Editor's evaluation: Genome editing of an African elite rice variety confers resistance against endemic and emerging Xanthomonas oryzae pv. oryzae strains
Year: 2023
Type: peer-review
Abstract: Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Methods Data availability References Decision letter Author response Article and author information Metrics Abstract Bacterial leaf blight (BB) of rice, caused by Xanthomonas oryzae pv. oryzae (Xoo), threatens global food security and the livelihood of small-scale rice producers. Analyses of Xoo collections from Asia, Africa and the Americas demonstrated complete continental segregation, despite robust global rice trade. Here, we report unprecedented BB outbreaks in Tanzania. The causative strains, unlike endemic African Xoo, carry Asian-type TAL effectors targeting the sucrose transporter SWEET11a and iTALes suppressing Xa1. Phylogenomics clustered these strains with Xoo from Southern-China. African rice varieties do not carry effective resistance. To protect African rice production against this emerging threat, we developed a hybrid CRISPR-Cas9/Cpf1 system to edit all known TALe-binding elements in three SWEET promoters of the East African elite variety Komboka. The edited lines show broad-spectrum resistance against Asian and African strains of Xoo, including strains recently discovered in Tanzania. The strategy could help to protect global rice crops from BB pandemics. Editor's evaluation This valuable study shows that new, virulent genotypes of Xanthomonas oryze pv. oryzae, that are similar to strains present in east Asia, cause outbreaks of bacterial blight of rice in Tanzania. The authors' use of CRISPR-based gene editing on multiple pathogen targets in an elite African rice variety to create lines resistant to both endemic and emerging pathogen strains in Africa makes for a compelling contribution to meet this alarming development. https://doi.org/10.7554/eLife.84864.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Rice is one of the most important staple foods for developing countries in Asia and Africa (Odongo et al., 2021). African consumers increasingly replace traditional staples such as sorghum, millet, and maize with rice. Today, African farmers, 90% of which are small-scale food producers with <1 ha of land (Pandey et al., 2010), produce ~60% of local rice demand. The demand will likely increase with the expected doubling of the population until 2050. Productivity is often hampered by diseases, such as Bacterial Leaf Blight (BB), Rice Yellow Mottle Virus (RYMV), and Rice Blast (RB) (Jiang et al., 2020; Longue et al., 2018; Mutiga et al., 2021). Breeding high-yielding varieties with resistance to these diseases will be an important factor needed for food security in Africa. BB, caused by the bacterium Xanthomonas oryzae pv. oryzae (Xoo), is a devastating rice disease in many rice-growing countries. Xoo comprises a wide spectrum of pathovars with diverse virulence mechanisms. In rice, resistance (R) genes for BB have been identified and are used extensively to breed resistant varieties. However, resistance based on single R genes was rapidly overcome by new pathovars (Vera Cruz et al., 2000). Regular monitoring of the virulence of current Xoo populations on a collection of rice tester lines carrying single or combinations of R genes has provided effective guidance for stacking different R genes to obtain broad-spectrum resistance in Asian rice varieties (Arra et al., 2017; Arra et al., 2018). In Africa, BB was first reported in Mali and Cameroon in the late 1970s, and later in several other West African countries (Buddenhagen et al., 1979; Verdier et al., 2012b). Recently, East African countries also reported BB epidemics (Duku et al., 2016). While information on the spread and severity of BB in Africa is scarce, BB is not yet considered a major threat in Africa; nevertheless, BB is an established disease. The situation is unstable, because climate change affects disease spread, and damage is expected to become more severe in the future due to climate change (Amos, 2013; Laha et al., 2016; Sere and Ouedraogo, 2005; Verdier et al., 2012b). Increased demand and the need to improve food security make it essential to generate broad-spectrum and durable resistance in local varieties specifically for Africa. Key aspects of the mechanisms underlying BB disease have been elucidated and provide an efficient roadmap for resistance breeding. Xoo strains secrete a suite of Transcription Activator-Like effectors (TALes) into rice xylem parenchyma cells via type III secretion systems. Once inside the host cell, TALes are targeted to the nucleus, where they bind to effector binding elements (EBEs) in the promoters of specific host genes via a unique domain of tandemly arranged 34 amino-acid long repeats. TALes function as transcriptional activators that trigger ectopic induction of target genes (Richter et al., 2014). Several Xoo TALes induce one or several host SWEET sucrose uniporter genes (SWEET11a, 13 and 14), presumably causing sucrose release from the xylem parenchyma into the apoplasm at the sites where the bacteria reside and reproduce (note that due to the discovery of a previously misannotated sucrose transporting paralog, SWEET11a was renamed formerly SWEET11 Wu et al., 2022). Sucrose is consumed by the bacteria, resulting in effective reproduction (Sadoine et al., 2021). Allelic EBE variants that cannot be recognized by TALes function as recessive gene-for-gene resistance genes. Without SWEET induction, Xoo is not a potent pathogen. To date, six target EBEs in SWEET promoters have been identified in a comprehensive Xoo collection. Notably, Asian and African Xoo strains are phylogenetically distinct and use distinct TALes to target SWEETs (Oliva et al., 2019; Tran et al., 2018). African Xoo strains exclusively use TalC and TalF, which both target SWEET14, while Asian strains encode PthXo1, PthXo2 (variants A, B, C), PthXo3 and AvrXa7, which target SWEET11a, 13, and 14, respectively (Eom et al., 2019; Oliva et al., 2019). American Xoo strains lack TALes and hence are poorly virulent (Verdier et al., 2012a). Based on the information of the TALe compendium and the SWEET target sites, broad-spectrum resistance to Asian strains was successfully introduced into Oryza sativa ssp. japonica cv. Kitaake and the spp. indica varieties IR64 and Ciherang-Sub1 by genome editing the EBEs in all three SWEET promoters (Eom et al., 2019; Oliva et al., 2019). The edited lines may prove to be valuable breeding materials that could benefit small-scale producers in Asia. To prepare for emerging strains with novel virulence mechanisms, the diagnostic SWEETR kit was developed, enabling effective characterization of causative strains and their SWEET-based disease mechanisms, and identification of suitable resistance strategies, such as the best possible SWEETR-gene combinations for deployment in the target country (Eom et al., 2019). First observations of BB in Africa were made as early as 1979 in Mali and Cameroon (Buddenhagen et al., 1979). Since then, the disease has been reported mostly in West Africa, including Senegal, Benin, Burkina Faso, Ivory Coast, Mali, and Niger (Afolabi et al., 2016; Gonzalez et al., 2007; Tall et al., 2020; Tekete et al., 2020). More recently, BB was also reported in the East African countries Uganda and Tanzania (Duku et al., 2016; Oliva et al., 2019). However, at present, little information on the genetic diversity and dynamics of Xoo populations is available for Africa. While pathogenic strains have sporadically been isolated from rice plants and characterized, there has been no systematic analysis of Xoo strains and BB disease severity across Africa. Moreover, diverse rice varieties are used across Africa, and there is only limited information on the acreage used for different varieties. These factors make it challenging to breed durably resistant lines for Africa. Numerous international partnerships have introduced varieties and cultivation techniques to major rice production areas in different countries in Africa. For instance, in 1975, one of the first irrigated perimeters with rice research labs and a training farm was developed in Dakawa, a major rice production zone in Tanzania, as a result of a multi-year partnership with North Korea (van et al., 2022). National (NAFCO) and international institutions (USAID, Cornell University) remain active in Dakawa to build irrigation schemes and implement The System of Rice Intensification (SRI). In 2011, a collaboration with China resulted in 50 ha of irrigated fields managed by local farmers. A Technology Demonstration Center that trains local farmers and technicians and grows Chinese varieties, including hybrid rice has been established (Makundi, 2017). Recent disease surveys in Tanzania, described here, identified an outbreak of BB in 2019 that has spread in subsequent years. We show that the causative strains have features that distinguish them from endemic African Xoo strains, and unravelled the virulence mechanisms that make them a major threat to African rice production. Recently, Kenya and Nigeria exempted SDN-1 genome-edited crops generated using site-directed nuclease without a transgene (SDN-1) from GMO regulations (Buchholzer and Frommer, 2023). This exemption provides a regulatory framework for the introduction of genome-edited, BB-resistant rice into African countries such as Kenya and Nigeria. As a first step towards the generation of BB-resistant rice cultivars for small-scale rice producers in these countries, we used an optimized transformation protocol to edit susceptibility genes in the rice cultivar Komboka, which is popular in East African countries (Luu et al., 2020). Komboka (IRO5N-221) is a relatively new high-yielding, semi-aromatic rice variety jointly developed by IRRI (International Rice Research Institute) and KALRO (Kenya Agricultural & Livestock Research Organization; BBSRC Varieties Kenya.pdf) that grows to an average height of 110cm and has a yield potential of~7 tons/ha, - almost twice that of Basmati, another popular variety planted in Kenya and Rwanda (The Star 2020). Komboka plants mature in~115days, respond well to fertilization, and are suitable for irrigated lowland cultivation in Africa (Kitilu et al., 2019). Komboka had been classified as moderately resistant to BB, RYMV, and RB; however, detailed resistance profiles for specific strains or races are not available, rendering this information of little value for breeding resistance in the context of specific pathovars present locally. Here, BB-resistance and susceptibility genes of Komboka were analyzed, and a combination of Cas9, Cpf1 and multiplexed gRNAs was used to edit all known EBEs in the SWEET genes, resulting in broad-spectrum resistance to not only previously known African and Asian strains, but also the newly introduced Xoo strains identified from the outbreak in Tanzania. Results Identification of an unprecedented disease outbreak in Tanzania Local breeders reported that although BB was found in Tanzania, it was not considered a major disease due to low incidence and low severity up until 2019 (RM, pers. comm.). Here, we report on an unprecedented BB outbreak identified in the irrigated schemes of Dakawa and Lukenge in 2019 and 2021, respectively (Figure 1; Figure 1—figure supplement 1, Figure 1—figure supplement 2). The two sites are ~60 km apart in the Morogoro region, an area that for decades has been a center of partnership initiatives that aim at increasing rice production (Makundi, 2017). We performed disease surveys in this area in 2019 and detected a severe outbreak in Dakawa on TXD 306 (SARO-5), a rice variety popular in irrigated ecologies in Tanzania (Figure 1—figure supplement 1). Later on, in 2021 in Lukenge, several fields also showed severe signs of infections (Figure 1; Figure 1—figure supplement 2). Leaves collected in 2019 and 2021 with typical BB symptoms were processed, resulting in the isolation of seven strains from Dakawa and 106 from Lukenge, all validated as Xoo using diagnostic primers (Lang et al., 2010; Table 1, Supplementary file 1a). In 2022, we surveyed BB on a larger scale and collected over 600 leaf samples from plants with typical BB symptoms from 37 fields in six different regions in Tanzania. The results of the survey show that BB has spread across Tanzania (Figure 1; Supplementary file 1b). Multiple Locus Variable Number of Tandem Repeat Analyses and/or whole genome SNP analysis is in progress and is expected to provide new insights into the genetic structure of the Xoo population in Tanzania. A careful analysis of yield losses has not yet been possible. Out of six fields surveyed in Dakawa in 2019, one was very severely infected (Figure 1—figure supplement 1). The yield loss in this field was estimated to be >60%. One of the fields surveyed in Lukenge in 2021 had an estimated yield loss of 10–15%. In 2022, several Dakawa fields had estimated yield losses of 5–10%, others <5%. Yield loss at other sites such as Igunga was estimated to 10%, in Mombo <5%. Note that estimates of severity and yield losses are snapshots, and careful and systematic multiyear analyses will be required for a reliable evaluation of severity, spread and yield losses. Figure 1 with 2 supplements see all Download asset Open asset Detection of an outbreak and survey of BB in Tanzania. (A) No reports of serious BB infections before 2019. (B) Detection in Dakawa; in 2019, (C) in Lukenge in 2021, (D–I) survey results from 2022 at different degrees of severity. Note that sampling and scoring occurred initially in the Morogoro region in Dakawa in 2019 and in Lukenge 2021, (no info on 2020 due to Covid-19); scoring and sampling was expanded to additional fields across Tanzania in 2022. Due to sampling bias early in the outbreak, data should be interpreted with caution. One may project that BB caused by the introduced strains will manifest in Tanzania and could spread to neighboring countries. Leaf area affected: yellow 4–6%, orange 7–12%, red 13–25% and purple 26–50%. Table 1 Efficiency of resistance genes in IRBB rice lines towards a diversity panel of 26 African Xoo strains. Quantitative scores of lesion length produced upon leaf-clip inoculation of a diversity panel consisting of 26 Xanthomonas oryzae pv. oryzae strains from West and East Africa on IRBB near-isogenic-lines (NILs) carrying different Xanthomonas resistance genes (Xa) and on the variety Azucena used as susceptible control (IRBB1: Xa1; IRBB3: Xa3; IRBB4: Xa4; IRBB5: xa5; IRBB7: Xa7; IRBB23: Xa23; IR24: recipient parent). Resistance or susceptibility of rice plants to Xoo was determined based on lesion length measured 14 days after inoculation: Resistant (R) <5 cm; moderately resistant (MR)=5–10 cm; moderately susceptible (MS)=10–15 cm; and susceptible (S) >15 cm. Strain NameCIX codeCountryAzucenaIR24IRBB1IRBB3IRBB4IRBB5IRBB7IRBB23KombokaNatiPark607BeninSRRRRRRRRKarfiguela13705Burkina FasoSRRRRRRRRN2-44482NigerSRRRRRRRRTanguieta3609BeninSMRRMRRRMRRRToula20629NigerSMRRMRRRMRRRBAI2504127Burkina FasoSMRRMRMRRRRRS62-2-222374SenegalSMRRMSRRMRRRCII-14083Ivory CoastSMSMRMSRRMRRRCII-21042Ivory CoastSMSMRMSMRRMRRRNAI92787NigerSMSMRMSMRMRMSRRAXO19471917CameroonSSRSRRMSRRMAI145894MaliSSRSMRMSSRRBAI34092Burkina FasoSMSMRSMRMRMSRRCFBP19482801CameroonSMSMRSMRMRSRRNAI54099NigerSSMRSMRMRSRRMAI734079MaliSSRSMSMRSRRS82-4-32976SenegalSMSMRSMSMRMSRRMAI1324517MaliSSRSSMSSRRiTzDak19-14457TanzaniaSSSSSSSSSiTzDak19-24458TanzaniaSSSSSSSSSiTzDak19-34462TanzaniaSSSSSSSSSiTzLuk21-14506TanzaniandndndndndndndndndiTzLuk21-24509TanzaniandndndndndndndndndiTzLuk21-34505TanzaniandndndndndndndndSiTzLuk21-44507TanzaniandndndndndndndndSiTzLuk21-54508Tanzaniandndndndndndndndnd n.d. not determined. To systematically analyze the newly isolated strains from Tanzania, and to compare them to the broader African Xoo landscape, we used 8 strains isolated from Dakawa in 2019 and Lukenge in 2021, as well as 18 representative strains from a collection of 833 strains sampled from rice fields in nine other African countries between 2003 and 2021. These endemic African strains had previously been identified as Xoo based on molecular diagnostic and pathogenicity assays (BS, unpubl. results; Supplementary file 1a). To evaluate the effectiveness of the resistance genes Xa1, Xa3, Xa4, xa5, Xa7, and Xa23, known to be efficient against African Xoo (Gonzalez et al., 2007; Tekete et al., 2020), six Near Isogenic Lines (NILs) of rice each harboring one of these single BB resistance gene, were inoculated with 18 endemic African strains and three strains isolated from Dakawa in 2019 (iTzDak19-1, iTzDak19-2 and iTzDak19-3; strain naming: Dak for Dakawa, Tz for Tanzania, and the novel strains collected in 2019 and 2021 obtained an 'i' for introduced) (Table 1; Figure 2). NILs harboring Xa1, Xa23, xa5 or Xa4 were resistant to most endemic African strains, but surprisingly, iTzDak19-1, iTzDak19-2, and iTzDak19-3 were virulent on all NILs tested. These unusual iTz strains were therefore characterized in more detail (see below). The strains from Lukenge were isolated much later and could so far not be included in the race-typing analysis, but were also characterized in more detail as decribed below. Figure 2 with 3 supplements see all Download asset Open asset Resistance spectrum of wild-type Komboka rice against African Xoo strains. Leaf-clip inoculation of wild-type Komboka rice plants with a panel of 21 Xoo strains originating from 8 African countries along with Asian reference strain PXO86 from the Philippines. Lesion length (in cm) was measured 14 days after Xoo inoculation. Data from three independent experiments are represented. Letters in the boxplot represent significant differences. As a representative of the isolates from Lukenge iTzLuk21-3 was tested for virulence on Komboka and shown to be fully susceptible (Figure 2—figure supplement 1). Resistance of Komboka to endemic African Xoo but not to new Tanzanian strains Komboka, an emerging elite variety released in several East African countries (Tanzania, Kenya, Uganda and Burundi) has been described as moderately resistant to Xoo, yet the nature of its resistance has not been elucidated. Using our African Xoo diversity panel, we found that most African strains were avirulent on Komboka, except for representatives of newly isolated strains from Dakawa (2019) and Lukenge (2021); (Figure 2, Figure 2—figure supplement 1; Table 1). By comparison, all strains were highly virulent on the susceptible variety Azucena (Figure 2—figure supplement 2; Table 1). Since Xa1, Xa4, xa5, and Xa23 confer resistance against Xoo (Table 1), we investigated R gene presence in the Komboka genome. Mining of the IRRI QTL database (https://rbi.irri.org/resources-and-tools/qtl-profiles) revealed that Komboka contains genetic markers linked to Xa4 but not xa5 or Xa23, while there was no information on Xa1 in the database (Figure 2—figure supplement 3A). We confirmed the presence of Xa4 in Komboka by tracing an Xa4-associated marker and by sequencing (Figure 2—figure supplement 3B,C; Supplementary file 1c). Consistent with the broad-spectrum resistance of Xa1 against endemic African Xoo strains, we found that Komboka a Xa1 that is highly similar to et al., 2020; Figure 2—figure supplement The combination of Xa4 and in Komboka the resistance to most African Xoo strains. However, the R genes present in Komboka do not protect against the iTz strains recently isolated from the outbreak in Tanzania. Tanzanian Xoo strains with Asian Xoo via whole genome To the mechanisms underlying the virulence of the Tanzanian strains, three strains collected from Dakawa in 2019, and from Lukenge in 2021, were to sequencing and analyses file Supplementary file 2). Notably, all strains clustered with Asian African Xoo isolates (Figure 3A). By Tanzanian strains and which had been collected in Dakawa before 2019, with the endemic African with reports (Oliva et al., 2019). Based on the analysis of the the recently isolated strains only genome (Figure supplement 1), that they from a single introduction The analysis that the new strains 'i' for Tz for are most to strains from analyses will be to the of the relatively between the strains from and the iTz strains (Figure The new strains all carry genes similar to of (Figure and genes encode TALes that Xa1 but had so far only been reported in Asian Xoo In the strains a TALe highly similar to the Asian PthXo1, with the of two at the of the (Figure Supplementary file EBE that bind to the SWEET11a at a with the EBE that is targeted by (Figure Consistent with a from the diagnostic SWEETR kit were resistant to representative strains from the outbreak in Dakawa and Lukenge, and or susceptible (Figure Moreover, SWEET11a were in Kitaake after inoculation with and iTzLuk21-3 (Figure our data show that the iTz strains collected in in the Morogoro region are to each and phylogenetically to Asian Xoo strains. The strains both and unique to Asian Xoo insights into the virulence and a for the of that protect African rice varieties, and in Komboka, against the novel iTz strains found in the outbreak (Oliva et al., 2019). Figure 3 with 1 supplement see all Download asset Open asset of the of Tanzanian Xoo (A) genome Xanthomonas oryzae the of Tanzanian isolates are Xoo, oryzae pv. oryzae pv. oryzae pv. (B) of the of the in A, including the newly isolated Tanzanian strains and neighboring have a value to or The scale length in of the of in to the of (C) Multiple of the domain of iTALes and the iTALes from the newly isolated Tanzanian iTz strains. The domain of from was used as TALe represent are EBE on the SWEET11a Repeat Variable are along their along the their target in the with the best possible for this in with an in the the TALe to are provided in Figure data Figure data 1 with of Xoo strains and Download Figure Download asset Open asset of the new Tanzanian Xoo strains on the induction of (A) Lesion were measured 14 days after inoculation of Kitaake lines and in the cultivar Kitaake (Eom et al., with and Tanzanian strains and from the outbreaks by a red Results from two independent experiments are represented. (B) of SWEET11a, and in wild-type Kitaake upon by and Tanzanian Xoo strains. were collected Data from three independent experiments were and are were to the rice factor SWEET in Komboka to obtain resistance to Asian and African strains As a for editing SWEET promoters in Komboka, the regions of SWEET11a, 13 and 14 were The promoters EBEs for and AvrXa7, and TalF, respectively (Figure Figure supplement 1, Supplementary file To protect Komboka against endemic strains from Africa as well as introduced Asian strains, all six known EBEs in the promoters of SWEET11a, 13 and 14 were edited using a hybrid editing Due to while Cpf1 We that may produce more robust resistance because the more from the target Moreover, the of combinations of in all EBEs is expected to be to (Oliva et al., 2019). Due to the it was not possible to for Cpf1 at all Cpf1 was with gRNAs and were to target A and EBEs in SWEET11a and respectively (Figure supplements 1 and 2; Supplementary file Since the EBEs for AvrXa7, PthXo3 and in are one was to target the region of all three no was available the EBE for TalC for a required by a was to target the TalC a first of six representative lines were tested for resistance using six representative Xoo in PXO86 and (Figure Figure supplement Supplementary file six lines from this first of transformation were resistant to the tested strains. One was fully resistant to all strains, while lines were fully resistant to strains, but only moderately resistant to This in resistance is most likely by the presence of a in the EBE for AvrXa7, while the an additional (Figure Since the assays very bacterial it is that resistance will be for effective resistance in field et al., et al., 2016). However, TALes were reported to have specific binding at the hence the in could be overcome by of (Richter et al., 2014). To obtain more robust resistance to all known Xoo a of transformation was lines and which in all including and in the were resistant to all strains including PXO86 (Figure Lines and were also resistant to and with the and in the EBE (Figure As one may have induction of SWEET11a by and iTzLuk21-3 was in lines and (Figure supplement 1). Komboka lines with resistance to representative Asian and African Xoo strains were with resistance to two representative strains from the emerging iTz population from Dakawa and Figure with 3 supplements see all Download asset Open asset Resistance of Komboka lines against six representative Xoo strains. (A) of and six edited Komboka lines to by Xoo strains and harboring PthXo1, PthXo2 AvrXa7, TalC and/or is a strain with in and as a The Komboka lines were in all six EBE sites validated by of the EBEs targeted by and TalF, and in the six edited Figure Download asset Open asset and of two lines generated in a transformation (A) at the EBEs for PthXo1, TalF, and in two Komboka edited and of wild-type Komboka and the two edited lines to the
Source: eLife
Author: Detlef Weigel
Institution: Max Planck Institute for Biology
Language: English
Subfield: Plant Science
Domain: Life Sciences
SDG: Zero hunger
Open Access status: gold
Estimated APC paid: $2,500
OpenAlex ID: W4383268048