The Life Cycle of the Alphaviruses: From an Antiviral Perspective

Abstract

The alphaviruses are a widely distributed group of positive-sense, single stranded, RNA viruses. These viruses are largely arthropod-borne and can be found on all populated continents. These viruses cause significant human disease, and recently have begun to spread into new populations, such as the expansion of Chikungunya virus into southern Europe and the Caribbean, where it has established itself as endemic. The study of alphaviruses is an active and expanding field, due to their impacts on human health, their effects on agriculture, and the threat that some pose as potential agents of biological warfare and terrorism. In this systematic review we will summarize both historic knowledge in the field as well as recently published data that has potential to shift current theories in how alphaviruses are able to function. This review is comprehensive, covering all parts of the alphaviral life cycle as well as a brief overview of their pathology and the current state of research in regards to vaccines and therapeutics for alphaviral disease.

1. Introduction

Alphaviruses are positive sense, single stranded, RNA viruses in the family Togaviridae, which are classified as members of the domain Riboviria¹. The alphaviruses currently encompass more than thirty members that infect a wide range of host and vector species, both terrestrial and aquatic. These viruses are widely dispersed geographically as well, with at least one alphavirus being present on every populated continent^2–6. Alphaviruses are continually emerging into naïve populations, and there are currently no licensed treatments or vaccines for alphaviral disease, making these viruses important research targets.

The alphaviruses have historically been divided into two clades based upon the location of their isolation. The Old-World viruses, which were first isolated in the eastern hemisphere, and the New-World viruses which were first discovered in the Americas. The Old-World viruses generally cause arthralgia and fever, with some also causing a rash⁷. There is some recent evidence that Old-World members may be able to cause encephalitis as well, with the adaptation of neurologically invasive Sindbis virus (SINV) for use in mice, as well as its association with rare cases of viral encephalitis in Europe^8–10. The New-World viruses are known to cause encephalitis; the three most prominent members of this clade, Eastern (EEEV), Western (WEEV), and Venezuelan Equine Encephalitis viruses (VEEV) demonstrate high levels of neurological pathogenicity⁷. Recently it has been proposed that certain South American clades of EEEV be split into the new species Madariaga virus, as they are genetically distinct and are less pathogenic¹¹

Continuing efforts of alphavirus discovery, characterization, and sequencing have now indicated that the division between the Old and New-World viruses may be losing usefulness, as there are now several known New-World alphaviruses that don’t cause any disease¹², as well as the recent discovery of alphaviruses that are native to the New-World but have disease phenotypes that are similar to the Old-World clade, such as Mayaro virus^2,13.

2. Alphaviral Disease

The Old-World virus of most concern is Chikungunya virus (CHIKV), which has recently expanded into naive populations across Asia, southern Europe, and most dramatically, in the Caribbean^3–6. This has resulted in CHIKV becoming endemic in several of these regions. The primary risk of CHIKV is a sustained arthralgia that can last for months, with one study in Mexico indicating that over a third of confirmed CHIKV cases have arthralgia twelve months after acute disease¹⁴. Similar pathologies have also been reported after infection with Ross River virus¹⁵.

The New-World alphaviruses generally cause more severe disease than the Old-World viruses; however, the three most common (EEEV, WEEV, VEEV) are noted for a high rate of asymptomatic infection¹². This asymptomatic infection rate varies between the three viruses and in the two primary populations of interest, equids and humans, with equids having significantly higher rates of symptomatic disease¹². Disease is also generally more severe in equid hosts than in humans, with most equid cases being lethal⁷. EEEV has the highest reported rate of neurological involvement and lethality, VEEV has the least, and WEEV falls between the two¹². However, VEEV is the New-World virus of most concern as it has historically caused the largest and most frequent outbreaks affecting both human and equid populations, and resulting in many thousands of human cases and equid deaths^12,16. EEEV remains rare in the human population, but there was a significant increase in the number of cases reported in 2019. This increase retreated in 2020, the most recent year with data available from the CDC, with that year having an average number of reports¹⁷. WEEV has virtually disappeared from the human population, and also become much rarer in its enzootic hosts in North America^18,19

VEEV is also a high risk for accidental release and exposure due to it propensity to infect via aerosols²⁰. This led to VEEV being developed as a bioweapon by both the former USSR and the USA, making VEEV a select agent, a classification it shares with EEEV²¹. Select agents are those toxins and organisms that “could threaten public health and safety”²² and have additional security regulations in addition to those controls indicated by biosafety. VEEV generally causes a mild febrile illness that occasionally results in encephalitic infection, resulting in death in approximately 10% of cases when encephalitis is present⁷, however the total case fatality rate is only around 1%²³. Those patients that do survive neurological symptoms are likely to suffer from long term sequelae²⁴.

3. Natural Transmission of the Alphaviruses

Alphaviruses are vector-borne viruses that generally require the use of an intermediate species to transmit to a naïve vertebrate host⁷ (Figure 1). Due to this cycle, the viruses must efficiently infect and replicate in multiple species. Alphaviruses infect a wide variety of both vector and host species. Single species of alphavirus can often infect multiple different species of vector, and different vector species are typically responsible for endemic maintenance and epidemic/epizootic outbreaks of disease²³.

Figure 1.

The transmission cycle of encephalitic alphaviruses. A) Alphaviruses are maintained in nature by cycling between a host species, typically a bird or small mammal, and a mosquito vector species. B) Spillover events often occur into livestock, which reach high viral titers and readily transmit the virus to additional vectors. In the case of the New-World viruses this infection almost always leads to death²⁷. C) Typically, after infection of livestock, humans that work in close association with these animals can also be infected by vector species. Humans are regarded as dead-end hosts for most alphaviruses. In humans these infections may lead to disease, and, in severe cases, death.

The virus first enters the mosquito or other vector through a blood meal that is taken from an infected host. The virus then encounters the cells of the mosquito midgut, before passing into the haemocoel, the circulatory system of the mosquito. Eventually virus arrives in the salivary glands where it replicates to high levels and is transmitted to the next vertebrate host during a blood meal^29,30. Not only does vector transmission complicate control of these viruses, but infection of the mosquito is an important selection process, and different strains of these viruses can behave differently in the vector. In particular, epidemic VEEV strains behave very differently in the mosquito than those are isolated from enzootic infection^29–31.

Upon blood meal from an infected mosquito, the alphavirus is injected into the skin of a naïve host. As these viruses have various cellular tropisms that will result in differing pathologies; we will here outline the general series of events that occur during infection of a susceptible and permissive host cell.

4. Alphaviral replication

The steps outlined here are common to all alphaviruses unless otherwise indicated. After inoculation into the vertebrate host alphaviruses enter permissive and susceptible host cells to manufacture new virions. The alphaviruses are noted to have highly efficient infection. While in infection there is an excess of genomes produced compared to plaque forming units, as time goes on these numbers reach near parity³². The alphaviral replication process is here described in detail, and a summary can be found in Figure 2.

Figure 2.

The replication cycle of alphaviruses. The virion enters a susceptible cell via receptor mediated endocytosis, primarily mediated by clathrin (green) and due to pH changes of the endosome releases its RNA (purple) into the cytoplasm of the host cell. The positive sense genomic RNA is first used by ribosomes to translate the viral nsPs as a polyprotein (yellow = nsP123, green = nsP4). The polyprotein will undergo cleavage events that control the synthesis of the viral RNA species (individual nsPs represented as single green circles). This RNA synthesis occurs in membrane invaginations that are termed spherules. These spherules protect the viral RNA and nsPs from detection by the host cell. Late in infection the structural genes are synthesized (pink = the E proteins, black = capsid). The capsid will form into nucleocapsid cores as it packages the viral RNA, and the glycoproteins are transported to the cell membrane. The nucleocapsid cores translocate to the cellular membrane where they bud off, collecting their envelope and glycoproteins and forming new infectious virions.

4.1. Receptor-mediated endocytosis

The primary mechanism by which alphaviruses enter naïve host cells is via receptor mediated endocytosis, and the viruses are highly promiscuous³³. Of particular interest are DC-SIGN and L-SIGN, which may influence myeloid cell infection³⁴. Additional receptors continue to be discovered^35,36. Heparan sulfate is an important binding partner in cell culture, and this appears to be a specific adaptation that occurs in response to passaging virus repeatedly in cells^33,37–39. The exception to this is EEEV, which has been found to have affinity for heparan sulfate in naturally circulating strains⁴⁰

After receptor binding, the alphaviruses are then transported into the cell via clathrin mediated endocytosis^41,42, resulting in a virus-containing endosome passing through the stages of acidification and maturation. The New-World viruses remain in the vacuoles until they reach the endosome stage, whereas the Old-World viruses escape from the early endosomal compartment⁴³.

There is also now evidence that some alphaviruses may be able to utilize alternative entry strategies, such as direct entry at the host cell plasma membrane^44,45. This has been well characterized in CHIKV, with research indicating that even though an acidification step is required, it can occur in a manner that is independent of the activity of clathrin⁴⁶. However, the importance of this entry method remains unclear, but it shows one way these viruses may be able to display such wide cell tropisms. This does indicate that these viruses may have the potential to develop resistance to entry inhibitors rapidly.

4.2. Viral entry as a target for antiviral drug development.

Receptor attachment and entry are common drug targets, and an example of a drug that targets this activity in the alphaviruses is suramin. Suramin was initially discovered as a treatment for trypanosomal diseases in Africa, and it is also known by the names naganol, suramine, forneau, and germanin^47,48. Suramin has a long history of being tested for potential therapeutic effects in many different diseases.

Before it was tested in the treatment of alphaviral disease, suramin was already known to function to inhibit trypanosomal disease, to potentially act as an inhibitor of the HIV reverse transcriptase, and to have anticancer effects^47,49,50. Suramin has recently been studied in the treatment of CHIKV infection and was found to inhibit multiple stages of the replication cycle, including viral entry. Suramin was also mildly efficacious in the treatment of SINV and SFV. Treatment with suramin was also found to reduce viral load in a mouse model of infection, and reduced infection relation swelling in the foot^51,52. As it has a known safety profile, suramin is a promising lead compound for further refinement via modeling and medicinal chemistry. The structure of suramin can be found in table 2 below.

Table 2:

A summary of selected anti-alphaviral compounds and their mechanisms of action. All structures are published in PubChem.

4.3. Fusion/Uncoating and RNA release

Fusion of the viral and host cell membranes is achieved by the activity of the E1 protein, and expression of E1 without the other glycoproteins is enough to mediate viral membrane fusion^63,64. This fusogenic activity is initially prevented by the interaction of E1 with E2, but this interaction is disrupted at low pH^42,65.

After escape from the endosome, the nucleocapsid interacts with ribosomes, which disassemble the capsid in a non-catalytic manner, which is dependent on conserved capsid sequences^66,67. The disassembly of the nucleocapsid is enhanced by low pH, and the pore forming activities of the E proteins are implicated to induce these pH changes^68–70.

4.4. Translation and processing of the nsPs

Upon release the alphavirus genome is available as an mRNA for cellular ribosomes and recruits all the factors required for protein synthesis in a similar manner as a cellular RNA⁷. First to be translated is the nonstructural polyprotein, which contains the proteins that are responsible for the replication of the viral RNA. The viral nsPs are numbered in the order that they occur in the genome from 5’ to 3’, 1-4. The genomic organization of alphaviruses can be found in Figure 3 A. The initial polyprotein is translated as either nsP123 or nsP1234, depending on read through of a stop codon that may or may not be present in the genome depending on the alphavirus in question^71–73,.

Figure 3.

A) The genetic structure of the alphaviruses. The alphaviruses have an approximately 12kb, linear, positive-sense genome. The genome has two open reading frames, the nonstructural and the structural. The nonstructural open reading frame is here displayed in green, and encodes the four nonstructural proteins, which are responsible for replication of the viral RNA. The structural open reading frame is displayed here in yellow and encodes for the E proteins and capsid as well as the 6K and TF proteins. The capsid and E proteins form the structure of the viral particle. B) RNA synthesis of alphaviruses. This RNA synthesis activity is carried out in spherules on the membranes of cellular organelles. After release into the cytoplasm the genomic RNA (purple) is used to synthesize the initial nonstructural polyprotein. nsP2 initially cleaves between nsP3 and 4 leading to nsP123/4 (nsP123 = yellow, nsP4 = green), which synthesizes primarily negative-sense template RNA (red). The protein undergoes rapid cleavage through intermediate states to reach the final replicase complex nsP1/2/3/4 (represented as individual green circles). This complex synthesizes new positive-sense genomic (purple) and subgenomic (blue) RNA and can no longer synthesize negative-sense RNA. The genomic RNA is used to synthesize additional genomes and is packaged into progeny virions. The subgenomic RNA is used to synthesize the structural genes that form the new virions.

After the initial translation of these proteins, they undergo tightly controlled autologous cleavage events which are independent from cellular processes and result in the formation of multiple intermediates as well as the final mature replicase complex nsP1/2/3/4^74,75. This fully cleaved, mature complex is highly stable. Control of this cleavage process is important as it controls the levels of viral RNA species that are present at different times during infection^76,77; this regulation is discussed in more detail in the following section. This cleavage process appears to have unique regulatory features such as having morphological cleavage recognition instead of sequence specificity⁷⁸. This regulatory process is highly important to viral biology as altering it leads to attenuation⁷⁹. Proper cleavage is also important to immune evasion, as viruses with incomplete cleavage result in alterations of the viral RNA species prevalence, increasing type I interferon induction as well as the sensitivity of the viruses to interferon⁸⁰.

4.5. Viral RNA Replication

The process of viral RNA synthesis is outlined in Figure 3 B. To make additional molecules of RNA genome, the virus is required to first transcribe the positive-sense genome into negative-sense template strand. This activity is performed by the partially cleaved polyprotein nsp123/4⁷⁷. However, the protein cleavage activity of nsP2 rapidly degrades the polyprotein into its constitutive parts. This initially produces intermediate forms of the replicase complex that are short lived and produce both negative and positive-sense RNA^76,77. The final cleavage between nsP2 and nsP3 leads to the formation of the mature replicase complex nsP1/2/3/4 which produces only positive-sense RNA^74,75,81. This self-proteolytic behavior creates a distinct expression profile of the viral RNA. Initially the immature forms of the complex produce higher levels of negative-sense RNA. As the complex is processed the synthesis of negative-sense RNA is reduced and eventually eliminated. This causes most negative-sense RNA to be produced early in infection, as well as less negative-sense RNA being produced overall⁶². Following cleavage and assembly of the mature replicase complex, RNA synthesis converts to the synthesis of positive-sense genomic and subgenomic RNA⁸².

The positive-sense genomic RNA functions primarily as the genetic material of the next generation of virus, as well as being translationally active in the cell to produce additional nonstructural proteins. However, recent work has indicated that the genomic RNA may have biological functions that are not dependent on its function as a viral mRNA, as increasing the amount of capped RNA decreased viral fitness, indicating that there is some important role played by the noncapped RNA which isn’t replicatively competent⁸³.

Late in infection an additional positive sense RNA is produced, the small subgenomic RNA, which is used to produce the structural proteins⁷¹. Additionally, when viral RNA is examined on agarose gels, there is a third RNA species that appears, the so called replication intermediate RNA ^62,84. This an RNA species that runs at very large size, however its significance is unknown, and there is no research indicating its role in the viral life cycle.

4.6. Viral RNA replication as a target of antiviral drug development.

One of the most common targets of antiviral drug development has been reducing replication of the viral genetic material, as this limits viral spread in the host and thus disease. A common drug class that has been found to have these effects is the nucleoside analogs. This class includes the drug ribavirin, used in the treatment of many viruses, but which is largely ineffective as an anti-alphaviral therapy⁸⁵. Multiple nucleoside analogues have been tested as treatments for the alphaviruses and two will be described here, favipiravir, and β-D-N4-hydroxycytidine.

Favipiravir is a nucleoside analogue that is approved for use by the Pharmaceuticals and Medical Devices Agency, the Japanese drug regulatory body, for the treatment of influenza under the tradename Avigan⁵⁴. Favipiravir has also been tested against a wide variety of other viruses, including the alphaviruses, both in vitro and in vivo. Favipiravir inhibits the activity of viral polymerases by competing with purines for incorporation in to the viral RNA. This activity locks the strand and prevents its use in further viral replication^55,86.

When tested in models of alphaviral infection, favipiravir has been found to be mildly efficacious against WEEV, and treatment with this drug resulted in the clearance of CHIKV from infected mice when given in the acute phase of infection, but had no effect in the chronic phase^86–88. However, favipiravir treatment for alphaviruses has also only been tested via intraperitoneal injection, which is not a preferred method of delivery in humans^87,89. There have been no tests with the oral formulation currently approved for use in humans for influenza. Sensitivity to favipiravir in the alphaviruses also appears to be strain dependent. With the strains of CHIKV that are of most concern being less sensitive to treatment than strains that are less involved in human outbreaks⁹⁰. While favipiravir is currently not clinically approved for use in the alphaviruses, the knowledge about its efficacy could lead to future compounds with enhanced anti-alphaviral properties.

Not all nucleoside analogues function by the same mechanism to inhibit viral replication. Another nucleoside analogue that has been examined in alphaviral infection is β-D-N4-hydroxycytidine or NHC⁶⁰. NHC has been tested for therapeutic efficacy against many viruses including influenza and respiratory syncytial virus, and is structurally highly similar to molnupiravir, one of the currently approved antiviral treatments for SARS-CoV-2^91,92. NHC functions by multiple mechanisms, one of which is the inducement of hypermutation during replication of the viral genome⁶⁰. This mutagenesis as well as the secondary effect of reduced viral infectivity results in inducing minimal resistance to treatment in the viral population⁶⁰ which indicates that NHC and its derivatives/related compounds show promise as potential treatments of the alphaviruses. The structures of both favipiravir and NHC can be found above in table 2.

4.7. Importance of the untranslated regions of alphaviruses

The genomes of alphaviruses have large 5’and 3’ untranslated regions (UTRs) which are biologically active. The 3’ UTR is important in avoiding the immune bottlenecks that exist in the arthropod phase of the life cycle⁹³, and the sequence variability that occurs in this region can have significant effects on transmission in mosquito vectors and on vector specificity⁹³. The main variation in this region is due to size, which directly relates to the number of repeated regions that occur in the sequence^94,95. This variation in the 3’ UTR does not only occur between different viruses but can also vary significantly within viral species and has been well documented to differ in the differently pathogenic strains of CHIKV⁹⁶. The 3’ UTR is also involved in pathogenicity of alphaviruses, such as in EEEV where it binds to micro RNA and promotes neurological disease⁹⁷, and is involved in the synthesis of the negative sense RNA⁹⁸.

The 5’ UTR functions in many ways to promote translation of the viral RNA both via its structure and sequence^99,100. These structures can vary significantly between species and aren’t necessarily interchangeable¹⁰¹. The structure of the 5’ UTR is also important to evasion of the interferon response in alphaviruses, with the stability of the structure playing an important role in preventing recognition of the cap structure of the viral RNA^100,102. The importance of this region to virulence is seen in the attenuation of VEEV strain TC-83, a mutation in this region alters the ratios of viral RNA types and results in a significant increase in sensitivity to interferon compared to wild-type viruses^103,104

4.8. Localization of genome replication

Alphaviruses demonstrate a sequestration of their replication to intracellular membranes, which is similar to other RNA viruses which also largely replicate in and on membranous structures^105–107. The alphaviruses utilize microinvaginations called spherules¹⁰⁵. These are sites where the viral RNA has been found to localize in infected cells^108,109. It has been confirmed in vitro that these structures contain viral RNA synthetic activity through the use of purified spherules to produce viral RNA¹¹⁰. It has been recently determined that the initial formation of the spherules is dependent solely on the activity of the nsPs with no requirement for viral RNA being present¹¹¹. However, the size of the individual spherules is dependent on the length of the RNA that is transcribed within, which appears to be a feature unique to alphaviruses¹¹². These spherules have been suggested to play a role in viral immune evasion by isolating viral double-stranded RNA away from cytoplasmic pattern recognition receptors^113,114. There is also evidence from the flaviviruses that in general, membranous association of viral RNA replication can protect RNA from enzymatic digestion¹¹⁵. Many of these phenotypes have also now been confirmed in CHIKV using cutting edge microscopy and structural biology methods¹¹⁶

Spherules were initially identified on large, endosomal-like compartments in infected cells. In several of the alphaviruses these spherules form at the plasma membrane and later traffic to intracellular compartments¹¹⁷. In vertebrate cells, recent work has indicated the movement of the spherules away from the plasma membrane is dependent on the activation of PI3K-Akt-mTOR, and reduction of this activation is associated with an increased proportion of the spherules remaining at the cellular membrane¹¹⁸. However, inhibition of this activity has no effect on viral titer, indicating that localization of replication may not be important to replicative success¹¹⁹

4.9. Translation of the structural genes

The structural genes of the alphaviruses are produced via translation of the subgenomic RNA. The synthesis of this RNA initiates independently from the full-length genomic RNA, but how this secondary initiation happens remains unknown. The initial gene product is a polyprotein that contains the capsid, E proteins, 6K, and TF proteins⁷,⁷¹. This translation is carried out in the cytosol and the structural proteins are synthesized at very high levels and are found throughout the cell^120,121. Production of the structural genes occurs primarily later in infection due to the increased expression of the subgenomic RNA⁷. The capsid protein contains a serine protease domain and uses this to rapidly cleave itself from the other structural genes after it is translated⁷¹.

After cleavage of the capsid protein, the glycoproteins pass into the endoplasmic reticulum, and move through the Golgi apparatus before being embedded into the plasma membrane of the cell⁷¹. The glycoproteins are also highly post-translationally modified via glycosylation and palmitoyaltion⁷¹.

4.10. Packaging of the viral RNA and release of the virion

Unlike some other viruses, the alphaviruses don’t readily produce empty particles¹²². However, they are known to produce defective interfering (DI) particles when passaged at high concentrations repeatedly, although this occurs less frequently than in other viruses such as vesicular stomatitis virus and influenza¹²³. These particles have a range of sizes and are often smaller than normal virions, and they usually incorporate deletions in the viral RNA. The deletions in these DI particles get larger with increased passages¹²³. The alphaviruses are noted for generally producing highly uniform, icosahedral particles¹²⁴ though this can be altered by mutating the structural proteins¹²⁵. Current data indicate that while multicore, and other irregular virion shapes do occur with these viruses, these are highly selected against, and particles largely only contain a single unit of genomic RNA⁷ and capsid cores select for carrying only single cargo units¹²².

After translation of the structural proteins, the viral RNA and capsid undergo interactions based on molecule size and charge, resulting in nucleocapsid like structures occurring in the cytoplasm^71,126,127. Alphaviruses bud directly from the plasma membrane of the infected cell⁷. However, it is unclear how this budding process is initiated¹²⁷. It has been found that both the preformed nucleocapsid like structures and the glycoproteins are able to drive budding^127,128. However, when either of these functions occurs independently of the other, there is a marked reduction in efficiency, indicating that it is likely that these two mechanisms interact to allow for the maximal budding of virions¹²⁸.

Transport of the structural proteins to the plasma membrane requires the host secretory system¹²⁸. However the exact proteins that are used remain unknown¹²⁸. Release of virions can also be inhibited by host proteins. In particular tetherin has been shown to prevent the release of virions from infected cells¹²⁸. The general replication scheme of alphaviruses is outlined in Figure 2.

Reducing viral release and decreasing the infectivity of viral particles is an activity of some anti-alphaviral compounds, in particular ß-D-N4-hydroxycytidine, a compound discussed earlier in section 4.6.

5. Functions of the alphaviral nonstructural proteins

The alphaviruses make four nonstructural proteins. These proteins are responsible for viral RNA replication as well as many other enzymatic functions. The nonstructural proteins are also intimately involved in the pathogenesis of the alphaviruses. The functions of these proteins will now be described in greater detail. While the functions of the nsPs are highly conserved, differences between the Old and New-World viruses will be indicated when necessary.

5.1. Nonstructural protein 1

NsP1 is the capping enzyme for the viral genomic RNA, and this activity occurs independent of the activities of the other nsPs¹²⁹. The activity of this protein has recently been examined in VEEV, having previously been studied only in Old-World viruses. This was also the first time that each individual step, including the final guanyl transfer, has been described¹³⁰. The steps occur as follows. 1) The transfer of a methyl group from S-adenosylmethionine to position N7 of a molecule of GTP is catalyzed; 2) nsP1 receives the methyl-GTP becoming guanylated, releasing pyrophosphate in the process; 3) the 7 methyl-GMP is transferred to the 5’ end of the target RNA^130,131. For this reaction to occur properly, the RNA being capped must have had its 5’ terminal phosphate removed by nsP2¹³².

NsP1 is also responsible for the anchoring of the viral replicase complex to cellular membranes which are the site of RNA replication, and this activity is required for capping to be carried out as well^133–135. Very recently a cryo-em structure was published that showed how nsP1 influences the structure of the membrane spherules and potentially controls entry and exit of materials¹³⁶. NsP1 was found to form a ring-like structure that appears to act as a gate and controls movement to and from the compartment¹³⁶. This detailed structure has now also been used to examine the capping activity of CHIVK in depth as well, as can be seen in recent publication from the Law lab¹³⁷

5.2. Nonstructural protein 2

NsP2 is a multifunctional protein with several distinct domains with discreet enzymatic activities. First, nsP2 is responsible for host cell transcriptional shutoff in the Old-World viruses, and loss of this phenotype reduces viral cytotoxicity^138,139. In the New-World viruses this activity is instead carried out by the capsid protein, and nsP2 is responsible for shutoff of host cell protein synthesis and may have a role in packaging of viral RNA^{138,140–142}. In VEEV this translational shutdown has been shown to mediate resistance to a pre-existing antiviral state¹⁴¹. Interestingly, a recent publication has indicated that both expression of a heterologous nsP2 as well overexpression of an nsP2 in the presence of infection is inhibitory for the alphaviruses in mosquito cells, and is likely one way that the infected cells resist superinfection¹⁴³

There are three recognized domains in nsP2. The N-terminal region contains a helicase domain and NTPase activity that serves to provide energy for the helicase^144,145. This same region also has RNA 5’-triphosphatase activity which prepares RNA for capping, allowing for translation and packaging in virions¹³². The previous 2 years have seen significant advances in the understanding of the structure of alphaviral nsP2 protein. First, the N terminal region of nsP2 from CHIKV, including the helicase domain, has recently been crystalized¹⁴⁶, and this portion of the protein was used to characterize RNA binding activity¹⁴⁷. Following this, there is now a full length crystallization of nsP2 available for CHIKV released by the same research group¹⁴⁸. Here it was revealed that the N and C termini of this protein are connected by a flexible linker, and that this linker is highly important for normal protein function and viral pathogenicity¹⁴⁸. Structures from this group have already proven useful for prediction of the structures of the nsP2 proteins of other alphaviruses⁶². These structures are also already being used to further research into potential therapeutics, in particular using computer modeling for drug selection and binding prediction^149,150.

The most interesting feature of the nsP2 N-terminal crystal structure was the large number of accessory domains that were present, as these domains had not previously been predicted by structural modeling. Of particular interest is the so called stalk domain, which based upon recent research appears to have an important function in viral RNA synthesis as shown by the activity of the recently characterized antiviral compound ML336⁶². This compound is discussed later in section 5.5

Large portions of the N-terminal region of nsP2 remain poorly characterized. Studies have implicated that in VEEV this region may be important to packaging of the viral genome¹⁴². However, in SINV a transposon insertion approach using the sequence for GFP found that this region was involved in the cleavage between nsP2 and nsP3, controlling the ratio of genomic and subgenomic RNA, and regulation of RNA synthesis¹⁵¹. This range of phenotypes indicates that this region is highly important to these viruses, but further characterization and research are needed.

NsP2 also contains a cysteine protease domain that is responsible for the cleavage of the nsPs from the polyprotein into its constitutive members^152–154. As described earlier, this cleavage is responsible for the transition from the synthesis of negative-sense viral RNA to positive-sense viral RNA^77,155. The protease has also been shown to target cellular proteins, a common feature of viral proteases, and this is related to resistance to innate immune responses¹⁵⁶.

Lastly, nsP2 contains a putative methyltransferase domain that was predicted using homology modeling¹⁵⁴. However, it is currently thought to be inactive as it lacks a key active site residue¹²⁹. Recently there has been work that has shown a potential alternative activity for this region. There is data that indicates this domain may play a role in interferon shutoff due to its interaction with signal transducer and activator of transcription proteins 1 (STAT1)¹⁵⁷. This activity appears to be mediated by enhancing the nuclear export of STAT1, which prevents the magnification of downstream immune signaling, including the interferon response¹⁵⁷. Work with CHIKV nsP2 has also continued with groups now using the complete crystal structure to predict potential interactions with innate immune proteins¹⁵⁸, and this structure can now be utilized for studies altering the structures and components of the nsP2 protein to further elucidate its functions¹⁵⁹

5.3. Nonstructural protein 3

NsP3 is poorly understood, but mutations within this protein have resulted in defects in both negative-sense and subgenomic RNA synthesis¹⁶⁰.

NsP3 contains a macrodomain with both adenosine diphosphate ribose (ADPr) binding and hydrolase activity^129,161,162, and these activities have begun to be characterized in the context of infection models. The ADP ribosylase activity is necessary for infection of neural cells and the hydrolase activity results in an increase in replicase complexes¹⁶³. In a SINV model, reductions in hydrolase activity led to reduced neurovirulence while increases in ADP ribosylase activity increased neurovirulence¹⁶⁴.

NsP3 also contains the highly conserved alphavirus unique domain, or AUD¹²⁹. This domain is maintained across all alphaviruses¹²⁹. Recent work using CHIKV has indicated that it potentially has many functions, particularly in subgenomic RNA replication¹⁶⁵. Disruption of the AUD resulted in decreased infectivity, potentially due to decreased interaction with the viral RNA and the subgenomic promoter¹⁶⁵.

The last feature of note in nsP3 is the hypervariable domain or HVD. This domain is so varied that it can be distinct between strains of a single viral species, such as in VEEV¹⁶⁶. This region is tolerant of significant mutation and even deletion, which is unique compared to the rest of the nsPs and their domains¹⁶⁷. Natural duplications and insertions in this region can even have positive effects on viral fitness¹⁶⁸. The HVD is also involved in interaction with host cell proteins, resulting in the formation of distinct protein complexes in Old and New-World viruses¹⁶⁹. These interactions include the cellular machinery responsible for the formation of stress granules, which alphaviruses utilize to their own replicative benefit^170–173. These interactions are highly specific to viral species as well, and may partially drive the differences in pathogenesis seen between species^173–175.

5.4. Nonstructural protein 4

NsP4 is produced in small amounts by most alphaviruses due to the inclusion of a stop codon between nsP3 and nsP4^71,176. Altering expression levels of nsP4 decreases viral fitness, indicating that tight control of expression is highly important¹⁷⁷. The tight limit on expression of nsP4 is also promoted by it being targeted by N-end rule degradation¹⁷⁸. Recent work performed using trans replication systems has elegantly shown how nsP4 is involved in RNA selectivity, yet this occurs in tandem with the other nsPs¹⁷⁹. Interestingly these authors have shown that in trans and in a transfection system, increasing nsP4 levels does not have these same detrimental effects¹⁷⁹

NsP4 is the RNA dependent RNA polymerase (RDRP) of the alphaviruses and is active in both positive and negative-sense RNA synthesis, with the specificity being determined by the cleavage state of the other nsPs^77,129,180. NsP4 can display RNA synthesis activity alone, but its activity is enhanced by the presence of the other nsPs^180–182. There has been recent publication of crystal structures of the RDRP domains of both RRV and SINV¹⁸³. This work indicated that the RDRP pocket of these viruses was highly dynamic and flexible, as well as being well conserved between the two viruses¹⁸³.

NsP4 also has a large N terminal region that lacks predicted structure or function. Recent work has predicted that this region is somehow involved in viral RNA synthesis, as mutations in this region result in resistance to the effects of a drug that inhibits the production of new viral RNA^62,184. However, the function of this region remains unclear. Work by others has also shown that mutations in this region has a broad range of effects on viral RNA synthesis¹⁸⁵. This work, as well as the antiviral resistant VEEV isolates that have been recovered¹⁸⁴, indicate that this region plays an important role in RNA synthesis of these viruses, potentially in tandem with nsP2. This indicates that these proteins have additional, complex interactions and roles in viral biology that remain to be understood. The functions of the nsPs are summarized in Table 3.

Table 3:

A summary of the structures and functions of the alphaviral nsPs

Protein	Structures	Functions
nsP1	methyl transferase domain, guanyl transferase domain, membrane association domains	caps viral RNA making it usable by ribosomes, anchors the replication machinery to cellular membranes111,129–131,133–136
nsP2	helicase domain, ADP binding region, cysteine protease, methyl transferase like domain	unwinds viral RNA for replication, cleaves the polyprotein into its constitutive parts, digests host cell proteins77,132,138–142,144–146,152–157
nsP3	macrodomain, alphavirus unique domain, hypervariable domain	poorly described, necessary for replication, highly involved in host cell interactions129,160–168,170–172,174,175
nsP4	RNA dependent RNA polymerase domain	synthesizes new viral RNA71,77,129,176–178,180–182,185

5.5. The nsPs as potential antiviral drug targets

In recent years there has been increasing interest in viral proteins as targets for antiviral drug development. This is primarily due to their being largely distinct from cellular proteins, as well as being required for viral replication. Additionally, many nucleoside and non-nucleoside analogs broadly target replication of RNA and DNA resulting in significant side effects and toxicity, making novel targets attractive. Targeting viral proteins directly theoretically allows for increased specificity and reduced side effect potential. The approval of sofosbuvir/ledipasvir combination treatment for hepatitis C infection and its incredible success in the clinic has increased interest in these targets, as it has proved that viral protein targets are therapeutically viable. Of particular note is that sofosbuvir/ledipasvir combination treatment has fewer side effects and higher cure rates than the previous standard of care using ribavirin/interferon combination treatment¹⁸⁶.

With regards to the alphaviruses, researchers have so far reported a novel class of anti-VEEV drugs based around a highly aromatized core¹⁸⁴. These compounds were then further developed into the benzamidine drug ML336, which is highly effective at inhibiting RNA synthesis, likely due to interactions with nsP2 and nsP4⁶². This compound is highly specific as well, having only minimal effects on cellular RNA synthesis⁶², and this combined with its low EC₅₀ value indicate that it is unlikely to have significant side effects. While ML336 is highly specific to VEEV, further manipulation of these drugs has potential to lead to treatments for additional alpahviruses^62,184. The structure of ML336 can be found in table 2 above.

There is also work looking at the other nsPs for potential antiviral targets. These include nsP3 in CHIKV, which has been used for modeling studies examining small molecule libraries¹⁸⁷. The other nsPs of CHIKV have been examined as well, and a review summarizing these investigations has been published by Sundar, Piramnanyagam, and Natarajan, and it also compares these efforts to dengue and zika¹⁸⁸. Further review of CHIKV drug discovery can be found in Kovacikova and Hemert¹⁸⁹. A review of promising pre-clinical antivirals against the encephalitic alphaviruses can be found in Kehn-Hall and Bradfute¹⁹⁰.

6. Functions of the alphavirus structural proteins

As outlined above, the alphaviruses manufacture their structural proteins through a second open reading frame derived from the genomic RNA termed the subgenomic RNA⁷. The subgenome encodes six proteins: capsid, E3/2, 6K/TF, and E1. These proteins are produced as a polyprotein similar to the nonstructural proteins, and the capsid has self-cleavage functionality to release itself from the polyprotein⁷¹. The function and processing of these proteins will be further described below.

6.1. Capsid protein

The capsid is the first structural protein that is translated after subgenomic RNA production. This protein is responsible for forming the primary structure of the virion and selects for packaging of the viral genomic RNA by recognition of a conserved packaging signal¹⁹¹, but the formation of nucleocapsid cores is not dependent on this signal^192–194. Encapsidation selects specifically for only single units of cargo to be packaged¹²², and in infected cells, this packaging activity is highly specific for viral RNA. After packaging the RNA, the nucleocapsid cores translocate to the plasma membrane where they will bud into the extracellular environment while collecting a membrane as well as the glycoprotein spikes⁷. The initiation of budding remains poorly defined and there are contributions from both the nucleocapsid core and the glycoproteins. While either can initiate budding independently, they likely function in a synergistic manner^127,128.

The capsid protein is one of the most produced proteins during alphaviral infection, and it has several pathogenic roles in addition to its structural function. In the New-World alphaviruses the capsid protein is able to block the nuclear pores and thus prevent the translation of new cellular proteins, enhancing viral pathogenesis, cytopathic effect, and assisting in immune evasion^138,140,195. This protein synthesis inhibition functions in tandem with the nsPs which actively inhibit the synthesis of cellular proteins^141,196,197. While several mechanisms have been proposed for this activity there is currently no definitive evidence to support one hypothesis over another and it is likely multifactorial. The capsid of the New-World viruses also contains sequences which result in its importation to the nucleus itself, which inhibits the transcription of cellular RNA¹⁹⁵. Exactly how this activity is caried out remains unknown.

While the capsid of the Old-World viruses contains a nuclear transport signal and is able to enter the nucleus, why it does this remains unclear as it is not involved in transcriptional shutoff in infection¹⁹⁸. This transcriptional shutoff in the Old-World viruses is carried out by nsP2, which ubiquinates subunits of the cellular RNA polymerase, targeting them for degradation^195,199.

These non-packaging activities of the capsid protein have been targeted by certain preclinical anti-alphaviral compounds, such as mifepristone and compounds derived from it, by inhibiting the normal trafficking of the capsid protein²⁰⁰.

6.2. The glycoproteins

The alphaviruses manufacture three glycoproteins during infection, E1, E2, and E3. E3 and 2 are made as a single fused peptide that undergoes later processing after translation in the endoplasmic reticulum. E1 is encoded singly⁷. These proteins are translated in the ER and then traffic through the Golgi apparatus and undergo various glycosylation and palmytoilation events^201–207. E2 is cleaved from E3 although the two remain in close association. E2 and E3 also may not always fully separate and this appears to be pH dependent^124,208. The proteins are then arrayed at the plasma membrane and form heterotrimers of E1 and E2²⁰⁹ where they are picked up by budding nucleocapsid cores and integrated into the mature virions⁷.

As mentioned above, the glycoproteins are involved in the fusion of the viral and host cell membranes in the endosomal pathway, which mediates release of the viral genome into the cytoplasm. E1 alone is sufficient to induce this fusogenic activity^63,64. E1 is also able to induce membrane pores, and this pore forming activity is likely to contribute to pH dependent viral particle disassembly as well as induce various physiological changes in the cells due to these membrane disruptions^68–70,210.

E3 is known to function as a signal sequence that guides the structural polyprotein into the endoplasmic reticulum after capsid cleavage²¹¹. E3 remains associated to E2 until furin cleavage separates the two proteins in the Golgi apparatus^211,212. However, E3 must have additional functions as replacing it with an endoplasmic reticulum signal sequence results in the other glycoproteins being trapped in the ER²¹³. Swapping the E3 sequences between alphaviral clades also results in attenuation while swapping within a clade does not have the same effect, and this phenotype is dependent on interactions that occur between E2 and E3²¹⁴. Finally E3 has an important role in protecting E1 from the acidic pH of the secretory pathway and allowing for its secretion to the cell surface²¹⁵.

6.3. 6K/TF protein

The 6K gene splits the E2 and E1 genes and forms two different protein products due to a frameshift that occurs in response to read through of a so called slippery sequence in the middle of the protein^7,216. This frameshift is actively enhanced by the folding of the polyprotein that occurs simultaneously with translation of the viral RNA²¹⁷.

The more common form of the protein is simply termed 6K as it has a size of roughly six kilodaltons²¹⁶. 6K is known to be important to budding of the mature virions and is hypothesized to act as a spacer for the glycoproteins^216,218. There is also strong evidence that the 6K protein directly associates with the E2 glycoprotein, and a loss of this association reduced budding efficiency^{216,219–222}. However, this relationship and the resulting reduction in budding has yet to be fully described.

The second protein is called TF for “trans frame,” as it is generated from a one nucleotide frame shift^216,223. This protein was relatively recently identified as a separate entity from 6K²¹⁶. TF is found incorporated into the virion structure after release from the cell^216,224. Recent work has also indicated that there is a role for TF protein in the evasion of the interferon response and that this is due to specific palmitoylation patterns²²⁵, with mutant viruses causing no disease in interferon intact animals, but disease pathology being restored in ifnar knockout mice²²⁵. Additionally palmitoylation of this protein is self-regulated, with specific domains of TF protein affecting its own palmitoylation levels²²⁶

One or both of these proteins also function as a viroporin ,and when either or both of these proteins is over expressed in bacteria they are cytotoxic to the cultured cells²¹⁶.

It is currently understood that many of the phenotypes attributed to 6K are likely mediated instead by the TF protein; however, the exact differences in their activities remain poorly defined. Both are important for efficient release of progeny virus, both are known to be highly cytotoxic when expressed ectopically, and both are important to viral virulence, with viruses that lack these proteins causing less severe disease phenotypes²¹⁶.

Perhaps the most interesting finding about these two proteins is that they are the only alphaviral proteins that are not necessary to produce viable viruses in cell culture²¹⁶. A deletion of both proteins does result in reduced viral titers in the supernatant of infected cells, but virus is produced^216,227. If any of the other proteins coded for by the alphaviruses are deleted, however, it results in nonviability. The functions of the structural proteins are summarized in table 4.

Table 4:

Summary of the functions of the structural proteins of alphaviruses

Protein	Function
Capsid	Forms the virion structure, selects for packaging of viral RNA, involved in initiating budding from the cell, in the New-World viruses inhibits cellular translation and transcription127,128,140,191–194,198,228
E1	Forms heterotrimers with E2 to form the glycoprotein spikes, fuses the viral and host cell membranes, forms pores in the endosome to induce pH changes63,68–70,210,229
E2	Forms heterotrimers with E1 to from the glycoprotein spikes209
E3	Functions as the signal sequence for the structural polypeptide, protects E1 from acidic pH, additional functions involved in trafficking and virulence213–215
6K	Involved in budding, may help space the glycoproteins on the virion, may function as a viroporin216,218
TF	Involved in evading interferon due to specific palmitoylation, may be a viroporin, incorporated into the virion216,224,225

7. Screening assays for antiviral drug development

As highlighted in the above sections, there is extensive ongoing work on discovery and testing of anti-alphaviral therapeutics. As with all antiviral development, this work has seen an explosion in productivity thanks to the development of high-throughput drug discovery assays and technology. The introduction of these techniques has allowed for an exponential increase in the number of compounds that can be screened, and has played roles in most of the already mentioned antiviral compounds. A summary of selected assays can be found in table 5 below. For a thorough review of the state of antiviral development studies more information can be found in Andersen et al.²³⁰.

Table 5:

Selected antiviral screening assays

antiviral assay	readout type	Use	references
CelTiter-Glo	luminescence	cell viability	62,231,232
DNA staining and TUNEL	fluorescence	cell viability	233–235
MTT	colorimetric	cell viability	236,237
Crystal violet	cell staining	cell viability	238
viral reporter systems	fluorescence or luminescence	tracking viral protein production tracking viral replication determining stages of antiviral interference	233,237,239–243

8. The study of alphaviruses in animal models

A critical part of studying infectious agents and the development of drug treatments is the availability of animal models that recapitulate disease as seen in human patients. The alphaviruses have been extensively studied since their initial discovery in mid 20^th century. Due to this there have been a large number of animal models investigated for potential study of these viruses. VEEV in particular was widely studied in a variety of animal models in the mid-20^th century in an attempt to determine a model that would closely mimic human infection²⁴⁴. Currently mice and non-human primates are the most common alphavirus models. A summary of commonly used and well described animal models for alphaviral infection can be found in table 6 below.

Table 6:

Animal models of alphavirus infection

virus	model	Pathology	references
VEEV	guinea pig	high lethality, no encephalitis symptoms	244–246
	hamster	high lethality, no encephalitis symptoms	244–246
	mice (various immune competent strains)	similar to human infection, route dependent pathology	247–255
	cynomolgus macaques	Similar to human infection lymphatic pathology, myocarditis develop CNS lesions regardless of encephalitis symptoms	244,246,254,256,257
	various equids	high lethality but distinct from human disease	244
EEEV	mice (various immune competent strains)	develop seizures, no vasculitis	247,254,258,259
	hamsters	high lethality, no encephalitis symptoms	254
	guinea pigs	lethal with no obvious encephalitis	254
	macaques	similar to human infection	254,260–263
	owl monkeys	similar to human infection	262
	marmosets	similar to human infection	261
	NHP models	additional readouts such as brainwaves under development	264
WEEV	mice (Swiss, Balb/c, others)	Similar to human infection Also develop myocarditis	254,265–268
	hamsters	high mortality when infected in the periphery, rescuable with interferon	258,269–271
	macaques	Similar to human infection, Also develop myocarditis, lethality is strain and experiment dependent	254,257,272,273
CHIKV	neonatal mice	high lethality	274–277
	interferon deficient mice	high lethality	275
	mice (various immune competent strains)	similar to human arthralgia	275,278–280
	Rag knockout mice and similar	Similar to human arthralgia increased viral persistence	275,281–283
	adult macaques	similar to mild human disease, high dose mimics severe disease	275,284–288
	aged macaques	Similar to human disease, developed to study disease in specific human populations	275,289
	pregnant macaques	Similar to human disease, developed to study disease in specific human populations	287
SINV	mice (various immune competent strains)	specialized viral strains develop encephalitis can also be used for arthralgia, which is more similar to human pathology	8,290
	neonate and weanling mice	high lethality, some encephalitis	291–294
RRV	mice (various immune competent strains)	similar to human disease	295–297
BFV	mice (various immune competent strains)	similar to human disease	298
ONNV	mice (various immune competent strains)	similar to human disease	299
MAYV	mice (various immune competent strains)	similar to human disease	300
	cynomolgus macaques	similar to human disease	284
SFV	mice (various immune competent strains)	similar to human disease	301

9. The Development of alphaviral vaccines

While there are no approved vaccines against alphaviruses for human use, there are veterinary vaccines available for livestock. These vaccines are available as a trivalent dose for the three Equine Encephalitis viruses and are widely used in North, Central, and South America³⁰². While this vaccine is highly effective at disease prevention it does require annual boosting³⁰². The livestock vaccine is based on inactivated virus and is not replicatively competent³⁰³;this is due to previous vaccination with the live TC-83 vaccine strain of VEEV resulting in detectable virus occurring in mosquitos in the area of testing, indicating that use of live attenuated virus could potentially result in outbreaks of disease if there were reversion events³⁰⁴.

There have been many attempts at creating vaccines for the alphaviruses for human use. The most advanced vaccine for VEEV is the strain TC-83, a cell culture attenuated VEEV that was developed by the United States Army³⁰⁵. However, this vaccine is poorly immunogenic and has less than an 85% seroconversion rate when given as a single dose³⁰⁶. It also has a very high side effect rate with more than 20%, and in some studies more than 80%, of treated individuals reporting a side effect upon use^306,307. TC-83 has also been noted to have the potential to revert to wild-type, epidemic strain VEEV^308,309. There is also a specific booster for T-83, C-84, a formalin inactivated vaccine for those who have received TC-83 but did not seroconvert³⁰⁶.

Despite its shortcomings, TC-83 remains in use for certain high-risk individuals such as those who frequently work with wild-type VEEV in high-risk research applications³¹⁰. A rationally designed attenuated strain of VEEV termed V3526 was also tested as a vaccine, but it was found to have some remaining neurovirulence in a non-human primate model and was later abandoned early in clinical trials due to a high rate of side effects^311,312. There are also candidate vaccines available for human use under special circumstances for both EEEV and WEEV, inactivated PE-6 and inactivated CM-4884, respectively. Though like TC-83 these vaccines have low response rates and these two inactivated vaccines often fail to produce durable immunity³¹³.

There has also been recent work examining the safety profile and immunogenicity of a trivalent virus like particle vaccine against VEEV, WEEV, and EEEV³¹⁴. While this work is in the early stages of clinical development, the vaccine was extremely safe with only minimal and primarily localized side effects, and also demonstrated the production of a durable immune response as measure by neutralizing antibody titers against the three viruses. One caveat is that as time went on subjects became less likely to respond to all three of the viruses and instead responded preferentially to only one or two of the viruses³¹⁴. A more detailed review of the state of vaccine development for the encephalitic alphaviruses can be found in Stromberg et al.³¹⁵.

With the spread of CHIKV across the tropics, it has also become a target of intense vaccine development research. The first candidate was the 181/25 strain of CHIKV which was abandoned due to a high rate of side effects^316,317. Recently there has been significant promise shown by a variety of CHIKV vaccines based on live attenuated, vector launched, and subunit platforms^318–321. Some of these have shown great promise for clinical approval³²¹.

10. Remaining questions.

While the alphaviruses have been well studied for many years, they remain significant public health threats due to challenges in the development of either vaccines or antiviral treatments. While vaccines are available for veterinary use^322,323, none of the available vaccine candidates have been found to meet the more stringent standards for humans due to significant side effects and relatively poor immunogenicity^324–326. This lack of treatment means that these viruses require significant ongoing study.

A major challenge that remains in the field is the characterization of the remaining regions of the nsPs that have no predicted function. These proteins remain challenging to study, however recent advances in protein expression have ameliorated this somewhat^146,154,327. There is evidence that these uncharacterized regions are important to viral replication^62,231 , which makes them promising targets for antiviral drug development.

In addition to being potential drug targets, further characterization of these uncharacterized protein regions has the potential to further our understanding of positive-sense RNA viruses generally. While these viruses are highly variable in their biology and pathogenesis, they retain highly similar replication strategies, as exemplified by the similarities of viral RNA dependent RNA polymerase proteins³²⁸ and the consistent use of host cell membranes as scaffolding to develop their replication centers^329,330.

Lastly, there has been some recent work to show that the alphaviral RNAs themselves are likely to have biological activities in addition to their use for translation of the viral proteins. This can be seen in the necessarily tight control of capping⁸³, as well as in the highly complex structures that form in the RNA and affect viral replication^331,332. This all indicates that there remain many outstanding questions about these viruses, and that further research will be needed as they remain significant threats to public health and for emergence into naïve populations.

Table 1:

A comparison of representative alphaviruses.

Virus	Old or New World	Arthritic	Neurologic	Mortality	Location
Chikungunya	OLD	+	−	Reports vary, as low as 0.03%. Some claim this is underreport ed.25	Tropics world wide
Semliki Forest	OLD	+	rarely	extremely rare with neurologic involvement26	Africa
Sindbis	OLD	+	rarely	extremely rare with neurological involvement8–10	Africa and Europe
Ross River	OLD	+	−	none27	Oceania
O’nyong-nyong	OLD	+	−	none28	Sub-Saharan Africa
Venezuelan Equine Encephalitis	NEW	−	+	≤1%23	South and Central America
Western Equine Encephalitis	NEW	−	+	3-7%23	Western North America and parts of South America
Eastern Equine Encephalitis	NEW	−	+	50-75%23	Eastern North and South America

Highlights:

• A broad review of all aspects of the alphaviral life cycle

• Particularly focused on the functions of the structural and nonstructural proteins

• Highlights recent discoveries in alphaviral replication

• Summarizes the current state of vaccine and antiviral development targeting the alphaviruses

• Points out current knowledge gaps in relation to alphaviral replication

• Poses additional areas of research focus to further the field