Discussion

The general result of the simulations suggests that the increase in the rate of evolutionary novelty emergence, reproductive rate, and propagule size influence positively the success of colonization of new hosts by a novel pathogen population (Fig. 2). However, within the scope of the simulations, the different values of the explored parameters resulted in distinct impacts on this success. By far, the most significant impact was observed for the number of propagules; even not adapting to the selection of the new host, due to the complete absence of emergence of novelties imposed by the model, an initial population composed of 10 colonizers resulted in a significant increase of the probability to thrive and persist under suboptimal fitness, even in hosts representing relatively small compatibility (d0 \(\approx\)1).
Indeed, propagule pressure is extensively known to positively influence the colonization of new host species (May et al. 2001; Hatcher et al. 2012) - or geographic areas and corresponding communities in the case of invasive species (Sax et al. 2007; Lockwood et al. 2009; Cassey et al. 2018). Large propagule sizes are usually linked with the reduction of consequences of demographic (e.g. stochasticity and Allee effects) (Hufbauer et al. 2013) and genetic (founder’s effect) (Simberloff 2009; Roman and Darling 2007) processes observed in small population size during changes in ecologic and geographic distribution. Since every simulation involving variation in propagule size used a low rate of emergence of evolutionary novelty and the relative fitness of propagules were kept unchanged (same d0 despite qualitative differences in the combination of loci), the advantage conferred by increasing propagule sizes during colonization appears to be associated with demographic issues, most likely stochastic, as we did not model social collaborative processes nor limitation in the encounter of mates during reproduction (see Hufbauer et al. 2013). However, the rate of increase in colonization success according to the imposed distance of the compound phenotype optimum of the new host diminishes with increasing values of the propagule size, not displaying a direct linear relationship (Fig. 2c).
Although less evident than the simulations with variable propagule size, increases in reproductive rate in less than 10 – fold (from 1.5 to 7.5) resulted in a more expressive increase in the probability of successful colonization than 1000-fold increases in the rate of emergence of novelties (from μ =10-6 toμ =10-3). This is an unexpected result, especially considering that the emergence of new associations - such as infectious diseases - is often linked by many to high mutation rates of the consumer associate (Pepin et al. 2008; Selman et al. 2012; Viana et al. 2015). Hence, our results indicate that the rate of emergence of evolutionary novelties alone (e.g. mutation rates for simple organisms such as viruses) has secondary importance in the colonization of new host species, as suggested in Araujo et al. (2015) and implicitly by the Stockholm Paradigm (Brooks and Hoberg 2007; Brooks and Boeger 2019; Brooks et al. 2019; Agosta and Brooks 2020). The accumulation of accessible historical information - termed the information spaceby Brooks and Agosta (2012) and Agosta and Brooks (2020) - is of greater importance for the events of host-repertoire expansion (i.e. the evolutionarily process that precedes what is known as host-switching; see Braga et al. 2018). It is the accumulation of heritable information by preceding generations (and ancestors) and its retention in the biological entities (i.e. populations, species) through time (=phylogenetic conservatism) that will determine the ability of lineages to endure ecological and environmental changes or to take advantage of opportunities (e.g. explore new resources, new habitats). Since compatibility (i.e. the distance to the actual host optimum) varies within individuals of a diverse pathogen population, regions of suboptimal fitness in the ancestral host - albeit potentially at low frequency in the population - may contain pathogen variants that are capable of reaching more distant (= more different) resources (new hosts) than originally higher-fitness variants (see also Araujo et al. 2015; Brooks et al. 2019). Consequently, under this scenario, actual rates of emergence of new inheritable evolutionary novelties (e.g. mutations) are less important than the number of individuals colonizing the new host (=propagule pressure), the rate of reproduction, and the degree of the variability in the original donor population.
When all three parameters considered are maximized, the simulations generate pronounced synergism (grey line in Fig. 2). The fact that this combination of values likely compares to those observed for viruses, particularly among RNA-viruses (Holmes 2009) is especially significant in understanding the evolution of this group of organisms and the corresponding emergence of infectious diseases. This outcome is compatible with the conclusions of Geoghegan, Duchêne, and Holmes (2017) that “cross-species transmission is a near universal feature of the viruses …, with virus-host co-divergence occurring less frequently…” For instance, continuous oscillations of host species were suggested as an intrinsically biological feature of coronaviruses (Menachery et al. 2017), but it is likely a property of viruses in general and perhaps of pathogenic bacteria as well. It is, thus, understandable that viruses and bacteria are the most common groups of organisms associated with emergent infectious diseases (Cleaveland et al. 2001; Woolhouse and Gaunt 2007; Gubler 2010; Pękala-Safińska 2018; Duarte-Neto 2019).
Since we expect that in the real-world representatives of the variants of pathogens are continuously exploring accessible resources (e.g. host species) (Brooks et al. 2019; Agosta and Brooks 2020) the emergence of new associations - or colonization of new environments - is expected when suitable matching (likely imperfect rather than perfect) between requirements of the pathogen, the resource (i.e. host properties), and/or environmental conditions meet. Therefore, the original host species represents an imperfect reference - but, perhaps, the only one accessible at this time - to describe the relative quality and the distance of the new resources to the pathogen. Phylogenetic distance between the host species involved in the host range expansion appears, within limits, to estimate the multidimensional space of traits that influence the compatibility of host and a specific pathogen lineage (Martiny et al. 2013; Braga et al. 2015; Streicker et al. 2010; Gilbert and Webb 2007). Since the resources defining compatibility vary according to both host and pathogen species, phylogenetic distances appear to be the only accessible proxy for the value ofd0 , but it should be considered parsimoniously because evolutionary convergence of resources (Brooks and McLennan 2002) and the variability of the pathogen and hosts may influence also the outcome of the colonization attempts (see for instance Boeger et al. 2005; Araujo et al. 2015).
The results of the present simulations are also fundamental to expand the understanding of the role of ecological fitting (Janzen 1985b; Agosta 2006; Agosta and Klemens 2008) on the evolution of ecological changes. As suggested previously by Araujo et al. (2015), newly established populations of pathogens may survive for many generations in a host even in the absence of adaptations. This is a more extreme scenario of what Darwin called the survival of the adequate (Agosta and Brooks 2020; Brooks and Agosta 2012). By surviving under these “suboptimal” conditions, pathogens may expand their temporal window for the “right” novelty to present itself and allow an increase in the population’s fitness (adaptation) following the ecological change. For instance, Antia et al. (2003), modeling a scenario of colonization similar to the present simulations, suggested that early values of R0 of a new pathogen may evolve towards an R0>1 subsequently, under the selective pressure of the newly colonized host. However, after exploring the available Sloppy Fitness Space of the pathogen population (Agosta and Klemens 2008; Agosta et al. 2010; Brooks et al. 2019; Agosta and Brooks 2020), evolutionary novelties emerge randomly in the consumer species (the pathogen). Thus, the perfect match may never happen (i.e. a perfectly fit association) despite the influence of selection and the consumer may remain in a situation of continuous suboptimal fitness regarding its host species, a scenario analogous to that proposed by Sax et al. (2007) for invasive species.
Another additional perspective is that the newly established population of pathogens, although unchanged in its diversity due to the absence or limited emergence of novelties (phenotypic or genetic), may also expand the window of opportunity to encounter additional hosts representing more or simply adequate resources solely by inhabiting a host species with distinct ecological interactions with the surrounding environment. For instance, by depicting dissimilar behaviors, the adequate new host species may increase the probability of the new pathogen population to encounter other potential hosts not previously available (considering the ecology of the original host species) through a process that likely comprises one of the mechanisms of colonization of new hosts species. This is an empirically recognized process associated with many cases of emergence of new symbiotic associations - contemporary (Brown 2001) and historical (Braga et al. 2015) - including one of the possible pathways of SARS-CoV-2 to humans during the emergence (Ji et al. 2020; Zhang et al. 2020).
In the case of SARS-CoV-2, the scenario is even more worrisome since humans became one of the “stones” in the process of host-repertoire expansion by stepping stones (Braga et al. 2015; Brooks et al. 2019). COVID19 has rapidly expanded to almost every part of the planet, providing opportunities for the virus to colonize other human populations and animal species. Presently, pets – ferrets, cats, and dogs – and captive wild animals – such as minks, tigers, lions, macaques, Syrian hamsters, tree shrews, marmosets, and Egyptian fruit bats (Gryseels et al. 2020; Lin et al. 2020) are known empirically to be compatible hosts while a much greater range of host species has been suggested through modeling (Damas et al. 2020) - from old-world monkeys to anteaters. While many of the presently known compatible host species are not seriously affected by the virus, they certainly represent unique selective pressures and opportunities for broader dissemination through ecological fitting (as suggested above). Hence, we may anticipate influences connected with the acquisition of new host species on the genetic make-up of SARS-CoV-2 – among others, it may result in an increase in its overall genetic variability and/or on the emergence of unique haplotypes in isolated host populations (as suggested also by Franklin and Bevins 2020). Indeed, the nature of RNA-viruses replication influences by host and geographic expansion and isolation are already known to generate new variants (Franklin and Bevins 2020) with dissimilar potential virulence to humans. Such evolutionary changes may result in new strains of the viruses with the ability to generate diseases with symptomatic, virulence, and epidemiological characteristics distinct from the original strains (see Jerzak et al. 2007; Bordería, Stapleford, e Vignuzzi 2011). This epidemiological scenario is complicated by the accumulation of evidence suggesting that SARS-CoV-2 may take the opposite path (retro-colonizing humans), a situation already recorded among other coronaviruses for the Siberian musk deer (Moschus moschiferus) and ferrets (Mustela lutreola ) (Hadfield et al., 2018; Van Der Hoek et al., 2004). Hence, despite the recognition that these retro-colonization events are likely rare (de Morais et al. 2020), they cannot be simply ignored in epidemiological surveillance systems. The significance of such scenarios and outcomes is further heightened given the currently expected limited capacity for viruses to re-infect humans from domestic or synanthropic wildlife sources. Thus, potential pathways are not under active surveillance.
The simulations revealed yet another aspect of this host-exploration dynamics that makes the above-proposed scenario of retro-colonization of humans particularly important in health surveillance for EIDs. The simulations strongly suggest that at higher values of the rate of emergence of evolutionary novelties (e.g. mutation rates for viruses), the genetic profile of the pathogen - although changing qualitatively and quantitatively under the selective pressure of the new host resource - putatively retain “ancestral” variants at low frequency despite lower supposed fitness (Fig. 4B, C). This outcome provides theoretical support for the retention of the capacity of fast-evolving lineages to retro-colonize their previous host species/lineage by ecological fitting (Janz and Nylin 1998; Janz et al. 2001; Brooks et al. 2019 ). RNA viruses, such as SARS-CoV-2, are well known to evolve rapidly by mutation and hybridization (Holland et al. 1982), and the retention of variants may facilitate retro-colonizing of humans from recent spillover into other animal species. This perspective in phylogenetic conservatism analyzed along with available empirical data (such as that of Celorio-Mancera et al. 2016) and the real nature of novelty emergence (i.e. which includes more than just the idea that adaptation is solely associated with random mutations - Jablonka and Lamb 1995; Jablonka et al. 2014; Agosta and Brooks 2020) may provide a better understanding on the process of retro-colonization of the original host species/lineages. Hence, retro-colonization should be an important element in epidemiological monitoring (as suggested by Favoretto et al. 2019, Franklin e Bevins 2020, and González-Salazar, Stephens, e Sánchez-Cordero 2017), especially in cases of recent emergence and re-emergence of EIDs.
The combined results of this study provide further theoretical support for the assertion that “emerging infectious diseases are evolutionary accidents waiting to happen” (Brooks and Ferrao 2005). An increase in host-repertoire by pathogens, potentially associated with the emergence of a new infectious disease, is most likely to occur among closely related species of hosts, but it is also possible among distantly related hosts when the resource(s) is(are) convergent (see discussion on specilization in Brooks and McLennan 2002). Capacity is much larger than we can anticipate, and it is the opportunity of encounter (i.e. the breakdown in mechanisms for ecological isolation) that is a more essential determinant to the emergence of new associations (Araujo et al. 2015; Brooks et al. 2019; Agosta and Brooks 2020). And opportunities are more frequent during periods of environmental disruptions, many of which are associated with climatological fluctuations in the past (Hoberg and Klassen 2002; Brooks and Hoberg 2007; Hoberg and Brooks 2008, 2015; Hoberg et al. 2017). Climatological fluctuations usually change the permeability of pre-existing ecological barriers and promote shuffling in the composition of organismic communities, augmenting the rate of encounter of different host lineages, many of the same clade, fostering intense exchange in pathogens.
In general, climate oscillation and independent or accompanying environmental disruptions over evolutionary time have been a central determinant of opportunities for faunal mixing and pathogen exchange that have structured complex associations (Hoberg and Brooks, 2008). Climate and environmental disruption occur across scales and historically have had a substantial episodic behavior in the past (Hoberg et al. 2017). However, during what is now characterized as the Anthropocene, the outcomes of environmental disruption have become significantly more prevalent due to globalization, other human-associated actions, and also to climate change, which promote movements of wildlife, humans, and domestic species into new geographic range (Wilson 1995; Brooks and Boeger 2019). As a consequence, we expect EID’s to become even more frequent in the years to come (Brooks et al. 2014). We have little control over capacity, but we can, to a certain level, monitor, avoid, and minimize the opportunity of encounter between parasites and compatible host species. This is the principle of the D.A.M.A. protocol (Brooks et al. 2014, 2019; Hoberg and Brooks 2015; Brooks and Boeger 2019).
However, even with an effective D.A.M.A. protocol established, the task to avoid the emergence of new diseases is especially difficult, considering available empirical information. Many of the most significant events in the history of life, and in the history of EID’s, are likely the result of unpredictable incidents when compatible biological entities unexpectedly meet (opportunity). Attempts to generate new associations (hosts and pathogens, in this case) likely occur continuously, most being unsuccessful. However, a single successful event may perpetuate the emerged association through evolution and have a significant influence on the future diversifications of the associates. That was likely the case for well-known symbioses, such as those of proto-eukaryotic cells and mitochondria, eukaryotic cells, and chloroplasts but also for many recent EIDs, such as HIV, Ebola, Dengue, Zika, Chikungunya, and, of course, Covid19.
Perhaps the final message from the empirical information accumulated from the recent emergence of infectious diseases and the dynamics revealed from the theoretical framework of the Stockholm Paradigm (Brooks et al. 2019) and associated evolutionary models (Araujo et al. 2015) is that we cannot “lower our guard”. These events are evolutionarily dynamic processes, with pathogens incessantly exploring the space of compatible host species (Brooks et al. 2019). And we – and other domesticated species -are certainly one of the most abundant, available, ecologically diverse, and widespread species of potential host on this planet.