Inoculated Host Range and Effect of Host on Morphology and Size of Macroconidia Produced by Claviceps africana and Claviceps sorghi

Twenty graminaceous plant species were evaluated for their susceptibility to the two sorghum ergot pathogens Claviceps sorghi and Claviceps africana. Five species viz., Sorghum arundinaceum, Sorghum halepense, Sorghum versicolor, Sorghum virgatum and Pennisetum glaucum were found to become infected by both pathogens via inoculation with 10 conidia/ml. Species which did not become infected under these conditions included Pennisetum pedicellatum, Zea mays, and species of Panicum, Brachiaria, Cenchrus, Andropogon, Dichanthium, Chrysopogon, Iseilema, Bothriochloa and Chloris. Honeydew secretions were observed from infected flowers of susceptible plant species. There was marked variation in size of macroconidia of both C. sorghi and C. africana on different hosts on which the pathogens were able to establish symptoms. Dimorphism was observed for macroconidia produced on P. glaucum, as elliptical and spindle shaped macroconidia were observed. Based on inoculation under greenhouse conditions, we conclude that C. sorghi and C. africana may have similar host ranges. Introduction Ergot is a serious constraint in sorghum [Sorghum bicolor (L.) Moench] hybrid seed production as the disease curtails seed set, which may render seed production uneconomical. The most obvious sign of the disease is the appearance of a spore-laden sticky fluid called honeydew, which exudes from sphacelia that replace infected ovaries. In India, ergot is caused by two different pathogens, Claviceps sorghi Kulkarni, Seshadri and Hegde, the native pathogen of India and Claviceps africana Frederickson, Mantle and de Milliano, an exotic pathogen first reported from Africa (Bandyopadhyay et al., 2002). Splash-dispersed macroconidia and airborne secondary conidia play a vital role in pathogen dissemination (Bandyopadhyay et al., 1998). Collateral hosts (Futrell and Webster, 1966; Chinnadurai and Govindaswamy, 1971; Bandyopadhyay et al., 1991), sclerotia (Mantle, 1968) and ergotinfected seed lots (Bandyopadhyay et al., 1998) may act as sources of inoculum for new growing seasons. Sclerotial germination is low, therefore collateral hosts may serve as significant inoculum reservoirs for initiation and perpetuation of disease (Bandyopadhyay et al., 1998). The host range of C. sorghi (Reddy et al., 1968; Chinnadurai and Govindaswamy, 1971; Sundaram, 1974; Sangitrao and Moghe, 1995) is variable and contrasting (Bandyopadhyay et al., 1998). No research has been conducted either on the host range of C. africana in India (Bandyopadhyay et al., 2002) or on host range comparisons between the two pathogens. Our goal was to identify and evaluate potential collateral hosts of C. sorghi and C. africana by performing greenhouse inoculations under controlled environmental conditions. Materials and Methods These studies were performed at the International Crops Research Institute for Semi-Arid Tropics (ICRISAT), Patancheru, India. Eight representative isolates of C. africana and two isolates of C. sorghi were selected from a group of 89 isolates collected during disease surveys in India (Bandyopadhyay et al., 2002). Selected isolates of C. africana (and the names of geopolitical states of origin in parenthesis) included NI2 (Uttar Pradesh), NI5 (Uttaranchal), NI12 (Rajasthan), Guj6 (Gujarat),MH71 (Maharashtra), SK-20-24 (Karnataka), AP17 (Andhra Pradesh) and TN13 (Tamil Nadu), while those of C. sorghi were NAP7 (Andhra Pradesh) and MH74 (Maharashtra). To establish initial inoculum of each isolate, sclerotia and sclerotia–sphacelia were removed from infected U. S. Copyright Clearance Centre Code Statement: 0931–1785/2005/1531–0001 $ 15.00/0 www.blackwell-synergy.com J. Phytopathology 153, 1–4 (2005) 2005 Blackwell Verlag, Berlin ISSN 0931-1785 panicles collected during field surveys, macerated in a mortar and pestle, and suspended in water to release macroconidia produced in the sphacelial cavities. The resultant inoculum suspension was sprayed onto panicles of male-sterile sorghum genotype 296A at the 50% flowering stage that had been bagged after emergence from boot leaves to avoid contamination with external sources of inoculum. The bags were briefly removed prior to inoculation and replaced immediately after inoculation, and the plants were incubated in a greenhouse at 25 C. Bags were finally removed from the panicles when the first sign of honeydew exudation was noticed. Removal of bags and exposure of panicles to less than 85% relative humidity suppressed secondary conidiation (Bandyopadhyay et al., 1990) that could otherwise contaminate plants. Plants infected with each isolate were maintained in separate greenhouses to further avoid cross-contamination. All inoculation studies for host range and cross-inoculation studies were performed using the bagging and incubation conditions. Infected rachis containing honeydew were immersed in sterile water for approximately 1 min to allow dispersal of conidia. The suspension was filtered through two layers of cheesecloth and diluted to produce a suspension of 10 macroconidia/ml (Puranik and Mathre, 1971; Frederickson et al., 1989). When stigmas of the top 50% spikelets had emerged, plants of sorghum male-sterile line 296A were sprayed to runoff with the conidial suspension (Puranik and Mathre, 1971; Frederickson et al., 1989). Inoculated panicles were covered with paper bags to maintain high relative humidity and avoid external contamination. Inoculated plants were placed in dew chambers at 25 C and 100% relative humidity for 24 h to allow infection. Following dew chamber incubation, the plants were transferred to a greenhouse at 25 C and greater than 80% relative humidity, with paper bags enclosing the panicles. Observation of honeydew formation was made on the eighth day after inoculation by removing the paper bags. Upon commencement of honeydew secretion the infected panicles were kept uncovered to avoid secondary conidiation, which may contaminate the purity of the isolates within the greenhouse. The infected plants were used for further studies pertaining to inoculated host range. Honeydew less than 5-day-old was used as the inoculum source for host range studies. Twenty graminaceous plant species, including four wild sorghum species were evaluated as potential hosts of C. africana and C. sorghi. The plant species included: Sorghum arundinaceum (Desv.) Stapf, Sorghum halepense (L.) Pers., Sorghum versicolor (Steud.) Stapf, Sorghum virgatum (Hack.) Stapf, Pennisetum glaucum (L.) R. Br., Pennisetum pedicellatum Trin., Zea mays L., Panicum maximum Jacq., (cv. Guinea grass), Panicum maximum Jacq. (cv. Giant guinea), Panicum antidotale Retz., Brachiaria mutica (Forsk) Stapf., Brachiaria decumbens (Forsk) Stapf, Cenchrus ciliaris L., Cenchrus setigerus Vahl., Andropogon gayanus Kunth, Dichanthium annulatum (Forsk.) Stapf, Chrysopogon fulvus (Spring.) Chiov., Iseilema laxum Hack., Bothriochloa pertusa (L.) A. Camus and Chloris gayana Kunth. Test plants were grown in 25-cm diameter plastic pots. Two plants were maintained in each pot and 10 panicles of each test plant species were inoculated with each pathogen isolate following the same bagging and incubation conditions. When stigmas emerged from the first few florets at the tip of the panicles, wild sorghum genotypes and pearl millet were sprayed to runoff with a honeydew suspension containing 10 conidia/ml (Puranik and Mathre, 1971; Frederickson et al., 1989), while maize was inoculated by dipping the silk in conidial suspension prior to pollen shed. The other grass hosts were dip-inoculated when the stigmas emerged from the spikelets but before the anthers shed pollen. To determine the infectivity of conidia produced on grass hosts, honeydew was collected from each of the infected host panicles and cross-inoculated to sorghum male-sterile line 296A when flowering reached 50%. Further, to corroborate the initial observed pathogenicity on test plants, honeydew from cross-inoculated 296A sorghum was again used as inoculum to infect respective test plants. The male-sterile genotype 296A was included as a control in all tests to determine if external sources of inoculum interfered with the experiments. Non-inoculated control plants were also maintained for all species. The experiments were repeated three times using new sets of maize and pearl millet plants in each experiment. For grass hosts and wild sorghum species, inoculated tillers were removed from the pots after each experiment and a new set of panicles emerging from tillers in the same pot were inoculated in subsequent experiments. Honeydew collected from inoculated host plants infected with a representative isolate of C. africana (NI2) and C. sorghi (NAP7) were used to determine the dimensions of macroconidia and microconidia. Honeydew of 3–4-day-old from the test plants was collected for observing macroconidial dimensions, while honeydew of 15–20-day-old was collected for observing microconidial dimensions. In each experiment, shape, size and colour of 50 macroconidia and 50 microconidia from each host infected by isolates of C. africana and C. sorghi were compared. Conidial dimensions determined from three replicate experiments were analysed using sas software (SAS Institute, 1999) in a completely randomized design. Results Of the 20 plant species tested, only S. arundinaceum, S. halepense, S. versicolor, S. virgatum and P. glaucum were infected by all isolates of both pathogen species in the three replicate experiments. All infected plants showed typical honeydew symptoms. On all wild sorghum genotypes, honeydew exudation was observed 7 days after inoculation, while on P. glaucum, honeydew exudation was observed 10 days after inoculation 2 Muthusubramanian et al.