Epizootiological Modeling of Pandora neoaphidis Mycosis Transmission in Myzus persicae Colonies Initiated by Primarily Infected Alates

Pandora neoaphidis transmission was monitored within progenycolonies initiated by infected Myzus periscae alates individuallyflown for 1 to 5 h. Mycosis progress in the colonies was wellfitted (r² = 0.97) to a modified logistic or Gompertz modelthat included their flight distance, postflight survival time,premycosis fecundity, and primary infection rate as influentialvariables.

INTRODUCTION

Top

Introduction

Mycoses caused by aphid-pathogenic Entomophthorales such asPandora neoaphidis play important roles in natural control ofaphids worldwide (10, 14, 16, 20, 21, 24). Most of these fungimay survive adversity as resting spores in soil, whereas forP. neoaphidis, the most prevalent aphid pathogen, no restingspores have been discovered (9, 17). Although the resting sporesmay potentially initiate infection in aphids (18, 19, 22), itis difficult to understand the general phenomenon that P. neoaphidisis predominate in global aphid epizootics.

Aphids are highly able to disperse themselves by hovering flightover vegetation or by passive flight with winds in wide ranges(23). This ecological strategy enables holocyclic or anholocyclicspecies to readily locate suitable hosts (5). A hypothesis thatP. neoaphidis could be widely dispersed with the flight of theiralates was thus proposed (15) and has been proven by recentfindings that ca. 30% of thousands of air-trapped alates bearseveral species of fungal pathogens, including primarily P.neoaphidis (3, 7, 11). Infected aphids usually die from mycosisafter a few days of infection development inside host hemocoel(6, 12, 13). Whether or not infected alates during the limitedlatent period are capable of flying for dispersal and then survivingfor colonization and reproduction is of primary interest forunderstanding the process of mycosis transmission via contagiousinfection of progeny individuals in contact with the cadaversof infected mother alates or the spores actively dischargedfrom them. Computer-monitoring simulated flight experimentswith infected alates may help to measure those capabilities(4, 11).

As a holocyclic species in temperate areas or an anholocyclicspecies in tropical or subtropical zones, the green peach aphidMyzus persicae is globally distributed and is able to infestover 40 different plant families (2). These features make M.persicae an ideal host model for insight into the process ofmycosis transmission in aphid populations across vegetations.In this study, variables describing flight and postflight colonizationcapabilities of M. persicae alates infected by P. neoaphidiswere measured by means of numerous batches of simulated flights.Two epizootiological models were fitted to describe the developmentof their progeny colonies and within-colony mycosis transmission.

Flight, colonization, and mycosis transmission by infected alates.

Top

Flight, colonization, and...

Vigorous M. persicae alates ( $≤$ 2 days old) from caged plants at20 to 23°C and a 12 h-12 h light-dark cycle were exposedto a 1-h spore shower from sporulating mycelial mats of P. neoaphidisF98028 (8) and then were assumed to have received primary infection.Maintained overnight under moist conditions for infection development,the alates were individually fixed by their abdomens on a channelmill (11) with a dip of water-soluble arabic glue and then flownfor 1 to 5 h. The flight distance of each alate during a givenperiod was recorded automatically in a computer by a signalreceptor with data collected to a photoelectronic sensor onthe channel mill where the alate flew circumferentially as itswings vibrated (11). Immediately after flight, the alates weregently removed from the mills by dipping water onto the abdomento dissolve the glue and then were individually reared underthe same regimen for 14 days so as to initiate new colonieson detached cabbage leaves in petri dishes. Each leaf with hairyroots from a petiole stimulated by prior treatment with 0.1%naphthalene acetic acid may support an aphid colony for at least2 weeks. In this system, aphids were allowed to freely colonizeboth leaf sides, and the infected alates initiating new colonieswere considered the only source of primary infection.

Overall, 328 alates survived $≥$ 1 day after colonization, withinitiated colonies consisting of $≥$ 1 nymph, and allowed the processof mycosis transmission to be observed. All observations weregrouped based on survival days and are shown in Table 1. Observedtrends in the development of the grouped colonies and accompanyingmycosis transmission over days are plotted in Fig. 1. On average,the alates flew for 2.9 ± 1.7 (range, 1 to 5) h or 2.6± 2.2 (0.01 to 10.2) km after infection, survived for3.2 ± 0.9 days after colonization, and produced 5.3 ±3.2 nymphs during their survival periods. Of those, 98.5% weremycosed on days 2 to 5 and 50% were mycosed on day 3. In eachgroup, the rate of primary infection ranged from 0.083 to 0.5,varying with the numbers of nymphs left by the infected alateson their death days. Variations in flight time and distancedid not differ significantly among the groups (P > 0.05).Those surviving longer after colonization tended to producemore nymphs, but their fecundities prior to death were not alwayssignificantly different from one group to another (Table 1).

View this table:
[in this window]
[in a new window]
TABLE 1. Flight and postflight colonization capabilities of M. persicae alates infected with P. neoaphidis and subsequent mycosis transmission within progeny colonies, grouped by their survival days after flight

View larger version (28K):
[in this window]
[in a new window]
FIG. 1. Observed trends in the development of progeny colonies (A) and within-colony mycosis transmission (B) individually initiated by Pandora neoaphidis-infected Myzus persicae alates that were allowed to colonize cabbage leaves for 14 days after simulated flight. Symbols indicate survival periods (days) of the infected alates after colonization. The first observation, marked with an asterisk for each case in panel B, represents the primary infection rate (proportion of the alate cadaver in its progeny colony). Error bars indicate standard errors.

Contagious infection occurred in most colonies and developedover days after their mother alates were mycosed (Fig. 1B).Thirty-eight colonies were entirely mycosed on day 7, and 56were entirely mycosed on day 14. Overall, entirely or partiallymycosed colonies were 13.1% on day 5, 45.7% on day 7, 76.2%on day 10, and 80.2% on day 14. Mycosis transmission tendedto occur earlier, develop faster, and cause higher proportionsof mycosed individuals in the colonies when their mother alatesdied earlier or produced fewer nymphs prior to death.

Modeling of progeny colony development.

Top

Modeling of progeny colony...

When both progeny colony increase and within-colony mycosistransmission were assumed to be affected by variables associatedwith the flight and postflight colonization of the infectedalates and the primary infection rate at their death time, thenumber of live aphids per colony (N_i) over days (T_i; i = 1,2, ..., 14) fit to the complex logistic model N = K/{1 + exp[b₀+ (b₁D_f + b₂T_s + b₃P_m0 + b₄ln(1 + N_dd))T_i]} + b₅/[1 + exp(b₆+ b₇T_i)], where N = ln(N_i + 1), D_f = flight distance (km), T_s= survival time (days), P_m₀ = primary infection rate, and N_dd= number of nymphs produced by each infected mother alate onthe death day. The parameter K was interpreted as the maximalpotential for the increase of progeny colony initiated by theinfected alate, and b₀ was simply the intercept for the fittedcurve. The expression b₁D_f + b₂T_s + b₃P_m0 + b₄ln(1 + N_dd) wasinterpreted as the rate of increase of the colony over time,and the fitted curve was adjusted by b₅/[1 + exp(b₆ + b₇T_i)].Based on weighted modeling (weight = number of infected alatesin each group), the observations in Fig. 1A and Table 1 werewell fitted to the model (Fig. 2A) (r² = 0.97; F_{8, 89} = 354.5;P < 0.0001), yielding a colony increase rate of 0.05D_f +0.06T_s + 0.26P_m0 – 0.38ln(1 + N_dd) and a maximal colonypotential of 57 [= exp(4.06) – 1] aphids. The fitted expression0.74/[1 + exp(–9.31 + 2.03T_i)] reflected well distincttrends of reproduction by the infected alates within the firstweek and then by the apterous adults developed from their progenyduring the second week. Apparently, the rate of colony increasewas dependent upon the flight distance of the infected alates,their survival days after flight, the number of their nymphsprior to death, and the rate of primary infection at their deathtime. Since contagious infection occurred soon after their deaths,the number of progeny aphids decreased within the first week,varying with the death days of the alates (Fig. 1A and 2A).

View larger version (21K):
[in this window]
[in a new window]
FIG. 2. Fitted trends in the number of live aphids (N_i) in M. persicae colonies individually initiated by P. neoaphidis-infected alates and the proportion of mycosed aphids (P_mi) in the colonies. (A) N = 4.06/{1 + exp[1.92 + (0.05D_f + 0.06T_s + 0.26P_m₀ – 0.38ln(1 + N_dd))T_i]}+ 0.74/[1 + exp(–9.31 + 2.03T_i)], where N = ln(N_i + 1). (B) P_mi = exp{–8.19exp[–(0.04D_f – 0.02T_s + 1.34P_m₀ –1.12P_m0²N)T_i]}. Student t tests for all estimated parameters were significant (P $≤$ 0.0005).

Modeling of mycosis transmission.

Top

Modeling of mycosis...

Provided that the progress of P. neoaphidis mycosis caused bycontagious infection was biologically associated with availablehost individuals and the primary infection rate relied merelyupon the infected alates, the varying proportion of mycosedindividuals (P_mi) within the colonies fit to the Gompertz model,which has been widely used to describe epidemic features ofplant diseases with epidemiologically interpretable parameters,in a general form of y = K · exp[–a · exp(–rt)],where K is the maximal epidemic level to be achieved, a is anintercept for a curve to be fitted, and r is the epidemic rateof a plant disease over time t (1). Since mycosis transmissionin this study was likely to vary with the progeny colony sizeassociated with the flight and postflight colonizations of theinfected mother alates, the primary infection rate, and theirpossible interaction, the model was modified as P_mi = exp{–c₀exp[–(c₁D_f+ c₂T_s + c₃P_m0 + c₄P_m0²N)T_i]}, where N = ln(N_i + 1) and K =1 due to P_mi $≤$ 1. The expression c₁D_f + c₂T_s + c₃P_m0 + c4P_m0²Napparently represented the rate of mycosis transmission withinthe colonies. As a result, the fitted Gompertz model depictedwell the progress of mycosis transmission at the rate of 0.04D_f– 0.02T_s + 1.34P_m0 – 1.12P_m0²N over the colonizationdays (Fig. 2B) (r² = 0.97; F_{4, 93} = 679.8; P < 0.0001). Again,the variables D_f, T_s, and P_m₀ were determinants for the transmissionrate and thus the fluctuating P_mi. Interestingly, N was a prominentdeterminant for P_mi. This indicates that the mycosis transmissionvia contagious infection was density dependent.

Since both fitted models described well interactions betweenP. neoaphidis and M. persicae in the development of progenycolonies and mycosis transmission (Fig. 2), they may shed lightinto general mechanisms involved in the wide dispersal and localtransmission of aphid mycoses caused by Entomophthorales. Ourresults clarify that the P. neoaphidis-infected alates werecapable of flying for dispersing themselves, initiating progenycolonies on plants, and transmitting their primary infectionto progeny through contact. This supports our recent reportsfrom examination of air captures of several aphid species andsimulated flight studies with infected Sitobion avenae alates(3, 4, 11) and highlights the decisive role of alate-borne inoculain initiation of M. persicae mycosis.

The potential for dispersal of the Entomophthorales with theirmigratory flight is imaginable, because alates may passivelyfly with wind for over 1,000 km (23). Recently, 35 P. neoaphidisisolates derived from global aphid hosts have been found toshare a uniform 1,100-bp size of the amplified internal transcribedspacer region (25). This could be indirect, molecular evidencefor the source of the primary inocula of P. neoaphidis associatedwith migratory alates. Although active flight of the infectedalates in the simulated flight system could never be as faras their potential passive flight with wind in nature, the flightdistances of up to 10.2 km measured within 5 h in this studyguarantee efficient dissemination of primary inocula among aphidsacross areas or vegetations. Upon successful colonization byinfected alates on suitable plants, their primary infectioncan be contagiously transmitted within progeny colonies, asshown in this paper, and eventually across colonies if aphidpopulations develop in the field (14), because aphid coloniescontaining infected individuals hardly escape from the fateof an epizootic in suitable environments (8). Our findings wouldbe revelatory for exploring pathogen-aphid interactions andfor understanding mechanisms involved in natural control ofaphids by the obligate fungal pathogens.

ACKNOWLEDGMENTS

This study was jointly supported by the Natural Science Foundationof China (30430150), the National Frontier Research ProgramProject 973' (2003CB114203), the Special Fund for Graduate StudyPrograms in Chinese Universities (200203335041), and the CheungKong Scholars Programme, Ministry of Education, China.

FOOTNOTES

* Corresponding author. Mailing address: Institute of Microbiology, College of Life Sciences, Zhejiang University, Hangzhou 310029, People's Republic of China. Phone and fax: 86-571-86971129. E-mail: mgfeng{at}zju.edu.cn.

REFERENCES

Top