Red raspberry ( Rubus idaeus L.) and other Rubus species are hosts of more than 30 viruses and virus-like agents (Martin et al., 2013). Most of the viruses in­fecting raspberries do not cause symptoms on eco­nomically important cultivars, and only a few cause visible symptoms on different plant parts. Different types of symptoms may be expressed on the leaves and fruits of sensitive cultivars, particularly in mixed virus infections. Symptoms induced by viruses can be easily misinterpreted as symptoms caused by insects or other pathogens. Yellow patches and blotches on leaves were commonly found in red raspberry orchards and for a long time were described as infestation by the raspberry leaf and bud mite ( Phyllocoptes gracillis Nal.) (Gratwick, 1992). P. gracillis feeds on raspberries, causing pale green and yellow patches and blotches, twisting and distortion of the leaf margins. The fruits or individual drupelets may also be affected. McGavin et al. (2012) reported the presence of a new negative-strand RNA virus in the red raspberry plants with leaf blotch symptoms. The virus was also detected in the raspberry leaf and bud mite, suggesting its vector role, and was named Raspberry leaf blotch virus (RLBV). RLBV belongs to the genus Emaravirus with a geno­me that is 17,410 nucleotides (nt) long and consists of eight segmented negative sense RNAs, each enco­ding a single open reading frame (McGavin et al., 2012; Lu et al., 2015). RLBV has been confirmed only in Euro­pean countries: Great Britain, Serbia, Finland, Bulgaria, Montenegro and Poland (Jevremović et al., 2016).

Red raspberry is economically the most important small fruit in Serbian agriculture, ranking Serbia among the World’s leading producers and exporters (Petrović et al., 2017). Raspberry viral diseases were not studied extensively in the past decades in Serbia and were not considered a significant threat to red raspberry production. Most of the detected viruses were causing latent infections or yellow mosaic symptoms (Dulić-Marković & Ranković, 1997). In the past de­cade, a mass occurrence of symptoms that include ye­llow blotching and twisting of raspberry leaves was noticed in orchards throughout Serbia. These symp­toms were primarily attributed to infestation by the raspberry leaf and bud mite. After the detection of RLBV in symptomatic red raspberry samples from Serbia (Mc­Gavin et al., 2012) we started an intensive survey of the presence of RLBV in the country to determine the association between the expressed symptoms and viral infection.

The aim of our study was to evaluate the reliability of previously reported primer pairs for RLBV detection and to access the genetic diversity of the selected RLBV isolates from Serbia.

A total of 96 samples of raspberry leaves were analysed for the presence of RLBV using three reco­mmended RNA-specific RT-PCR primer pairs for RLBV detection targeting RNA1, RNA3, and RNA5 (McGavin et al., 2012). RT-PCR results are presented in Table 1. Significant differences were shown in the efficiency of the used primer pairs. RLBV RNA 1 was detected in 56 samples (58.3%), RNA 3 in 66 (68.7%), and RNA5 in 41 (43.7%). Primer pair 1287/1095, targeting RNA3, proved to be the most efficient for RLBV detection in red raspberry samples from Serbia. Pri­mer pairs targeting RNA1 and RNA5 failed to de­tect RLBV in 10 and 24 samples, respectively. RLBV was confirmed in 65 samples with leaf blotch symp­­toms, and only in 1 sample from the asymptomatic raspberry ’Willamette’. RLBV was confirmed in raspberries ‘Fertödi Zamatos’, ‘Glen Ample’, ‘Meeker’, ‘Polana’, ‘Tulameen’, and ‘Willamette’, but not in ‘Heritage’, ‘Polka’, and in two analysed wild raspberries with leaf chlorosis. There was no specific association between the type and intensity of the symptoms with the host cultivar. Mild or severe mottling and yellow blotches are the symptoms that were commonly found in all infected cultivars. RLBV presence was not confirmed in 31 samples, of which 16 were symptomless, and in 15 samples showing leaf yellows, leaf chlorosis, and vein chlorosis.

Nucleotide sequences of the 567 bp fragment of the nucleocapsid were determined for 21 selected iso­lates from 4 cultivars: ‘Fertödi Zamatos’ (1), ‘Glen Ample’ (1), ‘Meeker’ (4), and ‘Willamette’ (15). Nu­cleotide sequences were deposited in the GenBank sequence database ( https://www.ncbi.nlm.nih.gov/) and assigned the accession numbers MF136659−MF136679. The nucleotide and amino acid sequences of the analysed Serbian isolates were 93.2−100% and 94.1−100% identical, respectively. Three se­quenced isolates (RS-RLBV-4, RS-RLBV-104, and RS-RLBV-146) showed identical nt sequences. Iso­lates RS-RLBV-138 and RS-RLBV-147 showed the lowest percentage of nucleotide identity (93.2%). Phy­logenetic analysis of the sequen­ces of the Serbian isolates revealed the existence of two groups of isolates (Fig. 2).

Serbian isolates were then aligned with the availa­ble RLBV sequences retrieved from the GenBank sequen­ce database including 21 isolates from Fin­land (Acc. JQ684678, KP644139−45 and KP730587−99), two from Slovakia (Acc. KY513312−13) and one from Scotland (Acc. FR730598). In total, 45 nt sequences we­re analysed. A reconstructed phylogenetic tree u­sing the Maximum Likelihood method based on the Ta­mura 3-parameter model with gamma distributed sites revealed the existence of two clusters of closely rela­ted isolates: cluster I included 17 isolates from Serbia, 19 from Finland and 2 from Slovakia, while cluster II included 4 isolates from Serbia, 2 from Finland and 1 from Scotland (Fig. 3).

Using the sequences of 21 Serbian and 21 Finnish isolates retrieved from the GenBank database, genetic distances were estimated by implementing the Tamura 3-parameter model with MEGA7 soft­ware. Standard error estimates were obtained by a bootstrap procedure (1,000 replicates). The results showed that the overall mean value of nucleotide diversity was 0.026±0.004 (the number of base substitutions per site from ave­ra­ging the overall sequence pairs±standard error esti­mate). The value of nucleotide diversity of 21 Serbian RLBV isolates was 0.026±0.004, and was 0.024±0.004 for 21 Finish isolates. The value of nucleotide di­ver­si­ty between groups of Serbian and Finnish RLBV isolates was 0.027±0.004.

The presence of RLBV was confirmed in several European countries with more or less significant raspberry production (Jevremović et al., 2016). Raspberry leaf blotch symptoms were noticed in raspberry orchards in Serbia more than five decades ago and were in general attributed to the feeding damages caused by the raspberry leaf and bud mite (Dobrivojević & Petanović, 1985; Milenković & Marčić, 2011). The first RLBV-infected red raspberry samples in Serbia were reported by McGavin et al. (2012), but with no further details on the analysed samples. After this report, a large-scale study on the RLBV presence and distribution in Serbia has begun. The first analysis for RLBV presence in Serbia was performed by RT-PCR using specific primers for RNA5 (McGavin et al., 2012). These primers failed to detect RLBV in numerous samples with typical raspberry leaf blotch disease symptoms. Therefore, a part of this study was to evaluate the reliability of the different primer pairs for RLBV detection. Primer pair 1287/1095 targeting RNA3 proved to be the most efficient among the three primer pairs used for RLBV detection in red rasp­berry samples from Serbia and should be further used as the preferred primer pair for RLBV detection in the country. RNA3 was also recommended as the preferred target for RT-PCR diagnostics in the study of Dong et al. (2016). The RT-PCR detection with other primer pairs used in our study was less reliable, especially with the primer pair used for the detection of RNA5 (Table 1). RNA5 specific primers failed in the RLBV detec­tion in 24 samples that were positive with RNA3 primers. The high diversity of Serbian isolates in the region targeted by RNA5 primers may provide an explanation for the obtained results. Also, another possibility is that some RLBV isolates do not contain RNA5 (Dong et al., 2016). The RNA 5 and its encoded protein P5 are unique to RLBV and some other Emaraviruses (McGavin et al., 2012; Kumar et al., 2017). Possible functional role(s) of this viral protein are still unknown.

Our results confirmed the presence of RLBV in six cultivars in all surveyed localities and the strong association of leaf blotch symptoms with RLBV. The red raspberry ‘Willamette’ is the predominant floricane cultivar grown in Serbia for decades, with a 90% share of production. ‘Meeker’ accounts for 5%, with all other cultivars combined. In the past several years, many growers reported a high incidence of severe leaf blotch symptoms and significant yield decrease in ‘Willamette’ and ‘Meeker’ orchards grown according to organic farming. RLBV was confirmed in all analysed symptomatic samples from these orchards (Jevremović, unpublished results). Due to the forbidden use of common acaricides, the control of raspberry leaf and bud mite in organic farming is difficult.

Nucleotide sequence analysis of the selected isolates from different cultivars and locations revealed considerable nucleotide diversity among the isolates. The high value of genetic diversity may suggest the long-term presence of the RLBV in Serbia. Compared to the RLBV sequences from other countries, Serbian isolates showed the highest nucleotide identity with the Slovakian isolates. Isolate RS-RLBV-76 share 99.5% of the nt identity with the Slovakian isolates SK-RLBV-1 and SK-RLBV-2.

Phylogenetic analysis of the partial nucleocapsid protein gene sequences of 45 RLBV isolates from Serbia, Finland, Slovakia, and Scotland confirmed the existence of two clusters of isolates. The reconstructed ML tree did not provide strong evidence for a specific clustering of RLBV isolates in clusters I and II according to geographical origin (Fig. 3). However, some clustering at a smaller spatial scale was detected in some cases, but not supported by bootstrap analysis. There are no further speci­fic clustering of the Serbian isolates based on the local geographic origin or host cultivar. This may indicate the intense gene flow within the country through the latent infections of the planting material, and further virus transmission by the raspberry leaf and bud mite that was described as the most important secondary pest in Serbian commercial raspberry orchards (Milenković & Marčić, 2011). Three isolates (RS-RLBV-4, RS-RLBV-104, and RS-RLBV-146) were with identical nt sequences. These iso­lates were sampled from locations that are 40−50 km from each other. On the contrary, the most divergent isolates (RS-RLBV-138 and RS-RLBV-147) were located at approximately the same distance (50 km) in the same region. The practice of using the planting material from commercial orchards may contribute to the dispersal of RLBV isolates to close and distant areas.

RLBV is a pathogen that is not monitored during the production of the planting material, and not listed in the recommendations for the production of healthy planting material given by the EPPO (OEPP/EPPO, 2009). Therefore, it may be distributed via the infected material within and between countries. After RLBV discovery and its confirmed presence in several European countries, it should be expected that RLBV will be incorporated in the next revised version of the EPPO certification scheme for Rubus species.

The molecular analysis and characterisation done in this study provides information regarding the reliable detection of Raspberry leaf blotch emaravirus, and confirmed its wide distribution and significant genetic diversity in Serbia. Research into the impact of RLBV infection on vegetative growth and yield is underway.