INTRODUCTION

Fusarium wilt of lettuce, caused by the soil-borne fungus Fusarium oxysporum f. sp. lactucae (FOL), is a devastating disease that poses a significant threat to intensive production systems of soil-grown lettuce (Lactuca sativa L.) worldwide (Paugh and Gordon, 2019). Recent evidence indicates that FOL can also spread in hydroponic and soilless cultivation systems (Thongkamngam and Jaenaksorn, 2017), potentially posing an additional risk to large-scale lettuce production under these controlled conditions. Infected lettuces initially exhibit stunted growth accompanied by chlorosis (Paugh and Gordon, 2020). As the pathogen colonizes the plant’s vascular tissues, characteristic wilt symptoms―including progressive leaf yellowing, vascular browning, and eventual plant collapse―emerge in irregular patches, gradually spreading and eventually affecting extensive portions of the field(SCOTT et al., 2010;Garibaldi et al., 2004;Paugh and Gordon, 2019). Disease severity markedly intensifies under elevated temperatures (SCOTT et al., 2010), leading to more rapid and aggressive colonization of susceptible cultivars, particularly during warm growing seasons. Consequently, global climate warming is likely to accelerate both the spread and detrimental effects of this disease worldwide (SCOTT et al., 2010;Singh et al., 2023).

Currently, available methods for managing Fusarium wilt in lettuce are limited, as effective soil fumigation and sterilization practices are often economically impractical and technically challenging, particularly in intensive, large-scale production systems (Garibaldi et al., 2004). Moreover, chemical or physical disinfection strategies may only offer short-term relief due to the persistence of Fusarium spores in soil and growing media (Garibaldi et al., 2004, Thongkamngam and Jaenaksorn 2017). Consequently, the development and deployment of resistant lettuce cultivars have emerged as the most promising and sustainable strategy to mitigate damage caused by Fusarium wilt (Garibaldi et al., 2004;Kim et al., 2016). Breeding programs aimed at successfully identifying and incorporating genetic sources of resistance into commercial lettuce varieties are therefore essential, providing durable protection and significantly reducing disease impacts in both traditional soil-based and modern hydroponic cultivation systems (Garibaldi et al., 2004;Thongkamngam and Jaenaksorn, 2017).

A physiological race is a subset of a pathogen species able to overcome a specific combination of host resistance (R) genes, thereby producing a characteristic infection profile on a standardized differential cultivar set (Garibaldi et al., 2004). Because each known lettuce R gene protects against only one―or at most a few―physiological races of FOL, accurate race determination in every production area is essential for deploying durable resistant cultivars. Race typing has traditionally relied on greenhouse or growth‑chamber inoculation of the differential set, a biologically definitive but labour‑intensive procedure that is difficult to scale beyond a few hundred isolates per season, even when a robust pathosystem and ample space are available. To support highthroughput surveys and time‑sensitive breeding decisions, PCR‑based molecular diagnostics have therefore been adopted (FUJINAGA et al., 2014;Gilardi et al., 2016;Lin et al., 2014;Seki et al., 2021); they deliver rapid, automatable race calls from minute amounts of mycelial or soil DNA. Most assays target lineage‑specific genomic region—whose presence or absence is assumed to correlate with race identity. Yet this binary strategy is vulnerable to (i) genotype–phenotype decoupling caused by horizontal transfer, deletion, or silencing of effector genes (Henry et al., 2021;Li et al., 2020); (ii) race polyphyly and convergent evolution, whereby genetically distinct lineages acquire identical effector complements and are mistakenly assigned to the same race (Helmer and Panek, 2024;Nirmaladevi et al., 2016); and (iii) recombination on accessory chromosomes that removes primer‑binding sites, yielding false‑negative results (Wei et al., 2024). Overcoming these limitations requires genome‑scale information that captures the full repertoire of effectors and accessory sequences. In this study, we will present the draft genome sequence of FOL strain 16‑086, isolated from wilted lettuce in Iksan, Jeollabuk‑do, Republic of Korea. The result of this sequence draft can play as a foundational resource for refining diagnostic markers and elucidating the evolutionary dynamics of pathogenicity in this species complex.

MATERIALS AND METHODS

1. The fungal isolate preparation

The wilt pathogen FOL strain 16‑086 was isolated from vascular tissue of diseased lettuce collected in Iksan, Jeollabuk‑do, Republic of Korea. Tissue segments (≈ 5 mm) were surface‑disinfested in 1 % N aOCl for 1 m in, rinsed three times in sterile water, blotted dry, and placed on Komada’s selective medium. After 5 days at 25 °C, a single germinated microconidium was picked, transferred to potato dextrose agar (PDA), and sub‑cultured under a 12 h photoperiod at 25 °C for 10 d to obtain a monoconidial culture. The species identity was confirmed by BLAST analysis of the translation‑elongation‑factor 1‑α (EF‑1α) gene sequence. Conidia harvested from PDA were counted with a haemocytometer and adjusted to 1 × 10⁶ spores ml⁻¹. Inoculation of this suspension onto the susceptible lettuce accession ‘knou322’ reproduced characteristic Fusarium wilt symptoms.

2. Genomic DNA extraction, sequencing, assembly and annotation

High‑molecular‑weight genomic DNA of FOL 16‑086 was isolated from lyophilised hyphal mats using the HiGene™ Genomic DNA Prep Kit (BioFACT, Republic of Korea) according to the manufacturer’s protocol. A TruSeq DNA Nano library was constructed and sequenced on an Illumina HiSeq X Ten platform (2 × 151 bp paired‑end) at Macrogen Inc. (Seoul, Republic of Korea). Adapter trimming and quality filtering were performed with Trimmomatic v0.33(Bolger et al., 2014), and the cleaned reads were assembled de novo with SOAPdenovo v1.05 (k‑mer = 49). Scaffold gaps were closed with GapCloser v1.12. Structural gene models were predicted using AUGUSTUS v3.1.0 with the Fusarium training set, and functional domains were assigned with InterProScan v5.18‑57.0 (Jones et al., 2014). All analyses were executed within the Genome Information System (GeIS; http://geis. infoboss.co.kr/), previously applied in multiple genomic studies(Yun et al., 2025;Kim et al., 2021).

3. PCR‑based race determination of F. oxysporum f. sp. lactucae.

Four race‑specific PCR assays were performed (FUJINAGA et al., 2014;Gilardi et al., 2016;Lin et al., 2014;Seki et al., 2021) using the primer pairs FOLR1‑F/FOLR1‑R, FIGS11/ FOLR2‑R, FOLR3‑F/FOLR3‑R and FUP‑F/FUP‑R as described. Positive amplification with these pairs diagnoses races 1, 2, 3 and the recently characterized Dutch race (provisionally “new race”), respectively. Each 20 μL reaction contained 10 ng of genomic DNA, 0.25 mM of each dNTP, 1 mM MgCl₂, 0.5 μM of each primer, 1× PCR buffer and 1 U of Taq DNA polymerase (Qiagen). Amplification was carried out in a PTC‑100 thermal cycler (MJ Research) with an initial denaturation at 95 °C for 3 min; 25 cycles of 95 °C for 30 s, 60.6 °C for 30 s and 72 °C for 30 s; and a final extension at 72 °C for 5 min. PCR products were resolved on a 2 % agarose gel stained with ethidium bromide, alongside a 100 bp ladder, and scored for the presence or absence of the expected race‑specific amplicons.

RESULTS AND DISCUSSIONS

The draft genome of FOL strain 16‑086 comprises 55.0 Mb (55,044,679 bp) distributed over 16,636 scaffolds, with 1,941 bp of unresolved “N” sequence. This assembly is∼ 4.9 Mb smaller than the 59.9 Mb reference genome Fol4287(Ma et al. 2010), yet its GC content (47.50 %) falls squarely within the 47.5–48.2 % range of other F. oxysporum genomes. AUGUSTUS predicted 16,158 protein‑coding genes, near the lower end of the 14,694–21,460 genes (Wang et al., 2024;Yang et al., 2024;Ling et al., 2022;Hao et al., 2023). InterProScan assigned at least one functional domain to 12,111 of these genes (74.9 %), comparable to coverage reported for cotton isolates (91%) (Urbaniak et al., 2019).

Isolate 16‑086 formed a compact, white‑to‑pale‑violet aerial mycelium on potato dextrose agar (PDA) within ten days at 27 °C under dark condition (Fig. 1a, left). Lightmicroscope examination of the resulting sporulating culture revealed two conidial morphotypes. The predominant propagules were rod‑shaped, aseptate microconidia, visualised on the haemocytometer (Fig. 1a, right). Because these spores constitute the bulk of the harvest obtained by gently flooding and scraping the colony surface, their counts were used to standardise the working inoculum to 1 × 10⁶ conidia mL⁻¹. Less abundant—but diagnostically important— were falcate, consistent with typical Fusarium morphology. Root‑dip inoculation of the susceptible lettuce line ‘KNOU322’ with the microconidial suspension reproduced characteristic vascular‑wilt symptoms post‑infection, thereby fulfilling Koch’s postulates and confirming the pathogenicity of strain 16‑086 toward lettuce.

Race assignment was attempted using four published lineage‑specific PCR assays that discriminate FOL races 1–3 and the recently reported “new race” from the Netherlands (Figure 1c) (FUJINAGA et al., 2014;Gilardi et al., 2016;Lin et al., 2014;Seki et al., 2021). Under standard cycling conditions, strain 16‑086 produced a very faint, correctly sized amplicon only with the race‑3 primer set, whereas the race‑1, race‑2, and “new race” assays were negative.

To verify primer performance, we retested the race‑3 primers against all positive controls at 28 and 30 cycles (Fig. 1d). At 28 cycles, amplification was robust in the race‑3 control, absent in the race‑1 and race‑2 controls, and faint but visible in 16‑086 (Figure 1d, upper panel). At 30 cycles, both the race‑3 control and 16‑086 yielded strong bands, while a weak non‑specific band appeared in the race‑2 control. Thus, we conclude that the PCR screening with the four lineage‑specific assays produced a diagnostic pattern that is suggestive—yet not definitive—of race 3. These observations place the 16‑086 amplicon at the specificity/ sensitivity boundary of the assay, where sequence variability at the primer‑binding sites (primer–template mismatches) or non‑specific priming can generate misleading signals. Consequently, the data do not unambiguously place 16‑086 in race 3; instead, they are consistent with—but do not establish—the following plausible, testable hypotheses: (i) a non–race‑3 (“race 2–like”) background that has acquired or locally rearranged a race 3 marker locus (singlelocus introgression), (ii) an incipient variant diverging from canonical race 3, or (iii) a recombinant genotype with a mosaic accessory‑chromosome complement (genome‑scale mixing). Discriminating among these hypotheses will require multi‑loci race typing and comparative whole‑genome analysis, followed by differential‑host inoculation to determine whether the pathogenicity spectrum of 16‑086 matches—or deviates from—established race 3 behaviour.

To date, four physiological races have been recognised within FOL Race 1, first described in Japan in 1967, is now the most widely distributed lineage, with confirmed outbreaks in the USA (1993), Iran (1995), Taiwan (1998), Brazil (2000), Portugal (2004) and Argentina (2014) (Cabral et al., 2014; Herrero, Gilardi et al., 2016;Paugh and Gordon, 2020). Race 2 and 3 remain largely confined to Japan (Seki et al., 2021), although a highly aggressive race 3 isolate was recently reported from Taiwan (Lin et al., 2014). An additional “new race” was identified in the Netherlands (Gilardi et al., 2016) and provisionally separated from races 1–3 on the basis of differential‑cultivar reactions and single‑locus PCR markers. The rapid intercontinental spread of race 1 has often been attributed to seed‐borne transmission (Gilardi et al., 2016), yet large‑scale PCR surveys capable of detecting infestations as low as 0.1 % have so far failed to support this hypothesis. Alternative dissemination routes—such as contaminated farm machinery—have therefore been proposed, particularly for the rapid expansion of race 1 in the south‑western United State (Paugh and Gordon, 2020).

Molecular race‑diagnosis currently relies on a small set of single‑locus PCR assays that target mitochondrial polymorphisms or transposable‑element insertions presumed to track FOL evolution (Gilardi et al., 2016;Lin et al., 2014;Seki et al., 2021;FUJINAGA et al., 2014). Although useful for routine screening, these presence/absence tests interrogate only a minute fraction of the accessory genome on which host‑specificity genes reside and thus cannot fully resolve lineage boundaries or emerging variants. The ambiguous amplification pattern obtained for strain 16‑086 underscores the limitations of single‑locus diagnostics. High‑quality whole‑genome resources—such as the 55.0 Mb draft assembly of 16‑086—are therefore critical for clarifying FOL population structure; comparative genomics can test whether 16‑086 retains the race 3–associated accessory‑chromosome complement, has lost or truncated key marker loci, or represents an emerging lineage that no longer fits the current race framework. Integrating genome‑scale evidence with classical differential‑cultivar assays is essential to refine the race assignment of 16‑086 and to strengthen molecular diagnostics more broadly. Ultimately, these data will guide the deployment of lettuce resistance genes with durable efficacy across the breadth of FOL diversity.

적 요

DATA AVAILABILITY

Raw Illumina reads and the draft genome assembly of F. oxysporum f. sp. lactucae strain 16‑086 are available from NCBI under BioProject PRJNA758594. The corresponding SRA experiment is SRR15671714, and the BioSample accession is SAMN21031248. These identifiers provide direct links to the FASTQ files and associated metadata for independent re‑analysis.