Introduction

Aerobic rice production system is a new concept in rice cultivation to increase water use efficiency (Bouman et al. 2002). However, under aerobic cultivation systems, drought-stressed rice plants were observed to be more susceptible to disease especially rice blast (Piotti et al. 2005). Rice blast caused by Pyricularia oryzae, (teleomorph: Magnaporthe oryzae B. Couch), is the most destructive and cosmopolitan disease of rice (Oryza sativa L.) (Ou 1985). Rice blast disease is a significant threat to food security worldwide. In Brazil, yield losses up to 100 % have been reported in a newly released upland cultivar (Prabhu et al. 2009). Chemical pesticides are reported to be highly effective in blast control (Yamaguchi 2004). However, chemicals with residual toxicity can result in environmental pollution, detrimental effects on non-target organisms and resistance development.

Plant growth-promoting microorganisms (PGPM) defined as rhizosphere bacteria (Kloepper et al. 1999) or fungi (Hyakumachi 1994) exert beneficial effects; on growth, yield and protection in a range of crops like cereals, pulses, ornamentals, vegetables and plantation crops. PGPM also served as determinants in triggering induced systemic resistance (ISR) in plant systems as well as eliminating deleterious rhizobacteria and fungi (Dilantha Fernando et al. 2005; Pandya and Saraf 2010).

On the other hand, compost in agriculture has long been implicated to benefit plant growth and health. Compost is a unique substrate that provides fundamental interactions between plant pathogens, biocontrol agents, soil organic matter and plant roots (Hoitink et al. 1997) that helps to improve soil physical, chemical and biological properties (Alfano et al. 2009; Bastida et al. 2010). For instance, the application of rice straw compost high in silicon and nutrients have been well documented in benefiting plant growth, soil health and disease control (Singh et al. 2007; Watanabe et al. 2009). Besides, the capability and potential of various types of composts in controlling soil-borne pathogens in various crops has been well documented (Erhart et al. 1999; Yogev et al. 2009).

However, most biocontrol agents are destroyed by heat during the composting process and recolonization of effective microorganisms during the curing process does not always occur especially in an enclosed composting system (De Ceuster and Hoitink 1999). Therefore, fortification of mature compost with biocontrol agents or PGPM was suggested to improve sustainability as well as performance in disease suppression (Hoitink et al. 2006).

There is no information on the effect of microbial-fortified compost on blast disease severity and growth of aerobic rice in association with different P. oryzae inoculation timings. The aim of the current study was to evaluate the effect of rice straw compost fortified with a consortium of four plant growth-promoting rhizobacteria (PGPR) and two fungi (PGPF) in promoting growth, yield and suppressing rice blast severity at different P. oryzae inoculation timing.

Materials and methods

Isolates and microbial-fortified rice straw compost

The four PGPR: Pseudomonas aeruginosa (UPMP1), Corynebacterium agropyri (UPMP7), Enterobacter gergoviae (UPMP9) and Bacillus amyloliquefaciens (UPMS3), and two PGPF: Trichoderma harzianum (UPMT1) and Trichoderma virens (UPMT2) previously isolated from the rhizosphere soil of aerobic rice. All six isolates had the ability to produce indole-3-acetic acid (IAA), siderophore and chitinase, solubilize phosphate and were antagonistic against P. oryzae based on in vitro screenings conducted (our unpublished result). All PGPR were identified by BIOLOG system and the PGPF were identified based on morphological characteristics (Watanabe 2002) and sequence analysis of internal transcribed spacer (ITS) 1 and 2 (White et al. 1990).

Three month-old rice straw compost obtained from PPK (Pertubuhan Peladang Kawasan) B2, Kodiang, Perlis was used in the experiment. The physico-chemical and microbiological properties of the rice straw compost are presented in Table 1. Bacterial and fungal inoculums (108 cfu/ml) were aseptically inoculated into the rice straw compost at the ratio 1:20 (inoculum:rice straw compost v/w). The moisture content of the compost was adjusted to 50 % to maintain the biological activity of the microbes.

Table 1 Physico-chemical and microbiological properties of the raw rice straw compost (Source: PPK B2, Kodiang, Perlis)
Full size table

Plant materials and growing conditions

The greenhouse experiment was conducted at Agriculture Technology Unit of Universiti Putra Malaysia (UPM), Serdang, beginning end of December 2009 to March 2010. The daily temperature ranged from 25 to 36°C with 80 to 90 % RH. Rice variety M4 used in this study was obtained from Malaysian Agricultural Research and Development Institute (MARDI), which is suitable to grow under minimum water condition, high yielding but moderate resistance to rice blast. The growth duration of variety M4 was 110 days.

Rice straw compost or microbial-fortified rice straw compost was incorporated into the laterite soil with depth of 15 cm at a rate of 10 t/ha in pots (0.40 m diameter, 0.60 m height and with holes at the bottom to allow drainage). Rice seeds were sown and seedlings thinned at 7 days after sowing (DAS) to allow only 20 seedlings per pot. Fertiliser inputs consisted of N-P2O5-K2O applied at 180-75-60 kg/ha as urea, Christmas Island Rock Phosphate (CIRP) and muriate of potash (MOP). Thirty percent of N was applied at 15, 35 and 55 DAS, respectively and the final 10 % was applied at 75 DAS. Full dose of P2O5 was applied at 15 DAS and K2O was applied in two splits, 50 % each at 15 and 55 DAS. Each pot was watered twice daily (10 am and 6 pm) with an automatic intermittent 30 min of irrigation.

Inoculum preparation and disease assessment

The blast pathogen (Pyricularia oryzae) isolated from rice variety M4 was challenge inoculated to rice seedlings at three different growing stages: 14 (vegetative), 56 (panicle initiation) and 80 (heading) DAS. Pyricularia oryzae conidia were mass produced (Mackill and Bonman 1986) and conidia density was adjusted to 105 conidia per ml and sprayed until run-off using a hand sprayer with 100 ml per pot at 14 DAS and 300 ml per pot at 56 and 80 DAS. The P. oryzae conidia suspension was added with surfactant Tween 20 (0.02 % v/v) to increase the adherence capacity of the conidia on plant surface. The P. oryzae challenged inoculation was conducted with single infection under specific greenhouse conditions where the inoculated rice seedlings were covered with plastic bag for 12 h to retain moisture and mist sprayer was continued to apply for the following 3 days to maintain RH of 95–98 % and temperature at 25–27°C.

Ten plants in each pot were randomly selected and tagged for disease severity assessment. Plant disease severity was assessed on the tagged plants at 7 days after inoculation. Rice blast rating was scored on individual plant basis using the standard evaluation system for rice (SESR) developed by IRRI (2002). Percentage of panicle blast was recorded at the harvesting stage by counting the number of blast infected panicles divided by the total number of panicles assessed. The disease severity was calculated using the following formula (Shrestha and Mishra 1994).

$$ {\text{Disease Severity}}\left( \% \right) = \frac{{\sum {\left( {{\text{number of plants in the rating }} \times {\text{ severity rating}}} \right){ } \times { 1}00} { }}}{{\left( {{\text{total number of plants observed }} \times {\text{ maximum scale}}} \right)}} $$

Disease severity data recorded were used to calculate the disease progress curve [Area Under Disease Progress Curve (AUDPC)] (Shaner and Finney 1977).

Evaluation of plant growth-promoting traits

At harvest, all 20 rice plants in each replicate were harvested and rice yields were expressed in tons per ha. Productive tillers (number of tillers with panicle) and 1000 grain weight (1000 fully filled grains) were also determined. Comparative or relative yield (%) in each treatment was calculated as follows (Shim et al. 2005).

$$ {\text{Comparative yield}}\left( \% \right) = \frac{{{\text{yield in treatment}} \times {1}00}}{\text{yield in control}} $$

Percentage yield loss and AUDPC changes with P. oryzae inoculation between 14 and 80 DAS was calculated using the following formula.

$$ {\text{Loss }}\left( \% \right){ } = \frac{{{\text{value in 14 DAS}} - {\text{value in 8}}0{\text{ DAS }} \times { 1}00}}{\text{value in 14 DAS}} $$

Growth performances of rice were determined based on plant height (cm) and leaf area index (LAI). Plant height was measured at harvest using a mechanical ruler, whereas leaf area was determined using a leaf area meter (Model LICOR-3100).

Experimental design and data analysis

The experiment was conducted in a randomized complete block design with five replications. Treatments were: soils amended with microbial-fortified rice straw compost (C), rice straw compost (N) and laterite soil alone (S). The treatments were arranged in a split-plot design where with or without P. oryzae inoculation treatments were assigned as main plot and inoculation time as subplot. All data collected were subjected to analysis of variance and tested for significance using Fisher’s Protected Least Significant Difference Test with PC-SAS software (SAS Institute, Cary, NC, 2001). Correlation coefficients were analyzed for all parameters at the three different P. oryzae inoculation timing using Statistix 8 software.

Results

Correlation of rice yield, plant growth and blast infection

It was observed that soil amended with microbial-fortified rice straw compost, significantly increased rice yield (ton/ha) at all P. oryzae inoculation timings, except at 56 DAS (Tables 2). With P. oryzae inoculation at 56 DAS, the rice yield obtained was not significantly different between the soil amended with microbial-fortified rice straw compost (5.76 t/ha) and rice straw compost alone (5.00 t/ha). Pyricularia oryzae inoculation at later stage (80 DAS) did not reduced significantly the percentage of infected panicle in soil amended with microbial-fortified rice straw compost (25.25 %) and soil amended with rice straw compost alone (27.75 %).

Table 2 Effect of Pyricularia oryzae inoculation at 14, 56 and 80 DAS on yield and disease parameters of rice cultivated at different soil treatments
Full size table

The relationships between rice blast infection, yield and growth parameters at the three P. oryzae inoculation timings were presented in Table 3. Generally, rice yield at all three P. oryzae inoculation timings was positively and significantly correlated to productive tillers, 1000 grain weight, plant height and LAI. However, AUDPC and infected panicles were negatively and significantly associated with rice yield. Comparison between coefficients at the three P. oryzae inoculation timing showed that the correlation of rice yield was higher in productive tiller numbers per pot (r = 0.96), LAI (r = 0.96) and plant height (r = 0.97) when inoculated at 14 DAS. Besides, AUDPC (r = −0.62), infected panicles (r = −0.59) and 1000 grain weight (r = 0.96) showed greater correlation with rice yield for P. oryzae inoculated at 80 DAS as compared to inoculation at 14 and 56 DAS.

Table 3 Correlation coefficients among disease, yield and growth traits with inoculation of Pyricularia oryzae at 14, 56 and 80 DAS
Full size table

In terms of yield loss, P. oryzae infection at the later stage (80 DAS) caused more severe yield loss. Infection of P. oryzae at 80 DAS, resulted only 53.83 % of yield obtained for soil amended with microbial-fortified rice straw compost, 52.27 % for soil amended with rice straw compost and 21.61 % for laterite soil alone, as compared to the un-inoculated with P. oryzae treatment (100 %).

Estimation of growth performance, yield loss and blast development

The yield loss at different P. oryzae inoculation times were determined using the regression equations between rice yield and inoculation timing, where y = −0.03x + 7.58 (R 2 = 0.99) for soil amended with microbial-fortified rice straw compost, y = −0.04x + 7.06 (R 2 = 0.99) for soil amended with rice straw compost and y = −0.002x + 0.46 (R 2 = 0.20) for laterite soil alone. In all the cases, there was a negative correlation between rice productivity and time of blast initiation. Yield obtained in soil amended with microbial-fortified rice straw compost was significantly higher as compared to soil amended with rice straw compost alone at all P. oryzae inoculation timings (Fig. 1a).

Fig. 1
figure 1

Linear correlation between rice yield (a), 1000 grain weight (b), AUDPC (c), blast infected panicle incident (d), productive tiller per pot (e), leaf area index (f), plant height (g) and P. oryzae inoculation timings in different soil treatments (Each value represents the mean of five replicates; vertical bars represent standard errors)

Full size image

A negative correlation between 1000 grain weight and P. oryzae inoculation timings was observed (Fig. 1b). One thousand grain weight in all soil treatments decreased significantly when P. oryzae was inoculated at the later growth stages. The regression equations between 1000 grain weight loss and P. oryzae inoculation timings were y = −0.06x + 30.83 (R 2 = 0.84) for soil amended with microbial-fortified rice straw compost, y = −0.06x + 29.61 (R 2 = 0.92) for soil amended with rice straw compost and y = −0.02x + 22.31 (R 2 = 0.44) for laterite soil alone.

To estimate rice blast severity at different growth stages with the different soil treatments, the regression between AUDPC and P. oryzae inoculation timings was developed. The AUDPC derived from leaf blast disease severity data showed negative correlation with time of blast initiation. Rice blast occurrence at early growth stages gave higher AUDPC values compared to later growth stages. The regression equations for soil amended with microbial-fortified rice straw compost, rice straw compost and laterite soil alone were y = −19.27x + 1572.30 (R 2 = 0.92), y = −40.22x + 3456 (R 2 = 0.92) and y = −64.91x + 6149.6 (R 2 = 0.96), respectively (Fig. 1c). However, blast infected panicle (%) was found to be positively correlated with P. oryzae inoculation timing (Fig. 1d). In the treatment with laterite soil alone, when P. oryzae was inoculated at 14 DAS, incidence of panicle blast was observed at harvesting stage. However, no panicle blast was detected in soils amended with microbial-fortified or rice straw compost. Panicle blast was very severe in non-amended soil (laterite soil alone) (y = 0.68x + 1.13, R 2 = 0.99), followed by soil amended with rice straw compost (y = 0.42x – 6.19, R 2 = 0.99) and soil amended with microbial-fortified rice straw compost (y = 0.36x – 6.90, R 2 = 0.88).

Pyricularia oryzae inoculation at the later stages increased the number of productive tillers in all treatments (Fig. 1e). Plants in soil treated with microbial-fortified rice straw compost had significantly higher number of productive tillers than in soil amended with only rice straw compost. The regression equations for productive tiller numbers in relation to P. oryzae inoculation timings were y = 0.19x + 45.57 (R 2 = 0.98), in soil amended with microbial-fortified rice straw compost, y = 0.23x + 37.76 (R 2 = 0.88) in soil amended with rice straw compost and y = 0.14x + 12.54 (R 2 = 0.93) in laterite soil alone.

Leaf area index of rice grown in soil treated with microbial-fortified rice straw compost was significantly and positively correlated to P. oryzae inoculation timing (Fig. 1f). However, there was no significant correlation between LAI and P. oryzae inoculation timing in laterite soil alone, where the correlation equation was y = 0.11x + 18.14 (R 2 = 0.76). The linear regression equations for LAI and the P. oryzae occurrence timing were y = 2.50x + 220.94 (R 2 = 0.97) in soil amended with microbial-fortified rice straw compost and y = 1.28x + 186.04 (R 2 = 0.99) in soil amended with rice straw compost.

Plant height in soil amended with microbial-fortified rice straw compost ranged from 110.70 to 111.50 cm, but was weakly correlated with P. oryzae inoculation timing, where the regression equation was y = 0.01x + 110.66 (R 2 = 0.67) (Fig. 1g). However, in soil amended with rice straw compost, delaying P. oryzae inoculation increased plant height (y = 0.15x + 95.25; R 2 = 0.98). In the laterite soil alone, plant height was not significantly correlated to P. oryzae inoculation timing, where plant height recorded was consistent and ranged between 41.40 and 47.50 cm (y = 0.09x + 40.23; R 2 = 0.99).

Discussion

Rice blast disease is a major constraint in rice production world wide and panicle infection is capable of causing complete yield loss (Shim et al. 2005). In the current study, rice straw compost fortified with P. aeruginosa, B. amyloliquefaciens, C. agropyri, E. gergoviae, T. harzianum and T. virens demonstrated significant effects in promoting plant growth, controlling rice blast and resulting in yield increments especially at early P. oryzae inoculation. However, blast disease inoculation at the mature plant stage (56 DAS) did not cause significant yield loss between soil amended with microbial-fortified rice straw compost (5.76 t/ha) or rice straw compost (5 t/ha). This observation reflects the high resistance of mature plants against P. oryzae infection as compared to the seedling stage with higher leaf blast severity (14 DAS). Consequently, the low leaf blast severity with infection of mature plants (56 DAS) did not significantly disturb the carbohydrate supply of the canopy during the specific panicle initiation period and hence did not affect floret number and spikelet number which were determined before heading (Penning de Vries et al. 1990).

The potential of Pseudomonas sp. (Ryu et al. 2004), Bacillus sp. (Idriss et al. 2002), Enterobacter (Gupta et al. 1998), Corynebacterium (El-Banna and Winkelmann 1998) and Trichoderma (Pandya and Saraf 2010) in enhancing plant growth and improving plant health through various direct and indirect mechanisms have been reported. The positive results of PGPM in controlling leaf blast were also reported in irrigated rice (Lucas et al. 2009) and aerobic rice (Filippi et al. 2011). In our study, the capability of PGPM to enhance productivity was explained by the production of peroxidase, polyphenol oxidase and phenylalanine ammonia-lyase as inducers of systemic resistance in the rice plant (Filippi et al. 2011) and also the plant growth promoting effects of the fortified microbial which related to the production of IAA (Khalid et al. 2004) and solubilizing of phosphate (Vyas and Gulati 2009).

Efficiency in blast management is dependent on accurate and precise knowledge of disease incidence and relationship between disease severity and yield loss (Tsai 1988). In our current study, correlation and estimation of various plant growth and blast disease parameters in different soil treatments at different P. oryzae inoculation timings were stressed. Pyricularia oryzae inoculation at 14 DAS demonstrated highest positive correlation in rice yield (ton/ha) to LAI, plant height and productive tiller number. Early P. oryzae inoculation led to decline in leaf area formation, plant height and productive tillers. These results are in agreement with Crill et al. (1982); Yeh and Bonman (1986), where rice plants at seedling and early tillering stages were more susceptible to leaf blast infection. Leaf blast was reported to reduce yield by causing lesion coverage, premature leaf senescence, reducing leaf area formation and phothosynthetic efficiency (Bastiaans et al. 1994). From our observation, heavily blast infected leaf tissues were found dying, but with the formation of new green leaves there was a gradual decline in leaf blast infection attributed to increased resistance in mature plants. Blast infected panicles were not observed at harvest except in treatments with laterite soil alone. Despite the high correlation of rice productivity to LAI, plant height and productive tillers at early P. oryzae inoculation, the comparative yield scored was still the highest compared to late inoculation of P. oryzae. This was in agreement with Crill et al. (1982), where leaf blast was the most commonly found in seedling but least destructive as compared to panicle blast.

The low rice yield with late inoculation of P. oryzae was correlated with high AUDPC, blast infected panicles and low 1000 grain weight. Similar observations by Bastiaans (1993) led to the suggestion that leaf blast mainly affected grain yield indirectly. Despite the low AUDPC value in mature plants, yield reduction might also be affected by low efficiency of blast lesions on leaf photosynthesis and respiration causing reduced carbohydrate production (Bastiaans 1991). These observations are in line with those reported by Goto (1965), who also observed reduced spikelet number, 1000 grain weight and fraction of filled grains due to leaf blast in mature plants. In addition, AUDPC value at late P. oryzae inoculation (80 DAS) was highly correlated to panicle blast infection. This resulting percentage of panicle blast was not significantly different in soil amended with microbial-fortified rice straw compost or with rice straw compost alone. Panicle blast at late P. oryzae infection had also directly affects on yield loss (Shim et al. 2005) by causing sterile or poorly filled grain and resulting in light weight kernels (Crill et al. 1982).

The low comparative yield with late P. oryzae inoculation may be explained by the fact that P. oryzae infection at late growing stages increased blast panicle incidence and affected the 1000 grain weight. The later inoculation of P. oryzae at 80 DAS caused yield losses of 30.99 % and 36.28 % in soil amended with microbial-fortified rice straw compost and with rice straw compost alone as compared to early infection at 14 DAS. Yield reductions were caused by panicle blast which reduced 1000 grain weight, percentage of ripe spikelets and percentage of fully mature grains (Teng et al. 1990). This indicates that panicle infection was critical for significant economic losses.

The application of microbial-fortified rice straw compost was beneficial in the control of rice blast development, promotion of crop growth and improving productivity. However, reduced grain filling in blast infected panicles was the most critical factor contributing to yield loss in aerobic rice (variety M4) production. The ability of prediction equations to accurately estimate yield loss and various plant growth parameters due to blast disease development at different P. oryzae inoculation timings can be useful for blast management in aerobic rice cultivation systems.