Disease Development and Discovery of Anatomically Resistant Features towards Bacterial Leaf Streak in Rice

Abstract

Bacterial leaf streak (BLS) caused by bacterium Xanthomonas oryzae pv. oryzicola (Xoc) is one of the most prominent rice diseases. BLS causes a significant reduction in paddy yields. However, there are limited studies and a lack of information regarding the mechanisms and cells affected on leaf tissues severed from this disease. Therefore, in this study, sensitive paddy variety IR24 was inoculated against BLS, and the pathogen colonised mesophyll cells and some bundle sheath cells. The infection spreads rapidly towards the base and apex of the leaf, but rather slowly to the left and right sides of the leaf veins. Another experiment was performed to unravel anatomical characteristics in sensitive paddy varieties (TN1, IR24, IR5) and resistant paddy varieties (IR26, Dular, IR36) against BLS. Susceptible paddy varieties have less thick midrib and leaf lamina, a high number of bundle sheath cells at primary vascular tissue (midrib), one layer of sclerenchyma cells at the secondary vein, and more than two metaxylems at the primary vein. Resistant paddy varieties, on the other hand, consist of a relatively thickened midrib and leaf lamina, fewer bundle sheath cells at the primary vascular tissue (midrib), more than one sclerenchyma layers at the secondary vein, and two metaxylems at the primary vein. This study contributes new knowledge in identifying the level of infection in paddy fields, and helps breeders in producing resistant paddies to this disease.

1. Introduction

Oryza sativa L. or paddy rice is the leading staple food in Asia. However, paddy rice yields can fluctuate due to numerous factors such as pests or diseases. Bacterial diseases are significant constraints to rice production in Africa, and several humid tropical and subtropical areas of Asia [1]. Bacterial diseases are most destructive and cause significant yield loss [2]. Among them, the most prevalent diseases are bacterial leaf blight (BLB) and bacterial leaf streak (BLS) caused by Xanthomonas oryzae pv. oryzae (Xoo) and Xanthomonas oryzae pv. oryzicola (Xoc), respectively [3]. BLB and BLS can spread on paddy plants and reduce rice yield [4]. However, both bacteria are too difficult to distinguish morphologically and genetically [4,5,6]. Both Xanthomonas species have 90% similarity by DNA:DNA hybridisation [7].

From 1993 to 1998, a survey was conducted in Zhejiang, China (subtropical) and Luzon, Philippines (tropical) to identify the rice disease and its causal bacteria. A total of 3500 isolates of infected paddy plants conclusively showed that there are 208 pathogenic bacterial isolates screened, and Xanthomonas oryzae pv. oryzicola were the most common bacteria in the two areas [8]. There are several recent studies investigating BLS in Africa, as reported in Senegal, Nigeria, Madagascar, and Mali [9]. Previous research showed the early stage paddy symptoms in Burkina Faso in October 2009, where this disease is commonly spotted on wild rice species (O. longistaminata and O. barthii), cultivated Oryza sativa (TS2, FKR19, and FKR56N), and weeds [9]. In October 2013, Vietnamese researchers found a high prevalence of bacterial leaf diseases in the irrigated rice fields of the Red River delta (western coast of the Gulf of Tonkin) [10]. BLS symptoms were reported in some rice varieties in several states of Malaysia from March 2014 to May 2015, and 99% of all isolates were identical to Xoc strain NCPPB3949 isolated back in 1956 [11]. Several studies attempted to explain the yield effect of the disease on rice production. A report showed that the disease reduced crop yield production losses by up to 32% (under favourable conditions) [12,13].

Xoc is a Gram-negative bacterium with rod-shaped, round-ended, one-end flagella that is a member of the gamma subdivision of the Proteobacteria class [7,13]. This pathogen is prevalent in regions with high humidity and warm temperatures. It can be transmitted by infected seeds from one summer to another summer without being interrupted by the winter season, as the pathogen become inactive during winter [12]. This pathogen was colonised and proliferated in the plant’s apoplast of mesophyll parenchyma cells [14]. One of the recommendations for BLS management is to plant resistant rice varieties against the disease [15]. Thus, understanding the mechanism of plant–microbe interactions is needed to elucidate resistance factors in plants.

Host recognition by a microbe is also an essential aspect of host–pathogen interactions. Therefore, we applied the anatomical method in investigating interactions and identifying the exclusive characteristics of resistant varieties towards BLS. The anatomical approach was used in systematics and taxonomy classification [16,17,18,19]. As reported, numerous characters are used to differentiate plant species or subspecies, such as variation in petioles [20,21]], leaf venation [22], and stipe [23]. For that reason, this study aimes to decipher disease progression and discover potential resistance characteristics between rice cultivars towards Xanthomonas oryzae pv. oryzicola using the anatomical approach.

2. Materials and Methods

2.1. Study Site

The study was conducted in the PC2-Certified Greenhouse (RTPC2) at the Institute of Systems Biology, Universiti Kebangsaan Malaysia (UKM). The soil used for planting was a combination of soil, sand, and organic manure with a ratio of 3:2:1. Seeds of six paddy varieties (TN1, IR24, IR5, IR36, Dular, and IR26) were soaked in water, drained, and germinated for 24 h until their radicals emerged from the seeds. Seeds were then transferred to plastic pots with cotton and constantly kept moisturised before transplanting. After 10 days of sowing, seedlings were transplanted into the pots. All plants were grown under controlled conditions at 28 °C, 12 h day and 12 h night, with watering every 2 days (1–2 cm water depth).

2.2. Variety Selection

IR24 was evaluated against Xanthomonas oryzae pv. oryzicola (Xoc) in a PC2-Certified Greenhouse, under controlled conditions at 28 °C. Then, IR24 and 5 more varieties (IR5, IR26, IR36, Dular and TN1) were selected to investigate the variation in anatomical characters in susceptible and resistant varieties. Plant materials were selected on the basis of the International Rice Research Institute (IRRI). Malaysian Agriculture Research and Development Institute (MARDI), Serdang Selangor provided the plant seeds used in this study.

2.3. Pathogen and Inoculum Preparation

Xanthomonas oryzae pv. oryzicola (Xoc) used in this study was provided by MARDI Serdang, Selangor. The strain was stored in 15% glycerol at −80 °C. Xoc was grown on potato sucrose agar (sucrose 20 g/L, peptone 5 g/L, calcium nitrate 0.5 g, sodium phosphate 0.82 g/L and agar 17 g/L) for 24 h at 28 °C. The bacteria were then resuspended in nutrient broth and diluted to approximately 1 × 10⁸ CFU mL⁻¹ at an optical density of 0.10 and a wavelength of 600 nm using a spectrophotometer. Mechanical inoculation was conducted using a pipette. Each drop contained 1 μL of bacteria suspension; a pipette was used to transfer drops onto the leaves. Bacteria were inoculated at the vegetative stage of plants (Day 30). The experiment was repeated three times.

2.4. Data Collection and Analysis

Lesions induced by Xoc on paddy genotype leaves were measured three days after inoculation using a 300 mm calibrated ruler. Obtained data were analysed using the one-way analysis of variance (ANOVA) t-test in IBM SPSS 26 software. In addition, BLS severity was recorded and analysed continuously for 20 days after inoculation (DAI) (2, 6 10, 13, 17 and 20 DAI) using a progressive 1–9 scale (

Table 1. Disease rating scale and percentage of leaf area covered with bacterial leaf streak.

Disease Rating Scale	Percentage of Leaf Area Covered with Disease
1	No lesions were observed
2	Typical lesions covering 1–5% of leaf area
3	Lesions covering 6–10% of leaf area
4	Lesions covering 11–20% of leaf area
5	Lesions covering 21–30% of leaf area
6	Lesions covering 31–40% of leaf area
7	Lesions covering 41–50% of leaf area
8	Leaf covering 51–75% of leaf area
9	Lesions covering >75% of leaf area or all leaves dead.

) [24].

2.5. Anatomy of Infected Leaves IR24, Sensitive and Resistant Varieties

The cross- and transverse sections were cut by hand with a thickness of 25 µm using a sliding microtome (Leica SM 2000 R). Sections were then stained with safranin and Alcian blue for 5–10 min. Alcian blue and safranin were used to detect the presence of pectins (and other acidic polysaccharides, including mucilages) and to stain lignified tissues such as xylem [25,26]. This was followed by dehydration in a graded ethanol series (50%, 75%, 90% and 100%), and slides were mounted in Eupharal [27]. Anatomical leaf features were observed using an Olympus BX34 DP72 camera and a Canon EOS 700D. Anatomical structures were captured and analysed using Analysis Docu and EOS Utility 2 software. Photographs of leaf cells were studied at four different objective magnifications (4×, 10×, 20×, and 40×). According to Mutka et al. [28], the percentage of the infected area can be calculated on the basis of the following formula to observe disease progression in infected leaves:

(i)

Percentage of infected area = total width of infected area/total width of entire studied area × 100.

(ii)

Percentage of stain absorption (colour absorption) = total width of coloured area (infected)/total width of entire studied area × 100.

Healthy and infected leaves were compared to identify their differences.

3. Results

3.1. Disease Progression in IR24

Paddy leaves were long and flattened for the healthy plant sample (control), which comprised 11 to 12 parallel veins per leaf, and it was 88.9 cm in length and 12.5 cm wide on average. Although lesions started to appear within 2–3 days of infection and spread in parallel to veins, no lateral spread was found. Therefore, BLS severity was recorded and analysed continuously (2, 6 10, 13, 17, and 20 DAI) using a progressive 1–9 scale (Table 1).

Leaves on the second day were assigned Scale 2 with infections spreading only in the leaf area at the inoculation site (Figure 1; yellow arrow), 1–5% of the total leaf area. Leaves on the sixth day were assigned Scale 3, with the infection area representing 6–20% of the total leaf area (Figure 1; green arrow). The infection spread from the inoculation area to the leaves’ left side, then to the apex and base, while the right side was still intact. Leaves on day 10 were assigned Scale 5, showing the infection covering a leaf area of up to 21–40%, and elongated in parallel to the leaf veins. Infection was found at one-third of the leaf apex compared to the middle or leaf base (Figure 1; blue arrow). Leaves on day 13 were assigned Scale 7, as the infection had already spread to two-thirds of the leaves. The estimated disease area was 41–50% (Figure 1; grey arrow). Leaves on day 17 were assigned Scale 8, with the infection covering 51–75% of leaves (Figure 1; orange arrow). The infected area (shown as the yellow region in Figure 1) dominated the leaf area and decreased the healthy leaf area (shown as the green region in Figure 1). Leaves on day 20 were assigned Scale 9, where the infection area covered >75% of the leaf area. The left and right sides of leaves began to roll, affected by dead cells (Figure 1, red arrow). Leaves infected by bacteria turned white and grey, possibly due to the growth of saprophyte fungi that developed into bacterial mite disease. A similar result was reported by Rudolph [29]. Figure 1 shows the infection progress with the disease severity scale in IR24.

3.2. Anatomical Analysis of Infected IR24

3.2.1. Mesophyll Cell Analysis

In healthy leaves (Stage 1), three layers of mesophyll cells were observed, with ten mesophyll cells arranged horizontally between two veins (Figure 2). The total number of mesophyll cells was ∼30 cells and in a normal state. Mesophyll cells at Stage 2 began to lose shape; more than half of the mesophyll cells collapsed and could not be identified. Normal mesophyll cells declined to ∼30%. The ratio of length and width of the area between veins was 6:1. In Stage 3, the infected area lost its original state, and no mesophyll cells could be seen in the area, even though the epidermis layers were still intact (Figure 3, Stage 3). At this stage, mesophyll cells were unrecognisable and fully degenerated, and the ratio of length to width was 7:1.

3.2.2. Comparison of Healthy and Infected Midrib of IR24

Mesophyll cells were entirely affected after ten days of infection. Figure 3A,B show healthy and infected midrib, respectively. Vascular tissue, especially the metaxylem, was multiplied, and even mesophyll cells were not recognisable. In contrast, companion cells in phloem tissue were destroyed, with sieve tubes still being identifiable (Figure 3B). The sheath of several bundles could not be identified, and bulliform cells were also affected. Sclerenchyma cells could still be identified despite not absorbing the stain. A translucent area in the infected midrib showed that chlorophyll was totally depleted.

Table 2. Differences in structure and stain absorption between healthy and infected by Xoc IR24 midrib.

Cells and Tissue	Healthy IR24 (A)	Infected IR24 (B)
Mesophyll	Cells are in a healthy state and can be identified; stain absorption is ∼100%	Mesophyll cells degenerate 100%, unidentified; there were translucent areas showing chlorophyll degraded and affected, stain absorption decreased ∼90%
Bundle sheath	Stain absorption ∼100% and in a healthy state	Stain absorption was ∼30%
Xylem	Metaxylem-2	Metaxylem-3
Phloem	Stain absorption ∼100%	Stain absorption ∼60%
Sclerenchyma	Cell structure in healthy condition, stain absorption ∼100%	Cell structure is still intact; stain absorption decreased ∼100%
Bulliform	100% cells can be identified	90% of cells can’t be identified

shows differences in structure and stain absorption between uninfected (healthy) IR24 and infected midrib of IR24.

3.3. Comparative Anatomical Features of Resistant and Sensitive Paddy Varieties

3.3.1. Midrib

Resistant paddy varieties IR36, Dular, and IR26 had midrib thickness of ∼328.92, ∼321.79, and ∼263.92 µm, respectively. Results showed that there was a difference in midrib thickness between sensitive and resistant paddy groups. Midrib was thicker in the resistant paddy group, and the TN1 variety had the thickest midrib of ∼409.09 µm, while the two other sensitive paddy varieties, IR24 and IR5, had ∼248.44 and ∼205.41 µm, respectively. In addition, there were differences between the susceptible and resistance group varieties in the number of bundle sheaths that surround primary vascular bundles. The sensitive group had more cell layers (TN1, ∼26 cells; IR24, ∼29 cells; and IR5, ∼23 cells) compared to the number of bundle sheaths present in the resistant group of IR36, Dular, and IR26, which had ∼22 cells each.

Table 3. Variety in number of bundle sheath cells in resistant and sensitive paddy varieties.

V	Primary Vascular Bundles in Midrib Cross-Sections	A	V	Primary Vascular Bundles in Midrib Cross-Sections	A
Sensitive Paddy Varieties			Resistant Paddy Varieties
TN 1		∼26	IR36		∼22
IR24		∼29	Dular		∼22
IR5		∼23	IR26		∼22

Note: V: varieties; A: number of bundle sheath cells; yellow arrow: bundle sheath cells; mx: metaxylem; and p: phloem.

shows the number of bundle sheath cells in the midrib’s primary vascular bundles of the resistant and sensitive varieties. All varieties found in this study had the same number of lacunae (four) and a presence of trichomes, including papillae. The presence of a parenchyma cell extension and the number of sclerenchyma cell layers in the midrib did not show significant differences.

3.3.2. Lamina

Laminar thickness in TN1 was ∼78.33, IR24 ∼59.67, and IR5 ∼63.83 µm; in the sensitive paddy group, it was ∼72.22 µm for IR36, ∼69.79 µm for Dular, and ∼66.57 µm for IR26. There were several primary veins in the leaf lamina, and more than two metaxylems were present, with more abundance in susceptible paddy varieties. The metaxylem was exceeded in TN1, IR24, and IR5 (63.6%, 63.6%, and 54.5%); in the IR36, Dular, and IR26 resistance groups, it was 33.33%, 42.86% and 40%, respectively (

Table 4. Variation in percentage of metaxylem > 2 and ratio of metaxylem > 2 to metaxylem = 2 (primary vein) of sensitive and resistant paddy varieties.

V	I	A	B	V	I	A	B
Sensitive Paddy Varieties				Resistant Paddy Varieties
TN1		63.6%	4:2	IR36		33.33%	2:4
IR24		63.6%	4:2	Dular		42.86%	3:4
IR5		54.5%	4:3	IR26		40%	2:3

Note: V: variety; I: cross-section; A: percentage of metaxylem > 2; and B: ratio metaxylem > 2 to metaxylem =2.

).

The number of sclerenchyma cell layers was different in the resistant and sensitive paddy varieties, especially in secondary veins. Secondary veins were located within primary veins. There were four or five secondary veins attached to the interval of one primary vein. One layer of sclerenchyma cells was found in susceptible varieties TN1, IR24, and IR5. The ratio of secondary veins having one sclerenchyma layer to secondary veins having more than one sclerenchyma layer for TN1, IR24 and IR5 was 3:2, while for IR36, Dular and IR26, it was 1:4, 2:3, and 2:3 (

Table 5. Variation in number of sclerenchyma cells layers in sensitive and resistant paddy varieties.

V	I	A	B	C	V	I	A	B	C
Sensitive Paddy Varieties					Resistant Paddy Varieties
TN1		∼1–2	∼40%	3:2	IR36		∼1–2	∼80%	1:4
IR24		∼1–2	∼40%	3:2	Dular		∼1–2	∼60%	2:3
IR5		∼1–2	∼40%	3:2	IR26		∼1–3	∼60%	2:3

Note: V: variety; I: cross-section of secondary vein; A: layers of sclerenchyma cells; B: percentage of sclerenchyma layers > 1; C: ratio of sclerenchyma layers = 1 to sclerenchyma layers > 1; and white arrows: represent sclerenchyma layers.

).

All studied varieties had trichomes present on the abaxial and adaxial leaf epidermis. However, there were a few inconsistent features in the sensitive and resistant paddy varieties, such as the number of bundle sheath cells at secondary veins, the number of sclerenchyma cell layers at primary veins, the distance between bundle sheath cells, and the distance between vascular bundles. Statistical analysis revealed that laminar thickness, the number of metaxylems, the number of sclerenchyma layers, and the number of bundle sheath cells were significantly different at p $\le$0.05 between susceptible and resistant varieties. On the other hand, midrib thickness did not exhibit any significant difference at p > 0.05.

Table 6. Variation in anatomical characteristics between sensitive and resistant paddy varieties towards bacterial leaf streak.

	Sensitive Paddy Varieties			Resistant Paddy Varieties
	TN1	IR24	IR5	IR36	Dular	IR26
Midrib thickness	∼409.09 µm	∼248.44 µm	∼205.41 µm	∼328.92 µm	∼321.79 µm	∼263.92 µm
Laminar thickness	∼78.33 µm	∼59.67 µm	∼63.83 µm	∼72.22 µm	∼69.79 µm	∼66.57 µm
Number of metaxylems	∼3	∼4	∼4	∼2	∼2	∼2
Percentage of metaxylems	63.60%	63.60%	54.50%	33.33%	42.86%	40%
Number sclerenchyma layers	∼1–2	∼1–2	∼1–2	∼1–2	∼1–2	∼1–3
Percentage of sclerenchyma layers	∼40%	∼40%	∼40%	∼80%	∼60%	∼60%

shows the variation in anatomical features between BLS-resistant and -sensitive paddy varieties.

4. Discussion

4.1. Disease Progression in IR24

Xoc penetrates the leaf through wounded areas or natural openings such as stomata [13]. Once bacteria actively or passively reach the substomatal space, they multiply before symptoms start to appear. Most Xoo or Xoc strains that successfully enter mesophyll cells can cause waterborne infections (water-soaked lesions) within days 3–7 of susceptible varieties; the leaf turns yellow and is surrounded by brownish spots [30,31]. After a few days, lesions are elongated in parallel to the leaf vein. The infection process continues along the vein and spreads slowly to the left and right sides of the leaf [1]. On day 20, the infection worsens, with dead cells visible on the leaves.

As a biotrophic pathogen, Xanthomonas derives its nutrients from host cells; therefore, BLS disease takes time to appear on the leaf, and disease development appears to be slower compared to other diseases such as blast, which only takes nine days to reach scale 9 [24,32].

Safranin and Alcian blue staining revealed mesophyll cells between adaxial and abaxial epidermal layers. Since mesophyll cells are attached to cell wall fixtures, Xoc needs to break down cell walls attached to lobes and then penetrate mesophyll cells. Although Xoo and Xoc can multiply in the space between mesophyll cells, only Xoc can break cell wall fixtures between mesophyll cells to spread within the area between cells and thereby cause infection [30].

The cell wall is a relatively difficult barrier for plant pathogens to penetrate. Therefore, the ability to break down the wall barrier is a crucial feature of plant pathogenic bacteria. The degradation of cell walls produces a signal of infection in plants as a damage-associated molecular pattern (DAMP) process that results in the activation of a potent innate immune response [33,34,35]

Some cells, however, do not contain chloroplasts and can still become infected with this pathogen. These could be triggered due to the plants’ hypersensitive response (HR) that occurs when pathogens penetrate the plant tissues. The HR activates responses that suppress the resistance response. Nevertheless, this can only happen if a pathogen is able to suppress the plant’s immune system [31]. HR is a deliberate plant cell suicide system at the site of infection that prevents pathogens from continuously acquiring water and nutrients at any infected cells to spare other healthy and uninfected plant tissues. Different to a typical plant defence system, HR is usually more specific to the pathogen and only occurs when proteins in plant cells detect the presence of specific disease-causing effector molecules carried by the pathogen into the host.

This response results in water stress, which explains the deterioration of bulliform cells in the infected midrib found in the adaxial epidermis of leaves and in vascular bundles (xylem and phloem) [36,37]. Bulliform cells appear intercostally as long, broad stripes. In paddies, they are arranged in groups of multiple cells, with clusters not being next to one another [38]. Generally, in lack of water, bulliform cells lose their vigour (turgor) and induce adaxial leaf curling due to a change in osmotic pressure. When there is sufficient water, bulliform cells reabsorb water and become turgid again, resulting in leaves reopening [39,40].

On the other hand, the metaxylem continues to divide even when the plant is infected. To date, there is still no study on the xylem division response to pathogenic infections, which shows any infiltration to the vascular system. Xoc tends to colonise parenchyma rather than vascular tissue, which is why xylem and phloem do not show significant deterioration [30].

4.2. Comparative Anatomical Characteristics of Resistant and Sensitive Paddy Varieties

Information on resistance genes against BLS found in rice remains scarce [41]. For instance, gene X01 confers resistance only towards African strains of Xoc and not Asian strains [42]. Hence, the discovery of potential characters in resistant varieties through anatomical methods is necessary to add value to information on germplasm screening towards bacterial leaf streak.

This study showed that resistant paddy varieties have thicker leaf lamina, larger-lobed mesophyll cells, and wider cell wall surfaces than those of susceptible paddy varieties. In addition, differences in metaxylem number and sclerenchyma cell thickness were also reported. Lignin is the major component of the metaxylem and sclerenchyma cells. The lignin layer beneath the epidermis provides a better resistance effect than that of the lignin present in vascular tissue. Lignin under the epidermal layer is the first barrier in preventing the entry of pathogens into the leaves. Lignification was proposed as an essential mechanism in wheat resistance to diseases caused by various fungal pathogens when no phytoalexins are found in wheat Triticum aestivum L. [43]. In wheat, lignin acts as a defensive response during infection. For example, S-enriched lignin accumulates during hypersensitive reactions by wheat due to Puccinia graminis infection [44], and is synthesised in wheat epidermal cells infected with Fusarium proliferatum [45]. Lignin and phenolic polymers (which look most like lignin) accumulate in plant cell walls in response to biotic and abiotic stresses, and invasions to their structures [46].

Sensitive paddy varieties have more bundle sheath cells compared to resistant paddy varieties. However, to our knowledge, none of these previous studies explains how the bundle sheath of C₃ plants is preferentially related to sensitiveness to disease.

5. Conclusions

Our study focused on the mechanisms and cells impacted by Xoc’s BLS disease on leaf tissues of the IR24 paddy variety. Pathogens successfully colonised mesophyll cells and a certain bundle of sheath cells. The infection rapidly spread perpendicular to the base and apex of the leaf, but rather slowly towards both sides of the leaf veins. Anatomical characteristics between susceptible (TN1, IR25, IR5) and resistant (IR26, Dular, IR36) paddy varieties against BLS were compared. Resistant and sensitive paddy varieties can be distinguished by midrib thickness characteristics and the number of bundle sheath cells in the main vascular bundle. While the difference between resistant and susceptible paddy varieties in primary veins is determined by the leaf laminar thickness and the number of metaxylems, the difference between resistant and susceptible paddy varieties in secondary veins is determined by the number of sclerenchyma layers. Understanding the level of pathogen infection in paddy rice assists researchers in solving this issue and breeders in developing paddy varieties that are resistant to the disease.