/+ mice (Figure 2C).Thus, the ratios and numbers of erythroblast populations, while perturbed in uninfected BALB/c Cish −/− mice, were close to normal after infection and similar in number to Cish +/+ mice (Figure 3A; Supplementary Figure S2).
Infection resulted in a significant increase in CD11b + Gr1 + Ly6G + neutrophils in Cish +/+ mice but not CD11b + Gr1 + Ly6G − monocytes.While the frequency of the CD11b + Gr1 + Ly6G − monocyte lineage and the frequency and total number of CD11b + Gr1 + Ly6G + neutrophil lineages were significantly increased in uninfected Cish −/− mice compared to equivalent Cish +/+ mice (Figure 2D; Impact of CISH ablation on hematopoiesis dynamics in the bone marrow of P. berghei-infected mice.Analysis of (A) bone marrow cellularity and relative proportions of (B) Ter119 + cells, (C) specific erythroid cell populations, (D) CD11b + Gr1 + Ly6 − monocyte, CD11b + Gr1 + Ly6 + neutrophil, and CD61 + megakaryocyte cells, and (E) hematopoietic colony forming cells during a P. berghei infection of BALB/c Cish +/+ and Cish −/− mice.Data represent was also significantly higher in the Cish −/− mice at this time (Supplementary Figure S2).However, the ratios of erythroblast populations were relatively normal, with only the polychromatic:basophilic erythroblast ratio significantly altered in Cish +/+ mice (Figure 3B).Infection resulted in no change in the frequency of splenic CD11b + Gr1 + Ly6G − monocytes in either mouse genotype.The frequency of CD11b + Gr1 + Ly6G + neutrophils was basally elevated in Cish −/− mice, but significantly decreased in both genotypes with infection, such that at 8 dpi their frequency was comparable (Figure 4D; Supplementary Figure S3).The frequency of CD61 + cells was significantly decreased in Cish +/+ mice only and was significantly lower than in Cish −/− mice pre-and post-infection, and the number of CD61 + cells in Cish +/+ mice at 8 dpi was also significantly lower (Figure 4D; Supplementary Figure S2).
Infection led to a significant decrease in the frequency of splenic CFU-GEMM in Cish +/+ mice only and a significant decrease in the frequency of CFU-GM in both genotypes, although more rapidly in Cish +/+ mice (Figure 4E).Conversely, a significant increase was observed in the frequency of CFU-G, CFU-M, BFU-E and CFU-E in both genotypes.This occurred more rapidly in Cish +/+ mice such that their frequencies were significantly higher than in the Cish −/− mice at 5 dpi, but by 8 dpi the frequency of all these populations except for CFU-E was significantly higher in the Cish −/− mice, with this being the case basally just for BFU-E (Figure 4E).
As EPO is a key mediator of erythropoiesis (Jelkmann, 2007), serum EPO levels were also analyzed in Cish +/+ and Cish −/− mice.Infection resulted in significantly elevated EPO expression at 8 dpi in Cish +/+ mice, but not in the Cish −/− mice from a similar basal level (Table 1).

Cish ablation does not influence the outcome of Plasmodium berghei infection
To determine whether the differences in hematopoiesis impacted malaria infection, both female and male BALB/c Cish +/+ and Cish −/− littermates were infected by intraperitoneal injection with 1 × 10 6 P. berghei ANKA-infected RBCs.No significant differences in parasitemia levels or kinetics (Figures 5A,C) were observed between the genotypes nor in the amount of weight lost as a result of infection of either sex (Figures 5B,D).

Discussion
This study examined how the absence of CISH affected hematopoietic, cytokine and blood cell parameters of mice in response to acute infection with P. berghei.Uninfected BALB/c Cish −/− mice exhibited a notable difference in basal bone marrow erythropoiesis, with a lower frequency of BFU-E compared to Cish +/+ mice, as well as an altered frequency, number and ratio of specific erythroblast populations, likely underpinning the reduction in RBCs and diminished frequency of Ter119 + erythroid cells in the bone marrow.In contrast, erythropoiesis in the spleens of uninfected Cish −/− mice showed normal ratios, with the expected physiological progression of erythroid differentiation.However, as the bone marrow is mostly responsible for basal erythropoiesis, the Cish −/− mice exhibited reduced peripheral RBC count, hemoglobin levels and hematocrit.These results are in agreement with recent research on the role of CISH in normal erythropoiesis, with suppression of bone marrow erythropoiesis associated with altered expression of a set of erythroidrelated genes observed in Cish −/− mice, but increased proliferation in the spleen (Maymand et al., 2023).Uninfected Cish −/− mice also showed a significantly increased frequency of CFU-G in the bone marrow, consistent with elevated frequency of CD11b + Gr1 + Ly6G + neutrophil cells in this organ, while in the spleen, the frequency of CD11b + Gr1 + Ly6G + neutrophils was similarly increased.These findings are consistent with a recent study examining the role of CISH in myelopoiesis (Naser et al., 2023).In addition, CD61 + cells of the megakaryocytic lineage were also elevated in the spleen of Cish −/− mice.
Intriguingly, despite dysregulated basal erythropoiesis in uninfected Cish −/− mice, they were able to maintain relatively stable peripheral blood counts and hematocrit during infection.This correlated with the ability to maintain BFU-E and a normal pattern of erythroid differentiation in the bone marrow, although the frequency of CFU-E did decline in similar manner to infected Cish +/+ mice.In contrast, the orthochromatic:polychromatic erythroblast ratio could not be maintained in the Cish +/+ mice.Spleen erythropoiesis was less impacted in infected Cish −/− mice, but Cish +/+ mice exhibited a statistically significant decrease in the polychromatic:basophilic ratio compared to uninfected mice.The differences in basal erythropoiesis between the genotypes appeared to be independent of EPO, since uninfected Cish +/+ and Cish −/− mice exhibited similar levels of EPO.An increase in EPO levels was observed in infected Cish +/+ but not Cish −/− mice, most likely because severe anemia was observed only in Cish +/+ mice.We have recently shown that CISH expression is induced by EPO in both bone marrow and spleen.Cish −/− mice could also respond robustly overall to EPO injection, with bone marrow erythropoiesis somewhat blunted but with spleen erythropoiesis enhanced to compensate (Maymand et al., 2023).Collectively this indicates a differential impact of CISH ablation on erythropoiesis at these two sites and suggests that CISH acts as an inducible regulator of EPO signaling in vivo.
While there were clear differences in hematopoietic parameters between Cish genotypes, similar parasitemia kinetics were observed in infected female BALB/c Cish +/+ and Cish −/− mice, which also lost a comparable amount of bodyweight.Male BALB/c Cish +/+ and Cish −/− mice likewise exhibited equivalent infection outcomes, although hematopoietic parameters were not examined in these mice.Both female Cish genotypes exhibited similar elevations of specific cytokines, including those associated with anemia, such as TNF-α and IL-6 (Lyke et al., 2004;Chopra et al., 2015), with only IL-18 levels higher in Cish +/+ mice following infection.These results are consistent with there being a similar frequency of CD11b + Gr1 + Ly6G − monocyte and CD11b + Gr1 + Ly6G + neutrophil lineages in both genotypes at 8dpi.
In the C57BL/6 model of ECM, female Cish +/+ and Cish −/− mice displayed a similar course of parasitemia, change in bodyweight and kinetics of ECM induction after being infected with P. berghei, which was also seen in male Cish +/+ and Cish −/− mice.The inflammatory responses in P. berghei infected mice that succumb to ECM has been well studied, with CD4 + T cells (Yanez et al., 1996), CD8 + T cells (Nitcheu et al., 2003) and NK cells (Hansen et al., 2007), all previously shown to play a role in the development of ECM.Studies have shown the absence of CISH leads to an enhancement of NK cell survival and proliferation (Delconte et al., 2016) and CD8 + T cell expansion and cytokine polyfunctionality (Palmer et al., 2015;Delconte et al., 2016).However, both Cish +/+ and Cish −/− mice suffered a similar fate following infection with P. berghei despite this.It is important to note that in the absence of CISH, other SOCS proteins that also contribute to the negative regulation of cytokine signaling (Sobah et al., 2021), could potentially compensate to control the inflammatory responses that drive ECM.For example, SOCS2 and SOCS3 expression was found to elevated in the arcuate nucleus of Cish −/− mice, which suggests that compensatory mechanisms exist (Naser et al., 2022).Therefore, it would be interesting to examine whether the expression of other SOCS proteins in the bone marrow and in regions of the brain where parasites sequester is affected in the absence of CISH.
Collectively these results indicate that CISH in not a crucial contributor of parasite load and that an absence of CISH alone is unable to influence the outcome of an acute malaria infection.These findings, therefore, fail to explain the association between SNPs in human CISH and increased susceptibility to the severe manifestations of malaria infections (Khor et al., 2010).However, complete ablation of CISH is not the same as the more subtle impacts seen in the human gene that instead results in altered expression, and so efforts to recapitulate the relevant human SNPs in mice would be worthwhile.It is also likely that prior exposure of human participants to Plasmodium and other infectious agents may influence how CISH impacts the immune response compared to naïve mice.Indeed, the loss of CISH has been shown to lead to dysregulated immune responses to different pathogens (Sun et al., 2014;Kotas et al., 2021).It would therefore be interesting to examine whether repeated exposure of Cish −/− mice to Plasmodium dysregulates the immune response in a manner that might enhance the severity of disease.In addition, chronic infection likely places more stress on the maintenance of erythropoiesis and so investigation of whether the dysregulated basal erythropoietic response in Cish −/− mice affects the progression of such an infection would also be worthwhile.The author(s) declared that they were editorial board member of Frontiers, at the time of submission.This had no impact on the peer review process and the final decision.

FIGURE 6
FIGURE 6Analysis of cytokines and chemokines in P. berghei infection.Concentrations of the indicated cytokines and chemokines that were either significantly different between genotypes or were significantly altered in response to a P. berghei infection in BALB/c Cish +/+ and Cish −/− mice.Data represent individual mice and the mean ± SD (n = 6 mice).Statistical significance between the indicated groups was determined by an unpaired t-test.*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

TABLE 1
Impact of P. berghei infection on serum EPO levels.