Introduction

The first description of a mosaic disease of tobacco (Nicotiana tabacum L.) (Mayer 1882) was swiftly followed by the discovery of filterable plant pathogens (Ivanovsky 1892) and the conception of the virus as a “contagium vivium fluidum” (Latin: contagious living fluid; Beijerinck 1898). Similar mosaic diseases were later observed on tomato plants (Solanum lycopersicum L.) in the USA, with the first description by Woods (1902). Many pathogens that cause mosaic diseases in tobacco and tomato in Europe and America were found to have common characteristics (e.g., Ainsworth 1933; Johnson 1926). In Japan, tomato plants were introduced during the seventeenth century and mainly cultivated for ornamental purposes until the nineteenth century. Tomatoes were first cultivated as garden vegetables in Japan at the beginning of the twentieth century (Kamimura 1980); mosaic diseases spread nationwide soon afterward (Murayama 1936; Nakata and Takimoto 1940).

Tomato mosaic diseases caused by tobamoviruses

Tomato-infecting tobamoviruses

In the 1920s and 1930s, tobacco mosaic virus (TMV)-like viruses were isolated from tobacco and tomato plants, and back-inoculated plants were found to reproduce mosaic symptoms. During this era, the identification and distinction of virus strains was based only on host susceptibility and symptoms (Stanley 1946), such that typical names included ordinary tobacco mosaic (Johnson 1926), aucuba mosaic of tomato (in distinction from aucuba mosaic of potato) (Smith 1928), and glasshouse streak (after symptoms observed in glasshouse-grown tomato plants) (Ainsworth 1933; Jerrett 1930).

Tobacco mosaic strains mechanically inoculated onto N. sylvestris or N. tabacum cultivar ‘White Burley’, which have the N′ resistance gene (N′ tobacco) caused systemic mosaic disease, whereas strains collected from tomato produced necrotic spots on inoculated leaves, with no systemic infection (Broadbent 1962; Kassanis and Selman 1947; Kunkel 1934). Later studies revealed that in the presence of the N′ gene, expression of the coat protein (CP) of the TMV tomato isolate, but not that of the tobacco isolate, induced a hypersensitive reaction that led to formation of local necrotic lesions and inhibition of virus spread (Knorr and Dawson 1988; Saito et al. 1987; Sekine et al. 2012).

Using this distinction method, Broadbent (1962) investigated TMV strains collected from tobacco and tomato plants in the UK. Among 64 isolates obtained from smoking tobacco plants, 60 exhibited responses similar to tobacco mosaic infection (i.e. systemic mosaic), whereas 187 of 203 strains collected from tomato plants showed responses similar to tomato mosaic infection, including necrotic spots on inoculated leaves with no systemic infection. Thus, different TMV strains appeared to be infecting tobacco and tomato plants. Similarly, N′ tobacco showed different responses to inoculation with TMV isolates from tomato and tobacco plants in Japan (Komuro et al. 1966; Komuro and Iwaki 1968; Oshima et al. 1964). Following these observations, the first report of the International Committee on Nomenclature of Viruses established two species, Tobacco mosaic virus and Tomato mosaic virus in the TMV group (Harrison et al. 1971; Wildy 1971). In the 1980s, complete nucleotide sequences were determined for TMV-vulgare and tomato mosaic virus (ToMV)-L (formerly TMV-L), which shared only ~ 80% nucleotide sequence identity (Goelet et al. 1982; Ohno et al. 1984).

TMV and ToMV are presently classified in the genus Tobamovirus (family Virgaviridae), which includes 37 virus species (Walker et al. 2022). Tomato mosaic diseases caused by tobamovirus infection have been reported worldwide, nearly everywhere that tomato plants are grown (CABI 2021). Geographically and historically, TMV and ToMV are the most prevalent tobamoviruses that infect tomato plants worldwide (Hollings and Huttinga 1976; Panno et al. 2021; Zaitlin 2000; Zitter 2014). Li et al. (2021) performed a field survey in China in 2013–2017, and detected ToMV and another tobamovirus, tomato mottle mosaic virus (ToMMV), in 2.9% and 1.1% of 446 field-grown tomato samples showing virus-like disease symptoms, respectively. This survey showed that ToMV was the fifth most common virus, and that tobamovirus infection rates had declined compared to those reported in previous studies, perhaps due to the application of control measures such as virus-resistant varieties.

A global phylogenetic analysis of 75 ToMV isolates with available CP open reading frame nucleotide sequences subdivided the isolates into three clades (Rangel et al. 2011). Clade I consisted of 71 ToMV isolates with very low genetic diversity, and four isolates were categorized into Clades II and III. The Clade III sequences were identical or similar to the sequence of ToMMV (Li et al. 2013). Considering that these four isolates were found in Brazil, where ToMMV was detected from the sample in 1992 (Nagai et al. 2018), the Clade III viruses may have been unrecognized cases of ToMMV. The phylogeographic history of ToMV in the Eurasian continent was examined using CP gene sequences from a total of 102 ToMV isolates collected in East Asia, Europe, and the Middle East from 1975 to 2020; the results suggested that it first appeared in Europe in ~ 1750 and then was introduced in the Middle East in the 1920s, followed by East Asia (Xu et al. 2021). Notably, ToMV sequences detected by reverse-transcription polymerase chain reaction (RT-PCR) in glacial ice subcores from Greenland generated ~ 140,000 years ago had nucleotide sequences that were nearly identical to those found in the modern age (Castello et al. 1999).

Symptomatology and yield loss associated with tomato mosaic disease

The symptomatology and epidemiology of tomato mosaic disease were studied extensively from the 1950s to the 1970s (reviewed in Broadbent 1976). However, the causal viruses were not properly distinguished until the ~ 1960s, and ToMV was still thought to be a strain of TMV until ~ 1980 (Brunt 1986). In this subsection, we follow the nomenclature of the original literature.

Tomato mosaic diseases caused by ToMV or TMV infection generally do not result in plant death but inhibit the growth of shoots and roots (Broadbent and Cooper 1964) and cause various leaf, stem, and fruit symptoms (Fig. 1). The appearance and severity of symptoms varies considerably depending on many factors such as the combination of virus isolates and host genotypes, timing of infection, light intensity, and temperature (Broadbent 1976; Zitter 2014). Leaves exhibit light and dark green mottling, sometimes accompanied by yellowing, curling, or narrowing (Zitter 2014). Stripe or streak diseases that cause leaf, petiole, or stem necrosis are also produced by some TMV strains (Ainsworth 1933; Ainsworth et al. 1934; Komuro 1963; Komuro et al. 1966).

Fig. 1
figure 1

Leaf mosaic (left) and fruit abnormalities (right) of glasshouse-grown tomato plants infected with tomato mosaic virus

Full size image

Tomato plants with mosaic disease may also suffer from fruit abnormalities such as yellow rings (at high temperatures) or uneven ripening, including bronzing, blotchy ripening, graywall, brownwall, and internal browning symptoms, which reduce their quality (Broadbent 1976; Zitter 2014). The appearance of symptoms on the fruit of TMV-infected plants is affected by various factors including light intensity (Murakishi 1960) and the timing of infection (Boyle and Bergman 1967; Boyle and Wharton 1957; Broadbent 1964; Taylor et al. 1969).

TMV infection is also associated with yield reduction (Broadbent and Cooper 1964; Rast 1967). In field-grown, machine-harvested tomatoes in Italy, every 10% increase in the infection rate resulted in a 4.4% decrease in the average yield (Di Candilo et al. 1992). The number of tomato fruit set per cluster temporally decrease after TMV infection (Alexander and Campbell 1959; Komochi et al. 1966).

Transmission of tobamoviruses

Tobamoviruses are readily transmitted from infected plants to healthy plants through mechanical dispersal via contaminated scissors, hands, clothing, and soil (Broadbent 1976). Although birds and bumblebees transmit tobamoviruses among tomato plants (Broadbent 1965a; Levitzky et al. 2019; Okada et al. 2000), they are not major routes of ToMV transmission. By contrast, seed transmission is an important pathway for tobamoviruses. In tomato seeds contaminated with TMV or ToMV, the virus resides mainly on the surface or within the seed coat (testa) and sometimes in endosperm but is absent from the embryo (Broadbent 1965b; Taylor et al. 1961). Seed transmission occurs through infection from lesions mainly formed during transplantation. When highly contaminated tomato seeds were planted, seed transmission was often observed, but in most cases only after transplantation (Taylor et al. 1961). Thus, contaminated seed coats are considered a major source of seed transmission.

Tobamovirus genome organization

Tobamovirus virions are rod-shaped particles of approximately 300 nm in length and 18 nm in diameter (Fig. 2), each containing one genomic RNA and approximately 2,130 CP molecules. The tobamovirus genome is a non-segmented, single-stranded, messenger-sense RNA of approximately 6,400 nucleotides in length. It encodes at least four proteins: a 130-kDa protein, its read-through product of 180 kDa, a 30-kDa protein, and the CP (17.5 kDa) (Ishibashi and Ishikawa 2016) (Fig. 3). The 130-kDa and 180-kDa proteins are produced through the translation of genomic RNA, participate in viral RNA replication (Ishikawa et al. 1986), and are collectively referred to as replication proteins. The 130-kDa protein also acts as a suppressor of RNA silencing (Ding et al. 2004; Kubota et al. 2003). The other two proteins are dispensable in viral RNA replication; each is produced through the translation of subgenomic RNAs synthesized during viral RNA replication. The 30-kDa protein is required for cell-to-cell movement of the virus, and therefore is referred to as a movement protein (MP) (Meshi et al. 1987). The CP is required for systemic infection and virion formation. Replication proteins co-translationally bind the 5′-proximal region of genomic RNA to select a replication template, and form a replication complex on intracellular membranes that catalyzes the synthesis of negative-strand RNA, followed by the synthesis of positive-strand genomic and subgenomic RNAs (Ishibashi and Ishikawa 2016).

Fig. 2
figure 2

Electron micrograph of purified virion of the L strain of tomato mosaic virus. Scale bar 200 nm. Photograph by Yasuhiro Tomitaka

Full size image
Fig. 3
figure 3

Schematic representation of tobamovirus genome RNA replication and gene expression via transcription of subgenomic RNAs. Circle indicates a 5′-cap structure. Boxes indicate open reading frames

Full size image

Virus control using attenuated strains

Attenuated ToMV strains

The infection of a plant with a virus strain prevents or delays subsequent infection by another strain of the same virus or a closely related virus. This phenomenon, called cross-protection, was first observed in tobacco plants infected with a light-green mosaic virus that showed tolerance to re-inoculation with a yellow mosaic virus, whereas such protection was not observed when a dark-green mosaic virus was administered as the first inoculant (McKinney 1929). Since then, cross-protection has been widely used to control viral diseases (Ziebell and Carr 2010). For this purpose, the protecting virus must produce milder symptoms than those of common strains. Such isolates have been obtained by selection from natural strains or through artificial treatments with virulent strains, such as infection at higher or lower temperatures, incubation with nitrous acid, or ultraviolet irradiation. The masked strain of TMV was the first artificially produced attenuated strain; it was obtained by incubating tobacco stems at 34˚C following inoculation with a virulent strain (Holmes 1934).

To protect tomato plants from widespread tomato mosaic diseases in Japan (Komuro et al. 1966), Oshima et al. (1965) developed the attenuated strain ToMV-L11 (hereafter, L11) by incubating ToMV-L-infected tissues at high temperatures. Infection with L11 protected tomato plants from natural infection by ToMV or challenge inoculation by ToMV-L. However, its protection efficiency was insufficient for practical use, partly due to the occurrence of revertants from L11 to virulent strains (Goto and Nemoto 1971; Oshima et al. 1965). To improve its stability, L11 was further passaged to obtain ToMV-L11A (hereafter, L11A), which showed milder symptoms in tomato plants. Field tests in Chiba and Shizuoka Prefectures in Japan demonstrated that inoculation with L11A led to higher fruit yield than in a non-inoculated control (Aoki and Ogiwara 1974; Oshima 1981). In response to the increased demand for pre-inoculated tomato seedlings from tomato growers, a large-scale spray inoculation technology was developed (Nagai and Takeuchi 1979).

In the 1970s, tomato cultivars harboring the resistance gene Tm-1 were introduced and cultivated. However, ToMV strains that overcame the resistance were soon found throughout Japan (Noba and Kishi 1979). Because L11A was a derivative of ToMV-L, which was unable to overcome Tm-1, L11A was inapplicable for these cultivars. Oshima et al. (1978) conducted repeated inoculation of L11A into Tm-1 tomato plants and obtained ToMV-L11A237, which multiplies faster than L11A in these plants. L11A237 exhibited much greater protection efficiency than L11A in tomato cultivars with Tm-1 against challenge inoculation with a virulent Tm-1 resistance-breaking isolate or natural infection of ToMV in field-grown tomatoes.

L11A was provided to tomato growers and achieved good results in various regions of Japan. Over time, isolates used in Chiba, Akita, and Fukushima Prefectures were observed to have different levels of protection efficiency. L11A-Fukushima was superior, particularly during summer at high temperatures (Matsumoto et al. 1998). L11A-Fukushima accumulation in infected tissues was 2–3-fold higher than that of L11A, which may account for its higher protection efficiency (Matsumoto et al. 2002).

Molecular characterization of attenuated strains

Comparisons of the complete nucleotide sequences of ToMV-L and L11A identified 10 nucleotide substitutions, among which three cause amino acid substitutions in the replication protein (Nishiguchi et al. 1985). Because the first substitution (G1117 to A causing a Cys349 to Tyr substitution in the replication protein) was shared with L11, it was expected to be responsible for symptom attenuation. Around the same time, the reverse genetics system of ToMV was developed, and tomato and tobacco plants inoculated with infectious transcripts from cloned cDNA of ToMV-L and L11A reproduced symptoms similar to those of each parental isolate (Meshi et al. 1986).

The importance of RNA silencing and its suppression by viral suppressors in virus infection has been well established (Jin et al. 2021). Using transgenic tobacco in which green fluorescent protein (GFP) gene expression was spontaneously silenced, ToMV-L infection was shown to suppress GFP silencing, whereas L11A had much weaker suppression activity. A GFP agroinfiltration assay in N. benthamiana identified the 130-kDa protein of ToMV as an RNA silencing suppressor, and showed that the Cys349 to Tyr substitution in the 130-kDa protein is involved in symptom attenuation and deficiency in silencing suppression (Kubota et al. 2003). Thus, the deficient RNA silencing suppression of attenuated tobamovirus strains may explain the observed reduction in virus titer and symptom attenuation. RNA silencing also plays an important role in formation of mosaic symptoms in tobacco by ToMV infection, defining the marginal regions of dark green tissues in a mosaic leaf (Hirai et al. 2008).

Introducing the same mutation into another tobamovirus, pepper mild mottle virus (PMMoV) produced similar attenuated phenotypes (Tsuda et al. 2007). Other amino acid substitutions responsible for symptom attenuation were found also in the replication proteins of tobamoviruses including TMV (Lewandowski and Dawson 1993), ToMV-L11A-Fukushima (Yamamoto et al. 2002), PMMoV (Hagiwara et al. 2002; Ichiki et al. 2005), and cucumber green mottle mosaic virus (CGMMV) (Ali et al. 2016) (Table 1).

Table 1 Amino acid substitutions in the replication proteins of attenuated strains of tobamoviruses
Full size table

Two mutually nonexclusive molecular mechanisms have been proposed for symptom attenuation by amino acid substitutions in the replication proteins of tobamoviruses. Tobamovirus replication proteins bind the small RNA duplex and have been suggested to inhibit formation of the RNA-induced silencing complex (Csorba et al. 2007; Kurihara et al. 2007). Replication proteins of an attenuated strain of CGMMV, CGMMV-SH33b that have the Glu480 to Gly, Ala1124 to Val, Asn1157 to Asp, and Pro1397 to Ser substitutions, exhibited an impaired activity to bind siRNAs (Chen et al. 2020). In contrast, the Cys349 to Tyr substitution alters the subcellular localization of ToMV replication proteins, such that replication proteins of L11 tend to be membrane-bound, which causes a shortage of replication proteins in the soluble fraction. The ability to bind the small RNA duplex of a soluble replication protein is similar between mutant and wild-type proteins. Based on this observation, it was proposed that soluble replication protein acts as an RNA silencing suppressor and that mutation affects RNA silencing suppressor activity through a reduction in soluble replication proteins (Hagiwara-Komoda et al. 2008).

Serial inoculation of two ToMV variants tagged with different fluorescent proteins into tobacco protoplasts combined with a mathematical model revealed that a 2 h interval was sufficient for protection against a challenge-inoculated virus and that the occupation of possible replication sites by the pre-inoculated virus in each cell could be a mechanism for cross-protection (Miyashita et al. 2015).

Expanding development and use of attenuated strains

Following the successful application of L11A for the control of tomato mosaic disease, a similar attenuated strain was developed in the Netherlands, through nitrous acid treatment of a virulent TMV strain (Rast 1972). The attenuated virus, MII-16, was widely used in the Netherlands and the UK (Fletcher and Rowe 1975; Rast 1975). In Japan, three attenuated strains of PMMoV, PMMoV-C-1421, Pa18, and TPO-2–19, were obtained independently through heat treatment or natural selection from field-grown pepper plants and showed good protection ability (Goto et al. 1997, 1984; Nagai 1987). Mosaic symptoms and fruit abnormalities in muskmelon plants caused by CGMMV were controlled by the attenuated strain CGMMV-SH33b, which was developed from virulent strain CGMMV-SH through nitrous acid treatment and ultraviolet light irradiation (Motoyoshi and Nishiguchi 1988).

Recent control of tomato mosaic diseases has largely depended on resistant tomato varieties harboring the Tm-22 gene; however, occurrences of ToMV with Tm-22-breaking ability have continued in Japan (Kubota 2016; Kuroiwa et al. 2022), and Tm-22-insensitive tobamoviruses have spread worldwide. Moreover, some tomato growers cannot obtain resistant varieties. Therefore, it is likely that the control of viral diseases using attenuated strains is still useful, and the development of superior isolates and new application methods should continue. Recently, L11A was also applied to control tobamovirus diseases in the Chinese lantern plant (Physalis alkekengi) (Yoneda et al. 2019).

Genetic resistance against tobamoviruses

Plants have developed defense systems against diverse pathogens. Any single gene responsible for resistance to specific pathogens is termed a resistance gene. Resistance genes are either dominant or recessive; most dominant resistance genes encode similar proteins containing nucleotide-binding site and leucine-rich repeats (NB-LRR) (de Ronde et al. 2014). An NB-LRR protein elicits a strong defense reaction that often accompanies hypersensitive cell death upon the recognition of a pathogen infection. Recessive resistance genes encode loss-of-function alleles of host susceptibility genes, which are required for pathogen infection (Garcia-Ruiz 2018; Hashimoto et al. 2016). Two dominant resistance genes against tobamoviruses have been identified and introduced into tomato cultivars (Pelham 1966).

Tm-1

Tm-1 is an atypical resistance gene encoding a conserved but uncharacterized protein that binds ToMV replication proteins and inhibits viral RNA replication (Ishibashi et al. 2007; Ishibashi and Ishikawa 2014). Upon binding to replication proteins, Tm-1 inhibits the formation of the ToMV replication complex on host intracellular membranes before negative-strand RNA synthesis (Ishibashi and Ishikawa 2013). A Tm-1 allele from ToMV-susceptible tomato plants, tm-1, inhibits the replication of tobamoviruses that are not adapted to tomato plants (Ishibashi et al. 2009), suggesting that ToMV has adapted to tomato plants through escape from the antiviral protein tm-1. Tm-1 was derived from the wild tomato species Solanum habrochaites. Molecular evolutionary analyses of Tm-1 alleles in S. habrochaites have suggested that a small part of Tm-1 (encoding the 79th to 112th amino acid residues) is under positive selection, which constitutes an interface of protein–protein interaction with ToMV replication proteins in a crystal structure (Ishibashi et al. 2012, 2014). These findings suggest that Tm-1 has evolved under the selective pressure of ToMV infection in S. habrochaites.

ToMV mutants capable of overcoming Tm-1 emerged soon after the introduction of the gene into tomato cultivars (Pelham et al. 1970). Key amino acid residues of the resistance-breaking ToMV mutants were identified in the helicase domain of the replication proteins (Meshi et al. 1988; Strasser and Pfitzner 2007). These residues are exposed on the surface of the protein, where Tm-1 binds (Ishibashi et al. 2014; Nishikiori et al. 2012). Amino acid substitutions found in the resistance-breaking mutants weaken the affinity to Tm-1 (Ishibashi et al. 2007, 2014), which explains how ToMV escapes recognition by Tm-1 to break resistance. The conventionally used Tm-1 gene encodes Ile at the 91st residue, whereas Tm-1 alleles in several S. habrochaites accessions encode Thr. The Tm-1 variant with Thr91 exhibits stronger activity sufficiently inhibiting the RNA replication of a resistance-breaking ToMV mutant (Ishibashi et al. 2012). Thus, Tm-1 alleles with Thr91 may be a potential source for resistance breeding.

Tm-2

The Tm-2 locus on chromosome 9 has the dominant ToMV-resistance alleles Tm-2 and Tm-22, and the recessive ToMV-susceptible allele tm-2. Tm-2 and Tm-22 are derived from Solanum peruvianum and encode coiled-coil–NB-LRR proteins with four amino acid differences (Lanfermeijer et al. 2003, 2005). Although ToMV infection generally does not induce a hypersensitive reaction in Tm-2 or Tm-22 tomato plants, Tm-2 and Tm-22 induce cell death when transiently coexpressed with ToMV MP (Kobayashi et al. 2011). Similarly, systemic necrosis is often induced when Tm-22 scion/rootstock is grafted with ToMV-susceptible or Tm-1 rootstock/scions. Tm-22 protein localizes to the plasma membrane and is activated through self-association via the coiled-coil domain upon the recognition of ToMV MP (Chen et al. 2017; Wang et al. 2020).

Tm-2 and Tm-22 have different resistance spectra; Tm-2-resistance-breaking ToMV mutants do not overcome Tm-22 resistance and vice versa. The MP of Tm-2 resistance-breaking mutants ToMV-B7 has the Cys68 to Phe and Glu133 to Lys substitutions and ToMV1-2 has Glu52 to Lys and Glu133 to Lys (Meshi et al. 1989; Strasser and Pfitzner 2007), whereas the MP of Tm-22-breaking ToMV-N3 has the Ser238 to Arg and Lys244 to Glu substitutions and ToMV-KMT has Asp240 to Tyr (Kuroiwa et al. 2022; Weber et al. 1993). The emergence of Tm-22-breaking ToMV mutants is sporadic, and Tm-22 has long been used to protect tomato plants from ToMV.

Emerging tobamoviruses in tomato crops

Tobamovirus disease in tomato plants has been controlled for more than 40 years, since the introduction of resistant varieties harboring Tm-22. However, new tobamoviruses that affect Tm-22 tomatoes have emerged and have caused outbreaks worldwide.

ToMMV

ToMMV was first identified in Mexico (Li et al. 2013) and subsequently reported in the USA (Webster et al. 2014), Israel (Turina et al. 2016), Spain (Ambrós et al. 2017), China (Li et al. 2017), and Australia (Lovelock et al. 2020). Recent studies have revealed that ToMMV had been present in Brazil since 1992 (Nagai et al. 2018) and was widespread in Europe before 2007 (Schoen et al. 2023). ToMMV-infected tomato plants show leaf distortion, mottling, and necrosis. ToMMV multiplication is strongly suppressed in Tm-22/Tm-22 homozygous plants and Tm-22/Tm-2 heterozygous plants, but it can infect Tm-22/tm-2 heterozygous tomato plants (Tettey et al. 2023), suggesting that differences in the expression levels of resistant alleles that weakly recognize ToMMV MP may be a determinant of susceptibility to the virus. Because many ToMV-resistant commercial tomato cultivars are Tm-22/tm-2 heterozygotes, ToMMV can infect these plants to produce symptoms (Nagai et al. 2019).

Tomato brown rugose fruit virus (ToBRFV)

Another emerging tobamovirus, tomato brown rugose fruit virus (ToBRFV), was recently identified in Jordan and Israel (Luria et al. 2017; Salem et al. 2016) (Fig. 4). ToBRFV has caused devastating disease outbreaks, even pandemics, and is currently considered the most serious threat to tomato production worldwide (Salem et al. 2023; Zhang et al. 2022). ToBRFV multiplication is not affected in Tm-22/Tm-22 homozygous plants (Hak and Spiegelman 2021). Key amino acid residues in ToBRFV MP for escape from recognition by Tm-22 have been identified (Yan et al. 2021b). Interestingly, ToBRFV MP is less functional for viral cell-to-cell transport than TMV MP (Hak and Spiegelman 2021), suggesting that the possible evolution of ToBRFV via adaptation to Tm-22 tomato plants may have been associated with a fitness cost.

Fig. 4
figure 4

a Tomato brown rugose fruit virus (ToBRFV)-infected greenhouse tomato plants in Mexico. b Fruits of ToBRFV-infected tomato plants in Turkey

Full size image

ToBRFV has been reported in European countries including Albania (Orfanidou et al. 2022), France (Skelton et al. 2022), Germany (Menzel et al. 2019), Greece (Beris et al. 2020), Italy (Panno et al. 2019), the Netherlands (van de Vossenberg et al. 2020), Norway (Hamborg and Blystad 2022), Slovenia (Vučurović et al. 2022), Spain (Alfaro-Fernández et al. 2021), Switzerland (Mahillon et al. 2022), and the UK (Skelton et al. 2019). Cases have also been confirmed in Central America and North America, including Canada (Sarkes et al. 2020), Mexico (Camacho-Beltrán et al. 2019), and the USA (Ling et al. 2019). ToBRFV has even reached countries in Asia, including China (Yan et al. 2019), Iran (Ghorbani et al. 2021), Lebanon (Abou Kubaa et al. 2022), Saudi Arabia (Sabra et al. 2022), Syria (Hasan et al. 2022), and Turkey (Fidan et al. 2019). Temporary ToBRFV infections have also been reported in several countries (https://gd.eppo.int/taxon/ToBRFV/distribution). ToBRFV genomes obtained from samples collected around the world show very limited sequence diversity (Abrahamian et al. 2022; Çelik et al. 2022), supporting its recent emergence and rapid spread from a single origin.

ToBRFV detection and quarantine of infected plants

The global spread of tobamoviruses including ToBRFV and ToMMV, is thought to occur through untreated or inadequately disinfected contaminated seeds (Dombrovsky et al. 2017). Studies have demonstrated that seeds obtained from plants infected with ToBRFV are indeed contaminated (Davino et al. 2020; Salem et al. 2022). Due to the significant threat posed by ToBRFV to global tomato cultivation, several countries have implemented national seed quarantine systems. For example, the USA issued a federal order regarding the importation and inspection of tomato seeds (https://www.aphis.usda.gov/aphis/ourfocus/planthealth/import-information/federal-import-orders/tobrfv/tomato-brown-rugose-fruit-virus), and the European Union declared a quarantine issue for ToBRFV-infected plants (http://data.europa.eu/eli/reg_impl/2020/1191/oj). Consequently, research institutions and corporations worldwide are actively working to develop rapid detection methods and highly accurate identification techniques for the virus. An increased cost associated with disinfection and phytosanitary measures is another, often overlooked problem associated with the global spread of ToBRFV (Zhang et al. 2022).

Quantitative RT-PCR (qRT-PCR) has emerged as a widely used RNA-based approach for ToBRFV detection. This technique involves converting the viral RNA into complementary DNA (cDNA) through reverse transcription, followed by amplification and quantification using PCR. Numerous primer pairs, probes, and protocols have been developed for this purpose (Caruso et al. 2022). Various advancements in RT-PCR methodology have improved ToBRFV detection capabilities of these techniques (Salem et al. 2023). In particular, multiplex RT-PCR simultaneously detects multiple viruses that may induce similar symptoms in tomato plants, including ToBRFV and other tobamoviruses such as TMV, ToMV, and ToMMV. For example, Yan et al. (2021a) and Tiberini et al. (2022) have developed RT-PCR primers for the efficient simultaneous detection of these tobamoviruses. This multiplexing capability offers a significant advantage in screening and surveillance programs, where multiple viral pathogens must be monitored concurrently.

Another effective method for detecting ToBRFV is the enzyme-linked immunosorbent assay (ELISA) and its variants. ELISA utilizes specific antibodies that bind to viral proteins, allowing the identification and quantification of ToBRFV (Bernabé-Orts et al. 2021). Many reputable commercial providers offer antisera and kits specifically designed for ELISA-based ToBRFV detection. Immunostrips are also available for lateral flow assays for ToBRFV detection. Importantly, these assays may exhibit cross-reactivity with other tobamoviruses such as TMV, ToMV, and ToMMV. However, despite these limitations, the immunostrip method provides rapid, field-deployable assays that deliver results within minutes.

In addition to RT-PCR and ELISA, researchers have explored other innovative techniques for ToBRFV detection. For example, digital droplet PCR (ddPCR) offers increased precision and sensitivity compared to conventional PCR (Vargas-Hernández et al. 2022). This technique partitions the PCR reaction into thousands of individual droplets, allowing for absolute quantification of the target viral RNA molecules. CRISPR-Cas12a is also employed for ToBRFV detection (Alon et al. 2021; Bernabé-Orts et al. 2022). The combination of these and other emerging technologies will facilitate the effective management of evolving tobamoviruses.

In addition to mechanical and seed transmission, tomato fruits and bumblebees may be involved in the spread of ToBRFV (Klap et al. 2020; Levitzky et al. 2019). Addressing potential transmission routes and implementing appropriate measures are essential for safeguarding global tomato cultivation from the threats posed by ToBRFV and other tobamoviruses.

Exploring new resistance resources for ToBRFV

As ToBRFV overcomes Tm-22 resistance, the exploration of new genetic resources against ToBRFV is urgently needed. Screening of germplasm resources of cultivated or wild tomato plants for ToBRFV-resistant and tolerant lines has been conducted through inoculation tests (Jewehan et al. 2022a, b; Kabas et al. 2022; Zinger et al. 2021). Various research institutes and seed companies have identified quantitative trait loci (QTLs) and filed patents according to their findings (Ashkenazi et al. 2018, 2020; Hamelink et al. 2019; Kopeliovitch and Gilan 2022; Millenaar et al. 2021). Although some QTLs have been mapped to common chromosomal locations, it remains unclear whether they represent the same allele, different alleles, or adjacent but different genes. Unique resistance genes have also been identified. For example, Ykema et al. (2021) identified a locus from S. habrochaites on chromosome 8 encoding an NB-LRR protein. Kalisvaart et al. (2022) introgressed a ToBRFV resistance trait from S. pimpinellifolium to S. lycopersicum (Hamelink et al. 2019), and identified a QTL region on chromosome 8, along with the most likely candidate gene, TOM2A, which is a homolog of a tobamovirus susceptibility gene identified in Arabidopsis thaliana (Fujisaki et al. 2008; Tsujimoto et al. 2003). Interestingly, mutated TOM2A homologs were also identified as recessive resistance genes for TMV in N. tabacum, and artificial knockout of a TOM2A homolog in tomato plants was found to confer resistance to both ToMV and TMV (Hu et al. 2021).

Newly identified genetic materials hold promise as novel sources of ToBRFV resistance and can be utilized in resistance breeding programs. Although claims by various seed companies that trial F1 tomato plant varieties show resistance to ToBRFV are a positive development for farmers and consumers, excessive patent protection could deviate from their original purpose of addressing the threat of ToBRFV. A balanced approach is needed to ensure that these patents do not strongly hinder new entrants in the field, allowing for the wider dissemination and utilization of ToBRFV-resistant varieties to mitigate the impact of the virus.

As observed in ToMV resistance genes (Cirulli and Ciccarese 1975; Fraser and Loughlin 1982), the new resistant materials may allow increased accumulation of ToBRFV under high-temperature conditions (Jewehan et al. 2022b). The emergence of resistance-breaking mutants is also a concern in the production of resistant varieties. Therefore, sustainable tomato cultivation may require the continuous introduction and screening of new breeding materials.

Genetic engineering of tomato plants for tobamovirus resistance

Biotechnological achievements using transgenic or gene editing technologies include numerous virus-resistant plants. We describe various attempts to develop tobamovirus-resistant tomato plants.

Pathogen-derived resistance

The transgenic expression of parts of viral genomes in plants confers resistance against corresponding viruses. This phenomenon, called pathogen-derived resistance, was first reported by Beachy and colleagues in a TMV–tobacco system (Abel et al. 1986), and found to be applicable to many plant–virus combinations. Nelson et al. (1988) demonstrated that the transgenic expression of TMV CP conferred resistance to TMV and ToMV in tomato plants, and caused no significant yield loss under greenhouse or field conditions; because resistance was positively correlated with CP expression levels, the presence of the CP prior to infection would have perturbed the TMV and ToMV multiplication. However, the expression of non-translatable versions of viral CP mRNA can also induce resistance to the corresponding virus (Lindbo and Dougherty 1992). Further analyses have revealed that post-transcriptional gene silencing occurred in this and many other cases (Waterhouse et al. 1999). Irrespective of the underlying mechanisms, pathogen-derived resistance is useful for protecting crops from viral diseases (Khalid et al. 2017; Lindbo and Falk 2017).

Engineering resistance genes

Engineering existing resistance genes is also promising. Transgenic introduction of the TMV resistance gene N from tobacco into tomato confers resistance to TMV, ToMV, and youcai mosaic virus (YoMV) (Whitham et al. 1996). Spiegelman et al. (2022) established a screening system for Tm-22 variants that recognized ToBRFV MP and identified three single-nucleotide mutations that allowed the Tm-22 gene product to induce resistance against ToBRFV infection. Lindbo (2022) also found that variants or mutants with substitutions in the LRR domain of the Tm-22 protein could recognize or bind to the MP of various tobamoviruses, including TMV, ToMV, ToMMV, and ToBRFV. Using molecular docking, in silico 3D structure prediction, and computational affinity methods, Rivera-Márquez et al. (2022) also identified potential amino acid substitutions in Tm-22 that conferred resistance to ToBRFV. These engineered Tm-22 variants may be used as transgenic or gene-edited plants through targeted mutagenesis of the endogenous gene.

Disruption of susceptibility genes

Genetic engineering for virus resistance can also be achieved through the manipulation of host susceptibility genes that are utilized by viruses for their multiplication. The development of CRISPR-Cas9-mediated gene editing technology has greatly facilitated targeted gene disruption. Host multi-pass transmembrane proteins TOM1 (Yamanaka et al. 2000) and TOM2A (Tsujimoto et al. 2003), and the membrane-associated small GTP-binding protein ARL8 (Nishikiori et al. 2011) are required for efficient tobamovirus multiplication. In A. thaliana, the TOM1 protein is encoded by AtTOM1 and its two homologs, AtTOM3 and AtTHH1. The ARL8 protein is encoded by three homologous genes, AtARL8a, AtARL8b, and AtARL8c. Simultaneous knockout of AtTOM1 and AtTOM3 confers strong resistance to tobamoviruses including YoMV and ToMV without appreciable growth defects (Fujisaki et al. 2006; Yamanaka et al. 2002). Similarly, the simultaneous knockout of AtARL8a and AtARL8b leads to strong tobamovirus resistance with no negative impact on plant vigor (Nishikiori et al. 2011). Knockout of the TOM2A gene in A. thaliana (AtTOM2A) confers resistance to YoMV but has little effect on ToMV multiplication. In either case, the mutant plants are fully susceptible to cucumber mosaic virus of the genus Cucumovirus. TOM1 and ARL8 proteins interact with each other and with tobamovirus replication proteins (Nishikiori et al. 2011). TOM1 protein also interacts with TOM2A protein (Tsujimoto et al. 2003). Further analyses have suggested that TOM1 and ARL8 proteins are necessary to activate an enzymatic activity in tobamovirus replication proteins, without which tobamovirus negative-strand RNA synthesis does not occur (Nishikiori et al. 2011).

The tomato genome contains five putative genes that encode proteins with significant amino acid sequence similarity to AtTOM1. These genes have been named differently in three different studies (Ali et al. 2018; Ishikawa et al. 2022; Kravchik et al. 2022) as summarized in Table 2. In this review, we follow the nomenclature of Ishikawa et al. (2022). Among SlTOM1 genes, the expression levels of SlTOM1e are very low. Ali et al. (2018) knocked down SlTOM1a, SlTOM1b, or SlTOM1c genes by expressing inverted repeat RNA of the coding sequences of each gene in the tomato cultivar ‘Micro-Tom’, and found that these plants showed resistance to ToMV. Tobamovirus-resistant plants developed using this method as well as pathogen-derived resistance are regulated under genetically modified organism legislation, restricting their practical use in many countries.

Table 2 Nomenclature of TOM1 homologs in tomato
Full size table

Ishikawa et al. (2022) knocked out the genes SlTOM1a–d using the CRISPR-Cas9 system in tomato cultivar ‘GCR26’, which is susceptible to ToMV and ToBRFV, and constructed mutant lines carrying homozygous single or multiple Sltom1 mutations in all combinations. In Sltom1 single or double mutants, ToMV and ToBRFV CPs accumulated to levels similar to or only slightly lower than those of wild-type non-transgenic plants. In Sltom1 triple mutant plants, ToMV and ToBRFV CPs accumulated more slowly than in wild-type plants. Neither ToBRFV or ToMV CP accumulation nor disease symptoms were observed in Sltom1 quadruple-mutant plants, suggesting that the four SlTOM1 genes contribute in parallel to ToMV and ToBRFV multiplication, in the order SlTOM1a ≥ SlTOM1c > SlTOM1d > SlTOM1b. In ToBRFV-inoculated Sltom1acd triple mutant plants, which showed the strongest resistance among Sltom1 triple mutants, ToBRFV mutants emerged with more efficient multiplication than wild-type ToBRFV; however, the mutant ToBRFV was unable to multiply in Sltom1 quadruple mutant tomato plants. Tobamovirus mutants that can multiply in Sltom1 quadruple mutant plants have not yet emerged, which suggests that this resistance is durable. Loss of TOM1 function inhibits tobamovirus RNA synthesis, which may help minimize the risk of emergence of mutant viruses that can overcome resistance by the Sltom1 quadruple mutant. Thus, knockout of the four SlTOM1 genes is a promising strategy for protecting tomato plants from ToBRFV and other tobamoviruses.

Kravchik et al. (2022) knocked out SlTOM1a, SlTOM1b, and SlTOM1c in tomato cultivar ‘M82’ using the CRISPR-Cas9 system, and constructed Sltom1ac double and Sltom1abc triple mutant plants. They found that ToBRFV CP accumulated more slowly in both mutants than in the wild-type plants. ToMV CP accumulation was delayed in the triple mutant plants but not in the double mutant plants. These results are consistent with those of Ishikawa et al. (2022), with minor discrepancies that may have arisen from genetic background differences (‘GCR26’ vs. ‘M82’) and/or the conditions of virus multiplication assays.

Recently, Hu et al. (2021) knocked out a tomato homolog of AtTOM2A and showed that the plant was resistant to TMV and ToMV. Susceptibility of the Sltom2a mutant tomato to ToBRFV has not been reported. The tomato genome contains three genes with significant amino acid sequence similarity to AtARL8a and AtARL8b (Kravchik et al. 2022). Knockdown or knockout of these genes may also confer resistance against ToBRFV and other tobamoviruses.

Following the development of the GABA-enriched tomato (Waltz 2021), virus-resistant plants created using gene editing technologies are expected to be cultivatable beyond the reach of genetically modified organism legislation in several countries (Tripathi et al. 2020). However, loss-of-function mutations can be generated by other methods such as classical mutagenesis using chemical mutagens, followed by the screening of desired mutants by TILLING or high-throughput sequencing (Henikoff et al. 2004; Rigola et al. 2009). ToBRFV-resistant tomato lines established by these methods could be cultivated globally without any restrictions.

Conclusions

Tomato plants are cultivated worldwide, and tobamoviruses are found wherever tomatoes are grown. Before the 1970s, approximately 20% of global tomato production was lost by ToMV (Broadbent 1976). Successful resistance breeding with Tm-22 has markedly reduced the damage caused by ToMV, such that it is no longer a concern. However, the emergence and subsequent outbreaks of ToMMV and ToBRFV have demonstrated that the arms race between tomato plant cultivation and tobamoviruses has been ongoing. From a virological perspective, the emergence and subsequent spread of ToBRFV, and similarly of SARS-CoV-2, during the high-throughput sequencing era has provided unique opportunities for improving our understanding of viral ecology and epidemiology. The fact that a host susceptibility gene identified in A. thaliana ~ 20 years ago is useful for constructing newly emerged tobamovirus-resistant tomato plants highlights the importance of basic research for plant protection. Further understanding of virus–host interactions will provide new tools to combat viruses that may emerge in the future.