Serological and molecular analysis indicates the presence of distinct viral genotypes of Apple stem pitting virus in India

Abstract

Recombination leads to the generation of new viral progeny which remain undetected by routine testing procedures and may be a threat to the infected host. Here, we have characterised the complete genome sequences of two isolates of Apple stem pitting virus from apple cv. Red Chief (Palampur) and cv. Gold Spur (N) with distinct serological reactivities. The viral genomes consisted of 9267 nucleotides for isolate Palampur and 9254 nucleotides for isolate N, excluding the poly (A) tail and contained 5five open reading frames (ORFs). Isolate N shared 80.8% sequence identity with ASPV apple isolate GA2 from China, while isolate Palampur shared 81.4% sequence identity with ASPV apple isolate PB66 from the United Kingdom. The serological difference of isolates N and Palampur along with their low sequence identity indicated the existence of two distinct virus genotypes which was corroborated by evolutionary and genetic differentiation analyses. Recombination events were detected in the RdRp and CP sequences of Palampur isolate thereby suggesting the role of recombination in the evolution of distinct virus genotypes.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-021-02798-5.

Introduction

Apple stem pitting virus (ASPV), a widely distributed fruit tree filamentous virus, is the type species of the genus Foveavirus of the family Betaflexiviridae (Adams et al. 2004). ASPV is associated with some economically important diseases, such as quince fruit deformation, epinasty and decline of Spy 227 and pear vein yellows (Jelkmann 1994; Leone et al. 1998; Mathioudakis et al. 2006, 2009; Stouffer 1989). Green crinkle disease symptoms in apple and quince are associated to apple green crinkle virus (AGCaV), a distinct variant of ASPV, whereas for the former ASPV was recently proved as the causal agent of the disease (Li et al. 2020; James et al. 2013; Morelli et al. 2017). The ASPV natural host range is mainly restricted to commercial pome fruits, and recently was also reported from apple- and pear-rootstocks, ornamental and wild Rosaceae species (Constable et al. 2007; Jelkmann 1994; Ma et al. 2008; Mathioudakis et al. 2010, 2011) with an exception infecting cherry (Yang et al. 2017). In India, ASPV was first reported in apples and the survey results showed the highest infection rates of ASPV among the viruses infecting apple in Himachal Pradesh, India (Dhir et al. 2010; Kumar et al. 2012), in addition to its detection in pear (Brakta et al. 2013). The virus has a single-stranded positive-sense RNA genome of size 9.2–9.3 kb encompassing five open reading frames (ORFs). ORF1 encodes an alpha-like replication protein, while ORFs 2–4 encode the triple gene block (TGB) proteins for the movement function and ORF5 encodes the coat protein (CP) (Jelkmann 1994). To date, there are 25 completely sequenced ASPV isolates, both from apple and pear as hosts, which are deposited in the GenBank database.

Previous studies performed based on the partial as well as complete CP and RNA dependent RNA polymerase (RdRp) gene sequences have shown high genetic diversity (Gadiou et al. 2010; Komorowska et al. 2011, 2019; Liu et al. 2012; Ma et al. 2019; Mathioudakis et al. 2010; Yoon et al. 2014; Yoshikawa et al. 2001). Isolates remained undifferentiated based on the host they infect as well as based on their origin suggesting their polyphyletic nature. It is known that the members of the genus Foveavirus show a high level of genetic diversity, and ASPV is a highly variable virus (James et al. 2013). This high genetic variability among ASPV isolates affects the serological and molecular detection procedures for routine diagnostics (Jelkmann and Keim-Konrad 1997). Another reason for the reduced immunological reactivity of the antiserum could be due to the high variability at 5′-end of ASPV CP leading to an alteration in its structure (Chooi et al. 2013; Dhir et al. 2011; Komorowska et al. 2011; Ma et al. 2019). New strains and/or recombinant isolates remain a potential threat to the apple industry.

Recombination plays an important role in RNA virus evolution and may give rise to potential new isolates (Li et al. 2016; Nagy and Bujarski 1996). Recombination is quite prevalent in ASPV in the CP gene (Dhir et al. 2011; Komorowska et al. 2011, 2019; Li et al. 2016). Moreover, the recent study of Komorowska et al. 2019 revealed the existence of genetic variants within the viral population containing deletions of different lengths in the CP gene. This study revealed that mutations, drift, selection pressure and recombination are factors that shape the ASPV population structure. Moreover, it has been shown that ASPV creates quasispecies-like populations, which seemed to be uncorrelated with the host species but associated with the severity of symptoms (Komorowska et al. 2019).

In this study, two isolates of ASPV were characterised based on the difference in their interactions with commercial antiserum, and their complete genome sequences was determined and compared with others as described in the data. The goal of our research was to establish the phylogenetic relationship among ASPV isolates, including the new ones described in the present study, analyse the genetic structure of ASPV population and indicate the forces driving ASPV evolution. The results will contribute to further defining the possible evolutionary relationship of ASPV isolates from different hosts and origins, and also call for the need to raise a strain- specific antiserum or mixture for the detection of ASPV isolates in the field.

Materials and methods

Viral isolates, ASPV detection and single-strand conformation polymorphism (SSCP) analysis

Four samples each originating from apple cv. Gold Spur and Red Chief showing symptoms of vein yellows, chlorosis/necrosis and leaf curling, respectively, cultivated in the Council of Scientific & Industrial Research—Institute of Himalayan Bioresource Technology (CSIR-IHBT) apple field located in Kangra, were collected and analysed for the presence of ASPV with DAS-ELISA using commercial ASPV antibodies (Bioreba, Switzerland) and RT-PCR using common detection primers ASPVS and ASPVC (Table 1).

Table 1.

List and sequence of primers and PCR conditions used in this study

Primers	Primer sequences (5′-3′)	PCR conditions	Amplic (bp)	Region of the genome (gene)/ Ref. sequence	Viral isolates/Reference
ASPVF1	AAATTATAAAGGCAACTTCAAAGGAAC	Denaturation-95-30 s, 8 cycles Denaturation-94-20 s, Annealing-56-30 s, Extension-68-9′, 27 cycles Denaturation-94-20 s, Annealing-58-30 s, Extension-68-9′, Final extension 72-10′	~ 9.2 kb	1–27 (5′-UTR)/ FR694186	Palampur/ This study
ASPV-R	GAAAATCTAGTTAAAAAAAAAATAAAA			9240–9267 (3′-UTR)
ASPV F	ATGGCCTCCTTATGGGGTC	95-30 s, 32 cycles-94-20 s, Annealing-57-30 s, Extension-68-1′30 s, Final extension-72-10′	~ 1.2 kb	7942–7960 (5′-CP)/ FR694186	Palampur/ (Dhir et al. 2011) Palampur/ (Dhir et al. 2011)
ASPV C	TTGGGATCAACTTTACTAAAAA			9177–9204
ASPVF1(n)	AAATTATAAAGGCAACTTCATAA	Denaturation-95-30 s, 8 cycles Denaturation-94-20 s, Annealing-50-30 s, Extension-68-5′, 27 cycles Denaturation-94-20 s, Annealing-52-30 s, Extension-68-5′, Final extension 72-10′	~ 4.8 kb	1–23 (5′-UTR)/ LM999967	N/ This study
ASPV48k	CTTCTGAAAATTCAAACCAGTGGACT			4787–4812
ASPV3980f	GGGGCTGAAATTGATTACAGACCCTC	Denaturation-95-30 s, 8 cycles Denaturation-94-20 s, Annealing-63-30 s, Extension-68-4′, 27 cycles Denaturation-94-20 s, Annealing-65-30 s, Extension-68-4′, Final extension 72-10′	~ 3.8 kb	3952–3977 (Replicase)/ LM999967	N/ This study
TGB3R	AATATTAAGCTACCACACCT			7818–7836 (MP)
TGB1F	ATGGAAACTGTGCTTAGT	Denaturation-95-30 s, 8 cycles Denaturation-94-20 s, Annealing-48-30 s, Extension-68-2′ 30 s, 27 cycles Denaturation-94-20 s, Annealing-50-30 s, Extension-68-2′ 30 s, Final extension 72-10′	~ 2.5 kb	6661–6678 (Replicase)/ LM999967	N/ This study
ASPV-R	GAAAATCTAGTTAAAAAAAAAATAAAA			9228–9254 (3′-UTR)
TGB3F	ATGTTTCCGAAAAGTGGTCTTGG	95-30 s, 32 cycles-94-20 s, Annealing-58-30 s, Extension-68-1′30 s, Final extension-72-10′	~ 1.6 kb	7605–7626 (TGB3)/ LM999967	N/ This study
ASPVCPN	ATGGCTTCAAGTAGCCCAAA	95-30 s, 32 cycles-94-20 s, Annealing-54-30 s, Extension-68-1′30 s, Final extension-72-10′	~ 1.2 kb	7930–7951 (5′-CP)/LM999967	N/ This study
ASPVNC	TGGGGATCAACTTTACTAAAA			9171–9191 (3′-CP)/ LM999967
Oligo d(T) anchor primer	GACGACGCGTATCGATGTCGACTTTTTTT	Denaturation-95-1′, 10 cycles Denaturation-94-15 s, Annealing-55-30 s, Extension-72-40 s, 25 cycles-Denaturation-94-15 s, Annealing-55-30 s, Extension-72-40 s*, Final extension 72-7′ *-elongate each successive cycle by additional 20 s	~ 550 bp	-	Palampur & N/ 2nd Generation 5′/3′ RACE kit (Roche, USA)
GSPI (R)/5′RACE	TCCATGCATCCAGGTATGAGATCGACC			519–545 (Replicase)/FR694186 & LM999967	Palampur & N/ This study
GSPII (R)/5′ RACE	CCTTGGTCCAGTAATGTAGTTCATCG			573–598 (Replicase)/ FR694186 & LM999967	Palampur & N/ This study
ASPV Sense (Detection)	ATGTCTGGAACCTCATGCTGCAA	95-30 s, 32 cycles-94-20 s, Annealing-62-30 s, Extension-68-30 s, Final extension-72–10′	~ 370 bp	8869–8895 (CP/Accession D21828)	Palampur & N/ (Menzel et al. 2002)
ASPV anti sense	TTGGGATCAACTTTACTAAAAAGCATAA			9211–9238 (3′-UTR)

The total RNAs were isolated from infected leaves and further used to amplify complete CP gene but no amplification was obtained from cultivar Gold Spur using the common primer pair designed earlier, ASPVF and ASPVC (Dhir et al. 2011; Table 1). PCR was then carried out using TGB3 forward primer (TGB3F) and ASPVNC reverse primer and as a result an amplicon of the size ~ 1.6 kb was obtained and sequenced. Based on the new sequence, a primer set (ASPVCPN and ASPVNC) was designed for the CP gene amplification of cv. Gold Spur. The fragment encompassing the 5′-end of the virus using primers ASPVF1 and GSPIIR (containing 5′-UTR and partial replicase) was also amplified (Table 1). The samples were subjected to SSCP analysis to identify sequence variants at various points (Magome et al. 1999). Briefly, the PCR products obtained were mixed with loading buffer (95% formamide, 20 mM EDTA, 0.05% bromophenol blue and 0.05% xylene cyanol) and subjected to 12% PAGE for 13 h and stained with ethidium bromide. The clones with a distinct profile were selected and designated as isolates N (Gold Spur) and Palampur (Red Chief).

RNA extraction RT-PCR, cloning and sequencing

For complete genome characterisation, total RNA isolation, cDNA synthesis and PCR were carried out as described (Dhir et al. 2015). Briefly, powdered leaf tissue was homogenized using tissue tearer in an extraction buffer (4 M Guanidium isothiocyanate, 0.1 M Tris base, 0.1 M sodium acetate, 0.5% PVP-40 and 0.1% β-ME). Supernatant was processed further by adding 0.1% N-lauryl sarcosine and incubated at 70˚C for 10 min followed by RNA isolation using RNeasy plant mini kit (Qiagen, Hilden, Germany). The primers were designed using alignments of the available sequences in the NCBI GenBank. The primer names, oligonucleotide sequences and the PCR conditions are shown in Table 1. The amplified products were cloned into a pGEM-T easy vector as per manufacturer instructions (Promega, Madison, USA). The products were transformed into JM109 cells and recombinant DNA was purified and checked using EcoRI restriction enzyme to confirm the presence of insert (Sambrook et al. 1989). Three independent clones were sequenced by the primer walk method in outer company (Bioserve, India). 5′ and 3′ RACE was carried out using 2nd Generation 5′/3′ RACE kit (Roche, USA) according to the manufacturer instructions using gene specific primers GSPI and II (Table 1). Briefly, first strand cDNA was synthesised using GSPI primer, followed by its purification. Purified cDNA was A-tailed using dATP and amplified using the Oligo-dT anchor primer and a nested GSPII primer (Table 1). The amplified products were ligated with pGEM-T easy vector (Promega, Madison, USA) and transformed in JM109 competent cells. Recombinant plasmid DNA was purified and presence of insert was verified using EcoRI restriction enzyme. Positive clones were sequenced using vector-specific T7 and SP6 primers (in house sequencing facility at CSIR-IHBT, Palampur). The full-length sequences were assembled using sequences derived from above using the program Snapgene viewer version 3.0 and submitted to NCBI with accession numbers FR694186 and LM999967 for isolates Palampur and N, respectively. The 5′-UTR sequence was analysed for functional RNA motifs and sites using RegRNA 2.0 (http://regrna2.mbc.nctu.edu.tw/index.html). RNA secondary structure prediction was done using mfold software MFold (uchttp://rothlab.ucdavis.edu/genhelp/mfold.html).

Sequence and recombination analysis

Sequence comparisons were performed using the ClustalX software to estimate the identity percentages at nucleotide and amino acid (aa) levels (Thompson et al. 1997). For the bioinformatic analysis, the obtained sequences were aligned by codons with homologous deposited nucleotide viral sequences using the Muscle software as implemented in MEGA X (Kumar et al. 2018). The data set of the full ASPV genome sequences contained the 2two Indian isolates characterised in the present study with the 25 available ASPV isolates and its distinct variant, AGCaV. Similar data sets were constructed using the same 27 ASPV isolates, in order to be comparable, for each ORF (1–5) independently. Furthermore, for the CP analysis, a preliminary evaluation of the 280 available ASPV CP sequences in the NCBI was conducted in which all the isolates with nucleotide identity more than 90% were excluded along with the isolates of the same origin and host which were grouped in the same phylogenetic clusters. The final data set constituted by 61 diverse ASPV isolates of different hosts and origin, in addition to the 2two Indian isolates of this study. Prior to the phylogenetic relationships, the recombination occurrence and location of recombination breakpoints were investigated using RDP, GENECONV, Chimaera, MaxChi, BootScan, SiScan and 3Seq methods implemented in the RDP version 4 program (Martin et al. 2015). Recombination events were considered as significant if five or more of these methods had a p < 0.05 in addition to phylogenetic evidence of recombination. Furthermore, in order to obtain a split network representing ambiguous signals, a split decomposition analysis was performed with SplitsTree4 (Huson and Bryant 2006).

Phylogenetic analysis

Sequence alignments, of the seven data sets described herein above, were used for inferring phylogenetic relationships and the construction of maximum-likelihood (ML) trees was implemented in MEGA X. The best nucleotide substitution model was calculated for each set of sequences in combination with evolutionary rate differences among sites that had the lowest BIC (Bayesian Information Criterion) scores determined by the Model Test program of MEGA X. Specifically, the General Time Reversable (GTR) model (Felsentein 1985) with a Gamma (G) distribution (with parameter α = 1.7076) of variable sites was used to compute the ML tree of ASPV full genome sequences, the GTR model with a G distribution (parameter α = 0.5330) of variable and Invariant sites (G + I) (I = 0.2392) for RdRp sequences, the GTR + G + I (parameter α = 0.6093, I = 0.2113) for CP, the GTR + G + I (with a parameter α = 0.4843, I = 0.2706) for TGB1, the Tamura-3 parameter (T92) model with G + I (parameter α = 1.1428, I = 0.2617) for TGB2, and the T92 + G + I (parameter α = 1.0771, I = 0.2939) for TGB3. For the CP analysis of the 63 isolates, the GTR + G + I model (parameter α = 0.7638, I = 0.1656) was applied. Confidence in the branch points in the phylogenetic tree was assessed by the bootstrap method with 1,000 pseudo-random replicates (Felsentein 1985).

Results

Serological and molecular identification of ASPV and SSCP analysis

For the initial detection of the virus, we tested the samples with DAS-ELISA using commercial antibodies (Bioreba, Switzerland). Two samples of cv. Red Chief were found to be positive out of the four tested, whereas no sample was tested positive from cv. Gold Spur. However, in a virus detection by RT-PCR using common detection primers (Table 1) spanning extreme 3′-end of the viral genome, two samples from cv. Red Chief and one sample from Gold Spur resulted in virus-specific amplification (Fig. 1a). High sequence variation was observed in the highly conserved 370 bp PCR amplicons (CP gene 3′-terminus; Supplementary Fig. 1) from cv. Red Chief and cv. Gold Spur which prompted us for the characterisation of complete CP genes for further analyses. The primer set designed for the amplification of full-length CP gene, was used in RT-PCR however no reaction product was obtained from sample cv. Gold spur. Therefore, we amplified a region spanning 5′ end of TGB3 and complete CP gene using forward primer from triple gene block 3 (TGB3F; Table 1) and ASPVNC reverse primer. From the sequence analyses, it was found that the CP 5′-end is highly variable for isolate N and it matched with low sequence similarity at the nucleotide level (70–86%) to other reported isolates from India as well as other world isolates. SSCP analysis showed the presence of two profiles in the band pattern (Fig. 1b). There was an absence of any sequence variant in cultivar Gold Spur while two sequence variants were seen in case of cultivar Red Chief (RC), further corroborated by sequencing different clones (data not shown). The sequence of the isolate from Gold Spur cultivar matched 70% at the nucleotide level to the isolate from cultivar Red Chief. Based on these results, the viral isolates from the plants Red Chief and Gold Spur, isolates Palampur and N, respectively, were selected for complete genome amplification, sequencing and variability analyses.

Fig. 1.

a Agarose gel electrophoresis of products amplified by RT-PCR for ASPV detection from eight samples originating from cv. Red Chief and Gold Spur. Lane M: 100 bp DNA ladder (Thermo Scientific, USA), Lanes G1–4: samples 1–4 from cv. Gold Spur and Lanes R1–4: samples from cv. Red Chief and Lanes C1–2: Positive control for ASPV (virus-tested positive apple cvs. maintained by tissue culture at CSIR-IHBT). b Gel images showing SSCP results. Partial CP: Partial coat protein gene amplified encompassing region 5′-end of CP to 3′of CP and partial genomic 5′-end comprising of viral 5′-UTR and partial replicase gene 500 bp (approx.). Dots represent sequence variants from sample Gold Spur (N) and sample Red Chief (Palampur). c Predicted secondary structure of the 5′-UTR of an Indian isolates N and Palampur and a European isolate (PB66; Germany), showing the difference of the loop size at the extreme 5′-end

Complete genome characterisation and sequence analyses

The complete nucleotide sequence of isolates Palampur (cv. Red Chief) and N (cv. Gold Spur) was determined and consisted of 9267 and 9254 nts, respectively, excluding the poly (A) tail. Isolate N has comparatively the shortest genome sequence in case of the 26 known ASPV isolates. The 5′ and 3′ UTRs were 29, 135 and 33, 126 nts for Palampur and N ASPV isolates, respectively, with 5′-UTRs having high AU-content (75%). Asian isolates (Japan, China and India) have short 5′-UTRs compared to other world isolates. We analysed the basic difference between the 5′-UTR’s using the RegRNA 2.0 tool. From the analysis, it was found that the extra 30 nt towards the 5′-end in isolates other than the Asian ones have an exon splicing enhancer (nt 1–7). The predicted secondary structures do not have much difference, apart from the big loop at the extreme 5′-end (Fig. 1c). All the isolates have a conserved hairpin structure. ORF1 encodes a 243 kDa polyprotein and has six domains: viral methyltransferase, 20G of oxygenase superfamily, peptidase, helicase, RdRp and DNA/RNA polymerase. An intergenic region of 99 nt is present between ORF1 and ORF2 for both Indian isolates, whereas an intergenic region of 89 and 99 nt is present between ORF4 and ORF5 of Palampur and N isolates, respectively. The CP gene size was similar as obtained during the SSCP analysis. Table 2 summarizes in detail the lengths either of the full-genome or for each ORF and untranslated areas of the studied isolates Palampur and N, along with their comparison identity analysis with closest neighbouring full-genome isolates. It is noteworthy, that compared to other ASPV CP available sequences both N and Palampur isolates were found to share minimum nucleotide identity with the Polish pear isolate IV-16 (KY176820) of 61.8% and 68.8%, respectively, whereas at the amino acid level, both Indian isolates shared minimum identity with the Indian apple isolate Uttrakhand (KR815876). The lower sequence identities of isolate N indicated a possible existence of a virus variant based on CP variability.

Table 2.

Presentation of ASPV full-genome, genes and UTRs, infected host and nucleotide sequence identities among closest neighboring isolates to Palampur and N

Isolate	Region	Host	Size(nt)	Palampur	N	GA2	PB66	AKS	LYC
Palampur	Genome	Apple	9267	ID	79.9	79.5	81.4	77.4	76.3
	ORF1		6552	ID	83.9	81.3	82.0	79.8	78.5
	ORF2		672	ID	76.6	78.6	92.9	77.4	75.6
	ORF3		363	ID	75.8	78.5	93.2	79.9	79.3
	ORF4		228	ID	82.9	84.6	86.6	81.6	85.1
	ORF5		1191	ID	67.3	77.9	77.1	69.2	69.2
	5′-UTR		29	ID	72.7	40.0	38.3	40.7	40.7
	3′-UTR		135	ID	74.1	73.0	69.4	68.6	68.1
N	Genome	Apple	9254	79.9	ID	80.8	79.6	80.1	76.7
	ORF1		6528	83.9	ID	85.2	83.8	79.9	79.1
	ORF2		672	76.6	ID	79.0	78.5	83.6	78.0
	ORF3		363	75.8	ID	75.0	75.4	85.4	78.0
	ORF4		228	82.9	ID	84.2	80.1	86.8	85.5
	ORF5		1197	67.3	ID	68.3	67.6	81.3	66.8
	5′-UTR		33	72.7	ID	51.7	53.3	55.9	55.9
	3′-UTR		126	74.1	ID	79.3	74.6	76.3	72.3

Numbers in bold shows maximum value obtained for each isolate genomic region comparison and in italics is the minimum value

Isolates included in the analysis: Palampur (Accession no. FR694186), N (Accession no. LM999967), GA2 (Accession no. KY702580), PB66 (Accession no. KF321966), AKS (Accession no. KY242757), LYC (Accession no. MG763895)

Recombination analysis

The changes in the phylogeny of different regions of a virus genome are often indicative of recombination events. As recombination within the available CP sequences of ASPV has been previously reported, we analysed the potential recombination events using initially all the available sequences of ASPV full genomes. The analysis gave unclear results of the lot of small events between different encoding regions (data not shown), and the full genome sequences were subsequently divided into their five ORFs and the RDP analysis was conducted for each ORF independently. No recombination events were detected for TGB genes (ORF2–4) and CP (ORF5), whereas in RdRp (ORF1) some recombination events were detected. Taking into account those events detected by five or more methods, their p value and the sequence size of the recombination events, Palampur was the only variant which showed recombination in RdRp, with isolate N as the major parent and PA66 as the minor parent (D21829, Apple; Table 3). This recombination event was also confirmed by the SplitsTree4 decomposition phylogenetic network (Fig. 2). For CP, an additional analysis was conducted using sequences of 63 diverse isolates, including N and Palampur obtained in this study. Only those events indicated by five or more programs incorporated in RDP4 were considered as significant. The analysis revealed that Palampur isolate is a recombinant variant, with a minor parent indicated as KC791787 (Apple, South Korea) and a major parent FR694922 (Rohru, Apple, India). The SplitsTree decomposition phylogenetic networks were created based on the alignment of the full-length CP sequences and the presence of potential recombination events within the ASPV population, and specifically between Palampur and FR694922, was confirmed (Supplementary Fig. 2; Table 3).

Table 3.

List of potential recombination events involving all five ASPV gene sequences of the studied Indian isolates detected by RDP, GENECONV, Chimaera, MaxChi, BootScan, SiScan and 3Seq methods using RDP4 program

Host	Gene	Recombinant isolate	Major parent (host, country)	Minor parent (host, country)	Detection methodsa	P valueb
Apple	RdRp (ORF1)	Palampur	Indian isolate N (apple)	D21829 (apple)	RGCMBS	1.2 × 10–5
Apple	CP (ORF5)	Palampur	FR694922 (apple, India)	KC791787 (apple, South Korea)	RGCMBS3	2.2 × 10–9

^aRecombination methods are presented accoringaccording to the following letters, R for RDP, G for GENECOV, C for Chimaera, M for MaxChi, B for BootScan, S for SiScan and 3 for 3Seq

^bOnly the highest P value is indicated by bold of the underlined detection method

Fig. 2.

Decomposition phylogenetic networks constructed using an alignment of 25 full-length ASPV RdRp sequences in addition with the isolates reported in the present study. The network was constructed using SplitsTree 4.0

Phylogenetic relationships among variable ASPV isolates

Twenty-five available full-genome isolates of ASPV in addition to the two determined Indian isolates were included in order to construct a ML tree in MEGA X using the most likely substitution model GTR + G (Fig. 3). The isolates were grouped into two clusters, without any relation to their geographical origin. However, on the basis the hosts they infect, the isolates originated from apple were grouped separately with those from pear with minor exceptions. Cluster B grouped isolates into two sub-clusters, all coming from apple with only the SG isolate from pear grouped in the small sub-cluster, whereas Palampur grouped with other apple isolates in the second sub-cluster, closely related to AKS and LYC isolates. Cluster A was divided into three groups of apple and pear isolates. The N isolate was occupied in a single clade closely related with the AGCaV, PB66 and GA2 apple isolates and the other isolates were a mix of apple and pear isolates but grouped in separated clades. Phylogenetic analyses were also performed for each coding region (RdRp, TGB1-3, CP) separately and revealed that only in case of RdRp Palampur and N isolates clustered together (Supplementary Fig. 3) thereby confirming the occurrence of recombination in ORF1.

Fig. 3.

ML phylogenetic tree conducted in MEGA X using full-genome ASPV sequences. The analysis involved 25 ASPV characterized full-genome isolates and the 2two Indian identified herein, all reported with their accession numbers, isolate name, host type and country of origin. The isolates from this study are indicated by a red square. Branches with < 80% bootstraps were collapsed. The scale bar represents a genetic distance of 0.5

The analysis conducted to infer the phylogenetic relationships among ASPV isolates using complete CP sequences was based on 63 diverse virus isolates. The ML tree was constructed in MEGA X (Fig. 4), using the most likely substitution model GTR + G + I. Two main clusters (consisted by smaller sub-clusters) were formed in which the isolates were not grouped according to their geographical origin, whereas based on the host they infect, the grouping was not very clear. In cluster B, the 80% of isolates originated from apple and it was divided into 3three sub-groups, the upper consisted by 5five apple isolates (the GA4 isolate in a separate clade) and 1one pear, the middle with 17 apple isolates and 3three from pear and the lower sub-group consisted of 3three separate clades of 6six apples and 3three pears which showed low identity rates compared to all the other isolates. The Indian isolate N was clustered within this sub-group, and closely associated with apple isolate AKS from China, which is consistent with the sequence identity analysis. All the available Indian apple isolates were grouped in cluster B with Palampur (middle sub-group), and separately from isolate N. The cluster A contained an equal number of pear isolates and apple isolates. Eight apple isolates from different countries grouped together with two pear isolates, whereas the upper group was a mix of apple and pear isolates.

Fig. 4.

ML phylogenetic tree conducted in MEGA X using full-length ASPV CP sequences. The analysis involved 63 ASPV isolates, with a total of 1335 positions in the final dataset. All isolates are reported with their accession numbers, isolate name, host type and country of origin. The isolates from this study are indicated by a red square. Branches with < 80% bootstraps were collapsed. The scale bars represent a genetic distance of 0.1

Discussion

ASPV has been reported as the causal agent of several serious pome fruit diseases (Koganezawa and Yanase 1990; Mathioudakis et al. 2009; Wu et al. 2010). Symptoms like necrotic/chlorotic spots, leaf curling and vein yellowing were observed in an apple field in CSIR-IHBT and were found to be positive for ASPV by RT-PCR, but not by ELISA using commercial antiserum. The difference in the immunological reactivity is related to the CP amino acid sequences that are predicted to alter the CP structure (Chooi et al. 2013; Ma et al. 2019). The clones of sequence variants, identified by SSCP, were sequenced and showed a highly variable sequence of the partial CP gene. Moreover, the full-length genome sequence of isolates from cv. Red Chief (Palampur isolate) and Gold Spur (N isolate) were established.

The genomes have similar organisation in encoding five ORFs, with isolate N having the smallest genome size of 9254 nts among all the known ASPV full-genome sequences. Isolate Palampur has a genome size of 9267 nts which was similar to the Chinese pear isolates KL1 and KL9 (9265 nt) (Liu et al. 2012). The CP sequence length is highly variable with a range of 1125–1245 nt for most ASPV isolates, with Indian isolates possessing similar lengths (1191 and 1197 nt for Palampur and N isolates, respectively), whereas KR815875 isolate from India has the longest length of 1248 nt and KY176809 from Poland has the shortest CP isolate of 1097 nt. The 5′-UTR was 29 and 33 nts long for isolate Palampur and N, respectively, and it was observed that both Indian ASPV isolates with some Chinese isolates (YLX, YL, KL1, KL9) have shorter 5′-UTR (28–34 nt) compared to the other isolates which usually have a longer 5′-UTR sequence (46–60 nt). The 5′-UTR of both isolates have high AU content and formed a conserved hairpin structure which is known in the case of Foveavirus (James et al. 2013) and might be responsible for recombination events. High AU content is suggested to form an easy secondary structure which is favourable for ribosomal scanning (Basso et al. 1994). The Indian isolates shared 79% nucleotide identity at full-genome base and only 70% nucleotide identity in the CP gene, thereby suggesting the possible existence of two diverse isolates. This hypothesis is further supported by the negative reactivity of N isolate to an antiserum developed against the CP gene sequence of the Indian isolate Palampur (Dhir et al. 2021).

When conducting comparable phylogenetic analysis based on the 25 already available complete genome of the virus with the 2two isolates reported here, in addition with the usage of each ORF independently, the isolates grouped without any relation to their geographical origin. In the case of the full-genome tree, the isolates could be differentiated based on the host they infect into clusters or sub-clusters in case of mix clusters, with the exception of pear isolates SG and LYC which fall in an apple group. The analysis based on each of the five coding regions did not show a clear demarcation of formed host-specific clusters. In the cases of a full genome, TGB 1–3 and CP trees, the Palampur and N isolates clustered in different groups confirming the existence of two diverse ASPV isolates in India. Interestingly, in RdRp tree both isolates were closely related and grouped together indicating the occurrence of recombination events.

When the phylogenetic evolutionary relationships were studied using a larger dataset of 63 diverse CP gene sequences, the ASPV isolates again did not show any type of relation associated with their geographical origin, and according to the host they infect, the pattern was not very clear. Similarly, to the above CP tree produced by fewer ASPV isolates, the Palampur and N isolates localised different branches. This illustrates the close relationship of both Indian isolates with the isolates that showed a high percentage of identities described earlier. Moreover, here it is reported that isolates cannot be phylogenetically discriminated based on the host they infect. Although in each defined cluster the majority of isolates were originated from one host type (apple or pear), there were also present isolates of a different host. This finding further supports the recent study on ASPV variability among Polish isolates (Komorowska et al. 2019) in which a mixed grouping of apple and pear ASPV isolates were also present. The increasing national and international trade of propagating materials and using of top grafting for introduction of new apple or pear varieties, may have played a major role in the diversification of ASPV population and occurrence of mixed infections. It is very likely that, due to the commercial exchange of ASPV infected plant material, similar genetic variants can be found in different geographic regions.

These studies enriched our knowledge on if and how the genome variability can influence the shape of the virus population structure and epidemiology in contrast with previous reports where ASPV isolates were purely grouped phylogenetically according to the host either in apple or pear groups (Komorowska et al. 2011; Liu et al. 2012; Ma et al. 2019; Wu et al. 2010). This is probably due to the limited number of available ASPV isolates during that period. In all the phylogenetic analyses (full-genome, TGBs, CP), the diverse nature of isolate N from Palampur and previously reported Indian apple isolates is clearly evident, indicating the existence of four diverse ASPV apple isolates in India (N, Palampur, KR815876, KM873720). Moreover, in all the analyses, the diverse isolate N appears closely related or occupied within the same group as the diverse distinct variant of ASPV, the AGCaV. In addition, it is noteworthy that, in the analyses, no ASPV isolates from quince were included, as the available genetic information is only from the RdRp gene (Mathioudakis et al. 2009, 2010, 2011). Since no vector is known for ASPV, the high identity rates occurring between the isolates of the same hosts from different countries indicates the role of exchanging infected plant material during rafting in virus dissemination.

In ASPV, like in other Betaflexiviridae members such as apple stem grooving virus and apple chlorotic leaf spot virus, a virus isolate in an infected plant can also be the result of different variant populations (quasispecies) of the virus due to mutations or natural deletions (Candresse et al. 1995; Komorowska et al. 2019; Magome et al. 1997; Yoshikawa et al. 2001). Another reason can be the exchange of different virus variants during the grafting when both plant materials are infected, which probably can support the hypothesis of the close relationship of apple isolates and their phylogenetic grouping supported in previous studies and herein. Recombination also plays an important role in virus evolution by the creation of new variants which can be adopted better to new hosts and environments. The recombination analyses based on comparable sets of 27 ASPV isolates of each ORF independently and of 63 diverse CP sequences revealed that Palampur isolate is a recombinant variant, showing 2two recombination events which were in accordance with the ML trees and decomposition phylogenetic networks. The first event occurred in RdRp indicated Palampur as a recombinant variant of the reported here isolate N and PA66 apple isolate consisting of a novel finding of recombination events within the RdRp region of ASPV, enriching our knowledge as there is only one report available in pear from China (Liu et al. 2019). No recombination occurred in TGB 1–3, whereas there is only one study reporting recombination evidence in TGB1 of pear isolates (Ma et al. 2016). The second recombination event occurred in CP, indicating Palampur as a recombinant of the Indian isolate Rohru and the apple isolate KC791787 from South Korea. This finding further supports the existence of recombination events in the ASPV CP region in apple and pear hosts (Dhir et al. 2011; Komorowska et al. 2011, 2019; Ma et al. 2016; Yoon et al. 2014), indicating the important role of recombination in the ASPV divergence in the CP gene.

Interspecies recombination is possible as the intermixing of the viral population may have taken place via root grafting as apple and pear mixed plantations are common. This can explain the existence of the mix grouping of ASPV isolates due to the virus evolution to adopt in a new host (e.g., pear) but the progeny came from another host (e.g., apple). An example of this in this study can be the Indian pear isolate solan (KR815875) which is a recombination of two apple isolates from India (KM873720) and Germany (PM8; KF319056) or the pear isolate from Czech Republic (E17; KY176810) as a recombinant of two apple isolates from Poland (KY176826, KY176825). Recombination analyses clearly show the gene reassortment among the viral strains infecting two different hosts. The results reported herein indicate that recombination has an important role and local impact in India which has led to the generation of two highly divergent isolates, of which N shows higher divergency and has remained undetected by common detection procedures like ELISA by commercial ASPV antibodies. There is a strong need for the evaluation of the detection procedures used in virus-certification programmes for those novel viruses and viral strains which remain undetected as well as for developing strain specific antiserum for ASPV detection.

In summary, our study revealed that selection and recombination have played an important role in ASPV evolution by increasing genetic diversity and affecting the emergence of new viral variants. Our findings indicate a necessity of constant monitoring of ASPV occurrence and developing new diagnostics tools taking into accout the serological and genetic variability of virus population.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

SD and VH conceived and designed the study. SD collected the samples and performed the experiments. MMM and BHJ performed data analysis. SD, MMM, BHJ and VH wrote the manuscript.

Funding

Financial support from Department of Biotechnology, Govt. of India (BT/PR/11001PBD/16/803/2008) is duly acknowledged.

Declarations

Conflict of interest

Authors declare no conflict of interest.

Research involving human and animal participants

This article does not contain any studies with human participants or animals performed by any of the authors.