To analyze the major viral diseases affecting P. edulis and their incidence by cultivation year, we conducted surveys in Jeonbuk State, a major P. edulis-producing region in the Republic of Korea. Ten species of sub-tropical fruit trees are cultivated in the Republic of Korea across an area of 1,713,000 m², of which P. edulis accounts for the second largest area, after mangoes, of 348,000 m². Jeonbuk State is the second largest grower of P. edulis in the Republic of Korea, with the cultivation area of 91,400 m² being the largest of the 133,000 m² total cultivation area for sub-tropical fruit trees in the region (Rural Development Administration, 2023). Although it is not the optimal region for cultivating P. edulis, Jeonbuk State has proactively introduced greenhouse cultivation techniques to enhance farm income. Consequently, within Jeonbuk State, P. edulis is mostly cultivated in Kimje, Namwon, Iksan, Wanju, and Jeongeup, and is being established as a new income-generating crop (Rural Development Administration, 2023). Therefore, for the survey, between 2019 and 2021, we arbitrarily selected ten farms with P. edulis greenhouses in Namwon, Wanju, Iksan, Kimje, and Jeongeup, where farms are densely placed.

The survey was conducted by visiting the farms at two-month intervals (February, April, June, August, October, and December). For the survey method, we arbitrarily selected three sites in each plot, and randomly selected 100 P. edulis plants from each site. We directly inspected each specimen showing signs of viral disease in the field and calculated the incidence rates. We then sampled the diseased parts (leaf, flower, fruit) to identify the pathogen. Based on the results of three years of surveys, we performed a correlation analysis of the incidence rates with the P. edulis cultivation years (age), to explore patterns of disease occurrence and contributing factors. We performed correlation analysis on the four viral species that regularly appeared during the survey period (EAPV, CMV, PaLCuGdV, and EuLCV). To ensure reliability, we excluded PLV from the analysis, which only appeared in 2021. For the correlation analysis, we used Pearson’s correlation coefficients and IBM SPSS Statistics Ver. 21.0 (IBM Co., Armonk, NY, USA).

For virus diagnosis, we used the methods of polymerase chain reaction (PCR) and reverse transcription polymerase chain reaction (RT-PCR). We collected 0.1 g of leaf tissue for nucleic acid extraction, placed the sample in a 1.5 ml Eppendorf tube, and homogenized the sample in liquid nitrogen. We then used a Viral Gene-spin^™ Viral DNA/RNA Extraction Kit (iNtRON Biotechnology, Seongnam, Republic of Korea) to extract the total DNA/RNA. The purity and concentration of the extracted nucleic acids were measured using a NanoDrop spectrophotometer (Thermo Fisher SCIENTIFIC, Seoul, Republic of Korea). The nucleic acids were then divided into small portions by pipetting and stored in a deep freezer at –80°C until subsequent experiments. Viral infection was diagnosed using the specific primers described by Hong (2017) in the identification of the four species of EuLCV, PaLCuGdV, CMV, and EAPV ( Table 1 ).

For PCR diagnosis of DNA viruses (EuLCV, PaLCuGdV), we used GoTaq^® DNA Polymerase (Promega, Seoul, Republic of Korea), 2.5mM dNTP, 5× Taq polymerase buffer, and 25mM MgCl₂. We added 10 ng of DNA extracted from leaves with suspected viral infection, and 10 pmol each of the diagnostic primer sets specific to each virus to the reaction mixture. SimpliAmp^™ Thermal Cycler (ABI, Thermo Fisher SCIENTIFIC, Seoul, Republic of Korea) was used to perform denaturation for 3 mins at 95°C, followed by 35 cycles of 20 s at 94°C, 30 s at 55°C, and 1 min at 72°C. Finally, the mixture was subjected to final extension for 10 mins at 72°C.

For one-step RT-PCR to diagnose RNA viruses (CMV, EAPV), we used SuPrimescript RT-PCR Premix (Genetbio, Daejeon, Republic of Korea). After adding 10 ng of the extracted RNA and 10 pmol of each primer set to the reaction mixture, we performed the experiment according to the manufacturer’s guidelines. The RT-PCR conditions were 30 mins at 50°C and then 10 mins of reverse transcription at 95°C, followed by 35 cycles of 30 s at 95°C. 30 s at 55°C, and 1 min at 72°C. Finally, the mixture was subjected to a final extension for 5 mins at 72°C.

To check the sizes of the PCR and RT-PCR amplification products, we used 1× TBE buffer and 1.2% agarose gel containing a DNA stain (GoodView^™, Bratislava, Slovak Republic). We used a 100 bp DNA Ladder (Invitrogen, Carlsbad, CA, USA) and performed electrophoresis for 30 mins at 100 V. Finally, we verified the infection using an image analyzer (WSE-5300 Printgraph CMOS I, Atto, Tokyo, Japan).

Two experiments were designed to analyze how viral infections affect the quality characteristics of P. edulis fruit. The first experiment focuses on the infection and quality characteristics of fruit following viral inoculation at different growth stages (hereafter referred to as the growth stage experiment). The second experiment focuses on analyzing the quality characteristics of fruit by cultivation year (age) following viral infection (hereafter referred to as the cultivation year experiment). The experiments were conducted in a triple-span, double-layer greenhouse located in Jeonbuk State Agricultural Research & Extension Services (JSARES), and the total area was 200 m². The variety of P. edulis used in the experiment were Tai-nung No.1 cuttings obtained from the JSARES horticultural department, and 90 plants were planted on April 16th, 2019.

For the growth stage groups, to infect the plants with EAPV and CMV, which can be inoculated via the sap, we planted 45 plants in total, including 15 plants for each virus (5 plants each, in triplicate) and 15 uninfected (control) plants. Here, the data on fruit quality characteristics was used again in the cultivation year experiment. For the cultivation year groups, we planted a total of 45 plants, including 15 saplings each for EAPV and CMV inoculation, and 15 saplings for infection with PaLCuGdV, which was infected mechanically via cutting.

The plants were spaced 2 m × 1 m apart, and we used a T-shaped training method. The minimum winter temperature was maintained at 10°C by warming, and ventilation was provided at high temperatures of 30°C or higher. The experiment period was from April 2019 to October 2022. Fertilizer management and other cultivation techniques were performed in accordance with the P. edulis cultivation manual published by the Rural Development Administration (Rural Development Administration, 2019).

In this study, in order to analyze the effects of viral infection on fruit quality characteristics of P. edulis, we investigated EAPV, CMV, and PaLCuGdV. One limitation of this study is that we did not include EuLCV. EuLCV is one of the major P. edulis viruses in the Republic of Korea, and has been reported to be infected via cuttings or trialeurodes (Jeon et al., 2022). However, because we were unable to obtain specimens infected with EuLCV only, this was excluded from the experiment.

For EAPV and CMV, which can be inoculated via the sap, for virus inoculation and formation of the uninfected group, we planted saplings that had been confirmed healthy by PCR and RT-PCR. The times of virus inoculation for the growth stage experiment were early growth (May), flowering (June), and fruit enlargement (July). Meanwhile, for the experiment on annual changes in quality characteristics with viral infection, we inoculated EAPV and CMV in early growth (May). Using 600 mesh carborundum, we homogenized diseased tissue (v/w, 10/1) in 0.01 M phosphate buffer (pH 7.0), performed sap inoculation, and observed the presentation of disease symptoms. We then verified the infection by RT-PCR.

In the analysis of infection and quality characteristics with viral inoculation by growth stage, the infection rates were converted into percentages by dividing the number of confirmed infected trees by the total number of inoculated trees, and multiplying by 100. Here, because we could not guarantee infection with each virus, we calculated the infection rates after inoculating 5 trees. Next, we analyzed infection and fruit quality characteristics in P. edulis with virus inoculation by growth stage, and we analyzed the changes in fruit quality characteristics depending on viral infection status. The viral strains used for inoculation were obtained in the form of freeze-dried infected leaves from the Crop Protection Division of the National Institute of Agricultural Sciences, the Republic of Korea.

For PaLCuGdV, which is infected mechanically during cutting, we verified a single infection with PaLCuGdV using PCR and RT-PCR diagnosis before planting cuttings received from the department of horticulture at JSARES. After planting, we investigated the fruit quality characteristics by cultivation year depending on viral infection status. All experimental groups were isolated using grow tunnels with a mesh size of ≤0.4 mm to prevent contamination between viruses, and the movement of vectors was blocked. For the cultivation method, we adhered to the P. edulis cultivation manual (Rural Development Administration, 2019).

After verifying viral infection in each treatment group, we selected three plants each in the EAPV, CMV, and PaLCuGdV treatment groups and the uninfected group over 4 years, from 2019 to 2022. However, in EAPV inoculation in the flowering stage, infection was only confirmed in two individuals, so we only investigated fruit quality characteristics for these two individuals. Meanwhile, in the EAPV early growth and CMV fruit enlargement stages, the infection rates were 0%, meaning that no plants were infected. Therefore, the investigations of fruit quality characteristics were not performed in these stages, and they were not included in the analysis of the results.

In the virus inoculation by growth stage experiment, we investigated the fruit weight, Brix, titratable acidity, marketable fruit rate, and number of fruits per plant. Meanwhile, in the experiment on annual changes in fruit quality characteristics with viral infection, we investigated the fruit weight, Brix, titratable acidity, marketable fruit rate, number of fruits per plant, and the marketable yield per 10 ares. P. edulis is a sub-tropical fruit tree that can be harvested in the same year after planting, and is capable of annual flowering and fruit production when the total effective temperature is around 1,500 degree-days (Jeollanamdo Agricultural Research and Extension Services, 2015). Therefore, we cultivated individuals with confirmed viral infections and uninfected individuals in isolation within nets, and we investigated the fruit quality characteristics individually.

The marketable fruit rate was calculated, with reference to the marketable fruit criteria of the Jeollanamdo Agricultural Research and Extension Services (2015), as the percentage of total fruit that were commercially suitable, and this was converted to a marketable yield per 10 ares. For quality characteristics, during the harvesting period in each treatment group (July 20th to September 20th, each year), we analyzed the fruit weight, titratable acidity, and Brix of up to 10 fruit per tree, every day. Analysis was performed immediately after harvesting with no additional ripening period. Brix was measured by diluting 1 g of sample in 10 ml of distilled water, measuring using a digital Brix meter (Atago, PAL-1, Japan), and multiplying by the dilution ratio. Titratable acidity was measured by adding 1% phenolphthalein indicator to 10 ml of sample solution, measuring the volume of 0.1N NaOH required for the sample to reach pH 8.3, and converting to units of tartaric acid.

For the survey method, we adhered to the survey analysis standards for agricultural science and technology (Rural Development Administration, 2012). For the marketable fruit rate, we set marketable fruit criteria for P. edulis as fruit weight ≥ 60 g, smooth, purple skin, Brix of ≥16°Bx, and titratable acidity of 2.5–3.0% (Jeollanamdo Agricultural Research and Extension Services, 2015). To analyze the differences in fruit quality characteristics by growth stage in P. edulis fruit in each group, to compare annual changes in fruit quality characteristics with viral infection, and to investigate the significance of mean differences over 4 years, we used IBM SPSS Statistics Ver. 21.0 (IBM Co., Armonk, NY, USA) to perform one-way ANOVA with Duncan’s multiple range test for post-hoc analysis. Statistical significance was set at α = 0.05.

During the survey period, we observed a total of 5 viruses infecting P. edulis, and the main viral diseases were PaLCuGdV, CMV, EuLCV, and EAPV ( Table 2 ). These results are consistent with previous reports that P. edulis cultivation and production are affected by various viruses, and that these are especially important diseases for passion fruit woodiness (Fischer and Rezende, 2008).

Viral diseases lead to very serious injury that greatly reduces the yield of P. edulis (Chen et al., 2021). In the Guizhou region of China, the primary viruses detected were EAPV, PLV, and TeMV, with detection rates of 63.65%, 34.7%, and 1.59%, respectively (Ren et al., 2023). Jeon et al. (2022) investigated patterns of viral disease occurrence in P. edulis and found that PaLCuGdV had the highest infection rate (68.3%), followed by CMV (58.3%), EuLCV (48.3%), and EAPV (3.3%). The report by Jeon et al. (2022) noted that viral disease occurs persistently in the Republic of Korea, and that the patterns of viral infection are becoming more diverse, showing similar trends to the results of the present study.

The incidence of viral diseases in P. edulis is analyzed in Table 2 . However, in the field, mixed infections of viruses frequently occur. Jeon et al. (2022) reported that the mixed infection rate of P. edulis in the Republic of Korea reached 75%, and Choi (2025) stated that the mixed infection rate from 2017 to 2021 was 65.4%. Mixed infections exacerbate symptoms by allowing different viruses to act simultaneously, affecting the replication of each virus (Otsuki and Takebe, 1976). In Taiwan, mixed infection of P. edulis with EAPV, EuLCV, and PaLCuGdV resulted in more severe mosaic symptoms on leaves and fruit deformities (Chong et al., 2018). Therefore, the occurrence of mixed infections can further exacerbate economic losses in P. edulis cultivation.

Although several viruses are involved in major diseases during P. edulis cultivation, because processes such as pruning, trimming, and fertilization are usually repeated within plots for 4–5 years of cultivation, EAPV and CMV infections, which can spread via sap, showed increasing incidence with cultivation years ( Table 3 ). EAPV is one of the viruses in the genus Potyvirus, known for its wide host range infecting P. edulis fruit, and is primarily transmitted by aphids in a non-persistent manner (Wu et al., 2024). In contrast, CMV is a representative virus of the genus Cucumovirus, known for its wide host range and strong transmissibility, and is also transmitted by aphids (Chen et al., 2021). In particular, EAPV causes significant economic loss because it reduces production and marketable yield through serious deterioration of tree vitality, corking of stems and fruit, wrinkling, mottling, and mosaic symptoms on leaves, and fruit deformation (Chong et al., 2018; Figure 1 ).

During the study period, samples were collected from a plot in Iksan in 2021 showing leaf curling, yellowing, stunting, and fruit malformation, and after next-generation sequencing (NGS), RT-PCR, and pathogenicity testing, the pathogen was finally identified as Passiflora latent virus (PLV) of the genus Calavirus, marking the first report of this virus in the Republic of Korea (Choi and Ju, 2023). PLV can be transmitted via sap and diseased saplings. In addition, Cho et al. (2021) reported a case of asymptomatic PLV infection in persimmon (Diospyros kaki) in the Republic of Korea, indicating the need for further surveys to investigate its potential spread to surrounding crops.

When we analyzed the correlations between cultivation year (age) of P. edulis and the incidence rates of various viruses, we found that increasing cultivation years (i.e., older age) showed a significant positive correlation with increasing incidence rates. EAPV especially showed a high incidence rate of 30.0% in 5-year-old P. edulis, and showed the strongest correlations (r=0.47**). This was followed by EuLCV (r=0.42**), CMV (r=0.41**), and PaLCuGdV (r=0.32*; Table 3 ).

In P. edulis cultivation, viral diseases can severely impair productivity, and the useful lifespan of P. edulis, which is typically around 5 years, can be shortened by up to 1 year due to disease. For this reason, nomadic cultivation patterns have been observed for this crop in some regions of Brazil (Fischer and Rezende, 2008).

We inoculated EAPV to the sap of P. edulis at different growth stages and investigated the infection rates. During early growth, 25 days after planting, even when EAPV was inoculated to young leaves, we did not observe infection even up to 90 days after the date of inoculation. However, we observed infection rates of 40% in the flowering stage 55 days after planting, and 60% when inoculated during the fruit enlargement stage ( Table 4 ).

The time required for disease symptoms to appear decreased from 53 days at 28°C during the flowering stage to 25 days at 30°C in the fruit enlargement stage (data not shown), indicating that temperature significantly influences virus replication. EAPV, classified under the genus Potyvirus, behaves similarly to potato virus Y, where Choi et al. (2017) reported systemic infection times decreasing from 14 days at 20°C to 5.7 days at 28°C. These findings suggest that lower temperatures, such as 20°C during early growth, reduce replication rates and hinder the virus in overcoming plant defenses.

Infection with EAPV during the flowering stage led to worse fruit quality in P. edulis, with lower Brix and higher titratable acidity, significantly reducing marketable fruit rates compared to healthy fruit. In contrast, infections during the fruit enlargement stage resulted in similar fruit quality metrics to uninfected plants, but marketable fruit rates still decreased significantly (p=0.041; Table 4 ). Additionally, early growth stage inoculation led to poor marketable fruit rates, likely due to stress from physical inoculation affecting tree vigor and rooting.

We inoculated CMV into the sap of P. edulis at various growth stages. Infection rates were 60% during early growth (30 days after planting) and 80% during flowering (60 days after planting), with no infection during the fruit enlargement stage (105 days after planting) ( Table 5 ). The time required for symptoms to appear was 27 days at 20°C during early growth and 30 days at 25°C during flowering (data not shown).

Ko et al. (2009) inoculated zucchini yellow mosaic virus (ZYMV) to cucumbers at different growth stages, and reported that, when inoculated at ratios of 25%, 50%, and 100% during semi-forced cultivation in April to June, when the aphid density is high and temperatures are rising, the initial incidence rates started at 6.9%, 23.7%, and 48.7%, respectively, and continually increased until the later stages of growth, when all eventually showed a 100% incidence rate. Meanwhile, it has been reported that a spinach strain of CMV did not proliferate or cause symptoms in Nicotiana spp. at 30°C or higher, and that the count of aphids, which act as vectors, increases at certain temperatures in the range 20–25°C (Yin et al., 2024). When CMV was inoculated to P. edulis at different growth stages, the results were similar. Our findings align with these, showing high CMV infection rates in P. edulis during early growth and flowering stages, with no infection during fruit enlargement when temperatures were high ( Table 5 ).

When we analyzed fruit quality characteristics, fruit that had been infected in the early growth stage showed significantly lower Brix (0.8°Bx) and titratable acidity (0.3%) compared to uninfected fruit, and a lower mean number of fruits per plant (8.3). In the flowering stage, the mean number of fruits per plant (2.6) was lower than uninfected fruit, but the fruit weight was heavy (3.4 g), and there were no significant differences in other quality characteristics ( Table 5 ).

Compared to healthy P. edulis, virally infected plants showed a decline in marketability with increasing cultivation year (age), with significant decreases in Brix and number of fruits per plant. When infected with EAPV, older age was associated with a mean decrease of 36.3 fruits per plant compared to the uninfected group, the fruit weight was lighter, and the Brix also decreased, resulting in a decline of 26.8%p in the marketable fruit rate ( Table 6 ). Notably, the external appearance showed symptoms such as fruit corking and malformation, resulting in the loss of commercial value ( Figure 1C ). Iwai et al. (2006) reported that EAPV symptoms include vein necrosis and rugosity of the upper trifoliate leaves, misshapen, woody and pitted fruit, and stunted vegetative growth. We also found that P. edulis infected with EAPV showed more severely stunted growth with increasing cultivation year, suggesting that this could have negative effects on growth and development. According to previous studies, infection by TeMV, a member of the genus Potyvirus like EAPV, resulted in a 21.6% reduction in total fat content in P. edulis fruit, and levels of total acid and vitamin C were also lower compared to healthy fruit. In contrast, the total phenolic content, a secondary metabolite involved in host-pathogen interactions, increased by 19.1%. These findings suggest that viral infection can selectively modulate the chemical composition of plants, supporting the results of this study (Chen et al., 2018).

P. edulis infected with CMV exhibited a significant decrease in fruit weight with increasing cultivation years, resulting in a 19.1%p reduction in marketable fruit rate compared to uninfected plants ( Table 6 ). The infection also led to more pronounced symptoms, such as clear spots, yellowing, and mosaic patterns on leaves, along with yellowing and pigmentation defects on the fruit, diminishing commercial value ( Figure 2 ). These findings align with previous studies, which reported CMV-infected P. edulis showing mosaic patterns, necrosis, leaf curl, decreased chlorophyll content, and downregulation of genes related to photosynthesis and chloroplasts (Yeturu et al., 2018; Chen et al., 2021; Zhao, 2021).

P. edulis infected with PaLCuGdV did not show significant differences with the uninfected group in terms of external appearance or fruit quality characteristics, and the mean fruit weight was heavier, at 2.9 g. However, compared to the uninfected group, the mean number of fruits per plant was lower by 20.3, and the titratable acidity was higher, resulting in a mean decrease of 32.7% in marketable yield ( Table 6 ).

PaLCuGdV is in the genus Begomovirus, and there have been relatively few studies of Begomovirus infection in P. edulis compared to Potyvirus spp. This implies that the commercial cultivation of P. edulis is less affected currently by Begomovirus spp (Wu et al., 2024). Tree vigor was managed by identical pruning and trimming processes, in accordance with a standard cultivation manual (Rural Development Administration, 2019), in all treatment groups, but we still observed annual differences in the numbers of P. edulis fruits. Nevertheless, the years with the highest and lowest yields were the same in all treatment groups, and the damage due to viral infection in P. edulis was most severe for EAPV, followed by CMV, then PaLCuGdV.

The demand for P. edulis is increasing due to its unique flavor, prompting expanded cultivation. However, the plant is vulnerable to viral diseases that can significantly reduce yield and, in severe cases, cause complete cultivation failure. In the Republic of Korea, P. edulis was introduced from Taiwan, where cultivation areas decreased dramatically from 1,392 ha in 1982 to less than 100 ha in 1990 due to viral diseases (Chen, 1991; Wu et al., 2024). The introduction brought previously unreported viruses like EAPV, EuLCV, and PaLCuGdV. Healthy, uninfected P. edulis demonstrates excellent quality, stable fruit numbers, and a high marketable fruit rate over 80%, with sapling replanting typically occurring at 4–5-year intervals. Proper prevention and management of viral diseases may allow cultivation beyond 5 years.