Age has a role in driving host immunopathological response to alphavirus infection

Summary

Alphaviruses are a group of arthropod‐borne pathogens capable of causing a wide spectrum of clinical symptoms, ranging from milder symptoms like rashes, fever and polyarthralgia, to life‐threatening encephalitis. This genus of viruses is prevalent globally, and can infect patients across a wide age range. Interestingly, disease severity of virus‐infected patients is wide‐ranging. Definitions of the pathogenesis of alphaviruses, as well as the host factors influencing disease severity, remain limited. The innate and adaptive immune systems are important host defences against alphavirus infections. Several reports have highlighted the roles of specific immune subsets in contributing to the immune pathogenesis of these viruses. However, immunosenescence, a gradual deterioration of the immune system brought about by the natural advancement of age, affects the functional roles of these immune subsets. This phenomenon compromises the host's ability to defend against alphavirus infection and pathogenesis. In addition, the lack of maturity in the immune system in newborns and infants also results in more severe disease outcomes. In this review, we will summarize the subtle yet diverse physiological changes in the immune system during aging, and how these changes underlie the differences in disease severity for common alphaviruses.

Abbreviations

Aβ

amyoid β

AD

Alzheimer's disease

CHF

chronic heart failure

CHIKV

Chikungunya virus

dpi

days post infection

G‐CSF

granulocyte colony‐stimulating factor

IFN

interferon

IL

interleukin

MCP

monocyte chemoattractant protein

M‐CSF

macrophage colony‐stimulating factor

MIF

macrophage migration inhibitory factor

NK

natural killer

PFU

plaque‐forming unit

RANTES

Regulated on Activation, Normal T Cell Expressed and Secreted

RRV

Ross River virus

SINV

Sindbis virus

TNF

tumour necrosis factor

Introduction

Alphaviruses are small, enveloped pathogens belonging to the family Togaviridae. They are primarily transmitted by arthropods, with a majority of the vectors being mosquitoes.1 In addition to being human pathogens, they are also found to take part in enzootic transmission cycles in a variety of birds and mammals.2 The positive single‐stranded RNA genome of alphaviruses is typically made up of an untranslated region, flanked by two coding regions encoding for the non‐structural and structural proteins, respectively.1, 3 These proteins undergo different post‐translational pathways, eventually assembling into progeny nucleocapsids and budding from the cell membrane or intracellular vesicles to form enveloped viral progenies.1

This genus of viruses contains approximately 30 members, and can be categorized into Old World and New World alphaviruses, based on their origin of discovery.1 Although they are already widely distributed throughout the world, variances in disease severity can be attributed to the wide geographical distributions and evolutionary trajectories in the past. This may result in unique adaptations to a variety of vectors and hosts, as well as differences in pathogenesis.4 Old World alphaviruses, including Chikungunya virus (CHIKV), Sindbis virus (SINV), Ross River virus (RRV) and Semliki Forest virus typically cause mild complications such as arthralgia, rash and febrile illness, and are rarely life‐threatening.5, 6 On the other hand, new world alphaviruses, like the Eastern equine encephalitis virus, Venezuelan equine encephalitis virus and Western equine encephalitis virus, cause more severe life‐threatening disease complications such as encephalitis.7, 8

Manifestations of infection

CHIKV is an Old World alphavirus responsible for chikungunya fever, a febrile disease characterized by the hallmark acute joint pain, and in some patients, chronic debilitating polyarthralgia.9, 10 The virus is mainly transmitted by Aedes aegypti or Aedes albopictus mosquitoes.11, 12, 13 It was first recognized as a human pathogen in Tanzania, East Africa, in 1952, during an outbreak involving 115 hospitalized patients.14, 15 This outbreak gave rise to the name ‘Chikungunya’, which means ‘which bends up’ in Makonde, a language of southern Tanzania,16 drawing reference to the contorted postures of infected patients due to severe joint pains.17 The virus re‐emerged in 2004, when a major epidemic in Kenya started a rapid global transmission of CHIKV from Africa towards the eastern hemisphere, including islands in the Indian Ocean, Europe, Asia and then to the Americas.16, 18, 19, 20 The outbreak in Réunion island was most noteworthy, with approximately one‐third of its population infected with CHIKV. This epidemic was also associated with the first reported case of neurological complications and mortality, as well as vertical transmission from mother to newborns.21, 22, 23, 24

Another arthritogenic alphavirus, RRV, is known to contribute significantly to infectious rheumatic disease worldwide.25 Other RRV‐induced complications include fever, rash, myalgia, and pain and stiffness in the joints.26 This infectious agent is also a concern, as it has the potential to emerge and cause major epidemics globally. It caused an epidemic in the South Pacific in the 1970s, with more than 60 000 infected patients. RRV subsequently re‐emerged in Fiji in 2003, after the 1979–1980 outbreak.27

Sindbis virus is another arthropod‐borne alphavirus with genomic and structural properties similar to those of CHIKV.1 It was first isolated from a Culex mosquito in Sindbis, a village near Cairo in 1952.28 Symptoms of SINV infection were poorly described until 1961, when the first symptomatic case of SINV was reported.29 It remained a minor public health problem until larger outbreaks were recorded in South Africa.30 The virus was also recognized as the causative agent of Pogosta disease (Finland), Ockelbo disease (Sweden) and Karelian fever (Russia) in the Northern European countries in the mid‐1980s.31, 32, 33 Like CHIKV, SINV infection in humans typically results in febrile illness, rash, malaise and acute arthralgia.6

Serological surveys indicate that exposure of the human population to alphaviruses is not uncommon.34, 35 Fortunately, infected patients are usually asymptomatic and recover without giving rise to any observable clinical outcomes.34, 36 However, disease severity in symptomatic patients during the prodromal and acute phases, as well as the frequency of central nervous system involvement, mortality rates and permanent neurological complications in survivors, are disproportionately higher in children and the elderly.6, 37, 38, 39, 40 During the natural aging process from infancy to old age in humans, the host immune system undergoes a subtle yet progressive change – immunosenescence.41, 42 Although many studies have identified the possible effects of aging on the immune system, few have associated these effects with alphavirus disease severity. In this review, age‐related effects on the immune system and the consequential influence on pathogenesis of alphaviruses will be discussed.

Differential efficiency of innate antiviral responses in different age groups

Newborns and infants are at high risk of infection. However, susceptibility to infection decreases from pre‐term to full‐term newborns, and drops further in infancy.43, 44 At birth, newborns have a relatively immature immune system, so are highly susceptible to bacterial and viral infections.45, 46 The innate immune responses are increasingly recognized to be the most important defence in newborns against infections,47 but emerging evidence highlights the deficiency of such responses in newborns and pre‐term infants.45, 48 The elderly are also highly susceptible to infection, due to the progressive changes of the immune system during aging, a term known as immunosenescence.49, 50 There have been multiple reports on the extensive effects of aging on adaptive immunity.51, 52, 53 The importance of the changes in innate immunity with age is also gaining recognition.41, 54, 55

Differences in efficiency of the innate and adaptive immune systems during these two stages of life can lead to striking differences in clinical observations during alphavirus infections. Although the underlying mechanistic differences are still not clear, functional differences of the components in the respective immune systems could play a critical role in influencing alphavirus disease severity.

Maturity of the immune system affects alphavirus disease severity

The state of immune system maturity significantly affects the risk of alphavirus infection and severity of disease.43, 44, 45, 46 To study the pathophysiology of CHIKV infection and uncover the complicating factors affecting disease pathology, many groups have developed different infection models to study disease development in vivo. These groups independently identified age as a crucial factor affecting CHIKV disease severity. One early study reported that the age of the mice infected with CHIKV greatly affected disease phenotype.56 Neonate mice aged 6, 9 and 12 days were infected with CHIKV through the intradermal route and observed for lethality. Both 6‐day‐old and 9‐day‐old infected mice suffered from flaccid paralysis by 7 days post infection (dpi). In addition, 6‐day‐old animals succumbed to CHIKV infection by 12 dpi, whereas 12‐day‐old neonates did not suffer from any observable morbidities or mortality. The severity of disease was also correlated with higher viral loads in various organs, as well as viral dissemination to the CNS.56 Importantly, this correlation of disease severity with age was further observed in CHIKV‐infected patients, where a large proportion of patients < 20 years old had relatively higher viral load, especially for newborns (< 1 year old).57 In contrast, most patients between 20 and 60 years old had relatively lower levels of viraemia.57

Extensive bullous lesions, an unusual and severe cutaneous form of the disease, were also observed in infants from the outbreaks in Réunion Island and South India.58, 59 In another report by the Institut de Veille Sanitaire, 37 of 224 reported emerging paediatric cases during the Réunion Island outbreak had bullous dermatitis,38 compared with 3% of the reported adult cases. It is also interesting to note that most of the emerging paediatric cases suffering from lesions were mostly infants < 1 year old.

These differential disease outcomes seen in children, especially infants, could be attributed to different immune responses against CHIKV. Indeed, meta‐analysis studies of the paediatric cohort from the 2009 Sarawak, Malaysia outbreak and the adult cohort highlighted higher levels of several immune mediators compared with the adult cohort.37 Higher levels of immune mediators include pro‐inflammatory cytokines [tumour necrosis factor‐β (TNF‐β), tumour necrosis factor‐related apoptosis‐inducing ligand (TRAIL), interleukin‐5 (IL‐5), growth‐related oncogene‐α (GRO‐α), IL‐18, interferon‐α2 (IFN‐α2), interleukin 2 receptor alpha (IL‐2Ra)], chemokines [macrophage migration inhibitory factor (MIF), MIG (CXCL9), monocyte chemoattractant protein 3 (MCP‐3), granulocyte colony‐stimulating factor (G‐CSF)], growth factors and others (stem cell growth factor‐β, macrophage colony‐stimulating factor (M‐CSF), hepatocyte growth factor, stem cell factor, leukaemia inhibitory factor, IL‐3). However, lower levels of IL‐1β, Regulated on Activation, Normal T Cell Expressed and Secreted (RANTES), stromal cell‐derived factor‐1α, and β‐nerve growth factor were also detected in children.37 The more robust pro‐inflammatory response in children could create an anti‐viral environment, so virus clearance would be faster. However, this enhanced innate response could also lead to excessive inflammation and other disease complications like bullous lesions as seen in paediatric cases during the Réunion Island outbreak.38

Immaturity of the immune system could also potentially aggravate RRV pathogenesis. The establishment of an in vivo infection model for the study of RRV was first reported in consecutive articles in 1973.60, 61 Using the prototype T48 strain of RRV, the authors described the pathological in vivo studies in two parts: one focused primarily on RRV‐associated brain pathology,60 while the other focused on pathogenic effects on muscle, periosteum, perichondrium and brown fat.61 Both reports highlighted a marked age‐dependence on RRV‐induced pathology. Upon intracerebral injection of 3000 plaque‐forming units (PFU) of RRV, 87·5% and 13·3% of 1‐day‐old and 3‐day‐old mice developed hydrocephalus, respectively. However, when RRV infection was performed in 5‐day‐old mice, none suffered from hydrocephalus.60 In addition, mice showed a gradual fall in RRV susceptibility, as older mice had lower peak RRV titres in thigh muscles compared with younger mice.61 A subsequent study reported an adaptation of RRV field isolates into the aforementioned in vivo infection model.62 Using this adapted model, the study showed the effects of age on the severity of RRV disease. RRV replication was detected in both serum and muscle of infected 1‐week‐old neonatal mice at 1 dpi. These mice developed myositis, muscle necrosis and weakness.62 However, when similar infection was repeated in 4‐week‐old mice, no clinical signs were observed.62 This age‐related difference in disease severity was also consistent in later studies. RRV infection in 15‐day‐old C57BL/6 mice resulted in severe hind limb dysfunction, and 100% mortality about a week later.63 In contrast, infection in 24‐day‐old mice resulted in similarly severe morbidity, but appeared to completely recover from the disease by 30 dpi, as indicated by the weight gain and absence of observable disease signs.63

The pathogenesis of SINV and its associations with age have also been extensively shown in various mouse models. SINV infection causes the development of acute encephalitis with an age‐dependent mortality rate.64, 65, 66, 67 Infection with 1 PFU of SINV was lethal in 1‐day‐old mice. However, 22‐day‐old weanlings did not succumb to SINV infection, even with an inoculum as high as 10⁹ PFU.67 Age‐dependent resistance to SINV infection is also shown in other studies. In a SINV pathogenic mouse model of various ages, from 1 to 8 weeks old, mice were infected with SINV subcutaneously.66 Mice from all age groups presented transient viraemia, and invasion into the brain, although greater quantities of virus were found in younger mice. These mice also showed age‐dependent resistance to SINV‐associated mortality. All 1‐week‐old mice succumbed to fatal SINV‐associated encephalitis. However, only 28% of 2‐week‐old mice, and none of the mice 4 weeks old or older died as a result of SINV infection.66 A subsequent report showed a possible role of antibody‐mediated mechanisms in clearing infectious virus from neurons in older animals.64

The different mouse models used to study the associations between the maturity of immune system and alphavirus disease severity can be found in Table 1.

Table 1.

Age‐dependent resistance to alphavirus infection in mouse models

Virus	Route of inoculation	Strain of mice	Age	Pathology	Mortality	References
CHIKV	Intracerebral	Albino Swiss mice	6‐ to 12‐day‐old	–	100% at 4 dpi to 7 dpi	Ross68
	Intracerebral	Albino Swiss mice	8‐day‐old	Ruffled fur, lethargy, paralysis	100% at 6 dpi	Suckling et al.69
	Intradermal	C57BL/6J	6‐, 9‐ and 12‐day‐old	6‐ and 9‐day‐old: flaccid paralysis at 6 to 7 dpi.	6‐day‐old: 100% at 12 dpi	Couderc et al.56
					9‐day‐old: 33% at 15 dpi
				12‐day‐old: no morbidity	12‐day‐old: 0%
	Subcutaneous (back)	ICR, CD‐1	2‐ to 3‐day‐old	Lethargy, loss of balance, difficulty walking, alopecia, decreased weight gain	ICR:8%CD‐1: 17%	Ziegler et al.70
			14‐day‐old	Restricted virus replication in the brain	–
	Subcutaneous (joint)	C57BL/6J	6‐week‐old	Self‐limiting arthritis, tenosynovitis and myositis	0%	Gardner et al.71
	Subcutaneous (joint)	C57BL/6J	14‐day‐old	Swelling and oedema, inflammatory lesions, severe chronic active tenosynovitis, necrotizing myositis	0%	Morrison et al.72
RRV	Intracerebral	Walter and Eliza Hall Institute (WEHI) strain	1‐day‐old	87·5% Hydrocephalus (3000 PFU)	30% at 10 dpi	Mims et al.60
			3‐day‐old	13·3% Hydrocephalus (3000 PFU)	–
			5‐day‐old	0% Hydrocephalus (3000 PFU)	–
	Subcutaneous	Walter and Eliza Hall Institute (WEHI) strain	1‐day‐old to 56‐day‐old	Decreasing RRV titres in the thigh muscles	–	Murphy et al.61
	Subcutaneous	BALB/c	7‐day‐old	Myositis, muscle necrosis, weakness	–	Seay et al.62
			4‐week‐old	No clinical symptoms	–
	Subcutaneous (joint)	CD‐1	14‐ to 15‐day‐old	Severe hind limb dysfunction, loss of grip strength, failure to gain weight	16% at 10 dpi	Morrison et al. 200663
		C57BL/6	15‐day‐old		100% at 7 dpi
		C57BL/6	24‐day‐old		0%
SINV	Intracerebral	Outbred conventional white mice	1‐day‐old	–	10 PFU: 100%1 PFU:75%	Reinarz et al.67
			22‐day‐old	–	104 PFU: 0%105 PFU: 22%
	Intracerebral	Outbred Swiss mice	2‐day‐old	–	100% at 2 to 3 dpi	Johnson et al.65
		BALB/c	4‐week‐old	–	0%

CHIKV, Chikungunya virus; dpi, days post infection; PFU, plaque‐forming unit; RRV, Ross River virus; SINV, Sindbis virus.

Changes in innate immune system during immunosenescence

The innate immune system comprises multiple leucocytes, chemical messengers and mechanisms that offer the first line of defence against any invading exogenous substance.73 It is made up of a myriad of major myeloid components: granulocytes, monocytes/macrophages and dendritic cells.74 Over the last decade, the natural killer (NK) cell, an innate‐like lymphocyte, has been demonstrated to play a major role in host rejection of virus‐infected cells.75, 76, 77

Neutrophils, the most abundant type of leucocyte in humans, form a quintessential part of the innate immunity.78, 79 However, neutrophils have also been demonstrated to play a pathogenic role in alphavirus infection. CHIKV infection in CCR2‐deficient mice, where absence of CCR2, a chemokine receptor central to the monocyte/macrophage‐dominated inflammatory arthritic infiltrates, resulted in high neutrophil levels. This increase then led to more severe disease pathology, but did not affect the control of viraemia.80 The increased levels of neutrophil infiltration also resulted in elevated levels of several inflammatory mediators including CXCL1, CXCL2, G‐CSF and IL‐1β.80 This could have caused exacerbated virus‐induced joint inflammation and cartilage damage during CHIKV infection.80

Aging can lead to deteriorations in neutrophil effector functions, including chemotaxis, intracellular killing, phagocytosis and granulocyte–macrophage colony‐stimulating factor and triggering receptor expressed on myeloid cells 1 signalling.81, 82, 83, 84, 85 These changes in effector functions could be due to the alteration of signalling pathways with age.86, 87 Diminished efficiency of chemotaxis not only affects the neutrophil migration to the site of infection, but also results in defective neutrophil egress from inflamed tissues.88, 89 Inefficient neutrophil egress could lead to excessive localized inflammation as seen in CHIKV‐infected CCR2^−/− mice.80

Deterioration in neutrophil function with age could affect CHIKV pathogenesis, but neutrophil roles in RRV and SINV infection are less studied. However, it was shown that RRV infection in patients led to neutropenia.90 The degradation of neutrophil function during immunosenescence, coupled with neutropenia during RRV infection, could result in less effective anti‐viral response and more severe pathology in the elderly.

Monocytes and macrophages are agranular leucocytes arising from haematopoietic stem cells in the bone marrow.91 They are components of the innate system, and have been reported to circulate in peripheral blood and be recruited to tissues following infection.92 They are identified to be the main reservoirs of CHIKV, RRV and SINV infection.93, 94, 95 Macrophages have also been highlighted as the central players in alphavirus infection and persistence during the chronic phase of the disease.96 Upon exposure to CHIKV and RRV, these infected myeloid subsets trigger a robust and rapid anti‐viral response with the production of specific cytokines and chemokines, including IL‐8, macrophage inflammatory protein‐1β, RANTES, IL‐1Ra, IFN‐γ‐induced protein 10 (IP‐10), IFN‐α and IL‐12.93, 97, 98, 99 Similarly, infections in macaques have also shown a sharp increase in IFN‐α/β, MCP‐1 and IL‐6 during the acute phase of CHIKV infection.96

Monocytes and macrophages have also been implicated as the primary mediators of muscle damage in RRV infection. RRV antigens have been detected in the synovial monocytes soon after the onset of epidemic polyarthritis in RRV‐infected patients.94 Subsequent studies identified macrophages as primary mediators of muscle pathology, as absence of macrophages in mice treated with macrophage‐toxic agents completely abrogated RRV disease symptoms.100 Massive influx of macrophages into the site of virus inoculation was also detected in untreated mice, further supporting its pathogenic role in RRV infection.100

Monocytes and macrophages have also been demonstrated to play a pathogenic role in SINV infection. In a particular study, SINV infection in human macrophages promoted macrophage activation and led to the release of MIF, TNF‐α, IL‐1β and IL‐6. The release of these cytokines up‐regulated matrix metalloproteases 1 and 3, which could be involved in articular damage.95

Monocytes and macrophages cause age‐related deterioration in cytotoxicity, intracellular killing and antigen presentation.55, 101, 102 Encounter with pathogens results in the activation of macrophages, switching it to a pro‐inflammatory phenotype. This dynamic switch in inflammatory response is influenced by macrophage modulating ligands present in the tissue environment.103, 104 However, the dysregulation of macrophage plasticity and function due to imbalance of macrophage modulating ligands due to age could result in a less efficient anti‐viral innate response,105 leading to a more severe CHIKV‐, RRV‐ and SINV‐associated disease phenotype as seen in the elderly.

Natural killer cells, another crucial component of the innate immune system, play an indispensable role against virus infections.77 They are a class of lymphocytes that do not have antigen‐specific receptors, unlike the T cells and B cells.106 The involvement of NK cells during alphavirus infection has been reported in many studies. In the case of CHIKV, NK cells have been shown to play a pathogenic role in in vivo infection models.107 Mice infected with the LR2006 OPY1 CHIKV isolate from the Réunion Island outbreak in 2007, compared with the CNR20235 CHIKV isolate from the Caribbean Island outbreak in 2013, induced higher NK activity, higher levels of IL‐6 and other pro‐inflammatory mediators, such as IFN‐γ, TNF‐α, IP‐10, RANTES, soluble receptor activator of nuclear factor‐κB ligand, GRO‐α, MCP‐1 and IFN‐α. When mice depleted of NK cells were infected with CHIKV, a less severe footpad swelling was observed in the early acute phase of infection (2–3 dpi). The absence of NK cells was postulated to result in the decrease of CD11b⁺ cell activation, leading to a down‐regulation of IL‐6 and other pro‐inflammatory mediators.108

In RRV infection, RRV‐infected patients showed a decrease in NK cell activity during the acute phase of the disease.109 NK cell activity then returned to normal levels when the RRV‐infected patient recovered from RRV‐associated arthritis.109 RRV infection in a mouse model also showed NK cell infiltration into the skeletal muscle tissues at 5 dpi.63

In contrast to CHIKV and RRV infection, NK cells do not appear to play any role in SINV infection.110 NK cells were induced in both young and older mice during SINV infection, suggesting that they were not important in age‐related resistance to fatal SINV infection. SINV infection in NK‐cell‐deficient mice also did not show differences in SINV pathogenesis, suggesting that NK cells may not be important in the recovery of mice from SINV infection.110

Like other cell subsets in the innate immune system, NK cell number, function and phenotype are altered during aging. The level of CD56^dim mature NK cells is highly increased in the elderly, whereas there is a decrease of the CD56^bright immature NK cell subpopulation.111, 112, 113, 114 As NK cells could potentially play a pathogenic role in humans during CHIKV infection, an increase in the differentiated NK cell population in the elderly could result in more severe CHIKV pathology.107 Interestingly, NK cell cytotoxicity in elderly individuals was not changed,115 despite the increase in CD56^dim mature NK cells. This suggests an impairment of NK cell cytotoxic activity at the single‐cell level.116 The deterioration of NK cell function during RRV infection in the elderly could lead to a less efficient resolution of the virus and hence to aggravated pathology.

Changes in adaptive immunity with age

The adaptive immune system has two critical yet unique features: a large repertoire of naive lymphocytes capable of recognizing any potential foreign antigens; and long‐lived memory lymphocytes that provide a rapid and robust antigenic response to any previously encountered pathogens. Due to immunosenescence, the generation of naive lymphocytes will decline,42, 117 coupled with poor maintenance of memory lymphocytes that lowers the efficiency of adaptive immunity.52, 118 As such the elderly typically demonstrate an increased risk of morbidity and mortality when infected with alphaviruses.38, 119, 120, 121, 122, 123

The most prominent changes linked to aging are the deterioration of different components of adaptive immunity.124, 125, 126, 127 As adaptive immunity and antibody responses are important in the resolution of alphavirus infection,95, 128, 129, 130 the deterioration of adaptive immunity in the elderly would potentially lead to more severe pathology and chronic complications. Indeed, when a cohort study was performed on CHIKV‐infected patients from the 2006–2007 Réunion Island outbreak, 78% of the patients with chronic arthralgia were elderly (> 60 years old).131 In addition, these patients were associated with higher viral load during the time of referral to the clinic.131 The patients with chronic arthralgia also had weak T helper type 1 and 2 systemic cytokine responses, signifying a less activated adaptive immune response.131 Interestingly, a decrease in naive T‐cell numbers has been associated with aging.124 In addition, reports have also shown that naive CD4⁺ T cells isolated from both older people and mice are less responsive towards stimulation.125, 127 Moreover, the cytokine secretion profiles of these naive CD4⁺ T cells showed significant differences when compared with those isolated from younger hosts.125 The less efficient naive CD4⁺ T‐cell helper function also resulted in less B‐cell stimulation and reduced antibody production.125 In contrast, memory CD4⁺ T cells that were generated at a young age remain relatively competent with age.124 However, these memory CD4⁺ T cells offered a poor response to stimulation.126

An efficient and specific humoral response is also critical for the clearance of virus infection.132 The need for an alphavirus‐specific antibody response for controlling pathogenesis and mediating viral clearance has been demonstrated by many studies.128, 130, 133, 134, 135 Serological studies of CHIKV‐infected patients showed the importance of neutralizing antibody responses to control CHIKV pathogenesis.136 Neutralizing antibody C9 was isolated from a CHIKV‐infected and recovered individual. It was found to recognize and bind to amino acid 162 of the CHIKV E2 glycoprotein. The region in which C9 binds is an acid‐sensitive region, and is important for rearrangement of E2 during envelope fusion and viral entry into host cells.136 When administered prophylactically or therapeutically in a CHIKV pathogenic mouse model, C9 also completely protected CHIKV‐induced viraemia and arthritis. This highlights the importance of an efficient neutralizing antibody response in protecting against CHIKV infection. A subsequent cohort study on CHIKV‐infected patients of the Singapore outbreak in 2008 also identified key linear epitopes of the antigenic response, including the previously identified E2 glycoprotein, and linear regions of E3 glycoprotein, C‐terminus of capsid region and the nsP3 protein.134, 137 An in vivo infection study also showed the importance of a CHIKV‐specific antibody response in the clearance of CHIKV infection. Viraemia in CHIKV‐infected B‐cell knockout (μMT) mice persisted for more than a year, emphasizing the importance of B cells in mediating CHIKV clearance.128 In addition, mice with absence of B cells had more severe joint pathology during the acute phase of the disease, potentially due to higher CHIKV load during the acute phase of infection.128

In SINV, active neutralizing antibodies were shown to protect against intracerebral SINV infection in neonatal A129 mice, which lack the receptor for type I interferons. SINV infection resulted in 100% mortality after 6 dpi.130 However, when these mice were administered G5, a neutralizing antibody specific for SINV E2 glycoprotein, none of the animals succumbed to SINV infection.130 Viral titres from serum were also significantly lowered.130 This highlights the role of SINV‐specific antibodies in controlling SINV replication, and the importance of the humoral response in protecting the host against SINV‐induced mortality. Using the SINV infection pathogenic mouse model, antibody‐secreting cells were shown to peak in draining cervical lymph nodes during the convalescent phase of infection (5–7 dpi), with increasing SINV‐specific antibodies in the brain from 7 dpi. In addition, memory B cells were also present during later time‐points after infection, suggesting a role of the humoral response in controlling persistence of viral RNA in the brain.138

The importance of neutralizing antibodies in the control of RRV pathogenesis is less studied. However, in a pre‐clinical study, the vaccination of mice against RRV induced antibodies that were protective against subsequent RRV infection. These antibodies were later determined to be neutralizing against RRV isolates from different sources, although the degree of neutralization was different against different isolates.135

There have been associations of aging with the impairment of specific antibody responses,52, 53, 117, 127 which could account for the poor viral resolution and more severe alphavirus‐induced pathology in the elderly.38, 119, 120, 121, 122, 123 Reports have attributed the suboptimal humoral immune response to aging‐related changes in affinity, specificity and class of antibodies produced.139 The number and size of germinal centres are also shown to decrease with age,51 and this results in the consequential impairment of germinal centre functions and antibody maturation reactions like antibody class switch recombination, and somatic hypermutation.140, 141 Hence, antibody affinity maturation will be adversely affected, in addition to the decrease in recirculating antibody‐secreting plasma cells in the bone marrow.142 The adaptive immune response is critical in limiting alphavirus pathogenesis during infection, as well as mediating the neutralization and clearance of these pathogens. Due to the progressive degeneration of humoral response with age, the level and efficiency of antibody response may not be sufficient in dealing against alphavirus infection, leading to more severe disease complications as observed in aged individuals.53

Effects of immunosenescence in other age‐related diseases

Alphavirus‐induced infections are not the only forms of disease to have their outcomes affected by immunosenescence. There have been countless other diseases that have been known to be associated with the natural ageing process. The effect of age on neurodegenerative disease, cardiovascular disease and diabetes will be discussed for illustrating the significant effects of immunosenescence.

Alzheimer's disease (AD) is one of the most common neurodegenerative diseases. A type of dementia, it is associated with a clinical syndrome involving progressive loss of memory, distressing behaviour irregularities like hallucination and aggressiveness, and problems with sleeping and eating. Recent reports have established the link between age‐related occurrence of microglia‐mediated low‐grade inflammation and the neurodegenerative disease.143, 144, 145, 146 It is proposed that microglia, in their normal state, delay the progression of AD by mediating the clearance of amyloid‐β peptide (Aβ). These leucocytes also secrete neuroprotective growth factors and anti‐inflammatory cytokines.147 However, it is hypothesized that these resident macrophages gradually lose their ability to carry out these neuro‐protective functions and convert to a hyper‐reactive pro‐inflammatory state during immunosenescence,143 causing neuronal damage in its own right. Other reports have suggested that aging causes the microglia to lose their ability to degrade Aβ.148 In turn, the aggregating Aβ binds and stimulates the microglia via pattern recognition receptors to produce even more inflammatory cytokines,145, 149, 150 worsening the extent of neuronal damage. Higher incidence of encephalitis has been reported in the elderly infected with New World alphaviruses.151, 152 The underlying cause might be due to inefficient virus control in the brain by functionally altered microglia. Associations between immunosenescence and alphavirus‐induced encephalitis will be an area of interest in the coming years.

In addition, it was shown that the aggregation of Aβ has implications on adaptive immune leucocytes as well. Aβ primes T cells into an autoreactive phenotype that destroys neurons directly, or mediates neuronal destruction by releasing IFN‐γ. It was also shown that patients suffering from milder forms of AD have greatly reduced number of naive T cells and elevated memory T‐cell subsets. Increased proportions of CD4⁺ and CD8⁺ T cells also lacked the expression of CD28, a co‐stimulatory receptor important for T‐cell activation and survival. These findings suggest a dysregulated adaptive immunity in patients with AD, which can contribute to the pathogenesis of AD.153

Chronic heart failure (CHF), a form of cardiovascular disease prevalent in the elderly, has also been associated with a state of chronic inflammation.154 High levels of pro‐inflammatory cytokines: TNF, IL‐1β and IL‐6 have been detected in patient myocardium and peripheral tissues.155 Furthermore, it was shown that pro‐inflammatory cytokine levels in the plasma affect clinical outcomes, as patients with higher levels of IL‐6 were associated with worse CHF pathogenesis.155 In addition, the expression of Toll‐like receptor‐4 (TLR4) receptor, as well as TLR4‐associated pro‐inflammatory cytokines, were shown to be increased in hearts of patients with CHF.156 The contribution of TLR4 expression towards CHF pathogenesis was also reported in another finding, with increased levels of TLR4 expression on monocytes in patients with CHF.157 Moreover, the use of fluvastatin, a statin class drug, was also able to reduce the pro‐inflammatory innate immune response by inhibiting monocyte TLR signalling.157 Such comparisons could be made with the severe CHIKV‐induced pathologies observed in the elderly.

Diabetes mellitus type 2, one of the most common diseases in the elderly, is another disease associated with low‐grade chronic inflammation.158, 159, 160, 161 This long‐term metabolic disorder is characterized by high blood sugar, insulin resistance and lower levels of insulin. Like AD and CHF, elevated levels of pro‐inflammatory mediators such as IL‐6, TNF and C‐reactive protein are found in patients with type 2 diabetes,162 closely resembling inflammatory profiles observed in CHIKV‐infected elderly patients.

Conclusions

The innate and adaptive immune systems are evolved in vertebrates to protect and survive novel and recurrent infections throughout their lifespan. However, subtle changes in the immune system occur during the lifetime of an individual and contribute significantly to the development of diseases and their complications in humans. Although many comprehensive studies have been performed to elucidate the associations between aging and the disease development of age‐related neurodegenerative, cardiovascular and metabolic diseases, significantly less work of similar nature has been done for alphaviruses and their associated disease complications. Immunological studies of CHIKV, RRV and SINV in mouse models as well as patient samples reviewed several age‐dependent differences, which could influence disease severity and host recovery. It is important for future age‐related studies on alphaviruses to determine the array of age‐related changes that contribute to differential disease outcomes, including the shift to chronic alphavirus disease. This will aid in the design of age‐dependent anti‐viral and therapeutic strategies against alphaviruses.

Disclosure

The authors declare no conflict of interest.