Efficient assays to quantify the life history traits of algal viruses

ABSTRACT Although viral life history traits—the traits that determine reproductive and survival success—are critical to understanding the population dynamics and impact of viruses, they are understudied compared to molecular traits. This discrepancy is partly due to the challenge of phenotyping viral life histories. We developed improved methods to quantify the life cycles of chloroviruses, which are lytic double-stranded DNA viruses that infect unicellular “chlorella-like” algae. We modified classical one-step growth and survival assays by including flow cytometry and kinetic modeling, and applied these to four chlorovirus strains. Together, the modified assays quantified the full life cycle, including adsorption rate, probability of depolarizing the host cell, probability of releasing progeny virions, time until lysis, burst size, specific infectivity, and mortality rate of infectious virions. We also discovered that biphasic or “tailing” decay can occur in chloroviruses, and measured the fraction of infectious virions that resisted decay. The modified assays are more efficient than existing techniques, quantify more traits, and are applicable to other lytic viruses with quantifiable virions. The modified one-step growth assay is suitable for viruses with mutual exclusion, superinfection exclusion, or superinfection immunity; the modified survival assay can be applied to any lytic virus with quantifiable virions. IMPORTANCE Viruses play a crucial role in microbial ecosystems by liberating nutrients and regulating the growth of their hosts. These effects are governed by viral life history traits, i.e., by the traits determining viral reproduction and survival. Understanding these traits is essential to predicting viral effects, but measuring them is generally labor intensive. In this study, we present efficient methods to quantify the full life cycle of lytic viruses. We developed these methods for viruses infecting unicellular Chlorella algae but expect them to be applicable to other lytic viruses that can be quantified by flow cytometry. By making viral phenotypes accessible, our methods will support research into the diversity and ecological effects of microbial viruses.


Figure M1
. Life cycle of the chloroviruses, as assumed in our statistical models.See the text and Table 1 for definitions of the traits.Traits marked in blue describe independent life cycle steps; traits marked in red arise from combinations of two or more steps (gray box).Traits in bold are estimated from the modified one-step growth (mOSG) or modified survival (mS) assays; traits not in bold are derived by comparing the two assays.
Note that in this section, we distinguish the probability r (the probability that a depolarized cell releases infectious virions) from r' (the probability that a depolarized cell releases virions, infectious or not).This is done for completeness, but we assume the two probabilities are equal.Abbreviations: F, initial frequency; P, probability; TrN, truncated normal distribution.

Derivation of Eq. 1 (mOSG assay)
The free virion concentration at each time point was made up of unadsorbed virions and newly   ( ) released progeny virions, i.e.
(Eq.S1.1),   ( ) where δ is the dilution factor, V 0 is the concentration of unadsorbed virions at the end of the adsorption period, and V p is the concentration of progeny virions in the diluted suspension.V p is determined by the concentration of depolarized algae at the end of the adsorption period and after dilution A d /δ, the probability that a depolarized cell releases virions r', the proportion of cells that released virions over time F(t), and the average number of progeny virions per release b r (Fig. M1): (Eq.S1.2).   ( ) = This assumption is based on the finding that chloroviruses mutually exclude each other through depolarization (Greiner et al. 2009).Therefore we can assume that all depolarized cells behave like cells infected by one virion, no matter how many virions capable of depolarization virions are attached to them.This is supported by measurements of b r in single cells (Lievens et al. 2022), at least in the range of MOPs we use (very high MOIs can be associated with differences and dual infections, Van Etten et al. 1983, Tessman 1985, Chase et al. 1989).
V 0 is the concentration of unadsorbed virions at the end of the adsorption period (i.e. at t = 0, right before dilution).We assume that there are enough binding sites available for all virions (Meints et al. 1988), that all virions are capable of adsorbing, and that virions adsorb at a constant rate represented by the adsorption constant .The concentration of unadsorbed virions is then (Hyman and Abedon 2009, eq. 18.2), where A a is the algal concentration during the adsorption period, M is the MOP, A a *M is the free virion concentration at the start of the adsorption period, and t a is the duration of the adsorption period in minutes.
A d is the concentration of depolarized algae at the end of the adsorption period.We assume that adsorbed virions are randomly distributed across host cells, and that virions capable of depolarizing host cells adsorb at the same rate as virions incapable of depolarizing host cells.In that case, the average number of virions capable of depolarization that adsorb to each cell is This non-linear function reflects the fact that maximum 100% of the algae can be depolarized.For example, when d*M A is 0.5, the fraction of depolarized algae is expected to be 39% (Eq.S3.1).
Doubling the MOP (and thus doubling d*M A , Eq. S4) would increase that fraction to 63%.When the initial d*M A is 5, however, the fraction of depolarized algae is already 99% and doubling it would have little effect.When combined with the constant burst size (see above), this is the key insight that allows us to disentangle burst size from depolarization probability: b d determines the concentration of progeny virions at saturating MOPs, while d determines the saturation point (Supp.Fig. S10).
F(t), the proportion of lysed cells over time, deviates from an idealized 'step' pattern due to variation among the host cells (Rabinovitch et al. 1999).Determining the precise distribution of lysis time across the host population is not trivial (e.g.In this study, we set the lysis time truncation value to 0 for simplicity (including as a separate α α parameter had minimal effects on the parameter estimates, Supp.Fig. S7).

Derivation of Eq. 2 (mS assay)
The proportion of virus-positive wells over time follows a binomial distribution with 16 trials and a success probability of P(t).The success probability is the probability that at least one infectious virion was added to a well.It can be calculated using the Poisson distribution with expected number of events λ(t): If mortality follows the exponential decay model (following e.g.Cottrell and Suttle 1995, Noble and Fuhrman 1997, Demory et al. 2021), the expected number of infectious virions per well is Here the mortality term is e -m*t , where m is the constant mortality rate and t is the time in days.The number of infectious virions at t = 0 is given by the initial number of virions V (10 μl of 5 × 10 4 , 5 × 10 3 , 5 × 10 2 , and 5 × 10 1 virions/ml suspensions, i.e. 500, 50, 5, or 0.5 virions) and the initial proportion of infectious virions s.
However, in this and other work we found that Eq.S7 was a poor fit for many virus strains (Supp.Fig. S8).Upon investigation, these strains had higher mortality at the beginning of the assay than at the end.Therefore, we added an additional parameter p: (Eq.S8 decay exponentially.The probability that an infectious virion is persistent is p.This parametrization is the simplest extension of Eq.S7, and fit our data very well.We don't exclude that other possibilities, e.g. a model where the persistent fraction decays at a slower rate, could also fit the data well.
Plugging Eq.S8 into Eq.S6, the expected proportion of virus-positive wells is:

Derivation of Eqs. 3 & 4 (Comparison of mOSG and mS assays)
The probability that a virion is infections, i.e. that it can complete life cycle steps 1-3 (Fig. 1), is the product of three probabilities: the probability that a virion is capable of adsorption (step 1; assumed to be 1), the probability that an adsorbed virion is capable of depolarization d (step 2), and the probability that a depolarized cell is capable of releasing infectious virions r (step 3).The specific infectivity s describes the initial proportion of infectious virions in the mS assay, which is equivalent to the probability that any given virion is infectious.Thus (Eq.S9),  = 1 *  *  which produces Eq. 3.
The burst size per release of virions b r (see Eq. S1.2) can be calculated from the probability that a depolarized cell releases virions r' and the burst size per depolarized cell b d , since (Eq.S10.1)   = ' *   + (1 − ') * 0 (Eq.S10.2)   = ' *   (Fig. M1).Assuming that all bursts contain a mix of infectious and noninfectious virions, the probability that a depolarized cell releases virions r' is equal to the probability that a depolarized cell releases infectious virions r.Therefore (Eq.S10.3),   =  *   which produces Eq. 4.
b r reduces to b d , the burst size per depolarized cell (see Fig. M1)A d , and F(t) are described in more detail below.Note that we assume b r' and b d are constant, and not contingent on MOP.
where d is the proportion of virions capable of depolarizing a host cell (i.e. the depolarization probability when a virion is adsorbed) and M A is the multiplicity of adsorption (sensuHyman and Abedon 2009).The concentration of algae with at least one depolarization-capable virion attached can then be calculated using the Poisson distribution with expected number of events d*M A :The multiplicity of adsorption is the concentration of adsorbed virions (the initial virion concentration minus the unadsorbed virion concentration) divided by the concentration of algal cells: This function expresses the proportion of lysed cells at t hours after dilution, given a mean lysis time μ l , standard deviation σ l , and earliest possible lysis time α. ()