Tomato and Pepper Seeds as Pathways for the Dissemination of Phytopathogenic Bacteria: A Constant Challenge for the Seed Industry and the Sustainability of Crop Production

1. Introduction

“Seeds are a basic input for all crop production: all farmers need good seed, irrespective of their farming systems and the markets that they focus on” [1]. The seed sector is probably the most crucial and critical agricultural input that is fundamental for growers to produce crops for food, feed and non-food uses. Indeed, with the global population expected to reach 9.8 billion by 2050, access to quality seeds is critical for food security and nutrition, as it is often stressed by FAO. The seed industry plays a pivotal role in ensuring crop productivity by strengthening the seed value chain and implementing good seed availability worldwide in terms of improved, well-adapted, productive, nutritious, and resilient genotypes: this appears fundamental in food insecure parts of the world, but also in high income countries, to ensure a correct remuneration to dedicated farmers. According to the ISPM 5 [2], seed (in the botanical sense) is a commodity intended for planting [2], therefore living material that farmers may use to produce a crop plants, more frequently vegetables or cereals, but also ornamentals and weed species to be used in gardens, parks, and other public green. Seed is a true global commodity: in 2021, the market for commercial seeds grew by 7.1% over 2020, reaching an estimated value of $47,242 million [3], with a crucial role in promoting food security: indeed, almost 90% of the world’s food crops are grown from seeds [4]. In 2020, vegetable seeds had a market share of almost 17% in global seed sales of the seed industry, therefore, compared with other seed segments (e.g. maize, soybean and cotton), the vegetable seed market is highly fragmented [5].

As for other propagation materials, the international movement of seeds is thoroughly regulated: the main justification for such rules is to avoid the movement of plant pests that are associated with seeds from a region where they are established to areas afar that are still pest-free. Therefore, the FAO standard on the international movement of seeds was approved and published to provide guidance to national plant protection organisations (NPPOs) in identifying, assessing, and managing the pest risk associated with international seed trade [4].

Tomato, and pepper to a lesser extent, can be also considered global seed commodities: for instance, the global tomato seeds market is valued at $1.33 billion in 2023 and expected to be worth $1.78 billion by 2027 [6]. Very frequently, trade and plant health are tightly connected, when it comes to phytosanitary issues: for instance, in case of tomato, breeding parental lines is done in one European country (e.g. The Netherlands), whereas production of basic seeds is done in a second European country (e.g. Spain). Then, basic seeds are treated and manufactured in different European areas or countries to provide seeds to China, where the production of hybrid seeds is commonly done. From China, large seed lots are again shipped to Europe for treatment as commercial hybrid seeds; finally, commercial packaging, final sale and use may happen in the Americas [7]) (Figure 1).

Seeds, as any other plant material in trade, may be associated with one or more plant pests: therefore, seeds are known pathways for the introduction and spread of pests into new territories, when and wherever suitable hosts and environmental conditions are available. Tomato and pepper seeds are recognised pathways for a large number of pests and pathogens, both regulated and non-regulated. In the European Union, a notification and rapid alert system dealing with interceptions of pests (EUROPHYT, https://food.ec.europa.eu/plants/plant-health-and-biosecurity/europhyt_en) is in place since several years and is essential for the implementation of preventative measures based on robust and up-to-date data from trade in plants, included seeds. For instance, in 2022, 48 interception of tomato seeds infected by the Tomato Brown Rugose Fruit Virus (ToBRFV) were notified to the European Commission, whereas in 2019 such notifications were only 2. The known seed-associated and seed-transmitted bacteria affecting tomato and pepper seeds are five: the Gram positive Clavibacter michiganensis subsp. michiganensis (Cmm), and four Gram negative Xanthomonads: Xanthomonas euvesicatoria pv. euvesicatoria (Xee), X. e. pv. perforans (Xep), X. hortorum pv. gardneri (Xhg) and X. vesicatoria (Xv). An additional one, Clavibacter capsici (basonym C. michiganensis subsp. capsici; Cc) has been recently reported to affect pepper and not tomato [9]. Finally, Pseudomonas syringae pv. tomato (Pst), is a seed-borne tomato pathogen, but it has never been reported to affect pepper. Apart from the latter, all other bacteria are regulated in the European Union; the status of C. capsici is still to be defined, since it’s a very recent finding and, presumably, will be regulated as well.

2. Seed endophytes: recruitment and dynamics

Endophytic communities are defined as populations of microorganisms that establish themselves within the internal tissues of plants, without causing any apparent harm to the host plant [10]. Nevertheless, this is a debated definition as, theoretically, the microbiota within an apparently healthy plant could consist of a mix of mutualistic, commensal, and latent pathogenic strains [11]. These communities consist of a wide range of taxa, including fungi and bacteria, and have been studied in various plant species and various plant parts, from roots to any aerial organs, with the aim of understanding their interactions with their host. Seed associated microbial communities represent the initial inoculum source for the plant microbiota and assist seed conservation, germination, and seedling development through the production of suitable metabolites that are made available during the early stages of seed revitalisation [12]. Most endophytes appear to originate from the plant rhizosphere [13]. Initially, the germinating seeds absorb water that is available in the sowing bed and start to excrete some exudates that may attract bacteria from the surrounding, particularly from the spermosphere and the rhizoplane; such bacteria may enhance plant growth and vigour [14]. Microorganisms may also be transferred from plants' vegetative parts to the seeds [15] or through the male gametophyte and, as seed develops inside the fruit, it colonises the embryo and then the surrounding endosperm [16]. Additionally, plants recruit endophytes from their phyllosphere as well and, moreover, microbial epiphytes are also recruited horizontally, therefore moving from plant to plant during particular weather conditions like, e.g. rains or showers [17]. Finally, some endophytes may be vertically transmitted through the seed [18].

Seeds represent a significant phase in the endophyte biology, and specific microbial communities (core seed microbiome) can persist there for years in a dormant state, when the appropriate conditions are met; subsequently, such communities will again develop into the new plant originating from the germinating seed [14]. Endophytes found in different parts of the plant, including roots, stems, leaves, flowers, fruits, and seeds, may show a different taxonomic structure and a different functional behaviour [19]: some of them have been shown to exert beneficial effects on certain plant species, whereas they may exhibit a pathogenic behaviour towards other plant species. Therefore, the pathogenicity of some endophytes can be influenced by a number of biotic and environmental factors [20,21]. In general, endophytes play an important role in plant health and may positively influence plant growth and productivity and, particularly, numerous studies revealed that endophytic microorganisms play a crucial role in controlling plant pathogens [22]. Recent studies revealed that endophytes also play a crucial role in plant disease management by assisting in nutrient availability and uptake, enhancing stress tolerance and providing disease resistance [23]. Therefore, most of the published research focuses on the taxonomic identification and characterisation of putative beneficial microbes colonising the seeds, in order to: (i) describe the seed-associated microbiota; (ii) highlight the significance of seed microbiota in seed quality; (iii) describe the plant colonisation and the vertical transmission of seed endophytes; (iv) verify the application potential of selected beneficial microbes [24].

The microbial communities associated to tomato seeds have been more extensively studies than the corresponding communities associated to pepper seeds: most published research is focusing on beneficial endophytes as a significant portion of the total seed microbiome [25,26,27,28,29].

3. Phytopathogenic bacteria in tomato and pepper seeds

Tomato and pepper are affected by several phytopathogens, mainly viruses and fungi; a few phytopathogenic bacteria are also known to cause important diseases and may be regarded as a global threat for the seed industry, since most of them are seed transmitted. Two Gram positive germs are a true challenge, Cmm and Cc: the first may affect both tomato and pepper, whereas the latter is specifically associated with pepper [30]. Additionally, five Gram negative bacteria are also known to be associated to tomato and/or pepper; whereas Pst is specific to tomato, the four xanthomonads show a very different behaviour towards their respective hosts: Xv, Xep are pathogenic to tomato only, whereas Xhg and Xee may affect both tomato and pepper [31]. Other phytopathogenic bacteria may infect tomato and pepper, the most important of which are Ralstonia solanacearum and R. syzygii subsp. indonesiensis [32]: despite their importance as causal agents of destructive diseases, they are not known to be seed-transmitted. To understand the routes of seed infection by the phytopathogenic bacteria, the diseases caused by both Clavibacters, Pst and the Xanthomonads are briefly described.

3.1. Clavibacter michiganensis subsp. michiganensis (Smith) Davis et al. and Clavibacter capsici (Oh et al.) Li et al.

Clavibacter michiganensis subsp. michiganensis (Smith) (Cmm) is a highly destructive Gram-positive bacterium, with an economic relevance in terms of losses on tomato cultivations worldwide. Cmm is the causal agent of the bacterial canker and can produce significant yield losses and economic damage to the affected crops: indeed, during severe epidemics, up to 93% of plant death, and approximately 50% decrease in average fruit weight are reported [33,34]. Cmm is considered a high-risk pathogen and is included in the A2 list of quarantine organisms by the European and Mediterranean Plant Protection Organization (EPPO) [35]. Similarly, Cc causes the bacterial canker of pepper, as initially reported by [36] in Korea. Both pathogenic Clavibacters invade and proliferate within the xylem vessels of their respective hosts, causing a characteristic symptom of browning along the internal vasculature and the gradual degradation of vascular tissues. Cmm and Cc are vascular pathogens that systemically colonise the xylem of tomato or pepper plants, leading to a severe disease known as bacterial canker. The infection causes several characteristic symptoms, including unilateral leaf wilting, marginal leaf necrosis, stem cankers and, ultimately, plant death [34,36,37,38]. Both bacteria are commonly recognised as being seed transmitted, both internally in the seed and on the seed surface. Therefore, seed infection and colonisation are essential aspects of the epidemiology of bacterial canker in tomato and pepper. Despite the rate of seed transmission of the disease is quite low [39], recent studies revealed that, under favourable conditions, one infected seed in 10,000 can give rise to devastating epidemics [40,41]. When infected seeds are sown, the bacterium can move systemically through the emerging seedlings, leading to bacterial canker symptoms as the plants grow [42]. Other sources of primary inoculum are represented by infected plant debris that remains in the field after harvesting; therefore, plantlets may become infected once transplanted into a field where infected debris are present and phytopathogenic bacteria are still viable. Theoretically, secondary inocula may be produced along the growing season, when conducive conditions favour pathogen evasion and dissemination through rains or sprinkler irrigation: in such case, phytopathogenic bacteria may penetrate their respective hosts through stomata and fruit lenticels. Fruit infection is visible when typical bird’s eye spots are developing on the berries. Finally, Cmm can be easily horizontally transmitted when tomato plants are trimmed and/or pinched off: indeed, trimming blades or even operators’ hands can be easily contaminated by a single infected plant and, therefore, spread the inoculum to further plants during such agronomic practice.

Tomato seed contamination by Cmm may occur systemically, through the vascular tissue, or by its penetration through fruit lenticels. Cmm multiply and colonises the xylem vessels acropetally, thus reaching the fruit clusters and penetrating the developing fruits through the pedicels, moving until the placental tissue. Alternatively, Cmm entry through fruit stomata causes the development of circular lesions in the exo- and mesocarp where the bacteria multiply; such bacteria may then reach the seeds through the endocarp or the locular cavity when the tomatoes ripen, thus infecting the seed testa [43]. The bacterium is highly tolerant to desiccation and can survive on/in seeds for years [44].

3.2. Pseudomonas syringae pv. tomato (Okabe) Young, Dye & Wilkie

Pseudomonas syringae pv. tomato (Pst) is a Gram-negative bacterium and the causal agent of the bacterial speck of tomato. Pst is considered one of the most significant and widespread bacterial pathogens affecting tomato plants [45]. The bacterial speck is a parenchymatic disease and causes significant economic losses for tomato growers worldwide, as it can reduce fruit yield and quality [46]. Pst penetrates its host plant mainly through stomata and lenticels and forms necrotic spots on the sepals, necrosis along the pedicel and colonises symptomless flowers of different tomato varieties [47]. Typical symptoms of the bacterial speck consist in small and dark lesions on leaves, usually surrounded by a yellow halo, and necrotic spots along stems and on fruits, provoked by a coronatine toxin; from necrotic lesions the germs can evade the plant as bacterial exudates and spread around. The disease can lead to yield reductions as a result from defoliation, flower abortion and fruit lesions [48,49]. The damages caused by Pst can be remarkable in nurseries, greenhouses, and fields during warm and humid conditions [50]. Pst can survive in various environments, such as in plant debris, soil and, as an epiphyte, on leaf surfaces of the host plants or even weeds [51]: indeed, the bacterial speck is a polycyclic disease and the several secondary inocula provide an excellent means of pathogen spread in the field. While surviving in various environments for extended periods, the bacterium can act as a source of inoculum for new infections and continue its life cycle among susceptible tomato plants [52]. This bacterium is a seed-borne pathogen that primarily survives on the seed surface, attached to the desiccated pulp or internally within the seed: therefore, the bacterium is spread over long distances by infected seeds [53,54,55].

3.3. Xanthomonads: Xanthomonas vesicatoria (Doidge) Vauterin, Hoste, Kersters & Swings, X. euvesicatoria pv. euvesicatoria (Jones et al) Constantin et al.; X. euvesicatoria pv. perforans (Jones et al.) Constantin et al., and X. hortorum pv. gardneri (Jones et al.) Morinière et al.

Xanthomonads are Gram negative, yellow-pigmented bacteria affecting a wide host of plants around the world. The bacterial spot is an economically significant plant disease affecting tomato and different types of peppers. The disease is caused by four distinct bacteria belonging to the Xanthomonas genus: Xanthomonas vesicatoria, X. euvesicatoria pv. euvesicatoria, X. euvesicatoria pv. perforans, X. hortorum pv. gardneri. Xanthomonads cause necrotic lesions on leaves with polygonal shape, surrounded by a chlorotic tissue; symptoms on fruits are scab-like, raised, whitish lesions, which leads to their decreased market value [56]. Xanthomonads may enter their host plants primarily through stomata and lenticels, but also wounds (e.g. trimming/pinching lesions or hail wounds) can occasionally provide penetration sites into the host plants. Infection by Xanthomonads can cause yield losses of up to 50%. Barak et al. [57] and it has been reported that very few infected plants can lead to severe disease outbreaks [58,59]. The bacterial spot is primarily a parenchymatic disease; therefore, the germs are not supposed to efficiently colonise the xylem vessels and use them to move acropetally/basipetally within the host plants. Xanthomonads are considered seed-borne, or seed-associated pathogens, and the infected seeds are major means for long-distance dispersal of the diseases. Seeds are associated/contaminated by Xanthomonads, where bacteria can be deposited on the seed surface mixed with the infected pulp; internal seed infections are also reported [60]. Xanthomonads can survive on/in stored seed for up to 10 years [54,56]. Indeed, infested tomato and pepper seeds can serve as a significant source of primary inoculum in transplants and fruit production systems [61,62]. These pathogens can also be transmitted to healthy plants as secondary inocula through various agricultural practices, e.g. irrigation, pinching, contaminated tools, and other means [63]. Giovanardi et al. [64] showed that a seed contamination level higher than 100 CFU g^-1 is needed for a disease bacterial spot outbreak caused by Xee. Nevertheless, due to the polycyclic nature of the disease, it is important to emphasise that the threshold level may be variable, considering that pepper-growing areas have quite different climatic conditions and/or different agronomic practices (e.g. higher seeding rates). However, the long-distance spread of the bacterial spot in tomatoes and peppers Xanthomonas spp. can be also associated with the trade of infected transplants harbouring latent infections [65].

4. How phytopathogenic Clavibacters, Pst and Xanthomonads colonise tomato and pepper seeds

Seeds are home to the most diverse microbiota and the composition of their microbial communities is determined by plant genotype, environmental conditions, and agricultural practices [66]. In general, microbes can be transmitted to and colonise their host plants horizontally, via the environment from a suitable source (e.g. another host plant), or vertically, from within the parent plant to the offspring via the seed [67]. Phytopathogenic bacteria may be transmitted to and colonise their specific host plants in the same way and, as plant endophytes, they may colonise seeds in their different parts, including the embryo, and their precise localisation in seeds is related to their epidemiology, biology, penetration route and colonisation pattern [68]. The knowledge of the precise location of phytopathogenic bacteria in seed is pivotal during the selection of an appropriate method for seed disinfection: for instance, phytopathogenic bacteria colonising the seed externally may be easily removed by a washing in a disinfecting solution; this sanitation approach is useless, when the pathogens are located in the endosperm or the embryo. In general, bacteria can colonise seeds horizontally from the external environment via flowers or fruits: this pathway requires that the pathogen is able to produce secondary inocula, which are dispersed in the field, e.g. by wind-driven rain droplets or sprinkler irrigation. This is the case of Pst and Xanthomonads that enter the developing fruits through stomata or wounds and infect the pulp. More rarely this happens with Clavibacters, which are vascular pathogens that evade their hosts far more rarely than Pst and Xanthomonads, thus producing secondary inocula only under highly conducive agri-environmental conditions (e.g. extreme warm and humid conditions, favouring stem cankers development with subsequent disruption of tissues). Then, germs are able to penetrate their respective hosts through fruit lenticels, colonise the pulp extensively, and reach the seeds inside the endocarp; in such cases, phytopathogenic bacteria are mostly located above or under the seed coat and, rarely, in the endosperm [69,70]. Developing seeds can be colonised vertically, from the parent plant: for Clavibacters, plant to seed movement usually happens through the colonisation of vegetative parts of the plants to the developing seed via vascular connections to the endosperm and the embryo [43,71]. Theoretically, phytopathogenic bacteria may infect the seed through contaminated pollen: this is considered both horizontal and vertical transmission. Pollen, as pathway for fruit and seed infections, has been described in a few pathosystems [72,73]; in particular, Dutta et al. [74], described the ingress of Acidovorax citrulli into watermelon seeds by the pollen germ tubes through the stigmas, thus resulting in the colonisation of ovules. Regretfully, dedicated studies have never been attempted for the tomato/pepper – phytopathogenic bacteria pathosystems.

Summarising, Cmm and Cc are typically systemic bacteria, causing vascular diseases [38], Pst is a recognised parenchymatic bacterium, therefore with a limited vascular movement [75]. Similarly, phytopathogenic Xanthomonads, despite the development of symptoms frequently resembling those of Pst, have a parenchymatic colonisation pattern, but a limited vascular movement is also reported [76]. Therefore, the plant colonisation pattern described for the different phytopathogenic bacteria, from the penetration site to the seeds, is pivotal for understanding how and where the seeds are colonised.

Then, seeds represent the main source of primary bacterial inoculum in tomato and pepper seedlings production facilities: the bacteria are introduced in the glasshouses by infected seeds or, alternatively, disseminated by water, as in case of Pst and Xanthomonads (in case of poor hygienic conditions). Commercial cultivation of tomatoes and peppers starts with transplanting plantlets in open fields or in protected environments (e.g. tunnels, glasshouses): in such cases, transplants represent the main source of primary inoculum, through contaminated water, plant residues from a previous crop, or reservoir plants may provide an additional source of primary inoculum. Therefore, to ensure the highest phytosanitary standards along the tomato/pepper production chains, the key issue is phytosanitary certification of seeds (mandatory in case of regulated pathogens) and the possible analysis of transplants [56].

5. Detection of phytopathogenic bacteria in tomato and pepper seeds

5.1. Direct isolation on agar media

Direct isolation is a common detection technique in phytobacteriology that is applied to plant samples in order to multiply the target bacteria on a solid medium, thus making them visible in the form of a colony. This technique is highly valuable in diagnosing plant bacterial diseases and allows the identification and characterisation of specific bacterial strains. Various media have been developed to support the growth of specific types of bacterial strains while inhibiting the growth of others; they are based on knowledge of the nutritional requirements and physiological tolerances of the target bacterium. For direct isolation, semi-selective media are preferred, to exclude the growth of unwanted microorganisms, as fungi or saprophytes [83]: for instance, Wilbrink’s medium [84], NSCAA[85], MXV [86] are three of numerous agar media commonly used for the isolation of Xanthomonads. Performance criteria for bacteria isolation from seeds have been assessed and, depending on media and seed sample, the detectable concentrations of the target bacteria may vary from a few dozen germs up to approximately 10⁴ CFU ml^-1 [56]. Differently, for the direct isolation of Cmm, two different agar media, CMM1T and SCMF, performed much better than the previous ones, with an analytical sensitivity around 25 CFU ml^-1 of final concentrate [30].

Thus, the direct isolation permits the recovery and growth of living germs that could be sub-cultured to obtain an axenic culture suitable for microbial identification. Moreover, the bacterial culture can be stored at -80°C for further studies or strain characterization.

5.2. Serological detection

The serological methods in plant bacteriology diagnostics were set up, approved by the phytosanitary authorities and applied worldwide from the late ‘70s to the ‘90s of the last century, when reliable molecular methods were not available, to speed up the analyses and increase their specificity from both symptomatic and asymptomatic plant material, included seeds. The most popular serological methods applied are the indirect ImmunoFluorescence Antibody Staining (IFAS), with a sensitivity threshold between 10³–10⁴ CFU ml^-1, the Enzyme-Linked ImmunoSorbent Assay (ELISA), with a sensitivity threshold approximately at 10⁴ CFU ml^-1, the Dot-Blot Immunobinding Assay (DIBA), the Fluorescent In Situ Hybridization (FISH) and the more recently implemented – from half ‘90s to half of the 2000s – the Immuno Magnetic Separation (IMS). Interestingly, after the development and implementation of PCR, these immunological assays were studied in combination with molecular methods (i.e. PCR-FISH, PCR-ELISA, PCR-IMS) to improve the detection threshold up to few colonies per ml [87]. Nowadays, serological methods are basically used to confirm or confute the results of direct isolation or the response of molecular methods. Cross reactions are the main concern of serological methods, especially ELISA: indeed, unknown seed endophytes or saprobes may share same epitopes, or antigenic determinants, with Clavibacters or Xanthomonads [88,89]. This problem should be tackled by using a second detection method that is based on a different scientific principle, as previously highlighted. Cmm was one of the numerous bacterial pathogens, which was used to produce monoclonal antibodies, aiming its detection in seed samples [87]. Moreover, an IMS method was set up and coupled with an enrichment step, which allowed Cmm detection at low concentration (approx. 10 CFU ml^-1), similarly to pathogen concentration that can be found in naturally infected tomato seeds (approx. 10 - 10² CFU ml^-1) [56,90,91]. Serological methods for the detection of the Xanthomonas species for tomato and pepper are present just in earlier literature; monoclonal antibodies were produced for the detection of Xanthomonas campestris pv. vesicatoria, the older name that comprised the nowadays three species of the Xanthomonas genus causing bacterial spot in tomato and pepper. In that paper, the specificity assay failed by reacting with Pseudomonas gardneri, now called Xanthomonas hortorum pv. gardneri. Officially, EPPO Bulletin suggests IF or ELISA assays on bacterial axenic cultures as well, as identification methods, and not only on seed sample extracts as optional screening tests [56,92,93]. The NSHS protocol, the only official protocol for the detection and identification of Pst, allows the culture identification through biochemical methods and pathogenicity assay (NSHS So. 3.1, 2001; https://seedhealth.org/files/2022/03/So-3.1-Pseudomonas-syringae-pv-tomato-%E2%80%93-tomato-word.doc.pdf). In the recent past, monoclonal antibodies were obtained and used towards the lipopolysaccharide O-chains of Pseudomonas pathovars; the MAb Ps4e reacted against the Pst strain IPGR 140, but also reacted on other 9 pathovars, clearly showing its unspecificity [94].

5.3. Molecular detection

From the early ‘90s, at the beginning of the PCR era up to date, molecular methods have significantly increased the specificity and sensitivity of pathogen detection. Molecular detection methods sped up the analytical procedures that should ensure the phytosanitary quality of seed and, contemporarily, they gave the possibility to significantly increase the number of samples to be analysed: this appears particularly important for seed companies involved in the international trade of seeds. Available PCR-based detection methods are numerous nowadays, from the conventional end-point PCR to the qPCR, to the Loop-mediated isothermal amplification (LAMP) and to the last tool available nowadays, the Digital Droplet PCR (ddPCR). These faster methods are usually combined with pathogen extraction methods based on the use of commercial kits that have improved and standardised the yield and quality of the nucleic acids.

The phytosanitary quality of seeds may be checked at the origin, e.g. by the producing company, and/or at destination, e.g. by an accredited diagnostic laboratory in the importing country. Whereas the producing company usually analyses their own seeds prior to additional manipulations (e.g. addition of dyes, fungicide treatments, addition of inert pelleting material), commercial seeds lots to be sampled and analysed by labs in the importing countries frequently find “treated” seeds; it has been noted that any material added to commercial seeds prior to packaging may have a negative impact on the yield and quality of nucleic acids and may strongly inhibit any PCR reaction. Therefore, a nucleic acid cleaning step is highly recommended after the extraction. In general, molecular methods are used twice along the flow charts: as preliminary assay on the seed extract and as a test to confirm pathogen identity after isolating the putative target bacteria.

For the molecular detection of any phytogenic bacteria possibly present in tomato and pepper seeds only one sample is required for extraction: the same seed extract is then analysed, following the respective procedures, to check the possible presence of Clavibacters, Xanthomonads and Pst; then, there is no need to prepare multiple extracts, thus speeding up the phytosanitary analyses [30,56]. For the detection of Cmm in tomato seeds, molecular methods are officially validated and recently updated. Several PCR adapted protocols are available through probes designed on several conserved regions (e.g. 16S-23S rDNA). Starting from the DNA of Cmm axenic culture, end-point duplex-PCR protocol [95] can be applied and it is suggested also by EPPO [30]; a duplex-qPCR method on Cmm pure culture is also suggested [96]. Although the end-point PCR method may detect low numbers of Cmm germs (approx. 10² CFU g^-1 seed) with high specificity, the PCR-based detection from seed extracts is still to be validated [64]. The more updated ISHI-Veg method suggests a multiplex-qPCR protocol that was adapted (with endogenous control primers) from different molecular protocols [96,97]: this protocol may be reliably applied on crude seed extracts and proved its high specificity and sensitivity. Nevertheless, direct isolation and a PCR test on axenic cultures is mandatory to identify and confirm the isolates as Cmm. At present, this validated molecular method is the most recent and reliable tool for Cmm detection directly on seed extracts and it is also approved by NSHS (So 2.1). A tool for the identification of Cmm colonies is the barcoding on a specific 350 bp region in 16S rDNA gene, but more rapidly, a Rep-PCR can be enough to identify the isolates in an axenic culture [30,98]. A LAMP PCR for the detection of Cmm directly on seed extracts was set up by using primers designed on the gene micA, coding a toxin towards the other subspecies of C. michiganensis [99]. This assay allows the detection of Cmm up to 10⁴ CFU g^-1 seed. The study compared the detection ability of the LAMP method with Immunostrip: both were able to detect Cmm in the seed samples. In a previous work, the same authors set up the LAMP method on pure Cmm cultures and, then, they compared the results with the available methods: in some cases, the conventional and the qPCR failed the detection; on the contrary, LAMP PCR detected Cmm in all the samples prepared, also the avirulent strains [100]. Among the new molecular techniques requiring a validation, Morcia et al. [101] set up a ddPCR, conceptually similar to qPCR - but more sensitive - which is able to detect Cmm (and the Ralstonia solanacearum species complex, Rssc) directly from symptoms and from stems; nevertheless, the authors did not evaluate the robustness of such method directly on seeds. The primer for the ddPCR were designed on 16S-23S region and the did not amplify other bacterial pathogens of tomato, as the Xanthomonads or Pst. For the detection of Viable But Not Culturable (VBNC) Cmm strains, a study was carried out by improving a qPCR [102].

For the detection of Xanthomonads, two duplex end-point PCR methods are suggested, both together, to cover all pathogen species and pathovars causing the bacterial spot [56]. The method is specific, using probes designed on sequences from an unknown fragment obtained by the Amplified Fragment Length Polymorphism (AFLP), but it cannot be directly applied on seed extracts because of its high detection threshold (approx. 10⁵-10⁶ CFU ml^-1). However, it is suggested and adapted by EPPO [56] to perform it on Xanthomonas spp. axenic cultures [103]; in addition, the primers to detect Xhg can fail on Iranian strains, questioning the usability of these primers in all geographic areas [65]. The ISHI-Veg method, even if more updated, suggests two multiplex qPCR methods (without references) on axenic cultures as well. In both official protocols, the direct isolation and the PCR on the axenic culture can validate the analysis; moreover, though it is just recommended, a pathogenicity assay may be done to confirm the isolate identity. Although not yet validated, a multiplex qPCR was also developed, aiming at the detection of the whole set of Xanthomonas spp. that are described as the causal agents of the bacterial spot [104]; the primers were designed on the hrpB7 gene of the Type III Secretion System (T3SS), and they were highly specific but, again, the method was not tested on seed extracts. Moreover, an isothermal reaction, through a Recombinase Polymerase Amplification assay (RPA) was also developed and implemented by using primers designed on the hrpB (or hrcN) gene and evaluated for its use in the field (in situ) on symptomatic plants; such technique showed a sensitivity threshold of about 10⁶ CFU ml^-1, which is acceptable during the analysis of symptomatic tissues, though the primers for Xep did not detect their bacterial target [105].

Pst detection and identification lack validated molecular methods and official guidelines for a diagnostic analysis: the reason is that this organism is not regulated in large part of the world (EPPO Global Database, https://gd.eppo.int/). Nevertheless, a few methods based on end-point PCR and qPCR are reliable and available; in particular, Zaccardelli et al. [106] designed primers on the hrpZ gene and developed a specific assay to be used on pure cultures, symptomatic material, symptomless transplants and experimentally contaminated seeds. The calculated detection threshold evaluated on crude extracts of leaves and seeds resulted in approx. 10⁵ and 10⁶ CFU g^-1 for leaves and seed, respectively. A qPCR for Pst detection was studied using a molecular beacon (MB) PCR by Fanelli and collaborators [107]; the method, using primers designed from a RAPD unknown fragment, reached a sensitivity threshold of approx. 10² CFU ml^-1, both in a pure culture suspension and in seed extracts. Seeds may contain bacteria in their Viable But Not Culturable (VBNC) state [108]: this mainly because of the long quiescent phase, or because sanitation treatments that temporarily block the culturability of bacteria. Therefore, the isolation might result in the underestimation of bacterial contamination or, possibly, to a false negative detection. This impacts on the evaluation of the efficacy of disinfection methods, where it is not generally possible to discriminate dead from viable germs. Then, molecular methods could be improved by adding propidium monoazide (PMA) as was performed for Cmm, where the authors set up a new qPCR method to detect VBNC Cmm in seed samples [102]. Recently, a qPCR amended with PMA was also set up for the detection of Xanthomonads in tomato seed batches [109].

6. Seed treatments

Seed is a recognised and very efficient pathway for the introduction and dissemination of several pests and pathogens [110] and this is particularly true for tomato, as the most important vegetable [111]. Therefore, to ensure their phytosanitary quality and allow their safe trade worldwide, seeds should be treated (or disinfected or sanitised) to reduce the presence of pathogens to a minimum [44]. An excellent example of hygiene in tomato seed production and pathogen control of Cmm is published by the international business chain system, GSPP, to prevent tomato seed to be infected by Cmm GSPP [112]. Therefore, seed is commonly treated/disinfected before commercialization or movement to ensure the production of good seedlings/transplants, to minimise yield losses, to maintain and improve crop quality, and to avoid the spread of harmful organisms [113]. However, the production of healthy high value tomato and pepper seeds should consider all the phytosanitary challenges. Indeed, seed production with healthy stock seeds tested negative for seed borne pathogens and in confined production areas can promote quality and healthy seeds [112]. These production strategies can be also used to reduce the input of copper-based compounds in the field during the cropping season, thus allowing a more sustainable pest management. In fact, bacterial diseases are difficult to manage, once affecting the crops, partly because there are few effective pesticides [114,115]. Copper (Cu) is indeed a fundamental tool in conventional and Integrated Pest Management (IPM) farming systems, despite its several limitations in the European Union [116]. These limitations on Cu-based pesticides are applied for copper, as a heavy metal, accumulate in the environment (soil and water bodies) and, therefore, may have a deleterious impact on biodiversity. Moreover, the presence of copper-resistant or copper-tolerant phytopathogens, such as Cmm [117], Xanthomonas species [118,119] and Pst [120] is currently raising great concern and is making the control of these bacterial diseases quite cumbersome, once they are established in the field. As a matter of fact, the best way to ensure a fair and safe international trade of seed commodities, to avoid the burden to increase the use of copper-based pesticides, and to allow a sustainable and remunerative crop management and yield, is making “clean” seeds available to farmers. This goal may be reached by applying a set of disinfection methods along several, different sanitation procedures. Physical, chemical, or biological methods for seed treatment have been proposed, from time to time, for eliminating or reducing the bacterial seed-borne inoculum (Table 1), as a primary management strategy to prevent disease outbreaks or epidemics [121].

6.1. Tomato and pepper seeds extraction procedures

Botanically, tomatoes and peppers are berries but, though belonging to the same Solanaceae family, they show quite different fruit morphology and structure, and such features impact on methods and procedures to extract and obtain seeds. Basically, tomato seeds are attached to the placental tissue surrounded by a thick layer of jelly polysaccharides and immersed in the locular cavity that, in ripe fruits, is filled by such polysaccharide matrix. In peppers, seeds are also protruding into the locular cavity, but without such a surrounding polysaccharide matrix and, additionally, the locular cavity is and remains empty in ripe fruits. Therefore, after opening ripe peppers (manually or mechanically) seeds are easily detached from the placenta, whereas in tomatoes the cutting or breaking of fruits results in the production of a juicy mass containing the seeds. Harvested peppers can be kept at room temperature to increase seed maturity; later, the fleshy walls are cut and seeds are manually scraped off. In case of large batches, unscraped cores or small whole fruits are covered with water and broken up with a mixing blade; then, mature seeds precipitate and the broken pulp, together with the immature seeds, remain floating in the water and can be discarded. Mature seeds are then rinsed and dried at room temperature under circulating air flow [122].

Large scale production of basic, certified or commercial tomato seeds is, in general, fully mechanised; the ripe (frequently overripe) fruits are mechanically harvested and passed into a crusher that squashes the fruits into a mixture of pulp, juice and gelatinous seeds. The mixture is then passed into a revolving cylindrical screen that separates the juice and the seeds from the remaining fruit debris through an appropriate mesh. Juice and seeds are collected into separate containers and the gelatinous seeds are then separated from the remaining pulp and cleaned from the polysaccharide gel by (i) fermentation or by (ii) adding 10% solution of sodium carbonate or, alternatively, by (iii) addition of hydrochloric acid. After such extraction treatments, seeds are immediately washed and, finally, dried at room temperature under circulating air flow [123].

As previously described, phytopathogenic bacteria affecting tomato and pepper can be present on the seed coat, as well as within the innermost layer of the seed coat, where they are difficult to eradicate by treatments (i.e. chemicals) without damaging the embryo. These sources of infection have been the focus of numerous attempts to devise effective seed treatments for the elimination of bacterial pathogens.

Seed extraction from fruits is the initial challenging step during the seed production chain, where possible cross contaminations of seed borne pathogens may happen [124,125]. Indeed, bacterial spot (Xanthomonas spp.), bacterial speck (Pst) and bacterial canker (Cmm) produce lesions on the surface of fruits, and bacteria growing inside these lesions can contaminate seeds during the seed extraction process [63]. Then, through seed extraction, bacteria can easily contaminate the seed externally; indeed, tomato seeds have numerous epidermal cells that form hairs and crevices on the seed coat that may provide areas and niches in which the pathogen may reside and, later, evade disinfection treatments [126]. Then, the jellified, pectinaceous material surrounding seeds, covering the bacteria and protecting them once the seed has dried, are commonly removed through different manual or mechanical handling, such as fermentation, acid extraction, or pectinase addition to the washing solutions; those procedures are then followed by mechanical washing, drying, and winnowing the inert material from the seed [127,128]. Another important goal of seed extraction is to facilitate the separation of seeds from the surrounding pulp to increase their nutritive value, germination performance, seed quality parameters [129] and, finally, act as a seed sanitation/disinfection step.

Natural fermentation of tomato pulp at room temperature for 24-72 hours and manual seed extraction are commonly used for obtaining seeds [71], and such procedure has been found to decrease the proportion of infected seeds by Cmm in seed lots since many years [130]; regretfully, such an approach is not efficiently applicable for large scale seed production [129]. Nonetheless, Jones et al. [114] reported and stressed that tomato seed extraction by fermentation or acid extraction efficiently reduced the level of seed contamination by killing Xanthomonads. This was confirmed by Giovanardi et al. [64], who proved that the natural fermentation of tomato pulp obtained from symptomatic fruits generated healthy seeds that tested negative by agar plating and standard molecular assays [56]. Similar results, but referred to Pst, were reported by Chambers and Merrimans [131], where the phytopathogenic bacteria were isolated from tomato seeds mechanically extracted from symptomatic fruits, but not from seeds extracted from the pulp mass after either a natural or an acid fermentation process. Pradhanang and Collier [132] compared different fermentation techniques (i.e. natural fermentation and acid extraction, where pulp soaked in an equal volume of 5% HCl solution for 7 to 10 minutes, followed by immediate seed wash), and showed that this fermentation step was effective in decreasing the bacterial inoculum, but still not reliable by itself for completely eliminate Cmm, unless it is followed by a HCl treatment of dry seeds. More recently, Degwale et al. [129] compared natural fermentation to a HCl-based treatment of tomato seeds: their results highlighted the potential of acid extraction in 2% HCl for 60 min to control seed-borne pathogens, as confirmed by the absence of any detectable microbial epiphytes.

For pepper, due to the anatomical and physiological characteristics of fruits and seeds, the most common seed extraction methods manual or mechanical, with the possible use of a disinfectant during the extraction [133]. However, literature is very poor, or even absent, on the possible impact of the different extraction methods on bacterial seedborne pathogens affecting pepper seeds.

6.2. Seed sanitation methods and procedures

The use of “clean” seeds (e.g. sanitised and/or certified) is the most important preventive measure to control seed transmitted diseases, particularly those caused by phytopathogenic bacteria: this approach is considered the simplest, yet the cheapest and safest method of managing different seed-associated pests [113]. The need and necessity to provide clean seeds to the growers led to decades of research on developing and implementing seed treatments [134,135,136]. Nonetheless, the efficacy of currently available treatments for seed-borne bacteria (e.g. thermotherapy, use of chemicals, application of bio-inoculants) are often partially effective, depending on the location of the pathogen on or in the seed; therefore, the control of seed-transmitted bacterial diseases is still a challenge in both organic and conventional production [137]. Moreover, the application of any seed treatment should take into proper account the methods and procedures used for seed extraction and the specific conditions required to keep the embryo vitality and the quality of seeds; these are fundamental aspects in crop production, since uniform seed germination and high seedlings vigour contribute to the successful crop establishment and performance [138]. Finally, treatments should not be applied to pelleted seed, or to seeds that have been previously treated, or to old or poor-quality seed, because germination can be significantly reduced by the treatment and/or the pathogen may not be eliminated completely.

6.2.1. Chemical seed treatments

Chemical-based seed treatment methods, such as the application of generic disinfectants or bactericides, are frequently used to reduce the incidence (or the level of contamination) of pathogens in seeds [139]. Their application and efficacy for controlling bacterial pathogens are well known since decades [140]. Chemical treatments are often simpler and more convenient to use, though care must be taken to use the minimum concentration of chemical(s) required for the minimum exposure time, as these can also decrease seed vitality [141]. Chemical seed treatments are effective and commonly used for seedborne fungal diseases as well [142]. For the elimination of phytopathogenic bacteria associated to seeds, the most used chemicals for treatment are sodium hypochlorite in an aqueous solution (diluted NaHCl, or chlorine, or bleach) and different acids [134,143]: these were found to be efficient in disinfecting tomato seed from Cmm [144,145]. The use of chlorine for the disinfection of tomato and pepper seed is common, but often the used methods do not consider additional complex factors, which can affect results and compromise seed quality (e.g. the presence of a mass of organic matter, like the tomato pulp). One of the best sources for information on using chlorine as a disinfectant is “White’s Handbook of Chlorination and Alternative Disinfectants” a 1062-page compilation of chlorine information [146].

The application of diluted NaHCl in tomato seed treatments can vary from 0.5% to 20% (v/v) in a water mixture and the duration of soaking is from 2 to 40 min [54,147]. In all cases, the seeds should be thoroughly rinsed in water after soaking to minimise any negative impact on seed vitality [148]. A chlorine concentration equivalent to less than 1% bleach can kill bacterial spores within 5 to 10 minutes [149], therefore, since Clavibacters and Xanthomonads are non-sporigenic germs the same concentration is efficiently applied as disinfectant. Finally, McFarquhar [150] showed the potential of diluted NaHCl in disinfecting artificially contaminated pepper seeds by Xee.

Inorganic and organic acids are the most popular in organic seed production for their effectiveness as sanitation treatments for both tomato and pepper [151]. The hydrochloric acid (HCl) is also commonly used as a sanitising chemical in seed production [144]. In fact, tomato seed can be treated with a diluted solution of HCl at different concentrations (i.e. 1 to 5%) and for different soaking times (i.e. 1 to 60 minutes), to eliminate seed-borne bacterial pathogens such as Xanthomonas spp., Pst and Cmm [140,145].

Other than HCl, acetic and peracetic (peroxyacetic) acids [141,152], and hydrogen peroxide are often used in combination in commercial formulation (e.g. Tsunami 100; Oxidate 2.0), were found to reduce pepper (Xanthomonads) and tomato (Cmm, Pst and Xanthomonads) pathogens by 80.3% and 99%, respectively [149,153].

Another common chemical disinfectant for tomato seeds is the trisodium phosphate (TSP). Its application allowed to reduce or even sanitise epiphytic bacterial pathogens through seed soaking in a 10% aqueous solution (v/v) for 15 minutes, followed by rinsing in a cold-water bath for 5 minutes to remove the residual disinfectant prior to seed drying [154].

6.2.2. Physical seed treatments

Physical treatments of tomato and pepper seeds include thermal treatments with hot water, aerated steam, hot dry air, ozonisation and ultraviolet (UV) irradiation. These treatments have a number of advantages over other treatments: in most countries they (i) do not require registration or approval, (ii) have a wide spectrum of activity, (iii) do not leave any toxic or polluting residues, (iv) are considered sustainable and eco-friendly [155].

Phytopathogenic bacteria are, in general, non-sporing, and most of them do not grow at a temperature higher than 41°C: indeed, the most thermophilic phytopathogenic bacteria have an optimum growth temperature between 31-36°C [156]. Temperatures approaching 48-50°C are then very deleterious for non-sporing, phytopathogenic bacteria, which are not able to survive: therefore, the most common temperature range for seed disinfection is between 48-55°C [53,140]. For tomato, no effect on germination was observed when seeds were heat-treated up to 55°C in humid conditions [157].

6.2.2.1. Heat treatments

Heat can be applied to seeds or other plant material as humid treatment (hot water or hot steam) or in dry conditions (ventilated oven), or both. Hot water soaking is an old-age practice, which is usually effective for the sanitation of seeds of different plant species [152,158]. Although theoretically easy to perform, a heat treatment can be difficult to conduct because of the need to fine-tune the temperature required to kill the pathogen(s) without impacting the germination potential, seedling vigour and shelf life of the seed [141,144,159,160]. Moreover, treatment protocols implemented for small seed lots may be difficult or even impractical for large-scale production and there is no guarantee that all the seeds in a large treatment batch will attain the treatment temperature for the recommended duration [161].

Heat treatment may be applied for agricultural commodities by (i) immersion in hot water, (ii) exposure to vapour heat, (iii) exposure to hot dry air. Hot water treatments of seed and plant material are classical thermophysical methods of plant protection and are more eco-friendly and more effective than chemical treatments. In particular, heat can kill bacteria as seed endophytes, whereas chemical treatments are mostly effective against phytopathogens residing on the seed coat or, in a few cases, just below it. A compilation of hot water seed treatment conditions for different vegetables, including tomato and pepper, is publicly available [162]. Several examples of hot water treatment for seed disinfection by bacterial pathogens and their efficacy are listed in Table 1. The main disadvantage of hot water treatments is the need for post treatment drying.

The application of dry heat to seed batches is equally effective as wet heat; interestingly, the increase of temperature in dry conditions is higher than in humid conditions: Marinescu [163] demonstrated that 1 h at 53°C in wet conditions had a potential to disinfect tomato seeds from Cmm equal to 1 h at 80°C in dry air flow. Dry conditions are less dangerous for seeds and heat exposure can be much longer than in humid conditions: indeed, Murata et al. [164] showed that dry heat treatment at 70°C for 4 to 6 days allowed the disinfection of tomato seeds from Cmm. Therefore, also dry heat treatments, which have been developed and implemented several years ago, are common practices applied to solanaceous seeds, as reported by Li et al. [165] and by Akman [166] to control external and internal seed-borne pathogens such as fungi and bacteria.

6.2.2.2. Ozone

Ozone, O₃, is a reactive oxygen species (ROS) with a very high antimicrobial potential, and ozonisation is known to be an excellent method for disinfection, thus ensuring food hygiene, sterilisation of utensils, facilities and warehouses, and sanitation of drinkable water [167]. The strong potential of ozone in eliminating or reducing seed-borne bacteria and fungi in seeds has been demonstrated in recent years. Sterilisation performance of O₃ and its effects on seed germination closely depend on the concentration and duration of the specific treatment, the type and structure of the seed, as well as the target pathogen [168,169].

The mechanism behind the antimicrobial properties of ozone is attributed to either its strong oxidising capacity, or due to the generation of ROS during its decomposition [170]. Glycoproteins and lipids located in the bacterial and fungal cell membranes are destroyed directly by ozone or its byproducts [171]. The application of ozone to tomato seeds was attempted by Çetinkaya et al. [172]: they demonstrated the efficacy in disinfecting experimentally inoculated seeds with Cmm and Pst by a gaseous ozone treatment at 90 and 120 minutes, without any negative effect on seed germination rate. Therefore, ozonisation might be a very promising treatment, cheap and effective, to eliminate phytopathogenic bacteria without affecting seed quality.

6.2.2.3. UV-C light irradiation

Finally, other methods based on physical principles have been implemented for seed treatment: irradiation by UV-C light has been used as a way to sterilise seed surface [173] and there is evidence that some seed-transmitted fungi may be killed by an appropriate UV-C treatment [174,175]. Regretfully, UV-C irradiation does not have a significant, direct effect on seed endophytes, particularly bacteria, despite some authors reported that an UV-C treatment was able to induce a resistant state in cabbage against Xanthomonas campestris pv. campestris [176] but without showing a direct antibacterial effect against the pathogen. Therefore, such technique does not suit the purpose of sanitising/disinfecting seeds, as a possible pathway for pathogen dissemination.

6.2.3. Microorganisms and Natural Products

Chemical seed dressing against seed-borne pathogens or against soil microorganisms that may affect seed germination and seedlings development is mainly done against fungi, e.g. Fusarium spp. for cereals [177] or Ascochyta pisi for peas [178]. For seed-borne bacteria there are far less options for their control [86]. Effective pesticides for seed dressings, and fungicides among them, are available in the global markets, but the increasing trend and consciousness for a more sustainable agriculture are inducing the policy makers of several countries to revoke such active substances for their intended use. For instance, Thiram, a very popular fungicide that was used for several decades for seed dressing, is now prohibited in the European Union [179].

Nowadays, biological seed treatments represent one of the fastest growing sectors under seed treatment strategies worldwide. Bio-inoculants can protect plants against different seed-borne pathogens through numerous biological mechanisms, such as antibiosis, competitive exclusion or inducing systemic resistance in host plants [180,181,182]. They represent a cost effective, alternative technology to chemical-based plant protection with the potential for a large-scale application by maintaining seed quality and promoting healthy plant growth in sustainable agricultural systems [183]. Thus, the demand for biological solutions for seed treatment is increasing, also in view of consumers’ acceptance for chemical-free food.

Beneficial microorganisms or natural products can be applied through coatings onto the seed surface for the protection of seeds and seedlings from seed-borne and soil-borne pathogens [184]. For seed treatments, various methods are used, such as seed coating, seed pelleting, seed washing, or seed dressing, which usually consists of applications where liquid bio-inoculants (generally supplemented with adhesives) are sprayed onto the seeds either by hand, or rotating drums, or automated seed coaters, followed by drying with forced air using a seed drying equipment.

Recently, the application of bio-inoculants to seeds has increased because of increasing public concern for safety issues and awareness for a healthy environment, sustainable development, and health hazards caused by the excessive use of agrochemicals [185]. However, research on the commercial use of beneficial microorganisms as seed inoculants is limited, and there are only a few examples of their commercial development, implementation and use. Indeed, only few biopesticides formulations for seed coating are currently registered and specifically commercialised in Europe for seed-borne diseases, e.g.: Cerall (Pseudomonas chlororaphis) marketed by Serbios srl, Italy, and Mycostop (Streptomyces griseoviridis K61) marketed by CBC Bioplanet, Italy (http://ec.europa.eu/food/plant/pesticides/eu-pesticides-database/public/?event=homepage&language=EN). The latter is also true for botanical seed treatments, despite many reports of bactericidal and fungicidal effects of compounds from plants. The reason is possibly due to commercial constraints, like the high development and registration costs in relation to market size [183].

Biopesticides registered for tomato or pepper against seed-borne bacteria, such as Cmm, Pst or Xanthomonads are still few or even absent. In the European Union, the regulatory framework for the registration of microbial biocontrol agents derives from that one currently used for chemical pesticides [186]. Therefore, there are still safety issues to be tackled, e.g. the dynamics and fate of such microbes in the environment, their possible classification as GRAS (Generally Recognised as Safe) (https://www.fda.gov/food/food-ingredients-packaging/generally-recognized-safe-gras), their impact on biodiversity, etc. The cost of producing a thick dossier including the data required by the regulatory institutions may discourage the registration and marketing of biopesticides.

Nevertheless, different studies have been published reporting the potential of plant extracts or microbial biocontrol agents applied as seed bio-inoculants (Table 1). Among these studies, only a limited number of reports include all scales (e.g. laboratory, greenhouse, and commercial fields), even if different seed sanitation methods and procedures showed to be effective up to 100% in eradicating bacterial pathogens under controlled conditions. For instance, Kasselaki et al. [137] showed that the application of Bacillus sp. as seed bio-inoculant, was able to reduce the presence of Cmm up to 100%. Concerning natural products, Mbega et al. [136] found that 4 plant extracts (i.e. Aloe vera, Coffea arabica, Glycyrrhiza uralensis and Yucca schidigera) out of 84 were able to experimentally disinfect tomato seeds inoculated with Xep in both in vitro and in planta assays.

In other studies, the inconsistency of the seed testing results with the in planta performance can be considered as one of the main challenges for the wide application of seeds coated with bio-inoculant (Table 1). Thus, results that clearly demonstrate and validate the efficacy of the delivery system and the microbial application covering all stages of the process are essential. Among other constraints concerning the microbial application techniques, it may be assumed that seeds can be coated only with a limited amount of inoculant, which might be a limiting factor for such treatment, since a threshold of a mBCA may be needed for a successful biological control. Again, factors influencing microbial viability and survival on seeds are the release of toxic exudates from the seed coat or the incompatibility between the inoculant strains and other seed-applied chemicals [183]. Microbial viability and survival may be improved by the coating materials (e.g. carriers and binders), which could also foster the performance of the target crop [187]. Since different beneficial microorganisms may react differently to coating, the development of coating materials that are compatible with a wide range of bio-inoculants is then crucial to the industry. Finally, the formulation of bio-inoculants should also take into proper consideration the local agriculture conditions and practices (such as soil conditions, use and nature of pesticides applied, fertilisers, and irrigation management) to optimise their performance and survival [183].

7. Conclusions

In this review, we described the epidemiology, seed health assays and treatments of the most harmful seed-borne phytopathogenic bacteria affecting tomato and pepper. These bacteria are responsible for the emergence or re-emergence of disease epidemics, through their movement across international borders and their introduction into new areas, thus limiting crop production and threatening the seed industry worldwide.

According to the Farm to Fork Strategy, part of the European Green Deal, limitations on chemical pesticides foster the importance of an integrated disease management aimed to avoid the spread of these pathogens and to prevent the establishment of trade barriers between countries. Therefore, the implementation by the seed industry of international science-based standards for phytosanitary measures are essential to maintain seed quality and prevent the risk posed by the short to long dissemination of pathogens. Among these standards, seed production strategies (e.g. GSPP), diagnostic protocols for seed analysis and certification, seed extraction and treatments are playing a key role for the seed industry. Moreover, non-chemical seed treatments, which include physical methods and microbial applications, together with the use of other antimicrobial agents of natural origin, like plant extracts, meet the requirements of the new agricultural policies and the market's requests. In fact, the demand for biological seed treatment solutions is increasing in view of consumer acceptance for chemical-free food. Nevertheless, further research and investigations are needed to seek for new solutions for innovative, sustainable and field-applicable Integrated Pest Management (IPM) strategies against tomato and pepper seedborne phytopathogenic bacteria.