Ascospore discharge by Fusarium graminearum as affected by temperature and relative humidity
Fusarium head blight (FHB) is a disease of increasing concern in the production of wheat (Triticum aestivum). This work studied some of the factors affecting the density of airborne Gibberella zeae inoculum. Spore samplers were placed at the edge of a field in order to observe spore deposition over a period of 45 days and nights in September and October, the period that coincides with wheat flowering. Gibberella zeae colonies were counted for each period and values transformed to relative density. A stepwise regression procedure was used to identify weather variables helpful in predicting spore cloud density. In general, a predominant night-time spore deposition was observed. Precipitation and daily mean relative humidity over 90% were the factors most hightly associated with peak events of spores in the air. Models for predicting spore cloud density simulated reasonably well with the fluctuation of airborne propagules during both night and day, with potential to be integrated into an FHB risk model framework. Additional keywords: Fusarium graminearum, aerobiology, spore dispersal, epidemiology. RESUMO Fatores que influenciam a densidade de inóculo aéreo de Gibberella zeae A giberela, causada por Gibberella zeae, é uma das mais importantes doenças do trigo (Triticum aestivum). O presente trabalho objetivou estudar alguns aspectos que influenciam a densidade de inóculo aéreo. Armadilhas caçaesporos foram instaladas ao lado de parcelas experimentais de trigo com coletas realizadas durante o período da noite e do dia, por 45 dias nos meses de setembro a outubro, coincidindo com o florescimento do trigo. O número de colônias de G. zeae, avaliadas em cada período, foi transformado para valores relativos e correlacionado com variáveis ambientais. Uma análise de regressão foi utilizada para se identificar as variáveis climáticas úteis para a predição da densidade da nuvem de esporos. Foi observada uma predominância de deposição de inóculo aéreo no período da noite. Ocorrência de chuva e umidade relativa média diária superior a 90% estiveram fortemente associados com picos de concentração de inóculo aéreo. Os modelos simularam os incrementos na densidade da nuvem de esporos, apresentando potencial para serem integrados em um modelo de simulação do risco de giberela. Palavras-chave adicionais: Fusarium graminearum, aerobiolobia, dispersão de esporos, epidemiologia. *Parte da tese de doutorado do primeiro autor. Universidade Federal de Pelotas (2004). **Bolsista do CNPq 55 Fitopatol. bras. 30(1), jan fev 2005 INTRODUÇÃO Fusarium head blight (FHB) of wheat (Triticum aestivum L.), caused mainly by Gibberella zeae (Schwein.) Petch. (anamorph Fusarium graminearum Schwabe), is a disease of increasing concern in the production of wheat (Mcmullen et al., 1997). Contamination of the grain by mycotoxins such as deoxynivalenol (DON) (Jones & Mirocha, 1999) has caused economic losses due to decreased quality and yield of the grain. Severe epidemics have been reported in Brazil in recent years which have resulted in considerable yield losses (Panisson et al., 2003). Typically, FHB is a floral disease infecting wheat during extrusion of anthers, although infections are also likely to occur from flowering to grain filling stages (Sutton, 1982; McMullen et al., 1997). Ascospores and macroconidia of the pathogen are produced on over-wintered residues, and either propagule may infect wheat under favorable weather conditions. Ascospores are considered the most important spore type in terms of dispersal potential (Parry et al., 1995; Paulitz, 1999) although macroconidia may play a role in the epidemics especially in regions where other Fusarium spp., which do not form a sexual stage, are important FHB pathogens (Rossi et al., 2002). Several empirical weatherbased models have been developed for predicting disease incidence, epidemic risk, or DON levels (Moschini & Fortugno, 1996; Hooker et al., 2002; De Wolf et al., 2003). A realistic approach for modeling FHB risk should take into account the effect of inoculum density and host developments as well weather variables (Fernandes & Pavan, 2002). Studies on the DELPONTE, E.M., FERNANDES, J.M.C. & PIEROBOM, C.R. Factors affecting density of airborne Gibberella zeae inoculum. Fitopatologia Brasileira 30:55-60. 2005. E.M. Del Ponte et al. 56 Fitopatol. bras. 30(1), jan fev 2004 temporal and spatial aspects of G. zeae airborne inoculum aimed at clarifying the processes and mechanisms favoring dispersal of propagules over either short or long distance are found in the literature (Ayers et al., 1975; Reis, 1990; Paulitz, 1996; De Luna et al., 2002). The present study investigates some weather variables that affect the fluctuation of airborne G. zeae inoculum in a wheat field. Identification of relevant weather variables can be useful for predicting the density of a G. zeae spore cloud to be incorporated into a FHB risk model framework. MATERIAL AND METHODS This study was carried out from September 1st to October 15th, 2003 at Embrapa Trigo, Passo Fundo, RS. Airborne inoculum of G. zeae was detected on selective media for Fusarium species (Nash & Snyder, 1962) deposited in plastic Petri dishes (9 cm diameter) mounted on two spore sample types. The first was a wind-driven sampler previously used by Reis (1988) and the second, called platform, was an adaptation of the sampler used by Schmale et al. (2002) for detection of G. zeae spores by gravitational settling. Plates containing selective media were placed at 1.3 m above the soil surface. Samplers were located 1 m apart on the border of experimental wheat plots, in an area surrounded by wheat and other cereal crops cultivated under the no-till system, which exhibited high levels of residue on the soil surface. Two daily samplings were performed at 9:00 and 21:00 h. Plates were exposed in two periods of 12 h each, called nightand day-time sampling. After exposure to the environment, the plates were transported to the laboratory and incubated in a growth chamber (25 oC and 12 h of darkness) in order to promote fungi growth. The number of G. zeae colonies was recorded for each plate as CFU (colony forming units) / 63 cm2. Prior to evaluation, a sample of 20 confirmed true G. zeae colonies (according to Nelson et al., 1983) were plated on the medium for comparisons during the evaluations. Other Fusarium species were observed but not identified at the species level. Weather information was acquired at Passo Fundo weather station (latitude -28,25o, longitude -54,4o, altitude 684 m) located 10 m from the wheat plots. The SAS statistical package (SAS Institute, Inc., Cary, NC) was used to explore data, calculate summary statistics and calculate Pearson’s correlation between pairs of variables. To study the weather effect on spore cloud density, a set of weather variables was created based on biological criteria and previous studies. The CFU number was transformed into relative density by dividing the number of colonies in one day by the maximum number of CFU detected during the sampling period. Those values were square-root transformed to normalize data and used as dependent variables. A stepwise regression (0.1 significance level for entry into the model) was applied to select the smallest set of independent variables related to the dependent one. Results of the regression analysis were evaluated on the basis of the standard error of model parameters, residual distribution, coefficient of determination (R2), and adjusted R2. The analysis was applied to both a combination of nightand day-time observations and to only night-time observations. G. zeae colonies and weather information from October 16th to October 23rd was used to evaluate the model’s performance.