Abstract
Asian soybean rust (ASR) caused by Phakopsora pachyrhizi, an obligate biotrophic fungal pathogen, is the most devastating soybean production disease worldwide. Currently, timely fungicide application is the only means to control ASR in the field. We investigated cellulose nanofiber (CNF) application on ASR disease management. CNF-treated leaves showed reduced lesion number after P. pachyrhizi inoculation compared to control leaves, indicating that covering soybean leaves with CNF confers P. pachyrhizi resistance. We also demonstrated that formation of P. pachyrhizi pre-infection structures including germ-tubes and appressoria, and also gene expression related to these formations, such as chitin synthases (CHSs), were significantly suppressed in CNF-treated soybean leaves compared to control leaves. Moreover, contact angle measurement revealed that CNF converts soybean leaf surface properties from hydrophobic to hydrophilic. These results suggest that CNF can change soybean leaf surface hydrophobicity, conferring resistance against P. pachyrhizi, based on the reduced expression of CHSs, as well as reduced formation of pre-infection structures. This is the first study to investigate CNF application to control field disease.
Introduction
Diseases in important crop plants have a significant negative impact on agricultural productivity. For example, Asian soybean rust (ASR) caused by Phakopsora pachyrhizi, an obligate biotrophic fungal pathogen, is the most devastating soybean production disease worldwide, with an estimated crop yield loss of up to 90%. ASR has impacted the South American economy in recent years. Yorinori et al. (1) reported that the losses caused by ASR were ~2 billion US dollars in Brazil alone in 2003. Although most rust fungi have a high host specificity, the P. pachyrhizi host range is broad and can infect diverse leguminous plant leaves in the field (2). The infection process starts when urediniospores germinate to produce a single germ-tube with an appressorium. Unlike cereal rust fungi that penetrates through stomata (3), P. pachyrhizi directly penetrates into host plant epidermal cells by an appressorial peg. After penetration, P. pachyrhizi extends the infection hyphae and forms haustoria (feeding structures) in the mesophyll cells 24 to 48 hours after infection (4). At five to eight days after infection, P. pachyrhizi then produces urediniospores by asexual reproduction (4). Urediniospores can be dispersed by wind and germinate on other host plants.
There are several ASR control methods for soybean protection against P. pachyrhizi, including chemical control by fungicide application, growing ASR resistant soybean cultivars, and employing cultivation practices. Synthetic fungicides are the primary ASR disease control method. However, fungicide use can cause many problems such as environmental impacts (5), increased production costs (6), and the emergence of fungicide-resistant pathogens (7, 8). Another major and effective control method is breeding or engineering of ASR resistant soybean cultivars. Analysis of soybean accessions disclosed six dominant R genes conferring resistance to a particular P. pachyrhizi race, and these loci were referred to as the Rpp 1–6 genes (9–13). However, none of the soybean accessions in the world show resistance to all P. pachyrhizi races (14). Due to the limited resistance available in soybean cultivars, heterologous expression of resistance genes from other plant species in soybean has been investigated as an alternative source of ASR resistance. Kawashima et al. (15) reported that soybean plants expressing CcRpp1 (Cajanus cajan Resistance against Phakopsora pachyrhizi 1) from pigeon pea (Cajanus cajan) showed full resistance against P. pachyrhizi. Conversely, identifying resistance traits from non-host plant species has become an intelligent approach. Uppalapati et al. (16) screened Medicago truncatula Tnt1 mutant lines and identified an inhibitor of rust germ tube differentiation 1 (irg1) mutant with reduced formation of pre-infection structures, including germ-tubes and appressoria. They demonstrated that the loss of abaxial epicuticular wax accumulation resulting in reduced surface hydrophobicity inhibited formation of pre-infection structures on the irg1 mutant (16). Furthermore, Ishiga et al. (17) reported that gene expression related to pre-infection structure formation were activated on the hydrophobic surface of the M. truncatula wild-type, but not on the irg1 mutant, based on P. pachyrhizi transcriptome analysis, suggesting that leaf surface hydrophobicity can trigger gene expression related to formation of pre-infection structures. Based on these previous studies, we hypothesized that modification of leaf surface hydrophobicity might be a useful strategy to conferring resistance against P. pachyrhizi.
Cellulose is an organic polysaccharide consisting of a β-1,4 linked glucopyranose skeleton. Cellulose is an important structural component of plant primary cell walls and is essential in maintaining the plant structural phase. Due to the positive properties, cellulose has been investigated as an application in different research and development fields including energy, environmental, water, and biomedical related fields (18). Cellulose nanofiber (CNF), which can be derived from cellulose, is one of the most abundant and renewable biomasses in nature (19). Because CNF exhibits properties such as low weight, high aspect ratio, high strength, high stiffness, and large surface area, CNF potentially has wide areas of application. There are several CNF isolation methods, e.g. acid hydrolysis, enzymatic hydrolysis, and mechanical processes. The aqueous counter collision (ACC) method can make it possible to cleave interfacial interactions among cellulose molecules without any chemical modification (20). Both hydrophobic and hydrophilic sites co-exist in a cellulose molecule resulting in amphiphilic properties when CNF is derived from the ACC method. Kose et al. (21) reported that coating with CNF derived from the ACC method could switch surface hydrophilic and hydrophobic properties, depending on substrate characteristics. They demonstrated that coating a filter paper and polyethylene with CNF changed the surface property into hydrophobic and hydrophilic, respectively (21). To investigate the potential application of CNF in agriculture, we examined whether coating with CNF protected soybean plants against P. pachyrhizi. We show that a specific CNF property can change soybean leaf surface hydrophobicity, resulting in reduced formation of pre-infection structures associated with reduced P. pachyrhizi infection.
Materials & Methods
Plant growth conditions, CNF treatment and pathogen inoculation assay
Susceptible soybean cultivar seeds (Glycine max cv. Enrei) were germinated in a growth chamber at 25°C/20°C with 16-hrs-light/8-hrs-dark cycle (100-150 μ mol m−2 s−1) for 3 to 4 weeks.
Cellulose nanofiber (CNF, marketed as nanoforest®) supplied through the courtesy of the Chuetsu Pulp & Paper (Takaoka, Japan) was used. A bamboo-derived CNF (BC) and a needle-leaved tree-derived CNF (NC) were adjusted to a concentration of 0.1% including 0.02% Tween 20 (FUJIFILM, Tokyo, Japan) before treatment. Both adaxial and abaxial sides of soybean leaves were spray-treated with 0.1% CNF till runoff and then the treated soybean plants were dried at room temperature for 3 to 4 hours before inoculation.
An isolate of the ASR pathogen P. pachyrhizi T1-2 (22) was maintained on soybean leaves. Fresh urediniospores were collected and suspended in distilled water with 0.001% Tween 20. The 3-week-old soybean plants were spray-inoculated with 1 × 105 spores/ml using a hand sprayer for uniform spore deposition. The inoculated plants were maintained in a chamber for 24 hours with 90% to 95% humidity at 23°C; 0-hrs-light/24-hrs-dark cycle. The plants were then transferred to a growth chamber (22°C/20 °C with 16 hrs-light/8 hrs-dark cycle) and incubated further to allow symptom development.
To quantify lesion number on ASR on CNF-treated plants, soybean leaves were spray-inoculated with P. pachyrhizi. At 10 days after inoculation, photographs were taken, and lesions were counted to calculate the lesion number per cm2. Lesions were counted from 54 random fields on three independent leaves.
To quantify the formation of pre-infection structures including germ-tubes and appressoria on control and CNF-treated plants, soybean leaves were spray-inoculated with P. pachyrhizi 1 × 105 spores/ml. At 72 hours after inoculation, the leaves were observed and photographed with the desktop scanning electron microscope (HITACHI TM3000, Tokyo, Japan). The germ-tubes forming differentiated appressoria were counted as appressoria. The differentiated germ-tubes without appressoria that grew on the leaf surface were also counted from 54 random fields on three independent leaves.
Real-time quantitative RT-PCR analyses
To investigate the gene expression profiles related to pre-infection structures, approximately 100 P. pachyrhizi spores in 10 μl aliquots were placed on the abaxial surface of 4-week-old detached soybean leaves with or without 0.1% CNF and incubated in darkness overnight, and then transferred to a growth chamber (22°C/20°C with 16-h-light/8-h-dark cycle). At 24 and 48 hours after inoculation, total RNA was extracted from the inoculated leaf areas and purified using RNAiso Plus (TaKaRa, Otsu, Japan) according to the manufacture’s protocol. For soybean pathogenesis-related gene protein 1 (GmPR1) expression profiles, soybean leaves were treated with or without 0.1% CNF. At 24 hours after CNF treatment, total RNA was extracted from leaves and purified using RNAiso Plus (TaKaRa) according to the manufacture’s protocol. Two micrograms of total RNA were treated with gDNA Remover (TOYOBO, Osaka, Japan) to eliminate genomic DNA, and the DNase-treated RNA was reverse transcribed using the ReverTra Ace qPCR RT Master Mix (TOYOBO). The cDNA (1:10) was then used for RT-qPCR using the primers shown in Supplementary Table S1 with THUNDERBIRD SYBR qPCR Mix (TOYOBO) on a Thermal Cycler Dice Real Time System (TaKaRa). P. pachyrhizi Elongation factor 1α (PpEF1α) and Ubiquitin 5 (PpUBQ5) were used to normalize P. pachyrhizi gene expression. Soybean Actin 4 (GmAct4) was used as an internal control to normalize soybean GmPR1 gene expression.
Contact angle measurement on soybean leaves
The surface hydrophobicity on the CNF-treated leaves was investigated based on contact angle measurement using an automatic contact angle meter DM-31(Kyowa Interface Science, Niiza, Japan). The contact angle was measured by dropping 2 μl of water from a syringe attached to the DM-31 automatic contact angle meter. The contact angle was measured on the adaxial and abaxial leaf surfaces with or without 0.1% CNF treatments. The contact angle was analyzed using the multi-functional integrated analysis software FAMAS (Kyowa Interface Science).
Results
Covering soybean leaves with CNF confers resistance against P. pachyrhizi
To investigate the potential application of CNF in agriculture, especially disease resistance against pathogens, we first treated soybean leaves with two CNF types derived from bamboo (BC) and needle-leaved tree (NC). At 4 hours after spraying with 0.1% CNF, we challenged soybean leaves with P. pachyrhizi and observed lesion formation including uredinia at 10 days after inoculation. Both CNF-treated leaves showed reduced lesion area compared to control leaves (Fig. 1A). Both CNF-treated leaves showed significantly reduced lesion number compared to control leaves (Fig. 1B). These results indicate that covering soybean leaves with CNF confers resistance against P. pachyrhizi.
Nanofibers such as chitin nanofibers induce plant immune responses by activating defense-related gene expression (23). Therefore, one could argue that the CNF-induced resistance phenotype in soybean plants may result from defense response activation rather than from the direct effects of CNF treatments against P. pachyrhizi. To rule out this possibility, we investigated the expression profiles of the defense marker gene GmPR1 after CNF treatments. GmPR1 expression in CNF-treated leaves showed no significant induction compared to control leaves (Fig. S1). These results confirmed that the CNF-induced resistance phenotype against P. pachyrhizi is a direct effect of CNF treatment.
Covering soybean leaves with CNF suppresses formation of P. pachyrhizi pre-infection structures
Since both CNF-treatments suppressed the lesion number, we next investigated the formation of pre-infection structures including germ-tubes and appressoria on CNF-treated leaves. In control leaves, around 60% of urediniospores germinated, and ~15% and ~30% formed appressoria on adaxial and abaxial leaves, respectively (Fig. 2A and Fig. 2B). In CNF-treated leaves, around 60% of urediniospores germinated, and interestingly less than 5% of them formed appressoria on both adaxial and abaxial leaves (Fig. 2A and Fig. 2B). These results suggest that covering soybean leaves with CNF suppresses formation of pre-infection structures including germ-tubes and appressoria.
Covering soybean leaves with CNF changes gene expression profiles related to formation of pre-infection structures
Ishiga et al. (17) reported that gene expression related to formation of pre-infection structures was induced on the hydrophobic surface based on P. pachyrhizi transcriptome analysis. Since CNF-treatments suppressed formation of pre-infection structures including germ-tubes and appressoria, we next investigated gene expression profiles related to formation of pre-infection structures. The expression of chitin synthase 5-1 (CHS5-1) and TKL family protein kinase was suppressed in CNF-treated leaves at 48 hours after inoculation with P. pachyrhizi (Fig. 3A and Fig. 3B). Furthermore, the expression of metacaspase and NADH dehydrogenase was suppressed in CNF-treated leaves at 24 and 48 hours after inoculation with P. pachyrhizi (Fig. 3C and Fig. 3D). These results suggest that covering soybean leaves with CNF changes gene expression profiles related to formation of pre-infection structures.
Chitin synthases (CHSs) are key enzymes in the biosynthesis of the fungal cell wall structural component, chitin. Since CHS5-1 expression was suppressed in CNF-treated leaves, we next tested the expression profiles of other P. pachyrhizi CHS genes in CNF-treated leaves. Except CHS2-1, all CHS genes transcripts were not significantly suppressed in CNF-treated leaves (Fig. S2B, Fig. S2C, Fig. S2D, Fig. S2E, Fig. 2F, Fig. 2H and Fig. S2H). In addition to CHS5-1, CHS2-1 expression was suppressed in CNF-treated leaves (Fig. S2A). Together, these results suggest that CNF-treatments suppress the expression of CHS5-1 and CHS2-1, resulting in reduced chitin biosynthesis activity in the P. pachyrhizi cell wall.
CNF converts leaf surface properties from hydrophobic to hydrophilic
CNF has amphipathic properties, and thus can convert material surface properties from hydrophobic to hydrophilic, and vice versa (21). To confirm whether CNF-treatments can convert soybean leaf surface properties from hydrophobic to hydrophilic, we decided to quantify the differences in surface hydrophobicity by measuring the contact angle at the interface of a liquid (water) drop with the leaf surface. A greater contact angle (>90°) is indicative of poor wetting or hydrophobicity. Interestingly, significant differences in the contact angle were observed between control and CNF-treated adaxial leaf surfaces (Fig. S3A). The adaxial leaf surface of control leaves exhibited an average contact angle of 128°, whereas CNF-treated leaves showed a dramatic decrease in the contact angle (around 90°), which is indicative of a hydrophilic surface (Fig. 4A). Similarly, significant differences in the contact angle were observed between control and CNF-treated abaxial leaf surfaces (Fig. S3B). The abaxial leaf surface of control leaves exhibited an average contact angle of 127°, whereas CNF-treated leaves showed a dramatic decrease in contact angle (around 70°; Fig. 4B). These results clearly indicate that CNF-treatments can convert leaf surface properties from hydrophobic to hydrophilic.
Discussion
We investigated the potential application of CNF in agriculture, especially disease protection in soybean plants against the rust pathogen, P. pachyrhizi, and found that CNF-treated soybean leaves conferred resistance against P. pachyrhizi (Fig. 1A and Fig. 1B). CNF-treatments can convert the soybean leaf surface properties from hydrophobic to hydrophilic (Fig. 4A and Fig. 4B), resulting in suppression of P. pachyrhizi genes involved in the formation of pre-infection structures, including germ-tubes and appressoria (Fig. 3) associated with reduced appressoria formation (Fig. 2). These results provide new insights into CNF application on P. pachyrhizi disease management strategies.
We demonstrated that CNF-treatments conferred soybean resistance against P. pachyrhizi associated with reduced lesion formation (Fig. 1A and Fig. 1B). The application of chitin nanofibers for plant protection against pathogens has been investigated. Egusa et al. (23) reported that chitin nanofibers effectively reduced fungal and bacterial pathogen infections in Arabidopsis thaliana by activating plant defense responses, including reactive oxygen species (ROS) production and defense-related gene expression. Furthermore, chitin nanofiber treatment can reduce the occurrence of Fusarium wilt disease in tomato plants (24). These results suggest that chitin nanofibers activate plant immunity, resulting in reduced pathogen infection. However, we showed no elicitor activity of CNF based on the GmPR1 defense maker gene expression profiles (Fig. S1). Although there is no similarity to the mechanism by which nanofibers, including cellulose and chitin, function to protect plants against pathogens, both nanofibers will be able to provide eco-friendly disease control strategies in sustainable agriculture.
Formation of pre-infection structures including germ-tubes and appressoria was significantly suppressed in CNF-treated leaves compared to control leaves (Fig. 2). Consistent with our results, Uppalapati et al. (16) reported the reduced formation of pre-infection structures on a M. truncatula irg1 mutant, in which the epicuticular waxes were completely defective and the surface property was changed to hydrophilic. These results indicate that properties such as hydrophobicity are important to form P. pachyrhizi pre-infection structures during early infection stages. The importance of hydrophobicity and/or epicuticular waxes on the formation of germ-tubes and appressoria has also been reported for other fungal pathogens (25–27). Further characterization of the mechanisms by which fungal pathogens recognize plant surface properties and initiate infection behavior will be needed to develop effective and sustainable disease control methods.
CNF-treatments suppressed gene expression related to chitin formation, including CHS2 and CHS5, which are associated with reduced formation of pre-infection structures (Fig. S2, Fig. 2 and Fig. 3). CHS5 is important in cell wall formation in most filamentous fungi (28, 29). Treitschke et al. (30) reported that an Ustilago maydis CHS5 mutant Δmsc1 showed reduced virulence associated with abnormal hyphal morphology. Madrid et al. (31) also demonstrated that CHS5 in Fusarium oxysporum, a causal agent of tomato vascular wilt, has a crucial role in virulence and mediates the tomato protective response. A F. oxysporum CHS5 mutant could not infect tomato, exhibiting abnormal morphologies such as hyphal swelling, due to changes in the cell wall properties (31). These results suggest that CHS5 gene deficiency or mutation causes morphological abnormalities in fungal cell wall formation, leading to virulence suppression. Together, it is tempting to speculate that suppression of P. pachyrhizi CHS5 in CNF-treated leaves may result in changes in cell wall properties of P. pachyrhizi pre-infection structures. Further characterization of CHS5 based on dsRNA-mediated silencing such as spray-induced gene silencing (SIGS) and host-induced gene silencing (HIGS), in conjunction with analysis of P. pachyrhizi cell wall properties on CNF-treated leaves, will be necessary to understand CHS5 molecular function during formation of pre-infection structures.
We demonstrated that CNF-treatments suppressed ASR, one of the most important soybean diseases (Fig. 1A and Fig. 1B) associated with reduced formation of pre-infection structures (Fig. 2A and Fig. 2B). Because numerous rust and filamentous fungal pathogens form pre-infection structures during early infection stages, these results imply that CNF might be an additional disease management tool to prevent crop diseases against these pathogens. However, we tested the ability of CNF to protect plants against an obligate biotrophic pathogen, but not other pathogen types, including hemibiotrophs and necrotrophs. Therefore, further characterization of CNF effects on disease suppression not only against fungal pathogens, but also against bacterial pathogens will be needed.
In summary, CNF-treatments confer resistance against P. pachyrhizi, a causal agent of ASR. Moreover, CNF-treatments can change leaf surface hydrophobicity, resulting in gene suppression related to chitin synthase, which is associated with reduced formation of pre-infection structures including P. pachyrhizi germ-tubes and appressoria (Fig. 5). Since CNF is an abundant and renewable biomass in nature, CNF application for plant protection will provide a new avenue into eco-friendly and sustainable disease management.
Supplementary Figure S1. Expression of soybean defense marker gene GmPR1 in response to CNF
Soybean plants were treated with 0.1% cellulose nanofiber derived from bamboo (BC) and needle-leaved tree (NC). Total RNA was purified at 24 hours after treatment and expression profiles were evaluated using RT-qPCR. Soybean Actin4 (GmAct4) was used as an internal control to normalize gene expression. NS indicates not significant between control and CNF-treatments in a t test.
Supplementary Figure S2. Expression profiles of chitin synthases (CHSs), including CHS2-1 (A), CHS2-2 (B), CHS2-3 (C), CHS3-1 (D), CHS3-2 (E), CHS3-3 (F), CHS4 (G), and CHS5-2 (H) during the early P. pachyrhizi infection stage on the surface of control, leaves covered with 0.1% cellulose nanofiber derived from bamboo (BC) and needle leaf tree (NC)
Soybean plants were drop-inoculated with P. pachyrhizi (2 × 105 spores/ml). Total RNAs including soybean plants and P. pachyrhizi were purified at 24 and 48 hours after inoculation and expression profiles were evaluated using RT-qPCR. Elongation factor and Ubiquitin 5 were used to normalize the samples. Vertical bars indicate the standard error of the means (n = 4). Asterisks indicate a significant difference between control and CNF-treatments in a t test (* p < 0.05, ** p < 0.01).
Supplementary Figure S3. Reduction of contact angle and hydrophobicity on CNF-treated soybean leaves
Contact angles of water droplets on the adaxial (A) and abaxial (B) leaf surface of control, leaves covered with 0.1% cellulose nanofiber derived from bamboo (BC) and needle-leaved tree (NC). Contact angles were evaluated as described in the Methods.