A Guide to Cannabis Virology: From the Virome Investigation to the Development of Viral Biotechnological Tools

Abstract

1. History of Cannabis sativa Virology

The earliest observations of Cannabis sativa viral disease are reported in the 40s and 50s of the XX century in Europe [1,2] (Figure 1) when fiber and/or seed crops were widely cultivated. These first reports indicate two symptomatologic conditions that were correlated to supposed viral infections, never formally demonstrated: the hemp streak virus (HSV) and the hemp mosaic virus (HMV). However, these reports were purely descriptive of symptomatology since no pathogenicity test was ever performed, and the viral etiology was never confirmed. A comprehensive analysis of these conditions has never been conducted, probably because hemp cultivation faced a period of regression due to the development of synthetic fibers and the use of other plants as a source for vegetable oil production in later years. In 1970 Schmidt and Karl [3] identified three viruses able to infect C. sativa plants and cause symptoms: the cucumber mosaic virus (CMV), alfalfa mosaic virus (AMV), and hemp mottle virus. While CMV and AMV were confirmed as infectious and pathogenic in later studies, the latter was never reported again. In the 1970s, Hartowicz et al. (1971) [4] performed viral mechanical transmission in C. sativa using isolates of twenty-two different viruses, and half were proven to infect inoculated plants. While this work was presented at the “North Central Weed Control Conference”, to our knowledge, a comprehensive and peer-reviewed paper reporting the methodology alongside the results obtained has never been published. A later publication [5] mentions this work and lists those viruses that resulted in symptomatologies in inoculated plants: tobacco ringspot virus (TRSV), tomato ringspot virus (TomRSV), tobacco streak virus (TSV), and CMV have been associated with mosaic symptoms and dwarfism, while AMV and the so-called “eunoymous ringspot virus (ERSV)” were only associated with mosaic. Of these, only TSV, CMV, and AMV were reported again in later works on viruses that infect C. sativa, while the others were no longer mentioned.

No new manuscripts on the topic were then published until 1997, when Kegler and Sparr [6] performed inoculations of different viruses in several European industrial varieties for seeds and fiber production (hemp) in a controlled environment. After six weeks from the first inoculations, systemic leaves of infected hemp plants were used for re-inoculations on test plants (Nicotiana spp. and Chenopodium spp.) using serological tests to verify viral infections. In this way, it was possible to identify the viruses described in Table 1, including CMV and AMV, and describe a range of symptom severity in hemp (Table 2).

More than ten years passed until a new virus infecting hemp was accidentally identified in 2012 [9]. The new Partitiviridae identified was proven to be widely endemic since it was detected in almost all varieties tested, but no correlation with symptoms was observed, so it was reported as cannabis cryptic virus (CanCV) (Table 1). In a follow-up study, characterizing the virus in other hemp varieties, Righetti et al. (2018) [10] confirmed that the virus is not associated with any symptoms (Table 2) and described a variable viral load depending on the plant analyzed. Vertical transmission has been demonstrated crossing both infected male plants and healthy females and healthy males and infected females. The resulting progenies were found to be 100% positive. Otherwise, mechanical transmission in Chenopodium quinoa and Nicotiana benthamiana test plants was proven to be ineffective as was the transmission via cross-grafting between rootstocks and scions from infected and not-infected C. sativa plants. In fact, members of the Betapartitivirus genus have no movement proteins so, in plants, their tropism occurs only with cell division [16].

The hop latent viroid (HLVd) has recently been increasingly associated with severe cannabis symptoms [14,15,17] (Table 2) that impair seed and fiber production in industrial varieties and cannabinoid contents in medical varieties. Infected medical cannabis plants can suffer vigor loss and a 50% up to 70% reduction in THC content. Furthermore, a survey carried out in California on 200,000 samples verified the presence of HLVd in 90% of cannabis-growing facilities. The potential estimated losses can therefore reach up to USD 4 billion loss per year in the United States of America alone [17]. Infected mother plants for agamic propagation act as a reservoir for the viroid that easily spreads to the clonal progeny. Moreover, HLVd diffusion is higher than field inspections would indicate because symptoms develop only in the late stages of infection.

Where the decriminalization and legalization of C. sativa for medical (Israel) and recreational (in some USA states) use have taken place, cultivated areas have increased, leading to favorable conditions for the spread of new viral infections. Lettuce chlorosis virus (LCV) was identified as a causal agent of leaves yellowing and stunted growth in Israel [11] (Table 2), symptoms that were previously only associated with nutrition deficiencies. In the USA, apical leaf deformations have been correlated with the presence of beet curly top virus (BCTV) [12,13] (Table 2).

Recently, Pitt et al. (2022) [8] confirmed that potato virus Y (PVY) infects C. sativa and demonstrated that it can be transmitted by aphids in a non-persistent manner. This makes PVY-infected cannabis fields potentially dangerous from a phytosanitary point of view because of the possible spread of the virus to other crops.

C. sativa virome investigation has recently found a new boost thanks to high-throughput sequencing (HTS) techniques. Chiginsky et al. (2021) [7] performed an Illumina NextSeq500 sequencing of several C. sativa plants showing virus-like symptoms. While aiming to characterize different BCTV strains and describe their spread among symptomatic plants in Colorado (USA), several viruses have been identified (Table 1), some even never reported in C. sativa. The genome of a new strain of TSV was partially reconstructed, and grapevine line pattern virus (GLPV), citrus yellow vein-associated virus (CYVaV), and opuntia umbra-like virus (OULV) were identified and reported for the first time infecting cannabis. Recently, Jarugula et al. (2023) [18] reported co-infections of BCTV along with HLVd and CYVaV after canonical reverse transcription–PCR analysis and HTS in Washington State (USA).

All accession numbers of viruses that have been detected through HTS in cannabis are listed in Table S1 in the Supplementary material.

In recent years, in addition to expanding knowledge of the cannabis virome, cannabis virology has focused on exploiting viral vectors to compensate for the lack of established tools for in vivo functional genomic analysis. With the expansion of genomic and transcriptomic resources, significant advances have been made in the understanding of cannabis genetics allowing the identification of regulatory genes responsible for valuable medical or agronomical phenotypic traits [19,20,21,22,23]. However, reverse genetics studies are required to validate bioinformatic predictions or to conduct molecular characterizations of protein effectors. The basic requirement of reverse genetics is to possess tools for targeted gene editing to vary or abolish the expression of a particular gene or to obtain specific mutations to its product. The use of sequence-targeted DNA endonucleases underlies genome-editing technologies such as CRISPR/Cas9, which are widely used in other crops of agricultural interest [24,25,26]. The repair of DNA double-strand breaks by non-homologous end joining or homology-directed repair (in the occurrence of a donor template) makes it possible to modify a single DNA base pair, larger genomic portions, or the regulation of gene expression [27]. For this to occur, the CRISPR/Cas9 complex must be delivered or expressed in callus-forming cells (e.g., by Agrobacterium tumefaciens-mediated transformation), which will then be regenerated in an adult plant. However, cannabis plants are described as recalcitrant to callus regeneration even though they are effectively transformed by A. tumefaciens [28]. Indeed, the production of transgenic plants is still not easy to achieve [15], and despite their significant potential, transgenesis and genome editing have been successfully employed in only a few works [29,30]. In this context, virus-induced gene silencing (VIGS) represents an alternative system to genome editing for functional genomics studies because it allows the down expression of target genes through post-transcriptional gene silencing (PTGS). Viral vectors engineered from cotton leaf crumple virus (CLCrV) [31] and tobacco rattle virus (TRV) [32] have been successfully used for the purpose of harboring portions of C. sativa gene sequences in their genome and promoting the silencing of visible markers such as phytoene desaturase (PDS) and magnesium chelatase subunit I (ChlI). These promising results encourage the use of VIGS for rapid reverse genetic screening by silencing specific genes involved in the regulation of specialized metabolites, disease resistance, plant development, and abiotic stress in cannabis.

2. Categorizations of Viruses and Viroid According to the Level of Threat

Here, we propose a subdivision of viruses and viroids detected in C. sativa according to their threat level for cannabis cultivation, from the highest level A to the lowest level D (Table 2). The main pathological characteristics of each infectious agent were considered for classification into the four different threat levels: severity of symptom development and type of horizontal transmission mediated or not by vectors. Further criteria for classification among viruses is the natural occurrence of infection in cannabis by distinguishing infectious agents identified during field surveys from those inoculated artificially in lab environments (Figure 2). One should keep in mind that most of the available information on the biology of the viruses treated in this paper comes from studies conducted on hosts other than cannabis. Moreover, their role as causative agents of associated symptoms has never been formally demonstrated by means of infectious cDNA clones as suggested by Massart et al. (2017) [33] as an unequivocal method for defining the causal relationship between viral agents and diseases. Further studies would therefore be needed to characterize each member of the cannabis virome.

3. Fully Infectious Viral cDNA Clones and Derived Viral Vectors for Functional Genomic Studies in Cannabis

Viruses are natural vectors for delivering genetic material into cells, and, because of their ability to replicate and produce high quantities of virus-derived mRNA, for decades, they have been profitably used in fundamental virology studies or engineered as biotechnological tools. Several RNA or DNA phytoviruses were transformed into fully infectious cDNA clones and exploited as reverse genetics systems for the study of viral factors and host–pathogen interaction. Infectious cDNA clones can be easily modified and reproposed as vectors for the expression of heterologous proteins or to suppress host gene expression by inducing gene silencing (VIGS), aiming for the selective knock-down expression of endogenous genes in various plants [51,53,54].

No infectious cDNA clones derived from a virus that naturally infects cannabis has been used in this plant to date. However, two viral vectors derived from CLCrV [31] and TRV [32] have been developed to perform VIGS experiments in C. sativa.

DNA-A and -B of the bipartite CLCrV were cloned into two binary vectors for A. tumefaciens-mediated transformation, in the way that the genes of the two genomic DNAs were flanked by two common regions containing the origin of replication, and a multiple cloning site was inserted upstream AL3 gene to allow the introduction of foreign DNA sequences. Gene portions of 300 to 400 base pairs of phytoene desaturase (PDS) and magnesium chelatase subunit I (ChlI) were selected to track, through chlorophyll or carotenoids foliar bleaching, the occurrence of PTGS in the Finola variety. Specifically, the gene portions of PDS and ChlI were chosen using the method described by Xu et al. (2006) [55] to produce efficient siRNAs and, at the same time, with minimal off-target silencing effects. The VIGS capacity of CLCrV was then verified by analyzing the transcriptional levels of PDS and ChlI, which decreased by 73% and 70%, respectively, in agro-infiltrated cannabis plants. However, carotenoids and chlorophyll a and b levels decreased only between 27 and 40%causing faint green leaves with white and yellow spots and not a remarkable foliar photobleaching phenotype [31].

TRV-based vectors have been widely used for VIGS experiments in different plants in which the virus is infectious [56,57]. TRV is a bipartite RNA virus, which is why two different cDNA agroclones (pTRV1 and pTRV2) are used for VIGS: one expresses RNA1 that encodes the viral functions of replication and movement, while the other harbors RNA2 deprived of the 2b and 2c coding sequence and carrying an MCS downstream of the coat protein sequence. Alter et al. (2022) [32] inserted three different gene portions of PDS into pTRV2 and then performed VIGS examinations in medical cannabis genetic lines for cannabinoid productions. Despite the low ability of infectious TRV clones to induce systemic infection, observed mainly in glandular trichomes, intense and widespread symptoms of leaf photobleaching were observed in the MF-219 gene line, depending on the viral construct chosen, with transcript levels of PDS lower by more than 90% compared to control plants [32].

4. Concluding Remarks

Interest in C. sativa is rising in different economic sectors (e.g., medical, industrial, bio-building, cosmetic, food, etc.) [58,59], leading to higher acreage cultivation and to a higher exchange of plant materials for breeding purposes. In addition, plants for high-quality products (e.g., cannabinoids) are often obtained through agamic multiplication from mother plants that can potentially be infected. The combination of these conditions has facilitated the spread of viruses and viroids as evidenced by subsequent increases in published reports of viral infections (Figure 3), refuting the old assumption that cannabis is tolerant or even resistant to viruses [60]. The wide application of modern HTS will undoubtedly help to further characterize the cannabis virome by identifying unreported viruses and viroids that infect this plant. The ability of HTS to generate a large number of sequences and assemble multiple viral genomes from the same biological sample makes this technique an ideal solution for the generic identification of unknown or very different viruses. For the same reason, HTS can be used in viral population studies to analyze inter- or intra-host diversity and dynamics. However, the high investigative power of this technique alone is not sufficient to biologically characterize a pathogenic virus by distinguishing it from a commensal. Therefore, HTS studies must be followed by others aimed at determining the causal relationship between the symptoms described and the occurrence of the infectious agent [61]. This is essential to making effective management and risk assessment decisions. A phytosanitary harmonized diagnostic protocol for the early detection of viruses in C. sativa is not currently available, at least in Europe. Private companies can offer screening tests for the detection of the main viruses and viroids reported infecting this plant, but protocols are not shared and there are no standard or validated ones. Therefore, harmonization of protocols is highly needed to develop a reliable shared procedure for the early detection of known viruses, reducing the risk of circulation of infected plant materials. The proposed virus and viroid categorization system is intended to establish a guide for classification among viral infections in cannabis by indicating those that are the most threatening. Our opinion is that future breeding programs for the development of new varieties resistant or tolerant to viruses and viroids belonging to level A should be considered a priority.

From being considered as a possible phytosanitary threat to cannabis, viruses also represent an effective biotechnological tool for studies aimed at the genetic improvement of the plant. Indeed, the use of infectious cDNA clones of cannabis-infecting viruses allows the delivery of genetic material without the need for transformation and thus the creation of genetically modified plants. VIGS experiments have become a standard for studies of functional genetics through a reverse genetics approach. The main advantage of VIGS is that it is an economical and time-saving technique that ensures quick phenotype generation without the need for stable plant transformation. Therefore, even in C. sativa, it is possible to use infectious viral cDNA clones to induce the PTGS of target genes by overcoming the difficulty of applying other genomic editing techniques in this plant. Despite recent successes in applying VIGS in cannabis, several problems that may limit its use may remain. Indeed, the viral vector may alter the plant’s metabolism by distorting the resultant phenotype. Moreover, the insertion of exogenous sequences into the VIGS vector may limit the fitness of the recombinant virus, which can result in genomic deletions after a few replicative cycles. Despite these potential limitations, it is expected that the VIGS technique will increase its scope in the cannabis system by exploiting the ability of VIGS-mediated down expression to be inherited by the infected plant progeny through RNA-directed DNA methylation (RdDM) [62]. RdDM experiments are still in their very early days but may represent a future tool for the genetic improvement of cannabis through stable epigenetic modifications.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/v15071532/s1, Table S1: NCBI accession number of viruses and viroids that have been detected and sequenced in Cannabis sativa L.

Funding Statement

This research received no external funding.

Author Contributions

N.M. and M.D.: writing—original draft preparation, N.M., A.P., C.R., M.D. and P.C.: writing—review and editing, N.M. and A.P.: visualization. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

References