Disease assessment, phytohormone and toxin levels during Sarocladium oryzae infection
To study the interaction of S. oryzae with its host and the role of its toxins cerulenin and helvolic acid herein, rice plants were inoculated with four isolates (Table 1) that were earlier shown to differ in virulence and toxin production (Table S1; Peeters et al. 2020). The rice sheaths of the second youngest leaf, enclosing the colonized grain, were collected at 4 hours after inoculation (0 days post inoculation ((DPI)) and 2 and 4 DPI. Although all isolates were able to cause long, brown necrotic lesions on the sheath enclosing the young panicle, significant differences were observed in the lesion size at 4 DPI (Fig 1).
At 4 hours after infection, low amounts of helvolic acid were detected in rice sheaths inoculated with the three most pathogenic isolates IBNG0008, BDNG0025 and RFNG41 (Fig. 2a). During the first 48 hours, their helvolic acid production increased. Isolate BDNG0025 showed the highest helvolic acid production at every time point. The least pathogenic isolate RFNG30, on the contrary, produced trace amounts of helvolic acid only at one replicate at 4 DPI, while further no helvolic acid could be measured at all (Fig. 2a). Cerulenin was not detected at 4 hours after inoculation. At 2 and 4 DPI, high amounts of cerulenin were measured in sheaths infected with IBNG0008, while BDNG0025 and RFNG41 showed a lower production. At none of the stages of the infection process, cerulenin could be detected for the isolate with the lowest virulence (RFNG30) (Fig. 2b).
In addition to the toxins, the levels of the phytohormones ABA, JA, IAA and SA were measured (Fig. 2c-f). At 4 hours after inoculation (0 DPI), no changes in ABA concentration could be observed. At 2 DPI, ABA levels were elevated in plants inoculated with the two most pathogenic isolates (IBNG0008 and BDNG0025) and, by 4 DPI, ABA levels were further increased (Fig. 2c). Compared to the healthy control plants, a ten-fold increase of ABA was measured for the virulent isolates (IBNG0008 and BDNG0025) while ABA levels were only doubled in plants inoculated with RFNG41 and RFNG30 (Fig. 2c). JA, on the other hand, showed a transient increase in response to all S. oryzae isolates. By 2 DPI, JA levels again decreased in sheaths infected by isolates with low pathogenic potential (RFNG41 and RFNG30) and by 4 DPI, JA levels were still comparable to the concentration of JA in the healthy control plants. For the virulent isolates IBNG0008 and BDNG0025, on the contrary, JA levels stayed elevated during the rest of the infection (Fig. 2d). IAA levels were altered by S. oryzae in a similar pattern as ABA. At 4 hours after inoculation, IAA levels were equal in all treatments. By 2 DPI, the virulent isolates (IBNG0008 and BDNG0025) had caused an increase of IAA which stayed elevated. In less diseased plants, similar levels as in the healthy control plants were measured at all sampling points (Fig. 2e). SA levels did not change in response to S. oryzae infection, although a small, not significant decrease of SA was observed at 4 hours after inoculation of IBNG0008 (Fig. 2f).
To further elucidate the correlation between the hormonal response, the toxin production and the virulence, a wide selection of well characterized S. oryzae isolates was used (Table 1). Figure 3 shows the phytohormone levels in the rice sheaths of the second youngest leaf at 6 DPI. The corresponding virulence data and toxin levels measured in the sheaths have been reported in previously published work (Peeters et al. 2020) and the averages are listed in Table S1. In agreement with the results shown in Fig. 2, ABA, JA and IAA levels were the most elevated in rice sheaths infected with IBNG0008 and BDNG0025. Together with IBNG0009, to which phytohormone levels responded in a similar way, these isolates were the most pathogenic (Fig. 3a-c, Table S1). All three isolates produced high levels of helvolic acid, while their cerulenin production strongly differed. In agreement with Fig. 2, BDNG0025 produced significantly less cerulenin than IBNG0008 (Table S1). One isolate (BDNG0005) produced a similar amount of helvolic acid as BDNG0025 but did not trigger ABA, JA or IAA (Fig. 3a-c, Table S1). It did however cause a small but significant decrease of SA compared to the healthy control plants and this isolate was the least virulent of all (Peeters et al. 2020). The isolate RFBG3 elevated levels of IAA while it caused only minor symptoms and produced no cerulenin or helvolic acid (Fig. 3a-c, Table S1).
Disease assessment and phytohormone levels during Pseudomonas fuscovaginae infection
Two wild type strains were used to study the phytohormone response to P. fuscovaginae infection. For this, rice plants were injected with a bacterial solution and disease was evaluated by measuring the lesion size around the inoculation point (Fig. 4). The strains were equally virulent and caused similar brown, necrotic lesions on both the sheath and the stem of the rice plant (Fig. 4a).
At 4 hours after inoculation (0 DPI) and 2, 4 and 8 DPI, samples of the rice sheath around the junction point were collected. Figure 5 shows that the rice plants strongly responded to the inoculation method by elevating ABA. The strongest increase was observed in the control plants which were inoculated with saline solution (Fig. 5a). By 2 DPI, ABA levels had decreased in all conditions. A late response to both wild type strains was observed at 8 DPI for both ABA and JA (Fig. 5a,b). IAA, on the other hand, showed a small transient increase in response to P. fuscovaginae. At 2 DPI, the IAA increase had started and by 4 DPI, IAA levels had dropped again for MB194 while in UPB0736 infected plants, IAA was still elevated. By 8 DPI, IAA concentrations were equal to the healthy control plants (Fig. 5c). SA did not respond to P. fuscovaginae infection (Fig. 5d).
The role of the lipopeptide fuscopeptin in these hormonal responses was investigated by inoculating the rice plants with P. fuscovaginae wild type strain UPB0736 and its fuscopeptin mutant delta445. Preliminary results (Fig. S1) showed a stronger response in the stem and lesions were larger on the third youngest leaf sheath so the latter was used for disease evaluation and sampled for hormone measurements. The results in Fig. 5c show a response of the rice plants to P. fuscovaginae at 2 DPI. Therefore, disease was scored at 2, 4 and 8 DPI. At every time point, inoculation with the wild type strain UPB0736 caused similar patterns of infection, while the fuscopeptin mutant strain delta445 was significantly less virulent (Fig. 6a).
To reduce the effect of the inoculation procedure, samples were collected at 1 DPI, 2 DPI, 4 DPI and at 8 DPI (Fig. 6). At the first three sampling points (1, 2, 4 DPI), ABA levels were similar in all conditions. By 8 DPI, plants inoculated with UPB0736 showed elevated ABA levels while plants inoculated with the fuscopeptin mutant delta445 did not show any ABA response (Fig. 6a). By 2 DPI, JA levels were slightly elevated in response to both P. fuscovaginae strains. Only in plants inoculated with UPB0736,
JA levels stayed elevated during the rest of the infection process (Fig. 6b). For IAA, on the other hand, an early transient response to the wild type strain was observed. At 1 and 2 DPI, IAA levels in the wild type infected plants were more than doubled compared to the healthy control plants and plants inoculated with delta445 (Fig. c). By 8 DPI, SA levels were slightly elevated in UPB0736 infected plants but the levels did not significantly differ from the plants inoculated with delta445 (Fig. 6d).
Yield losses
The yield losses, as a result of sheath rot infection, were investigated in rice plants that were inoculated with isolates of S. oryzae and P. fuscovaginae that differed in virulence (Fig. 1, 5a). Disease severity and various yield parameters were recorded at 6 and 8 weeks post inoculation (WPI) with respectively S. oryzae and P. fuscovaginae. At 6 WPI, the infection by the pathogenic S. oryzae strain, IBNG0008, had strongly advanced (average 842 ± 209 mm²). Also the infection by the less pathogenic isolate RFNG30 had further progressed (average 96 ± 53 mm²). According to the observed lesions on the rice sheath, P. fuscovaginae infection did not seem to have advanced much in 8 weeks (UPB0736, average 213 ± 133 mm²; delta445, average 11 ± 10 mm²). However, inside the stem, the rice plants showed necrosis at 8 WPI of both P. fuscovaginae strains but also some healthy control plants, injected with sterile saline solution, were necrotic inside.
Both sheath rot pathogens affected the yield (Fig. 7a). Plants infected with S. oryzae IBNG0008 produced shorter panicles with less seeds, compared to healthy plants. RFNG30 infection also reduced panicle length (Fig. 7b). The seeds produced by the plants infected with the IBNG0008 were significantly more open and more empty, compared to the control plants. Plants infected with the less pathogenic strain RFNG30 produced as much seeds as healthy plants, but the seeds were more often open (Fig. 7a,c). Plants inoculated with P. fuscovaginae wild type strain UPB0736 did not produce panicles at all (Fig. 7d,e). Inoculation with the fuscopeptin mutant delta445 also led to a significant decrease in panicle and seed production compared to the control plants (Fig. 7f).