Morpho-phylogenetics and pathogenicity tests reveal a new species of Fusarium infecting Cannabis sativa L. (hemp) in Northern Thailand

Cannabis sativa L. (cannabis and hemp) is an important medicinal and industrial crop of the family Cannabaceae^[1]. Originating from Central Asia, this annual herb has been widely cultivated for food, fiber, and medicinal and narcotic uses^[2]. Cannabis or marijuana is grown primarily for psychoactive cannabinoids produced in glandular trichomes on female flowers^[3], whereas hemp is a sustainable multipurpose crop valued for its efficient resource use and high-quality biomass for fiber and construction materials^[4]. These two forms differ mainly in cannabinoid composition, particularly delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD), with THC being psychoactive and CBD non-psychoactive, but medically beneficial^[5]. Cannabis and industrial hemp are cultivated in more than 40 countries under diverse legal frameworks, with industrial hemp defined by low THC content (generally ≤ 0.2%–0.3%)^[6,7]. Despite its growing economic value, C. sativa production is constrained by pests and diseases, particularly fungal pathogens that affect yield, quality, and product safety during both pre- and post-harvest stages^[8]. Favorable environmental conditions, genotype susceptibility, and improper post-harvest handling can promote fungal contamination. Among these pathogens, Fusarium is a large and diverse genus of ubiquitous filamentous fungi, distributed worldwide in soil, air, water, and plant tissues. Species of Fusarium are among the most economically important fungi, acting as major plant pathogens that cause significant yield losses in agricultural crops; producing mycotoxins that contaminate food and feed which pose serious health risks to humans and livestock, and are opportunistic pathogens of humans and animals, particularly in immunocompromised individuals. Control of Fusarium diseases remains challenging due to their broad distribution, long-term survival in soil and plant debris, and their ability to overcome host resistance^[9,6].

Plant diseases, particularly root and crown rots and vascular wilts, are caused by members of the Fusarium solani species complex (FSSC), and affect a variety of plants such as soybeans, potatoes, cucurbits, peas, sweet potatoes, Chinese roses, and other legumes^[10−13]. Moreover, Fusarium species are reported to be some of the most destructive pathogens affecting C. sativa, especially during the root and vegetative growth phases, leading to decreased crop quality and potential total yield loss^[8]. For instance, brown rot of flower buds, foliar blight (which is characterized by mycelium covering inflorescences and leaves), and foliar and flower blight (which affects both flowers and leaves) are among the symptoms caused by several Fusarium species. These symptoms underscore the possibility of serious impacts on the plant's reproductive systems and overall health^[14−16]. Furthermore, symptoms of F. oxysporum infection include internal stem decay, crown rot, and yellowing. In addition, it may cause pith browning, brown rot of the roots, root necrosis, chlorosis, and irregular wilt. Notably, it also causes damping-off in cuttings, which negatively affects seedling survival^[16−19]. Similarly, F. solani and F. proliferatum are known to cause pith necrosis, crown infections, and tissue decay, all of which reduce plant health and productivity. At present, C. sativa productivity and health are seriously threatened by these Fusarium species, especially during crucial stages of growth^[8,14,20].

Our ongoing work focuses on fungal pathogens associated with C. sativa (hemp) cultivated in Northern Thailand. During our investigation, we have found more than 10 pathogenic fungi, including one species that was found to be different from extant ones, and deemed to be a novel species. The objectves of this study were: (1) to characterize the novel isolate through detailed morphological descriptions; (2) to determine its phylogenetic position within the genus Fusarium by analysing DNA sequences from regions of the tef-1α, rpb2, and cal genes; and (3) to access its pathogenicity. The resulting data provides a framework for accurate quarantine screening and field diagnostics, supporting improved management of Fusarium in hemp production.

Symptomatic parts of C. sativa (hemp) stems were collected from an outdoor plantation of Mae Sa Noi Village, Mae Rim District, Chiang Mai Province, Northern Thailand (18°51'50.06" N, 98°50'45.57" E; elevation 1,285 m/4,216 ft). Ten symptomatic stems were randomly collected from the site. The samples were individually kept at 4 °C in sterile polyethene bags and processed in the laboratory within 24 h of collection. The single-spore isolation method on water agar (WA) was used to obtain pure cultures for subsequent molecular work, following Senanayake et al.^[21]. All pure cultures were maintained at the Sustainable Development of Biological Resources Laboratory, Faculty of Science, Chiang Mai University, Chiang Mai, Thailand (SDBR-CMU), and the Kunming Institute of Botany Culture Collection, Kunming, China (KUNCC). One-month-old cultures on potato dextrose agar (PDA) were used to prepare dried fungal cultures, which were deposited as herbarium specimens in the Herbarium of the Department of Biology, Faculty of Science, Chiang Mai University, Chiang Mai, Thailand (CMUB).

Morphological study

The morphological analysis of Fusarium isolates was conducted as described by Wang et al.^[22]. Cultural characteristics, including colony shape, colour, and odour, were noted during a 7-d incubation in the dark at 30 °C on PDA, oatmeal agar (OA), yeast extract agar (YEA), V8 juice agar (V8), and synthetic nutrient-poor agar (SNA). Colony characteristics were further observed after incubation on PDA at 30 °C for 15–30 d until sporulation. Phenotypic characteristics such as colony shape, size, colour, exudates, colony margins, spore colour and size were observed, recorded, and photographed using a Nikon ECLIPSE Ni-U compound microscope equipped with a Nikon DS-Ri2 camera^[23]. For each anatomical structure (such as conidia, conidiophores, conidiogenous cells, and chlamydospores), at least 50 measurements were made using the Tarosoft (R) Image Framework program.

DNA extraction, PCR amplification, and sequencing

Genomic DNA from fungal mycelia was extracted using the Biospin Fungus Genomic DNA Extraction Kit (BioFlux, Hangzhou, China) based on manufacturer's protocols. The translation elongation factor-1 alpha (tef1-α), RNA polymerase second largest subunit (rpb2) and calmodulin (cal) were amplified by the polymerase chain reaction (PCR) using the primer pairs EF1/EF2 (EF1: 5'-A TGGGTAAGGAGGACAAGAC-3', EF2: 5'-GGAGGTACCAGTGATCATGTT-3')^[24,25]; RPB2-5f/RPB2-7cR (RPB2-5f: 5'-GGGGWGAYCAGAAGAAGGC-3', RPB2-7cR: 5'-CCCATRGCTTGYTTRCCCAT-3')^[26], and Cal-228F/Cal-2Rd (Cal-228F: 5'-GAGTTCAAGGAGGCCTTCTCCC-3', Cal-2Rd: 5'-TGRTCNGCCTCDCGGATCATCTC-3')^[25]. PCR amplification was performed in a 25 μL reaction volume, containing 12.5 μL Master Mix (mixture of EasyTaqTM DNA Polymerase, dNTPs, and optimised buffer; Beijing TransGen Biotech Co., Ltd., Chaoyang District, Beijing, China), 8.5 μL of double-distilled water (ddH₂O), 2 μL DNA template, and 1 μL each forward and reverse primer (10 μM). The PCR thermal cycle program used for gene amplification is shown in Table 1. Purification of PCR products and DNA sequencing were conducted by TsingKe Company (Kunming City, Yunnan Province, China).

Phylogenetic analyses

The reverse and forward sequences were combined and examined using BioEdit v7.2.5^[27]. The sequences were BLASTn searched in GenBank (https://blast.ncbi.nlm.nih.gov/) to identify taxa with the highest level of similarity (Table 2). The sequence alignment was completed using the MAFFT online server (https://mafft.cbrc.jp/alignment/server/index.html)^[28] and minor modifications were performed in BioEdit v7.2.5^[27] as needed. Maximum likelihood analysis (ML) was done on the CIPRES Science Gateway v3.3 (www.phylo.org/portal2)^[29] using RAxML-HPC2 on XSEDE (8.2.12)^[30] with the GTRGAMMA substitution model with 1,000 bootstrap iterations. The best-fit substitution models were evaluated using MrModeltest v2.3^[31]. Bayesian Inference (BI) analysis was carried out using MrBayes on XSEDE v3.2.7a^[32−34] using the same platform^[29]. Six Markov chains were run continuously for 10 million generations and automatically terminated when the standard deviation of split frequency exceeded 0.01. The trees were sampled every 100^th generation, and the first 25% of the sampled trees were eliminated during the burn-in phase of the analysis, which was assessed using Tracer v1.7^[35]. The phylograms were shown and changed using FigTree v1.4.0^[36] and Adobe Illustrator version 24.3 (Adobe Systems, San Jose, CA, USA).

Pathogenicity tests

Fungal inoculum preparation

The pathogenicity test was performed using the Fusarium flavens (SDBR-CMU869 = KUNCC 25-20161) for Koch's postulates. The fungal isolate was grown on PDA plates and incubated at 25 ± 2 °C for one week. To prepare for the suspension, 15 mL of sterile distilled water (DW) was added to PDA plates containing F. flavens. The spores were gently dislodged using a sterile rod, the resulting liquid was filtered through four layers of sterile gauze, and collected in a 50 mL tube. The filtrate was then centrifuged at 900 rpm, 4 °C for 10 min. After discarding the supernatant, the pellet was resuspended in 20 mL of sterile distilled water, centrifuged again, and decanted. The final pellet was suspended in 15 mL of sterile distilled water, and the spore concentration was determined using a hemocytometer under the Olympus light microscope (Olympus CH30RF200, Japan), then adjusted to 10⁶ spores·mL⁻¹.

Detached leaf and stem assay procedure

Before being tested, asymptomatic plant tissues (leaves and stems) were collected from a C. sativa (hemp) plantation (45-day-old plants), kept in sterile plastic bags, and carried to the laboratory. Leaves and stems were sterilized with a 0.2% sodium hypochlorite (NaOCl) solution (v/v) for 3 min, washed three times with DW, and air-dried. After that, a uniform wound (3 pores, 1 mm in width) was made at the equator of each leaf and stem using an aseptic needle. To fulfil Koch's postulates, inoculations were performed using a mycelial plug and a spore suspension. A 5-mm-diameter plug of F. flavens was taken from the actively growing edge of the fungal colonies and placed onto the surface of each healthy plant tissue. Sterile PDAC (potato dextrose agar supplemented with chloramphenicol) plugs served as negative controls. The inoculated tissues were positioned on a sterile plastic net inside sterile 90 mm petri dishes, which contained moistened sterile paper towels soaked with 1 mL of DW to maintain humidity. The 10 µL of spore suspension was evenly dropped into healthy plant tissues^[14]. Three replications of each treatment were performed. The samples were incubated at 25 ± 2 °C for 3–7 d until disease symptoms appeared. After the appearance of symptoms, the plates were examined for fungal growth and symptom progression.

Detached seedlings assay procedure

To perform a pathogenicity test on hemp seedlings, spore suspensions were prepared from cultures of F. flavens as previously described. Hemp seeds were surface-sterilized with 2.5% NaOCl solution for 5 min, washed three times with DW, and then germinated on moist filter paper in sterile plastic boxes for 3–5 d at 25 ± 2 °C. The seedlings were transferred into pots (7 cm high × 8 cm wide) containing a sterilized mixture of vermiculite : perlite : peat moss (1:1:2) for 2 weeks under greenhouse conditions. Fifteen two-week-old hemp seedlings were gently washed with DW to remove debris and then soaked in the spore suspensions, while the control seedlings were soaked in DW for 30 min. After inoculation, each seedling was transplanted into a pot (14.5 cm high × 14 cm wide) containing 800 g of sterilized soil^[14]. The plants were cultivated at ambient temperatures ranging from 25 to 34 °C and substrate moisture content within a range of 58% to 81% for 30 d. The pot experiment was conducted in a greenhouse located at the Faculty of Science, Chiang Mai University, Thailand. Throughout the experiment, plants were monitored for wilt symptoms, such as slow growth and yellowing. The level of disease was assessed based on plant growth rates and categorized into four classes: mild (1%–25%), moderate (26%–50%), severe (51%–75%), and very severe (76%–100%). Disease incidence (DI) was calculated as the percentage of diseased plants using the formula: DI (%) = (Number of diseased plants/Total number of plants) × 100. Each isolate was tested in triplicate to ensure consistency and reliability.

To fulfill Koch's postulate, fungi associated with disease symptoms were re-isolated from infected detached tissues and seedlings. The re-isolated fungi were identified based on morphological characteristics and DNA sequence analysis, thus fulfilling Koch's postulates.

The BLAST results revealed that our isolate had the highest similarity to members of the Fusarium incarnatum-equiseti species complex. Therefore, this complex was selected for detailed phylogenetic analysis. The combined sequence dataset of tef1-α, rpb2, and cal comprised 96 Fusarium strains in F. incarnatum-equiseti species complex (FIESC), and one outgroup taxon, F. concolor (NRRL 13459) (Fig. 1). The aligned dataset consisted of a total 2,954 characters with gaps (tef1-α: 1–623 bp, rpb2: 624–2,356 bp, cal: 2,357–2,954 bp). The best-scoring maximum likelihood (RAxML) tree was selected to represent the relationships among taxa, with a final likelihood value of –15,909.427308. The matrix contained 963 distinct alignment patterns, with 24.92% undetermined characters or gaps. Estimated base frequencies were as follows: A = 0.260395, C = 0.258686, G = 0.244464, and T = 0.236456; substitution rates were AC = 1.412250, AG = 4.053456, AT = 1.124038, CG = 0.659120, CT = 8.715739, and GT = 1.000000; gamma distribution shape parameter α = 0.305373, Tree Length = 1.140524. For Bayesian inference (BI) analysis, the best-fitting model of each locus and the final average standard deviation of split frequencies were determined as follows: the average standard deviation of split frequencies was 0.009952, while the best-fit substitution models were GTR + I + G for tef1-α, GTR + I + G for rpb2, and SYM + G for cal. Similar tree topologies were acquired from ML and BI analyses. In the phylogenetic tree (Fig. 1), the new species, F. flavens (SDBR-CMU869 = KUNCC 25-20161), formed an independent branch basal to F. caulendophyticum (CGMCC 3.25474 and GUCC 191050), and F. weifangense LC 18333 with 72% ML and 1.00 BYPP support (Fig. 1).

Figure 1. 

Phylogram constructed by maximum likelihood (ML) analyses of the combined tef1-α, rpb2, and cal sequence dataset of Fusarium. Bootstrap support valued for ML and Bayesian posterior probabilities (BYPP) equal to or greater than 70%, and 0.90, are shown above the nodes as ML/BYPP. The tree is rooted to F. concolor (NRRL 13459). The newly generated strain is indicated by black bold. T indicates the ex-type strains.

Taxonomy

Fusarium flavens. Ei, Phookamsak, Monkai & S. Lumyong, sp. nov. (Fig. 2)

Figure 2. 

Fusarium flavens (SDBR-CMU869 = KUNCC 25-20161, ex-type). (a)–(e) Seven-day-old culture on PDA, SNA, YEA, V8 juice agar, and OA incubated at 30 °C (top and reverse). (f) Natural symptoms of stem disease on Cannabis sativa (red arrow indicates the symptoms). (g), (h) Sporodochial conidiophores. (i) Phialides. (j)–(l) Macroconidia found on PDA media incubated at 30 °C. Scale bars: (g)–(i) = 100 μm; (k), (l) = 10 μm.

Index Fungorum number: IF 904383.

Etymology—name refers to the yellowish colour of cultures on the PDA and V8 juice agar.

Holotype—(CMUB 40081) isolated from Cannabis sativa (hemp) stems, Chiang Mai, Thailand.

Pathogenic fungus occurring on the stems of Cannabis sativa (hemp) in Northern Thailand. Sexual morph: not observed. Asexual morph: Sporodochia not observed. Conidiophores macronematous and developed on the aerial mycelium. Aerial phialides subulate to subcylindrical, smooth- and thin-walled, 6–33 × 2–5 µm ($\overline {\rm x} $ = 20.9 × 3.4 µm, n = 15). Microconidia not observed. Macroconidia formed by phialides on aerial mycelium, falcate, curved dorsiventrally with almost parallel sides tapering slightly towards both ends, 2–5-septate, hyaline, smooth- and thin-walled; 21–47 × 2–5 µm ($\overline {\rm x} $ = 34.1 × 4.2 µm, n = 30). Chlamydospores not observed.

Culture characteristics: Fusarium flavens (SDBR-CMU869 = KUNCC 25-20161) grew in darkness at 30 °C after 7 d on PDA, SNA, YEA, V8, and OA. Colonies on PDA reached 80 mm in diam. after 7 d of incubation, with abundant mycelial growth, the colony surface turned to creamy white, reverse pale yellow. Colonies on SNA reached 82 mm in diam. after 7 d of incubation, colonies are membranous to woolly, white, with abundant sporulation on the surface, giving a powdery appearance; reverse colourless. Colonies on YEA reached 82 mm in diam. after 7 d of incubation, white, with sparse to moderate aerial mycelium, slightly raised center, and the same reverse colour. Colonies on V8 juice agar medium reached 80 mm in diam. after 7 d of incubation, pale orange with floccose mycelium, reverse saffron yellow. Colonies on OA reached 83 mm in diam. after 7 d of incubation, white to orangish-white, dense and woolly, resembling a cotton-like mass, reverse creamy white colour (Fig. 2).

Material examined—Thailand, Chiang Mai Province, Mae Rim District, Mae Sa Noi Village, unhealthy stems of hemp (Cannabis sativa, Cannabaceae), 18°51'50.06" N, 98°50'45.57" E; elevation 1,285 m/4,216 ft, 12 December 2023, Toe Swe Zin Ei, PS4 (CMUB 40081, holotype), ex-type, SDBR-CMU869 = KUNCC 25-20161.

Notes: According to nucleotide BLAST search, the tef1-α, rpb2, and cal sequences of the new isolate Fusarium flavens (SDBR-CMU869 = KUNCC 25-20161) showed the highest similarity to Fusarium sequences available in GenBank. Details of the closest BLAST matches and percentage identities are presented in Table 3. Our phylogeny depicts a close association of F. flavens with F. humuli, as well as F. weifangense and F. caulendophyticum (Fig. 1). The nucleotide comparison among our new isolate (SDBR-CMU869 = KUNCC 25-20161), and the ex-type strains of F. caulendophyticum, F. weifangense, and F. humuli indicated reasonable nucleotide differences to justify the establishment of a novel species as recommended by Jeewon & Hyde^[37] (Table 4).

Fusarium caulendophyticum, described from healthy stems of Rosa roxburghii in Guizhou Province, China^[38], possesses irregularly branched conidiophores with lateral and terminal mono or polyphialides. In contrast, F. flavens develops simpler, subulate to subcylindrical aerial phialides, with larger conidiogenous cells (6–33 × 2–5 vs 10–24 × 2–3 μm) and larger macroconidia (21–47 × 2–5 μm vs 16–27 × 2.5–4 μm),^[38] compared with those in F. caulendophyticum.

Fusarium weifangense, isolated from symptomatic tissues of Triticum aestivum in Shandong Province, China^[39], forms densely branched sporodochial conidiophores with apical whorls of 3 to 4 phialides. In contrast, F. flavens produces macronematous aerial conidiophores with subulate to subcylindrical phialides and smaller conidia, compared with those in F. weifangense (21–47 × 2–5 µm vs 26.5–49.4 × 4.1–7.1 µm)^[39].

Fusarium humuli, described from leaves of Humulus scandens in Jiangsu Province, China^[22], produces pale orange sporodochia with densely packed, verticillately branched conidiophores bearing small subulate to subcylindrical phialides (6.3–11.9 × 2–3.4 µm). In contrast, F. flavens forms only aerial conidiophores with larger phialides (6–33 × 2–5 µm) and larger conidia (21–47 × 2–5 μm vs 21–35 × 2–3 µm)^[22]. Based on these morphological differences and phylogenetic evidence, F. flavens (SDBR-CMU869 = KUNCC 25-20161) is introduced here as a novel species.

Pathogenicity test on detached tissues

The pathogenicity test of F. flavens (SDBR-CMU869 = KUNCC 25-20161) on hemp was carried out using a detached leaves and stems assay (Fig. 3a–h). After inoculation with the fungal mycelium plug, the symptoms of brown lesions developed rapidly and spread onto the leaves and stem surface after 3 d of incubation (Fig. 3b, d). In contrast, control leaves and stems receiving treatment with sterile PDAC plugs remained healthy and symptom-free, as shown in (Fig. 3a, c). In the spore suspension inoculation method, brown lesions and wilting developed and spread across the surface of the leaves and stems (Fig. 3f, h). Conversely, the control leaves and stems receiving DW remained healthy and showed no disease symptoms (Fig. 3e, g).

Figure 3. 

Pathogenicity tests on detached hemp leaves, stems, and seedlings. (a) Control leaves, and (c) stems, inoculated with a PDAC plug. (b), (d) Symptoms after being inoculated with Fusarium flavens (SDBR-CMU869 = KUNCC 25-20161) mycelium plug for 7 d. (e) Control leaves, and (g) stems, inoculated with distilled water (DW). (f), (h) Symptoms after being inoculated with F. flavens (SDBR-CMU869 = KUNCC 25-20161) spore suspension for 7 d. Inoculation of hemp seedling control with (i) DW, compared with (j) that inoculated with F. flavens (SDBR-CMU869 = KUNCC 25-20161) spore suspension. (k) The roots of a plant inoculated with DW (control) shows no symptoms. (l) The roots of a plant inoculated with F. flavens (SDBR-CMU869 = KUNCC 25-20161) spore suspension shows root rot symptoms (30 d).

Pathogenicity test on detached seedlings

In the hemp seedling test, Fusarium flavens was also evaluated for its ability to cause disease (Fig. 3i–l). Four weeks after inoculation, seedlings inoculated with F. flavens exhibited slower growth compared to uninoculated controls. Slow development of symptoms was observed in approximately 73% of the seedling inoculated with F. flavens, respectively. The average plant height of control and inoculated plants is presented in Table 5. The remaining plants showed reduced and stunted development (Fig. 3i, j [right], and l), while the control, non-inoculated, saline-watered seedlings continued to grow healthily (Fig. 3i, j [left] and k). The results of isolated pathogenic strains of F. flavens caused varying symptoms in infected leaves, stems, and seedlings. The fungi were re-isolated from infected plant parts for Koch's postulate assays, and were confirmed based on their morphological characteristics that this fungus was the disease-causative agent.

The identification of Fusarium species, particularly within the Fusarium incarnatum–equiseti species complex (FIESC), requires a combined taxonomic approach integrating both morphological and molecular phylogenetic analyses. Morphological traits, such as colony growth, spore type, size, and septation are important but often insufficient due to overlapping characteristics and cryptic variation among closely related species^[22]. Ribosomal DNA markers (ITS and LSU) are widely used, but generally lack the resolution to distinguish among cryptic taxa^[40−43]. Therefore, protein-coding genes, including tef1-α, rpb2, and cal, are regarded as more informative and are increasingly adopted as standard markers for species delimitation in Fusarium^[38,44].

In practice, accurate identification involves combining morphological examination with multigene phylogenetic analysis, which provides robust species boundaries and resolves hidden diversity. This approach is particularly important in FIESC, which contains numerous cryptic species and newly described taxa^[38,44]. In the present study, Fusarium isolates were recovered from hemp stem rot in Northern Thailand. Based on both morphology and multigene phylogeny (tef1-α, rpb2, and cal), one isolate was determined to represent a novel species, F. flavens (SDBR-CMU869 = KUNCC 25-20161), in accordance with established taxonomic practices^[38,44]. This study aimed to isolate and identify fungal pathogens associated with C. sativa, and to confirm their pathogenicity by fulfilling Koch's postulates using detached plant tissues and attached seedlings. The causal agent of stem rot, F. flavens, was found on hemp stems. Several Fusarium species are significant pathogens of C. sativa in North America. For instance, F. falciforme has been identified as the causal agent of root rot in the United States^[45]. In Israel, soil-borne species such as F. oxysporum, F. proliferatum, and F. solani have been associated with disease development in cannabis^[14]. Additionally, F. proliferatum has been documented as the cause of stem rot in Canada^[46]. During our taxonomic investigation of pathogenic fungi, several Fusarium species were identified based on morphological characteristics, nucleotide pairwise comparisons, and phylogenetic evidence. One of them, F. flavens, a novel species, is reported here for the first time on C. sativa. The novel species, F. flavens was found to be phylogenetically related to F. caulendophyticum and F. weifangense (Fig. 1). However, the phylogenetic placement of F. flavens received relatively moderate bootstrap support (Fig. 1), indicating the need for further investigation. However, the independent lineage and the nucleotide differences support the establishment of a new species. In this study, pathogenicity was confirmed using both detached tissues and seedling tests, demonstrating that the F. flavens isolate obtained from hemp stems can also cause both stem and root rot symptoms. In pathogenicity tests, brown lesions and wilting developed in detached tissues, while seedling inoculations resulted in root rot symptoms. The severity varied among test conditions, indicating that the pathogenic potential of this isolate may be influenced by environmental factors.

In Thailand, Fusarium species have been documented to affect a variety of crops. For example, F. proliferatum, F. thapsinum, and F. verticillioides have been found on sorghum^[47]. Additionally, F. oxysporum has been reported in banana farms in Chiang Rai, causing significant crop losses and posing a serious threat to banana production^[48]. Moreover, two species of Fusarium, F. citrullicola and F. melonis, have been identified from rot lesions on watermelon and muskmelon fruits^[49]. Recent studies have also discovered three novel causal agents of muskmelon fruit rot: F. compactum, F. jinanense, and F. mianya^[50]. A recent study by Samarakoon et al.^[51] described F. pernambucanum as a causal agent of bud rot of hemp in northern Thailand. There is a need to do further research on this genus in Thailand, given its potential to cause diseases in many economically important crops.

In conclusion, these findings highlight the presence and potential impact of Fusarium species in C. sativa cultivation in Chiang Mai Province, Northern Thailand. The confirmation of their pathogenicity provides a valuable basis for future research for effective disease management, and the development of resistant cultivars. Overall, this study expands our understanding of the fungal diversity and pathogenic threats affecting C. sativa in Thailand and underscores the need for ongoing monitoring and integrated management strategies to reduce potential economic losses in hemp production.