Inter-species protein trafficking endows dodder (Cuscuta pentagona) with a host-specific herbicide-tolerant trait


  • Linjian Jiang,

    1. Department of Horticulture and Crop Science, Ohio Agricultural Research and Development Center, The Ohio State University, Wooster, OH, USA
    Search for more papers by this author
  • Feng Qu,

    1. Department of Plant Pathology, Ohio Agricultural Research and Development Center, The Ohio State University, Wooster, OH, USA
    Search for more papers by this author
  • Zhaohu Li,

    Corresponding author
    1. State Key Laboratory of Plant Physiology and Biochemistry, College of Agronomy and Biotechnology, China Agricultural University, Beijing, China
    • Department of Horticulture and Crop Science, Ohio Agricultural Research and Development Center, The Ohio State University, Wooster, OH, USA
    Search for more papers by this author
  • Douglas Doohan

    Corresponding author
    • Department of Horticulture and Crop Science, Ohio Agricultural Research and Development Center, The Ohio State University, Wooster, OH, USA
    Search for more papers by this author

Authors for correspondence:

Douglas Doohan

Tel: +1 330 202 3593


Zhaohu Li

Tel: +86 10 6273 3427



  • Besides photosynthates, dodder (Cuscuta spp.) acquires phloem-mobile proteins from host; however, whether this could mediate inter-species phenotype transfer was not demonstrated. Specifically, we test whether phosphinothricin acetyl transferase (PAT) that confers host plant glufosinate herbicide tolerance traffics and functions inter-specifically.
  • Dodder tendrils excised from hosts can grow in vitro for weeks or resume in vivo by parasitizing new hosts. The level of PAT in in vivo and in vitro dodder tendrils was quantified by enzyme-linked immunosorbent assay. The glufosinate sensitivity was examined by dipping the distal end of in vivo and in vitro tendrils, growing on or excised from LibertyLink® (LL; PAT-transgenic and glufosinate tolerant) and conventional (CN; glufosinate sensitive) soybean hosts, into glufosinate solutions for 5 s. After in vitro tendrils excised from LL hosts reparasitized new CN and LL hosts, the PAT level and the glufosinate sensitivity were also examined.
  • When growing on LL host, dodder tolerated glufosinate and contained PAT at a level of 0.3% of that encountered in LL soybean leaf. After PAT was largely degraded in dodders, they became glufosinate sensitive. PAT mRNA was not detected by reverse transcription PCR in dodders.
  • In conclusion, the results indicated that PAT inter-species trafficking confers dodder glufosinate tolerance.


With c. 170 different species distributed throughout the world, the obligate parasitic plant, dodder (Cuscuta spp.), reduces productivity of many crops (Lanini & Kogan, 2005; Li et al., 2008; Sandler, 2010). Dodder is devoid of roots or leaves, and relies entirely on host for water and nutrients obtained via haustoria that connect to host vascular bundles. The connection between host and dodder vascular systems is continuous (Birschwilks et al., 2006), and facilitates transport of not only water, minerals and photosynthates, but also viruses (Hiosford, 1967), proteins (Haupt et al., 2001) and mRNAs (Roney et al., 2007; David-Schwartz et al., 2008) from host to the parasite. Because plants possess hundreds of different phloem-mobile proteins and RNAs that play important roles in regulating plant development and stress responses (Turgeon & Wolf, 2009), it is expected that the development and stress tolerance of dodder could also be influenced by these host-derived mobile substances that are capable of inter-species trafficking. Indeed, Alakonya et al. (2012) recently demonstrated that host-derived small mobile RNAs can silence genes in dodder.

Dodder is difficult to control because of its long-lasting seed dormancy, wide host range, relatively short life cycle, and parasitic strategy (Dawson et al., 1994; Lanini & Kogan, 2005). Despite much anticipation for highly selective dodder control on transgenic herbicide-tolerant crops, dodder routinely survives and resumes growth after application of glufosinate and glyphosate to glufosinate- and glyphosate-tolerant crops, respectively (Guza, 2000; Nadler-Hassar & Rubin, 2003; Nadler-Hassar et al., 2009).

Glufosinate-tolerance in LibertyLink® (LL) crops is achieved by transgenically expressing phosphinothricin acetyl transferase (PAT) that detoxifies the herbicide (De Blok et al., 1987). We hypothesized that dodder, parasitized on glufosinate-tolerant hosts, withstands glufosinate by acquiring PAT from the host. This concept was bolstered by the knowledge that dodder can acquire the green fluorescent protein (GFP) with 238 amino acids from host plants (Haupt et al., 2001), and PAT is considerably smaller with only 183 amino acids (De Blok et al., 1987).

We tested this hypothesis by examining: whether dodder contains PAT enzyme when parasitizing LL crops and the dynamics of this enzyme in in vitro dodder tendrils; whether dodder growing on a LL host is tolerant to glufosinate compared with that growing on a conventional host; whether dodder loses glufosinate tolerance after PAT is degraded; whether PAT mRNA moves from host to dodder.

Materials and Methods

Dodder preparation

Conventional (CN) and LL soybean (Glycine max (L.) Merr.) seeds (Seed Consultants Inc., Washington Court House, OH, USA) along with 20 dodder (Cuscuta pentagona Engelm.) seeds were placed on the surface of water-saturated potting mix (Premier Tech., Riviere-du-Loup, QC, Canada) in a 10-cm wide 12-cm deep square pot. Seeds were then covered with a thin layer of potting mix. After emergence, soybean seedlings were thinned to two plants per pot. About four dodder seedlings in each pot succeeded in twisting around the stems of soybean seedlings and developed haustoria. One month after sowing, actively growing dodder tendrils were selected for glufosinate treatment.

PAT protein quantification

The relative concentration of PAT in dodder and soybean tissue was quantified by ELISA kit (Envirologix, Portland, ME, USA) using a diluted solution of unifoliate leaf extract from LL soybean as a standard curve. The Envirologix ELISA kit employed in these experiments uses two antibodies, ensuring high specificity for LL soybean PAT. The specificity was confirmed by testing the kit on LL canola. PAT from LL canola has the same length and shares 82% of amino acids with PAT in LL soybean (Herouet et al., 2005); however, the enzyme was not detected using this kit. Moreover, the Envirologix ELISA kit has been adopted and accepted as a reliable method to detect PAT in many previous studies (Montague et al., 2007; Schafer et al., 2008; Kim et al., 2010; Wright et al., 2010).

PAT protein extractions from dodder and soybean leaf samples were performed at room temperature using the extraction buffer provided with the ELISA kit (0.01 M phosphate buffered saline, NaCl 0.138 M, KCl 0.0027 M, TWEEN 20 0.05%, pH 7.4). The ratio of sample weight (g) to extraction buffer (ml) was 1 : 10. Plant extracts were centrifuged at 14 000 g for 5 min, supernatants‎ were transferred into new tubes and stored at −20°C until PAT quantification.

PAT content in the following samples was quantified by ELISA: in vivo dodder tendrils (= 4) from LL and CN soybeans, in vitro dodder tendrils (= 4) at 1, 2, 3, 4, 5, 6, and 7 d after excision from CN and LL hosts. When collecting dodder tendrils from LL hosts, potential PAT contamination due to physical contact with LL leaves was carefully avoided by selecting tendrils that were at least 5 cm away from LL host plants.

Glufosinate application to in vivo and in vitro dodder tendrils

Dodder tendrils attached to host plants were treated with glufosinate by immersing the distal 10-cm tip into a glufosinate-ammonium (Rely® 200; Bayer CropScience, Research Triangle Park, NC, USA) solution that included TWEEN 20 (0.1% v/v; FMC Industrial Chemicals Group, Philadelphia, PA, USA) for 5 s. Tendrils selected for treatment were c. 20-cm long from the closest haustorium and were all actively growing (= 6). Concentrations of glufosinate-ammonium evaluated were 0, 0.36, 0.72 and 3.6 g l−1. Using this method, only dodder tendrils were treated with glufosinate and glufosinate did not contact host tissues. The effect of glufosinate on treated in vivo tendrils was excised from CN and LL hosts and recorded photographically at 8 d after treatment (DAT).

Additional tendrils were excised from CN and LL hosts, and treated with glufosinate within 1 h after excision following the same procedure as in vivo dodder tendrils. The treated in vitro tendrils (= 4) were transferred to a growth chamber, at 25°C, relative humidity (r.h.) of 30%, 12 h photoperiod and light density of 30 μmol m−2 s−1. The effect of glufosinate on treated in vitro tendrils was recorded photographically every day for 19 d and photographs for 10 DAT are presented.

Persistence of glufosinate tolerance of in vivo and in vitro dodder tendrils

In vitro dodder tendrils (= 4) were treated with glufosinate at 6 d after excision from CN and LL hosts by the dipping method previously described. Excised dodder tendrils continue growing in the incubator at 25°C, r.h. of 30%, 12 h photoperiod and light density of 30 μmol m−2 s−1 for at least 20 d without parasitizing a host plant (L. Jiang, F. Qu, Z. Li, D. Doohan, unpublished data 2011). Herbicide injury was assessed daily for 10 d.

Excised dodder tendrils can also parasitize a new host and become in vivo in c. 7 d. Because of this characteristic, we were able to guide dodder tendrils previously excised from LL hosts to parasitize new CN and LL soybean seedlings. New tendrils (= 6) that developed following reattachment were then collected from CN and LL hosts and treated with glufosinate using the dipping method. Glufosinate injury was assessed daily for 10 d.

Detection of PAT mRNA

Dodder tendrils (= 8) parasitizing CN and LL soybean were collected and immediately frozen in liquid nitrogen. The leaves of LL soybean were also collected as positive controls. Total RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) following the provided protocol. Total RNA was treated with DNaseI and converted into cDNA using a first-strand cDNA synthesis kit (Invitrogen). The cDNA was subjected to PCR using HotStar Taq Plus DNA polymerase (Qiagen, Valencia, CA, USA) for 40 cycles at the annealing temperature of 46°C in 30 μl reaction volume. The primers were ‘aggacagagccacaaacacc’ and ‘tgggtaactggcctaactgg’ for the pat gene. The control primers were ‘cctagcaaaccaggggagtt’ and ‘ggcaagcagagtctttcagg’ targeting the phytoene desaturase gene in dodder. Both PAT and phytoene desaturase gene PCR products were examined on 1.7% agarose gel, purified from gel using a QIAquick gel extraction kit (Qiagen) and sent to Molecular and Cellular Imaging Center (Wooster, OH, USA) for sequencing. The detection limit of PAT mRNA by reverse transcription PCR was estimated using diluted first-strand cDNA, synthesized from a LL soybean leaf RNA sample, as templates.

Data analysis

The experiment was repeated at least twice. Comparison of daily PAT level between dodders excised from LL hosts and those excised from CN hosts was performed by t-test at the significance level of 0.05 (SAS 9.3; SAS Institute Inc., Cary, NC, USA).


Dodder tendrils excised from LL soybeans contained the transgenic glufosinate-detoxifying enzyme PAT

As PAT exclusively confers glufosinate tolerance in LL host by detoxifying glufosinate, we investigated whether dodder growing on LL soybean contains the PAT enzyme. A PAT standard curve was prepared by making a 1 : 10 dilution of LL soybean leaf extract in ELISA buffer. A linear regression (R2 = 0.98) was obtained with dilutions of 102, 103, 104, 105 and 106.

Based on this standard curve, dodder tendrils growing on a LL host contained PAT at 0.3% of that encountered in LL soybean leaf (Fig. 1), 21 times of the PAT level in dodder excised from CN host. After dodder tendrils were excised from LL hosts, the PAT level declined continuously during the period they grew in vitro, but was still higher (< 0.05) than the level found in in vitro controls during the first 5 d, and reached a similar level to controls at days 6 and 7.

Figure 1.

Phosphinothricin acetyl transferase (PAT) quantification in dodder (Cuscuta pentagona) by ELISA. The dynamics of PAT in dodder excised from conventional (CN; open bars) or LibertyLink® (LL; filled bars) hosts. Dodder tendrils were incubated in a growth chamber (25°C, photoperiod 12 h, r.h. of 30%, and light density of 30 μmol m−2 s−1) for a week and samples were taken for PAT quantification daily. Error bars are + standard error of the mean (= 4). Columns labeled with a different letter is significantly different from each other within the same day by t-test at < 0.05.

In vivo dodder on LL hosts tolerated glufosinate

In vivo dodder tendrils on CN soybean died 8 d after being dipped in a glufosinate-ammonium solution at 0.36 and 0.72 g l−1; whereas, dodder tendrils on LL soybean dipped in the same solutions were injury-free (Fig. 2a). However, glufosinate tolerance of dodder on LL soybean was limited as they were killed by glufosinate at 3.6 g l−1, equivalent to 0.84 kg ha−1, the recommended field application rate, applied in 234 l of water ha−1. Because the dipping method exclusively applied glufosinate to dodder tendrils, CN and LL soybean hosts were not in contact with the herbicide and remained injury free during the course of experiment. These results indicate that dodder acquired partial glufosinate tolerance from the LL host.

Figure 2.

Dodder (Cuscuta pentagona) acquired partial glufosinate tolerance from a LibertyLink (LL; phosphinothricin acetyl transferase (PAT)-transgenic) soybean host. (a) In vivo dodder growing on LL hosts became tolerant to glufosinate. Dodder tendrils were dipped into glufosinate-ammonium solutions for 5 s, while they were still parasitizing hosts. Treated dodder tendrils were excised from hosts 8 d after treatment for photography. (b) In vitro dodders excised from LL hosts became tolerant to glufosinate. Within 1 h after excision from conventional (CN) or LL hosts, tendrils were dipped in glufosinate–ammonium solutions for 5 s and incubated in a growth chamber (25°C, photoperiod 12 h, r.h. of 30%, and light density of 30 μmol m−2 s−1). Pictures were taken 10 d after treatment.

In vitro dodder tendrils excised from LL hosts tolerated glufosinate

Although glufosinate is considered to be a contact herbicide, Shelp et al. (1992) demonstrated that a small portion of glufosinate molecules are capable of long-distance trafficking. Therefore, a possible explanation of dodder's glufosinate tolerance while parasitizing LL hosts is that the herbicide traffics from treated dodder tendrils to the LL host where the herbicide is then detoxified. To exclude this possibility, tendrils were dipped in glufosinate solutions within 1 h after excision from CN and LL hosts. These excised tendrils were then incubated in a growth chamber.

Results similar to those of the in vivo trial were obtained: tendrils excised from LL soybeans were injury-free 10 d after dipping in glufosinate solutions of 0.36 and 0.72 g l−1, while dodder excised from CN hosts were killed (Fig. 2b). Tendrils excised from CN or LL soybeans were killed by glufosinate at 3.6 g l−1. As glufosinate applied to excised tendrils cannot move to the host, this result confirms that dodder acquired glufosinate tolerance from the LL host plants.

Dodder lost glufosinate tolerance after PAT was degraded

After 6 d of incubation in the growth chamber, the PAT level in excised dodder tendrils had dropped down to the PAT background in dodder excised from CN soybeans (Fig. 1). At this point dodder tendrils excised from CN and LL hosts were treated with glufosinate-ammonium by the dipping method following the same regime as with in vivo trials. Dodder tendrils excised from CN and from LL hosts were sensitive to all glufosinate treatments (Fig. 3a), indicating that in vitro tendrils lost glufosinate tolerance after PAT was extensively degraded.

Figure 3.

Dodder (Cuscuta pentagona) tendrils excised from LibertyLink® (LL) host became sensitive to glufosinate after phosphinothricin acetyl transferase (PAT) was extensively degraded. (a) Sensitivity of in vitro dodder tendrils from LL hosts after incubation for 6 d was similar to that of tendrils from conventional (CN) hosts. Incubation conditions were 25°C, photoperiod 12 h, r.h. of 30%, and light density of 30 μmol m−2 s−1. Pictures were taken 10 d after glufosinate treatment. (b) Dodder tendrils excised from LL host became sensitive to glufosinate after reparasitizing a CN host and remained tolerant even after reparasitizing a LL host. Pictures were taken at 10 d after glufosinate application to excised tendrils.

PAT degradation in excised dodder tendrils paralleling that described earlier was achieved under in vivo conditions by facilitating reattachment of dodder tendrils previously excised from LL soybeans, to CN soybeans. While those new dodder tendrils maintained previous PAT content after parasitizing new LL soybean hosts for 16 d, those growing on CN soybean hosts contained PAT at 0.002% of that encountered in LL soybean, which is equal to the background level in dodder that grew consistently on a CN host. These data indicated that PAT was completed degraded. As expected, these dodder tendrils no longer tolerated glufosinate (Fig. 3b). Cumulatively, results of these in vitro and in vivo PAT degradation experiments demonstrate that dodder's tolerance to glufosinate was dependent upon the presence of host PAT.

PAT mRNA was not detected in dodder

PAT mRNA was detected in LL soybean, but not detected in dodders growing on LL soybean (Fig. 4a). This method can detect the PAT cDNA at 10−4 of that encountered in the LL soybean leaf sample (Fig. 4b). Since the control primers amplified the phytoene desaturase gene using the same cDNA template of dodder samples and confirmed by DNA sequencing, this result suggested that PAT mRNA is not trafficking from LL host to dodder. This result adds additional evidence that the PAT in dodder was a result of direct PAT protein trafficking from the LL host.

Figure 4.

Detection of phosphinothricin acetyl transferase (PAT) mRNA by reverse transcription PCR. (a) Lanes 1–4, detection of PAT mRNA in PAT-transgenic soybean, nontransgenic soybean, dodder (Cuscuta pentagona) growing on PAT-transgenic soybean, and dodder growing on nontransgenic soybean, respectively. Lanes 5–6, detection of mRNA of phytoene desaturase gene in dodder growing on PAT-transgenic soybean, and dodder growing on nontransgenic soybean, respectively. (b) Lanes 1–6 are respectively 1, 10, 102, 103, 104 and 105 times dilutions of the first strand cDNA synthesized from a LibertyLink® (LL) soybean leaf RNA sample.


Glufosinate tolerance in dodder growing on LL hosts has been reported by other researchers. Guza (2000) observed that dodder growing on LL sugar beet readily survived glufosinate-ammonium foliar treatment at 0.8 kg ha−1, even when the herbicide application was repeated three times on a weekly interval. Nadler-Hassar et al. (2009) also reported that foliar application of glufosinate did not control dodder growing on LL canola host. The experimental design used in their studies did not separate death of the host from death of the parasite; therefore, attribution of dodder survival to the PAT status of the host could not be made with certainty. We addressed this problem by developing a new method to exclusively apply glufosinate to dodder tendrils. Because glufosinate is a contact herbicide, the CN host plants were not injured. The additional experiments using excised tendrils confirmed that dodder tolerance to glufosinate was entirely due to PAT acquired from the glufosinate-tolerant host plants.

In previous studies, PAT conferred tolerance to glufosinate at very low concentrations in plant tissue. For example PAT, in PAT-transgenic tobacco, constituted < 0.001% of total soluble protein, yet was sufficient to protect plants from glufosinate applied at 4 kg ha−1 (De Blok et al., 1987). Van der Hoeven et al. (1994) generated PAT-transgenic tobacco plants using a root-specific promoter that expressed even lower levels of PAT in shoot tissues than those reported by De Blok et al. (1987). These transgenic plants tolerated glufosinate applied at 1 kg ha−1. When the herbicide rate was increased to 4 kg ha−1 these plants were severely injured, but new injury-free leaves were generated. Similarly, Montague et al. (2007) reported that glufosinate at 0.5 kg ha−1 caused only ‘yellowing of leaf margins’ when applied to PAT-transgenic alfalfa within which the PAT level was 0.0002% of total soluble protein in the leaves.

These results align well with those generated in our experiments. Since 1 g of fresh LL soybean leaf tissue contains c. 17.5 μg PAT (EFSA Panel on Genetically Modified Organisms (GMO), 2011) and c. 16 mg total soluble protein in soybean leaves (Vu et al., 1982), the calculated PAT content in the leaves of LL soybeans used in these experiments was c. 0.1% of total soluble protein. Accordingly, the PAT level in dodder growing on LL hosts was equivalent to c. 0.0003% of total protein, comparable to the results Montague previously reported. Cumulatively, these results indicate that host-derived PAT is capable of conferring a modest level of glufosinate tolerance in dodder.

Our results show that a host-specific phenotype can be horizontally transferred from host to parasite as a result of inter-species protein trafficking. These results indicate that horizontal phenotype transfer between host and parasite should be thoroughly investigated as a critical aspect of host/parasite interactions since plants possess hundreds of phloem-mobile proteins that have important biological functions (Turgeon & Wolf, 2009). Moreover, such phenomena may open a new direction for plant parasite control in transgenic herbicide tolerant crops by preventing PAT inter-species trafficking.


The authors thank Tom Lanini, Seed Consultants Inc., and Bayer CropScience for providing dodder seeds, soybean seeds, and LibertyLink canola seeds, respectively. ELISA and PCR were conducted in Dr Stockinger laboratory at OARDC/OSU. This research was funded by Ohio Agriculture Research and Development Center SEEDS program and supported by the Chinese Universities Scientific Fund.