Immune response during hantavirus diseases: implications for immunotherapies and vaccine design

Summary

Orthohantaviruses, previously named hantaviruses, cause two emerging zoonotic diseases: haemorrhagic fever with renal syndrome (HFRS) in Eurasia and hantavirus cardiopulmonary syndrome (HCPS) in the Americas. Overall, over 200 000 cases are registered every year worldwide, with a fatality rate ranging between 0·1% and 15% for HFRS and between 20% and 40% for HCPS. No specific treatment or vaccines have been approved by the U.S. Food and Drug Administration (FDA) to treat or prevent hantavirus‐caused syndromes. Currently, little is known about the mechanisms at the basis of hantavirus‐induced disease. However, it has been hypothesized that an excessive inflammatory response plays an essential role in the course of the disease. Furthermore, the contributions of the cellular immune response to either viral clearance or pathology have not been fully elucidated. This article discusses recent findings relative to the immune responses elicited to hantaviruses in subjects suffering HFRS or HCPS, highlighting the similarities and differences between these two clinical diseases. Also, we summarize the most recent data about the cellular immune response that could be important for designing new vaccines to prevent this global public health problem.

In this review, we summarize the main findings relative to the immune response in humans and animal models for both HFRS and HCPS and discuss the immunological basis for the rational design of vaccines and immunotherapeutic interventions.

Abbreviations

ANDV

Andes orthohantavirus

CFR

case fatality rate

CRP

C‐reactive protein

CTL

cytotoxic T lymphocyte

DC

dendritic cell

DOBV

Dobrava–Belgrade orthohantavirus

dsRNA

double‐stranded RNA

EC

endothelial cell

FDA

Food and Drug Administration

GPC

glycoprotein precursor

HCPS

hantavirus cardiopulmonary syndrome

HFRS

haemorrhagic fever with renal syndrome

HTNV

Hantaan orthohantavirus

IL

interleukin

ISG

interferon‐stimulated gene

mAbs

monoclonal antibodies

MAVS

mitochondrial antiviral signalling protein

MDA5

melanoma differentiation‐associated protein 5

MMP‐9

matrix metalloproteinase 9

moDC

monocyte‐derived dendritic cell

nAb

neutralizing antibody

NE

nephropathia epidemica

NK

natural killer

NP

nucleoprotein

NSs

non‐structural proteins

PB

plasmablast

PBMC

peripheral blood mononuclear cell

PHV

Prospect Hill orthohantavirus

PKR

protein kinase R

PUUV

Puumala orthohantavirus

RIG‐I

retinoic acid‐inducible gene I

SEOV

Seoul orthohantavirus

SNV

Sin Nombre orthohantavirus

Th

T helper

TLR

Toll‐like receptor

TNF

tumour necrosis factor

Treg

regulatory T cell

TULV

Tula orthohantavirus

VEGF

vascular endothelial growth factor

VLP

virus‐like particle

VSV

vesicular stomatitis virus

Introduction

Orthohantaviruses (hantaviruses from now on in this review) are a genus of viruses belonging to the Bunyavirales order and Hantaviridae family. ¹ Hantaviruses contain a negative‐sense single‐stranded tri‐segmented RNA genome. The large (L) viral genomic segment encodes an RNA‐dependent RNA polymerase, the medium (M) segment encodes a glycoprotein precursor (GPC, which generates Gn and Gc glycoproteins in the host cell), and the small (S) segment encodes a nucleoprotein (NP) and, in some hantaviruses, a non‐structural protein (NSs). ² The pleomorphic virion is enveloped, with a size between 70 and 160 nm. ³ , ⁴ A tetrameric assembly of the Gn and Gc glycoproteins on the virus surface constitutes the spike complex that mediates cell entry and virus assembly. ³

The main animal reservoirs of hantaviruses are rodents, showing a strong degree of host‐virus specificity. ⁵ Hantavirus outbreaks are reported after dynamic changes in the host population, which are influenced by several factors, including environmental forces and anthropogenic disturbances. ⁶ , ⁷ , ⁸ Transmission to humans takes place mostly after inhalation of contaminated droplets from rodent excreta. ⁹ Humans are usually dead‐end hosts for these viruses ¹⁰ ; however, person‐to‐person transmission has been described with the highly pathogenic Andes orthohantavirus (ANDV). ¹¹ , ¹² , ¹³

The hantavirus species are the aetiologic agents of two different diseases in humans: haemorrhagic fever with renal syndrome (HFRS) and hantavirus cardiopulmonary syndrome (HCPS). Nearly 200 000 cases per year of hantavirus infection are reported worldwide, ¹⁴ and the number of hantavirus species is still on the rise, with 50 species described to date, of which over 24 are recognized as pathogenic. ¹⁵ HFRS‐related hantavirus species are distributed in the Eurasia region, causing case fatality rates (CFR) between 0·1% and 15%, mainly by Hantaan orthohantavirus (HTNV), Dobrava–Belgrade orthohantavirus (DOBV) and Puumala orthohantavirus (PUUV) infection. ¹⁶ On the other hand, Sin Nombre orthohantavirus (SNV) and ANDV are the leading agents of HCPS in the Americas, with case fatality rates of up to 20–40%. ¹⁷ , ¹⁸ , ¹⁹ To date, no specific treatment has been approved for hantavirus‐caused diseases, despite the constant efforts to develop effective therapies and vaccines against these viruses. ¹⁴ Early intravenous ribavirin treatment has been described as useful for HTNV‐infected patients with HFRS in Asia, ²⁰ , ²¹ but not for PUUV ²² or HCPS‐related viruses. ²³ Due to the lack of effective antiviral treatments or vaccines, clinical interventions rely mostly on treating symptoms and on supportive care to provide haemodynamic and oxygen support in patients suffering more severe clinical forms of these diseases. ²⁴

An early neutralizing antibody (nAb) response has been associated with a favourable clinical outcome in HCPS individuals. ²⁵ Despite the importance of T cells in controlling viral infections, little is known about their contribution to hantavirus infection and immunity. Both HFRS and HCPS are characterized by a robust cytotoxic T lymphocyte (CTLs) response, ²⁶ and clearance of ANDV‐RNA mediated by cytotoxic CD8⁺ T cells without an increase in nAbs titres has also been reported. ²⁷ These data suggest that these cells might be necessary for protective immunity against ANDV infection. ²⁷ However, no additional observations have been published to support this hypothesis. It is noteworthy that hantavirus‐infected patients exhibit an elevated proinflammatory cytokine profile in the serum, organs and tissues. ²⁸ , ²⁹ , ³⁰ It has been proposed that hantavirus‐specific CD4⁺ T cells and CD8⁺ T cells contribute to the cytokine storm observed and capillary leak, with the consequent pulmonary oedema and cardiogenic shock during HCPS. ² , ³¹

Several features of hantavirus pathogenesis and immunity, ³² , ³³ , ³⁴ as well as advances in vaccines and therapeutics, ³⁵ , ³⁶ have been recently reviewed and discussed elsewhere. Here, we summarize the main findings relative to the immune response in humans and animal models for both HFRS and HCPS, emphasizing the similarities and differences among both syndromes, and discussing the immunological basis for the rational design of vaccines and immunotherapeutic interventions. A more comprehensive understanding of the protective response is needed to develop efficacious vaccines to protect against each syndrome.

Hantavirus‐caused diseases: hantavirus cardiopulmonary syndrome and haemorrhagic fever with renal syndrome

Hantavirus infection causes two different zoonotic human diseases: HFRS, caused by Old World hantaviruses, and HCPS, produced by New World hantaviruses. ⁹ Exposure to hantaviruses is associated with domestic, recreational and occupational activities in areas with wild rodents. ³⁷ Environmental factors, such as precipitations, temperature and landscape disturbances, might also impact hantavirus dynamics in rodents and contribute to human disease outbreaks. ⁸

Human infection occurs after inhalation of aerosolized contaminated faeces, urine or saliva. Microvascular endothelial cells (ECs) are the main target for hantaviruses, ³⁸ and although infections are systemic, viral dissemination within the body in humans has not been conclusively demonstrated. ¹⁸ The incubation period ranges from 7 to 39 days in HCPS ³⁹ and 14 to 28 days in HFRS cases. ² These syndromes share initial symptoms, such as headache, myalgia, vomiting and abdominal pain. In HCPS, following a short prodromal phase with flu‐like symptoms, patients develop a severe acute cardiopulmonary stage, with dyspnoea, pulmonary oedema, hypotension and shock. ⁴⁰ , ⁴¹ It is noteworthy that cardiogenic shock and severe respiratory failure are frequent in lethal outcomes by HCPS. ⁴² Furthermore, haemorrhagic and renal manifestations such as petechiae, haemorrhages and higher creatinine levels in the blood might also be present during ANDV infections. ⁴³ Respiratory sequelae, such as dyspnoea, have been registered even after 3 years in HCPS survivors infected by Choclo orthohantavirus and SNV. ⁴⁴

On the other hand, the clinical manifestations of HFRS might involve five phases: febrile, hypotensive, oliguric, polyuric and convalescent. ⁴⁵ However, the absence or overlapping of clinical stages is frequent. ⁴⁶ Severe cases develop renal dysfunction, with proteinuria and haematuria. ⁴⁶ Renal involvement is the most frequent condition in patients with HFRS, but pulmonary dysfunction might also present in the severe cases of HFRS, ⁴⁷ being life‐threatening conditions. ⁴⁸ Nevertheless, most of the individuals present a full recovery in the convalescent stage. ⁴⁹

The pathophysiological mechanisms of these syndromes remain unclear, and probably multiple pathways contribute to the development of these two diseases. Viral loads have not been associated with disease severity in PUUV, SNV or ANDV infections. ⁵⁰ , ⁵¹ , ⁵² However, due to the high viral RNA levels in patients with HCPS, it has been suggested that viraemia might trigger the immunopathology. ⁵⁰ Along these lines, viral replication elicits EC dysfunction in the vascular niche, with an increase in the secretion of vascular endothelial growth factor A (VEGF‐A) and an alteration of vascular permeability that ultimately leads to pulmonary oedema. ¹⁷ , ⁵³ Changes in plasma VEGF levels might not be evident in patients; however, the presence of VEGF in pulmonary oedema fluid has also been associated with disease and severity in patients with HCPS. ⁵³ Along these lines, studies in DOBV‐affected patients indicate that plasma levels of VEGF remain normal during the disease. ⁵⁴ On the other hand, another study reported that severe cases of DOBV had higher and prolonged VEGF levels as compared to moderate illness, suggesting that VEGF could be more associated with tissue repair than tissue dysfunction. ⁵⁵ Additional studies are required to elucidate the specific roles of VEGF in the pathogenesis or endothelial repair in the hantavirus disease context.

Thrombocytopenia, leucocyte activation and release of proinflammatory cytokines and chemokines are inherent changes that might be related to pathogenesis. ⁵³ , ⁵⁶ Several reports have demonstrated an upregulation of VEGF‐A and proinflammatory markers, such as nitric oxide, C‐reactive protein (CRP) and proinflammatory cytokines, including IL‐1β, IL‐2, IL‐6, IL‐8, TNF and IFN‐γ, among others. ³⁰ , ⁵³ , ⁵⁷ , ⁵⁸ , ⁵⁹ , ⁶⁰ , ⁶¹ This ‘cytokine storm’ contributes to the illness associated with the viral infection. A study in cases of HCPS or nephropathia epidemica (NE, a mild form of HFRS caused by PUUV) revealed that HCPS is characterized by a massive upregulation of proinflammatory cytokines and chemokines, such as IL‐6, IL‐18, CXCL9, CXCL10 and MIF. ⁶² Instead, NE cases display an increase in IL‐6 and IL‐12p40, and downregulation of IL‐18 compared with healthy controls. ⁶² Further, a more robust innate immune response and an earlier secretion of cytokines related to a T‐helper (Th) 1 response are observed during HCPS as compared to NE patients. ⁶² Besides, DOBV‐infected cases are related to higher serum cytokine expression in comparison with cases of PUUV, ⁵⁸ suggesting that the elevated cytokine expression correlates with disease severity. Consistent with this notion is the observation that high IL‐6 levels have been associated with more severe forms of HFRS and HCPS ⁶³ , ⁶⁴ , ⁶⁵ and with fatal outcomes of HCPS. ⁶⁴ The upregulation of these chemo‐attractant cytokines might be responsible for the recruitment of hantavirus‐specific immune cells and extensive bystander activation of innate and cytotoxic cells that destroy ECs. ³⁴ This scenario, combined with the cell‐to‐cell junction disruption, could promote platelet migration at the injury site, with the consequent thrombocytopenia commonly found in hantavirus disease. ⁵⁶ , ⁶⁶ Additionally, alterations between platelets and their ligands are reported during hantavirus infection, with high levels of fibrinogen and von Willebrand factor (VWF) in the acute stage of NE, ⁶⁷ both directly involved in the initial steps of clot formation. Furthermore, β3 integrin, the receptor of pathogenic hantaviruses, is expressed in platelets and is involved in vascular permeability and platelet function. ⁶⁸ Thus, although hantavirus infection does not produce cytopathic effects, current studies suggest that several pathogenic effects can be induced by hantavirus infection, ultimately leading to vascular leakage (Fig. 1).

Figure 1.

Cellular responses associated with hantavirus diseases. Hantaviruses infect endothelial cells, monocytes and macrophages without cytopathic effect but producing an endothelial disruption. The detection of virions activates the immune response, with the primary production of cytokines from innate immune cells such as types I and III IFNs, IL‐1β, TNF, IL‐6 and IL‐15, and chemokines such as CCL5 and CXCL10, among others. Later, in addition to NK cells and monocytes, CD8⁺ and CD4⁺ T cells are recruited to infected tissue, releasing granules as perforins, granzyme and specific Th1/Th2 cytokines inducing a ‘cytokine storm’. Due to the disruption of the endothelial layer, massive oedema is present in severe cases. Furthermore, a significant increase in plasmablasts (PBs) is present early in the disease with an early IgM/IgG production.

Innate immunity elicited by hantavirus infection

Hantavirus recognition by innate immunity

Hantaviruses replicate predominantly in ECs, macrophages and dendritic cells (DCs). ⁹ Upon viral infection, pathogen‐associated molecular patterns are recognized by host cell receptors at extra‐ and intracellular levels as Toll‐like receptors (TLR), retinoic acid‐inducible gene I (RIG‐I), melanoma differentiation‐associated protein 5 (MDA5) and protein kinases (PKR). ⁶⁹ , ⁷⁰ , ⁷¹ , ⁷² , ⁷³ , ⁷⁴ , ⁷⁵ , ⁷⁶

TLR3 recognizes the double‐stranded RNA (dsRNA) viral replication intermediate, promoting the synthesis of alpha/beta interferon (IFN‐α/β) and interferon‐stimulated genes (ISGs) to fight the viral infection. ⁶⁹ Nevertheless, pathogenic hantaviruses apparently alter the activation of TLR3 and its downstream effectors. In vitro studies have shown that HTNV infection delays the secretion of antiviral effectors, which may contribute to the innate evasion by delaying the inflammatory response, subsequently allowing a higher viral replication. ⁶⁹

After recognizing viral RNA motifs, RIG‐I and MDA5 activate mitochondrial antiviral signalling protein (MAVS), inducing the expression of type I IFN and proinflammatory cytokines at the early stages of infection. ⁷⁷ In this context, according to studies on human ECs, RIG‐I‐like receptors are fundamental to elicit an antiviral response against HTNV infection through the IFN production and ISG expression to limit the viral replication in vitro. ⁷⁰ It is noteworthy that ANDV presents various ways to inhibit both RIG‐I and MDA‐5 pathways. NP‐ANDV interferes with interferon regulatory factor 3 phosphorylation and TANK‐binding kinase 1 autophosphorylation, ⁷⁴ and NSs‐ANDV interacts with MAVS decreasing its ubiquitination and therefore downregulating the innate immune response. ⁷¹ Additionally, the pathogenic hantaviruses ANDV, Tula orthohantavirus (TULV) and New York 1 (NY‐1V) can modulate the IFN pathway through the cytoplasmic tail of the Gn protein, while, on the other hand, non‐pathogenic Prospect Hill orthohantavirus (PHV) does not inhibit the IFN pathway but induces a strong IFN‐ β response in human ECs. ⁷⁵ , ⁷⁶

On the other hand, PKR senses dsRNA leading to the attenuation of mRNA translation and inhibition of the protein synthesis and assembly of RNA stress granules (SGs). ⁷² PUUV and ANDV actively inhibit PKR‐dependent SG formation despite causing a transient, limited SG formation in the early stages of infection. ⁷² Interestingly, when NP expression increases in the late stages of infection, these proteins inhibit PKR autophosphorylation, keeping the translation process active. ⁷³

Apoptosis is another pathway triggered by infected cells against viral infection; nevertheless, hantaviruses can prevent cell death by the overexpression of NP in the late stages of the infectious cycle. ⁷⁸ NP motifs are recognized by caspase‐3, which leads to their degradation over caspase‐3 targets, ⁷⁸ thus preventing the activation of intrinsic apoptosis effectors. Moreover, HTNV has also been shown to inhibit the extrinsic pathway of apoptosis through the downregulation of dead receptor 5, in a TRAIL‐mediated mechanism activated by cytotoxic granules. ⁷⁹

Despite the viral strategies to avoid the innate immune system, in the late stages of infection type I IFN production has been reported in pathogenic hantavirus species. ⁸⁰ , ⁸¹ The activation of immune cells and secretion of proinflammatory cytokines (such as IL‐1β, TNF and IL‐6) and chemokines (such as CCL5 and CXCL10) ²⁸ , ⁴¹ are likely responsible for the recruitment of further mononuclear cells at the injury site, promoting a sustained proinflammatory state where cytokines and cytolytic activity may perpetuate endothelial damage. ²⁸ , ⁸¹ All these components may contribute to the development of the disease.

Natural killer cells

Natural killer cells are innate lymphoid cells with the ability to rapidly respond during viral infections by releasing cytotoxic granules that contain granzymes and perforin and the secretion of cytokines, such as IFN‐γ, TNF, IL‐12, IL‐15, IL‐18 and IL‐21. ⁵⁹ The dynamics of NK cells during hantavirus infection are poorly understood. Particularly, studies have focused on PUUV infection, where peripheral NK cell counts change along the infection time. During the febrile stage, NK cell counts from NE patients are lower than in healthy donors, ⁸² probably caused by extravasation. One week after this stage, NK cells increase and remain high for 2 months after NE onset, and then return to normal levels after a year. ⁸³ CD56^dim NK cells are increased in infected patients, showing elevated expression of activation markers, such as CD69, NKG2D, 2B4 and cytotoxicity receptors that include NKp30 and NKp46, with granzyme B and perforin secretion during the acute phase of HFRS. ⁸⁴ This active and proliferative phenotype is associated with an elevated IL‐15 secretion by infected epithelial cells. ⁸⁴ Similarly, elevated levels of IL‐15 in serum were reported in SNV‐infected rhesus macaques, ⁸⁵ and increased levels of IL‐15 are associated with high severity course and fatal outcome in ANDV‐infected patients with HCPS, ⁶⁵ but not in patients with HCPS from North America. ⁶⁰

It is noteworthy that in vitro studies have shown that HTNV‐infected ECs, but not uninfected ECs, are protected from NK cell cytotoxicity. ⁷⁸ , ⁸⁴ Despite significant TRAIL upregulation in infected ECs and NK cells, HTNV‐infected cells are protected by TRAIL‐mediated killing, as previously discussed. ⁷⁹ Moreover, inhibition of cell‐mediated apoptosis has been observed in pathogenic and non‐pathogenic hantaviruses, associated with NP‐driven inhibition of granzyme B and caspase‐3. ⁸⁶ Those events would explain why infected ECs are not damaged despite strong cytotoxic lymphocyte activation in patients, according to the authors. ⁷⁸ , ⁸⁶

Dendritic cells and macrophages

Upon HTNV infection, DCs mature and upregulate the expression of class I and class II major histocompatibility complex molecules, as well as costimulatory and adhesion molecules. ⁸⁷ DC activation also induces a proinflammatory response with TNF and IFN‐γ secretion. ⁸⁷ Moreover, HTNV infection of human monocyte‐derived DCs (moDCs) causes their maturation and promotes their antigen cross‐presentation capacity, ⁸⁸ an essential process for the activation of the antiviral activity mediated by CD8⁺ T cells. ⁸⁹ According to these findings, in vitro ANDV infection of moDCs promotes their maturation and a proinflammatory cytokine secretion profile, with high levels of TNF, and diminished production of IL‐10 and TGF‐β. Additionally, these cells increase the secretion of matrix metalloproteinase 9 (MMP‐9), whose activity is associated with higher endothelial permeability, a characteristic finding in HCPS cases. ⁹⁰

Interestingly, the ability of hantaviruses to replicate in mononuclear phagocyte system cells has been suggested to be related to their pathogenicity. Opposite to HNTV, the low‐pathogenic Tula orthohantavirus (TULV) does not replicate in human monocytes when inoculated in vitro. ⁸¹ Similarly, non‐pathogenic PHV was unable to replicate in inflammatory DCs. ⁸¹ Although mononuclear phagocytic system cells express both β1 and β3 integrins, the expression of CD86 is less sustained in TULV than in HTNV infection. ⁸¹ Additionally, experiments using immature DCs with defective signalling capacity of integrins prove that the capability of pathogenic hantavirus to replicate is dependent on the integrin signalling pathway. ⁸¹ Analyses of peripheral blood cells from HFRS patients with respiratory problems showed a significant decrease in mononuclear cells with an increase in CCR7 expression, suggesting a concomitant migration of these cells to the airways. ⁴⁷ These findings indicate that the activation of the immune response mediated by infected DCs could contribute to the pathogenesis during viral infection. ³⁴

Moreover, analyses of human tissue samples from ANDV fatal cases and infected wild Oligoryzomys longicaudatus rodents identified ANDV in alveolar macrophages and submandibular glands. ⁹¹ It has been proposed that replication in human salivary glands and expectoration of alveolar macrophages could contribute to person‐to‐person transmission. ⁹¹ Therefore, it is of great importance to understand the role of macrophages during hantavirus infection, not only as part of the immune response against the virus but also their potential role in viral transmission. ⁹¹

Adaptive immunity to hantavirus infection

CD8⁺ T‐cell response

The specific role of T cells during hantavirus disease has not been elucidated. Hantavirus infection does not produce a cytopathic effect on the vascular endothelium. Therefore, it has been suggested that strong immune activation of CTLs is involved in capillary leakage during a severe clinical presentation. ⁹² Additionally, NP‐specific CD8⁺ T cells could persist for more than 15 years after PUUV infection, suggesting strong, long‐lasting CD8⁺ T‐cell responses. ⁹³ Furthermore, there is an early increase in circulating CD8⁺ T cells, specifically in the effector population (Ki67⁺ CD38⁺ HLA‐DR⁺). ⁹⁴

A vigorous virus‐specific CD8⁺ T‐cell response has been observed with the onset of symptoms, with high levels of granzyme B and perforin accompanied by elevated expression of the inhibitory receptor CTLA‐4. ⁹⁵ Besides, 10 days after symptoms had started, a concomitant decline in the CD8⁺ T‐cell effector population and viral load is observed, suggesting an essential role for the CTLs in PUUV clearance and patient recovery. ⁹⁵ In contrast, a lack of correlation between CD8⁺ T‐cell effector population response and clinical severity has been described. ⁹⁶ Discrepancies between results studying PUUV could be due to differences among the immunological panel analysed and the incorporation of clinical and biodemographic variables.

CD8⁺ T cells might also be involved in the clinical HCPS outcome; however, current evidence is inconclusive. Immunoblasts are present in peripheral blood, and CD3⁺ infiltrates are found in lung necropsies from patients with HCPS. ⁹⁷ Furthermore, patients with severe HCPS by SNV infection might have higher specific circulating CD8⁺ T cells than those with a mild course of the disease. ²⁶ Although a case report showed a late viral clearance mediated by CD8⁺ T cells secreting granzyme B and IFN‐γ, ANDV‐RNA remained in blood cells even 67 days after viral infection. ²⁷ Later, a strong memory CD8⁺ T‐cell response against Gn‐ANDV with an effector phenotype (CD27⁻ CD28⁻ CCR7⁻ CD127⁻) that lasted up to 13 years after HCPS onset was described. ⁹⁸ Along these lines, it was observed that CD8⁺ T‐cell responses are long‐lasting during hantavirus infections. ⁹³ , ⁹⁸

The Syrian golden hamster model of fatal ANDV infection also exhibited a CD8⁺ T‐cell increase in peripheral blood and lungs. However, no differences in severity or outcome were observed when the disease was evaluated in T‐cell‐depleted hamsters as compared to control animals, suggesting that T cells play neither pathological nor protective roles in the hamster model of fatal ANDV infection. ⁹⁹ A study in rhesus macaques infected with SNV showed a high proportion of activated cytotoxic T cells during early infection and the development of a memory effector phenotype with granzyme B expression after disease, ⁸⁵ similar to the findings in humans. ⁹⁸ Despite evidence for the implication of CD8⁺ T cells during and after hantaviruses infections, their participation in pathology or clearance remains unclear.

CD4⁺ T‐cell response

CD4⁺ T‐cell responses in natural infections have been less addressed in studies than CD8⁺ T‐cell responses. Nevertheless, in hantavirus infection, CD4⁺ T cells may also be a part of the complex immune response in the affected organ in humans. During early PUUV infection, a decrease in CD4⁺/CD8⁺ ratios has been reported, which returned to baseline levels during the convalescent stage. ⁹⁴ Moreover, CD4⁺ T‐cell responses against hantaviruses are described as mixed Th1/Th2 responses based on sera cytokine profiles ³⁰ without a correlation between effector CD4⁺ T cells and clinical parameters. ⁹⁶ Analyses performed in DOBV‐affected patients show a negative correlation between total Th cells expressing CD69 and urea, creatinine and CRP levels, suggesting that greater activation of T‐cell subsets might contribute to ameliorate disease. ¹⁰⁰ Nevertheless, the roles of each Th subtype during hantavirus diseases remain unclear.

Regulatory T cells (Tregs) are the most studied subpopulation of CD4⁺ T cells, due to the notion of an exacerbated immune response involved in hantavirus pathogenesis. It has been suggested that Tregs are essential for establishing persistent hantavirus infection in rodent hosts and are involved in modulating the immunopathology. ¹⁰¹ , ¹⁰² , ¹⁰³ Increased levels of Tregs are observed in lungs of rats during Seoul orthohantavirus (SEOV) infection. ¹⁰¹ Interestingly, inactivation of Tregs not only led to a reduction in viral RNA in lungs and virus shedding in rats, but also led to reduced amounts of acute multifocal lesions in lungs. ¹⁰¹ Besides, FOXP3 mRNA levels remain static in infected hamsters during HCPS. ¹⁰⁴

The contribution of Tregs during human infection and clinical disease has not been elucidated yet, and studies addressing this T‐cell subset differ in methodological approaches. In the course of HFRS, it has been observed that FOXP3 expression levels are upregulated during PUUV infection and are directly correlated with hospitalization days, suggesting that Tregs play a role in disease severity. ⁹⁶ On the contrary, suppressive cytokines, such as TGF‐β1 and TGF‐β2, are decreased in severe PUUV cases as compared to mild clinical courses. ¹⁰⁵ Despite no differences in Treg frequency in PUUV cases in comparison with healthy donors, two main immune checkpoints involved in T‐cell regulation, PD1⁺ and CTLA4⁺, are elevated in CD4⁺ T cells. ⁹⁵ Also, an increased frequency of Tregs is observed early in PUUV cases as compared to healthy controls and DOBV‐infected individuals, ¹⁰⁰ suggesting that regulatory mechanisms might be promoting a more balanced and less harmful immune response. This observation is in line with the recent findings in an HCPS survivor cohort after ANDV infection. In this group, the frequency of PD‐1 expression in Tregs remains elevated years after the infection, suggesting that immune‐suppressive mechanisms might be involved in regulating HCPS disease, leaving an immunological signature. ¹⁰⁶ Furthermore, the dissection of regulatory subtypes indicates a downregulation of a Th1‐like Treg response, with a consequent increase in the Th2‐like Treg population, possibly mediated by CXCR3 downregulation through GP‐ANDV. ¹⁰⁶

B‐cell response

A robust IgM response emerges early after hantavirus infection, with a subsequent increase in IgG antibodies. ¹⁰⁷ , ¹⁰⁸ , ¹⁰⁹ Although surface glycoproteins first engage cellular receptors, the initial antibody response seems to target NP protein predominantly, being NP‐specific antibodies detectable shortly after symptom onset. ¹¹⁰ , ¹¹¹ A low specific IgG response is associated with severe disease in PUUV‐infected patients, ⁵¹ and higher nAbs titres have been detected in survivors when compared to deceased individuals, suggesting that nAbs production is directly associated with the chance of survival. ²⁵ Furthermore, nAbs have been detected years after PUUV, SNV and ANDV infections, implying a long‐lasting immune response. ¹¹² , ¹¹³ Surprisingly, ANDV survivors also show an increase in nAbs titres during time, even years after disease, suggesting that viral antigens might be present at later times after infection. ⁹⁸

Moreover, studies have suggested differences between New and Old World hantavirus species regarding the IgG subclass produced by the infected individual during the disease. IgG1 is the most prevalent subclass in patients infected by PUUV, ¹¹⁴ while IgG3 is the main subclass in HCPS by SNV. ¹¹⁵ However, no association between IgG subclass and clinical severity has been reported. An in‐depth characterization of IgG responses against different hantaviruses is required to better understand the relevance of antibody‐mediated responses for patient survival.

Despite the importance of B cells in antibody production, little is known about the dynamic changes in B‐cell populations after hantavirus infection. Recently, a 100‐fold increase in the plasmablast (PB) population during acute ANDV infection was described. ¹¹⁶ Remarkably, they found a virus‐specific population of specific antibody‐secreting cells and a significant rise in reactivity against virus‐unrelated antigens. ¹¹⁶ Nevertheless, none of these observations were associated with disease severity. It is noteworthy that this is not the first report that detects the presence of unrelated antibodies during hantavirus disease. This finding is consistent with a previous article describing autoantibodies due to PUUV infection. ¹¹⁷ Moreover, this finding is consistent with research that showed a significant risk of Hodgkin's lymphoma after HFRS by PUUV, similar to other viral infections, such as with the Epstein–Barr virus and the human immunodeficiency virus type 1. ¹¹⁸ However, additional research is required to elucidate whether those autoreactive antibodies have any role in hantavirus disease or are associated with a higher risk of autoimmune illness.

The current knowledge about the immune responses during hantaviruses disease is summarized in Table 1. Furthermore, according to published data, immune mechanisms associated with a positive or negative prognosis for hantavirus diseases are proposed in Fig. 2.

Table 1.

Main findings associated with cellular immune response during hantavirus diseases

HCPS	References	HFRS	References
Natural killer cells
No data published	‐	Lower peripheral blood levels in comparison with healthy donors during febrile stage. Peak during renal phase. Elevated for at least 2 months after infection High expression of activation markers	82, 83, 84
Dendritic cells/monocytes
Release of active MMP‐9 Proinflammatory cytokine profile: elevated TNF, and decreased IL‐10 and TGF‐β secretion (in vitro).	90	Decreased antigen uptake HLA‐I upregulation after HTNV challenge Increase in CCR7 in peripheral monocytes Differential expression of CD86 after the infection of low and high pathogenic viruses	47, 81, 87, 88
CD8+ T cells
Higher frequency in mild HCPS course Memory effector phenotype with Gn predominance over 4 years Not involved in pathogenesis according to the Syrian hamster model	26, 98, 99	Effector population increase during acute stage. High levels of perforin and granzyme. NP‐specific response persists more than 15 years	93, 94, 95
CD4+ T cells
Th1/Th2 profile Static mRNA levels of FOXP3. High frequency of PD‐1 in Tregs from survivors Treg Th1‐like downregulation	30, 104, 106	Higher expression of FOXP3 in severe course. Suppressor phenotype Increase in Treg population	83, 96, 100
B cells
Exacerbated PB response Higher titres of nAbs in mild disease Long‐lasting immune response IgG3 as the main subclass	25, 113, 115, 116	Long‐lasting immune response IgG1 as the main subclass	112, 114

Summary of cellular immune responses with emphasis in correlation with mild and severe course of each hantavirus cardiopulmonary syndrome (HCPS) and haemorrhagic fever with renal syndrome (HFRS).

Figure 2.

Immune mechanisms involvement in hantavirus prognosis disease. Early nAb production is broadly associated with a positive prognosis in both HCPS and HFRS cases. Further, an increase in serum TGF‐β levels has been related to mild HFRS presentation. Higher levels of the proinflammatory IL‐6 cytokine have been found in deceased patients with HCPS in comparison with survivors. Similarly, it has been associated with a more severe course of HFRS. Moreover, lung compromise could be associated with a higher expression of CCR7 during HFRS disease.

Current therapeutic approaches for treating hantavirus diseases

At the clinical level, current therapies targeting hantavirus diseases are based mostly on supportive care, with extracorporeal membrane oxygen therapy as a unique option in HCPS life‐threatening conditions. ¹¹⁹ As high nAbs titres are related to a favourable outcome, fresh frozen plasma from recovered individuals has been used as a therapy in patients with HCPS. ¹²⁰ This treatment decreased CFR from 32% to 14% in a non‐randomized multicenter trial ¹²¹ and is currently being administered to patients with HCPS, even though standardization of this intervention is still required.

Recombinant antibodies are a promising treatment for hantavirus diseases. At the practical level, the use of purified monoclonal and polyclonal nAbs has also been explored with promising results in pre‐clinical stages. Mapping and isolation of two monoclonal antibodies (mAbs) from an HCPS‐recovered individual with high nAbs titres showed high neutralizing activity in vitro and protected against lethal HCPS in Syrian hamsters when administered as a post‐infection treatment. ¹²² More recently, neutralizing mAbs targeting both Gn and Gc ANDV proteins were developed through murine hybridomas, after vaccinating mice with a VSV‐DNA vaccine expressing ANDV glycoproteins. Besides the neutralizing functions of these mAbs, they displayed antibody‐dependent cellular cytotoxicity and were also protective in the lethal ANDV Syrian hamster model. ¹²³

Additionally, purified polyclonal human IgG has been produced after DNA vaccination of transchromosomal bovines against ANDV, SNV, and, more recently, HTNV and PUUV. ¹²⁴ , ¹²⁵ All these purified polyclonal human IgG displayed high neutralizing activity and were protective against lethal HCPS. As was the case for mAbs, rationally designed human polyclonal IgGs require further testing and should be considered promising prophylaxis therapy.

Nevertheless, nAbs are not the only promising immune treatments, as some strategies employed in cancer therapies could successfully be applied to treat hantavirus diseases. IL‐6, as described above, is elevated in HCPS and HFRS, being associated with a severe outcome in HCPS. ⁶² , ⁶³ , ⁶⁴ , ⁶⁵ Moreover, the increase in other cytokines is also involved in hantavirus diseases. The cytokine storm phenomena have also been observed in other viral illness, as coronavirus disease 19 (COVID‐19). Like hantavirus diseases, severe cases of COVID‐19 are related to pneumonia with cellular infiltration and cytokine storm. ¹²⁶ Due to the global emergency and impact of COVID‐19, many therapeutic approaches have been tested simultaneously. Among them, the use of commercial mAbs to block the cytokine release syndrome is currently a strategy being tested under clinical trials to treat critical cases of COVID‐19, and successful results have been reported when a cocktail of antibodies is administered to COVID‐19‐infected patients. ¹²⁷ Along these lines, data obtained from these clinical trials could help dampen the hyperreactive response after SARS‐CoV‐2 infection and provide baseline information on whether the decrease in proinflammatory cytokines could be considered as a successful therapeutic strategy for severe hantavirus cases.

Impact of the current knowledge for the vaccine development

Up to date, there are no FDA‐approved vaccines for HFRS and HCPS. Considering the worldwide impact and the high fatality rate of both diseases, the development of protective vaccines for these syndromes is imperative. The high eco‐epidemiological complexity of these emerging infectious diseases makes it difficult to rely only on control measures to prevent human–reservoir contacts. However, this approach should be considered as a required complement to achieve a successful immunization scheme. ¹²⁸ Vaccine development for hantaviruses has been hindered by several factors, such as a lack of animal models that recap critical aspects of the illness, the limited knowledge on the immunopathology and protective immunity mechanisms, and other obstacles inherent to clinical studies. ¹²⁹ However, during the past years, notorious advances have been made with several types of vaccines at the experimental stage and, most notably, with promising DNA vaccine prototypes being tested in clinical studies. ³⁵ This section discusses vaccine approaches under development, as well as some critical missing aspects of immunity, and implications of humoral and cellular immunity for the rational design of vaccines.

Hantavirus vaccines at experimental and clinical stages

A formalin‐inactivated HTNV vaccine (Hantavax) has been used since 1990 in parts of Asia. ³⁶ Also, bivalent vaccines based on inactivated HTNV and SEOV have been used for more than 10 years in China, supported by the extended immunization programme applied in that country. ¹³⁰ , ¹³¹ These vaccines elicit nAbs titres that might last 2–3 years when used in three‐dose schedules. ¹³² Although these vaccines are considered safe, ¹³² studies assessing their efficacy are scarce, with a 10‐year retrospective cohort study failing to demonstrate the protective effect of Hantavax in Korea. ¹³³

Several strategies have been tested to develop a vaccine against one or more hantaviruses species, including purified recombinant proteins, ¹³⁴ , ¹³⁵ virus‐vectored antigens, ¹³⁶ , ¹³⁷ , ¹³⁸ virus‐like particle (VLPs) ¹³⁹ , ¹⁴⁰ and nucleic acid‐based vaccines. ¹⁴¹ , ¹⁴² , ¹⁴³ , ¹⁴⁴ , ¹⁴⁵

Studies using recombinant proteins indicate that purified antigens are enough to elicit immunogenicity and provide partial or total protection against hantavirus infection, including cross‐protective responses. ¹⁴⁶ Immunization of hamsters with recombinant viruses expressing both Gn and Gc provided protection from infection after an HTNV challenge inducing nAbs. It is noteworthy that immunization with baculovirus encoding for the NP of HTNV also protected hamsters in this model. ¹⁴⁷ Other groups showed that vaccination with recombinant PUUV NP expressed in yeast and E. coli provides protection in the bank vole model, and cross‐protective responses could be achieved using NP from other hantaviruses, including ANDV. Interestingly, the degree of protection was neither related to the amino acid sequence identity nor the cross‐reactive humoral responses of the different NPs. ¹⁴⁶ In mouse models, administration of DOBV‐NP in a three‐dose immunization scheme elicited robust and long‐lasting NP‐specific IgG antibodies with a cross‐reactive response against HNTV and PUUV. ¹⁴⁸ Similarly, a three‐dose immunization of C57BL/6 mice with DOBV‐NP provided protection from infection in 75% of challenged animals when administered with Freund's complete adjuvant but only of 12·5% when the antigen was administered with alum. When evaluating cytokine secretion of PBMCs in ELISpot assays, the immune response towards the NP antigen was associated with a strong IL‐4 secretion, suggesting that a Th2 response would not be optimal to confer protection to infection. ¹⁴⁹

Regarding virus‐vectored antigens, there are promising cross‐protective vaccine candidates aimed to protect from HCPS. Initial studies using a non‐replicating adenovirus (Ad), a single dose of Ad vaccine encoding Gn, Gc, Gn + Gc or NP individually or in combination, confer protection in the Syrian hamster model for ANDV infection. ¹³⁶ Vaccines carrying Gn or Gc were able to provide sterile immunity despite a lack of nAbs; similarly, Ad‐expressing NP also protected in the absence of nAbs. ¹³⁶ Vesicular stomatitis virus (VSV) pseudotypes encoding hantavirus glycoproteins have been explored as a vaccine approach against HFRS and HCPS. Regarding HCPS, a single dose of a VSV that expressed GPC‐ANDV (rVSVΔG/SNVGPC) provided sterilizing immunity when administered 28 days before a lethal dose of ANDV in the Syrian hamster model, with a robust nAb response. ¹³⁷

Furthermore, the cross‐reactive antibody response was observed after testing VSV expressing ANDV or SNV GPC in both lethal ANDV and non‐lethal hamster‐adapted SNV infection models. ¹³⁸ According to in vitro‐neutralizing assays on primary human ECs, the protective response elicited by these VSV‐based vaccines is associated with a humoral response that blocks the interaction between GPC and protocadherin‐1, a membrane protein that has been recently recognized as a critical host factor for ANDV entry and infection, ¹³⁸ , ¹⁵⁰ supporting the importance of that molecule for ANDV infection, and as an intervention target. Long‐term protection studies with the rVSV∆G/ANDVGPC vaccine suggest that protective immunity can last up to 6 months, but not for a year. ¹³⁷ Interestingly, this vaccine also provided protection when administered 1 week before lethal ANDV challenge, in the absence of specific nAbs, suggesting other mechanisms of immunogenicity. ¹⁵¹ Moreover, oral and intramuscular (IM) vaccination with a single dose of rVSVΔG/SNVGPC reduced viral loads in the lungs and blood of Peromyscus maniculatus. ¹⁵² Remarkably, partial protection was achieved without inducing significant nAbs, but was enough to prevent SNV infection of uninfected/vaccinated animals when directly exposed to infected deer mice. ¹⁵² These data suggest the relevance of cellular immune mechanisms involved in the protection against hantavirus infection.

Some of the most promising delivery approaches to date are nucleic acid‐based vaccines, with several candidates currently advancing to clinical trials, as recently reviewed. ³⁶ Nucleic acid vaccines have advantages over viral vector‐based vaccines, such as not interfering with pre‐existing immunity and lacking irrelevant viral antigens. Initial studies in Syrian hamsters showing increased protection after delivering SEOV‐M segment compared with SEOV‐S segment ¹⁵³ led to the development of HNTV‐M segment DNA vaccines. These candidate vaccines were tested in rodent and non‐human primate (NHP) models for hantavirus infection with promising results on disease protection and nAb responses. ¹⁴¹ Up to date, and after successful pre‐clinical studies, several DNA vaccines targeting PUUV, HTNV and ANDV are being tested in clinical trials. ³⁵ Clinical studies using a 4‐week interval with three‐dose scheme indicate that PUUV and HNTV DNA vaccines delivered through IM electroporation increase immunogenicity as compared to gene gun delivery, rising detectable nAbs titres in near of 78% of volunteers. ²⁴ Current clinical trials focus on vaccine dosage and different delivery methods to optimize immunogenicity with a successful safety profile. ³⁵ DNA vaccines have also been experimentally tested as a pan‐hantavirus vaccine candidate, showing that rabbits immunized with a plasmid mix targeting SNV, ANDV, PUUV and HNTV develop nAbs. ¹⁵⁴ Further research is required to validate this approach as up to date, no DNA vaccines have been approved for use in humans.

Cellular immune response induced by hantavirus vaccines

A common approach for determining the efficacy of some vaccines is quantifying the presence of nAbs. ¹⁵⁵ As stated before, because nAbs are enough to confer protection against hantavirus‐related diseases, the evaluation of immunogenicity of hantavirus vaccine candidates has relied mainly on this determination. However, evidence from diverse models and prototypes suggests that cellular immunity may be necessary for clearance and protective immunity against these viruses (summarized in Table 2). During the development of a PUUV DNA vaccine, cross‐protection against DOBV and ANDV was observed without the detection of nAbs, suggesting that protection could be mediated directly by immune cells. ¹⁴⁴ As discussed above, T cells, but not the neutralizing activity of antibodies, have been associated with homologous and heterologous protection against hantavirus infection in mice. ¹⁴⁶ In addition, CD8⁺ T‐cell cytotoxic activity, as well as IFN‐γ and TNF secretion, seems to be required for the clearance of HTNV. ¹⁴⁶ More recently, an HTNV‐VLP vaccine (VLP‐CD40L; VLP‐GM‐CSF) was developed, which elicited high titres of nAbs with a specific cytotoxic response mediated by T CD8⁺ cells in mice. ¹⁴⁰ Regarding HCPS, protection in the absence of neutralizing activity of antibodies has been reported after immunization with adenoviral vectors encoding for Gn, Gc or NP. ¹³⁶ Moreover, a pan‐hantavirus DNA vaccine against HTNV/PUUV/SNV/ANDV was reported to protect against ANDV infection despite low nAbs titres. ¹⁵⁴

Table 2.

Multiple strategies for the development of a hantavirus vaccine

Hantavirus vaccine candidates
Vaccine type	Vaccine/antigens	Animal model	Immunogenicity evaluation	References
Recombinant proteins	Nucleoprotein from ANDV, TOPV, DOBV or PUUV	Bank voles	Specific CD8+ T‐cell production Cross‐reactive response against PUUV	146
	Yeast‐expressed DOBV nucleoprotein	Mice	NP‐specific IgG response, with IgG1, IgG2a, IgG2b and IgG3 subclass production Th1/Th2 response Cross‐reactivity with HTNV and PUUV	134
	Truncated recombinant PUUV nucleoprotein linked to bacterial membrane protein	Mice	NP IgG response CD8+ T‐cell response	135
DNA vaccines	HTNV and ANDV M gene segments	Rhesus macaques	Neutralizing antibodies	142
	HTNV M segment	Rhesus macaques	Neutralizing antibodies Cross‐reactivity with SEOV and DOBV	141
	SNV M gene segment	Syrian hamsters	Neutralizing antibodies	154
	HTNV/PUUV/SNV/ANDV M gene segment mix	Rabbits	Neutralizing antibodies	154
	PUUV M gene segment	Syrian hamsters	Neutralizing antibodies Protection against lethal ANDV infection, without nAbs	144
Virus‐vectored	Replication‐competent VSV‐vectored ANDV glycoproteins	Syrian hamsters	Neutralizing antibodies	137
	Replication‐competent VSV‐vectored ANDV or SNV glycoproteins	Syrian hamsters	Cross‐reactive IgG response Neutralizing antibodies	138
	Non‐replicating Ad vector expressing N, Gn, Gc or Gn/Gc	Syrian hamsters (protection studies) Mice (cytotoxicity assays)	Neutralizing antibodies in SHs after challenge Specific T CD8+ cell response	136
Virus‐like particles (VLPs)	HTNV‐VLP with CD40L or GM‐CSF incorporation	Mice	Neutralizing antibodies Antigen‐specific IFN‐γ production CTL response	140
Inactivated virus	Hantavax (formalin‐inactivated HNTV)	Humans	Neutralizing antibodies B‐cell response Th1 response Cytotoxic response	156

Published approaches for vaccine development against haemorrhagic fever with renal syndrome and hantavirus cardiopulmonary syndrome.

These observations suggest that T‐cell immunity, particularly cytotoxic T‐cell activity, can be important for generating cross‐protection against hantavirus species. However, mechanistic studies should be carried out in hantavirus animal models to support these conclusions.

In humans, preliminary results of a phase 3 study with the Hantavax vaccine (NCT02553837) revealed that high responders (described as volunteers with higher nAb titres) present a differential expression gene profile related to T and NK responses, upregulating the activation markers CD69 and CD83, and the CXCR4 chemokine after the second and third boost. Additionally, after the fourth boost, Khan et al. observed a later immunoglobulin‐related signature, including the upregulation of IGLV1‐40, IGLV2‐11, IGKV3D‐20, IGLC2 and IGHG1. However, this correlation was not observed in low responders, suggesting that this differential expression gene profile might have ultimately led to an enhanced neutralizing response. ¹⁵⁶ This study indicates that several interrelated pathways of the immune response, such as phagocytosis, B cells, T cells and NK responses, are involved in the protective mechanisms induced by Hantavax.

Discussion

Despite the worldwide distribution of pathogenic hantaviruses and the constant efforts invested in vaccine development, to date there are no approved vaccines against these viruses. The only vaccine in use, which is based on an inactivated virus (Hantavax), does not provide an effective, long‐lasting immune response. ¹⁵⁷ Vaccine availability is a necessary complement to currently applied preventive actions to avoid incidences, such as the recent HCPS outbreaks in South American countries, with fatality rates above 30%. ¹⁵⁸

Many investigations have demonstrated differences between hantavirus species in many aspects, such as the biological receptors used, the immune responses induced, the clinical disease course, and fatality rates. The clinical course of HCPS could evolve faster in comparison with HFRS, and therefore, samples for longitudinal studies are not always available. A common finding in hantavirus survivors is the early nAb response. ²⁵ Furthermore, it has been demonstrated that nAbs can persist years after infection, ¹¹³ and their presence is enough to protect against lethal challenge in several animal models. For this reason, most of the therapeutic or prophylactic strategies are currently based on nAb production to provide protection. However, other humoral immune mechanisms are, in concert with the cellular immune response, an essential arm of the immune system and therefore need to be considered in evaluating immunogenicity and protection against these viruses. ¹⁵⁹

The study of hantavirus involves many challenges, starting from the critical requirement of biosafety and biosecurity standards that limit animal studies to laboratories with the highest biosafety levels, which are frequently absent in areas with elevated HCPS endemicity cases. In this line, animal models that accurately recapitulate clinical disease are insufficient for comprehensive mechanistic studies of the many pathogenic species of hantavirus. Additionally, immune response analyses in hamsters are limited due to a lack of reagents to measure specific immune populations. Thus, the diversification of animal models and the development of immunological tools for studying cellular immunogenicity are key aspects to be reached to advance towards efficient, protective vaccines against hantaviruses.

Studies on the hamster model of HCPS have suggested that the immunopathogenesis of HCPS might not be related to T cells ⁹⁹ ; however, analysis of human cases indicate that T CD8⁺ cells could be involved in the viral clearance without nAb participation, ²⁷ and similar results are observed in the cross‐protection against different hantavirus species. CD8⁺ memory cells maintain an effector phenotype years after infection, ⁹³ , ⁹⁸ and thus, vaccine studies should consider the response of effector memory profiles in the design of future vaccine candidates. Remarkably, cross‐reactivity between different species of hantaviruses has been observed in immunization studies indicating that a pan‐hantavirus vaccine could be a feasible, successful approach.

Disclosures

The authors declare no conflict of interest.

Author contributions

Farides Saavedra and Alexis M. Kalergis performed conceptualization. Farides Saavedra, Fabián E. Díaz, Angello Retamal‐Díaz and Camila Covián wrote the original draft. Camila Covián, Angello Retamal‐Díaz, Farides Saavedra and Fabián E. Díaz prepared figures and tables. Farides Saavedra, Fabián E. Díaz, Camila Covián, Angello Retamal‐Díaz, Pablo A. González and Alexis M. Kalergis wrote, reviewed and edited the manuscript.

Data availability statement

No new data sets were generated for this manuscript.