The common fig (Ficus carica) is a species of deciduous tree or shrub that are native to the Mediterranean and were likely first domesticated in the Jordan Valley about 11,400 years ago, preceding cereal domestication (Kislev et al. 2006). Figs are now widely grown throughout the world, both for its fruit and as an ornamental plant. Figs are well adapted to dry areas with a Mediterranean climate with characteristic high temperature and low humidity (Stover et al. 2007). Fig production in South Africa is mainly based in the dry warm regions of the Western Cape (SAFPA 2021).

The commercial propagation of figs is through grafting or self-rooted cuttings that favour the spread of various diseases, including fig mosaic disease (FMD) (Preising et al. 2021). FMD is the most common graft transmittable disease of fig trees and is characterised by a wide variety of symptoms mainly on the leaves, in the form of yellow and chlorotic mottling, interveinal chlorosis, vein blotching, discolorations, deformations, mosaic patterns and ringspot patterns (Elbeaino et al. 2009; Preising et al. 2021). In the 1950s it was shown that the causative agent of the disease is spread by the mite, Aceria ficus (Cotte) (Flock and Wallace 1957) and in 2009, it was demonstrated that fig mosaic virus (FMV, species Emaravirus fici, genus Emaravirus) can cause FMD (Elbeaino et al. 2009). In addition to FMV, numerous viruses have been reported to be associated with FMD. In 2006 and 2007, fig leaf mottle-associated virus 1 (FLMaV1) and fig leaf mottle-associated virus 2 (FLMaV2) were detected, respectively, and proposed to be members of the family Closteroviridae (Elbeaino et al. 2006, 2007). Fig badnavirus 1 (FBV1) (Tzanetakis et al. 2010; Preising et al. 2021), fig latent virus 1 (FLV1) (Gattoni et al. 2009), fig mild mottle-associated virus (FMMaV) (Elbeaino et al. 2010), fig fleck-associated virus (FFkaV) (Elbeaino et al. 2011a), and fig cryptic virus (FCV) (Elbeaino et al. 2011b) have been detected in mixed infections in FMD affected fig trees. A divergent variant of grapevine badnavirus 1 (GBV1) was recently discovered to infect different fig species (Chirkov et al. 2022) and named fig-grapevine badnavirus 1 (fGBV1). FMD-associated viruses are only spread via vegetative propagation of infected plant material, except for FLV1 which is seed transmissible and FCV proposed to be seed transmissible (Castellano et al. 2009; Elbeaino et al. 2011b). Not all the above mentioned viruses are necessarily associated with symptoms and many infections have also been identified in asymptomatic plant material (Preising et al. 2021; Chirkov et al. 2022).

High-throughput sequencing (HTS) has become the method of choice to obtain a snapshot of viruses infecting a certain host plant in a single assay (Massart et al. 2014; Rott et al. 2017; Navarro et al. 2018; Olmos et al. 2018; Maliogka et al. 2018; Maree et al. 2018; Villamor et al. 2019; Bester et al. 2020, 2021a; Bester and Maree 2022). Not only is HTS very efficient to identify disease associated agents, it also contributes to generating more complete genome sequences (Coetzee et al. 2010; Bester et al. 2021c, d) and can also allow for variant detection or genotyping (Bester et al. 2021b).

To date, FLMaV1 is the only virus officially identified to infect fig trees in South Africa (Castellano et al. 2007). The goal of the current study was to identify viruses found in South African fig trees using HTS and virus-specific reverse transcription polymerase chain reaction (RT-PCR) assays.

In July 2021 fig leaves displaying mottling and mosaic patterns were collected from three garden trees in the Simondium district, Western Cape, South Africa for HTS analysis (Fig. 1). Total RNA was extracted from one gram of leaf tissue from each of the samples using a CTAB extraction protocol (Ruiz-García et al. 2019). The integrity and purity of the RNA was assessed using Nanodrop 2000 spectrophotometry (Thermo Scientific, Massachusetts, USA) and 1% agarose Tris–acetate-EDTA (TAE) gel electrophoresis. The three total RNA extracts were pooled, and a ribo-depleted RNA library was constructed using the Illumina TruSeq Stranded Total RNA Sample Preparation kit with Plant Ribo-Zero (Macrogen, South Korea). The RNA library was sequenced on an Illumina NovaSeq 6000 instrument (Illumina, California, USA) (2 × 100 bp). The adapter sequences were removed and the data trimmed for quality (SLIDINGWINDOW:3:20, MINLEN:20) using Trimmomatic (Bolger et al. 2014). The trimmed data were de novo assembled using SPAdes 3.13.0 with default parameters (Nurk et al. 2013). Nucleotide BLAST (BLASTn) analysis of the de novo assembled contigs against a local copy of the NCBI GenBank database identified viral contigs with high nucleotide (nt) identity to FMV (98%), FLMaV1 (87%), FLMaV2 (83%), FBV1 (99%), GBV1 (99%) and FLV1 (84%). For each identified virus, the GenBank reference accession with the highest nt identity to the assembled contig was selected. Reference sequences for the viral contigs were retrieved from GenBank and read mapping to these reference sequences were performed using CLC Genomics Workbench 11.0.1 (Qiagen), with default parameters, individually as well as simultaneously to evaluate the degree of non-target read mapping. No non-target read mapping was observed (Table 1). Due to the relatedness between FBV1 and GBV1, the un-mapped reads of the individual read mapping against each reference sequence were collected and mapped to the other reference sequence to evaluate if data for both FBV1 and GBV1 were present. The number of reads mapped, and the fraction of the reference covered for FBV1 and GBV1-remained constant independently of how the read mapping was performed (Table 1). It was concluded that based on HTS analyses both FBV1 and GBV1 were present in the HTS data. Reads spanning the breakpoint in the circular genome or the start and stop of the accession in its linear form of both badnaviruses were also identified by mapping the reads to two concatenated copies of each reference accession for each virus. This is an indication that the episomal form of both these viruses were present in the sample and that the viruses were replicating in the plant (Geering 2021).

Fig. 1
figure 1

Fig leaves displaying virus-like symptoms including mottling and mosaic patterns observed on the three fig leaf samples pooled for high-throughput sequencing analyses (a) and representative leaf symptoms collected as samples during the survey including deformed leaves (b), mosaic (c), chlorotic mottling (d), and ringspots (e) compared to an asymptomatic leaf (f)

Full size image
Table 1 Read mapping results obtained using CLC genomics workbench (default parameters) after simultaneously and individually mapping the trimmed reads to each of the selected virus accessions
Full size table

In April 2022, an RT-PCR survey was conducted for the following viruses: FMV, FLMaV1, FLMaV2, FLV1, FBV1, FFkaV, FMMaV, FCV and the genus Badnavirus. Twenty-four leaf samples were collected from fig trees in the same garden as where the original HTS sample was collected. These samples included four asymptomatic and 20 symptomatic samples from nine different fig varieties (Tiger, Greta, Brown Turkey, Cape White, Adams, Black Mission, Smyrna, Eva). The symptomatic trees had FMD-like symptoms including 13 samples with mosaic patterns, three samples with ringspot patterns, and four samples with deformed leaves (Fig. 1). Total RNA was extracted from one gram of leaf tissue from each of the samples using a CTAB extraction protocol (Ruiz-García et al. 2019) and the total RNA extracts were subjected to multiple virus-specific two-step RT-PCRs. Complementary DNA (cDNA) was synthesized from 1 µg of total RNA using 0.15 µg of random hexamer primers (Promega) and Maxima Reverse Transcriptase (Thermo Scientific) in a final reaction volume of 20 µl according to the manufacturer’s instructions. A 2 µl aliquot of cDNA was added to a 23 µl PCR reaction mixture containing 1 X KAPA Taq buffer A (Mg +) (KAPA Biosystems), 0.2 mM dNTPs (Thermo Scientific), 0.4 µM forward and reverse primer (IDT) (Table 2) and 1.25 U/µl KAPA Taq DNA polymerase (KAPA Biosystems). Cycle conditions included an initial denaturation step at 94 °C for 3 min, followed by 35 cycles of 94 °C for 30 s, 50–55 °C for 30 s (dependent on the specific primer set, Table 2), elongation at 72 °C for 10–60 s (dependent on the specific primer set, Table 2) and final elongation at 72 °C for 7 min. Amplicons were visualised on ethidium bromide-stained 1% agarose TAE gels, except for FLV1 and FFkaV which were visualized on 2% agarose TAE gels. The HTS pooled RNA sample was included in all RT-PCR assays. Representative positive amplicons were excised from the agarose gels and DNA was recovered using the Zymoclean Gel DNA Recovery Kit (Zymo Research). Amplicon DNA was bi-directional Sanger sequenced using the amplicon specific primers (Central Analytical Facility (CAF) at Stellenbosch University).

Table 2 Fig-infecting virus-specific reverse transcription polymerase chain reaction (RT-PCR) assays
Full size table

RT-PCR results indicated that FMV, FLMaV1, FLMaV2, FLV1, FBV1, and FFkaV were present to varying frequencies in the 24 fig leaf samples (Table 3). The presence of FMV, FLMaV1, FLMaV2, FLV1 and FBV1was confirmed in the HTS sample with the RT-PCRs corroborating the HTS data. FMMaV and FCV were absent in all samples according to the RT-PCRs and since the HTS sample was negative for FMMaV, FCV and FFkaV, no RT-PCR positive control was available to verify the validity of these three RT-PCR assays. FFkaV was however detected in three samples other than the HTS sample and Sanger sequencing confirmed the origin of these amplicons. The resulting FFkaV amplicons had an 89–92% sequence identity to FM200426 from Italy, indicating the presence of a potential divergent variant of FFkaV in South Africa. However more sequence information is needed as the amplicon of FFkaV only included 270 nts. FFkaV was not detected in the original sample subjected to HTS, but since FFKaV was previously detected in FMD symptomatic trees, a RT-PCR assay targeting FFKaV was included in the survey. The same rationale was followed for FMMaV and FCV, however no positive samples were identified in this study.

Table 3 Prevalence of fig-infecting viruses in Simondium, Western Cape as assessed by virus-specific reverse transcription polymerase chain reactions (RT-PCRs). A positive infection is indicated with a plus sign (+)
Full size table

The presence of a GBV1 variant was confirmed in the HTS sample and in seven additional samples (Table 3) using the universal badnavirus primer set RT-PCR (Table 2). Sanger sequencing of the resulting amplicons resulted in the GBV1 annotation. The universal badnavirus RT-PCR amplicons had an identical and highest nt identity to the divergent variant of GBV1 identified in figs (Chirkov et al. 2022) and fig badnavirus 2 (FBV2) (MW842910.1), a sequence recently added to GenBank. Based on nt pairwise comparisons the complete genomes of GBV1 (OP087317.1) and FBV2 (MW842910.1) are 99.48% identical and compared to FBV1 (KT809307.1) the nt identity is 72.71% and 72.88%, respectively. This suggest that FBV2 and the divergent fig variant of GBV1 is the same species and variant, and the names should be consolidated.

FMD is a complex disease associated with mixed virus infections, and the association of a specific virus or virus combination with specific FMD symptoms remains unclear. In the present study, the aim was to determine the prevalence of nine known fig-infecting viruses in FMD affected trees. The results of the survey showed the prevalence of FMV, FLMaV1, FLMaV2, FLV1, FFkaV, FBV1 and GBV1 in garden fig trees with varying infection rates. Mixed infections were present in 100% of the symptomatic and asymptomatic samples evaluated in this study with a minimum of three viruses present per sample (Table 3) with no apparent association with specific disease symptoms.

Based on RT-PCR results, FBV1, GBV1 and FMV were the most common viruses infecting the sampled fig trees. Whether the badnaviruses play a significant role in symptom development will need to be investigated, since no single infections of a badnavirus was identified in this study. The results from this study correlates with previous reports identifying a strong association between FMV and FMD symptoms (Elbeaino et al. 2009; Latinović et al. 2019; Preising et al. 2021). Two of the four asymptomatic samples also tested positive for FMV, suggesting that an additional trigger may be needed for symptom or disease expression or that it is a new infection not yet established in the plant. Fig trees can also display uneven symptom distribution and symptoms can be limited to only one branch. It is therefore possible that a different part of the tree may have been symptomatic.

In this study, the prevalence, and the possible association with FMD-symptoms of seven viruses were investigated. This is the first report of the presence of FMV, FLMaV2, FLV1, FFkaV, FBV1 and GBV1 infecting fig trees in South Africa expanding the data on the incidence and distribution of these fig-infecting viruses and that can aid the propagation of virus free plant material in South Africa.