Temperature dependent viral tropism: understanding viral seasonality and pathogenicity as applied to the avoidance and treatment of endemic viral respiratory illnesses

Summary This review seeks to explain three features of viral respiratory illnesses that have perplexed generations of virologists: (1) the seasonal timing of respiratory illness and the rapid response of outbreaks to weather, specifically temperature; (2) the common viruses causing respiratory illness worldwide, including year‐round disease in the Tropics; (3) the rapid arrival and termination of epidemics caused by influenza and other viruses. The inadequacy of the popular explanations of seasonality is discussed, and a simple hypothesis is proposed, called temperature dependent viral tropism (TDVT), that is compatible with the above features of respiratory illness. TDVT notes that viruses can spread more effectively if they moderate their pathogenicity (thereby maintaining host mobility) and suggests that endemic respiratory viruses accomplish this by developing thermal sensitivity within a range that supports organ‐specific viral tropism within the human body, whereby they replicate most rapidly at temperatures below body temperature. This can confine them to the upper respiratory tract and allow them to avoid infecting the lungs, heart, gut etc. Biochemical and tissue‐culture studies show that ‘wild’ respiratory viruses show such natural thermal sensitivity. The typical early autumn surge of colds and the occurrence of respiratory illness in the Tropics year‐round at intermediate levels are explained by the tendency for strains to adapt their thermal sensitivity to their local climate and season. TDVT has important practical implications for preventing and treating respiratory illness including Covid‐19. It is testable with many options for experiments to increase our understanding of viral seasonality and pathogenicity.


| DEFINITION OF TERMS
In this review, 'dormant' means biochemically inactive, or active at low levels. It may imply no viral replication, or low levels of replication. As discussed below, dormant virions may be invisible to the host immune system. (The word 'latent' is avoided because this refers to retroviruses and herpesviruses whose genomes persist within the host, from which they can subsequently reactivate, which are not considered here.) The locations of dormant viruses in the respiratory tract (in, on or between cells) is unknown, and may vary.
A 'mild', 'attenuated' or 'less pathogenic' respiratory virus is one that is more thermally sensitive (ts). Such viruses possess biochemistry that is most active at temperatures below normal body temperature.
A 'virulent' or 'highly pathogenic' virus is one that is less ts, or active at both body temperature and below. A highly pathogenic virus is likely to cause respiratory illness at any ambient temperature. A mild respiratory virus is likely to cause respiratory illness only at low ambient temperature. For a mild respiratory virus, the 'permissive' temperature is the temperature of the nose and throat, while the temperature of the lungs is 'restrictive'. For a highly pathogenic virus the temperature of all parts of the body is permissive.

| INTRODUCTION: CONVENTIONAL EXPLANATIONS OF THE SEASONALITY OF VIRAL ILLNESS, AND THEIR PROBLEMS
Scientists have for decades recognised the tendency of viral illnesses to worsen during predictable seasons. 1,2 This review focuses on four features of the respiratory viral illness seasonality that lack rigorous and convincing scientific explanations: A. Seasonal timing and weather influence.
B. Common viruses worldwide.
C. Rapid arrival and termination.

| Seasonal timing and weather influence
The seasonal occurrence of viral respiratory illness in temperate regions, with more illness during colder seasons (Figure 1), including a surge in respiratory illness that frequently appears in early autumn ( Figure 2), demands scientific explanation. The shared winter seasonality of so many unrelated viral species is extraordinary. 3,4 The diverse characteristics of some important human respiratory viruses that cause illness with winter seasonality are shown in Table 1. Crowding (E1) would predict that summer sporting events and festivals would be associated with increased respiratory illness, which is not the case. Tamerius et al. showed this lack of association for influenza. 2 Figure 3 shows that 'Google flu search' did not increase in any country in the Northern Hemisphere during the FIFA 2014 football World Cup, b despite spectators and fans packing into crowded bars, restaurants and homes to view World Cup matches.
School vacations are not well-correlated with respiratory illness. For example, children are on vacation in Singapore in December, which coincides with the yearly peak of influenza. 2 The crowding hypothesis E1 would also predict that colds and flu would be much more prevalent in large cities than in the countryside, which is not the case. 5 Survival (E2) focuses on the suggestion that cold and dry conditions increase the survival of the virus outside the body, thereby encouraging transmission. The corollary is that E2 would also predict that colds and flu would be rare in warm, wet or humid weather.
However, in tropical locations that have rainy seasons, colds and flu are most prevalent during those seasons, when humidity is likely to be high. 2,6 It has been suggested that air conditioning in the Tropics may enhance viral survival by cooling and drying indoor air. 7 However, in some Tropical locations such as Fortaleza (Brazil), temperatures during the rainy season are slightly lower than in the dry months, so air-conditioning use is unlikely to be higher, yet colds and flu are much more prevalent in the rainy season. 2,6 Also, World Health Organization reports show that influenza was common yearround in the Tropics during the period 1964-1975, a time when most people in the Tropics did not have air conditioning. 8 The effect of humidity on virus survival can be examined experimentally. Table 2  improved in dry conditions, but when influenza A viruses were supplemented with material from the apical surface of human airway epithelial cells they remained equally infectious when relative humidity was varied from 23% to 98%. 9 Moreover, other viruses, including rhinovirus-14, rhinovirus-16, and adenovirus, were more stable in humid conditions. [10][11][12] The illnesses caused by these viruses are nevertheless seasonal in temperate regions in the cold, dry air of winter, with virtually all other viral respiratory illnesses. None of this information supports the idea that it is the dry conditions of winter specifically that are ideal for the survival of respiratory viruses and thus explain the timing of cold and flu seasons.
Studies also show that the onset of respiratory illness outbreaks in response to temperature change is too rapid to be caused by  15 The authors also tracked relative humidity but found only 'a weak non-linear relationship' with health outcomes, indicating that increased virus survival with low humidity was likely not a cause. Fluctuations in ambient temperature are what appear to be crucial for most respiratory viruses, with rapid starts to outbreaks that cannot be accounted for by increased transmission.
A recent review of Covid-19 data subscribes to the virus survival explanation, but mostly focuses on absolute temperature and humidity, not weather changes, and concludes that much of the data is confounded with public policies implemented in the reaction to the pandemic. 16 However, Covid-19 outbreaks in many areas have F I G U R E 2 Four datasets that show an 'autumn surge' in colds (red ovals). (a) Graphs show the average yearly incidence of colds for each period of 10 days þ/À 1 day 13 A1: London, 1951London, -1955A2: Newcastle, 1951A2: Newcastle, -1957A3: Cirencester, 1954A3: Cirencester, -1956A4: Chalke valley, 1948A4: Chalke valley, -1949. For London and Newcastle the number at the end of September was greater than or equal to the number in midwinter. In Cirencester, A3, and Chalke valley, A4, smaller autumn increases were seen. (b) At the beginning of van Loghem's study in September 1925 one third of respondents in Amsterdam had colds, with high numbers in other regions (red oval). 14 For more information see followed regional weather patterns. The autumn 2020 wave of Covid-19 in the USA began in the Midwestern states, 17,18 which experienced two severe, record-breaking cold snaps in September and October 2020. 19 SARS-CoV-2 data will continue to stream in, but it (already) appears to be affected by weather changes too rapid to be accounted for by increased transmission.
The immune defences (E3) explanation assumes that the immune system does not function as well in the colder, darker conditions of winter in temperate regions. However, Paynter et al.
found that vaccines were slightly more effective in winter than in summer. 20 An early study on an isolated tropical island found that colds increased in September when the temperature dropped by only 1°C-2°C, from around 24°C at night to around 22°C. 21 It seems implausible that such a small temperature drop, starting at a relatively high temperature, could meaningfully depress human immunity. Experimental studies by Andrewes, and Dowling et al., suggested that immune defences are not made meaningfully weaker by experimental chilling of individuals (although these studies used 'pedigree' viral strains that may behave differently from wild-type strains, discussed below). 22,23 F I G U R E 3 Google 'flu search' activity during FIFA 2014. Google flu search measured the number of people who search for terms related to colds and flu using the Google search engine. During major sporting competitions people often crowd together to watch the events. During the 2014 FIFA football World Cup, in the 18 countries in the Northern Hemisphere where Google reported flu search activity, there was no increase in the number of Google searches for related terms. This suggests that the winter epidemics displayed are not mainly driven by crowding (E1). Notice the epidemics of colds in September 2014 (red oval). Autumn epidemics are common when the temperature dips at the end of summer ( Figure 2) T A B L E 1 Characteristics of common winter season respiratory viruses (exception: parainfluenza type 3)

Family
Viruses causing human respiratory illness   Table 3 lists the viruses that most frequently caused hospitalization of children in three studies based at hospitals at different latitudes. 3,4,27 There are striking similarities. Moreover, viral respiratory illnesses are rare in mid-summer in temperate regions, but they usually occur throughout the year at moderate levels in the Tropics (sometimes with a surge in the rainy season 6 ). For example, Figure 6 shows that influenza is almost absent during the summer in Northern USA and Sydney (Australia), but it is present

| Common viruses worldwide
year-round in Singapore and during the rainy season in Fortaleza (Brazil). The same trends can be seen in the prevalence of influenza reported by the World Health Organization for the period 1964-1975. 8

| Rapid arrival and termination
As characterized by Hope-Simpson, an 'explosive' arrival and rapid termination is observed for many epidemics of influenza and other endemic respiratory viruses ( Figure 7). Influenza frequently arrives simultaneously throughout large geographical areas, in isolated remote rural settings and nearby towns and cities, and even widely separated locations at similar latitudes. 5

| THE FOUR PREMISES OF TEMPERATURE DEPENDENT VIRAL TROPISM (TDVT)
The hypothesis explains the four features of seasonal respiratory outbreaks listed above. It is built on four simple premises:  Based on these premises, this review concludes that the seasonality of viral outbreaks is mainly a consequence of viral thermal sensitivity. Natural selection causes the level of thermal sensitivity to adapt to local conditions. Thermal sensitivity therefore varies with the seasons and in different climates as discussed below.
The Nobel laureate André Lwoff suggested part of the hypothesis in 1959, when he noted that the degree of virulence of viruses is often related to their level of thermal sensitivity. 33 In 1979, Richman and Murphy developed this further, discussing many examples of thermal sensitivity in natural and lab-made viral strains, and noting that the near-universal attenuation of ts strains made them good candidates for vaccines. 34 The full hypothesis was proposed by Shaw Stewart and discussed at length in 2016, focusing on seasonality and the natural selection of strains with varying degrees of thermal sensitivity and pathogenicity. 28 The same hypothesis was put forward by Eccles in 2020, this time concentrating on the advantages of thermal sensitivity to the virus. 35 The current shorter review summarises the evidence for TDVT, including recent studies, and discusses its implications for avoiding and treating respiratory illness including Covid-19.  29 Most countries experienced only one main peak, occurring in the same two-month period. Epidemics the following winter were much smaller and only apparent in four countries. Mortality in Finland is difficult to interpret during and after the Finnish Civil War (January1918 to May 1918). A common explanation for the rapid termination of this epidemic is that almost the whole population was exposed to the virus, and that individuals either died or acquired immunity. However, recent studies have suggested that the extreme pathogenicity of Spanish flu arose from the unusual selective conditions that existed in the armed forces during World War I, 29  The moderation of viral pathogenicity is predicted by the 'transmission-virulence trade-off' hypothesis. 37 This states that the benefits of increased replication (including increased shedding) must be balanced against the reduction of time during which shedding takes place, and the reduced mobility of hosts. The trade-off hypothesis was introduced to explain patterns in myxomatosis data. 37 Myxomatosis has been studied intensely because it provides a classic reference for the rapid evolution of virulence. It is a highly pathogenic viral disease, normally spread by mosquitos, that jumped from New World rabbits to European rabbits. During the first year after its introduction to Australia, it is estimated to have killed 99.5% of the rabbits that it infected. 38 Within 'a few years' genetic resistance to the disease in rabbits increased in some areas, but independent testing with laboratory rabbits showed that milder viral strains emerged more rapidly.    Tables 4 and 5  TDVT also can shed light on the febrile response to infection. The response is well-preserved across the animal kingdom, and cues delivered by the thermal element of fever are known to stimulate both innate and adaptive immune responses. 76 It is, however, not clear why raised temperature is used as a protective signal in addition to chemical signals (cytokines and hormones). If, as suggested by TDVT, most respiratory viruses are inhibited by higher temperatures, fever may slow their replication throughout the body. It is interesting that reducing fever by the administration of antipyretics in a randomised prospective study in the ICU was associated with negative outcomes and increased mortality (although the trend was not statistically significant, p = 0.06). 77  84 Shaman and Galanti found that 50% of infections caused by the endemic coronaviruses within families were asymptomatic. 83 Table 6 includes studies of respiratory viruses in Antarctica that showed dormancy. [86][87][88] In one example, parainfluenza shedding occurred throughout the winter isolation period, with two episodes of respiratory illness. 88 Other studies in the community identified asymptomatic individuals who had not seroconverted, but who were F I G U R E 1 1 'Covid toes' on the feet of a 16-year-old boy, who usually did his schoolwork wearing socks without shoes. He felt completely well, but a blood test showed the presence of a viral infection. So-called Covid toes are blemishes that sometimes appear on the hands and feet, usually pink or purple, after exposure to SARS-CoV-2, usually in older children and adolescents. Note the viral-plaque-like appearance of blemishes of varying sizes. TDVT suggests that CoV-2 virions that may reach the bloodstream are more likely to bind to and invade cells in colder parts of the body, possibly causing these visible symptoms shedding influenza A 78,80,81 or influenza B. 79 Thai et al. commented that this 'may indicate that viral RNA remained in the respiratory tract without being internalised and eliciting an immune response'. 78 Since dormant respiratory viruses are often present in the respiratory tracts of otherwise healthy individuals, and since TDVT suggests that such viruses will be activated by local temperature drops, the hypothesis predicts that cold-weather snaps will cause surges in respiratory disease (Figure 4). In another Antarctic example, several men developed colds after the air temperature fell from 0°C to 24°C. 87   presumably now visible to the immune system -can be destroyed by our defences. Note that the Eurowinter Group found that frequency of going outside was not significantly correlated with reduced mortality, suggesting that this apparent protective effect is not solely driven by the regular activation, by cold, of virions in small batches. 90 Several studies from the 1950s and 1960s, however, have been interpreted as showing that body chilling does not increase the chance that individuals will suffer from respiratory illness (Table 7).

| Premise #4: ts respiratory viruses often remain dormant in the respiratory tracts of their hosts. If ambient temperature drops or hosts are chilled the viruses may become active and cause sickness
Two influential studies by Dowling et al. and Andrewes were performed with volunteers who were inoculated nasally with

F I G U R E 1 2
The regression coefficients (R), and their significance (p), for cause-specific indices of respiratory disease-related mortality on personal cold-exposure factors standardised at 7°C mean daily temperature in eight European regions, ranging from northern Finland to Athens. 90 The Eurowinter Group used market research techniques to analyse the climate and the measures taken to protect individuals from low temperatures. The activities that are most strongly correlated with dying from a respiratory illness are at the top left of the plot, while those that appear to be protective are at the bottom left. Activities that are not related to the chance of death are on the right or at the middle level suspensions of recycled cold viruses (of unknown species) taken from previous volunteers. 22,23 When individuals were subsequently chilled, the studies found that they were no more likely to get colds than unchilled individuals, and the authors concluded that chilling does not bring on colds. Since these experiments were designed to be carried out within a limited time-frame, often one week, and infections had to appear quickly if they were to be recorded, it is possible that unusual fast-acting viral strains that had lost their natural thermal sensitivity were selected over time. 28 Another study using a type-15 rhinovirus did find that 16% more of the chilled group developed colds than the unchilled one, but, because the results did not reach statistical significance, the authors reported a 'failure to demonstrate effect'. 91 Moreover, the strain used was isolated eight years before the study and grown in WI-26 cells, a cell-line that is routinely propagated at 37°C. 92 It is therefore likely that this strain, like many other laboratory strains, had lost much of its natural thermal sensitivity by the time the study was carried out. Therefore, it is possible that chilling increases the risk of getting a cold caused by a 'wild' virus, but not a cold caused by a virus such as was used in these studies.

T A B L E 7 Studies investigating the effect of chilling on respiratory illness in human volunteers
The   1. Dress warmly to avoid chilling. 90 Chilling reduces the temperature of the lining of the respiratory tract, allowing dormant respiratory virions to become active and replicate. 89 2. Avoid standing still outdoors in cold temperatures. 90 3. Take regular outdoor exercise sufficient to cause sweating. 90 It seems likely that breathing cold air activates dormant virionsmaking them visible to the immune system -while dressing warmly improves immune defences in the respiratory tract. 28 Note that the combination of cold air and warm clothing seems to be protective. , several studies showed that the production of influenza viral proteins increased at temperatures F I G U R E 1 4 Global movement of influenza strains. The authors divided the world into 9 regions, and, by analysing the sequences of viruses that were sampled worldwide over three years, they charted the movement of influenza strains. The region where each virus was sampled is colour coded, and the colours show where the ancestors of current viruses (in each region) came from. The right-hand side of each box is only coloured with the region's own colour because strains have had no time to move. As you go to the left the colours show where the ancestral viruses were at the time indicated. For example, 'Yamagata-like' influenza B viruses in India (fourth row, first column) moved very little during the study, with most viruses at the end being descended from strains that were also in India at earlier times. The figure shows that influenza tends to move from hotter locations to colder ones.

| Future work for the validation of TDVT
The hypothesis is testable at many levels, and suggestions for experiments have already been made. 28 Important suggestions include: 1. Systematically survey the best temperatures for propagating wild-type respiratory viruses in tissue cultures.
2. Infect animals with labelled respiratory viruses and determine the positions of replicating viruses in the respiratory tracts of animals that have experienced different temperature regimes (i.e., temperature-shift time-course experiments, see Figure 9(b)).

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analysed in this study. service operated by Google that attempted to make predictions about influenza activity. After Google ceased publishing Flu Trends in August 2015, the graphical data for related searches was still available as 'Google flu search'. c Epidemiological studies show that vitamin D deficiency is associated with increased incidence of cancer and cardiovascular disease, but a very large, randomised, placebo-controlled trial of supplementation with vitamin D (cholecalciferol) did not result in a lower incidence of either disease. It has been suggested that other important effects of sunshine such as the photorelease of nitric oxides in the skin may explain the link with reduced cardiovascular disease and similar considerations may apply to the incidence of both cancer and respiratory illness. d During the rainy season in Tropical locations increased humidity reduces evaporation in the respiratory tract and its temperature will be closer to body temperature. When this temperature difference is small any strain that can be transmitted is more likely to infect both the upper and mid (or lower) respiratory tract, causing symptomatic illness.