• Loading metrics

Cucumber mosaic virus infection as a potential selective pressure on Arabidopsis thaliana populations

  • Nuria Montes,

    Roles Formal analysis, Investigation, Writing – original draft

    Current address: Fisiología Vegetal, Departamento Ciencias Farmacéuticas y de la Salud, Facultad de Farmacia, Universidad San Pablo-CEU, Boadilla del Monte (Madrid), Spain and Servicio de Reumatología, Hospital Universitario de la Princesa, Instituto de Investigación Sanitaria (IIS-IP), Madrid, Spain.

    Affiliation Centro de Biotecnología y Genómica de Plantas (UPM-INIA), and E.T.S.I. Agronómica, Alimentaria y de Biosistemas, Campus de Montegancedo, Universidad Politécnica de Madrid, Pozuelo de Alarcón (Madrid), Spain

  • Carlos Alonso-Blanco,

    Roles Conceptualization, Investigation, Writing – review & editing

    Affiliation Departamento de Genética Molecular de Plantas, Centro Nacional de Biotecnología, Consejo Superior de Investigaciones Científicas (CNB-CSIC), Campus Universidad Autónoma, Cantoblanco, Madrid, Spain

  • Fernando García-Arenal

    Roles Conceptualization, Funding acquisition, Investigation, Writing – review & editing

    Affiliation Centro de Biotecnología y Genómica de Plantas (UPM-INIA), and E.T.S.I. Agronómica, Alimentaria y de Biosistemas, Campus de Montegancedo, Universidad Politécnica de Madrid, Pozuelo de Alarcón (Madrid), Spain

Cucumber mosaic virus infection as a potential selective pressure on Arabidopsis thaliana populations

  • Nuria Montes, 
  • Carlos Alonso-Blanco, 
  • Fernando García-Arenal


It has been proposed that in wild ecosystems viruses are often plant mutualists, whereas agroecosystems favour pathogenicity. We seek evidence for virus pathogenicity in wild ecosystems through the analysis of plant-virus coevolution, which requires a negative effect of infection on the host fitness. We focus on the interaction between Arabidopsis thaliana and Cucumber mosaic virus (CMV), which is significant in nature. We studied the genetic diversity of A. thaliana for two defence traits, resistance and tolerance, to CMV. A set of 185 individuals collected in 76 A. thaliana Iberian wild populations were inoculated with different CMV strains. Resistance was estimated from the level of virus multiplication in infected plants, and tolerance from the effect of infection on host progeny production. Resistance and tolerance to CMV showed substantial genetic variation within and between host populations, and depended on the virus x host genotype interaction, two conditions for coevolution. Resistance and tolerance were co-occurring independent traits that have evolved independently from related life-history traits involved in adaptation to climate. The comparison of the genetic structure for resistance and tolerance with that for neutral traits (QST/FST analyses) indicated that both defence traits are likely under uniform selection. These results strongly suggest that CMV infection selects for defence on A. thaliana populations, and support plant-virus coevolution. Thus, we propose that CMV infection reduces host fitness under the field conditions of the wild A. thaliana populations studied.

Author summary

Plant-virus coevolution has not been demonstrated in any wild system, and it has been proposed that viruses often would be mutualistic symbionts, rather than pathogens, in wild plant ecosystems. We analyse if viruses are virulent pathogens of plants in wild ecosystems and, consequently, plants have evolved defences against virus infection. To test this hypothesis, we studied the genetic diversity of Arabidopsis thaliana for two defence traits, resistance and tolerance, to Cucumber mosaic virus (CMV) at a regional scale in the Iberian Peninsula. Resistance and tolerance to CMV showed substantial genetic variation within and between host populations, and depended on the virus x host genotype interaction, two conditions for coevolution. Resistance and tolerance were independent traits that co-occurred at the population and regional scales, and that have evolved independently from other adaptive life-history traits. Analyses also indicated that resistance and tolerance are likely under selection, most likely due to virus infection. These results support a hypothesis of plant-virus coevolution and contribute to demonstrate that plant viruses may be virulent parasites of plants in wild ecosystems.


It is commonly accepted that hosts and pathogens coevolve [1]. This concept rests on the assumption that pathogens are virulent parasites, defining virulence as the negative impact of infection on the host fitness. As a consequence, hosts will evolve defences to limit pathogen infection, or to compensate for its costs [2]. In plants, the two major defences against pathogens are resistance, defined as the ability of the host to limit infection and/or parasite multiplication and tolerance, which limits the fitness effect of a given parasite burden, i.e., specifically decreases virulence [3,4]. As host defences may reduce the parasite’s fitness, hosts and parasites may coevolve, coevolution being the process of reciprocally adaptive genetic change in two or more species [1]. Evidence for host pathogen coevolution is not abundant. For plants it derives mostly from studies of agroecosystems in which the pathogen evolves in response to the deployment of resistance in the host population [5]. These studies have provided the bases for theory on host-pathogen coevolution, including the gene-for-gene model of host-pathogen interaction [6,7]. However, coevolution requires certain conditions to be met [1]: i) genetic variation in the relevant host (e.g., resistance, tolerance) and pathogen (e.g., infectivity, virulence) traits; ii) reciprocal effects of the relevant traits of the interaction on the fitness of host and pathogen; iii) dependence of the outcome of the host-pathogen interaction on the combination of host and pathogen genotypes involved. These conditions must be analysed in wild systems, in which the host may evolve in response to environmental pressures, including pathogen infection, at odds with agricultural systems. Evidence for plant-pathogen coevolution from wild pathosystems is limited to a few instances, all involving fungal, oomycete or bacterial pathogens [8,9].

To our knowledge, plant-virus coevolution has not been demonstrated in any wild system. In fact, it has been proposed that viruses often would be mutualistic symbionts, rather than pathogens, in wild plant ecosystems [1012], and it has been shown that virus infection may be detrimental or positive for the host depending on the environment [13,14]. Hence the interest in seeking evidence about whether viruses are plant pathogens in wild ecosystems and viruses and plants co-evolve, or if virus virulence is the result of the specific conditions of agroecosystems. Reports of negative effects of virus infection in wild plants in their natural habitats are not abundant [e.g., [1522] and indicate that effects may largely depend on site or host population [23], but the genetic variation of defence and virulence has not been analysed in these systems.

To analyse plant-virus coevolution in wild ecosystems we have chosen the system Arabidopsis thaliana L. Heynh. (Brassicaceae)-Cucumber mosaic virus (Cucumovirus, Bromoviridae), (CMV). A. thaliana is an annual semelparous species with two distinct developmental periods: the vegetative growth period, producing a rosette of leaves, and the reproductive period in which the inflorescence grows, new flowers are produced continuously, and older flowers develop into siliques [24]. It is a cosmopolitan species, with a broad native distribution in Eurasia and Africa [2527]. The Iberian Peninsula has been shown to contain the largest A. thaliana diversity in Eurasia due to its colonization from different refugia after the last glaciations [2629]. In Iberia, A. thaliana occurs in a variety of habitats, and substantial genetic variation has been described, within and among populations, for relevant adaptive traits including phenological traits like flowering time and seed dormancy [3033]. Although wild populations of A. thaliana have been shown to contain ample genetic and phenotypic diversity for responses to herbivores and pathogens [3438], the diversity for traits related to plant-virus interactions has not been systematically analysed.

CMV is an RNA virus with the broadest host range including about 1,200 species in more than 100 plant families. CMV is horizontally transmitted by many species of aphids in a non-persistent manner, and through the seed with efficiencies that depend on the genotypes of CMV and the plant species [39]. In A. thaliana, seed transmission rates vary between 2 and 8% [40, 41]. CMV isolates are highly diverse and have been classified into subgroups IA, IB and II, based on the nucleotide sequence similarity of their genomic RNAs [39,42].

Analyses of the incidence of five viruses in six wild A. thaliana populations from central Spain during 10 years showed that CMV was most prevalent, up to 80% according to population and year [43,44], indicating that the A. thaliana–CMV interaction is significant in nature. As in other hosts monitored in the Iberian Peninsula, Subgroup IA isolates are most prevalent [4446]. Our group has analysed the role of resistance and tolerance in this interaction. The infection of 21 wild genotypes of A. thaliana representing the variation of the species in Eurasia with three CMV strains, showed that quantitative resistance to CMV depended on the interaction between host genotype x virus strain, and was a host trait with moderate to high heritability [47]. Virulence, estimated as the effect of infection on viable seed production, did not correlate with virus load, due to host genotype x virus strain-specific tolerance and, again, tolerance was a host trait with moderate to high heritability [47]. Interestingly, tolerance was positively correlated with the length of post-embryonic development (life span) of the host genotypes [47], and was due, at least in part, to host life-history trait modification upon infection: long life span genotypes delayed flowering upon infection and re-allocated resources from vegetative growth to reproduction, thus decreasing the effects of infection on progeny production, i.e., attaining tolerance [48]. It remains to be shown that defence polymorphisms result from the selection applied by CMV infection, and not by any other environmental factor known to modulate plant developmental architecture and phenology, such as life span, flowering time and plant size, which are known to have a role in adaptation of A. thaliana to the abiotic environment [4951].

In this work we study the genetic diversity of A. thaliana for resistance and tolerance to CMV at a regional scale in the Iberian Peninsula. To this end, we exploit a collection of 76 natural populations covering the wide ecological, environmental and genetic diversities of A. thaliana in this region [32,49]. We address the following questions: i) Which is the amount of genetic diversity for resistance and tolerance to different CMV genotypes? ii) Are the geographic and environmental climatic patterns of resistance and tolerance similar or different from those of related adaptive life history traits of A. thaliana? iii) Are resistance and tolerance traits under natural selection? Addressing these questions is crucial to determine if CMV infection has a negative impact on its host fitness under natural field conditions and, consequently, if there is coevolution between A. thaliana and CMV.


Genetic variation for resistance and tolerance to CMV in A. thaliana

To estimate the genetic diversity for resistance and tolerance to CMV, 76 A. thaliana wild genotypes collected in different natural populations from the Iberian Peninsula (Fig 1 and S1 Table) were assayed. We consider here a population as the set of A. thaliana plants growing in a specific geographical site. Plants were inoculated after eight-week vernalisation, using two CMV isolates from Iberian A. thaliana populations, Cdc-CMV and Lro-CMV. All 76 genotypes were systemically infected by both CMV isolates, no immunity or hypersensitive resistance reaction being detected.

Fig 1. Geographic distribution of Arabidopsis thaliana populations analysed in this study.

Circles indicate population locations. Circles with asterisk indicate the 12 populations used for within/between population analyses, their names appearing next to them.

Resistance was estimated from the levels of virus multiplication, quantified as virus RNA accumulation. RNA accumulation varied between 0.26 and 56.28 μg of virus RNA g fwt-1 (Table 1, Fig 2 and S2 and S3 Tables), variation significantly depending on the A. thaliana genotype (F75,816 = 2.98, P<10−4), the virus isolate (F1,816 = 113.30, P<10−4) and the interaction virus isolate x host genotype (F75,816 = 75.82, P<10−4), which together explained 88% of the variance (24.6% for A. thaliana genotype, 38.9% for virus isolate and 24.5% for their interaction). On average, Cdc-CMV accumulated to higher levels than Lro-CMV (19.28±1.37 and 7.53±0.71 μg virus RNA g fwt-1, respectively) (Fig 2 and Table 1). Virus accumulation in the host plant showed high heritability for both CMV isolates, with an average of 0.8 (Table 1), heritability being defined as the genetic component of the variance of the trait.

Fig 2. Variation for resistance and tolerance to CMV in A. thaliana.

Frequency distributions are for accumulation (μg RNA g fresh leaf weight-1) of (A) Cdc-CMV and (B) Lro-CMV RNA, and of the effect of infection by (C) Cdc-CMV and (D) to Lro-CMV on seed production.

Table 1. Values (mean and range of variation), heritability and autocorrelation, of life history and defence traits to CMV in the Iberian population of A. thaliana.

Tolerance was estimated from the effect of virus infection on progeny production. Since CMV infection does not affect seed viability or the weight of single seeds in a large number of A. thaliana genotypes [14,47,52], tolerance was estimated as the ratio of seed weight in infected to mock-inoculated plants (SWi/SWm), which varied between 0.03 and 0.79 (Table 1, Fig 2, S2 and S3 Tables). As for resistance, tolerance significantly depended on A. thaliana genotype (F76,816 = 3.10, P<10−4), virus isolate (F1,816 = 71.40, P<10−4) and their interaction (F75,816 = 5.18, P<10−4), which together explained 73% of the variance (26.9% for A. thaliana genotype, 25.0% for virus isolate and 21.2% for their interaction). Tolerance to Cdc-CMV (0.35±0.02) was lower than tolerance to Lro-CMV (0.51±0.02) (Fig 2 and Table 1). Tolerance in the host showed medium to high heritability, between 0.70 and 0.54 for Cdc-CMV and Lro-CMV, respectively.

No significant relationship was detected between virus RNA accumulation and SWi/SWm across genotypes nor within genotypes (r≤-0.11, P≥0.358). Together, results show that natural populations of A. thaliana contain substantial but independent genetic variation for resistance and tolerance to CMV.

Geographic and climatic patterns of A. thaliana resistance and tolerance to CMV

To determine if there is a geographic pattern for the genetic diversity for CMV resistance or tolerance, we first analysed the spatial autocorrelation of both variables. Neither virus accumulation nor SWi/SWm showed significant spatial autocorrelation at any geographic scale (P>0.050) (Table 1).

Second, to find if these defence traits might be associated with the climate, we analysed their relationship with climatic variables from the original local populations. Neither the accumulation of any of the two CMV isolates, nor the SWi/SWm ratio in plants infected by any of them, significantly correlated with any of the analysed abiotic variables (see Methods and S4 Table) according to Dutilleul’s t-tests (r≤-0.30, P≥0.039), SARs (F≤7.185, P≥0.009), or Mantel tests (r≤-0.12, P≥0.019) (S4 Table).

Finally, we analysed if there is a relationship between the genetic diversity for CMV defence traits and the overall genetic diversity of the A. thaliana wild genotypes estimated from neutral markers (250 SNPs). Mantel tests between pair-wise genetic distances estimated from neutral markers and pair-wise differences for virus RNA accumulation or SWi/SWm did not detect any significant correlation in relation to Cdc-CMV or Lro-CMV (r≤-0.04, P≥0.403).

Genetic diversity and geographic or climatic patterns of life history traits in A. thaliana

Tolerance to CMV in A. thaliana is related to the host life history traits, as tolerance is due in part to a reallocation of resources from growth to reproduction, which depends on the allometry of the vegetative to reproduction organs, (SW+IW)/RW [48]. In the 76 wild genotypes, resource reallocation upon infection by both CMV isolates also depended on plant allometry, being more efficient in genotypes with lower (SW+IW)/RW (F1,76>14.58, P<10−4), and the effect of infection by both CMV isolates on RW correlated positively with LP and RW of mock-inoculated plants (r≤0.29, P≥0.022). Neither resistance nor tolerance correlated with viable seed production of mock-inoculated controls (r≤0.02, P≥0.540).

We then analysed several life history traits related with growth and phenology in eight-week vernalised mock-inoculated plants of the 76 wild genotypes. Rosette weight (RW), inflorescence without seeds weight (IW) and seed weight (SW), growth period (GP) and life-span (LP) significantly differed between genotypes (F75,399≥9.23, P<10−4) (S2 and S3 Tables). Heritability of these traits was high, between 0.61 and 0.96 (Table 1). Overall, A. thaliana populations display a large genetic variation for the analysed life history traits, as in previous works [33,49,50].

Rosette weight (RW), inflorescence without seeds weight (IW) and seed weight (SW) significantly differed between genotypes (F75,399≥9.23, P<10−4) (S2 Table). To determine if life history traits related to tolerance showed similar or different geographic patterns than CMV defences, we analysed their autocorrelation. All growth and phenological traits showed significant spatial autocorrelation up to 153 km (Table 1). Furthermore, we analysed the relationship between these life history traits and climate using Dutilleul’s t-test, univariate SAR models and Mantel tests (S4 Table). Overall, RW, IW, SW and GP, but not LP, were positively correlated with altitude and negatively correlated with most climatic variables, including annual mean, minimal and maximal temperature, and precipitation seasonality (Fig 3 and S4 Table). These analyses showed A. thaliana genotypes from higher altitude, lower temperatures and higher precipitation seasonality flowered later, developed larger rosettes and inflorescences and produced more seeds. Moreover, Mantel tests showed that genetic distance was positively correlated with IW, SW, GP (r>0.15, P<0.008) and marginally with RW (r = 0.08, P = 0.078) but not with LP (r<-0.02, P>0.700).

Fig 3. Relationships between life-history traits and geographic or climatic factors.

Correlations are shown (A, B, C) for rosette weight (RW), (D, E, F) inflorescence weight (IW), (G, H, I) seed weight (SW) and (J, K, L) growth period (GP) with altitude, annual mean temperature and precipitation seasonality. Values are means of at least five replicates per plant genotype.

In contrast to CMV defence traits, life history traits related with resource allocation vary according to the climatic environment where populations evolved. Therefore, resistance and tolerance to CMV show different evolutionary histories than life history traits, likely reflecting distinct abiotic and biotic environmental selective forces acting on each group of traits.

A. thaliana population differentiation for resistance and tolerance to CMV

To quantify the distribution of genetic diversity for CMV defence traits within and among A. thaliana populations we analysed ten randomly sampled individual plants (henceforth named as “individuals”) from 10 or 12 Iberian populations, which were tested for their resistance and tolerance to two CMV isolates. One isolate from an Iberian population of A. thaliana (Cdc-CMV) and a reference isolate (Fny-CMV) were chosen because they had been used in previous work [43,47,48]. Since CMV resistance and tolerance depend on the environment [14,48,53], two experiments were performed with different vernalisation period lengths, as vernalisation affects life history traits relevant to tolerance such as rosette size, rosette leaf number, flowering time [33,49,54], and seed germination [55]. An eight-week vernalisation treatment simulated a cold winter, whereas a four-week vernalisation simulated a mild winter, as often occur across years in the original population locations. In both experiments, all individuals were systemically infected by both CMV isolates. The two experiments yielded similar results. For clarity only results of the long vernalisation treatment experiment are presented in the text, but results of the short-vernalisation are shown in S5 Table.

Virus accumulation varied considerably among individuals within populations (Fig 4, Table 2 and S6 Table). The average virus accumulation in each population ranged from 2.62 to 32.02 μg virus RNA g fwt-1. The heritability of virus accumulation varied between 0.23 and 0.90 depending on CMV isolate and host population (Table 2). Virus accumulation significantly depended on the virus isolate (F1,884 = 67.88, P<104), the A. thaliana population (F9,884 = 2.40, P = 0.059) (Fig 4 and Table 2), the A. thaliana individual nested to population (F88,884 = 2.54, P<10-4), and on the interactions CMV isolate x A. thaliana population (F9,884 = 4.11, P<10-4) and CMV isolate x A. thaliana individual nested to population (F88,884 = 7.08, P<10-4). CMV isolate, A. thaliana individual nested to population and their interaction explained 49.7, 3.3 and 14.6% of the variance, respectively, while A. thaliana population and the interaction CMV isolate x population explained 7.0 and 5.5%. The average accumulation was higher for Cdc-CMV than for Fny-CMV (18.61±2.76 and 7.37±1.74 μg virus RNA g fwt-1, respectively). Besides, average values of Cdc-CMV accumulation over individuals and populations correlated significantly, or marginally, with the corresponding values of Fny-CMV accumulation (rs = 0.32, P = 0.001; rs = 0.60, P = 0.067, respectively), indicating that, in general, individuals and populations that were more resistant to Cdc-CMV were also more resistant to Fny-CMV.

Fig 4. Variation for resistance and tolerance to CMV within and among wild A. thaliana populations.

Frequency distributions of resistance (virus accumulation, μg RNA g fresh leaf weight-1) and tolerance (SWi/SWm) to Cdc-CMV (blue) and Fny-CMV (red). Three-letter code for each population is as in Fig 1.

Table 2. Values (mean and range of variation) and heritability of defence traits to CMV in Iberian populations of A. thalianaa.

CMV tolerance also showed substantial variation among individuals within populations, SWi/SWm values ranging between 0.02 and 0.93 for Cdc-CMV, and between 0.03 and 0.81, for Fny-CMV-infected plants (Fig 4, Table 2 and S6 Table). Average SWi/SWm values in each population varied from 0.16 to 0.47 (Fig 4 and Table 2), indicating a lower range of variation among than within populations. Heritability of tolerance varied between 0.10 and 0.80 depending on CMV isolate and host population (Table 2). SWi/SWm varied significantly depending on virus isolate (F1,884 = 5.30, P = 0.046), A. thaliana population (F9,884 = 2.40, P = 0.059) (Fig 4 and Table 2), A. thaliana individual nested within population (F88,884 = 2.445, P = 0.023) and the interaction CMV isolate x A. thaliana individual (F88,884 = 2.65, P<10−4). A. thaliana individual and the interaction CMV isolate x A. thaliana individual explained a larger proportion of SWi/SWm variance (32.2%, 16.1%, respectively) than CMV isolate or A. thaliana population (1.6%, 6.4%, respectively). When averaged over all individuals, tolerance to Cdc-CMV (0.36±0.06) and Fny-CMV (0.32±0.06) were similar. Average values of SWi/SWm in Cdc-CMV-infected plants correlated across individuals and populations with those of Fny-CMV-infected plants (rs = 0.73, P = 0.016; rs = 0.60, P<10−4, respectively). Thus, as for resistance, the individuals and populations more tolerant to Cdc-CMV were also, in general, more tolerant to Fny-CMV. However, values of virus accumulation and SWi/SWm did not correlate over individuals for any CMV isolate (rs≤0.09, P≥0.129). Together, these results indicate that A. thaliana defences against CMV infection depend on the host genetic variation determining the specific defences.

Comparison of genetic differentiation among A. thaliana populations between quantitative traits and neutrals markers

To find out if CMV defence traits of A. thaliana might be under selection, we compared the genetic differentiation among populations for CMV resistance and tolerance, with that for neutral genetic variation (Fig 5). Two-hundred and fifty genome-wide SNPs, distinguished 74 different genotypes among the 120 individuals analysed (S7 Table) and were used to calculate FST values. As in the previous section, analyses are based on the results of the long vernalisation treatment experiment, analyses based on the short-vernalisation treatments gave similar results and are shown in S5 Table.

Fig 5. Genetic differentiation among wild A. thaliana populations for CMV resistance and tolerance traits, or for neutral markers.

Fig shows values of quantitative genetic differentiation (QST) for resistance (squares) and tolerance (triangles), to Cdc-CMV (blue) or Fny-CMV (red), and for neutral genetic differentiation (FST) (black circle), among ten A. thaliana populations. 95%CI are indicated.

The estimated average genetic differentiation among populations for neutral markers was 0.58 (95%CI = 0.54–0.62), which is presumed to reflect the demographic history of the populations. Genetic differentiation among populations for quantitative traits was estimated by QST values, leading to a similar average value of 0.31 (95%CI = 0.22–0.52) for accumulation of both CMV isolates. The average QST estimated for SWi/SWm was 0.18 (95%CI = 0.12–0.35) for Cdc-CMV and 0.10 (95%CI = 0.06–0.21) for Fny-CMV-infected plants. Therefore, the genetic variation of A. thaliana for both defence traits is distributed mostly within populations. QST values for resistance and tolerance to Cdc-CMV and Fny-CMV were significantly smaller than FST values (Fig 5), thus indicating that A. thaliana populations are genetically less differentiated for resistance and tolerance to two CMV isolates than for neutral markers.

Furthermore, we analysed if FST and QST values followed a pattern of isolation-by-distance. Mantel tests detected a significant correlation between FST values and geographic distance between pairs of populations (r = 0.50, P<10−4), a likely result of their demographic history. By contrast, no significant correlation was found between the pair-wise QST values for virus accumulation or SWi/SWm, for any CMV isolate, and their geographic distances (r≤0.24, P≥0.258). Therefore, factors other than demography contribute to the population differentiation patterns observed for resistance and tolerance to CMV.


Host-pathogen coevolution determines the dynamics and genetics of infectious disease, and may shape the genetic structure of host and pathogen populations [1]. Understanding this topic, central in pathology and evolutionary biology, requires knowledge on the genetics of defence and pathogenicity, and the dynamics of their change in populations [1]. The abundance of theoretical analyses of host-pathogen coevolution (e.g., [1,56,57]) is not matched by a similar amount of empirical and experimental studies. While there is abundant information compatible with host-pathogen coevolution in plant systems (e.g. [58]), it mostly derives from crops, in which the genetic composition of the host plant is manipulated by humans. Studies from wild systems, in which the genetic composition of host and pathogen populations may be determined by reciprocal selection, are much scarcer, particularly for plant-virus interactions [5,59]. To address this question we challenged A. thaliana individuals collected from a high number of local populations with different CMV isolates. We chose to inoculate plants mechanically rather than by aphid transmission, which is the natural means of horizontal transmission [39]. Mechanical inoculation ensures a high rate of infection and minimises inoculum dose effects on virus accumulation, as opposed to aphid transmission, which is highly inefficient for CMV [39, 40]. Moreover, there is no information on the aphid species that transmit CMV in A. thaliana populations in central Spain. Also, A. thaliana genotypes were assayed under common controlled conditions, which are not necessarily the same as in the field.

The challenge with different CMV strains of 185 individuals collected in 76 A. thaliana local populations from the Iberian Peninsula, showed large differences in quantitative resistance, as estimated from the level of virus multiplication in infected plants. Similarly, large differences were found for tolerance to CMV, estimated as the effect of infection on host progeny production. A large part of the variation of resistance and tolerance was explained by the virus isolate, in agreement with previous results showing that CMV isolates vary largely in multiplication rate and virulence in A. thaliana [41,47,48,53]. Variation for resistance and tolerance occurred at all analysed spatial scales: among individuals from a local population, among local populations, and within the whole Iberian Peninsula region. These conclusions held for assays conducted in different environments, a result to be underscored, as resistance and tolerance of A. thaliana to CMV can be modulated by the abiotic environment [14]. The observed variation in resistance and tolerance had a significant genetic component, as they showed medium to high heritability values (0.23–0.81 for resistance, and 0.10–0.70 for tolerance) depending on the spatial scale of the analysis and the isolate of CMV. Thus, our results show genetic variation for presumably defence traits in the host population, a condition for host-pathogen coevolution. Regardless of spatial scales, our data also show that resistance and tolerance significantly depend on the interactions between virus genotype and host genotype, another condition for host-pathogen coevolution.

It has been reported that A. thaliana genotypes showing high tolerance to CMV have a long life span, and that tolerance is, at least in part, the result of resource re-allocation from vegetative growth to reproduction, which is more efficient in long-lived genotypes [14,47,48,52]. It also has been shown that these life history and phenological traits have evolved as (direct or indirect) responses to climatic conditions [33,49,50,60]. Accordingly, the analysis of 76 A. thaliana genotypes from different Iberian populations showed that resource reallocation upon infection depended on the allometric ratio (SW+IW)/RW, and the effect of CMV infection on vegetative growth correlated positively with LP and RW of mock-inoculated plants. Life span, vegetative growth and seed production in non-infected plants were correlated with climatic variables. In contrast, the genetic variation for CMV multiplication in the infected host, and for the effect of CMV infection on seed production, was unrelated to those climatic factors. Therefore, the CMV defence traits have evolved, at least partly, independently from those other adaptive traits which have evolved in response to climate. Accordingly, the evolution of defence traits is not the result of A. thaliana responses to climate. These differential evolutionary histories strongly suggest that these traits are true resistance and tolerance defence responses that may have evolved in response to CMV infection (see below). This conclusion also agrees with the fact that resistance and tolerance are virus-specific traits of A. thaliana, and not unspecific responses to the stress of virus infection [52]. Thus, our study shows substantial genetic variation for resistance and tolerance to CMV within and between populations of A. thaliana. Genetic variation within or/and between populations for resistance to a variety of pathogens has been reported for a limited number of wild plants, including Amphicarpaea bracteata, Eucalyptus globulus, Podophyllum peltatum, Linum marginale, Silene latifolia, Phaseolus vulgaris, Plantago laceolata or A. thaliana [6178]. Analyses of the variation for plant tolerance to pathogens are much rarer [47,75,79], and none of them has analysed within population variation. All these studies refer to resistance or tolerance to fungi, oomycetes or bacteria, and the only report we are aware of on variation for resistance to a virus, is our previous analysis in the A. thaliana-CMV system [43], which involved a much more limited sample of host populations.

The analyses in this study also showed that resistance and tolerance display different evolutionary histories than neutral genetic variation, since QST values for resistance and tolerance are lower than FST values. Accordingly, resistance and tolerance are traits likely under uniform selection, i.e., a selection which favours a higher diversity of traits within than among populations. This conclusion is supported by results for two CMV isolates and in different environmental conditions. Also, neutral genetic differentiation follows a pattern of isolation by distance, which is not detected for the genetic differentiation for resistance or tolerance. By contrast, similar analyses have previously suggested diversifying selection on the genetic variation for quantitative traits such as flowering time, leaf number, specific leaf area or leaf succulence in A. thaliana [32,80,81]. The most parsimonious explanation for uniform selection on CMV defence traits is that CMV infection plays a role as selective factor. Although we cannot discard that the analysed defence traits may have evolved in response to selection from other pathogens or pests, several arguments make a strong case for selection pressure due to CMV infection: i) the high prevalence of CMV (up to 80%) in wild A. thaliana populations in the Iberian peninsula [43,44]; ii) the fact that the analysed defence traits are virus-specific in A. thaliana [52]; and iii) the lack of correlation between CMV multiplication and virulence [47], virus multiplication being highly dependent on virus genotype and environment [14,47]. Selection for resistance to pathogens has been best documented in populations of Linum marginale in response to the fungus Melampsora lini, of Plantago lanceolata in response to the fungus Podosphaera plantaginis, or of A. thaliana in response to the oomycete Hyaloperonospora arabidopsidis or the bacterium Pseudomonas syringae [37,68,69,7274,78,82,83]. Our results extend these observations to plant-virus interactions. If defence in A. thaliana against CMV is under selection, a corollary is that CMV is a virulent pathogen of this plant under natural conditions, a relevant conclusion that contributes significantly to understanding plant-virus interactions.

The observed pattern of genetic diversity for resistance and tolerance to CMV, higher within than between populations, can be explained by the features of the pathosystem. CMV is ubiquitous in Iberia [45,84] and has been found in all monitored wild populations of A. thaliana, prevalence differing among populations and years [43]. Also, our present and past results [47,48] show that CMV isolates differ in virulence to A. thaliana, and that virulence is modulated by environmental factors as diverse as temperature, light intensity or host plant density [14,53].Variation in the genetic composition of CMV populations would also result in variation for CMV infection-associated selection, as the outcome of the interaction depends on the A. thaliana and CMV genotypes involved. These factors would explain the maintenance of genetic variation in defence traits within host populations and the limited differentiation among populations. Furthermore, this explanation suggests that resistance and tolerance to CMV involve fitness penalties for A. thaliana, which would hinder fixation of resistance/tolerance alleles and would contribute to the maintenance of defence polymorphisms within populations [85]. We have not found evidence for such costs under the assayed conditions, as neither resistance nor tolerance were negatively correlated with viable seed production of mock-inoculated controls. However, fitness costs might not be detectable under our experimental conditions, and/or might be unveiled under the less favourable environment of the field.

Another interesting result is that resistance and tolerance to CMV co-occur in wild A. thaliana populations. Theory predicts that resources being limited, hosts would not invest in both resistance and tolerance, which would be mutually exclusive defences. The conditions that should favour the evolution of resistance or tolerance have been much modelled, and a negative correlation between both traits across host genotypes is expected [4,86,87]. We found no correlation between resistance and tolerance in any experiment, indicating that they evolve as independent traits. It has been proposed that resistance and tolerance could coexist if costs and benefits of each defence were different and non-additive [8891]. Models also propose that tolerance alleles should become fixed in host populations, which would not be polymorphic for this trait under most assumptions [9193]. A report on the tolerance to CMV in Mimulus guttatus conforms to these predictions, as tolerance had no costs, but showed little genetic variation [94,95]. On the contrary, our results do not agree with model predictions, as we found large genetic variation for tolerance and evidence for coexistence of resistance and tolerance.

In conclusion, this work shows that in A. thaliana there is genetic variation, within and among populations, for defences to CMV that result in lower virus multiplication or in lower impact of infection on the plant fitness. Genetic variation for defence to CMV is not associated with variation for climatic factors, in contrast to variation for other adaptive life-history traits of A. thaliana. In addition, we found evidence that these two defence traits are under uniform selection. The results of this study are compatible with CMV infection acting as a selection pressure for defence on populations of A. thaliana and, hence, we propose that CMV infection likely reduces the host fitness under the field conditions of the analysed wild A. thaliana populations, although field experiments would be required to prove this fact. The results presented here also show that some of the conditions for coevolution are met in the system A. thaliana-CMV, but more work on the virus side is necessary to prove if coevolution occurs. These results raise two challenging questions: what are the mechanisms that maintain polymorphisms for resistance and tolerance within A. thaliana populations, and what is the negative impact of CMV infection on the host in nature and how does such an impact vary according to field conditions.

Materials and methods

Plant material and environmental data

Two different sets of A. thaliana samples from the Iberian Peninsula were analysed. First, 76 accessions or wild genotypes collected from different populations were selected to cover the genetic and environmental diversity of the species in that region [30,49] (Fig 1 and S1 Table). This collection spanned 800 x 700 km, populations being spaced in the average 384.9±3.7 (20.2–1,038.1 km). Altitudes ranged from 123 to 1,670 m above sea level. Each sample was genetically different based on previous SNP genotyping and genome sequences [51,96]. Second, ten individuals plants (individuals) randomly sampled from 12 of these populations were selected for intra and inter-population analyses (Fig 1 and S1 Table). Samples from eight of these populations have been previously genotyped for 250 genome-wide SNPs that were segregating in these populations [30,32]. For the remaining four populations (Bis, Mer, Moc, Pob) 10 individuals/population were genotyped in this study for the same set of SNPs.

The climatic information from the locations of A. thaliana populations was obtained from the digital climatic atlas of the Iberian Peninsula at 1 km2 resolution [50,97]. Thirty-three variables were used, related to temperature, precipitation and solar radiation (S3 Table). In addition, 19 bioclimatic variables derived by combination of annual trends, seasonality and extreme conditions were also included ( Altitude was also analysed as a proxy for climate. Annual mean temperature of the populations ranged 6.1–17.4°C (12.5±0.3) and annual precipitation ranged 405.7–1695.8 mm (753.8±33.9) (S8 Table).

All accessions or individuals used in this study were propagated by selfing during two generations by the single seed descent procedure, in a glasshouse supplemented with lamps to provide a long-day photoperiod. This allowed reducing residual heterozygosity that might contain some wild individuals but also removing any potential maternal and grand-mother effects. Seeds were stratified (darkness, 4ºC) for 7 days before germination at 25/20ºC day/night, 16 h light. Ten day-old seedlings were transferred to 4ºC, 8 h light, for vernalisation during 4 or 8 weeks, depending on the experiment. After vernalisation, plants were transplanted to 0.43 L pots and returned to the greenhouse, where they were kept (25/20ºC day/night, 16 h light) until the end of the experiment.

Virus isolates and inoculations

Three subgroup IA CMV isolates were used, Fny-CMV, Cdc-CMV and Lro-CMV, which differ in the sequence of their genomic RNAs in about 1% of positions. Fny-CMV is a well-characterized reference isolate [98]. Cdc-CMV and Lro-CMV were isolated from field-infected A. thaliana plants of the Cdc and Cho populations, respectively, in 2008 and 2011, Cdc-CMV was named At-CMV in [43]. Isolates were multiplied in Nicotiana clevelandii, Fny-CMV from transcripts of cDNA clones and Cdc-CMV and Lro-CMV from biological clones derived from local lesions in Chenopodium quinoa. Virions were purified as in [99]. A. thaliana plants were mechanically inoculated at the five-leaf stage (stage 1.05, [100]) with 15 μl of sap from infected N. clevelandii leaves in 0.01 M phosphate buffer pH 7.0, 0.2% sodium diethyldithiocarbamate. Fifteen μl of buffer were applied to mock-inoculated controls. The (unkown) virus concentration in leaf sap ensured infection of 100% of inoculated plants. Each treatment (virus-inoculated or buffer mock-inoculated) involved at least five replicated plants from each original sample, that is at least five plants derived from the same genotype or individual. All plants in each experiment were grown in a completely randomized design.

Quantification of CMV multiplication

Virus multiplication in plants was estimated from virus RNA accumulation as described in Pagán et al., (2014) [41]. Briefly, at fifteen days post-inoculation 0.01 g fresh weight (fwt) of leaf tissue was harvested from four different systemically infected leaves. Nucleic acids were extracted from the pooled leaf tissue using TRI-reagent (Sigma-Aldrich, St Louis, MO, USA). Virus RNA was then quantified by dot-blot hybridization with 32P-labelled RNA probes complementary to nucleotides 1933–2215 of Fny-CMV RNA3 (GeneBank Acc. No. D10538). In each blot, internal standards for Fny-CMV, Cdc-CMV or Lro-CMV RNA were included as a two-fold dilution series (1–0.001 μg) of purified virion RNA in nucleic acid extracts from non-inoculated plants. Mock-inoculated samples served as negative controls. Nucleic acid extracts were blotted at different dilutions to ensure that hybridization signal was on the linear portion of the RNA concentration-hybridization curve. As loading controls, parallel membranes were hybridized with a cDNA probe of β-tubulin chain 2 (TUB2) mRNA of A. thaliana (1086–1568 nt, GeneBank Acc. No. NM_125664.4).

Quantification of life-history traits and tolerance to CMV

Rosette weight was used to estimate vegetative growth effort, inflorescence plus seed weight to estimate total reproductive effort, and seed weight to estimate progeny production [101]. Previous work has shown that CMV infection does not affect seed viability, nor the weight of individual seeds, in a broad range of A. thaliana genotypes [14,47,52]. Plants were harvested at complete senescence and dry weight of rosettes (rosette weight, RW), inflorescence structures without seeds (inflorescence weight, IW) and seeds (seed weight, SW) were measured separately (g). Two phenological parameters of A. thaliana life cycle were quantified: Growth period (GP) and life-span (LP) were measured as the time (days) between planting seedlings in soil and opening of the first flower (GP), or complete senescence (LP). Tolerance was measured by the effect of virus infection on progeny production: SWi/SWm, where i and m denote infected and mock inoculated plants, respectively [48].

Genetic analyses

Broad sense heritability of each trait was estimated as h2b = VG/(VG+VE), where VG is the among-genotypes or among-populations variance component and VE is the residual variance. Variance components were determined using the REML method [102] of SPSS 20 package (SPSS Inc., Chicago, USA).

Genetic differentiation between populations for quantitative traits was measured by QST values [103], estimated as VB/(VB+VW) [104,105], where VB is the between-population variance and VW is the within-population variance. VB and VW were estimated by the REML method from a nested analysis of variance performed using population and individual or genotype (nested within populations) as random factors. The 95% confidence intervals (95%CI) for QST values were estimated as [106]. Genetic differentiation for neutral markers was estimated as FST [107] using the analysis of molecular variance (AMOVA) as implemented in ARLEQUIN v3.5.1.2 [108]. AMOVA were performed using multilocus genotypes for 250 segregating SNPs [30,32] and their significances were estimated from 1,000 permutations.

The relationships between Euclidean geographical distance and FST or QST values among population pairs were determined by Mantel correlation test using PASSaGE v.2 [109] with 1,000 permutations. Genetic distances between individuals were calculated as the proportion of allelic differences over the total number of alleles in the corresponding set of polymorphic loci, using GGT v. 2.0 [110].

Statistical analyses

Differences in RW, IW, SW, GP, LP, virus accumulation or tolerance to CMV, according to host individual/genotype and virus isolate, were analysed by general linear models (GLM) considering host individual/genotype as a random factor, and virus isolate as a fixed factor. Differences in RW, IW, SW, GP, LP, virus accumulation or tolerance to CMV according to population, host individual/genotype, and virus isolate, were analysed by GLM considering isolate as a fixed factor, and population and individual/genotype nested to population, as random factors. Relationships between values of different traits were tested using Spearman’s correlation test. GLMs and Spearman’s correlation tests were performed using SPSS 20 software package.

Spatial autocorrelation patterns of environmental variables, life-history traits, virus accumulation and tolerance to virus, were analysed using correlograms [111] generated with PASSaGE v.2. For each variable, Moran’s I autocorrelation coefficients [112] were calculated and their significance tested from 1,000 permutations. Correlation between pairs of environmental variables, between pairs of different traits and between environmental variable and different traits were tested with Dutilleul’s modified t-test using SAM v.4 [113,114]. Simultaneous autoregressive models (SAR) [115] were performed to test the relationship between environmental variables and different traits using SAM v.4. Bonferroni correction was applied for multiple comparisons.

The relationships between environmental variables and life-history or defence traits were analysed by partial Mantel tests controlling for the location of populations given by the geographic distance matrix using PASSaGE v.2. For that, matrices of euclidean distances were derived for each environmental variable and phenotypic trait and significance was evaluated from 1,000 permutations.

Supporting information

S1 Table. Geographic origin and location of Iberian A. thaliana populations.


S2 Table. GLM analysis of values of life history traits of mock-inoculated A. thaliana plants, of tolerance to CMV, and of virus accumulation, using "host genotype" as a random factor.

For each trait, heritability (h2b), mean, minimum and maximum values, are shown.


S3 Table. Values of climatic variables for 76 Iberian A. thaliana populations, and for the phenotypic traits of representative individuals.


S4 Table. Coefficients of the correlation/regression between values of mock-inoculated A. thaliana life-history or defence to CMV traits, and those of the environmental variables of the corresponding population obtained by SARs, Correlation with Dutilleul´s correction, and Mantel test.

A threshold significance value (α = 0.0014) was set after applying Bonferroni's correction (1– [1– a] 1/n) for multiple comparisons.


S5 Table. Mean, minimum and maximum values of resistance and tolerance to CMV of A. thaliana individuals inoculated after a four-week vernalisation period.

Mean, minimum and maximum values, heritability and QST of resistance and tolerance to CMV for twelve A. thaliana populations are also shown.


S6 Table. Mean values of resistance and tolerance to CMV for A. thaliana individuals inoculated after an eight-week vernalisation period.


S7 Table. Distribution of genotypes (H) and individuals (lower case three letter codes) in 12 analysed A. thaliana populations.


S8 Table. Values (mean and range of variation) of the environmental variables used in the study.

Values are the averages of 76 population.



Antolín López Quirós and Miguel Ángel Mora provided excellent technical support.


  1. 1. Woolhouse MEJ, Webster JP, Domingo E, Charlesworth B, Levin BR. Biological and biomedical implications of the co-evolution of pathogens and their hosts. Nat Genet. 2002; 32: 569–577. pmid:12457190
  2. 2. Agnew P, Koella JC, Michalakis Y. Host life history responses to parasitism. Microb Infect. 2000; 2: 891–896.
  3. 3. Råberg L. How to Live with the Enemy: Understanding Tolerance to Parasites. PLoS Biol. 2014; 12: e1001989. pmid:25369060
  4. 4. Pagán I, García-Arenal F. Tolerance to plant pathogens: theory and experimental evidences. Int J Mol Sci. 2018; 19: 810.
  5. 5. Fraile A, García-Arenal F. The coevolution of plants and viruses: Resistance and pathogenicity. Adv Virus Res. 2010; 76: 1–32. pmid:20965070
  6. 6. Brown JKM, Tellier A. Plant-Parasite Coevolution: Bridging the Gap between Genetics and Ecology. Annu Rev Phytopathol. 2011; 49: 345–367. pmid:21513455
  7. 7. Brown JKM. Durable resistance of crops to disease: a Drawinian perspective. Annu Rev Phytopathol. 2015; 53: 513–539. pmid:26077539
  8. 8. Salvaudon L, Giraud T, Shykoff JA. Genetic diversity in natural populations: a fundamental component of plant–microbe interactions. Curr Opin Plant Biol. 2008; 11: 135–143. pmid:18329329
  9. 9. Burdon JJ, Thrall PH. Coevolution of Plants and Their Pathogens in Natural Habitats. Science. 2009; 324: 755–756. pmid:19423818
  10. 10. Roossinck MJ. Symbiosis versus competition in plant virus evolution. Nat Rev Microbiol. 2005; 3: 917–924. pmid:16322740
  11. 11. Roossinck MJ, García-Arenal F. Ecosystem simplification, biodiversity loss and plant virus emergence. Curr Opin Virol. 2015; 10: 56–62. pmid:25638504
  12. 12. Fraile A, García-Arenal F. Environment and evolution modulate plant virus pathogenesis. Curr Opin Virol. 2016; 17: 50–56. pmid:26827164
  13. 13. Xu P, Chen F, Mannas JP, Feldman T, Sumner LW, Roossinck MJ. Virus infection improves drought tolerance. New Phytol. 2008; 180: 911–921. pmid:18823313
  14. 14. Hily JM, Poulicard N, Mora MÁ, Pagán I, García-Arenal F. Environment and host genotype determine the outcome of a plant-virus interaction: from antagonism to mutualism. New Phytol. 2016; 209: 812–822. pmid:26365599
  15. 15. Maskell LC, Raybould AF, Cooper JI, Edwards ML, Gray AJ. Effects of turnip mosaic virus and turnip yellow mosaic virus on the survival, growth and reproduction of wild cabbage (Brassica oleracea). Ann Appl Biol. 1999; 135: 401–407.
  16. 16. Funayama S, Terashima I, Yahara T. Effects of virus infection and light environment on population dynamics of Eupatorium makinoi (Asteraceae). Am J Bot. 2001; 88: 616–622. pmid:11302846
  17. 17. Power AG, Mitchell CE. Pathogen spillover in disease epidemics. Am Nat. 2004; 164: S79–S89. pmid:15540144
  18. 18. Malmstrom CM, Hughes CC, Newton LA, Stoner CJ. Virus infection in remnant native bunchgrasses from invaded California grasslands. New Phytol. 2005; 168: 217–230. pmid:16159335
  19. 19. Malmstrom CM, Stoner CJ, Brandenburg S, Newton LA. Virus infection and grazing exert counteracting influences on survivorship of native bunchgrass seedlings competing with invasive exotics. J Ecol. 2006; 94: 264–275. pmid:18784788
  20. 20. Prendeville HR, Tenhumberg B. Pilson D. Effects of virus on plant fecundity and population dynamics. New Phytol. 2014; 202: 1346–1356. pmid:24571200
  21. 21. Fraile A, McLeish M, Pagán I, González-Jara P, Piñero D, García-Arenal F. Environmental heterogeneity and the evolution of plant-virus interactions: Viruses in wild pepper populations. Virus Res. 2017; 15: 68–76.
  22. 22. Rodríguez-Nevado C, Montes N, Pagán I. Ecological Factors Affecting Infection Risk and Population Genetic Diversity of a Novel Potyvirus in Its Native Wild Ecosystem. Front Plant Sci. 2017; 8: 1958. pmid:29184567
  23. 23. Pallett DW, Thurston MI, Cortina-Borja M, Edwards ML, Alexander M, Mitchell E, et al. The incidence of viruses in wild Brassica rapa ssp. sylvestris in southern England. Ann Appl Biol. 2002; 141: 163–170.
  24. 24. Ausin I, Alonso-Blanco C, Martinez-Zapater JM. Environmental regulation of flowering. Int J Dev Biol. 2005; 49: 689–705. pmid:16096975
  25. 25. Hoffmann MH. Biogeography of Arabidopsis thaliana (L.) Heynh. (Brassicaceae) J Biogeogr. 2002; 29: 125–134.
  26. 26. Brennan AC, Méndez-Vigo B, Haddioui A, Martínez-Zapater JM, Picó FX, Alonso-Blanco C. The genetic structure of Arabidopsis thaliana in the south-western Mediterranean range reveals a shared history between North Africa and southern Europe. BMC Plant Biol. 2014; 14: 17. pmid:24411008
  27. 27. Durvasula A, Fulgione A, Gutaker RM, Alacakaptan SI, Flood PJ, Neto C, et al. African genomes illuminate the early history and transition to selfing in Arabidopsis thaliana. Proc Natl Acad Sci USA. 2017; 114: 5213–5218. pmid:28473417
  28. 28. The 1001 Genomes Consortium. 1,135 Genomes reveal the global pattern of polymorphism in Arabidopsis thaliana. Cell. 2016; 166: 481–491. pmid:27293186
  29. 29. Lee CR, Svardal H, Farlow A, Exposito-Alonso M, Ding W, Novikova P, et al. On the post-glacial spread of human commensal Arabidopsis thaliana. Nat Commun. 2017; 8: 14458. pmid:28181519
  30. 30. Picó FX, Mendez-Vigo B, Martinez-Zapater JM, Alonso-Blanco C. Natural genetic variation of Arabidopsis thaliana is geographically structured in the Iberian Peninsula. Genetics. 2008; 180: 1009–1021. pmid:18716334
  31. 31. Picó FX. Demographic fate of Arabidopsis thaliana cohorts of autumn‐and spring‐germinated plants along an altitudinal gradient. J Appl Ecol. 2012; 100: 1009–1018.
  32. 32. Méndez-Vigo B, Gomaa NH, Alonso-Blanco C, Picó FX. Among- and within-population variation in flowering time of Iberian Arabidopsis thaliana estimated in field and glasshouse conditions. New Phytol. 2013; 197: 1332–1343. pmid:23252608
  33. 33. Vidigal DS, Marques AC, Willems LA, Buijs G, Méndez-Vigo B, Hilhorst HW, et al. Altitudinal and climatic associations of seed dormancy and flowering traits evidence adaptation of annual life cycle timing in Arabidopsis thaliana. Plant Cell Environ. 2016; 39: 1737–1748. pmid:26991665
  34. 34. Mauricio R, Rausher MD. Experimental manipulation of putative selective agents provides evidence for the role of natural enemies in the evolution of plant defense. Evolution. 1997; 51: 1435–1444. pmid:28568624
  35. 35. Jorgensen TH, Emerson BC. Functional variation in a disease resistance gene in populations of Arabidopsis thaliana. Mol Ecol. 2008; 17: 4912–4923. pmid:19140981
  36. 36. Huard-Chauveau C, Perchepied L, Debieu M, Rivas S, Kroj T, Kars I, et al. An atypical kinase under balancing selection confers broad-spectrum disease resistance in Arabidopsis. PLoS Genet. 2013; 9: e1003766. pmid:24068949
  37. 37. Karasov TL, Kniskern JM, Gao L, DeYoung BJ, Ding J, Dubiella U, et al. The long-term maintenance of a resistance polymorphism through diffuse interactions. Nature. 2014; 512: 436–440. pmid:25043057
  38. 38. Brachi B, Meyer CG, Villoutreix R, Platt A, Morton TC, Roux F, et al. Coselected genes determine adaptive variation in herbivore resistance throughout the native range of Arabidopsis thaliana. Proc Natl Acad Sci USA. 2015; 112: 4032–4037. pmid:25775585
  39. 39. Jacquemond M. Cucumber mosaic virus. Adv Virus Res. 2012; 84: 439–504. pmid:22682176
  40. 40. Hily JM, García A, Moreno A, Plaza M, Wilkinson MD, Fereres A, Fraile A, García-Arenal F. The Relationship between Host Lifespan and Pathogen Reservoir Potential: An Analysis in the System Arabidopsis thaliana-Cucumber mosaic virus. PLoS Pathog. 2014; 10: e1004492. pmid:25375140
  41. 41. Pagán I, Montes N, Milgroom MG, García-Arenal F. Vertical Transmission Selects for Reduced Virulence in a Plant Virus and for Increased Resistance in the Host. PLoS Pathog. 2014; 10: e1004293. pmid:25077948
  42. 42. Palukaitis P, García-Arenal F. Cucumoviruses. Adv Virus Res. 2003; 62: 241–323. pmid:14719367
  43. 43. Pagán I, Fraile A, Fernández-Fueyo E, Montes N, Alonso-Blanco C, García-Arenal F. Arabidopsis thaliana as a model for the study of plant-virus co-evolution. Philos Trans R Soc Lond, B, Biol Sci. 2010; 365: 1983–1995. pmid:20478893
  44. 44. Montes N. Arabidopsis thaliana como modelo de estudio de la coevolución planta-virus Thesis, Universidad Politécnica de Madrid (UPM). 2016; Available from:
  45. 45. Fraile A, Alonso-Prados JL, Aranda MA, Bernal JJ, Malpica JM, García-Arenal F. Genetic exchange by recombination or reassortment is infrequent in natural populations of a tripartite RNA plant virus. J Virol. 1997; 71: 934–940. pmid:8995610
  46. 46. Bonnet J, Fraile A, Sacristán S, Malpica JM, García-Arenal F. Role of recombination in the evolution of natural populations of Cucumber mosaic virus, a tripartite RNA plant virus. Virology. 2005; 332: 359–368. pmid:15661167
  47. 47. Pagán I, Alonso-Blanco C, García-Arenal F. The Relationship of Within-Host Multiplication and Virulence in a Plant-Virus System. PLoS ONE. 2007; 2: e786. pmid:17726516
  48. 48. Pagán I, Alonso-Blanco C, García-Arenal F. Host Responses in Life-History Traits and Tolerance to Virus Infection in Arabidopsis thaliana. PLoS Pathog. 2008; 4: e1000124. pmid:18704166
  49. 49. Méndez-Vigo B, Picó FX, Ramiro M, Martinez-Zapater JM, Alonso-Blanco C. Altitudinal and Climatic Adaptation Is Mediated by Flowering Traits and FRI, FLC, and PHYC Genes in Arabidopsis. Plant Physiol. 2011; 157: 1942–1955. pmid:21988878
  50. 50. Manzano-Piedras E, Marcer A, Alonso-Blanco C, Picó FX. Deciphering the Adjustment between Environment and Life History in Annuals: Lessons from a Geographically-Explicit Approach in Arabidopsis thaliana. PLoS ONE. 2014; 9: e87836. pmid:24498381
  51. 51. Tabas-Madrid D, Méndez-Vigo B, Arteaga N, Marcer A, Pascual-Montano A, Weigel D, et al. Genome-wide signatures of flowering adaptation to climate temperature: regional analyses in a highly diverse native range of Arabidopsis thaliana. Plant Cell Environ. 2018; 41: 1806–1820. pmid:29520809
  52. 52. Shukla A, Pagán I, García-Arenal F. Effective tolerance based on resource reallocation is a virus-specific defence in Arabidopsis thaliana. Mol Plant Pathol. 2018; 19: 1454–1465. pmid:29027740
  53. 53. Pagán I, Alonso-Blanco C, García-Arenal F. Differential tolerance to direct and indirect density-dependent costs of viral infection in Arabidopsis thaliana. PLoS Pathog. 2009; 5: e1000531. pmid:19649316
  54. 54. Björn HK, Gavin RS, Nienhuis J. Effects of photoperiod and vernalization on the number of leaves at flowering in 32 Arabidopsis thaliana (Brassicaceae) ecotypes. Am J Bot. 1993; 80: 646–648.
  55. 55. Auge GA, Leverett D, Edwards BR, Donohue K. Adjusting phenotypes via within- and across-generational plasticity. New Phytol. 2017; 216: 343–349. pmid:28262950
  56. 56. Agrawal A, Lively CM. Infection genetics: gene-for-gene versus matching-alleles models and all points in between. Evol Ecol Res. 2002; 4: 79–90.
  57. 57. Agrawal A, Lively CM. Modelling infection as a two-step process combining gene-for-gene and matching-allele genetics. Proc R Sci. 2003; 270: 323–34.
  58. 58. Sacristán S, García-Arenal F. The evolution of virulence and pathogenicity in plant pathogen populations. Mol Plant Pathol. 2008; 9: 369–384. pmid:18705877
  59. 59. Alexander H, Mauck K, Whitfield A, Garrett K, Malmstrom C. Plant-virus interactions and the agro-ecological interface. Eur J Plant Pathol. 2014; 138: 529–547.
  60. 60. Bergelson J, Roux F. Towards identifying genes underlying ecologically relevant traits in Arabidopsis thaliana. Nat Rev Genet. 2010; 11: 867–879. pmid:21085205
  61. 61. Parker MA. Disease impact and local genetic diversity in the clonal plant Podophyllum peltatum. Evolution. 1989; 43: 540–547. pmid:28568395
  62. 62. Jarosz AM, Burdon JJ. Predominance of a single major gene for resistance to Phakopsora pachyrhizi and Glycine argyrea: Phenotypic and genetic patterns. Heredity. 1990; 64: 347–353.
  63. 63. Jarosz AM, Burdon JJ. Host-pathogen interactions in natural populations of Linum marginale and Melampsora lini. II. Local and regional variation in patterns of resistance and racial structure. Evolution. 1991; 45: 1618–27. pmid:28564135
  64. 64. Antonovics J, Thrall PH, Jarosz A, Stratton D. Ecological genetics of metapopulations: the Silene–Ustilago plant–pathogen system. In: Real L, ed. Ecological genetics. Princeton, US, Princeton University Press; 1994. pp 146–170.
  65. 65. Burdon JJ, Jarosz AM. Host–pathogen interactions in natural populations of Linum marginale and Melampsora lini. I. Patterns of resistance and racial variation in a large host population. Evolution. 1991; 45: 205–217. pmid:28564067
  66. 66. Hamilton MG, Williams DR, Tilyard PA, Pinkard EA, Wardlaw TJ, Glen M, et al. A latitudinal cline in disease resistance of a host tree. Heredity. 2013; 110: 372–379. pmid:23211794
  67. 67. Stahl EA, Dwyer G, Mauricio R, Kreitman M, Bergelson J. Dynamics of disease resistance polymorphism at the Rpm1 locus of Arabidopsis. Nature. 1999; 400: 667–671. pmid:10458161
  68. 68. Thrall PH, Burdon JJ, Young A. Variation in resistance and virulence among demes of a plant host-pathogen metapopulation. J. Ecol. 2001; 89: 736–748.
  69. 69. Thrall PH, Burdon JJ. Evolution of gene-for-gene systems in metapopulations: the effect of spatial scale of host and pathogen dispersal. Plant Pathol. 2002; 51: 169–184.
  70. 70. Kaltz O, Shykoff JA. Local adaptation in host–parasite systems. Heredity. 1998; 81: 361–370.
  71. 71. de Meaux J, Mitchell-Olds T. Evolution of Plant Resistance at the Molecular Level: Ecological Context of Species Interactions. Heredity. 2003; 91: 345–52. pmid:14512948
  72. 72. Laine AL. Resistance variation within and among host populations in a plant-pathogen metapopulation-implications for regional pathogen dynamics. J. Ecol. 2004; 92: 990–1000.
  73. 73. Laine AL. Spatial scale of local adaptation in a plant-pathogen. J Evol Biol. 2005; 18: 930–938. pmid:16033565
  74. 74. Laine AL. Evolutionary of host resistance: looking for coevolutionry hotspot at small spatial scales. Philos Trans R Soc Lond B Biol Sci. 2006; 273: 267–273.
  75. 75. Salvaudon L, Héraudet V, Shykoff JA. Parasite-host fitness trade-offs change with parasite identity: genotype-specific interactions in a plant-pathogen system. Evolution. 2005; 59: 2518–2524. pmid:16526500
  76. 76. Salvaudon L, Heraudet V, Shykoff JA. Arabidopsis thaliana and the Robin Hood parasite: a chivalrous oomycete that steals fitness from fecund hosts and benefits the poorest one?. Biol. Lett. 2008; 4: 526–529. pmid:18682359
  77. 77. Kniskern JM, Barrett LG, Bergelson J. Maladaptation in wild populations of the generalist plant pathogen Pseudomonas syringae. Evolution. 2011; 65: 818–830. pmid:21044058
  78. 78. Salvaudon L, Shykoff JA. Variation in Arabidopsis developmental responses to oomycete infection: resilience vs changes in life history traits. New Phytol. 2013; 197: 919–926. pmid:23231447
  79. 79. Kover PX, Schaal BA. Genetic variation for disease resistance and tolerance among Arabidopsis thaliana accessions. Proc Natl Acad Sci USA. 202; 99: 11270–11274.
  80. 80. Mitchell-Olds T. Arabidopsis thaliana and its wild relatives: a model system for ecology and evolution. Trends Ecol Evol. 2011; 16: 693–700.
  81. 81. Luo Y, Widmer A, Karrenberg. The roles of genetic drift and natural selection in quantitative trait divergence along an altitudinal gradient in Arabidopsis thaliana. Heredity. 2015; 114: 220–228. pmid:25293874
  82. 82. Jakob K, Goss EM, Araki H, Van T, Kreitman M, Bergelson J. Pseudomonas viridiflava and P. syringae—Natural Pathogens of Arabidopsis thaliana. Mol Plant Microbe Interact. 2002; 15: 1195–1203. pmid:12481991
  83. 83. Goss EM, Bergelson J. Variation in resistance and virulence in the interaction between Arabidopsis thaliana and a bacterial pathogen. Evolution. 2006; 60: 1562–1573. pmid:17017057
  84. 84. Alonso-Prados JL, Aranda MA, Malpica JM, García-Arenal F, Fraile A. Satellite RNA of Cucumber mosaic virus spreads epidemically in natural populations of its helper virus. Phytopathology. 1998; 88: 520–524. pmid:18944903
  85. 85. Antonovics J, Thrall P. The cost of resistance and the maintenance of genetic polymorphism in host-parasite systems. Philos Trans R Soc Lond B Biol Sci. 1994; 257: 105–110.
  86. 86. Herms DA, Mattson WJ. The dilemma of plants: To grow or defend. Q Rev Biol. 1992; 67: 283–335.
  87. 87. Manzaneda AJ, Prasad KVSK, Mitchell-Olds T. Variation and fitness costs for tolerance to different types of herbivore damage in Boechera stricta genotypes with contrasting glucosinolate structures. New Phytol. 2010; 188: 464–477. pmid:20663059
  88. 88. Mauricio R, Rausher MD, Burdick DS. Variation in the defense strategies of plants: are resistance and tolerance mutually exclusive?. Ecology. 1997; 78: 1301–1311.
  89. 89. Boots M, Bowers RG. Three mechanisms of host resistance to microparasites–avoidance, recovery and tolerance–show different evolutionary dynamics. J Theor Biol. 1999; 201: 13–23. pmid:10534432
  90. 90. Restif O, Koella JC. Shared control of epidemiological traits in a coevolutionary model of host-parasite interactions. Am Nat. 2003; 161: 827–836. pmid:12858269
  91. 91. Restif O, Koella JC. Concurrent Evolution of Resistance and Tolerance to Pathogens. Am Nat. 2004; 164: E90–E102. pmid:15459887
  92. 92. Roy BA, Kirchner JW. Evolutionary dynamics of pathogen resistance and tolerance. Evolution. 2000; 54: 51–63. pmid:10937183
  93. 93. Miller MR, White A, Boots M. The evolution of host resistance: tolerance and control as distinct strategies. J Theor Biol. 2005; 236: 198–207. pmid:16005309
  94. 94. Carr DE, Murphy JF, Eubanks MD. The susceptibility and response of inbred and outbred Mimulus guttatus to infection by Cucumber mosaic virus. Evol Ecol. 2003; 17: 85–103.
  95. 95. Carr DE, Murphy JF, Eubanks MD. Genetic variation and covariation for resistance and tolerance to Cucumber mosaic virus in Mimulus guttatus (Phrymaceae): a test for costs and constraints. Heredity. 2006; 96: 29–38. pmid:16189544
  96. 96. Gomaa NH, Montesinos-Navarro A, Alonso-Blanco C, Picó FX. Temporal variation in genetic diversity and effective population size of Mediterranean and subalpine Arabidopsis thaliana populations. Mol Ecol. 2001; 20: 3540–3554.
  97. 97. Ninyerola M, Pons X, Roure JM. Atlas Climático Digital de la Península Ibérica. Metodología y aplicaciones en bioclimatología y geobotánica. Universidad Autónoma de Barcelona: Bellaterra; 2005.
  98. 98. Palukaitis P, Roossinck MJ, Dietzgen RG, Francki RIB. Cucumber Mosaic Virus. Adv Virus Res. 1992; 41: 281–348. pmid:1575085
  99. 99. Lot H, Marrou J, Quiot JB, Esvan C. Contribution à l’étude du virus de la mosaïque du concombre (CMV). Méthode de purification rapide du virus. Ann Phytopathol. 1975, 4: 25–38.
  100. 100. Boyes DC, Zayed AM, Ascenzi R, McCaskill AJ, Hoffman NE, Davis KR, Görlach J. Growth stage-based phenotypic analysis of Arabidopsis: A model for high throughput functional genomics in plants. Plant Cell. 2001; 13: 1499–1510. pmid:11449047
  101. 101. Thompson K, Stewart AJA. The Measurement and Meaning of Reproductive Effort in Plants. Am Nat. 1981; 117: 205–211.
  102. 102. Lynch M, Walsh B. Genetics and analysis of quantitative traits. Sunderland, MA, USA: Sinauer Associates Inc; 1998.
  103. 103. Spitze K. Population structure in Daphnia obtusa: quantitative genetic and allozymic variation. Genetics. 1993; 135: 367–374. pmid:8244001
  104. 104. Bonnin I, Prosperi JM, Olivieri I. Genetic markers and quantitative genetic variation in Medicago truncatula (Leguminosae): A comparative analysis of population structure. Genetics. 1996; 143: 1795–1805. pmid:8844165
  105. 105. Le Corre V. Variation at two flowering time genes within and among populations of Arabidopsis thaliana: comparison with markers and traits. Mol Ecol. 2005; 14: 4181–4192. pmid:16262868
  106. 106. Quinn GP, Keough MJ. Experimental Design and Data Analysis for Biologists. UK: Cambridge University Press; 2002.
  107. 107. Weir BS, Cockerham CC. Estimating F-statistics for the analysis of population structure. Evolution. 1984; 38: 1358–1370. pmid:28563791
  108. 108. Excoffier L, Lischer HEL. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Resour. 2010; 10: 564–567. pmid:21565059
  109. 109. Rosenberg MS, Anderson CD. PASSaGE: pattern analysis, spatial statistics and geographic exegesis. Version 2. Methods Ecol Evol. 2011; 2: 229–232.
  110. 110. van Berloo R. GGT 2.0: versatile software for visualization and analysis of genetic data. J. Hered. 2008; 99: 232–236. pmid:18222930
  111. 111. Sokal RR, Oden NL. Spatial autocorrelation in biology 1. Methodology. Biological Journal of the Linnean Society, 10: 199–228.
  112. 112. Moran PAP. Notes on continuous stochastic phenomena. Biometrika. 1950; 37: 17–23. pmid:15420245
  113. 113. Dutilleul P, Clifford P, Richardson S, Hemon D. Modifying the t test for assessing the correlation between two spatial processes. Biometrics. 1993; 49: 305–314.
  114. 114. Rangel TF, Diniz-Filho JAF, Bini LM. SAM: A comprehensive aplication for Spatail Analysis in Macroecology. Ecography. 2010; 33: 1–5.
  115. 115. Beale CM, Lennon JJ, Yearsley JM, Brewer MJ, Elston DA. Regression analysis of spatial data. Ecol Lett. 2010; 13: 246–264. pmid:20102373