Introduction

Tomato spotted wilt virus (TSWV), the type-member of the Tospovirus genus, family Bunyaviridae, is one of the most harmful viral pathogens. It ranks second on the list of the most important plant viruses worldwide (Scholthof et al. 2011). TSWV has a wide host range including more than 1000 plant species among weed species, ornamental and horticultural crops (Parrella et al. 2003; Hanssen et al. 2010), such as pepper (Capsicum annuum) and tomato (Solanum lycopersicum) (Persley et al. 2006; Pappu et al. 2009). The virus is naturally transmitted by several species of thrips (Thysanoptera: Thripidae) in a persistent and propagative manner with Frankliniella occidentalis being the most effective vector (Prins and Goldbach 1998). The genome of TSWV consists of three negative-sense or ambisense RNA segments: large (L, 8.9 kb), medium (M, 4.9 kb) and small (S, 2.9 kb). Segment L encodes a putative RNA-dependent RNA polymerase (de Haan et al. 1991); segment M encodes the cell-to-cell movement protein, NSm (Li et al. 2009), and the precursor of the surface glycoproteins, GN/GC, involved in TSWV transmission by thrips (Sin et al. 2005; Naidu et al. 2008); and segment S encodes the silencing suppressor, NSs (Takeda et al. 2002) and the nucleocapsid, N (de Haan et al. 1990; Pappu et al. 2009).

In tomato and pepper, the best strategy to control the disease has been to use the natural host resistance found in wild Solanum and Capsicum species (Stevens et al. 1992; Boiteux and de Ávila 1994). Sw-5 from Solanum peruvianum and Tsw from Capsicum chinense are the most effective resistance genes in tomato and pepper, respectively, and they are now deployed in commercial cultivars worldwide (Pappu et al. 2009). Tsw has been found to confer resistance to a wide spectrum of TSWV isolates (Moury et al. 1997), although a partial inefficiency by high temperatures after inoculation with wild type isolates on resistant peppers at an early stage of plant development has been described (Moury et al. 1998; Turina et al. 2012). However, the main setback has been the emergence of TSWV isolates able to overcome this resistance. Several resistance-breaking (RB) isolates of TSWV have been reported from pepper cultivars containing the Tsw gene in Louisiana, USA (Black et al. 1996), Italy (Roggero et al. 2002), Australia (Thomas-Carroll and Jones 2003) and Spain (Margaria et al. 2004).

The genetic determinant responsible for Tsw resistance breakdown in pepper was mapped to the S RNA segment by analysis of reassortants between RB and non-resistance-breaking (NRB) TSWV isolates (Jahn et al. 2000; Margaria et al. 2007). Contradictory evidences involving NSs or N proteins in breaking the Tsw gene resistance in pepper have been published (Margaria et al. 2007; Lovato et al. 2008; Tentchev et al. 2011), although recent studies have demonstrated that the NSs protein is the avirulence determinant (de Ronde et al. 2013, 2014).

Inoculation of NRB TSWV isolates in resistant pepper triggers a hypersensitive response (HR) inducing localized necrosis at the site of virus infection which prevents systemic infection, whereas RB isolates of TSWV produce systemic symptoms consisting of chlorosis and stunting. However, unusual biological behavior of the RB isolates have been observed, for example, the existence of RB isolates inducing severe necrotic lesions on uninoculated upper leaves in C. chinense plants (Margaria et al. 2007), or evidences of partial reversion from RB to NRB isolates after a few cycles of subculture in susceptible peppers (Thomas-Carroll and Jones 2003). Some of these unusual pathologic alterations could be attributed to different causes: (i) reassortment by exchange of entire genome segments between different TSWV isolates (Qiu and Moyer 1999; Tentchev et al. 2011); (ii) existence of a new TSWV pathotype generated by mutations in NSs or another gene and the selective pressure exerted by the resistance genes on TSWV isolates with different fitness (Aramburu et al. 2010; López et al. 2011); (iii) the effect of temperature changes on the HR expression provided by Tsw gene (Moury et al. 1998; Soler et al. 1998); and (iv) synergistic or antagonistic interactions caused by mixed infections (García-Cano et al. 2006; Murphy and Bowen 2006; Syller, 2012).

In this article, we analyzed the possible causes of the systemic necrosis found in field samples of resistant peppers infected by TSWV. We reproduced these symptoms by simultaneous inoculation of RB and NRB isolates of TSWV showing that they were produced by synergistic interactions between both types of TSWV isolates.

Materials and methods

Samples collection and TSWV detection

Samples of pepper plants carrying the Tsw gene showing systemic necrosis apparently caused by TSWV infection were collected during 2009 growing season from a commercial crop in northeastern Spain. Selected field samples were tested for TSWV infection by enzyme-linked immunosorbent assay (ELISA) (Clark and Adams 1977), using a specific polyclonal antiserum to TSWV, at recommended dilutions (Loewe Biochemica Gmbh, Sauerlach, Germany), as previously described (Aramburu et al. 2010). Samples were also tested by ELISA for the presence of Cucumber mosaic virus (CMV), Parietaria mottle virus (PMoV) and Potato virus Y (PVY), using specific antisera (Aramburu et al. 2010) to discard possible infection mixtures with the most frequent viruses found in pepper crops of this area.

Biological cloning

Two field samples positive to TSWV were mechanically inoculated on Nicotiana glutinosa plants to obtain homogeneous intra-isolate population of genetic variants. Several biological clones were obtained after three serial passages of a single local lesion. Mechanical inoculation was performed on seedlings at the 2-4 leaf stage by rubbing infected leaf extracts diluted (1:20, w:v) in 0.05 M phosphate buffer, pH 7.2, containing 0.2% 2-mercaptoethanol, 1% polyvinyl pyrrolidone, molecular weight 10.000 and activated charcoal. These TSWV clones were subsequently amplified in Nicotiana benthamiana plants and systemically infected leaves were stored at -80°C until use. TSWV isolates were then mechanically inoculated to pepper cv ‘Segura’ heterozygous for the Tsw resistance gene and pepper cv ‘Delfos’ without Tsw gene (both provided by FITO S.A., Barcelona, Spain) to evaluate the ability to overcome the resistance. In addition to ELISA tests, TSWV infection was monitored by symptom observation every day since 4 to 21 days post-inoculation (dpi).

Additionally, five TSWV isolates of our collection, previously characterized by their capacity to induce or break the resistance conferred by the Tsw gene were used in this work (Table 1). The isolates Da1NL2, Mon1NL2 and GRAU from tomato plants collected in north-east of Spain were grouped as NRB isolates, and the isolates Can2PC2 and Mu2PC2 from resistant pepper collected in Canary Island and in southeastern Spain, respectively, were grouped as RB isolates.

Table 1 Symptoms on uninoculated upper leaves of Nicotiana benthamiana and pepper plants 12 days post inoculation singly with resistance-breaking (RB) and non-resistance-breaking (NRB) isolates or co-inoculated with different pairs of RB + NRB isolates of Tomato spotted wilt virus (TSWV).
Full size table

Biological assays to test the effect of temperature

TSWV resistant pepper plants cv. “Segura” were used to test the effect of temperature on Tsw resistance. The symptom expression by RB and NRB isolates was conducted at two constant temperature regimes: i) 35°C, to inactivate the HR and ii) 25°C to activate the HR. A series of 10 pepper plants was singly inoculated with either NRB (Da1NL2 and Mon1NL2) or RB isolates (Can2PC2 and Mu2PC2). Inoculated plants were maintained during 12 dpi in temperature-controlled growth chambers (Havell Sylvania SA, Madrid, Spain), with 16-h photoperiod. Four plants were included as control in each serie of plants, two N. benthamiana plants used as positive controls of TSWV infection, and two uninoculated pepper plants were used to evaluate the effect of temperature on plant growth.

Biological assays to test the interaction of RB and NRB isolates of TSWV

To test the interaction of RB and NRB isolates, pairs of RB + NRB isolates were co-inoculated on a series of 10 pepper cv. ‘Segura’ plants. Inoculated plants were kept in a thrips-free growth chamber during 12 dpi with 16/8 h light/dark cycle and constant temperature of 25°C to activate the HR. Inocula of the six possible pairs of RB + NRB isolates were prepared mixing equivalent amounts of each isolate. The amount of each isolate was determined by ELISA and by the number of lesions produced on N. glutinosa as previously described (Aramburu et al. 2010). Pepper cv. ‘Segura’ and N. benthamiana plants singly inoculated with RB and NRB isolates and pepper cv “Delfos” and N. benthamiana plants co-inoculated with the same pairs of RB + NRB isolates were used as controls. Plants were monitored for symptom development every day since 4 to 12 dpi.

Identification of RB and NRB isolates in co-inoculated resistant peppers plants

Both, molecular and biological analysis at 12 dpi were carried out on resistant pepper plants co-inoculated with Da1NL2 + Mu2PC2 isolates to determine the presence or absence of each isolate. The molecular analysis included: (i) a reverse transcription coupled with polymerase chain reaction (RT-PCR) combined with restriction fragment length polymorphism (RFLP) analysis of the M and S segments, and (ii) partial sequence analysis of the RT-PCR products of the M and S segments. RT-PCR products were amplified using the primer pairs M1F 5´-GTTATAGGATAATTATCTTGTGTC-3´ and M1R 5´-AGAGCAATCAGTGCAAACAAAAACCTTAATCC-3´ and S1F 5´-GATCGAGATGTGCTATAATCAAGC-3´ and S1R 5´- GAACCTGTGCAAAAGATGTGTGAG-3´ for M and S segments, respectively. They were designed from conserved sequences after comparing the full-length sequence of the M and S segments of the two isolates. Differences in the sequence of these RT-PCR products allowed to be selectively digested with EcoRI (MBI Fermentas, Madrid, Spain). Ten microlitres of PCR products and 10 U of the restriction enzyme were incubated for 2 h at 37°C. PCR products or their restriction fragments were separated by electrophoresis in agarose gels, stained with ethidium bromide, and the DNA was visualized under UV lighting.

Biological analysis was performed following the next steps: (i) leaf extracts of resistant pepper plants showing necrosis on apical leaves were inoculated on N. benthamiana plants, a non selective host used to favor the multiplication of the infectious mixture; (ii) leaf extracts showing symptoms of systemic infection were inoculated on N. glutinosa plants to obtain biological clones; (iii) leaf tissue disks of 30 local lesions obtained 4 dpi were inoculated onto individual N. benthamiana plants to multiply the different clones; (iv) each biological clone was inoculated to three resistant pepper plants to evaluate the symptoms; and (v) finally, infectious tissues of N. benthamiana corresponding to the clones that induced only HR on inoculated pepper leaves were analyzed by molecular techniques, as described above, to identify the TSWV isolate present in the infection.

Results

Biological characterization of TSWV causing severe symptoms in field resistant peppers

Pepper samples with the Tsw gene showing severe and unusual symptoms consisting in necrotic lesions, chlorosis, stunting and apical necrosis which later on evolved to petiole and stem collapse were collected from the field (Figure 1). Two pepper samples that tested positive for TSWV and negative for CMV, PMoV and PVY by ELISA were selected to obtain biological TSWV clones by isolating local lesions in N. glutinosa plants. After transmission by mechanical inoculation to pepper plants carrying the Tsw gene, these biological TSWV isolates segregated into two different types of symptoms: (i) severe symptoms consisting in necrotic lesions on inoculated leaves at 4-6 dpi, chlorosis, stunting, systemic necrosis with asymmetrical distribution on uninoculated leaves and petiole collapse of some leaves at 8-10 dpi, and stem collapse at 12-14 dpi, that eventually induced plant death at 16 dpi (Figure 2A), and (ii) typical symptoms of RB isolates on resistant peppers characterized by chlorosis and stunting of apical leaves (Figure 2B).

Fig. 1
figure 1

Field resistant pepper plant showing unusually severe symptoms of tomato spotted wilt virus characterized by local lesions, chlorosis and apical necrosis

Full size image
Fig. 2
figure 2

Symptoms in tomato spotted wilt virus resistant pepper plants after inoculation with biological clones of the virus obtained from field samples. A local lesions on inoculated leaves, chlorosis, stunting and local necrosis on uninoculated leaves and stem collapse at 21 days post inoculation (dpi). B image showing chlorosis and stunting of apical leaves at 10 dpi

Full size image

Effect of temperature on symptom expression in pepper plants carrying the Tsw gene

To assess whether the effect of temperature could be involved in the expression of systemic necrosis, two regimes of constant temperature, 25 and 35°C, were assayed after inoculation of pepper plants carrying the Tsw resistance gene with RB (Can2PC2 and Mu2PC2) or NRB (Mon1NL2, Da1NL2 and GRAU) isolates. Results were consistent in five independent experiments for each assay condition. Uninoculated pepper plants kept at 35°C did not show any significant change in growth compared to those maintained at 25°C after 12 days. Plants inoculated with NRB isolates only showed HR lesions on inoculated leaves at 25°C, whereas plants got systemically infected at 35°C. Plants inoculated with RB isolates were systemically infected at both 25°C and 35°C and without HR in inoculated leaves. In all cases, systemic symptoms consisted of chlorosis and stunting, but none showed the apical necrosis observed in field peppers (data not shown).

Synergism between NRB and RB isolates of TSWV in resistant pepper plants

Pepper plants carrying the Tsw resistance gene were co-inoculated with pairs of RB + NRB isolates to assess whether mixed infections could be involved in the expression of systemic necrosis. Plants were grown at constant temperature of 25°C to maintain the HR. These plants showed local lesions on inoculated leaves and systemic infection, which in several plants (Table 1) consisted of typical chlorosis and stunting (as shown in Figure 2B), whereas a clear systemic necrosis was observed in other plants (as shown in Figure 2A). The proportion of infected pepper plants showing systemic necrosis for each RB + NRB combination was: Can2PC2 + Mon1NL2 (0/10), Can2PC2 + Da1NL2 (3/10), Can2PC2 + GRAU (4/6), Mu2PC2 + Mon1NL2 (10/10), Mu2PC2 + Da1NL2 (5/10) and Mu2PC2 + GRAU (1/10). Furthermore, new lots of 10 resistant pepper plants inoculated with leaf extracts from eight different pepper plants showing systemic necrosis segregated into the two types of symptoms previously described (data not shown).

Identification of TSWV isolates present in co-inoculated pepper plants

To confirm the presence of RB + NRB isolates in co-inoculated resistant peppers, the Mu2PC2 + Da1NL2 combination was randomly selected and upper leaves showing necrotic symptoms were subjected to RFLP and sequence analysis. For this purpose, primer pairs M1F/M1R and S1F/S1R were designed to amplify, respectively, the last 692 nt of the M segment and a specific region of 545 nt of the NSs gene (avirulence determinant) of the S segment. Amplified products of M and S segments from samples inoculated with Da1NL2 or Mu2PC2 isolates alone could be digested with EcoRI. The amplicon from the M segment of Da1NL2-inoculated plants resulted into two bands of 93 bp and 599 bp each after digestion with EcoRI (Figure 3A, lane 1), whereas no bands digestion were obtained for the amplicon from the S segment with this enzyme (Figure 3B, lane 1). In contrast, the amplicon from the M segment of Mu2PC2-inoculated plants could not digested with EcoRI (Figure 3A, lane 2), whereas two bands of 127 bp and 418 bp each were obtained after digestion of the amplicon from the S segment (Figure 3B, lane 2).

Fig. 3
figure 3

RT-PCR-RFLP pattern of M A and S B segments amplified from apical necrotic leaves of resistant peppers individually inoculated with tomato spotted wilt virus Mu2PC2 Da1NL2, or a mixture of both isolates. Lane M: 100 bp DNA ladder marker; lanes 1 and 2 EcoRI digestion products of Da1NL2 and Mu2PC2 isolates respectively; lanes 3–7 EcoRI digestion products of Mu2PC2 + Da1NL2 co-inoculation

Full size image

In resistant peppers co-inoculated with the combination of isolates Mu2PC2 + Da1NL2, only the M and S segments corresponding to the Mu2PC2 isolate were detected at 12 dpi by RFLP assays (Figure 3A, lanes 3-7 and Figure 3B, lanes 3-7, respectively). Sequence analysis of these RT-PCR products from five plants confirmed that only Mu2PC2 isolate was present in the upper leaves (data not shown).

On the other hand, leaf extracts from resistant peppers co-inoculated with a combination of Mu2PC2 and Da1NL2 were inoculated on N .glutinosa plants to obtain biological clones. Thirty biological clones were analyzed by inoculation on resistant pepper plants which induced diverse symptomatology: two clones induced the typical symptoms of NRB isolates characterized by local lesions on inoculated leaves (HR), 19 clones induced the typical symptoms of RB isolates characterized by chlorosis and stunting and nine clones reproduced the severe systemic necrosis (Figure 4, left, middle and right, respectively). Sequence comparisons of the RT-PCR products from N. benthamiana plants infected with the two clones that only induced HR showed a 100% homology with the M and S segments of the Da1NL2 NRB isolate (data not shown).

Fig. 4
figure 4

Symptoms in leaves of resistant pepper inoculated with tomato spotted wilt virus isolates and maintained at constant temperature of 25°C. Left, local lesions induced by hypersensitive response in inoculated leaves with a non-resistance-breaking (NRB) isolate at 6 days post inoculation (dpi). Middle, chlorosis at 10 dpi on apical uninoculated leaves caused by systemic infection of a resistance-breaking (RB) isolate. Right, chlorosis and necrotic lesions on apical uninoculated leaves induced by systemic infection of a mixture of RB + NRB isolates at 10 dpi

Full size image

Discussion

Pepper plants carrying the Tsw resistance gene showing unusually severe systemic necrosis were collected in a field survey during 2009 growing season in northeastern Spain. Similar symptoms were reproduced when leaves extracts of these samples were directly inoculated on resistant peppers plants under controlled conditions. After ruling out the presence of other viruses in the field samples, we concluded that these symptoms could be due to the existence of infrequent TSWV RB isolates. These isolates would produce necrotic lesions on the upper uninoculated leaves as a consequence of an inefficient HR, as it has been previously described (Moury et al. 1997; Margaria et al. 2007). However, TSWV biological clones obtained from field infected samples segregated into two types of symptoms in resistant peppers: (i) the typical symptomatology caused by RB isolates characterized by chlorosis and stunting, and (ii) the severe systemic necrosis described above. This segregation suggested the possible existence of mixed infection with different types of TSWV isolates in the field samples.

The influence of elevated temperatures on TSWV-induced HR has been widely studied (Black et al. 1991; Gil and Luis 1994; Moury et al. 1997, 1998; Soler et al. 1998, 1999), but inconclusive results have been reported. One explanation could be that these experiments were performed by alternating high and low temperatures according to daily cycles (night/day). In fact, tests conducted by Roggero et al. (1996) at constant temperature of 33°C showed that resistant pepper plants developed systemic infection after inoculation with wild-type isolates, while all the inoculated plants grown at lower temperatures of 18 to 24°C showed necrotic local lesions by HR independently of plant age. For this reason, HR evaluation in our study was performed using regimes of constant temperature at 25°C or 35°C, which provided homogeneous results on each repetition. We found that HR was inactivated at 35°C on resistant peppers and both NRB and RB isolates induced indistinguishable systemic infection, whereas at 25°C, HR blocked the multiplication of NRB isolates on inoculated leaves. Considering these results, the systemic necrosis observed at constant temperature of 25°C in some resistant pepper plants, after co-inoculation with mixtures of RB + NRB isolates, should not be a consequence of HR inactivation against the NRB isolate. These symptoms could suggest the presence of the NRB isolate on uninoculated apical leaves. However, the detection of the NRB isolate in several uninoculated apical leaves of resistant peppers co-inoculated with RB + NRB isolates was not possible by RT-PCR-RFLP assays and nucleotide sequence analysis. In both cases, it was only possible to detect M and S segments of the RB isolate. It should be noted that RT-PCR-RFLP assays are useful for determining the prevalence of one isolate in a mixture of isolates, but fails to detect one isolate when its presence in the mixture is at very low titer (Aramburu et al. 2010). Finally, the presence of the NRB isolate in the uninoculated apical leaves of resistant peppers was demonstrated using biological clones from individual local lesions obtained on N. glutinosa. These biological clones were obtained after mechanical inoculation of extracts from apical necrotic leaves, followed by sub-culturing in the non selective host N. benthamiana to promote viral replication. Of a total of thirty clones, only two of them induced the typical symptoms of the NRB isolates consisting in HR on inoculated leaves and absence of systemic infection. The nucleotide sequence analysis of M and S segments from these two clones confirmed that both belonged to the NRB isolate.

Our finding show that NRB isolates can infect resistant peppers in the presence of RB isolates inducing severe symptoms consisting of systemic necrosis due to HR. Our results demonstrate that this type of synergistic interaction would occur frequently and besides, it is facilitated when NRB isolates were inoculated 3 to 6 days after the inoculation with RB isolates (data not shown). The continuous effect of HR in the whole plant against the NRB isolate causes systemic necrosis and sometimes collapse, which indirectly affects the multiplication of the RB isolate as well. The low proportion of the NRB isolates in resistant pepper plants infected with a mixture of RB + NRB isolates could have been the cause of some erroneous conclusions. Margaria and co-workers (2007) reported the existence of some TSWV RB isolates that induced systemic HR on resistant peppers. Subsequently, these observations were explained as a result of a partial recognition of the avirulence determinant of RB isolates (de Ronde et al. 2013), as described for potato virus X and the Rx resistance gene. However, in view of our results, the systemic HR could be also explained as consequence of a mixed infection between RB and NRB isolates. Thomas-Carroll and Jones (2003) also showed evidences of a partial reversion of TSWV RB isolates to NRB after five serial passages in susceptible peppers. However, it is highly unlikely that mutations able to break the resistance conferred by the Tsw gene repetitively can revert to wild-type behavior after a few passages in a susceptible host. This result could be also explained by a random selection of NRB isolates after several passages in a non selective host from a mixture of RB + NRB isolates able to break the Tsw resistance due to the synergistic interaction.

Resistant pepper plants co-infected with both types of isolates showing severe symtomatology could have epidemiological implications. As a consequence of a continuous HR in whole plant, virus multiplication, acquisition by thrips or even the insect reproduction could decrease the secondary spread of the virus.

Although synergistic interactions are known to be produced predominantly by unrelated viruses that infect the same host, for example breakdown of resistance to TSWV in tomato (García-Cano et al. 2006), they have also been reported for more or less closely related virus species belonging to the same family (Syller, 2012). However, to our knowledge, this is the first description of a synergic interaction occurring between isolates of the same virus species. Synergistic interactions have a facilitative effect on both, or at least one of the viruses, manifested by an increase in virus(es) replication (Syller, 2012). This is true in our case, since NRB isolates are able to systemically infect pepper plants with the help of RB isolates. However, in a second phase HR triggered by the NRB isolates, hampers the replication of RB isolates due to widespread necrosis, therefore, this interaction could become antagonistic for RB isolates.

Currently, the exact mechanism that makes possible this interaction remains unknown. It could be a consequence of a reduced HR, which could trigger programmed cell death not fast enough to block the multiplication of the virus mixture, which would allow that a few NRB virions continuously escape of the local lesions for inducing a systemic HR in the whole plant.