Thrombocytopenia, a condition in which the blood has a lower than normal number of platelets, is one of the most consistent findings among human patients and experimental animal models of AVHF; thrombocytopenia is used as a major diagnostic feature in patients with AVHF [33
]. In Venezuelan HF, for example, most patients showed thrombocytopenia, and although the clinical courses of these patients varied, the gross and histopathological necropsy findings were remarkably similar and generally showed evidence of bleeding [35
]. In contrast to Lassa fever, the bleeding that occurs with severe thrombocytopenia is more common in Argentine and Bolivian HFs [37
The causes of the thrombocytopenia associated with AVHF remain poorly understood. In this regard, DIC could explain platelet consumption; nevertheless, the occurrence of DIC in AVHF infections is inconclusive, at least for the arenaviridae
]. Furthermore, the occurrence of thrombocytopenia before the appearance of antibody or complement activation does not support immunologically mediated mechanisms of platelet destruction [39
]. Therefore, a high level of splenic sequestration or impaired megakaryo/thrombopoiesis could be the major physiopathogenic mechanisms responsible for the low platelet count.
Conflicting data were obtained from AHF patients in the 1970s. While Gallardo et al
. found hypocellularity in bone marrow samples of AHF patients, particularly in the erythroid and megakaryocytic lineages [40
], Ponzinibbio et al.
could not show any megakaryocytic anomalies [41
]. However, infected megakaryocytes has been observed in JUNV-infected guinea pigs [42
Recent findings show that JUNV not only replicates in human megakaryocytes and their precursor CD34+ cells but also that viral infection selectively impairs thrombopoiesis by decreasing in vitro
proplatelet formation and platelet release [43
]. The decrease in platelet release was shown to be TfR1-dependent and mimicked by poly(I:C); additionally, type I interferon (IFN I) was implicated as a key paracrine mediator. Although the molecular basis governing the IFN I-mediated reduction of in vitro
platelet production is still unknown, a low content of NF-E2 (a transcription factor that plays a major role in terminal differentiation of megakaryocytes and platelet release) was found in megakaryocytes treated with IFN I. Moreover, an ultrastructural analysis revealed that in the IFN I-treated megakaryocytes, the distinctive demarcation membrane system was almost absent and lacked organization and platelet territories [43
]. Interestingly, a correlation between high levels of circulating IFN α and both virulence and prognosis has been described in clinical [44
] and experimental [45
] AHF. Overall, these data support an emerging role for IFN I as a pathogenic factor for the thrombocytopenia observed in VHF patients and maybe in other diseases associated with increased bone marrow IFN I levels [43
In addition to a low platelet number, platelet dysfunction might also be a major contributor to the hemorrhagic phenotype observed in AVHF patients. Platelet dysfunction has several potential causes, including circulating fibrin degradation products, activated platelets (exhausted platelets syndrome), or specific inhibitors. In the case of inhibitors, a plasma inhibitor of platelet function was found in 80% of Lassa fever patients with a hemorrhage but in only 16% of those without a hemorrhage and was significantly associated with disease severity [46
]. Furthermore, a continuous rise in the inhibitory activity correlated with clinical deterioration, whereas a decline corresponded to clinical improvement. A similar inhibitor of platelet function has been demonstrated in patients with AHF [47
]. This inhibitor has in vitro
effects similar to those observed in patients with Lassa fever; however, it appears to be more thermolabile, and the inhibitory activity was not neutralized by convalescent plasma containing a high titer of protective antibodies against JUNV [48
]. Although the presence of a platelet inhibitor could account for the bleeding diathesis, there has been no report describing abnormalities of platelet function in AVHF infected patients, perhaps related to the absence of on-site adequate equipped laboratories of hemostasis ( and ).
In 2008, two major advances using experimental mouse models contributed to the current knowledge regarding the role of platelets in AVHF. First, it was demonstrated that mice rendered thrombocytopenic only suffered localized hemorrhages at the sites undergoing non-infectious inflammatory processes and that low numbers of circulating platelets were able to prevent these inflammation-induced hemorrhages [49
]. Second, Iannacone et al.
reported that platelet-depleted mice infected with LCMV (Armstrong strain) developed a syndrome characterized by mucocutaneous bleeding, vascular leakage, anemia, uncontrolled viral replication, suboptimal immune responses, and subsequent death. Remarkably, a lethal hemorrhage was less associated with thrombocytopenia and instead was more closely associated with the platelet dysfunction mediated by high IFN I levels [50
]. Interestingly, as neither Interferon-α nor β inhibited platelet responses in-vitro
, the authors suggested that IFN I could directly affect megakaryocytes rather than platelets. The recent description of the functional IFN I receptor in human megakaryocytes further strengthens this hypothesis [51
Another major issue of the Iannacone et al.
study was the observation that in addition to having life-threatening hemorrhagic anemia, the platelet-depleted mice failed to mount an efficient cytotoxic T lymphocyte (CTL) response and were unable to clear the LCMV. Transfusion of functional platelets into these animals reduced the hemorrhage, prevented death and restored the CTL-induced viral clearance in a manner partially dependent on the CD40 ligand (CD40L). These results indicate that, upon activation, the platelets expressing integrin β3 and CD40L are required to protect the host against the induction of an IFN I-dependent lethal hemorrhagic diathesis and for clearing the LCMV infection through CTLs. In the same line of evidence, Loria et al.
recently showed that mice profoundly depleted of platelets (>95% depletion) and infected with the Armstrong LCMV strain developed hemorrhagic spots in several organs along with high viral titers, generalized splenic necrosis, and increased mortality. Interestingly, the authors also found that the presence of the remaining 15% of the platelets was sufficient to prevent vascular damage but not viral replication, necrotic destruction of innate and adaptive immune splenocytes, or CTL exhaustion [52
]. These observations not only confirm the novel notion that platelets are necessary to protect vascular integrity and are critical mediators of viral clearance but also underscore an underappreciated relationship between platelet mediated-hemostasis, viral infection, and immunosuppression. Furthermore, the authors perceptively suggested that the higher circulating platelet levels in mice compared to other species may explain why mice are not susceptible to AVHF and offered a simple alternative model to study the pathophysiology of AVHF [52
]. This new experimental strategy, together with other recently described models of genetically modified mice, will help to clarify the issues regarding the pathogenesis of AVHF [53