Quantitative Trait Locus Analysis of Microscopic Phenotypic Characteristic Data Obtained Using Optical Coherence Tomography Imaging of Rice Bacterial Leaf Blight Infection in the Field

: Rapid climate change has increased the incidence of various pests and diseases, and these threaten global food security. In particular, BLB (bacterial leaf blight) is caused by Xoo ( Xanthomonas oryzae pv. oryzae ) and its main characteristic is that the rice suddenly dries and withers. Recently, omics have been effectively used in agriculture. In particular, it is a key technology that can accurately diagnose diseases in the ﬁeld. Until now, QTL (quantitative trait loci) mapping has been analyzed using only subjective phenotypic data by experts. However, in this study, diseases were accurately diagnosed using OCT (optical coherence tomography), and QTL mapping was performed using leaf thickness and leaf angles after Xoo inoculation. After Xoo inoculation of a 120 Cheongcheong/Nagdong double haploid (CNDH) population, QTL mapping was performed using the changing leaf angle, and OsWRKY34q1 was detected in RM811-RM14323 of chromosome 1. OsWRKY34q1 always had a higher expression level in the BLB-resistant population than in the susceptible population after Xoo inoculation. OsWRKY34q1 belongs to the WRKY family of genes. OsWRKY34q1 could be effectively used to develop BLB-resistant rice varieties in response to the current era of unpredictable climate change.


Introduction
Rice is currently the world's most cultivated and stable food for over 3.5 billion people worldwide. In particular, current rice production and consumption is the highest in Asia, supplying up to 50% or more of Asians' dietary calories [1]. The rapidly increasing population along with agricultural problems caused by global warming and increases in pesticide-resistant pests and pathogens have threatened global food security; thus, improving rice yield and breeding programs are necessary. Rice research has focused on identifying useful crop genes using molecular genetic tools and cultivating new varieties via molecular breeding strategies. Bacterial leaf blight (BLB), a disease caused by Xanthomonas oryzae pv. oryzae (Xoo), is a major problem in rice cultivation areas in Southeast Asia, including Korea [2]. The crop damage caused by Xoo varies depending on the region, weather, variety, cultivation method, time of onset, and outbreak severity and can significantly reduce the quantity and quality of rice yields. In severe cases, more than

K3 Strain Culture Media, Inoculation, and Infection Assessment
K3 was used as the pathogen inoculated with white leaf blight. The strain was stored at −80 • C using 10% sterile glycerol before use, and recovered by culturing at −80 • C for 72 h at 27 • C in peptone sucrose agar (PSA) medium before inoculation into rice. (Akhtar et al., 2008). Vitality was restored, and the pathogen was incubated in a sterilized medium at 28 • C at 130 RPM, in darkness using a solution of 8 g each of bacto nutrient broth powder in 1 L of distilled water (DW) using an autoclave. Leaf clipping was performed following the method described by Kauffman et al. (1973) [21]. The rice seedlings were inoculated 40 d after transplantation. The middle 10 plants in each row were inoculated with K3, and five leaves on each plant were inoculated. A pair of scissors was dipped into the K3 suspension and then used to cut 3-4 cm from the leaf tip. Fourteen days after the pathogenic inoculation, the infected leaf was measured using OCT. Data were generated to evaluate BLB defense-related genes using QTL analysis. According to the Rural Development Administration's Agricultural Science and Technology Research and Analysis Standard (2003), the lesion length was classified as susceptibility, and plants with a lesion length of less than 5.0 cm were classified as susceptible.

Optical Coherence Tomography (OCT) Analysis of Infected Rice Leaves
An SS-OCT system was used to scan the top, middle, and bottom of the front and back sides of the rice leaf after Xoo inoculation, and real-time scan images were obtained. Supplemental Figure S1 shows a schematic design of the SS-OCT system (OCS1310V1, Thorlabs, Inc., Newton, NJ, USA) that was utilized for rice leaf imaging. A swept-source laser engine with a center wavelength of 1310 nm and full width half maximum bandwidth of >97 nm (−10 dB cut-off point) was used in this commercially available system. The source has an average laser output power of >20 mW and a 100 kHz axial scan rate. The lateral and axial resolutions of the system are 25 µm and <16/12 µm (air/water), respectively. The sensitivity is 111 dB at zero optical-path length difference. A detailed explanation of the system specifications can be found in previous literature [22].
OCT technology was used to scan and measure the opening and closing angles of the rice leaf surface to assess the rice leaf tissue damage. We measured and analyzed the leaf thickness before and after infection and the number of opening and closing angles in the OCT scan images of all infected leaf surfaces in the CNDH population. We created distribution diagrams of the leaf thickness and the derivative distribution of the leaf opening and closing angles. These data were then used as QTL microscopic quantitative trait data in the corresponding QTL analysis [17]. Rice infection with BLB will inevitably lead to necrosis and shrinkage of the internal leaf tissues. As a result, the efficiency of rice photosynthesis and nutrient transport pipelines decline and eventually fail completely.
Rice photosynthesis mainly depends on chlorophyll levels and the opening and closing angles of the rice leaf surface. Therefore, we used OCT technology to scan and measure the opening and closing angles of the rice leaf surface to assess the rice leaf tissue damage. We observed significant differences in the thickness and opening and closing angles of the inoculated and the uninoculated rice leaves.

Quantitative Trait Loci (QTL) Analysis
QTL mapping of BLB-resistant candidate regions was performed using Window QTL cartographer 2.5 (Zeng, 1994). The data for QTL mapping included the angle of the leaf analyzed using OCT technology. For the genetic map of the CNDH population, a genetic map with an average distance between markers of 10.6 cM was constructed using 222 SSR (Simple Sequence Repeats Markers) (Lander et al., 1987). The physical position of the SSR marker on the chromosome was analyzed using RAP-DB (The Rice Annotation Project Database). For OCT data analysis values, Composite Interval Mapping (CIM) was used in the Kosambi function of Window QTL cartographer 2.5. Also, the accuracy of QTL was improved by detecting only regions with a LOD score of 3.0 or higher among the mapped QTL regions (Zeng, 1994). The identified QTLs were named using the method adapted by McCough et al. (1997) [23].

Target Gene Selection and Physical Mapping
BLB resistance candidate genes were screened in the regions detected through QTL mapping. SSR marker information present in the QTL mapped region was analyzed using the Rice Annotation Project Database (RAP-DB) and Rice Expression Profile Database (Rice X Pro). All ORFs (open reading frames) existing in the detected region were searched using the analyzed SSR marker. Candidate genes related to BLB resistance were filtered out of ORFs present in the SSR marker region using the National Center for Biotechnology Information (NCBI) database.

RNA Extraction
RNA extraction from rice leaves was performed using RNeasy plant mini kits (QIA-GEN, Hilden, Germany), and instructions were followed through the handbook provided with the kits. After inoculating the leaves with pathogens, the collected rice leaves were sampled at various time points (0, 1,2,4,8,16,24,48, and 72 h) for RNA extraction. The samples were rapidly cooled using liquid nitrogen to minimize RNA denaturation. The samples collected at each time point were immediately placed in liquid nitrogen and then manually ground using a mortar and pestle. According to the manufacturer's instructions, the rice powder was suspended in 450 µL Buffer RLT with β-mercaptoethanol by vortexing. The lysate was transferred to a QIAshredder spin column, centrifuged at 13,000 rpm for 1 min, and then transferred into a new tube. Next, 0.5 volume of (96-100%) ethanol was added. The mixture was moved into an RNeasy spin column (pink) and centrifuged for 15 s at 10,000 rpm to allow RNA binding within the spin column, and then the residue was discarded. The column was washed by adding 700 µL Buffer RW1, and centrifuged for 1 min at 10,000 rpm. Finally, 500 µL Buffer RPE was used twice to remove any residual liquid in the spin column, followed by centrifugation at 10,000 rpm for 1 and 2 min, respectively. RNA was eluted by adding 40 µL RNase-free water into a new collection tube up to a final volume of 1.5 mL, and then the mixture was centrifuged for 2 min at 13,000 rpm.

Quantitative RT-PCR Analyses
For quantitative RT-PCR, the concentration of RNA extracted according to the manufacturer's instructions of the RNeasy plant mini kit (QIAGEN, Germany) was analyzed using a Nano Drop 2000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA). Quantitative RT-PCR was performed using Eco Real-Time PCR system (Illumina, Inc., San Diego, CA, USA). Total RNA 1 µg, 2X qPCRBIO SyGreen (www.pcrbio.com, London, UK), and the specific primers for the examined gene expression 10 pmol and DW was used to make a final volume of 20 µL. WRKY transcription factor 34 gene sequence were used for design of primer set (forward 5 -ATGGCGGCGGCGATGATTCTC-3 , reverse 5 -TCA GGCATTGCAAGTTCGAATCC-3 ). Among the CNDH populations, WRKY transcription factor 34 expression levels after Xoo K3 inoculation were compared among BLB resistance lines (Nagdong, CNDH23, CNDH100, CNDH105, CNDH112) and susceptible lines (Cheongcheong, CNDH49, CNDH99, CNDH106, CNDH115). The chorismate mutaseoverexpressing (OxCM) line was provided by Jan et al. (2020); OxCM is BLB resistant. The relative expression levels of WRKY transcription factor 34 were analyzed and compared among OxCM and CNDH populations. Inoculation by the K3 strain of Xoo was conducted in the field, and after inoculation, the phenotype was checked.

Results
Rice infection with BLB will inevitably lead to necrosis and shrinkage of the internal leaf tissues. As a result, the efficiency of rice photosynthesis and nutrient transport pipelines decline and eventually fail. Rice photosynthesis mainly depends on chlorophyll levels and the opening and closing angles of the rice leaf surface. The opening and closing angles of the rice leaf surface were scanned and measured through OCT to assess the rice leaf tissue damage. We observed significant differences in the thickness and opening and closing angles of the inoculated and the uninoculated rice leaves. Figure 1 shows the 2D crosssectional OCT images of rice leaf. OCT was used 40 days after inoculation with Xoo, which induces BLB, in the 120 CNDH population. The adaxial and abaxial leaves were divided into three parts, top, middle, and lower, and phenotypes such as leaf thickness and angle were analyzed (Table 1). Under normal conditions without Xoo inoculation, the thicknesses of the adaxial top, middle, and lower cheongcheong were 1.97 ± 0.83 mm, 2.02 ± 0.81 mm, and 2.10 ± 0.93 mm, respectively. After Xoo inoculation, the thicknesses of the top, middle, and lower adaxial of Cheongcheong were 1.10 ± 0.81 mm, 1.14 ± 0.90 mm, and 1.13 ± 0.86 mm, respectively. In Nagdong, the thicknesses of the adaxial top, middle, and lower parts of the leaves were 1.40 ± 0.68 mm, 1.20 ± 0.69 mm, and 1.17 ± 0.92 mm, respectively, under normal conditions. However, after Xoo inoculation, the thickness of the adaxial top, middle, and lower parts of the leaves decreased to 0.68 ± 0.74 mm, 0.69 ± 0.92 mm, and 0.64 ± 0.67 mm, respectively. In the CNDH population, the thicknesses of the adaxial top, middle, and lower parts of the leaves were 0.93 ± 0.88 mm, 0.93 ± 0.88 mm, and 0.95 ± 0.90 mm, respectively, under normal conditions. After Xoo inoculation, the adaxial top, middle, and lower parts of the leaves thicknesses of the CNDH population were 0.48 ± 0.45 mm, 0.47 ± 0.45 mm, and 0.49 ± 0.46 mm, respectively. Under normal conditions of Cheongcheong, the thicknesses of abaxial top, middle, and lower parts of the leaves were 2.08 ± 0.79 mm, 2.08 ± 0.81 mm, and 2.06 ± 0.89 mm, respectively, and when Xoo was inoculated, they were 1.31 ± 0.81 mm, 1.50 ± 0.91 mm, and 1.16 ± 0.73 mm, respectively. In Nagdong, abaxial top, middle, and lower parts of the leaves thicknesses were 1.76 ± 0.79 mm, 1.38 ± 0.62 mm, 1.44 ± 0.91 mm, respectively, under normal conditions, and 0.60 ± 0.91 mm, 0.63 ± 0.84 mm, and 0.71 ± 0.69 mm when Xoo was inoculated, respectively. In the CNDH population, the abaxial top, middle, and lower parts of the leaves thicknesses were 1.04 ± 0.98 mm, 1.01 ± 0.95 mm, and 0.97 ± 0.91 mm, respectively, under normal conditions, and 0.49 ± 0.47 mm, 0.51 ± 0.50 mm, and 0.49 ± 0.46 mm when Xoo was inoculated, respectively. The angles of the top, middle, and lower cheongcheong leaves in normal conditions were 72.83 ± 0.92 • , 71.90 ± 0.82 • , and 106.98 ± 0.74 • , respectively, and when inoculated with Xoo they were 108.06 ± 0.79 • , 105.64 ± 0.81 • , and 78.13 ± 0.85 • , respectively. The leaf top, middle, and lower parts of the leaves angles of Nagdong were 139.0 ± 0.80 • , 158. 13  and abaxial leaf thickness and leaf angle were analyzed, they were all continuous variations and normally distributed ( Figure 2     When the 120 CNDH population was QTL mapped on the basis of adaxial and abaxial thickness and angle using OCT, QTLs with the LOD score of 3.0 or higher were detected on chromosomes 1, 6, 8, and 12 ( Figure 3). The adaxial of BLB-infected leaves was analyzed. The qio-1t-8 was detected when using the adaxial thickness data at the top position of the adaxial. The qio-1t-8 is a region detected with the LOD score of 3.52 and an explainable phenotypic variation in RM6999-RM22334 of chromosome 8 of 36%. Also, qio-1t-8 was derived from the allele of Nagdong. When the thickness data were used at the middle position of the adaxial, qio-2t-8 was detected. The qio-2t-8 could explain the phenotypic variation with the LOD score of 4.81, 39% in RM6999-RM22334 of chromosome 8, and was derived from the allele of Nagdong. At the lower position of the adaxial, qio-3t-6 was detected. The qio-3t-6 was located in RM20158-RM217 of chromosome 6. And the LOD score of RM20158-RM217 was 2.96, 31%, which could explain the phenotypic variation, and was derived from the allele of Cheongcheong. QTLs were analyzed by dividing the abaxial thickness of BLB-infected leaves into three categories: top, middle, and lower. When QTLs of thickness were analyzed on the abaxial top, qio-1bt-1 was detected in RM8111-RM14323 of chromosome 1 and qio-1bt-8 was detected in RM22499-RM22334 of chromosome 8. The LOD score of qio-1bt-1 was 34.85, and 93% of the explainable phenotypic variation was derived from the allele of Cheongcheong. The LOD score was qio-1bt-8 was 3.86 and the explainable phenotypic variation of 36%, and it was derived from the allele of Nagdong. In the abaxial middle, qio-2bt-8 was detected. The qio-2bt-8 was detected with an LOD score of 3.48 in RM6999-RM23314 of chromosome 8, which could explain the phenotypic variation of 31%, and was derived from the allele of Nagdong. In the abaxial lower, qio-3bt-8 was detected in RM6999-RM23314 of chromosome 8. The qio-3bt-8 had the LOD score of 5.09 and an explainable phenotypic variation of 37%, derived from the allele of Nagdong. The angles of BLB-infected leaves were analyzed. Leaf angle data of the analyzed 120 CNDH population were used for QTLs mapping. When weaving leaf angle, leaves were analyzed by dividing them into three areas: top, middle, and lower. When using angle data of the top region of BLB-infected leaves, qio-1ba-8 was detected with the LOD score of 4.58 in RM22499-RM22334 of chromosome 8. The qio-1ba-8 was able to explain the phenotypic variation of 36%, and was derived from the allele of Nagdong. In the middle region, qio-2ba-8 and qio-2ba-12 were detected in RM22499-RM22334 of chromosome 8 and RM12-RM247 of chromosome 12, respectively. The LOD score of qio-2ba-8 was 3.39, 35%, which When the 120 CNDH population was QTL mapped on the basis of adaxial and abaxial thickness and angle using OCT, QTLs with the LOD score of 3.0 or higher were detected on chromosomes 1, 6, 8, and 12 ( Figure 3). The adaxial of BLB-infected leaves was analyzed. The qio-1t-8 was detected when using the adaxial thickness data at the top position of the adaxial. The qio-1t-8 is a region detected with the LOD score of 3.52 and an explainable phenotypic variation in RM6999-RM22334 of chromosome 8 of 36%. Also, qio-1t-8 was derived from the allele of Nagdong. When the thickness data were used at the middle position of the adaxial, qio-2t-8 was detected. The qio-2t-8 could explain the phenotypic variation with the LOD score of 4.81, 39% in RM6999-RM22334 of chromosome 8, and was derived from the allele of Nagdong. At the lower position of the adaxial, qio-3t-6 was detected. The qio-3t-6 was located in RM20158-RM217 of chromosome 6. And the LOD score of RM20158-RM217 was 2.96, 31%, which could explain the phenotypic variation, and was derived from the allele of Cheongcheong. QTLs were analyzed by dividing the abaxial thickness of BLB-infected leaves into three categories: top, middle, and lower. When QTLs of thickness were analyzed on the abaxial top, qio-1bt-1 was detected in RM8111-RM14323 of chromosome 1 and qio-1bt-8 was detected in RM22499-RM22334 of chromosome 8. The LOD score of qio-1bt-1 was 34.85, and 93% of the explainable phenotypic variation was derived from the allele of Cheongcheong. The LOD score was qio-1bt-8 was 3.86 and the explainable phenotypic variation of 36%, and it was derived from the allele of Nagdong. In the abaxial middle, qio-2bt-8 was detected. The qio-2bt-8 was detected with an LOD score of 3.48 in RM6999-RM23314 of chromosome 8, which could explain the phenotypic variation of 31%, and was derived from the allele of Nagdong. In the abaxial lower, qio-3bt-8 was detected in RM6999-RM23314 of chromosome 8. The qio-3bt-8 had the LOD score of 5.09 and an explainable phenotypic variation of 37%, derived from the allele of Nagdong. The angles of BLB-infected leaves were analyzed. Leaf angle data of the analyzed 120 CNDH population were used for QTLs mapping. When weaving leaf angle, leaves were analyzed by dividing them into three areas: top, middle, and lower. When using angle data of the top region of BLB-infected leaves, qio-1ba-8 was detected with the LOD score of 4.58 in RM22499-RM22334 of chromosome 8. The qio-1ba-8 was able to explain the phenotypic variation of 36%, and was derived from the allele of Nagdong. In the middle region, qio-2ba-8 and qio-2ba-12 were detected in RM22499-RM22334 of chromosome 8 and RM12-RM247 of chromosome 12, respectively. The LOD score of qio-2ba-8 was 3.39, 35%, which could explain the phenotypic variation, and was derived from the allele of Nagdong. The LOD score of qio-2ba-12 was 59.32, and the explainable phenotypic variation was 97%, derived from the allele of Cheongcheong. In the lower, qio-3ba-8 was detected. The qio-3ba-8 had the LOD score of 3.66 in RM6999-RM22334 of chromosome 8, 35% of the explainable phenotypic variation, and was derived from a Nagdong allele (Table 2).
Agronomy 2021, 11, x FOR PEER REVIEW 8 of 15 could explain the phenotypic variation, and was derived from the allele of Nagdong. The LOD score of qio-2ba-12 was 59.32, and the explainable phenotypic variation was 97%, derived from the allele of Cheongcheong. In the lower, qio-3ba-8 was detected. The qio-3ba-8 had the LOD score of 3.66 in RM6999-RM22334 of chromosome 8, 35% of the explainable phenotypic variation, and was derived from a Nagdong allele (Table 2).    A physical map was constructed focusing on regions detected with an LOD score of 3.0 or higher when QTL mapping was performed using the results of analyzing the thickness and angle of BLB-infected leaves by dividing them into abaxial and adaxial ( Figure 4). Various open reading frames (ORFs) related to plant defense were screened in RM811-RM14323 of chromosome 1, RM22499-RM23314 of chromosome 8, and RM12-RM247 region of chromosome 12. 3 WRKY family genes, 5 plant defense genes, 2 hormone signaling genes, and 3 secondary metabolites genes were searched for the regions repeatedly detected with an LOD score of 3.0 or higher. In the WRKY family genes, genes related to WRKY transcription factor were detected such as Similar to WRKY transcription factor 34 gene, Similar to WRKY 1 gene, and Similar to WRKY transcription factor 10 gene. In the plant defense genes group, the Similar to Resistance gene analog PIC23 gene, Similar to Resistance protein candidate gene, Glyoxalase/bleomycin resistance protein/dioxygenase gene, Disease resistance protein family gene, Ferredoxin I gene, and chloroplast precursor gene were searched. Zinc finger gene and Similar to Fertility restorer gene were searched in the group related to hormone signaling. In the secondary metabolites group, Similar to Cytochrome P450 71C1 gene, Similar to Cytochrome P450 71C4 gene, and Cytochrome P450 family gene were searched (Table 3). Interval markers are those within the significance threshold on each border of the quantitative trait loci (QTL) range. y Positive values of the additive effect indicate that alleles from Cheongcheong are in the direction of increasing the traits. x The proportion of evaluated phenotype variation attributable to a particular QTL was estimated by the coefficient of determination (R 2 ). w Increased allele is the source of the allele causing an increase in the measured trait.
A physical map was constructed focusing on regions detected with an LOD score of 3.0 or higher when QTL mapping was performed using the results of analyzing the thickness and angle of BLB-infected leaves by dividing them into abaxial and adaxial ( Figure 4). Various open reading frames (ORFs) related to plant defense were screened in RM811-RM14323 of chromosome 1, RM22499-RM23314 of chromosome 8, and RM12-RM247 region of chromosome 12. 3 WRKY family genes, 5 plant defense genes, 2 hormone signaling genes, and 3 secondary metabolites genes were searched for the regions repeatedly detected with an LOD score of 3.0 or higher. In the WRKY family genes, genes related to WRKY transcription factor were detected such as Similar to WRKY transcription factor 34 gene, Similar to WRKY 1 gene, and Similar to WRKY transcription factor 10 gene. In the plant defense genes group, the Similar to Resistance gene analog PIC23 gene, Similar to Resistance protein candidate gene, Glyoxalase/bleomycin resistance protein/dioxygenase gene, Disease resistance protein family gene, Ferredoxin I gene, and chloroplast precursor gene were searched. Zinc finger gene and Similar to Fertility restorer gene were searched in the group related to hormone signaling. In the secondary metabolites group, Similar to Cytochrome P450 71C1 gene, Similar to Cytochrome P450 71C4 gene, and Cytochrome P450 family gene were searched (Table 3).   To classify the 120 CNDH population into BLB-resistant and susceptible groups, the lesion length that occurred in leaves after Xoo inoculation was analyzed ( Figure 5A). And finally, Nagdong, CNDH23, CNDH100, CNDHJ105, and CNDH112, which had little or very short lesion length, were classified as resistant group, and Cheongcheong, CNDH49, CNDH99, CNDH106, and CNDH115, whose leaves were completely browned or completely withered, were classified as a susceptible group. Using these groups, the relative expression level after Xoo inoculation was analyzed. The BLB-resistant lines had a higher relative expression level than all BLB-susceptible lines from 1 h to 72 h after inoculation ( Figure 5B). Similar to Chloride channel protein CLC-d (AtCLC-d) Os03g0696300 CCAAT-binding transcription factor, subunit B family protein Os03g0698800 Zinc finger, CCCH-type domain containing protein Os03g0698900 Alkaline phytoceramidase family protein Os03g0701200 Similar to Sugar-starvation induced protein (Fragment) Os03g0702000 UDP-glucuronosyl/UDP-glucosyltransferase family protein Os03g0702500 UDP-glucuronosyl/UDP-glucosyltransferase family protein Os03g0703000 Similar to Beta-glucosidase Os03g0703100 Similar to Beta-glucosidase Os03g0703200 Protein kinase-like domain containing protein Os03g0704700 Oxysterol-binding protein family protein Os03g0706900 Zinc finger, RING-type domain containing protein Os03g0707600 OsGAI Os03g0708100 Phytanoyl-CoA dioxygenase family protein Os03g0708900 Zinc finger, RanBP2-type domain containing protein Os03g0710100 Protein kinase-like domain containing protein Os03g0710500 Similar to Luminal binding protein 2 precursor (BiP2) 8 RM149-RM23191 Os08g0439000 Phosphofructokinase family protein Os08g0439900 Mitochondrial glycoprotein family protein Os08g0440100 Similar to Temperature stress-induced lipocalin Os08g0442300 Similar to Calcineurin-like protein Os08g0452500 Auxin responsive SAUR protein family protein Os08g0452900 Non-protein coding transcript, unclassifiable transcript Os08g0453200 Dormancyauxin associated family protein Os08g0459700 Similar to Adenosine diphosphate glucose pyrophosphatase precursor very short lesion length, were classified as resistant group, and Cheongcheong, CNDH49, CNDH99, CNDH106, and CNDH115, whose leaves were completely browned or completely withered, were classified as a susceptible group. Using these groups, the relative expression level after Xoo inoculation was analyzed. The BLB-resistant lines had a higher relative expression level than all BLB-susceptible lines from 1 h to 72 h after inoculation ( Figure 5B).

Discussion
BLB causes serious damage not only to rice but also to various crops and is a major cause of decline in yields of major crops worldwide [24]. Xanthomonas oryzae pv. oryzae is a bacterial pest that causes BLB, and it proliferates in the xylem and phloem of the plant and interferes with the movement of nutrients and water. In the end, nutrients cannot move in plants, and leaves turn white and photosynthesis is disturbed, which is a major cause of reduction in yield and deterioration of taste. Pathogen that cause BLB continues to mutate. Therefore, research related to the current continuous tracking of races and the discovery of resistance genes corresponding to each race is essential for solving future food problems in response to current climate change. The most effective method for BLB control currently reported is to cultivate resistant varieties. However, since the causative bacteria inducing BLB have a very fast appearance of races, it is essential to develop resistant varieties using new resistance genetic resources to match their speed [25,26]. Therefore, it is essential to analyze the characteristics of the rice BLB progenitor and to analyze the various phenotypic changes of the plant in response to the pathogen inoculation to explore the resistance gene [25]. Recently, an accurate, efficient, and fast breeding system has been essential to develop high-yield varieties in response to unpredictable climate change and use them to develop rice varieties that can adapt to various climates and stressful environments. It is phenomics that makes these technologies a reality [27]. Phenomics is a key base technology for innovative development of agricultural biotechnology, and it is beneficially used for analysis of valuable gene functions of crops, molecular markers, and early selection of excellent lines for molecular breeding through phenotype-based trait evaluation [28]. In particular, image analysis for phenotyping was used in various ways to select disease-resistant varieties of plants [29,30]. Because accurate evaluation of each trait is very important to improve crop traits, it is very important to reliably and accurately determine disease resistance and specific traits in large quantities. To date, various phenotypic data have been used to map genes related to BLB resistance, but in this study, OCT was utilized for phenotyping that minimized subjective factors. The microscopic phenotype data of the rice leaf surface extracted from the OCT image results showed that the rice CNDH population was infected with BLB. Necrosis of the internal cells and tissues of the rice leaf was apparent, which reduced the rice leaf thickness. The closing angle was significantly larger in inoculated plants, demonstrating that the internal rice leaf tissue and cells were damaged by pathogens. BLB inoculation impacted basic plant functions, such as reducing the photosynthesis rate [31].
Recently, OCT has visualized important plant tissues using optical methods and has allowed the morphological and functional status of plant tissues to be evaluated in vivo [32]. This analytical method enables real-time monitoring of the morphological and functional state of the tissue in a time of 2-5 s in intact plants without removing them from the habitat [9]. In this study, QTLs related to BLB resistance were mapped using OCT. The 120 CNDH population is a population created by another culture of F 1 made by crossing Cheongcheong and Nagdong. Generation has progressed in the field of Kyungpook National University since 2010, and it has a very diverse phenotype and genotype. The 120 CNDH population is currently used as an intermediate model and is also used to analyze the functions of newly discovered genes [33,34]. After inoculating the 120 CNDH population with Xoo, which induced BLB, leaf analysis was performed using OCT. When the phenotype of the 120 CNDH population after BLB inoculation was analyzed using OCT, there was a difference in leaf thickness and leaf angle between the group resistant to BLB and the group susceptible to BLB.
Leaves were classified into adaxial and abaxial, and QTLs related to BLB resistance were mapped using the data analyzed for leaf thickness and leaf angle of the 120 CNDH population after Xoo inoculation. When the normal distributions of leaf adaxial thickness, leaf abaxial thickness, and leaf abaxial angle were analyzed after Xoo inoculation of 120 CNDH populations, either the normal distribution followed or the normal distribution was skewed to the right. Following a normal distribution means that no one gene related to BLB resistance acts, but various genes interact and participate [35].
QTLs related to BLB resistance were mapped using phenotyping data using OCT. QTLs with the LOD score of 3.0 or higher were detected on chromosomes 1, 6, 8, and 12. Among them, qio-3t-6 of chromosome 6 is a region where only the thickness data of the adaxial lower region were detected, and the remaining QTLs were regions where various traits were repeatedly detected. ORFs related to resistance to various stresses, including BLB, were searched around the commonly detected region. These regions contain 23% of WRKY family genes associated with BLB resistance, 38% of plant defense genes, 15% of hormone signaling genes, and 23% of secondary metabolite genes. When plants are exposed to various stressful environments, various gene expression changes occur [36,37]. Plants are also immobile and must protect themselves from attack by various pathogens [38]. Therefore, plants have a more developed defense system than other species.
In particular, numerous genes involved in stress are regulated at the transcription level, and genes are up-regulated and down-regulated by various transcription factors including the WRKY transcription factor [39]. Plant hormones act quickly and precisely to confer resistance to stressful environments. In addition, since plants cannot move, various structures and histologies have been evolved to overcome stress, and the production of secondary metabolites is particularly important among them. Plants cannot move, and they live in the first place. Therefore, secondary metabolites that play an important function in plant resistance are as essential as primary metabolites [40]. In this study, key genes related to plant stress resistance and plant defense were detected in the regions mapped with QTLs related to BLB resistance. Therefore, it can also be used as a disease-resistance-related molecular marker centered on the detected region. We also focused on WRKY family genes among BLB resistance candidate genes searched for BLB resistance. Among these WRKY family genes, WRKY transcription factor 34 (OsWRKY34q1) was used to analyze the relative expression level using the resistant or susceptible group after Xoo inoculation.
After Xoo inoculation in the 120 CNDH population, the lesion length was analyzed and divided into resistant and susceptible groups. The relative expression level of Os-WRKY34q1 was high in the resistant group, and there was no difference in the sequence of OsWRKY34q1 in the resistant and susceptible group (data not shown). Currently, numerous QTL mappings have been attempted to discover genes associated with BLB resistance. In 2008 Piz-5, pi1, and pita associated with BLB resistance were mapped on chromosomes 6, 11, and 12, respectively [41]. In 2000, BLB resistance-related genes were mapped for two consecutive years, and QTLs related to BLB resistance were repeatedly mapped on chromosomes 1, 2, 3, and 5 [42]. The reason why QTLs related to BLB resistance were mapped at similar locations or various other chromosomes in previous studies was because the number, size, and environmental factors of the population used in the study were different [43].
WRKY family genes are responsible for very important functions such as development stage, defense system, and signaling in plants [44]. Therefore, OsWRKY34q1 screened in this study can be effectively used to develop BLB resistant varieties. In particular, since it was phenotyping data using OCT, candidate genes related to BLB resistance have been additionally discovered using accurate and objective data. These can be used very effectively to develop key resistant varieties in response to the differentiation of existing Xoo species.

Conclusions
For centuries, humans have consciously domesticated animals and plants according to their needs, and the history of phenotype investigations predates those of the genotypes. In recent years, the rapid development of high-throughput sequencing technology has facilitated simple and rapid genotype analyses. However, due to the dynamics and complexity of plant phenotypes, phenotype research lags behind genotype research. Agricultural breeding to select for pathogen resistance is necessary to combat the continuous deterioration of environmental resources and the increasingly severe food crisis. However, we observed a recent bottleneck in research on macroscopic visualization traits. Thus, we developed high-throughput OCT technology to observe the microscopic traits of plants. Using micro-traits data for QTL analysis, we successfully screened out WRKY transcription factor 34, which confers resistance to BLB. Additionally, other candidate genes for related resistance in different intervals were screened out. Moreover, our QTL results showed that the target interval identified using the QTL results before and after the BLB inoculation coincided perfectly with those obtained using OCT microscopic trait data. These findings demonstrate that the application of microscopic traits is effective and accurate for QTL analysis and molecular breeding.

Data Availability Statement:
The data presented in this study are available from the authors on request.