SHORT COMMUNICATIONOpen Access icon OPENOpen Access license

Hemp Abnormal Growth Is Attributed to Mono-, Di-, or Tri-Infections of Spiroplasma citri, ‘Candidatus Phytoplasma trifolii’, and Beet Curly Top Virus

    Authors and Affiliations
    • Jennifer L. Schoener
    • Shouhua Wang
    1. Plant Pathology and Molecular Diagnostic Laboratory, Nevada Department of Agriculture, Sparks, NV


    As hemp (Cannabis sativa) emerges as a commercial crop in Nevada, mild to severe abnormal growth has been observed from many plants in commercial fields throughout the growing season. Affected plants exhibited a group of symptoms typically including stunting, leaf yellowing, excessive apical branching, clusters of witches’ broom, leaf rolling upwards, and leaf mottling and mosaic. These symptoms mostly showed up together in a plant or a crop and are defined here as abnormal growth syndrome (AGS). Between 2017 and 2019, the Nevada State Plant Pathology and Molecular Diagnostic Lab received 67 symptomatic hemp samples from Nevada for diagnosis, and ‘Candidatus Phytoplasma trifolii’ was detected in 14 samples (21%). To investigate additional biotic agents associated with AGS, PCR products generated by primers P1/P7 were cloned into a pGEM-T vector and sequenced, and Spiroplasma citri DNA was found in two samples (3%). All 67 DNA samples were further tested for beet curly top virus (BCTV), and 57 samples (85%) were found to be infected by BCTV. Twelve samples (18%) were coinfected by both ‘Ca. Phytoplasma trifolii’ and BCTV, and two (3%) were coinfected by S. citri, ‘Ca. Phytoplasma trifolii’, and BCTV. The findings suggest that BCTV is the most prevalent pathogen causing the hemp abnormal growth in Nevada, but ‘Ca. Phytoplasma trifolii’ and S. citri may also contribute to the severity and complexity of symptoms. Thus, hemp abnormal growth can be attributed to single, dual, or triple infections of these three leafhopper-vectored mollicutes and virus.

    Copyright © 2023 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.

    Congress first approved the cultivation of hemp (Cannabis sativa) under the Farm Bill Section 7606 of the Agricultural Act of 2014. Since then, hemp has emerged as a commercial crop in Nevada and other parts of the United States. Accordingly, many diseases, primarily caused by fungi and oomycetes, have been found in hemp crops, with some pathogens having destroyed an entire crop (Schoener et al. 2017; Wang 2021).

    The abnormal growth of hemp became a major detriment for some hemp growers during the early years of hemp production, particularly in certain CBD strains that suffered severe damage and yield losses. It appeared to be widespread in Nevada hemp crops (Feng et al. 2019; Schoener and Wang 2019), and it has also been a significant issue for hemp production in other western states, including California (Melgarejo et al. 2022), Oregon (Rivedal et al. 2022), Arizona (Hu 2021; Hu et al. 2021), and Colorado (Giladi et al. 2020). Similar symptoms were also noted in some hemp crops in Maryland (Behnam Khatabi, personal communication). Most reports suggested that hemp plants with abnormal growth were infected by a single pathogen, predominantly beet curly top virus (BCTV), followed by ‘Candidatus Phytoplasma trifolii’, whereas other reports suggested that mixed infection occurred in hemp by multiple viruses and viroids (Chiginsky et al. 2021; Jarugula et al. 2023) or by different strains of BCTV (Melgarejo et al. 2022). All these reported detections provided a better understanding of the etiology for a broad range of symptoms observed from hemp crops.

    However, positive detections of a specific pathogen may not explain the actual or complete causes of hemp abnormal growth, especially when a pathogen is only detected from a certain percentage of plants with similar symptoms. Furthermore, the expression of abnormal growth in hemp caused by different pathogens may vary among plants or varieties, and the symptoms caused by viruses, viroids, or mollicutes may overlap or be indistinguishable. Between 2015 and 2022, the Nevada State Plant Pathology and Molecular Diagnostic Lab received over 297 hemp and cannabis samples submitted from 11 counties for cause diagnosis, with 146 (49%) expressing severe abnormal growth symptoms. Field visits to the sites from which samples were collected revealed that certain fields were seriously impacted (Schoener and Wang 2019). The symptoms were classified into four categories: (i) severe stunting and leaf yellowing (Fig. 1A); (ii) leaf mottling, chlorosis, or mosaic (Fig. 1B); (iii) leaf rolling upward and/or twisting (Fig. 1C and F-1); and (iv) witches’ broom-like excessive apical branching (Fig. 1D and E). These symptoms were commonly expressed together in a plant (Fig. 1F) and simultaneously occur in a crop with a commonality of plant growth alteration. Therefore, we define these symptoms as abnormal growth syndrome (AGS).

    FIGURE 1

    FIGURE 1 Symptom components of hemp abnormal growth syndrome (AGS) observed in production fields. A, Plant stunting with leaf yellowing. B, Leaf chlorosis, mottling, and mosaic. C, Leaf curling upward. D, Witches’ broom with excessive apical branching. E, Magnified view of a cluster under a stereo microscope. F, A plant with AGS exhibiting all components of symptoms shown in A to D, specified by arrows: 1, Leaf curling and twisting. 2, Witches’ broom. 3, Leaf mottling and mosaic. 4, Leaf yellowing and plant stunting.

    Download as PowerPoint
    FIGURE 2

    FIGURE 2 Detections of beet curly top virus (BCTV) and ‘Candidatus Phytoplasma trifolii’ from hemp plants exhibiting abnormal growth syndrome. A, Positive amplifications of BCTV DNA (1,133 bp) from 11 symptomatic hemp plants (3 to 13) and no detection in 2 asymptomatic hemp plants (1 and 2) using primers BGv377/BGc1509 (1:10 dilution of sample DNA). Plants #8 and 12 had visible bands in the PCR run with an undiluted sample DNA (not shown). B, Positive amplification of approximately 1.8-kb fragments of ‘Ca. Phytoplasma trifolii’ DNA (11 samples) and Spiroplasma citri DNA (samples #57 and 129) from symptomatic plants using P1/P7 primers. C, Positive amplification of a 1,250-bp fragment of ‘Ca. Phytoplasma trifolii’ DNA from 11 symptomatic plants in a nested PCR using primers P1/P7 (first run) and R16F2N/R16R2 (second run). L = exACTGene 1-kb DNA ladder (Thermo Fisher Scientific, Waltham, MA). C = Non-template control. Numerical labels = A case number assigned to each plant sample.

    Download as PowerPoint

    Initially, we examined each hemp sample with AGS received from 2017 to 2019 and whole plants in fields to better understand the symptomatology. One of the most striking symptoms observed was the excessive apical branching and crowded growth of underdeveloped leaves, giving a shoot proliferation or clusters of witches’ broom appearance. Leaves in the cluster were yellowish, often distorted, and small (Fig. 1D and E). These symptoms suggest a phytoplasma infection. To confirm this hypothesis, we took a subsample (200 mg) of midvein and petiole tissue from five to six leaves of each symptomatic plant and performed a total DNA extraction using the Qiagen DNeasy Plant Mini Kit (Qiagen INC, Germantown, MD). A duplex real-time PCR assay was performed using phytoplasma-specific primers JH-F1/JH-F all/JH-R and probe JH-P uni (Hodgetts et al. 2009) to detect phytoplasma DNA. Plant specific primers (18S F/18S R) and a probe (18S P) were used to detect plant DNA as an internal control (USDA-PPQ PPCDL 2016). Both detections were done in the same reaction. The sequences of the primers and probes with 5′ fluorescent dye and 3′ or internal quencher are listed in Table 1. The reagents and volumes used for the qPCR reaction were as follows (per sample): 15.15 µl of molecular-grade water, 2.5 µl of Platinum Taq buffer (10×), 3.0 µl of MgCl₂ (50 mM), 0.5 µl of dNTP mixture (10 mM), 0.75 µl of primer mix (JH-F 1/JH-F all/JH-R) (10 µM), 0.25 µl of JH-P uni (10 µM), 0.5 µl of primer mix 18S F/18S R (2 µM), 0.25 µl of 18S P (2 µM), 0.1 µl of Platinum Taq polymerase, and 2 µl of sample DNA or controls. A DNA sample was considered positive for ‘Candidatus Phytoplasma’ DNA if (i) its FAM Ct value was between 13 and 38, (ii) the plant DNA had a positive TET Ct value, and (iii) the positive control (phytoplasma DNA from a pear tree used in this test) had a positive FAM Ct value (USDA-PPQ PPCDL 2016). Phytoplasma DNA was detected in 14 of 67 symptomatic samples, with Ct values ranging from 14 to 33.

    TABLE 1 DNA sequences of primers and probes used in this study

    To confirm the detection and identify the species of phytoplasmas by their DNA sequences of the 16S rRNA gene, conventional PCR was then performed for all qPCR-positive DNA samples using primer pair P1/P7 (Smart et al. 1996). The reagents and volumes used for the PCR reaction were based on the Platinum Taq DNA Polymerase user guide (Thermo Fisher Scientific, Waltham, MA). PCR was run in a thermocycler with an initial denaturation at 94°C for 2 min followed by 35 cycles of 94°C for 1 min, 55°C for 2 min, and 72°C for 3 min, with a final extension stage at 72°C for 10 min. Out of 14 qPCR positives, 10 samples had strong amplification of approximately 1.8-kb fragments, whereas others had relatively weak or double bands (Fig. 2B). In samples #50, 56, and 57, an approximately 2.0-kb fragment was also amplified, which may be a signal of amplification of DNA from a related organism by the universal primers P1/P7. Then, we ran a nested PCR for all qPCR-positive samples with a second primer pair R16F2n /R16R2 (Gundersen and Lee 1996). To do this, the P1/P7 PCR product was diluted 10 times, and 1 µl of the diluted product was used for the R16F2n /R16R2 PCR run. The master mix formula and PCR parameters were the same as the P1/P7 run, except this time using R16F2n /R16R2n primers. The nested PCR generated a 1,250-bp fragment in all 14 qPCR-positive samples (Fig. 2C).

    Next, we attempted to examine PCR products generated by P1/P7 and R16F2n /R16R2 because both primer sets are universal to detect various species or types of mollicutes. Following the manufacturer's instruction, we purified each PCR product (gel band) using the Qiagen MinElute Gel Extraction Kit and subcloned it into a pGEM-T vector using the pGEM-T vector system (Promega, Madison, WI). Transformed colonies were cultured in 3 ml of super broth overnight, and their plasmid DNA was extracted using the Qiagen Plasmid Miniprep Kit. For each PCR product's transformation experiment, we picked 5 to 10 colonies for plasmid extraction and sequenced the insert (PCR amplicon) using a T7 promoter primer and SP6 upstream primer (Table 1). Because both sequencing primers originate from the T-vector, we were able to obtain complete and accurate 5′ and 3′ ends of both P1/P7 and R16F2n /R16R2 amplicons. Both amplicons were fully assembled manually for accuracy with the aid of DNASTAR lasergene software (DNASTAR, Madison, WI). We analyzed over 140 sequencing reads and found that all 14 samples contain ‘Ca. Phytoplasma trifolii’ DNA that are almost identical, with only 18 nucleotide variants among P1/P7 fragments and 7 variants among R16F2n/R16R2 fragments. Complete 1,810-bp P1/P7 fragments (OQ597527 to OQ597531) and 1,250-bp R16F2n/R16R2 fragments (OQ597719 to OQ597726) were deposited into GenBank. This result was consistent with the report of Feng et al. (2019), which was based on hemp samples submitted from Nevada. Interestingly, we also found Spiroplasma citri DNA sequences (amplified by P1/P7 primers) from four clones of two samples, whereas ‘Ca. Phytoplasma trifolii’ DNAs were found in other clones of the same samples. This result suggests that P1/P7 can amplify the respective rDNA region of both ‘Ca. Phytoplasma trifolii’ and S. citri and demonstrates that both pathogens infected the hemp plants. Next, we attempted to obtain an S. citri full amplicon sequence generated by P1/P7. Unlike ‘Ca. Phytoplasma trifolii’ P1/P7 amplicons that were assembled using R16F2n/R16R2 fragments as a linker, S. citri P1/P7 amplicons only had 5′- and 3′-end sequences available from T7 promoter and SP6 upstream sequencing reads. We used the Primer3Plus program to pick seven primers from their 3′ and 5′ ends of sequencing reads and used them as sequencing primers to sequence those plasmid DNA containing the S. citri DNA insert. These sequencing reads provided sufficient overlapping for us to assemble a S. citri P1/P7 amplicon sequence (1,845 bp) from each clone, all of which were deposited into GenBank (OQ572327 to OQ572330).

    Not all symptomatic samples were positive for ‘Ca. Phytoplasma trifolii’ and/or S. citri, which implies that additional pathogens were associated with AGS. Therefore, we tested the same set of DNA samples extracted from the midvein and petiole tissue for BCTV using primers BGv377/BGc1509 (Soto and Gilbertson 2003). These primers amplified a 1,133-bp DNA fragment from the BCTV capsid protein gene, and 57 out of 67 samples tested had positive amplifications (Fig. 2A). To obtain a complete genome of BCTV, additional primers OutR/2213F (1,191 bp) and 1278F/2609R (1,340 bp) (Rondon et al. 2016; Strausbaugh et al. 2008) were used to amplify the rest of the genome from two samples coinfected with both ‘Ca. Phytoplasma trifolii’ and S. citri . Through the same strategy of cloning and sequencing, a complete genome of BCTV was obtained (GenBank OQ628293). The 2,932-bp complete genome is mostly identical to that of BCTV identified from hemp in Arizona (Hu et al. 2021), with only three base insertions at positions 16 and 1,327-1,328 (Jiahuai Hu, personal communication). Of these BCTV-infected samples, 12 were also infected with ‘Ca. Phytoplasma trifolii’ and two with both ‘Ca. Phytoplasma trifolii’ and S. citri. Thus, these samples were found to have been infected by at least one of these three pathogens, with BCTV being the most prevalent.

    Notably, these three pathogens detected from the phloem tissue of hemp plants are all transmitted by the phloem-feeding beet leafhopper, Circulifer tenellus (Hemiptera: Cicadellidae) (Bennett 1967). It is reasonable to speculate that the composition of these pathogens in hemp plants may largely depend on this insect vector population that feeds on the hemp crop, as they may carry one or more viruses or mollicutes and transmit them to hemp plants. Therefore, it is not a surprise that hemp plants are coinfected with S. citri, ‘Ca. Phytoplasma trifolii’, and BCTV if the leafhopper carries all these pathogens.

    S. citri is a pathogen causing citrus stubborn disease with typical symptoms of leaf mottling, stunting, and yield reduction (McNeil et al. 2023). It infects carrot, causing purple leaf discoloration, stunting, and fibrous secondary roots, with evidence of coinfections with S. citri and a phytoplasma in some symptomatic carrot plants (Lee et al. 2006). It coinfects with phytoplasma in sesame plants (Salehi et al. 2022). These reports suggest that coinfections with S. citri and phytoplasma occur commonly in crops, but the impact of coinfections on the host plant, especially in the symptom development, remains to be investigated. The positive finding of S. citri in hemp suggests that S. citri can infect hemp and potentially impact hemp growth. In fact, many hemp plants with AGS exhibit leaf mottling and mosaic. These symptoms are not well explained by phytoplasma and BCTV infections, and S. citri could be a potential etiological factor that contributes to those symptoms observed in hemp AGS. More importantly, it will stimulate further research on S. citri detection in hemp and on its impact on hemp growth with or without coinfections of other pathogens.

    Although we demonstrated that three unrelated pathogens solely or collectively infected hemp plants, it is not completely known how and to what extent each or any combination of them contribute to the symptoms we have observed in hemp plants. This can be answered through experimentally inoculating hemp plants with one or multiple pathogens. However, phytoplasmas are well known for their inducing characteristic symptoms, such as excessive shoot proliferation, witches’ broom, yellowing and stunting, and little leaf (Bertaccini 2022), which are all major symptoms of hemp AGS. Therefore, we believe that ‘Ca. phytoplasma trifolii’ is a key pathogen contributing to the AGS. Similarly, BCTV is a well-documented pathogen associated with crops in the western United States (Strausbaugh et al. 2008), and it was confirmed to be the cause of curly top disease in hemp in California (Melgarejo et al. 2022). These reports support that BCTV contributes to hemp AGS, especially for leaf curling, twisting, and stunting symptoms. In contrast to phytoplasma and BCTV, S. citri is less understood in terms of its impact on hemp growth. Nevertheless, we believe it plays a role in hemp AGS. The demonstration of coinfections in commercial hemp crops helps to understand this unique syndrome and suggests that hemp AGS may not be diagnosed under the “one pathogen/factor – one disease” concept. Additionally, growers may need to address multiple pathogens that are associated with various abnormal growth symptoms in hemp. Considering that all three pathogens reported here are transmitted by hoppers, the presence of each or combinations of these pathogens in a hemp crop may depend on viruliferous or mollicute-carrying populations of hoppers and other insect vectors in the environment of hemp production.

    The author(s) declare no conflict of interest.


    Funding: This work was supported in part by the funding from the Western Plant Diagnostic Network.

    The author(s) declare no conflict of interest.