- Email: [email protected]
Why does size matter for viruses – A new paradigm on viral size
Medical Hypotheses 73 (2009) 133–137
Contents lists available at ScienceDirect
Medical Hypotheses journal homepage: www.elsevier.com/locate/mehy
Why does size matter for viruses – A new paradigm on viral size Gnanadurai John Fletcher a,*, Solomon Christopher b, Manu Gnanamony a,1 a b
Department of Clinical Virology, Christian Medical College, Vellore 632004, Tamil Nadu, India Department of Biostatistics, Christian Medical College, Vellore 632002, Tamil Nadu, India
a r t i c l e
i n f o
Article history: Received 17 February 2009 Accepted 21 February 2009
s u m m a r y Size and shape are the two immutable laws that govern all life forms including viruses. In this study we postulate and evaluate the hypothesis that there exists a strong association between viral geometry and features of viral disease outbreaks. Data on viral disease outbreaks were retrieved from WHO and CDC public domains for a period of twelve years to assess the relationship between viral size and epidemiological factors such as number of outbreaks, case fatality rate, proportion of emerging infectious diseases and transmission routes. We observed a significant correlation between viral size and frequency of disease outbreaks (q = 0.82, p = 0.004), case fatality rate (q = 0.48, p = 0.03) and genome size (r = 0.79, p < 0.001). Viral sizes were significantly different among diverse transmission routes (p < 0.001). The proportion of emerging infectious diseases were significantly different between viruses with size <105 and P105 nm3 (21% vs 64%, p = 0.046). In conclusion, this preliminary evidence shows that viral size plays a substantial role in the epidemiology of viral diseases. Our data suggests that small size viruses are associated with more number of outbreaks than large size viruses. Large size viruses are associated with high case fatality rate and can be potential emerging pathogens. Viral size may be crucial for niche selection and specified transmission routes in the susceptible host. Hence, viral geometry should not be neglected in epidemiology and modeling of viral diseases, and planning vaccine strategies. Ó 2009 Elsevier Ltd. All rights reserved.
Introduction
Development of hypothesis
Our biosphere represents assorted size ranges of organic forms from the smallest virus and bacteria to the largest metazoans (whales and sequoias). Today, we recognize more than 1550 virus species with a size range of 25–250 200 200 nm, which display a wide range of host specificity [1]. What intrigue us about viruses are their specific size, symmetry and shape. What led parvovirus to be small and poxvirus to be a giant? Do diverse viral sizes demand any teleological explanation in biology? Modern biology deals with diverse viral sizes for tertiary reasons and has completely ignored its biological importance. However, recent studies on viral structures have revolutionized the field of virology [2–4]. Viruses are known to cause wide ranges of diseases from selflimited infection to fatal disease. Epidemiology of viral diseases are determined by immunity, proportion susceptible, proportion infected, incubation period, generation time and transmission patterns [5]. Here, we present a strong case for association of viral sizes with frequency of disease outbreaks, case fatality rate (CFR), emerging infectious diseases (EID) and transmission routes.
Viruses display wide ranges of sizes and shapes which exist so to their advantage by natural selection. Viral size has been known to influence its entry and exit in susceptible cells. The concept on viral size has also been exploited for size dependent vaccine delivery systems. In addition, we observed that all gastroenteritis causing viruses are small in size. Hence, we hypothesize that there is a strong association between viral geometry and features of viral disease outbreaks.
* Corresponding author. Tel.: +91 416 2282616, mobile: +91 9487 629620; fax: +91 416 2212102. E-mail addresses: fl[email protected] (G.J. Fletcher), Solomon.christopher@ gmail.com (S. Christopher). 1 Tel.: +91 416 2282616. 0306-9877/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.mehy.2009.02.029
Evaluation of hypothesis To ascertain the association between the viral geometry and certain features of viral disease outbreaks, we compared viral size (volume) with frequency of disease outbreaks, case fatality ratio (CFR), ability to cause emerging infectious diseases and transmission routes. Data collection The data on disease outbreaks for a period of 12 years (January 1996–December 2007) were collected from WHO [6]. Information was specifically collected on viral disease outbreaks with reference to the number of outbreaks, number infected per outbreak and number of deaths. The total number of infections and deaths were
134
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
collected from all successive updates provided for each outbreak. Information on viral size were obtained from ICTV [1]. Data on food borne outbreaks were obtained from CDC (http://www.cdc.gov/ mmwr/preview/mmwrhtml/ss5510a1.htm?s_cid=ss5510a1_e) for a period of five years (1998–2002) for which complete information was available. Data on viruses causing emerging infectious diseases for a period of 11 years (1995–2005) were collected from Field’s Virology [7]. Transmission routes for 22 medically important viral families were collected and categorized into feaco-oral, vector borne, blood borne, respiratory, sexual, vertical and zoonotic (Field’s Virology, 2007). Viruses that shared more than one transmission route were represented in all the respective routes. To standardize viral size for different shapes, volume of viral particle was calculated based on the shape of virus. Since viral size is provided as a range, mid size value was used in the calculation of volume. The following geometric formulae were used for calculation: for spherical virus, V = 4/3pr3; for cylindrical virus, V = pr2l; p and for icosahedral virus, V = 5/12(3 + 5)a3. An overall CFR was calculated for each virus as the ratio of number of deaths to the total number infected by combining data from all outbreaks. Viruses that caused outbreaks resulting in at least one fatal case were included for analysis. Due to this reason, buffalo pox (April 1996), coxsackie virus (1996, 1997), chikungunya (2006), rotavirus (2000), and astrovirus (2002) were removed from CFR analysis. Statistical analyses All statistical analyses were performed using SPSS 11.0 for windows. Spearman’s rank correlation was used to assess the relationship between viral size and, frequency of disease outbreaks and case fatality rate. Pearson’s product–moment correlation was used for assessing the relationship between viral and genomic size on logarithmic scale. Kruskal Wallis test was used to test for differences in viral size among various transmission routes. Fisher exact test was used to assess the association between emerging viral pathogens and viral sizes. Results To study the association of viral size with disease outbreaks, data for a period of 12 from WHO and five years from CDC were analyzed. During this period, viruses belonging to 13 families caused 927 outbreaks. List of viruses in each family implicated in outbreaks is shown in Table 1. In these outbreaks, there was a significant negative correlation between viral size (mid-volume in nm3) and frequency of disease outbreaks (q = 0.86, p = 0.002) (Fig. 1). We observed a similar trend in sub-group analysis of
Fig. 1. Relationship between viral size and frequency of disease outbreaks.
viruses that shared same transmission route (arthropod and rodent borne viruses) (q = 0.700, p = 0.188) (Fig. 2). The astrovirus outbreak (single) was removed from analysis (viral size vs frequency of outbreaks) due to underreporting. However, sensitivity analysis including this outbreak did not alter the results. To understand the prominence of outbreak frequencies with viral sizes, a cut-off of 14 (median) outbreaks was set. In this, Picornaviridae, Flaviviridae, Caliciviridae, Orthomyxoviridae and Bunyaviridae were small in size (104–105 nm3) except Filoviridae (106 nm3). We observed a significant negative correlation between genome size (in base pairs) and frequency of disease outbreaks (q = 0.82, p = 0.004). Significant linear relationship was observed between viral size and genome size (in logarithmic scale) among viruses that caused outbreaks (r = 0.79, p < 0.001) (Fig. 3). There was a positive correlation between viral size and case fatality ratio (q = 0.479, p = 0.03). Highest CFR was observed in outbreaks caused by marburg (84%), nipah (60%) and ebola (52%) viruses and the lowest with norovirus (0.004%), hantavirus (0.08%) and dengue virus (0.2%). The proportion of emerging infectious diseases between the two viral size groups (<105 and P105 nm3) were significantly different (21% vs 64%, p = 0.046) (Fig. 4). Majority of viruses that were implicated as emerging viral pathogens had size P105 nm3 which include SARS, ebola, marbug, hendra, nipah and monkey pox. The distribution of viral sizes among different transmission routes is shown in Fig. 5. Viral sizes were significantly different among diverse transmission routes (p < 0.001). Almost all viruses that had enteric route of transmission were smaller than 105 nm3 and most viruses which had zoonotic mode of transmission were
Table 1 List of viruses under each viral family for 12 year outbreak. Viral family
Virus
Picornaviridae Astroviriridae Flaviviridae Caliciviridae Orthomyxoviridae Bunyaviridae Reoviridae Arenaviridae Togaviridae Coronaviridae Filoviridae Paramyxoviridae Poxviridae
Enterovirus (3), poliovirus (20), coxsackie virus (2), hepatitis A virus (50) Astrovirus (1) Dengue virus (21), yellow fever (55), west Nile fever virus (4), Japanese encaphilitis (4), St. Louis encaphilitis (1) Norovirus (657), hepatitis E virus (2) Influenza (11), avian influenza (23) Crimean Congo virus (8), Hantaan virus (4), Rift valley virus (10) Rotavirus (1) Lassa fever (9) Chikungunya (7), O’nyong nyong (1) SARS (6) Ebola virus (16), margbug virus (3) Measles virus (4), hendra virus (1), nipah virus (1) Buffalo poxvirus (1), monkey poxvirus (1)
The numbers in brackets indicate the number of outbreaks.
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
135
Fig. 2. Relationship between viral size and frequency of outbreaks in Arboviruses.
Fig. 5. Association between viral size and transmission routes.
Viral size and frequency of outbreaks
Fig. 3. Comparison of viral size and genome size.
Fig. 4. Association between viral size and emerging infectious diseases.
above a size of 105 nm3. However, viruses that had respiratory mode of transmission showed diverse size ranges (104–106 nm3). Consequences of hypothesis and discussion Viruses are known to infect all life forms and cause a wide spectrum of diseases. The epidemiology of viral disease is complex and multifaceted. The quantification of disease occurrence in relation to viral geometry may offer useful insights into the epidemiology of these diseases. Data on disease outbreaks were retrieved from WHO because of documentation accuracy, accessibility and wide geographic coverage. Lack of WHO data on enterically transmitted viruses were compensated by data from CDC website.
Of the five viral families that caused large outbreaks which clustered on the left wall of the graph (Fig. 1), Picornaviridae and Caliciviridae were two enterically transmitted viral families that were small. At this juncture, we are unable to recall any large size viruses that can be definitely transmitted through enteric route and cause gastroenteritis. To cause such large frequencies of enteric outbreaks, viruses (Picornaviridae and Caliciviridae) must be small, non-enveloped and retain infectivity at pH 3.0 and 2.7, respectively. Hence mutation and recombination are the only strategies that are viable for small viruses to maximize their diversity and adaptability. Like other RNA viruses, picornaviruses relish in their high error rates and survive on threshold of error catastrophe [8]. This made us speculate that all enterically transmitted viruses have undergone size dependent niche selection to infect the small intestine which is the largest lymphoid organ in the body. Parvovirus is the smallest virus that is transmitted by non-enteric route (respiratory) in humans and causes fifth disease, transient aplastic crisis and other systemic diseases [9]. However, studies have shown that canine parvoviruses, a distant relative of human parvovirus is transmitted enterically [10]. In addition, bocavirus, a classified member of Parvoviridae is known to be transmitted by respiratory and enteric routes [11]. Viral agents like picobirnavirus, aichi virus and enterovirus 22 that are rarely implicated in gastroenteritis are small in size [12] which corroborates our findings. Hence, the independent etiological role of torovirus (large size) in gastroenteritis is doubtful. Astrovirus is another small, enterically transmitted virus associated with outbreaks of gastroenteritis [13]. But, both in WHO and CDC outbreak information, the frequencies of astroviruses outbreaks were under reported, possibly due to limitations in screening, diagnostic transition and occurrence of co-infections. Additionally, we speculate that competition among many small size viruses for its niche (small intestine) may be an additional factor for such low outbreak frequency. In this context, any future vaccine intervention for major gastroenteritis viruses (rotavirus and Norwalk) may be a tipping point for astrovirus outbreaks. Flaviviridae and Orthomyxoviridae were other viral families that had outbreak frequencies above the calculated cut-off (median) and clustered on the left wall of the graph. The size ranges of these viral families were 104 and 105 nm3, respectively. Flaviviridae is a small arthropod borne virus family responsible for 84 independent outbreaks. On subset analysis of arboviruses, same size restricted
136
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
trend with outbreak frequencies were observed (q = 0.700) which strengthens our findings. Despite vaccination, yellow fever is still an emerging threat because of high sequence diversity, selection pressure with differing vectors and other ecological factors. Dengue virus, a member of Flaviviridae shows high genetic variability, variation in vector population sizes and strain replacement/extinction potential [14] which are factors commonly associated with large outbreaks. In addition to climatic seasonality, arthropod biology (quiescence, synchrony, vectorial capacity) and human behavior; geometry (viral size) may additionally determine the complex arbovirus epidemiology. Orthomyxovirus, a relatively small respiratory pathogen (105 nm3) is of special concern for its pandemicity and fatality. In addition to reassortment and recombination, the size advantage of influenza virus can have additive effects on influenza pandemics [15]. A parallel drawn from nanoparticle study [16] made us speculate that size of influenza virus may influence replication and nature of immune response in influenza diseases. All large size viruses (Togaviridae, Coronaviridae, Paramyxoviridae and Poxviridae) caused very low outbreaks except Filoviridae (Fig. 1). Ebola and marburg are known to cause haemorrhagic infections with high fatality rates [17]. Geographical location, elusive reservoir, other ecological factors are shown to be associated with Filoviridae outbreaks [18]. We speculate that unique viral size, shape and pleomorphism may additionally determine certain virological properties. Eradication of small pox was possible due to the absence of non-human reservoir, single stable serotype, availability of effective vaccine and no asymptomatic carriage of virus. We speculate, in addition to these factors, geometric (big size) disadvantage could have been a crucial factor for small pox extinction. Though large size viruses can cause fatal outbreaks, size advantage for control is biologically possible, with vaccine as an effective arsenal. Virus specific neutralizing antibodies are crucial for effective humoral immune response in viral infections [19]. Stoichiometric determination showed that poliovirus require about an average of 9 monoclonal antibody (Ab) molecules and rhinovirus require 10–20 Ab molecules per virion to show 90% neutralization. Influenza and rabies virus require 100 and 130–350 IgG molecules for 90% neutralization [20]. In the occupancy model of multi-hit hypothesis, we speculate that viral size (surface area, nm2) may be crucial for efficient neutralization. This shows a possible relationship between geometry of viral size and the number of antibodies obligatory to confer neutralization. The concept of virus-like particles has revolutionized the field of vaccinology. Animal viruses have evolved to enter and exit the susceptible cells by generic cellular and virus-specific mechanisms. The basic differences of viral entries apart from obligate receptor–ligand interaction and alternate entry routes (caveolae and antibody) can be better explained by additional size differences seen in diverse viruses. Studies have shown that particle size influence the route of entry [21], dynamics of endocytosis and viral budding [22]. Particles less than 200 nm prefer clathrin mediated endocytosis whereas particles of 500 nm show caveolae-mediated entry [23]. Besides, studies have shown the biological relevance of viral sizes with modern applications of virus sized vaccine delivery system in vaccinology [24]. Studies have also shown that an optimal particle size exists for clathrin or caveolin independent cell entry or exit [25]. Viral size dependent theoretical and mathematical models should be developed to understand the challenging pathogenesis of hepatitis B virus, hepatitis C virus and human immunodeficiency virus. Hence, such curious qualitative and quantitative humoral and cell mediated response on diverse viral sizes may shed light into disease susceptibility and pathogenesis of challenging viral diseases.
Viral size and case fatality rate (CFR) CFR is a crucial indicator of disease severity during outbreaks which can be influenced by host factors and viral factors. In our study, high CFR is seen with large size viruses (marburg, nipah and ebola) possibly due to the presence of virulent genes. Ebola virus encodes VP35, a viral encoded type I antagonist that inhibits interferon regulatory factor (IRF-3) activation [26], VP24 that blocks interferon (IFN) signaling [27] and niche specific virulent genes can also contribute to its fatal pathogenesis. In nipah, the non structural proteins (P, V and W) have been shown to down regulate IFN signaling cascade [28]. Majority of large size viruses that were associated with emerging infectious diseases also had high CFR. Hence considering the genome economy, expansion of virulent genes is the only possibility for fatal pathogenesis associated with large size viruses. This is partly explained by our data that revealed a linear relationship between viral size and genome size (Fig. 3). However, the possibility of horizontal gene transfer cannot be ruled out in large size DNA viruses despite overestimation [29]. Hence we conclude that large size viruses can cause fatal outbreaks with low frequency. Viral size and emerging infectious diseases Viruses with size P105 nm3 (SARS, ebola, marbug, hendra, nipah and monkey pox) have a greater potential to cause emerging infectious diseases than viruses with size <105 nm3. Our information on emerging infectious diseases did not include influenza virus despite its association with pandemicity. In the view of viral size, inclusion of influenza virus (>105 nm3) will only strengthen our hypothesis. The emergence of infectious diseases is determined by genetic variation of virus, environmental and social factors [30]. In our study, the genome size is an important viral factor that had a strong correlation with viral size. This made us hypothesize that the niche for these viruses (size P105 nm3) may not be vacant for re-emergence. Viral size and transmission route Majority of the enteric pathogens are restricted to small size probably due to protective geometric response to lethal gastric acid and other antiviral defenses in the alimentary tract. Respiratory route is the most promiscuous transitory route for diverse viral sizes (Fig. 5). Hence explosive outbreaks by respiratory pathogens are conceivable. In future, size restricted transmission routes should be considered in epidemiological outbreak modeling and control strategies. In conclusion, viral geometry plays a substantial role in transmission dynamics, disease outbreaks and outcomes. Hence, appropriate experimental models and functional studies are essential to unravel geometric implications of viruses. Acknowledgements We owe a great debt to our unseen teachers: Prof. Stephen Jay Gould (life’s Grandeur), Prof. Richard Dawkins (Ancestor’s tale) and Prof. Brian Goodwin (How the leopard changed its spots) who inspired us to conceive this idea through their books. We thank Dr. Susan, Dr. Sitara Rao and Dr. Mahesh Moorthy for their perusal, appreciation and admonishment. We thank the department of Clinical Virology and Christian Medical College, Vellore for their support at every stages of this manuscript preparation.
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
References [1] Buchen-Osmond C. Taxonomy and classification of viruses. Man Clin Microbiol 2003;8:1217–26. [2] Nandhagopal N, Simpson AA, Gurnon JR, et al. The structure and evolution of the major capsid protein of a large. Lipid-containing DNA virus. Proc Natl Acad Sci USA 2002;99:14758–63. [3] Kuhn RJ, Zhang W, Rossmann MG, et al. Structure of dengue virus: implications for flavivirus organization, maturation, and fusion. Cell 2002;108:717–25. 0 [4] Zhao R, Pevear DC, Kremer MJ, et al. Human rhinovirus 3 at 3.0 A Å resolution. Structure 1996;4:1205–20. [5] Nathanson N. Epidemiology. In: Knipe DMH, Howley PM, editors. Field’s virology, vol. 1. Philadelphia: Wolters Kluwer; 2007. p. 423–46. [6]. [7] Peters CJ. Emerging viral diseases. In: Knipe DMH, Howley PM, editors. Field’s virology, vol. 1. Philadelphia: Wolters Kluwer; 2007. p. 605–25. [8] Racaniello VR. Picarnaviridae: the viruses and their replication. In: Knipe DMH, Howley PM, editors. Field’s virology, vol. 1. Philadelphia: Wolters Kluwer; 2007. p. 795–838. [9] Young NS, Brown KE. Parvovirus B19. N Engl J Med 2004;350:586–97. [10] Hammond MM, Timoney PJ. An electron microscopic study of viruses associated with canine gastroenteritis. Cornell Vet 1983;73:82–97. [11] Lindner J, Modrow S. Human bocavirus – a novel parvovirus to infect humans. Intervirology 2008;51:116–22. [12] Glass RI, Bresee J, Jiang B, et al. Gastroenteritis viruses: an overview. Novartis Found Symp 2001;238:5–19 [discussion 19–25]. [13] Wilhelmi I, Roman E, Sanchez-Fauquier A. Viruses causing gastroenteritis. Clin Microbiol Infect 2003;9:247–62. [14] Holmes EC, Twiddy SS. The origin, emergence and evolutionary genetics of dengue virus. Infect Genet Evol 2003;3:19–28. [15] Nelson MI, Viboud C, Simonsen L, et al. Multiple reassortment events in the evolutionary history of H1N1 influenza A virus since 1918. PLoS Pathog 2008;4:e1000012.
137
[16] Sharma S, Mukkur TK, Benson HA, Chen Y. Pharmaceutical aspects of intranasal delivery of vaccines using particulate systems. J Pharm Sci 2008. [17] Ascenzi P, Bocedi A, Heptonstall J, et al. Ebolavirus and marburgvirus: insight the Filoviridae family. Mol Aspects Med 2008;29:151–85. [18] Feldmann H, Wahl-Jensen V, Jones SM, Stroher U. Ebola virus ecology: a continuing mystery. Trends Microbiol 2004;12:433–7. [19] Bachmann MF, Kundig TM, Kalberer CP, Hengartner H, Zinkernagel RM. How many specific B cells are needed to protect against a virus? J Immunol 1994;152:4235–41. [20] Klasse PJ, Sattentau QJ. Occupancy and mechanism in antibody-mediated neutralization of animal viruses. J Gen Virol 2002;83:2091–108. [21] Xiang SD, Scholzen A, Minigo G, et al. Pathogen recognition and development of particulate vaccines: does size matter? Methods (San Diego, Calif) 2006;40:1–9. [22] Bao G, Bao XR. Shedding light on the dynamics of endocytosis and viral budding. Proc Natl Acad Sci USA 2005;102:9997–8. [23] Rejman J, Oberle V, Zuhorn IS, Hoekstra D. Size-dependent internalization of particles via the pathways of clathrin- and caveolae-mediated endocytosis. Biochem J 2004;377:159–69. [24] Scheerlinck JP, Greenwood DL. Virus-sized vaccine delivery systems. Drug Discov Today 2008;13:882–7. [25] Gao H, Shi W, Freund LB. Mechanics of receptor-mediated endocytosis. Proc Natl Acad Sci USA 2005;102:9469–74. [26] Basler CF, Mikulasova A, Martinez-Sobrido L, et al. The Ebola virus VP35 protein inhibits activation of interferon regulatory factor 3. J Virol 2003;77:7945–56. [27] Reid SP, Leung LW, Hartman AL, et al. Ebola virus VP24 binds karyopherin alpha1 and blocks STAT1 nuclear accumulation. J Virol 2006;80:5156–67. [28] Eaton BT, Broder CC, Middleton D, Wang LF. Hendra and nipah viruses: different and dangerous. Nat Rev 2006;4:23–35. [29] Monier A, Claverie JM, Ogata H. Horizontal gene transfer and nucleotide compositional anomaly in large DNA viruses. BMC Genomics 2007;8:456. [30] Nichol ST, Arikawa J, Kawaoka Y. Emerging viral diseases. Proc Natl Acad Sci USA 2000;97:12411–2.
Contents lists available at ScienceDirect
Medical Hypotheses journal homepage: www.elsevier.com/locate/mehy
Why does size matter for viruses – A new paradigm on viral size Gnanadurai John Fletcher a,*, Solomon Christopher b, Manu Gnanamony a,1 a b
Department of Clinical Virology, Christian Medical College, Vellore 632004, Tamil Nadu, India Department of Biostatistics, Christian Medical College, Vellore 632002, Tamil Nadu, India
a r t i c l e
i n f o
Article history: Received 17 February 2009 Accepted 21 February 2009
s u m m a r y Size and shape are the two immutable laws that govern all life forms including viruses. In this study we postulate and evaluate the hypothesis that there exists a strong association between viral geometry and features of viral disease outbreaks. Data on viral disease outbreaks were retrieved from WHO and CDC public domains for a period of twelve years to assess the relationship between viral size and epidemiological factors such as number of outbreaks, case fatality rate, proportion of emerging infectious diseases and transmission routes. We observed a significant correlation between viral size and frequency of disease outbreaks (q = 0.82, p = 0.004), case fatality rate (q = 0.48, p = 0.03) and genome size (r = 0.79, p < 0.001). Viral sizes were significantly different among diverse transmission routes (p < 0.001). The proportion of emerging infectious diseases were significantly different between viruses with size <105 and P105 nm3 (21% vs 64%, p = 0.046). In conclusion, this preliminary evidence shows that viral size plays a substantial role in the epidemiology of viral diseases. Our data suggests that small size viruses are associated with more number of outbreaks than large size viruses. Large size viruses are associated with high case fatality rate and can be potential emerging pathogens. Viral size may be crucial for niche selection and specified transmission routes in the susceptible host. Hence, viral geometry should not be neglected in epidemiology and modeling of viral diseases, and planning vaccine strategies. Ó 2009 Elsevier Ltd. All rights reserved.
Introduction
Development of hypothesis
Our biosphere represents assorted size ranges of organic forms from the smallest virus and bacteria to the largest metazoans (whales and sequoias). Today, we recognize more than 1550 virus species with a size range of 25–250 200 200 nm, which display a wide range of host specificity [1]. What intrigue us about viruses are their specific size, symmetry and shape. What led parvovirus to be small and poxvirus to be a giant? Do diverse viral sizes demand any teleological explanation in biology? Modern biology deals with diverse viral sizes for tertiary reasons and has completely ignored its biological importance. However, recent studies on viral structures have revolutionized the field of virology [2–4]. Viruses are known to cause wide ranges of diseases from selflimited infection to fatal disease. Epidemiology of viral diseases are determined by immunity, proportion susceptible, proportion infected, incubation period, generation time and transmission patterns [5]. Here, we present a strong case for association of viral sizes with frequency of disease outbreaks, case fatality rate (CFR), emerging infectious diseases (EID) and transmission routes.
Viruses display wide ranges of sizes and shapes which exist so to their advantage by natural selection. Viral size has been known to influence its entry and exit in susceptible cells. The concept on viral size has also been exploited for size dependent vaccine delivery systems. In addition, we observed that all gastroenteritis causing viruses are small in size. Hence, we hypothesize that there is a strong association between viral geometry and features of viral disease outbreaks.
* Corresponding author. Tel.: +91 416 2282616, mobile: +91 9487 629620; fax: +91 416 2212102. E-mail addresses: fl[email protected] (G.J. Fletcher), Solomon.christopher@ gmail.com (S. Christopher). 1 Tel.: +91 416 2282616. 0306-9877/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.mehy.2009.02.029
Evaluation of hypothesis To ascertain the association between the viral geometry and certain features of viral disease outbreaks, we compared viral size (volume) with frequency of disease outbreaks, case fatality ratio (CFR), ability to cause emerging infectious diseases and transmission routes. Data collection The data on disease outbreaks for a period of 12 years (January 1996–December 2007) were collected from WHO [6]. Information was specifically collected on viral disease outbreaks with reference to the number of outbreaks, number infected per outbreak and number of deaths. The total number of infections and deaths were
134
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
collected from all successive updates provided for each outbreak. Information on viral size were obtained from ICTV [1]. Data on food borne outbreaks were obtained from CDC (http://www.cdc.gov/ mmwr/preview/mmwrhtml/ss5510a1.htm?s_cid=ss5510a1_e) for a period of five years (1998–2002) for which complete information was available. Data on viruses causing emerging infectious diseases for a period of 11 years (1995–2005) were collected from Field’s Virology [7]. Transmission routes for 22 medically important viral families were collected and categorized into feaco-oral, vector borne, blood borne, respiratory, sexual, vertical and zoonotic (Field’s Virology, 2007). Viruses that shared more than one transmission route were represented in all the respective routes. To standardize viral size for different shapes, volume of viral particle was calculated based on the shape of virus. Since viral size is provided as a range, mid size value was used in the calculation of volume. The following geometric formulae were used for calculation: for spherical virus, V = 4/3pr3; for cylindrical virus, V = pr2l; p and for icosahedral virus, V = 5/12(3 + 5)a3. An overall CFR was calculated for each virus as the ratio of number of deaths to the total number infected by combining data from all outbreaks. Viruses that caused outbreaks resulting in at least one fatal case were included for analysis. Due to this reason, buffalo pox (April 1996), coxsackie virus (1996, 1997), chikungunya (2006), rotavirus (2000), and astrovirus (2002) were removed from CFR analysis. Statistical analyses All statistical analyses were performed using SPSS 11.0 for windows. Spearman’s rank correlation was used to assess the relationship between viral size and, frequency of disease outbreaks and case fatality rate. Pearson’s product–moment correlation was used for assessing the relationship between viral and genomic size on logarithmic scale. Kruskal Wallis test was used to test for differences in viral size among various transmission routes. Fisher exact test was used to assess the association between emerging viral pathogens and viral sizes. Results To study the association of viral size with disease outbreaks, data for a period of 12 from WHO and five years from CDC were analyzed. During this period, viruses belonging to 13 families caused 927 outbreaks. List of viruses in each family implicated in outbreaks is shown in Table 1. In these outbreaks, there was a significant negative correlation between viral size (mid-volume in nm3) and frequency of disease outbreaks (q = 0.86, p = 0.002) (Fig. 1). We observed a similar trend in sub-group analysis of
Fig. 1. Relationship between viral size and frequency of disease outbreaks.
viruses that shared same transmission route (arthropod and rodent borne viruses) (q = 0.700, p = 0.188) (Fig. 2). The astrovirus outbreak (single) was removed from analysis (viral size vs frequency of outbreaks) due to underreporting. However, sensitivity analysis including this outbreak did not alter the results. To understand the prominence of outbreak frequencies with viral sizes, a cut-off of 14 (median) outbreaks was set. In this, Picornaviridae, Flaviviridae, Caliciviridae, Orthomyxoviridae and Bunyaviridae were small in size (104–105 nm3) except Filoviridae (106 nm3). We observed a significant negative correlation between genome size (in base pairs) and frequency of disease outbreaks (q = 0.82, p = 0.004). Significant linear relationship was observed between viral size and genome size (in logarithmic scale) among viruses that caused outbreaks (r = 0.79, p < 0.001) (Fig. 3). There was a positive correlation between viral size and case fatality ratio (q = 0.479, p = 0.03). Highest CFR was observed in outbreaks caused by marburg (84%), nipah (60%) and ebola (52%) viruses and the lowest with norovirus (0.004%), hantavirus (0.08%) and dengue virus (0.2%). The proportion of emerging infectious diseases between the two viral size groups (<105 and P105 nm3) were significantly different (21% vs 64%, p = 0.046) (Fig. 4). Majority of viruses that were implicated as emerging viral pathogens had size P105 nm3 which include SARS, ebola, marbug, hendra, nipah and monkey pox. The distribution of viral sizes among different transmission routes is shown in Fig. 5. Viral sizes were significantly different among diverse transmission routes (p < 0.001). Almost all viruses that had enteric route of transmission were smaller than 105 nm3 and most viruses which had zoonotic mode of transmission were
Table 1 List of viruses under each viral family for 12 year outbreak. Viral family
Virus
Picornaviridae Astroviriridae Flaviviridae Caliciviridae Orthomyxoviridae Bunyaviridae Reoviridae Arenaviridae Togaviridae Coronaviridae Filoviridae Paramyxoviridae Poxviridae
Enterovirus (3), poliovirus (20), coxsackie virus (2), hepatitis A virus (50) Astrovirus (1) Dengue virus (21), yellow fever (55), west Nile fever virus (4), Japanese encaphilitis (4), St. Louis encaphilitis (1) Norovirus (657), hepatitis E virus (2) Influenza (11), avian influenza (23) Crimean Congo virus (8), Hantaan virus (4), Rift valley virus (10) Rotavirus (1) Lassa fever (9) Chikungunya (7), O’nyong nyong (1) SARS (6) Ebola virus (16), margbug virus (3) Measles virus (4), hendra virus (1), nipah virus (1) Buffalo poxvirus (1), monkey poxvirus (1)
The numbers in brackets indicate the number of outbreaks.
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
135
Fig. 2. Relationship between viral size and frequency of outbreaks in Arboviruses.
Fig. 5. Association between viral size and transmission routes.
Viral size and frequency of outbreaks
Fig. 3. Comparison of viral size and genome size.
Fig. 4. Association between viral size and emerging infectious diseases.
above a size of 105 nm3. However, viruses that had respiratory mode of transmission showed diverse size ranges (104–106 nm3). Consequences of hypothesis and discussion Viruses are known to infect all life forms and cause a wide spectrum of diseases. The epidemiology of viral disease is complex and multifaceted. The quantification of disease occurrence in relation to viral geometry may offer useful insights into the epidemiology of these diseases. Data on disease outbreaks were retrieved from WHO because of documentation accuracy, accessibility and wide geographic coverage. Lack of WHO data on enterically transmitted viruses were compensated by data from CDC website.
Of the five viral families that caused large outbreaks which clustered on the left wall of the graph (Fig. 1), Picornaviridae and Caliciviridae were two enterically transmitted viral families that were small. At this juncture, we are unable to recall any large size viruses that can be definitely transmitted through enteric route and cause gastroenteritis. To cause such large frequencies of enteric outbreaks, viruses (Picornaviridae and Caliciviridae) must be small, non-enveloped and retain infectivity at pH 3.0 and 2.7, respectively. Hence mutation and recombination are the only strategies that are viable for small viruses to maximize their diversity and adaptability. Like other RNA viruses, picornaviruses relish in their high error rates and survive on threshold of error catastrophe [8]. This made us speculate that all enterically transmitted viruses have undergone size dependent niche selection to infect the small intestine which is the largest lymphoid organ in the body. Parvovirus is the smallest virus that is transmitted by non-enteric route (respiratory) in humans and causes fifth disease, transient aplastic crisis and other systemic diseases [9]. However, studies have shown that canine parvoviruses, a distant relative of human parvovirus is transmitted enterically [10]. In addition, bocavirus, a classified member of Parvoviridae is known to be transmitted by respiratory and enteric routes [11]. Viral agents like picobirnavirus, aichi virus and enterovirus 22 that are rarely implicated in gastroenteritis are small in size [12] which corroborates our findings. Hence, the independent etiological role of torovirus (large size) in gastroenteritis is doubtful. Astrovirus is another small, enterically transmitted virus associated with outbreaks of gastroenteritis [13]. But, both in WHO and CDC outbreak information, the frequencies of astroviruses outbreaks were under reported, possibly due to limitations in screening, diagnostic transition and occurrence of co-infections. Additionally, we speculate that competition among many small size viruses for its niche (small intestine) may be an additional factor for such low outbreak frequency. In this context, any future vaccine intervention for major gastroenteritis viruses (rotavirus and Norwalk) may be a tipping point for astrovirus outbreaks. Flaviviridae and Orthomyxoviridae were other viral families that had outbreak frequencies above the calculated cut-off (median) and clustered on the left wall of the graph. The size ranges of these viral families were 104 and 105 nm3, respectively. Flaviviridae is a small arthropod borne virus family responsible for 84 independent outbreaks. On subset analysis of arboviruses, same size restricted
136
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
trend with outbreak frequencies were observed (q = 0.700) which strengthens our findings. Despite vaccination, yellow fever is still an emerging threat because of high sequence diversity, selection pressure with differing vectors and other ecological factors. Dengue virus, a member of Flaviviridae shows high genetic variability, variation in vector population sizes and strain replacement/extinction potential [14] which are factors commonly associated with large outbreaks. In addition to climatic seasonality, arthropod biology (quiescence, synchrony, vectorial capacity) and human behavior; geometry (viral size) may additionally determine the complex arbovirus epidemiology. Orthomyxovirus, a relatively small respiratory pathogen (105 nm3) is of special concern for its pandemicity and fatality. In addition to reassortment and recombination, the size advantage of influenza virus can have additive effects on influenza pandemics [15]. A parallel drawn from nanoparticle study [16] made us speculate that size of influenza virus may influence replication and nature of immune response in influenza diseases. All large size viruses (Togaviridae, Coronaviridae, Paramyxoviridae and Poxviridae) caused very low outbreaks except Filoviridae (Fig. 1). Ebola and marburg are known to cause haemorrhagic infections with high fatality rates [17]. Geographical location, elusive reservoir, other ecological factors are shown to be associated with Filoviridae outbreaks [18]. We speculate that unique viral size, shape and pleomorphism may additionally determine certain virological properties. Eradication of small pox was possible due to the absence of non-human reservoir, single stable serotype, availability of effective vaccine and no asymptomatic carriage of virus. We speculate, in addition to these factors, geometric (big size) disadvantage could have been a crucial factor for small pox extinction. Though large size viruses can cause fatal outbreaks, size advantage for control is biologically possible, with vaccine as an effective arsenal. Virus specific neutralizing antibodies are crucial for effective humoral immune response in viral infections [19]. Stoichiometric determination showed that poliovirus require about an average of 9 monoclonal antibody (Ab) molecules and rhinovirus require 10–20 Ab molecules per virion to show 90% neutralization. Influenza and rabies virus require 100 and 130–350 IgG molecules for 90% neutralization [20]. In the occupancy model of multi-hit hypothesis, we speculate that viral size (surface area, nm2) may be crucial for efficient neutralization. This shows a possible relationship between geometry of viral size and the number of antibodies obligatory to confer neutralization. The concept of virus-like particles has revolutionized the field of vaccinology. Animal viruses have evolved to enter and exit the susceptible cells by generic cellular and virus-specific mechanisms. The basic differences of viral entries apart from obligate receptor–ligand interaction and alternate entry routes (caveolae and antibody) can be better explained by additional size differences seen in diverse viruses. Studies have shown that particle size influence the route of entry [21], dynamics of endocytosis and viral budding [22]. Particles less than 200 nm prefer clathrin mediated endocytosis whereas particles of 500 nm show caveolae-mediated entry [23]. Besides, studies have shown the biological relevance of viral sizes with modern applications of virus sized vaccine delivery system in vaccinology [24]. Studies have also shown that an optimal particle size exists for clathrin or caveolin independent cell entry or exit [25]. Viral size dependent theoretical and mathematical models should be developed to understand the challenging pathogenesis of hepatitis B virus, hepatitis C virus and human immunodeficiency virus. Hence, such curious qualitative and quantitative humoral and cell mediated response on diverse viral sizes may shed light into disease susceptibility and pathogenesis of challenging viral diseases.
Viral size and case fatality rate (CFR) CFR is a crucial indicator of disease severity during outbreaks which can be influenced by host factors and viral factors. In our study, high CFR is seen with large size viruses (marburg, nipah and ebola) possibly due to the presence of virulent genes. Ebola virus encodes VP35, a viral encoded type I antagonist that inhibits interferon regulatory factor (IRF-3) activation [26], VP24 that blocks interferon (IFN) signaling [27] and niche specific virulent genes can also contribute to its fatal pathogenesis. In nipah, the non structural proteins (P, V and W) have been shown to down regulate IFN signaling cascade [28]. Majority of large size viruses that were associated with emerging infectious diseases also had high CFR. Hence considering the genome economy, expansion of virulent genes is the only possibility for fatal pathogenesis associated with large size viruses. This is partly explained by our data that revealed a linear relationship between viral size and genome size (Fig. 3). However, the possibility of horizontal gene transfer cannot be ruled out in large size DNA viruses despite overestimation [29]. Hence we conclude that large size viruses can cause fatal outbreaks with low frequency. Viral size and emerging infectious diseases Viruses with size P105 nm3 (SARS, ebola, marbug, hendra, nipah and monkey pox) have a greater potential to cause emerging infectious diseases than viruses with size <105 nm3. Our information on emerging infectious diseases did not include influenza virus despite its association with pandemicity. In the view of viral size, inclusion of influenza virus (>105 nm3) will only strengthen our hypothesis. The emergence of infectious diseases is determined by genetic variation of virus, environmental and social factors [30]. In our study, the genome size is an important viral factor that had a strong correlation with viral size. This made us hypothesize that the niche for these viruses (size P105 nm3) may not be vacant for re-emergence. Viral size and transmission route Majority of the enteric pathogens are restricted to small size probably due to protective geometric response to lethal gastric acid and other antiviral defenses in the alimentary tract. Respiratory route is the most promiscuous transitory route for diverse viral sizes (Fig. 5). Hence explosive outbreaks by respiratory pathogens are conceivable. In future, size restricted transmission routes should be considered in epidemiological outbreak modeling and control strategies. In conclusion, viral geometry plays a substantial role in transmission dynamics, disease outbreaks and outcomes. Hence, appropriate experimental models and functional studies are essential to unravel geometric implications of viruses. Acknowledgements We owe a great debt to our unseen teachers: Prof. Stephen Jay Gould (life’s Grandeur), Prof. Richard Dawkins (Ancestor’s tale) and Prof. Brian Goodwin (How the leopard changed its spots) who inspired us to conceive this idea through their books. We thank Dr. Susan, Dr. Sitara Rao and Dr. Mahesh Moorthy for their perusal, appreciation and admonishment. We thank the department of Clinical Virology and Christian Medical College, Vellore for their support at every stages of this manuscript preparation.
G.J. Fletcher et al. / Medical Hypotheses 73 (2009) 133–137
References [1] Buchen-Osmond C. Taxonomy and classification of viruses. Man Clin Microbiol 2003;8:1217–26. [2] Nandhagopal N, Simpson AA, Gurnon JR, et al. The structure and evolution of the major capsid protein of a large. Lipid-containing DNA virus. Proc Natl Acad Sci USA 2002;99:14758–63. [3] Kuhn RJ, Zhang W, Rossmann MG, et al. Structure of dengue virus: implications for flavivirus organization, maturation, and fusion. Cell 2002;108:717–25. 0 [4] Zhao R, Pevear DC, Kremer MJ, et al. Human rhinovirus 3 at 3.0 A Å resolution. Structure 1996;4:1205–20. [5] Nathanson N. Epidemiology. In: Knipe DMH, Howley PM, editors. Field’s virology, vol. 1. Philadelphia: Wolters Kluwer; 2007. p. 423–46. [6]
137
[16] Sharma S, Mukkur TK, Benson HA, Chen Y. Pharmaceutical aspects of intranasal delivery of vaccines using particulate systems. J Pharm Sci 2008. [17] Ascenzi P, Bocedi A, Heptonstall J, et al. Ebolavirus and marburgvirus: insight the Filoviridae family. Mol Aspects Med 2008;29:151–85. [18] Feldmann H, Wahl-Jensen V, Jones SM, Stroher U. Ebola virus ecology: a continuing mystery. Trends Microbiol 2004;12:433–7. [19] Bachmann MF, Kundig TM, Kalberer CP, Hengartner H, Zinkernagel RM. How many specific B cells are needed to protect against a virus? J Immunol 1994;152:4235–41. [20] Klasse PJ, Sattentau QJ. Occupancy and mechanism in antibody-mediated neutralization of animal viruses. J Gen Virol 2002;83:2091–108. [21] Xiang SD, Scholzen A, Minigo G, et al. Pathogen recognition and development of particulate vaccines: does size matter? Methods (San Diego, Calif) 2006;40:1–9. [22] Bao G, Bao XR. Shedding light on the dynamics of endocytosis and viral budding. Proc Natl Acad Sci USA 2005;102:9997–8. [23] Rejman J, Oberle V, Zuhorn IS, Hoekstra D. Size-dependent internalization of particles via the pathways of clathrin- and caveolae-mediated endocytosis. Biochem J 2004;377:159–69. [24] Scheerlinck JP, Greenwood DL. Virus-sized vaccine delivery systems. Drug Discov Today 2008;13:882–7. [25] Gao H, Shi W, Freund LB. Mechanics of receptor-mediated endocytosis. Proc Natl Acad Sci USA 2005;102:9469–74. [26] Basler CF, Mikulasova A, Martinez-Sobrido L, et al. The Ebola virus VP35 protein inhibits activation of interferon regulatory factor 3. J Virol 2003;77:7945–56. [27] Reid SP, Leung LW, Hartman AL, et al. Ebola virus VP24 binds karyopherin alpha1 and blocks STAT1 nuclear accumulation. J Virol 2006;80:5156–67. [28] Eaton BT, Broder CC, Middleton D, Wang LF. Hendra and nipah viruses: different and dangerous. Nat Rev 2006;4:23–35. [29] Monier A, Claverie JM, Ogata H. Horizontal gene transfer and nucleotide compositional anomaly in large DNA viruses. BMC Genomics 2007;8:456. [30] Nichol ST, Arikawa J, Kawaoka Y. Emerging viral diseases. Proc Natl Acad Sci USA 2000;97:12411–2.