Introduction

Members of genus Cuscuta (dodders), family Convolvulaceae (previously Cuscutaceae) represent approximately 150–200 species of parasitic flowering plants with global distribution. They are holoparasites, e.g., their lifecycle is fully dependent on the presence of host plant from which, through haustoria, establishing a xylem and phloem connection they acquire mineral elements, water, and organic compounds (McNeal et al., 2007). The genus is represented by 10 species in the Bulgarian flora: Cuscuta approximata Bab., C. campestris Yunck., C. cesatiana Bertol., C. epilinum Weihe, C. epithymum (L.) L., C. europaea L., C. lupuliformis Krock., C. monogyna Vahl, C. planiflora Ten. and C. trifolii Bab. (Assyov et al., 2012). Reports on the distribution and host range are fragmented, incomplete, and focused on agricultural systems. Dikova (2006) reported infestation of sugarbeet (Beta vulgaris L.) by C. campestris around Svishtov. Two species, C. campestris (sub. C. arvensis Beyr. ex Engelm.) and C. epithymum were registered as alfalfa (Medicago sativa L.) parasites in the region of Rouse (Zhekova et al., 2014). An unidentified dodder species (Dimitrova, 2011) and C. epithymum (Dimitrova, 2004) were also found on alfalfa around Pleven. Although other Cuscuta species could be regarded as an economically significant threat (Parker, 2012) and all dodder species are considered Federal noxious weeds (United States Department of Agriculture, https://www.plants.usda.gov), the greatest concerns globally are related to C. campestris. The species is introduced as invasive species in Bulgaria, with distribution in all floristic regions and within a wide range of elevations (0–1800 m.a.s.l.), being parasitic on a wide variety of crop plants and wild species (Petrova et al., 2013). However, most of the studies were morphologically based and lacked molecular determination of the species.

The main threat, posed by Cuscuta spp. is its potency to negatively affect plant growth and development, thus leading to significant crop plant yield losses (Parker, 2012). The impact of parasitic plants of genus Cuscuta on their host plants is regarded as direct growth and development inhibition due to exhaustion of organic carbon and nitrogen (Hibberd & Dieter Jeschke, 2001). Parasitic flowering plants have a higher negative osmotic potential of cell sap that allows them to uptake organic nutrients from the host plant (Hibberd & Dieter Jeschke, 2001). As Cuscuta has no roots and no effective photosynthesis system, most of the nutrients come from the host phloem, but their haustoria reach into the xylem too for nutrients such as calcium. This makes Cuscuta a phloem feeder (Haupt et al., 2001). Although nutrient exhaustion of the host is significant, parasitism may have a wide spectrum of other impacts, which are not so obvious. For example, it was reported that Cuscuta infection substantially inhibited the photosynthetic activity of the hosts (Shen et al., 2007; Zagorchev et al., 2020).

Besides the said direct aspects of parasite-host interactions, members of genus Cuscuta were also established as vectors of different pathogens, including plant viruses. Examples for this are Cuscuta campestris, found to be a potential vector of Tobacco Rattle Virus (TRV) in Bulgaria (Dikova, 2006), while other authors (Mikona & Jelkmann, 2010) reported transfer of Grapevine leafroll-associated virus-7 (GLRaV-7). However, it was shown that two out of three species of Cuscuta were able to transfer the virus and the transfer was not equally effective with different host species. The passage of viral particles from the host to the parasite and vice versa through the haustorium connection was established (Birschwilks et al., 2006; Haupt et al., 2001), thus raising concerns that dodders might play a substantial role in plant virus spread. However, few studies were published about the global distribution of plant viruses among Cuscuta species and their role as potential vectors. The aim of the present study was to monitor the distribution of four major plant viruses among Cuscuta spp. in Bulgaria and to further study the possible transfer of these viruses into host plant species.

Material and methods

Sampling

Samples from 36 populations of Cuscuta spp. were collected in the summer (June–September) of 2020. All populations with respective host plants and locations are enlisted in Table 1. Plant material was transferred to the lab. Cuscuta and host species were determined morphologically and voucher herbarium material was prepared to be deposited in Herbarium SO (Sofia University St Kliment Ohridski). Vegetative material was frozen in liquid nitrogen and stored at -80 °C until further analyses, while seeds (if available) were air-dried and stored at 4 °C with silica gel.

Table 1 List of investigated Cuscuta spp. with GenBank accession number, GPS coordinated of the locations, host plants and distribution of infections with Tobacco Mosaic Virus (TMV), Alfalfa Mosaic Virus (AMV), Tomato Yellow Leaf Curl Virus (TYLCV) and Cucumber Mosaic Virus (CMV)
Full size table

Molecular taxonomy

Sequencing analysis of ribosomal DNA (rDNA) region was used for taxonomy status determination. Total DNA from different Cuscuta species’ vegetative material was extracted with GeneMATRIX Plant & Fungi DNA Purification Kit (EURX®) according to the manufacturer’s instructions. The concentration and purity of the obtained DNA were estimated spectrophotometrically by 260/280 nm ratio absorbance measurements. PCR amplification was performed using primers ITSL (forward; ITSL (5’-TCGTAACAAGGTTTCCGT AGGTG-3’ (Hsiao et al., 1998) and ITS4 (reverse; ITS4 (5’-TCCTCCGCTT ATTGAT ATGC-3’ (White et al., 1990) for the region of ribosomal DNA, covering partial fragment of the small subunit ribosomal DNA gene ITS1, 5.8S, ITS2 and a partial fragment of large subunit ribosomal DNA gene. PCR conditions were as follows: initial denaturation at 94 ºC for 5 min; 35 cycles (denaturation at 94 ºC for 45 s; primer annealing at 50 ºC for 45 s; extension at 72 ºC for 1 min) and final extension at 72 ºC for 10 min. PCR products were visualized on 1% agarose gel with CSL-Runsafe (Cleaver Scientific Ltd) under UV light. Direct sequencing (bidirectional) after purification of amplified products (GeneMATRIX PCR / DNA Clean-Up Purification Kit, EURX®) was performed by Sanger Sequencing Services of Eurofins Genomics. The forward and reverse sequences were assembled with ChromasPro 2.1.10 (Technelysium Pty Ltd.) to avoid sequencing errors. The MEGA X software (https://www.megasoftware.net/) was used for alignments, BLAST, and Maximum Likelihood phylogenetic tree construction, Tamura-Nei model (Tamura & Nei, 1993). All GenBank sequences of rDNA regions for Cuscuta spp. were used for the comparison. rDNA/ITS sequences were deposited in GenBank (https://www.ncbi.nlm.nih.gov/).

Virus detection by DAS-ELISA

Four commercially available DAS-ELISA-based detection kits for Tobacco Mosaic Virus (TMV), Cucumber Mosaic Virus (CMV), Alfalfa Mosaic Virus (AMV), and Tomato Yellow Leaf Curl Virus (TYLCV), provided by LOEWE® were used. Three of the studied viruses – TMV, CMV, and TYLCV were listed among the top ten economically important plant viruses (Scholthof et al., 2011). Alfalfa mosaic virus is widely spread and damaging to alfalfa, but also economically important crop plants as soybean (Malapi-Nelson et al., 2009) and potato (Xu & Nie, 2006). Experiments were performed essentially according to the producer’s manual with the exception that a single ELISA plate was partitioned for the simultaneous detection of all four viruses instead of using a single plate for a single virus. One hundred mg of plant material was ground in liquid nitrogen and resuspended in 2.2 mL of the provided sample buffer. After centrifugation at 10 000 g for 10 min, the supernatant was aliquoted into 12 wells (0.2 mL in each) to be simultaneously tested for each of the four viruses in triplicates. Positive and negative controls were used for each virus on every plate. The development of the colored signal was followed visually and confirmed spectrophotometrically (Microplate reader DR-200B, Diatek Instruments, Wuxi, China) at 405 nm for one hour. Positives were considered all samples in which all three repetitions gave absorbance at 405 nm, at least two times greater than the signal of the negative control.

Tissue printing and Western Blot detection

Seeds from Cuscuta campestris with proven CMV or TYLCV infection were scarified with concentrated sulfuric acid for 30 min, extensively washed with water, and placed in close vicinity to fully grown Arabidopsis plants Columbia ecotype, Col-0. After successful infection and development of the parasite (e.g., when at least 50 mg of C. campestris fresh mass are available), the stem was cross-sectioned and immediately manually pressed onto nitrocellulose paper (0.22 µm, Applichem) for 5 s. Four types of sections were “printed” – 1) control (non-infected) Arabidopsis; 2) C. campestris stem; 3) cross-section through the haustorial connection between Arabidopsis and C. campestris and 4) Arabidopsis approximately 1 cm above the infection site. This is a simple procedure, widely employed for the localization of enzymatic activities and different antigens in plant tissues (Varner & Ye, 1994). Membranes were shortly dried in the air and blocked for one hour, room temperature (RT), in non-fat dried milk, 2% (w/v) in Washing buffer (both provided with the ELISA kits, LOEWE®). After triple washing (Washing buffer), immunoblot detection was performed by successive incubations with primary antibody (1:50 dilution, polyclonal rabbit anti-CMV or anti-TYLCV, provided with the ELISA kits, LOEWE®) and secondary antibody (1:1000 dilution, anti-rabbit alkaline phosphatase-conjugated, Sigma Aldrich, A3687), both for one hour. Finally, specific staining was developed with SigmaFast™ BCIP/NBT tablet and afterward, non-specific protein staining was performed with Ponceau S (Merck).

For Western Blot analysis fresh plant material (leaves from either control or parasitized Arabidopsis and C. campestris) was ground in liquid nitrogen, the powder was resuspended in phosphate-buffered saline (PBS), pH 7.2, centrifuged at 10 000 g, 4 °C for 15 min, and protein concentration in supernatants was estimated with Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). An equal amount of protein (15 µg) was separated on 12.5% T SDS PAGE (Laemmli, 1970) on Cleaver Scientific VS 10 electrophoresis system, using Page Ruler™ prestained protein ladder (Thermo Fischer Scientific) as molecular weight standard. After completion of electrophoresis, proteins were transferred to nitrocellulose membrane (0.22 µm, Applichem), using Hoefer Semi-Dry Blot system in Towbin buffer (Towbin et al., 1979) for 1 h at 0.8 mA per cm2. Subsequent Western Blot analysis was performed essentially in the same way as in the tissue printing method, except for Ponceau S staining. The same samples were also used for DAS-ELISA detection of CMV according to the procedure described above.

Results

Cuscuta species identification

All PCR products gave a single band of approximately 700 bp (Fig. 1), which was sequenced. The BLAST analysis confirmed that the predominant Cuscuta species found was C. campestris – 27 out of 36 populations. It was distinct by the yellowish-orange color of the vegetative tissues and ITS identification matched the morphological identification in all cases. It was mostly found near roads (Fig. 2a) and in small private gardens, where the most frequent host plants were Polygonum aviculare, Portulaca oleraceae, and Convolvulus arvensis, as well as several crop species. Interestingly, three populations were found on sand dunes near the sea coast (Fig. 2b), where host plants were Centaurea arenaria, Peucedanum obtusifolium and Medicago marina. Another three Cuscuta species were identified, all with the reddish color of the stem. Morphological and molecular identification often did not match, due to significant difficulties in distinguishing Cuscuta species in general (Ho, 2017). Therefore, we stick to molecular identification. A total of two populations of Cuscuta europaea (Fig. 2c), four populations of Cuscuta epithymum (Fig. 2d), and two populations of Cuscuta approximata were identified.

Fig. 1
figure 1

1% agarose gel electrophoresis of amplified rDNA region. 1–5: different Cuscuta spp. samples; 6: negative control; M: Perfect™ 100–1000 bp DNA Ladder (EurX). RunSAFE staining

Full size image
Fig. 4
figure 2

Maximum Likelihood phylogenetic tree, based on rDNA region sequences

Full size image

rDNA/ITS sequences were deposited in GenBank (https://www.ncbi.nlm.nih.gov/) and accession numbers are shown in the same Table 1 together with species names. The distribution of all populations is shown in Fig. 3 on a grid map, revealing a good coverage of the country’s territory. A Maximum Likelihood phylogenetic tree (Fig. 4) was constructed, revealing the grouping of C. epithymum, C. approximata, and C. europaea into a distinct cluster, differing from C. campestris. Within C. campestris, substantial heterogeneity was identified.

Fig. 2
figure 3

Different Cuscuta species – Cuscuta campestris (a and b), Cuscuta europaea (c) and Cuscuta epithymum (d)

Full size image
Fig. 3
figure 4

An UTM grid map of Bulgaria showing the location of investigated species mentioned in the text (Scale bar 1: 1 500 000) – Cuscuta campestris (●), C. epithymum (○), C. europaea (◊), Cuscuta approximata (♦)

Full size image

Occurrence of the viral infections

Two of the studied viruses – TYLCV and CMV, were found among the studied populations (Table 1) using DAS-ELISA detection. A single population of C. epithymum (Maglizh) was found double-positive for TYLCV and CMV. All other infected populations belonged to C. campestris. Eight (30%) were positive for CMV, 5 (19%) were positive for TYLCV and 3 (11%) were positive for both viruses (Table 1).

Tissue printing of Arabidopsis, infected with C. campestris var. Telish in laboratory conditions (Fig. 5a) showed characteristic granular staining of CMV viral particles in C. campestris stem (Fig. 5b), Arabidopsis stem above the Cuscuta infection (Fig. 5c), and in the section of haustorial connection (Fig. 5d). Notably, the staining was concentrated in the vascular elements (Fig. 3c). No staining was observed in control Arabidopsis plants (Fig. 5e).

Fig. 5
figure 5

Immunolocalization of CMV on tissue prints of C. campestris infected Arabidopsis (a). Arrows indicate localization of immunostaining in cross sections of C. campestris stem (b), infected Arabidopsis stem (c), site of infection (d) and control Arabidopsis plants (e)

Full size image

Both the lab-grown Cuscuta (A405 = 0.403 ± 0.09, Mean ± SD) and the wild population (A405 = 0.936 ± 0.11, Mean ± SD) also tested positive for CMV on DAS-ELISA, but the infected Arabidopsis did not (A405 = 0.276 ± 0.02, Mean ± SD) (Fig. 6a, the same picture is exemplary for the design of DAS-ELISA experiments, described in the Material and Methods section). The absorbance of positive control, negative control, and blank were 2.484, 0.200, and 0.148, respectively. Western blot analysis (Fig. 6b) revealed a single band of around 52 kDa, detected in C. campestris and Cuscuta-infected Arabidopsis, but not in the control Arabidopsis. This band matched the molecular weight of the dimer of the CMV major coat protein (Gildow et al., 2008). However, the monomeric form (at around 26 kDa) was not detected. The same experiments were performed for TYLCV with seeds from C. campestris var. Dolni Bogrov, but the control Arabidopsis tested positive (either non-specific staining or internal infection with TYLCV), so we do not report these results.

Fig. 6
figure 6

DAS-ELISA of AMV, TMV, TYLCV and CMV and Western blot detection of CMV in Cuscuta campestris and Arabidopsis. On the left panel (a): + (positive control);—(negative control); b (blank); 1 – 4 (Cuscuta spp. population, negative for the studied viruses); 5 (wild C. campestris—Telish); 6 (Cuscuta-infected Arabidopsis); 7 (lab-grown C. campestris – Telish); AMV – alfalfa mosaic virus; TMV – tobacco mosaic virus; TYLCV – tomato yellow leaf curl virus; CMV – cucumber mosaic virus; On the right panel (b): 1 (protein ladder); 2 (control Arabidopsis); 3 (Cuscuta-infected Arabidopsis); 4 (Cuscuta). The arrow on the right indicates immuno-stained 52 kDa protein band

Full size image

Discussion

The present study aimed to assess the occurrence of four of the most damaging and agriculturally important plant viruses (Scholthof et al., 2011) among the parasitic plant genus Cuscuta in Bulgaria. However, it is also one of the first efforts to assess the distribution and host range of these parasitic plants in the country. The predominance of C. campestris, which accounted for 75% of the found populations is in agreement with previous reports about its worldwide distribution and status of global weed (Parker, 2012), as well as wide host species range (Baráth, 2021). Although mostly found in human-affected areas such as gardens and along roads, where common hosts include common weeds such as P. aviculare, C. arvensis, both reported as preferred host species (Baráth, 2021) and P. oleraceae, it was also found parasitizing on Balkan endemic plants (Petrova & Vladimirov, 2010) such as Achillea clypeolata, as well as in sensitive habitats like sand dunes (Table 1). The genetic heterogeneity among the studied populations is substantial and not geographically consistent as for example the populations from Sofia were grouped into different clusters. Similarly, the occurrence of CMV and TYLCV was not restricted to genetically close populations. Unlike C. campestris, other species of the genus, considered native, appeared to have a much narrower distribution. However, this is somehow contradictory to similar studies in another European country – Hungary, where C. approximata was considered rare, but C. europea and C. epithymum were also widely distributed with equal or broader host species range (Barath & Csiky, 2012).

The four plant viruses tested for their presence among the Cuscuta spp. populations were chosen based on their agricultural significance (Scholthof et al., 2011). Since now, we were not able to find reports on the occurrence of these four viruses in dodders, although these parasitic plants are known hosts and vectors of other viruses such as Potato virus Y (Birschwilks et al., 2006), Tobacco rattle virus (Dikova, 2006) and Grapevine leafroll-associated virus-7 (Mikona & Jelkmann, 2010). While TMV and AMV were not detected in any of the studied samples, TYLCV and CMV were fairly well established among the C. campestris populations. Since the number of samples of C. approximata, C. epithymum, and C. europeae is comparatively small, we would avoid being conclusive, stating they are less affected by viral diseases. However, the broader distribution of C. campestris, along with its more common occurrence in agricultural lands may contribute to the wider exposure to viral diseases. Vice versa, the occurrence of CMV and TYLCV in C. campestris is an alarming finding, suggesting this species may represent a reservoir and vector, in addition to their common insect vectors such as aphids Aphis gossypii and Myzus persicae for CMV (Ali et al., 2006) and whitefly Bemisia tabaci for TYLCV (Li et al., 2010), for the spreading of these viruses among both natural and agricultural habitats. Still, the origin of infection in C. campestris itself needs to be studied, as both virus acquisition of parasitic plants from host plants (Gal-On et al., 2009) was reported, but dodders are also hosts for aphids (Zhuang et al., 2018). It is also noteworthy, that since dodders do not possess functional leaves (Parker, 2012), visual symptoms of viral infection were not observed, as previously reported (Mikona & Jelkmann, 2010). Therefore, it is also difficult to assess whether the parasitic plant itself suffers from undesired effects of the viral infection.

While the transmission of the virus in the host-parasite direction was not studied in the present experiment, the transmission in the parasite-host direction was proven at least for the CMV. This is in agreement with previous findings for other viruses (Birschwilks et al., 2006; Mikona & Jelkmann, 2010) and further demonstrated the potential of parasitic plants as viral vectors.

Conclusions

As clearly indicated, the introduced and invasive parasitic plant species Cuscuta campestris represent a substantial reservoir for the economically important Tomato Yellow Leaf Curl Virus and Cucumber Mosaic Virus in Bulgaria. The C. campestris populations were highly genetically heterogenic and with a wide spectrum of host plants. Unlike it, the native species C. epithymum, C. europaea, and C. approximata were found to be less abundant in the studied locations and much less affected by viral infection. Furthermore, it was confirmed that C. campestris could serve as a vector for Cucumber Mosaic Virus, transmitting the infection to host plants. Thus, its agricultural and ecological effect could extend beyond its direct negative impact on the growth and development of host plants. This is a substantial finding as although there are reports of such potential in dodders, there is a lack of systemic studies on the occurrence of viruses among Cuscuta spp. populations.