Broomcorn millet grain cultures of the entomophthoralean fungus Zoophthora radicans: sporulation capacity and infectivity to Plutella xylostella

Mycol. Res. 109 (3): 319–325 (March 2005). f The British Mycological Society

319

doi:10.1017/S095375620500239X Printed in the United Kingdom.

Broomcorn millet grain cultures of the entomophthoralean fungus Zoophthora radicans : sporulation capacity and infectivity to Plutella xylostella

Li HUA1 and Ming-Gung FENG1,2* 1

Institute of Microbiology, College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, 310029, Peoples’ Republic of China. 2 Institute of Applied Entomology, College of Agricultural Science and Biotechnology, Zhejiang University, Hangzhou, Zhejiang, 310029, Peoples’ Republic of China. E-mail : [email protected] Received 30 August 2004; accepted 16 December 2004.

The shelled grains of glutinous broomcorn millet, Panicum miliaceum, were used as solid substrate to prepare granular cultures of Zoophthora radicans, an entomophthoralean biocontrol agent against numerous insect pests. Steamed millet grains were inoculated by mixing 15 g millet grains (D.W.) with mashed pieces of half a 60-mm-dish colony in 3 ml modified Sabouraud dextrose broth and incubated at 15 xC and L :D 12 :12 for up to 24 d. 20 grains were sampled at 3 d intervals from day six onwards and individually assessed for their sporulation capacity using a self-designed device for spore collection. The millet cultures after o12 d incubation produced 12.0–14.9r104 spores grainx1 during a 7 d period. The maximal sporulation capacity associated with the 21 d-old culture was about half of that of Z. radicans-killed Plutella xylostella larvae (28.7r104 spores cadaverx1), which individually were at least three times larger than the millet grains. Based on the time–concentration–mortality responses of second-instar P. xylostella larvae to Z. radicans in three independent bioassays, the spores ejected from the cultured millet grains, from the mycelial mats from liquid culture, and from larval cadavers displayed insignificant variations in infectivity to the host species, and yielded similar LC50 and LT50 estimates. Conclusively, the millet-based technology for production of granular cultures of Z. radicans was easy, inexpensive and highly efficient, and it could be superior to previous methods used in mass production of myceliumbased preparations of Entomophthorales since this new approach requires no special additives, drying, freezing and milling. This technology may suit to mass production of culturable but nutritionally fastidious entomopathogens from the Entomophthorales.

INTRODUCTION Zoophthora radicans (Zygomycota : Entomophthorales) is an obligate entomopathogen with a wide range of insect hosts, particularly Homoptera and Lepidoptera (McGuire, Maddox & Armbrust 1987a, Humber 1989, Feng, Johnson & Kish 1990, Li 2000). This and other entomophthoralean biocontrol agents disperse widely in association with the migratory flights of aphids (Chen & Feng 2004, Feng, Chen & Chen 2004), infect insects with spores (primary conidia or ballistoconidia) forcibly ejected from host cadavers and suppress host populations as mycoses develop to epizootics (McGuire et al. 1987b, Wraight et al. 1990, Feng, Johnson & Halbert 1991, Galaini-Wraight et al. 1991, Feng et al. 1992, Feng & Li 2003). Intensive studies * Corresponding author.

have been directed toward the utilization of Z. radicans for biocontrol by means of releasing mycosis-killed cadavers or cultures in vitro into host populations (Milner, Soper & Lutton 1982, Wraight et al. 1986, McGuire, Maddox & Armbrust 1987c, Pell & Wilding 1994, Furlong et al. 1995). Despite certain successes in previous studies, in vivo propagation of entomophthoralean fungi such as Z. radicans is generally considered to be too laborious and expensive for practical purpose (Wilding & Latteur 1987). The challenge of finding a suitable technique for the mass production of Entomophthorales for application in biocontrol remains. As obligatory insect pathogens, some entomophthoralean fungi can be cultured in vitro, but grow slowly even on highly nutritive media such as Sabouraud dextrose agar plus egg yolk and milk (SEMA ; Glare, Milner & Chilvers 1986). A milestone

Millet grain cultures of Zoophthora radicans ‘marcescence process ’ method was developed to generate dry mats of Z. radicans mycelia grown in liquid culture (McCabe & Soper 1985). Ground mycelial mats produced as above, or with more or less modification, can be stored for some time at low temperatures and then rehydrated in a moist environment (Wraight et al. 1986, 2003, Li et al. 1993, Pell et al. 1998). Spores (ballistoconidia) discharged from the rehydrated mycelial mats may induce or augment epizootics in pest populations under field or greenhouse conditions (Wraight et al. 1986, Wraight & Roberts 1987, Pell & Wilding 1994). However, mycelial mats of Z. radicans and other entomophthoralean fungi produced by submerged fermentation and subsequent desiccation are difficult to formulate further because post-production processes including milling, freezing and storage are deleterious to mycelial viability (Li et al. 1993). Moreover, desiccated states of the fungus are not infectious until rehydration and sporulation occurs (Wraight et al. 2003), and thus must be properly formulated to facilitate field application. A novel method has been developed to immobilize Pandora neoaphidis (syn. Erynia neoaphidis) hyphae (cultured in liquid medium) in an alginate matrix for production of granules or beads (Shah, Aebi & Tuor 1998). Additives such as sucrose, starch and chitin incorporated into the alginate granules may enhance their sporulation capacity (Shah, Aebi & Tuor 1999). Despite this encouraging progress, technical development in mass production and formulation to date has not been sufficient to facilitate an effective strategy to incorporate Entomophthorales into pest management systems. The most important problem is related to the high cost of production and adaptability of a formulation to variable environmental conditions whether mycelial mats or alginate granules are used for the final formulated product. Recently, shelled grains of glutinous broomcorn millet, Panicum miliaceum, a drought-tolerant crop grown worldwide and common in northern or northwestern China, have been successfully used as solid substrate for direct production of granular cultures of Pandora delphacis and P. neoaphidis (Feng & Liang 2003, Hua & Feng 2003). Cultured small grains eject spores infectious to aphids just as do mycotized cadavers, but sporulate much more abundantly and longer than the cadavers. This new technology is very likely to be applied to other entomophthoralean fungi being used as biocontrol agents. We sought to standardize procedures for the millet grain cultures of a Z. radicans isolate highly infectious to the larvae of diamondback moth Plutella xylostella (Lepidoptera : Yponomeutidae) (Liu, Xu & Feng 2003), to quantify sporulation capacity of the cultured grains vs Z. radicans-killed P. xylostella larvae, and to compare variation in the fungal infectivity to P. xylostella larvae among the spores ejected from the mycelial mats grown on millet grains and in liquid medium, and from the larval cadavers of the pest.

320 MATERIALS AND METHODS Fungal isolate and culture Zoophthora radicans (ARSEF 1100; USDA-ARS Collection of Entomopathogenic Fungal Cultures, Ithaca, NY) was isolated from a Plutella xylostella cadaver in Malaysia in April 1981. This isolate was maintained after arrival to our laboratory on slants of Sabouraud dextrose agar supplemented with 0.5 % (v/v) sesame oil and 0.1 % (w/v) sucrose fatty acid esters (Sanliu Food Chemical Industry, Sanliu, Guangxi Province) at 3 xC in dark and recovered twice a year (Feng & Xu 2001). A slant culture was grown for ca 20 d on SEMA medium (Glare et al. 1986) in 60 mm diam Petri dishes at 15 xC and L :D 12 :12, yielding fungal colonies for the following use. Preparation of granular cultures Shelled broomcorn millet grains (purchased from a local grocery store ; cultivar unknown) were used as solid substrate to prepare granular cultures of Zoophthora radicans following a protocol described elsewhere (Feng & Liang 2003, Hua & Feng 2003). Briefly, millet grains (15 g per 100 ml flask) were soaked for y30 min in water at y80 x. After rinsing to remove dust, the grains were autoclaved for 15 min at 121 x and cooled to ambient temperature. Each flask of the autoclaved grains was inoculated with half a plate colony homogenized in 3 ml Sabouraud dextrose broth supplemented with 0.5 % (v/v) sesame oil and 0.1 % (w/v) sucrose fatty acid esters. After plugging with vent stoppers, all flasks were incubated for up to 24 d at 15 x and L :D 12 :12. No agitation measures for aeration were taken during the incubation period. Measurement of sporulation capacity During incubation, 20 mycotized grains were arbitrarily taken from a flask at 3 d intervals from the sixth day onwards and individually measured for sporulation capacity using spore collection chambers (Hua & Feng 2003). Each chamber was composed of upper plate and lower concave bottom containing 200 ml of 0.5 % (w/v) dodecyl sodium sulfate solution. Each cultured millet grain was attached to the centre of the upper plate (13 mm diam) containing 2 % (w/v) agar only. The upper plate was then inverted onto the concave bottom (13 mm diam, 10 mm tall). The solution of 0.5 % dodecyl sodium sulfate that functions as a surfactant may inactivate fungal spores but has no effect on their morphology (Ying & Feng 2001). Thus, the spores ejected from each grain fell freely down to the surfactant solution in which they were well suspended and inactivated without germination or morphological change. All sampled millet grains in the spore collection chambers were then incubated at 15 x and light/dark 12/12 h. The lower concave bottoms were replaced daily for spore counts until no further spores appeared

L. Hua and M.-G. Feng in the solution. From each of the concave bottoms with collected spores, 15 1 ml suspension samples were pipetted into counting chambers of standard hemocytometers for spore counts under microscope. To assure uniform samples, the spore suspension was shaken before being pipetted. A sum of daily spore counts was an estimate of overall sporulation capacity for each grain. For comparison, 20 cadavers of Plutella xylostella larvae killed by the same fungal isolate (inoculated at the second instar stage) were evaluated for sporulation capacity with the same protocol. Bioassays Millet grains cultured with Zoopthora radicans for 21 d (assay 1) were uniformly distributed on 2 % agar in a 90 mm diam Petri dish and incubated at 15 x and light/ dark 12/12 h for 24–48 h for abundant sporulation. For inoculation, second instar Plutella xylostella larvae on detached cabbage leaves in Petri dishes were exposed for different periods to spore showers from the sporulating grains, generating spore concentrations of 15.1¡2.2, 39.2¡3.1, 53.8¡3.7 and 81.1¡5.1 spores mmx2, which were determined with a coverslip method (Feng & Johnson 1991, Feng et al. 1998, Xu & Feng 2000). For comparison, P. xylostella larvae were also exposed to the showers of spores ejected from sporulating P. xylostella cadavers (assay 2 : 15.6¡2.5, 34.3¡1.4, and 87.8¡15.9 spores mmx2) and mycelial mats produced in Sabouraud dextrose broth (assay 3: 4.2¡0.2, 11.7¡5.6, 38.7¡4.4, and 91.0¡0.4 spores mmx2). Each spore concentration included 3–5 replicates (30–53 larvae per replicate). Larvae in three Pertri dishes unexposed to spore showers were included as blank controls for each of the three bioassays. After exposure to the spore showers, all larvae on detached leaves in Petri dishes were maintained for 9 d at 20 x and L :D 12 :12 and examined daily for mortality. Larval cadavers were individually examined microscopically to verify Z. radicans infection immediately or after overnight maintenance in moist Petri dishes if Z. radicans was not present. Data analysis Variances among the daily and cumulative counts of the spores ejected from the millet grains at different culture ages were analyzed using the one-way ANOVA procedure. Data from each bioassay were fitted to a time–concentration–mortality model (Nowierski et al. 1996, Feng et al. 1998, Feng & Poprawski 1999, Xu & Feng 2000). Parameters for the effects of spore concentration and time and associated variances and covariances were estimated from the modeling and were then used to compute values for LC50 (lethal concentration to cause 50% mortality at a given time after exposure) and LT50 (time to cause 50% mortality at a given spore concentration). DPS software (Tang & Feng 2002) was used in all analyses.

321 RESULTS Granular millet cultures The ovate, yellowish grains of the broomcorn millet were 1.4–2.0r2.0–2.6r2.5–3.2 mm in size (Figs 1–2) and had an average weight of 5.14¡0.02 mg. When soaked and steamed in a regular autoclave procedure, the grains became swollen and had a rubber-like firmness with a water content of 36.5 % (measured after drying at 80 x for 48 h). When incubated at 15 x and light/dark 12/12 h, the inoculated millet grains gradually turned whitish within the first two or three days due to mycelial growth of Zoophthora radicans on the grains. Mycelial growth became dense and thick, and ultimately covered the grains after one week or more (Fig. 3). Mature grain cultures had a water content of 47.1 % and sporulated very well on non-nutritional substrate (2 % water agar) under moist conditions (Figs 4–5). The spores ejected from the millet culture formed a dust-like ‘spore halo ’ identical to that around a larval cadaver killed by Z. radicans (Fig. 6). Each cultured millet grain biologically was equivalent to an entomophthorosis-killed cadaver. Sporulation capacity and time pattern Spore production of the sampled millet grains cultured with Zoophthora radicans was monitored for 7 d (Table 1). The sums of daily spore counts differed significantly according to the age of the cultures (F6, 114=15.97, P<0.001), but insignificantly among the 20 grains sampled at a given culture age (F19, 114=1.50, P=0.099). The 21 d-old millet culture sporulated best and discharged up to 14.9 (¡0.9)r104 spores grainx1. The cultures at the ages of 12–18 d yielded ca 12r104 spores grainx1 but did not differ significantly from the 21 d-old culture. The cultures younger than 12 d displayed significantly less sporulation capacity but produced apparently more spores with the increasing culture age. In comparison, cadavers of Plutella xylostella larvae killed by Z. radicans infection (inoculated at the second instar stage) produced an average of 28.7 (¡8.9)r104 spores (n=20). Thus, the sporulation capacity of each cadaver roughly doubled that of each cultured millet grain. This is perhaps largely due to the much larger size of each larval cadaver than each cultured grain (Figs 4, 6). Based on the individual size of P. xylostella cadavers (y10 mm in length and y2 mm in breadth), the cultured millet grains produced at least as many spores as the larval cadavers on the basis of the volume of mycotized substrate. The timing patterns of sporulation from the millet grains with different culture ages were similar on nonnutritive agar (Table 1) ; each cultured grain sporulated for up to 7 d. Generally, 66–88 % of spores produced were discharged in the first 3 d with 30–40 % of total sporulation on the second day. Younger cultures tended to sporulate most heavily during the first 3 d, whereas the older cultures tended to discharge more

Millet grain cultures of Zoophthora radicans

322

Table 1. Comparison of sporulation capacity and timing pattern between Zoophthora radicans cultures grown on broomcorn millet grains and cadavers of Plutella xylostella larvae killed by the same fungal species. Sporulation capacity of millet cultures, no. spores grainx1 (mean¡S.E.r104)a Sporulation day

6d

9d

12 d

15 d

18 d

21 d

24 d

No. spores cadaverx1 (¡S.E.r104)

1 2 3 4 5 6 7 Total

1.70¡0.20a 1.93¡0.25a 0.50¡0.06c 0.26¡0.03c 0.14¡0.02d 0.13¡0.02b 0.11¡0.01b 4.77¡0.42b

2.78¡0.30a 3.12¡0.39ab 0.77¡0.07c 0.26¡0.05c 0.22¡0.05cd 0.19¡0.06b 0.19¡0.04b 7.54¡0.73b

2.27¡0.36a 4.30¡0.45a 2.61¡0.26ab 1.42¡0.15ab 0.68¡0.07bc 0.43¡0.04b 0.32¡0.02b 12.03¡0.96a

2.06¡0.29a 3.61¡0.35a 2.26¡0.27ab 1.37¡0.18b 1.18¡0.22ab 0.91¡0.12a 0.71¡0.10a 12.09¡0.99a

1.60¡0.29a 3.50¡0.45a 2.87¡0.27a 1.57¡0.13ab 1.04¡0.11ab 0.84¡0.07a 0.60¡0.05a 12.02¡0.89a

2.75¡0.30a 4.62¡0.38a 2.88¡0.31a 2.04¡0.23a 1.24¡0.11a 0.82¡0.10a 0.58¡0.04a 14.92¡0.88a

1.73¡0.28a 3.26¡0.33ab 1.90¡0.23b 1.57¡0.22ab 1.23¡0.17a 0.90¡0.11a 0.80¡0.11a 11.39¡1.22a

12.09¡2.27 10.57¡3.98 5.61¡4.57 0.38¡0.24 0.03¡0.02 0.01¡0.00 0.00¡0.00 28.69¡8.88

a

Means with different lowercase letters in each line differed significantly (Tukey’s HSD, P<0.05).

Figs 1–6. Zoophthora radicans cultures grown on broomcorn millet grains steamed in a regular autoclaving procedure. Fig. 1. Uncooked, shelled broomcorn millet grains. Fig. 2. The millet grains after being soaked in hot water and autoclaved. Fig. 3. The millet grains heavily covered by a mature mycelial layer. Fig. 4. A dust-like ‘ halo’ formed around a cultured millet grain placed on a glass slide in humid Petri dish. Fig. 5. Discharged primary conidia typical for Z. radicans from the dust-like halo surrounding the cultured millet grain. Fig. 6. The halo of spores discharged from a Z. radicans-killed cadaver of P. xylostella on a glass slide in humid Petri dish. Bars: 1 mm (Figs 1–4, and 6), and 20 mm (Fig. 5).

spores during the last 4 d. For instance, the 15 d-old millet culture released its sporulation capacity by 17, 29.8, 18.7, 11.3, 9.7, 7.5, and 5.9 % on days 1–7, respectively. Up to several thousand spores were discharged from each cultured grain, even on the last sporulation day, despite the small proportion (5.9 %) in total. In contrast, cadavers of P. xylostella larvae killed by Z. radicans released 98.5 % of their sporulation potential within the first 3 d (i.e. 42.1, 36.8 and 19.6 % on days 1–3, respectively) and had another 1.3 % released on day 4. Obviously, the sporulation of Z. radicans millet cultures lasted much longer than that of the larval cadavers.

Infectivity The time–concentration–mortality trends observed in assays 1–3 are displayed in Fig. 7. Deaths of Plutella xylostella larvae attributed to Zoophthora radicans mycosis did not appear until day 5 after exposure to spore showers but mostly occurred on days 6–8. Larval mortalities increased with the spore concentrations. All cadavers displayed the typical course of development for Z. radicans mycoses. At 9 d after exposure, larval mortalities were 36.9–91 % at 15.1–81.1 spores mmx2 in assay 1, 33.6–91.5 % at 15.6–87.8 spores mmx2 in assay 2, and 11.3–93.6 % at 4.2–91.0 spores mmx2 in

L. Hua and M.-G. Feng 1.0

323 2.5

CK 15.1

0.8

39.2 LC50 (log10)

53.8

0.6

81.1 0.4

2.0

0.2 1.5 0.0 4 1.0

5

6

7

8

9

5

6

7 Days after exposure

8

9

CK 9

15.6 34.3 87.8

0.6

8 LT50 (days)

Cumulative mortality

0.8

0.4

7

0.2 6

0.0 4 1.0

5

6

7

8

9

60

70

80

90

100

Fig. 8. Trends in the estimates of logarithm-scaled LC50 (no. spores mmx2 ; upper) and LT50 (lower) based on the modeling of the time–concentration–mortality data from the bioassays of Zoophthora radicans on Plutella xylostella larvae exposed to spore showers from sporulating cultures grown on broomcorn millet grains (dotted line) or in liquid medium (dashed line) and from mycosis-killed P. xylostella cadavers (solid line) . Error bars : S.E.

11.7 38.7 91.0

0.4 0.2 0.0 4

50

No. spores mm

4.2

0.6

40

–2

CK

0.8

30

5

6 7 Days after exposure

8

9

Fig. 7. Trends in the cumulative mortalities of Plutella xylostella larvae after exposure to different concentrations of spore showers (no. spores mmx2 ; CK : unexposed blank control) from Zoophthora radicans cultures grown on broomcorn millet grains (upper) or in liquid medium (middle) and from cadavers of P. xylostella larvae killed by Z. radicans (lower). Error bars : S.E.

assay 3, respectively. The background mortality in the unexposed blank controls ranged from 2.2–3.4% in assays 1–3, but no fungal infection was associated with the cadavers. The data in Fig. 7 fits well to the time– concentration–mortality model (Nowierski et al. 1996, Feng et al. 1998, Feng & Poprawski 1999). The concentration-effect parameter (b¡S.E.) was estimated to be 2.15¡0.03 for assay 1, 2.76¡0.05 for assay 2, and

2.30¡0.03 for assay 3. The conditional time-effect parameters for days 5–9 after exposure (c5– 9¡S.E.) were estimated as x5.72¡0.09, x4.52¡0.08, x3.95¡0.08, x3.67¡0.08 and x3.49¡0.08 for assay 1 ; x7.04¡ 0.16, x5.78¡0.15, x5.05¡0.14, x4.65¡0.13 and x4.45¡0.13 for assay 2 ; and x6.04¡0.11, x4.80¡ 0.09, x4.15¡0.08, x3.78¡0.08 and x3.68¡0.08 for assay 3, respectively. All Student’s t tests for the estimates of the parameters were significant (P<0.01). The Hosmer–Lemeshow test for heterogeneity of the goodness of fit was insignificant for the modeling of each data set (assay 1: df=8, C=4.29, P=0.83 ; assay 2 : df=7, C=7.96, P=0.34 ; assay 3 : df=9, C=9.72, P=0.37). Thus, each set of the estimated parameters determined a time–concentration–mortality relationship with an accepted homogeneity. Based on the established time–concentration– mortality relationship for each assay, LC50 and LT50 were computed as functions of post-exposure time and spore concentration, respectively (Fig. 8). The LC50 with 95% confidence intervals decreased from 310.3 (223.1–431.6) spores mmx2 on day 5 to 28.2 (25.3–31.5)

Millet grain cultures of Zoophthora radicans spores mmx2 on day 9 after exposure in assay 1, from 264.0 (188.9–368.8) to 30.4 (27.1–34.2) spores mmx2 in assay 2, and from 293.4 (204.8–420.3) to 27.5 (23.6–32.1) spores mmx2 in assay 3, respectively. Apparently, differences in LC50 were insignificant among the three bioassays during the period of mycosis development. On the other hand, a maximal LT50 (9 d) corresponded to the spore concentrations of 28.3, 30.5 and 27.6 spores mmx2 in assays 1–3, respectively. As the concentrations increased, the LT50 estimates for the three bioassays tended to approach to each other, for example, 6.9, 7, and 6.8 d at 50 spores mmx2, and 5.8, 5.9, and 5.8 d at 100 spores mmx2. Thus, Z. radicans spores discharged from the cultured millet grains, larval cadavers, and liquid cultured mycelium, had a similar infectivity to P. xylostella larvae.

DISCUSSION The broomcorn millet grains proved an excellent solid substrate to make granules of Zoophthora radicans cultures by simply mixing a small amount of inoculum with properly steamed grains and then incubating them under optimal conditions for the fungal isolate. The resultant grain cultures, with a water content of less than 50 %, generate spores as infective to Plutella xylostella larvae as those from mycosis-killed cadavers or mycelial mats (Figs 2–3). Thus, the growth, sporulation and virulence of Z. radicans grown in this manner are fully equivalent to the millet cultures of Pandora spp. (Feng & Liang 2003, Hua & Feng 2003). The millet cultures of Z. radicans incubated at 15 x for 12–24 d had overall sporulation capacities of 12–15r104 spores grainx1, with the maximal production in 21 d-old cultures (Table 1). With the weight of 5.14 mg grainx1 taken into consideration, the sporulation capacities of the millet cultures are equivalent to >2.3r107 spores gx1 if the incubation is prolonged beyond 12 d. In other studies, the millet cultures of P. delphacis and P. neoaphidis incubated at 25 x for 5–11 d, and 20 x for 6–15 d, were able to produce 13.0–17.1r104 and 16.8–23.4r104 spores gx1, respectively (Feng & Liang 2003, Hua & Feng 2003). Both maximal estimates resulting from the 5 d and 15 d-old millet cultures of the two Pandora species are equal to 3.3r107 and 4.6r107 spores gx1. These millet cultures of Zoophthora and Pandora produce as many spores as maltose-treated, unmilled, pure mycelial mats of Z. radicans (4.0r107 spores gx1), but far more than the milled mycelial mats (<0.1r107 spores gx1) (Li et al. 1993). However, the sporulation capacities of the millet cultures in our studies cannot be compared with those of P. neoaphidis alginate granules of differing methods for estimating overall sporulation levels (Shah et al. 1998, 1999). In contrast, P. neoaphidis mycelial pellets produced from shake-flasks possess lower sporulation potential than aphid cadavers (Bonner, Pell & Gray 2003).

324 We conclude that the use of broomcorn millet grains for the production of granular cultures of Zoophthora and Pandora is superior to previous methods for the mass production of mycelium-based preparations of entomophthoralean fungi reported. This technology is preferable to the alternatives due to its simplicity, low cost, and the lack of any special needs for additives, drying, freezing, and milling prior to storage and use. The millet grains are readily and inexpensively available (øUS $ 0.72 kgx1). All that is needed to prepare the granular cultures is to steam the millet grains, inoculate, and incubate them for certain periods, which largely depend on the optimal growth temperature for a given fungal species or isolate. Since the genera Zoophthora and Pandora include numerous obligate insect pathogens with fastidious requirements for nutrients, the millet-based technology developed in our laboratory may be suitable for in vitro cultures of many of the most readily cultured Entomophthorales. If this proves to be true, basic and applied studies on this large group of fungal biocontrol agents for utilization in pest control will be greatly facilitated. The potential to store and to use mycotized millets to initiate epizootics under controlled or field conditions is being actively investigated.

ACKNOWLEDGEMENTS This study was jointly supported by the grants from the Natural Science Foundation of China (30430150), the National Frontier Research Program ‘Project 973 ’ (2003CB114203), the Special Fund for Graduate Study Programs in Chinese Universities (200203335041) and the ‘Cheung Kong Scholars Programme’, Ministry of Education, Peoples’ Republic of China.

REFERENCES Bonner, T. J., Pell, J. K. & Gray, S. N. (2003) A novel computerized image analysis method for the measurement of production of conidia from the aphid pathogenic fungus Erynia neoaphidis. FEMS Microbiology Letters 220 : 75–80. Chen, C. & Feng, M. G. (2004) Sitiobion avenae alatae infected by Pandora neoaphidis: their flight ability, post-flight colonization and mycosis transmission to progeny colonies. Journal of Invertebrate Pathology 86: 117–123. Feng, M. G., Chen, C. & Chen, B. (2004) Wide dispersal of aphidpathogenic Entomophthorales among aphids relies upon migratory alates. Environmental Microbiology 6: 510–516. Feng, M. G. & Johnson, J. B. (1991) Bioassay of four entomophthoralean fungi (Entomophthorales) against Diuraphis noxia and Metopolophium dirhodum (Homoptera: Aphididae). Environmental Entomology 20: 338–345. Feng, M. G., Johnson, J. B. & Halbert, S. E. (1991) Natural control of cereal aphids (Homoptera: Aphididae) by entomopathogenic fungi (Zygomycetes: Entomophthorales) and parasitoids (Hymenoptera : Braconidae and Encyrtidae) on irrigated spring wheat in southwestern Idaho. Environmental Entomology 20: 1699–1710. Feng, M. G., Johnson, J. B. & Kish, L. P. (1990) Survey of entomopathogenic fungi naturally infecting cereal aphids in irrigated cereal crops in southwestern Idaho. Environmental Entomology 19: 1534–1542.

L. Hua and M.-G. Feng Feng, M. G. & Li, H. P. (2003) Experimental epizootiology of Zoophthora anhuiensis (Entomophthorales) against Myzus persicae (Homoptera: Aphididae) with a description of a modified Gompertz model for aphid epizootics. Environmental Microbiology 5: 1203–1211. Feng, M. G. & Liang, Y. (2003) Biological aspects on the cultures of the entomophthoralean fungus Pandora delphacis grown on broomcorn millets. Chinese Science Bulletin 48: 1816–1821. Feng, M. G., Liu, C. L., Xu, J. H. & Xu, Q. (1998) Modeling and biological implication of time-dose-mortality data for the Entomophthoralean fungus, Zoophthora anhuiensis, on the green peach aphid Myzus persicae. Journal of Invertebrate Pathology 72: 246–251. Feng, M. G., Nowierski, R. M., Johnson, J. B. & Poprawski, T. J. (1992) Epizootics caused by entomophthoralean fungi (Zygomycetes: Entomophthorales) in populations of cereal aphids (Homoptera: Aphididae) in irrigated small grains of southwestern Idaho, USA. Journal of Applied Entomology 113: 376–390. Feng, M. G. & Poprawski, T. J. (1999) Robustness for the processing of bioassay data of microbial control agents based on time-dosemortality model. Subtropical Plant Science 51: 36–38. Feng, M. G. & Xu, Q. (2001) A simple method for routine maintenance and preservation of entomophthoraceous cultures. Journal of Invertebrate Pathology 77: 141–143. Furlong, M. J., Pell, J. K., One, P. C. & Syed, A. R. (1995) Field and laboratory evaluation of a sex pheromone trap for the autodissemination of the fungal entomopathogen Zoophthora radicans (Entomophthorales) by the diamond-back moth, Plutella xylostella (Lepidoptera: Yponomeutidae). Bulletin of Entomological Research 85: 331–337. Galaini-Wraight, S., Wraight, S. P., Carruthers, R. I., Magalhaes, B. P. & Robert, D. W. (1991) Description of a Zoophthora radicans (Zygomycetes: Entomophthorales) epizootic in a population of Empoasca kraemeri (Homopera: Cicadellidae) on beans in central Brazil. Journal of Invertebrate Pathology 58: 311–326. Glare, T. R., Milner, R. J. & Chilvers, G. A. (1986) The effect of environmental factors on the production, discharge, and germination of primary conidia of Zoophthora phalloides Batko. Journal of Invertebrate Pathology 48: 275–283. Hua, L. & Feng, M. G. (2003) New use of broomcorn millets for production of granular cultures of aphid-pathogenic fungus Pandora neoaphidis for high sporulation potential and infectivity to Myzus persicae. FEMS Microbiology Letters 227: 311–317. Humber, R. A. (1989) Synopsis of a revised classification for the Entomophthorales (Zygomycotina). Mycotaxon 34: 441–460. Li, Z. Z. (2000) Flora Fungorum Sinicorum. Vol. 13. Entomophthorales. Science Press, Beijing. Li, Z., Butt, T. M., Beckett, A. & Wilding, N. (1993) The structure of dry mycelia of the entomophthoralean fungi Zoophthora radicans and Erynia neoaphidis following different preparatory treatments. Mycological Research 97: 1315–1323. Liu, Q. E., Xu, J. H. & Feng, M. G. (2003) Comparative virulence of Zoophthora radicans isolates from different hosts against Plutella xylostella larvae. Acta Entomologica Sinica 46: 447–453. McCabe, D. & Soper, R. S. (1985) Preparation of an entomopathogenic fungal insect control agent. U. S. Patent 4530834. McGuire, M. R., Maddox, J. V. & Armbrust, E. J. (1987a) Host range studies of an Erynia radicans strain (Zygomycetes: Entomophthoraceae) isolated from Empoasca fabae (Homoptera: Cicadellidae). Journal of Invertebrate Pathology 50: 75–77. McGuire, M. R., Morris, M. J., Armbrust, E. J. & Maddox, J. V. (1987b) An epizootic caused by Erynia radicans (Zygomycetes: Entomophthoraceae) isolated from Empoasca fabae (Homoptera: Cicadellidae). Journal of Invertebrate Pathology 50: 78–80. McGuire, M. R., Maddox, J. V. & Armbrust, E. J. (1987c) Effect of temperature on distribution and success of introduction of an Empoasca fabae (Homoptera: Cicadellidae) isolate of Erynia

325 radicans (Zygomycetes: Entomophthoraceae). Journal of Invertebrate Pathology 50: 291–301. Milner, R. J., Soper, R. S. & Lutton, G. G. (1982) Field release of an Israeli strain of the fungus Zoophthora radicans (Brefeld) Batko for biological control of Therioaphis trifolii (Monell) f. maculata. Journal of Australian Entomological Society 21: 113–118. Nowierski, R. M., Zeng, Z., Jaronski, S., Delgado, F. & Swearingen, W. (1996) Analysis and modeling of time-dose-mortality of Melanoplus sanguinipes, Locusta migratoria mgratorioides, and Schistocerca gregaria (Othoptera: Acrididae) from Beauveria, Metarhizium, and Paecilomyces isolates from Madagascar. Journal of Invertebrate Pathology 67: 236–252. Pell, J. K., Barker, A. D. P., Clark, S. J., Wilding, N. & Alderson, P. G. (1998) Use of a novel sporulation monitor to quantify the effects of formulation and storage on conidiation by dried mycelia of the entomopathogenic fungus Zoophthora radicans. Biocontrol Science and Technology 8: 13–21. Pell, J. K. & Wilding, N. (1994) Preliminary caged field trial using the fungal pathogen, Zoophthora radicans Brefeld (Zygomycetes: Entomophthorales) against the diamondback moth, Plutella xylostella L. (Lepidoptera : Yponomeutidae) in the UK. Biocontrol Science and Technology 4: 71–75. Shah, P. A., Aebi, M. & Tuor, U. (1998) Methods to immobilize the aphid-pathogenic fungus Erynia neoaphidis in an alginate matrix for biocontrol. Applied and Environmental Microbiology 64: 4260–4263. Shah, P. A., Aebi, M. & Tuor, U. (1999) Production factors involved in the formulation of Erynia neoaphidis as alginate granules. Biocontrol Science and Technology 9: 19–28. Tang, Q. Y. & Feng, M. G. (2002) DPS Data Processing System for Practical Statistics. Science Press, Beijing. Wilding, N. & Latteur, G. (1987) The Entomophthorales-problems relative to their mass production and their utilization. Mededelingen van de Faculteit Landbouwwetenschap- pen Rijksuniversiteit Gent 52: 159–164. Wraight, S. P., Butt, T. M., Galaini-Wraight, S., Allee, L. L., Soper, R. S. & Roberts, D. W. (1990) Germination and infection processes of the entomophthoralean fungus Erynia radicans on the potato leafhopper, Empoasca fabae. Journal of Invertebrate Pathology 56: 157–174. Wraight, S. P., Galaini-Wraight, S., Carruthers, R. I. & Roberts, D. W. (2003) Zoophthora radicans (Zygomycetes : Entomophthorales) conidia production from naturally infected Empoasca kraemeri and dry-formulated mycelium under laboratory and field conditions. Biological Control 28 : 60–77. Wraight, S. P., Galaini-Wraight, S., Carruthers, R. I. & Roberts, D. W. (1986) Field transmission of Erynia radicans to Empoasca leafhoppers in alfalfa following application of a dry mycelial preparation. In Fundamental and Applied Aspects of Invertebrate Pathology (R. A. Samson, J. M. Vlak & D. Peters, eds): 233. Foundation for the Fourth International Colloquium of Invertebrate Pathology, Wageningen. Wraight, S. P. & Roberts, D. W. (1987) Insect control efforts with fungi. Developments in Industrial Microbiology 28: 77–87. Xu, J. H. & Feng, M. G. (2000) The time-dose-mortality modeling and virulence indices for two entomophthoralean species, Pandora delphacis and P. neoaphidis, against the green peach aphid, Myzus persicae. Biological Control 17: 29–34. Ying, S. H. & Feng, M. G. (2001) Optional components for emulsifiable suspensions of Beauveria bassiana conidia. Acta Phytophylacica Sinica 28: 345–351.

Corresponding Editor: R. A. Humber