Recently, great advance in the knowledge of genetic aspects of TRM has been made with the availability of its complete genome, and that can now be explored to support the development of control strategies (Greenhalgh et al. 2020). Despite this, so far, there was not any information on the genetic variability or structure of this invasive pest along its occurrence areas. This can be crucial for pest management programs, since populations can develop different bioecological traits, such as host preference, susceptibility to biotic and abiotic stressors and cause different responses in the host plant (Porreta et al. 2007; Remais et al. 2011; Migeon et al. 2021). Furthermore, there were still doubts about the taxonomic status of the taxon as cryptic species could occur among populations associated with different host plants. For the first time, here, these crucial aspects were explored by accessing populations from different continents, and from cultivated and wild solanaceous host plants by novel approach to understanding evolutionary aspects and to support the development of TRM management practices.
TRM confirmed as an oligophagous eriophyid mite taxon
Both genetic diversity and phylogeny showed that TRM studied populations, associated with different solanaceous host plants, are co-specific. The divergence for the mithocondrial and nuclear studied genomic regions among studied populations/haplotypes was even lower than that observed for intraspecific variability in other eriophyid taxa. The highest COI divergence among haplotypes was 0.43% (i.e. < 0.5%) (Table 5). Such distance corresponded with COI intraspecific genetic levels identified by Duarte et al. (2019) for seven Abacarus species (the same tribe that Aculops) in the same gene, e.g., 0.0 to 0.7%. and close that observed by Skoracka et al. (2012) for Aceria tosichella Keifer, 1969 species complex (0.4%). The usual limits of COI intraspecific divergence in more than 13 000 congeneric pairs including representatives from 11 phyla have usually been lower than 2% (Herbert et al. 2003) and most was less than 1% (Avise 2000). Concerning interspecific distances for Prostigmata mites (Eriophyidae sub-order) it has been higher than 4.3% for species in different families (Ros and Breeuwer 2007; Matsuda et al. 2012; Duarte et al. 2019; Pérez-Sayas et al. 2022). Similar results were obtained for the ITS: the average divergence between TRM sequence variants/populations was 0.02% (Table 6), absolutely corresponding to intraspecific distances when in comparison to divergences observed for other eriophyid mites. Duarte et al. (2019) observed ITS intraspecific diversity in Abacarus species ranging from 0.1 to 0.3%, and interespecific distances ranging from 4–17.9%. ITS divergence of 2% have been interpreted as discriminating lineages within A. tosichella populations (Skoracka et al. 2012). In this study, no variability in the nuclear D2 region was observed among the TRM populations, similar to that observed in Abacarus species, which reached 0.3% (Duarte et al. 2019).
Phylogenies, inferred from the three DNA fragments separately and from the Bayesian combined analysis, emphasize the close relationship among the TRM populations studied. The low support to nodes shows that populations cannot be consistently distinguished. Therefore, the hypothesis of the occurrence of cryptic species among TRM populations associated with different host plants was not corroborated for the studied populations.
Although populations from all plants reported as TRM hosts have not been evaluated in this study, the results obtained allow us to confirm that A. lycopersici is an oligophagous eriophyid mite that can infest at least two Solanaceae genera- Solanum (with at least six species) and Physallis.
High majority of eriophyid mites inhabits a single host plant species indicating close host-relationship. According to Skoracka et al. (2010), about 80% of eriophyoid species occupy one host plant species, about 95% infest plant species belonging to one genus, and about 99% to one host family. TRM can be ascribed in 5% of Eriophyoidea mites infesting host plants in two different genera. Further studies including populations from Convolvulus (family Polygonaceae), also reported as a host, should be performed to define if TRM is among the 1% of species that can infest plants in more than one family. The evolutionary and molecular aspects that allow eriophyoid mites to adapt to different host plants could be studied by comparing TRM genome with those of eriophyid mites presenting high host specificity though considering that the whole genome of eriophyoid mites is currently available only for TRM (Greenhalgh et al. 2020).
For many years, Aceria tulipae Keifer, 1938 had been treated as one exceptionally generalist eriophyid. Keifer (1969) and Shevtchenko et al. (1970), showed, that Ac. tulipae found on Liliaceae was, both morphologically and biologically, different from Ac. tulipae inhabiting wheat. As a consequence, a monocot-infesting species Ac. tosichella has been described as a separate taxon from Ac. tulipae inhabiting Liliaceae plants. Species in the Ac. tosichella complex have different host range or preferences (Carew et al. 2009; Skoracka et al. 2012, 2018; Navia et al. 2013), and abilities to transmit plant viruses (Schiffer and Lachmuth 2009; Skoracka et al. 2014; Wosula et al. 2016). Similar observations were made on other supposed generalists inhabiting monocots: Abacarus species (Nalepa, 1896) infesting grasses (Skoracka and Dabert 2010); Trisetacus species infesting conifers (Lewandowski et al. 2014); Retracrus species infesting palm trees and heliconias (Navia et al. 2015). Therefore, many monocot-associated eriophyid mites, previously considered generalists, have showed to constitute complexes of cryptic species. However, the supposed low-specificity in eriophyid mites associated with dicotyledons plants has not been investigated. For instance, no studies investigated the occurrence of cryptic species on Calacarus citrifolii Keifer, 1955, the species with the wider host range of all superfamily reported on 21 plant families (de Lillo and Amrine - Computerized Database for Eriophyoidea, Filemaker Pro). Among species of economic importance, TRM does stand out to be an oligophagous species.
TRM genetic homogeneity in Europe revealing a highly invasive haplotype
Analysis of the diversity of sequences of the COI mitochondrial region showed that haplotype 1 (cH1) is dominant in Brazil, is associated with all host plants considered, and that it is the only one present in European populations (France, Italy, Poland and The Netherlands). The genetic homogeneity among TRM European populations highlights that cH1 is a highly invasive and not host-specific haplotype. This adaptation to infest a high number of plants, which enhances the range of pathways in mites’ transport, certainly has favored TRM invasiveness and wide distribution.
A new introduced population can be composed of a sub-sample of genotypes, from one or several populations in the origin range. No intraspecific variation in the invaded range suggests i) single small initial invasive population, ii) multiple invasions of the same source, or iii) adaptative selection of the haplotype that has been established. The relationship between the success of invasive populations and their genetic diversity has been discussed for a long (Lee 2002; Estoup et al. 2016), by assessing genetic diversity as positive factor influencing survival and adaptation in invaded areas (Lee 2002; Petit et al. 2004; Puillandre et al. 2008). However, absence or low genetic diversity, caused by a population bottleneck, has been reported for many invasive species (Le Page et al. 2000; Sax and Brown 2000; Martel et al. 2004; Novak and Mack 2005; Puillandre et al. 2008), including mites (Navia et al. 2005; Soulignac et al. 2005; Boubou et al. 2012; Dowling et al. 2012) similarly to observed to TRM in this study. These studies suggested that even genetically homogeneous founder populations may retain the ability to respond to natural selection, adapt and expand in the invaded area. In such cases, the success of the invasion has been considered as the ‘genetic paradox of invasions’ (Sax and Brown 2000). For some invasive species, the paradox has showed to be spurious, as seen in introduced populations with low diversity in neutral markers that maintain high genetic variation in ecologically relevant traits. However, in other cases, it can be considered genuine (see Estoup et al. 2016): compensatory mechanisms maybe acting to counter the loss of genetic variation and unique aspects of the species’ biology, as well as environmental interactions that could allow an invasive population to thrive, have been showed or proposed (Estoup et al. 2016; Schrieber and Lachmuth 2016; Eyer et al. 2018; Marin et al. 2019). Further studies need to be conducted to confirm genuine genetic paradox in TRM invasion and to better understanding evolutionary strategies that allowed its invasion success with no genetic variability.
The genetic homogeneity evidenced among TRM European populations do not primarily corroborate the hypothesis that a differentiated host response observed on different host plant species/varieties could be due to the genetic diversity of the associated mite pest. The results suggest that the differentiated symptomatology and damage intensity caused by TRM infestation is due to other biotic or associated abiotic factors related to host plant physiology- e.g. metabolic pathways and mechanisms in plant defense in different hosts (see Glas et al. 2014; Kant et al. 2015), or host response to drought stress, that in some cases can promotes the colonization success of eriophyid mites (Ximénez-Embún et al. 2017).
Towards TRM origin, host adaptation and invasion process
Most of the knowledge about the introduction routes of invasive species is derived from historical and observational data, which are often sparse, incomplete and, sometimes, misleading (Estoup and Guillemaud 2010). This difficulty is remarkable for the tiny eriophyoid mites, for which relevant historical records sometimes are scarce and often incomplete, since they can go unnoticed for a long time and be reported, out of time, when outbreaks occur and in taxonomical confused context (Navia et al. 2010). One such case is A. lycopersici, for which worldwide historical reports and taxonomic mistakes have not been allowed to timely and thoroughly understand its geographic expansion. The mite was described from Australia in 1917 (Tryon 1917); then it was erroneously described as new in 1937 in North Africa (Morocco) (Massee 1937); in 1940s it was for first reported in North America (USA) (Keifer 1940) and in Europe (Spain) (Planes 1941); in 1950s in the Middle East (Lebanon) (Talholk 1950) and in Asia (Georgia) (Tukalevskii and Rogachev 1959) and in the 1960s in South America (Brazil) (Costa and Carvalho 1962) and South Africa (Ryke and Meyer 1960). Another junior synonym of A. lycopersici was described in 2005 in China as Tetra lycopersici Xue and Hong, 2005 (Amrine and de Lillo, personal communication). In this context of historical untraceability patterns genetic data can be explored for tracing invasion routes (Estoup and Guillemaud 2010).
Although the objective of this work was not to trace invasion routes, since a better representation of populations worldwide would be necessary for this purpose, genetic diversity and a phylogeographical analysis based on both mitochondrial and nuclear markers analyzed jointly with the main host plant history corroborate the hypothesis of a South American origin for TRM. In general, the highest genetic diversity for a species occurs in its area of origin. Our results showed a highest TRM genetic diversity in Central Brazil; four haplotypes and five sequence variants were present in the only studied area of Central Brazil while just one haplotype and two ITS sequence variants were present in populations of four European countries. Also, it is interesting to note that the haplotype cH2 was found exclusively associated with non-tomato solanaceous- S. americanum and Physalis- in Brazil.
Under the assumption that tomato does not constitute the original TRM host plant, but some other wild solanaceous plants (Oldfield 1996; Michalska et al. 2010; Navia et al. 2010) and that currently this mite presents a worldwide distribution (CABI 2022) the most likely is that TRM adopted tomato as an alternate host plant in areas of co-occurrence of the original wild host plant and the cultivated tomato ant that this host adaptation occurred before expansion of tomato as a cultivated crop in the 16th century. The cultivated tomato has its origin in central South America (Blanca et al. 2012, 2015; Razifard et al. 2020), and then it presented a complex domestication history in South America and Mesoamerica (Razifard et al. 2020); by 500 BC, it was already being cultivated in southern Mexico (Smith 1994), and Aztecs raised several varieties of tomato (Townsend 2000). However, expansion of tomato cultivation to other colonies in the Caribbean and to the old continent (firstly Asia and Europe) only occurred at the beginning of the 16th century by Spanish colonizers (Smith 1994). The results of this first study support with genetic data the Neotropical origin of TRM however, it is not possible to know whether the mite was disseminated during the colonization period or afterwards, through the exchange of plant material. Phylogenomic analysis, including populations along wide tomato range in South and Central America from as many wild host plants as possible will enable to enlighten evolutionary history of this intriguing eriophyid mite that became a tomato and solanaceous pest. It is possible that the tomato domestication and breeding process may have unintentionally selected materials with higher susceptible to TRM. Determination of the original TRM host plant, supposedly better adapted to TRM herbivory, may be useful for revisiting breeding programs focusing in resistance.