Letter | Published:

MicroRNAs from the parasitic plant Cuscuta campestris target host messenger RNAs

Nature volume 553, pages 8285 (04 January 2018) | Download Citation


Dodders (Cuscuta spp.) are obligate parasitic plants that obtain water and nutrients from the stems of host plants via specialized feeding structures called haustoria. Dodder haustoria facilitate bidirectional movement of viruses, proteins and mRNAs between host and parasite1, but the functional effects of these movements are not known. Here we show that Cuscuta campestris haustoria accumulate high levels of many novel microRNAs (miRNAs) while parasitizing Arabidopsis thaliana. Many of these miRNAs are 22 nucleotides in length. Plant miRNAs of this length are uncommon, and are associated with amplification of target silencing through secondary short interfering RNA (siRNA) production2. Several A. thaliana mRNAs are targeted by 22-nucleotide C. campestris miRNAs during parasitism, resulting in mRNA cleavage, secondary siRNA production, and decreased mRNA accumulation. Hosts with mutations in two of the loci that encode target mRNAs supported significantly higher growth of C. campestris. The same miRNAs that are expressed and active when C. campestris parasitizes A. thaliana are also expressed and active when it infects Nicotiana benthamiana. Homologues of target mRNAs from many other plant species also contain the predicted target sites for the induced C. campestris miRNAs. These data show that C. campestris miRNAs act as trans-species regulators of host-gene expression, and suggest that they may act as virulence factors during parasitism.


In host-induced gene silencing (HIGS), siRNA-producing transgenes silence targeted pathogen and parasite mRNAs in trans3,4. Plant-based HIGS is effective against fungi5, nematodes6, insects7 and the parasitic plant Cuscuta pentagona8. The ease with which HIGS can be introduced into plants suggests that they might exchange naturally occurring small RNAs with parasites. Consistent with this hypothesis, small RNAs from the plant pathogenic fungus Botrytis cinerea target host mRNAs during infection9, and HIGS targeting of dicer-like mRNAs in B. cinerea reduces pathogen virulence10. Conversely, host miRNAs are exported from cotton into the fungal pathogen Verticillium dahliae11. However, to our knowledge, no examples of naturally occurring trans-species miRNAs have been described for plant–plant interactions.

Cuscuta haustoria facilitate bidirectional movement of viruses, proteins, and mRNAs1, but the functional effects of these movements are unclear. Cuscuta is susceptible to HIGS, so we hypothesized that naturally occurring small RNAs might be exchanged across the C. campestris haustorium and affect gene expression in the recipient species. We profiled small-RNA expression in C. campestris grown on A. thaliana hosts using high-throughput small-RNA sequencing (small-RNA-seq). Two biological replicates each from three tissues were analysed: parasite stem, comprising a section of C. campestris stem above the site of haustorium formation; interface, comprising C. campestris stem with haustoria with associated A. thaliana stem tissue; and host stem, comprising sections of A. thaliana stems above the interface region, as previously described12. Small-RNA-producing loci from both organisms were identified, classified, and subjected to differential-expression analyses (Supplementary Data 1).

As expected, owing to dilution of parasite RNA with host RNA, C. campestris small-RNA loci were generally downregulated in the interface relative to the parasite stem (Fig. 1a). However, 76 C. campestris small-RNA species were significantly upregulated in the interface relative to the parasite stem (false discovery rate (FDR) ≤0.05). Of these interface-enriched species, 43 (57%) were miRNA species with canonical accumulation of discrete miRNA–miRNA* pairs (the expected processing intermediates of miRNA biogenesis) from predicted stem–loop precursors (Fig. 1b, Supplementary Data 2–4). RNA blots confirmed interface-specific expression of specific miRNAs (Fig. 1c). One of the 43 miRNAs is a member of the conserved MIR164 family; the other 42 upregulated miRNAs have low sequence similarity to known miRNA loci, and none of the mature miRNAs or miRNA* aligned perfectly with the A. thaliana genome (Supplementary Data 5). Several of the miRNA loci were detected by PCR of C. campestris genomic DNA prepared from four-day old seedlings that had never interacted with a host plant (Extended Data Fig. 1). The majority of the induced C. campestris miRNAs (26 out of 43) produced a 22-nucleotide mature miRNA. Such 22-nucleotide miRNAs occur less frequently than 21-nucleotide miRNAs in plants, and they are strongly associated with accumulation of secondary siRNA from their targets13,14. Secondary siRNAs are thought to amplify miRNA-directed gene silencing2.

Figure 1: C. campestris miRNAs induced at the haustorial interface.
Figure 1

a, Mean abundance plot of C. campestris small-RNA loci comparing interface (I) to parasite stem (PS). Significantly upregulated (Up) loci are highlighted (alternative hypothesis: true difference > 2-fold, FDR ≤ 0.05 after Benjamini–Hochberg correction for multiple testing). nt, nucleotide. b, Predicted secondary structures of induced C. campestris miRNA hairpin precursors with colour-coded small-RNA-seq coverage per nucleotide. c, RNA blots of 22-nucleotide interface-induced miRNAs. HS, host stem; CS, control stem. U6, small nuclear RNA loading control. The experiment was performed twice with similar results. Full gels are shown in Supplementary Fig. 1.

We hypothesized that the induced 22-nucleotide miRNAs would cause formation of secondary siRNA from targeted host mRNAs. Therefore, we searched small-RNA-seq data for A. thaliana mRNAs that both contained plausible miRNA-complementary sites and shared sequences with siRNAs that accumulated specifically at the interface. Six A. thaliana mRNAs were found that met both criteria: TIR1, AFB2 and AFB3, which encode partially redundant auxin receptors15; BIK1, which encodes a plasma-membrane-localized kinase required for both pathogen-induced and developmental signalling16,17; SEOR1, which encodes an abundant phloem protein that reduces photosynthate loss from the phloem after injury18,19; and HSFB4 (also known as SCZ), which encodes a predicted transcriptional repressor that is required for the formation of ground-tissue stem cells in roots20,21,22. The siRNAs produced from these mRNAs resembled other examples of secondary siRNAs in their size distributions, double-stranded accumulation, and phasing (Fig. 2a, b, Extended Data Fig. 2). TIR1, AFB2 and AFB3 are also known to be targeted by the 22-nucleotide miRNA miR393, and to produce secondary siRNAs downstream of the miR393-complementary site23. In parasitized stems, the location and phase register of the TIR1, AFB2 and AFB3 secondary siRNAs shift upstream, proximal to the sites that are complementary to the C. campestris miRNAs (Extended Data Fig. 2), implying that the C. campestris miRNAs, and not miR393, are triggering the interface-specific secondary siRNAs. The predominant 21-nucleotide phase register at several loci was shifted by +1 to +2 nucleotides relative to predictions. This is consistent with the ‘phase drift’ seen at other phased siRNA loci24,25 that cause the register to be shifted forward, and is probably due to the presence of low levels of 22-nucleotide siRNAs. Analysis of uncapped mRNA fragments showed strong evidence for miRNA-directed cleavage at all of the sites complementary to C. campestris miRNAs, specifically those from interface samples but not from control stem samples (Fig. 2, Extended Data Fig. 2). We did not find any induced miRNAs or siRNAs from the A. thaliana host that were capable of targeting these six mRNAs. We also did not find any endogenous C. campestris secondary siRNA loci corresponding to any of the induced miRNAs. Some C. campestris orthologues of TIR1, HSFB4 and BIK1 had possible, but very poorly complementary, miRNA target sites (Extended Data Fig. 3). These observations suggest that the induced C. campestris miRNAs have evolved to avoid targeting ‘self’ transcripts. We conclude that 22-nucleotide miRNAs from C. campestris act in a trans-species manner to target A. thaliana mRNAs.

Figure 2: C. campestris miRNAs cause slicing and phased siRNA production from host mRNAs.
Figure 2

a, Small-RNA-seq coverage for A. thaliana SEOR1 for host stem, interface and parasite stem. n = 2 biologically independent samples each; miRNA-complementarity is shown by RNA-ligase-mediated 5′-rapid amplification of cDNA ends (5′-RLM-RACE) and expected phase register; r.p.m., reads per million. b, Length and polarity distribution of SEOR1-mapped siRNAs from interface samples. c, Radar chart showing fraction of interface-derived siRNAs in each possible 21-nucleotide phasing register. d, 5′-RLM-RACE products from nested amplifications. ARF17, positive control. The experiment was performed once. Full gels are shown in Supplementary Fig. 1.

Accumulation of five of the six targets was significantly reduced in parasitized stems compared to control stems (Fig. 3a). The true magnitude of repression of these targets could be even greater, since many miRNAs also direct translational repression. Accumulation of A. thaliana secondary siRNAs is partially dependent on the endonuclease DCL4 (DICER-LIKE 4) and wholly dependent on RDR6 (RNA-DEPENDENT RNA POLYMERASE 6, also known as SGS2 or SDE1)2. Accumulation of an abundant secondary siRNA from TIR1 was eliminated entirely in the sgs2-1 mutant, and reduced in the dcl4-2t mutant (Fig. 3b). Thus, host DCL4 and RDR6 are required for secondary siRNA production. This implies that the C. campestris-derived miRNAs are active inside host cells and hijack the host’s own silencing machinery to produce secondary siRNAs.

Figure 3: Effects of C. campestris miRNAs and their targets.
Figure 3

a, Accumulation of A. thaliana mRNA in interface versus control stems, shown by quantitative reverse-transcriptase polymerase chain reaction (qRT–PCR). Control stems, n = 8; interface, n = 7 biologically independent samples. Box plots show the median, box edges represent the first and third quartiles, and the whiskers extend to 1.5× interquartile range. P values are displayed above the x axis; Wilcoxon rank-sum tests, unpaired, one-tailed. AT4G34270 (also known as TIP41-like protein), control. b, RNA blots from C. campestris infestations of the indicated A. thaliana genotypes. Full gels are shown in Supplementary Fig. 1. The experiment was performed twice with similar results. c, C. campestris biomass on A. thaliana hosts of the indicated genotypes. P values and plotting conventions as in a, except two-tailed tests were used; n = 11, 8, 11, 10, 14, and 7 biologically independent samples (left to right).

In repeated trials, we did not observe consistent significant differences in parasite fresh weight using dcl4-2t and sgs2-1 mutants as hosts (Extended Data Fig. 4). Thus, loss of induced secondary siRNAs is not sufficient to affect parasite growth in this assay. Similarly, there were no significant differences in biomass of C. campestris grown on scz2 or tir1-1/afb2-3-mutant hosts (Fig. 3c). Significantly less (P < 0.05) C. campestris biomass was observed using the bik1 mutant as host (Fig. 3c). However, interpretation of this result was complicated by the weak, frequently lodging stems of the bik1 mutant16. BIK1 is involved in both plant development and immunity, and its developmental functions may mask its role in the C. campestris interaction. Significantly more (P < 0.05) C. campestris biomass was observed on seor1 or afb3–4 mutants (Fig. 3c). Therefore, both SEOR1 and AFB3 function to restrict C. campestris growth on A. thaliana. This observation is consistent with the hypothesis that both SEOR1 and AFB3 are trans-species miRNA targets of biological relevance in A. thaliana.

C. campestris has a broad host range among eudicots26. Therefore, we searched for sites in eudicot orthologues of the targeted A. thaliana mRNAs that were complementary to the C. campestris miRNAs induced specifically at the interface. Probable orthologues of BIK1, SEOR1, TIR1 and HSFB4 were identified as predicted targets of interface-induced miRNAs in many eudicot species, while only one species had predicted targets in the orthologues of the negative control GAPDH (Fig. 4a, Extended Data Table 1). We conclude that the induced C. campestris miRNAs would be able to collectively target TIR1, SEOR1, HSFB4 and BIK1 orthologues in many eudicot species.

Figure 4: Conservation of host mRNA targeting by C. campestris.
Figure 4

a, Predicted targets of interface-induced C. campestris miRNA–miRNA*. NA, no orthologous genes found. b, RNA blots from interface and control stem of C. campestris-infested N. benthamiana (Nb), A. thaliana (At), and C. campestris pre-haustoria (PH). The experiment was performed once. c, 5′-RLM-RACE products for the indicated N. benthamiana cDNAs. ARF, positive control. The image was cropped to remove irrelevant lanes. Full gels are shown in Supplementary Fig. 1. The experiment was performed once. d, Complementary site and 5′-RLM-RACE results from a N. benthamiana TIR1 orthologue.

We performed additional small-RNA-seq from C. campestris on A. thaliana hosts, and from C. campestris on N. benthamiana hosts. Both sets of experiments were designed identically to the original small-RNA-seq study (two biological replicates each of host stem, interface and parasite stem samples). The interface-induced set of C. campestris miRNA loci was highly reproducible across both of the A. thaliana experiments as well as the N. benthamiana experiment (Extended Data Fig. 5). Induction of several C. campestris miRNAs during N. benthamiana parasitism was confirmed by RNA blots (Fig. 4b). Several N. benthamiana mRNAs both contained plausible target sites for C. campestris miRNAs and showed accumulation of phased, secondary siRNAs in the interface samples, including orthologues of TIR1 and BIK1 (Extended Data Fig. 6). Analysis of uncapped RNA ends provided strong evidence for miRNA-directed cleavage of one of the N. benthamiana TIR1 orthologues (Fig. 4c, d). This is direct evidence that the same C. campestris miRNAs target orthologous host mRNAs in multiple species. None of the interface-induced miRNAs we tested were detectable in C. campestris pre-haustoria from seedling tips that had coiled around dead bamboo stakes instead of a live host (Fig. 4b, Extended Data Fig. 7). This suggests that contact with a living host is a requirement for expression of these miRNAs.

These data demonstrate that C. campestris induces a large number of miRNAs at the haustorium, and that some of these miRNAs target and reduce accumulation of host mRNAs. Many of the induced miRNAs are 22 nucleotides in length, and are associated with secondary siRNA production from their host targets using the host’s secondary siRNA machinery. Several of the targets are linked to plant pathogenesis: manipulation of levels of TIR1, AFB2, and AFB3 mRNA affects bacterial pathogenesis and defence signalling27, and BIK1 is a central regulator of pathogen-induced signalling28. Perhaps the most intriguing target is SEOR1, which encodes a very abundant protein that is present in large agglomerations in phloem sieve-tube elements18. seor1 mutants show an increased loss of sugars from detached leaves19, and our data show that seor1 mutants also support increased C. campestris growth. A key function of the haustorium is to capture nutrients from the host phloem; targeting SEOR1 could be a strategy to increase sugar uptake from host phloem. Overall, these data suggest that C. campestris trans-species miRNAs might function as virulence factors to remodel host gene expression to the parasite’s advantage. Further experiments that directly disrupt the delivery or function of these miRNAs will be needed to test this hypothesis directly.


Cuscuta was initially obtained from a tomato field in California, and seed stocks were derived from self-pollination through several generations in the Westwood laboratory. The isolate was initially previously identified as Cuscuta pentagona. C. pentagona is very closely related to C. campestris, and the two are distinguished by microscopic differences in floral morphology; because of this they have often been confused29. We subsequently determined that our isolate is indeed C. campestris. A. thaliana sgs2-1 mutants30 were a gift from H. Vaucheret (INRA Versailles). A. thaliana dcl4-2t mutants (GABI_160G0531) were obtained from the Arabidopsis Biological Resource Center (Ohio State University). A. thaliana seor mutants (GABI-KAT 609F0418) were a gift from M. Knoblauch (Washington State University). The A. thaliana tir1-1/afb2- and afb3-4 mutants32 were a gift from G. Monshausen (Pennsylvania State University). The bik1 mutant16 was a gift from T. Mengiste (Purdue University). The scz2 mutant22 was a gift from R. Heidstra (Wageningen University). All A. thaliana mutants were on the Col-0 background.

Growth conditions and RNA extractions

For initial experiments (small-RNA-seq and RNA blots in Fig. 1), A. thaliana (Col-0) plants were grown in a growth room at 18–20 °C with 12 h light per day, illuminated (200 μmol m−2s−1) with metal-halide (400 W, GE multi-vapour lamp) and spot-gro (65 W, Sylvania) lamps. C. campestris seeds were scarified in concentrated sulfuric acid for 45 min, rinsed 5–6 times with distilled water and dried. The seeds were placed in potting medium at the base of four-week-old A. thaliana seedlings and allowed to germinate and attach to hosts. The C. campestris plants were allowed to grow and spread on host plants for an additional three weeks to generate a supply of uniform shoots for use in the experiment. Sections of C. campestris shoot tip (~10 cm long) were placed on the floral stems of a fresh set of A. thaliana plants. Parasite shoots coiled around the host stems and formed haustorial connections. Tissues from plants that had established C. campestris with at least two coils around healthy host stems and clear parasite growth were used in these studies. Control plants were grown under the same conditions as parasitized plants, but were not exposed to C. campestris.

For the preparation of tissue-specific small-RNA libraries, tissues were harvested after C. campestris cuttings had formed active haustorial connections to the host. This was evidenced by growth of the C. campestris shoot to a length of at least 10 cm beyond the region of host attachment (7–10 days after infection). Three tissues were harvested from the A. thalianaC. campestris associations: 2.5 cm of A. thaliana stem above the region of attachment, A. thaliana and C. campestris stems in the region of attachment (referred to as the interface), 2.5 cm of the parasite stem near the point of attachment. To remove any possible cross-contamination between A. thaliana and C. campestris, harvested regions of the parasite and host stem were taken 1 cm away from the interface region and the surface of each harvested tissue cleaned by immersion for 5 min in 70% ethanol, the ethanol was decanted and replaced, the process was repeated three times and the stems were blotted dry with a Kimwipe after the final rinse. All three sections of tissue were harvested at the same time, and material from 20 attachments was pooled for small-RNA extraction. Small RNA was extracted from ~100 mg of each tissue using the mirPremier microRNA Isolation Kit (Sigma-Aldrich) according to the manufacturer’s protocol. Small RNA was analysed using an small-RNA kit (Agilent) on a 2100 Bioanalyzer platform.

Samples used for 5′-RLM-RACE (Fig. 2d) and qRT–PCR (Fig. 3a) analyses of A. thaliana targets were prepared as described above with the following modifications: Col-0 A. thaliana hosts were cultivated in a growth room with 16-h days, 8-h nights, at ~23 °C under cool-white-fluorescent lamps. Attachment of C. campestris cuttings was promoted by illumination with far-red LED lighting for 3–5 days, and total RNA was extracted using Tri-reagent (Sigma) per the manufacturer’s suggestions, followed by a second sodium-acetate–ethanol precipitation and wash step. Samples used for RNA blots of secondary siRNA accumulation from A. thaliana mutants and replicate small-RNA-seq libraries were obtained similarly, except that the samples were derived from the primary attachments of C. campestris seedlings on the hosts instead of from cuttings. In these experiments, scarified C. campestris seedlings were first germinated on moistened paper towels for three days at ~28 °C, then placed adjacent to the host plants with their radicles submerged in a water-filled 0.125-ml tube.

C. campestris pre-haustoria (Extended Data Fig. 7) were obtained by scarifying, germinating and placing seedlings as described above, next to bamboo stakes in soil, under illumination from cool-white fluorescent lights and far-red-emitting LEDs. Seedlings coiled and produced pre-haustoria four days after being placed, and were harvested and used for total-RNA extraction (used for RNA blots in Fig. 4b) using Tri-reagent as described above. N. benthamiana was grown in a growth room with 16-h days, 8-h nights, at ~23 °C, under cool-white fluorescent lamps. Three-to-four-week-old plants served as hosts for scarified and germinated C. campestris seedlings. Attachments were promoted by three-to-six days with supplementation by far-red-emitting LEDs. Under these conditions, C. campestris attached to the petioles, and not the stems, of the N. benthamiana hosts. Interfaces and control petioles from un-parasitized hosts were collected seven-to-eight days after successful attachments, and total RNA (used for RNA blots in Fig. 4b and small-RNA-seq libraries) was recovered using Tri-reagent as described above.


The initial small-RNA-seq libraries were constructed using the Tru-Seq small-RNA kit (Illumina) per the manufacturer's protocol and sequenced on an HiSeq2500 instrument (Illumina). Subsequent small-RNA-seq libraries (replicate two using A. thaliana hosts, and the N. benthamiana experiments) used the NEBnext small-RNA library kit (New England Biolabs), following the manufacturer’s instructions. Raw small-RNA-seq reads were trimmed to remove 3′-adapters and filtered for quality and trimmed length ≥16 nucleotides using cutadapt33 version 1.9.1 with the settings “-a TGGAATTCTCGGGTGCCAAGG–discard-untrimmed -m 16–max-n = 0”. For experiments where A. thaliana was the host, trimmed reads that aligned with zero or one mismatch (using bowtie34 version 1.1.2, settings “-v 1”) to the A. thaliana plastid genome, the Cuscuta gronovii plastid genome (C. gronovii was the closest relative to C. campestris for which a completed plastid-genome assembly was publicly available), A. thaliana rRNAs, tRNAs, small nuclear RNAs (snRNAs), and small nucleolar RNAs (snoRNAs) were removed. Similarly, for experiments where N. benthamiana was the host, the reads were cleaned against the C. gronovii plastid genome, the N. tabacum plastid genome and rRNAs, and a set of tRNAs predicted from the N. benthamiana genome using tRNAscanSE.

For the original A. thaliana host data, the clean reads were aligned and analysed with reference to the combined TAIR10 A. thaliana reference genome and a preliminary version 0.1 draft genome assembly of C. campestris using ShortStack35 (version 3.8.3) with default settings. The resulting annotated small-RNA loci (Supplementary Data 1) were analysed for differential expression (interface versus parasite stem) using DESeq236, with a log2 fold threshold of 1, alternative hypothesis of ‘greaterAbs’, and alpha of 0.05. P values were adjusted for multiple testing using the Benjamini–Hochberg procedure, and loci with an adjusted P value of ≤0.05 (equivalent to an FDR of ≤0.05) were denoted upregulated in interfaces relative to parasite stem. Among the upregulated loci, those annotated by ShortStack as miRNAs deriving from the C. campestris genome which produced either a 21- or 22-nucleotide mature miRNA (Supplementary Data 2) were retained and further analysed. The predicted secondary structures and observed small-RNA-seq read coverage was visualized (Supplementary Data 3, 4) using strucVis (version 0.3; https://github.com/MikeAxtell/strucVis).

For analysis of mRNA-derived secondary siRNAs, the clean small-RNA-seq reads from the original A. thaliana experiment were aligned to the combined TAIR10 representative cDNAs from A. thaliana and our preliminary version 0.1 transcriptome assembly for C. campestris, using ShortStack35 v3.8.3, with the settings –mismatches 0,–nohp, and defining the full length of each mRNA as a locus using the option –locifile. The resulting counts of small-RNA alignments for each mRNA were used for differential-expression analysis, comparing interface to host stem, using DESeq236 as described above. A. thaliana mRNAs with significantly upregulated (FDR ≤0.05) small RNAs in the interface compared to host stem were retained for further analysis. The cDNA sequences of these loci were retrieved and used for miRNA target predictions using GSTAr (v1.0; https://github.com/MikeAxtell/GSTAr); the full set of mature miRNA and miRNA* (Supplementary Data 2) from the interface-induced C. campestris miRNA loci were used as queries.

Analysis of the second set of A. thalianaC. campestris small-RNA-seq data aligned the cleaned reads to the combined A. thaliana and C. campestris reference genomes as described above, except that the list of loci derived in the analysis of the original data (Supplementary Data 1) was used as a –locifile in the ShortStack analysis. Differential-expression analysis was then performed using DESeq2 as described above. Analysis of the N. benthamiana–C. campestris small-RNA-seq data began with a ShortStack analysis of the cleaned reads against the combined N. benthamiana (v0.4.4) genome and the preliminary assembly of the C. campestris genome, using default settings. The de novo N. benthamiana loci obtained from this run were retained. The resulting alignments were used to quantify abundance of small RNAs from the C. campestris small-RNA loci defined with the original data. The resulting read counts were then used for differential-expression analysis with DESeq2 as described above. Analysis of secondary siRNAs derived from N. benthamiana mRNAs was performed in a similar way to the A. thaliana mRNA analysis described above, except that the combined transcriptomes were from C. campestris and N. benthamiana (v0.4.4 annotations).

RNA blots

Small RNA gel blots were performed as previously described37 with modifications. For the blots shown in Fig. 1b, 1.8 μg of small RNA from each sample was separated on 15% TBE–Urea Precast gels (Bio-Rad), transblotted onto the Hybond NX membrane and cross-linked using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide38. Hybridization was carried out in 5× SSC, 2× Denhardt’s solution, 20 mM sodium phosphate (pH 7.2), 7% SDS with 100 μg ml−1 salmon testes DNA (Sigma-Aldrich). Probe labelling, hybridization and washing were performed as described37. Radioactive signals were detected using Typhoon FLA 7000 (GE Healthcare). Membranes were stripped between hybridizations by washing with 1% SDS for 15 min at 80 °C and exposed for at least 24 h to verify complete removal of probe before re-hybridization. Blots in Figs 3b and 4b were performed similarly, except that 12 μg of total RNA was used. Probe sequences are listed in Supplementary Data 6.

5′-RNA ligase-mediated rapid amplification of cDNA ends

Five micrograms total RNA was ligated to 1 μg of a 44-nucleotide RNA adaptor (Supplementary Data 6) using a 20 μl T4 RNA ligase 1 reaction (NEB) per the manufacturer’s instructions for a 1 h incubation at 37 °C. The reaction was then diluted with 68 μl water and 2 μl 0.5 M EDTA pH 8.0, and incubated at 65 °C for 15 min to inactivate the ligase. Sodium acetate pH 5.2 was added to a final concentration of 0.3 M, and the RNA was precipitated with ethanol. The precipitated and washed RNA was resuspended in 10 μl water; 3.33 μl of this sample was used as template in a reverse transcription reaction using random primers and Protoscript II reverse transcriptase (NEB) per the manufacturer’s instructions. The resulting cDNA was used as template in first round PCR using a 5′ primer matching the RNA adaptor and a 3′ gene-specific primer (Supplementary Data 6); 1 μl of the product was used as template for nested PCR with nested primers (Supplementary Data 6). Gene-specific primers for A. thaliana cDNAs were based on the representative TAIR10 transcript models, while those for N. benthamiana cDNAs were based on the v0.4.4 transcripts (Sol Genomics Network39). In Fig. 4c, N. benthamiana TIR1 is transcript ID NbS00011315g0112.1; N. benthamiana ARF is transcript ID NbS00059497g0003.1. Bands were purified from agarose gels and cloned into pCR4-TOPO (Life Tech). Inserts from individual clones were recovered by colony PCR and analysed by Sanger sequencing.

Quantitative reverse-transcription–PCR

Total RNA used for qRT–PCR was first treated with DNaseI (RNase-free; NEB) per the manufacturer’s instructions, ethanol precipitated and resuspended. The treated total RNA (2 μg) was used for cDNA synthesis using the High Capacity cDNA Synthesis Kit (Applied Biosystems) per the manufacturer’s instructions. PCR reactions used PerfeCTa SYBR Green FastMix (Quantabio) on a StepONE-Plus quantitative PCR system (Applied Biosystems) per the manufacturer’s instructions. Primers (Supplementary Data 6) were designed to span the miRNA target sites to ensure that only uncleaved mRNAs were measured. Three reference mRNAs were used: ACT2, AT1G13320 (which encodes PDF2, a subunit of PP2A), and AT4G3427040. Raw Ct values were used to calculate relative normalized expression values to each reference mRNA separately, and the final analysis used the median relative expression values between the ACT12 and AT4G34270-normalized data.

C. campestris growth assays

C. campestris seedlings were scarified, pre-germinated, and placed next to hosts in 0.125 ml water-filled tubes under cool-white fluorescent lighting supplemented with far-red-emitting LEDs (16-h day, 8-h night) at ~23 °C as described above. After a single attachment formed (four days), far-red light supplementation was removed to prevent secondary attachments. After 18 more days of growth, entire C. campestris vines were removed and weighed (Fig. 3c). Multiple additional growth trials were performed specifically on the dcl4-2t and sgs2-1 mutant hosts under varying conditions (Extended Data Fig. 4).

miRNA target predictions

To find probable orthologues for A. thaliana genes of interest, the A. thaliana protein sequences were used as queries for a BLASTP analysis of the 31 eudicot proteomes available on Phytozome 11 (https://phytozome.jgi.doe.gov/pz/portal.html#). Transcript sequences for the top 100 hits were retrieved. In some cases no hits were found in a particular species; these are shown as ‘NA’ in Fig. 4a. The miRNA query set was all mature miRNA and miRNA* from the interface-induced, C. campestris-derived 21- or 22-nucleotide miRNAs (Supplementary Data 2). Probable targets from the 31 species were identified as those having a score of up to 4.5 using targetfinder.pl v0.1 (https://github.com/MikeAxtell/TargetFinder/).

N. benthamiana orthologues of A. thaliana proteins were found based on BLASTP searches against the v0.4.4 N. benthamiana protein models at Sol Genomics Network39, and miRNA target sites predicted using targetfinder.pl as above.

Statistics and reproducibility

No statistical methods were used to predetermine sample size. The experiments were not randomized. 95% confidence intervals from Fig. 3a: 0.249 to 0.611 (BIK1), 0.267 to 0.781 (SEOR1), −0.122 to 0.649 (HSFB4), 0.385 to 0.894 (TIR1), 0.083 to 0.724 (AFB2), 0.071 to 0.678 (AFB3), −0.461 to −0.120 (AT4G34270). Note that these confidence intervals from unpaired Wilcoxon rank-sum tests are the estimators of the median of control stem minus interface for each gene. 95% confidence intervals from Fig. 3c: −0.580 to −0.200 (seor1), 0.220 to 0.400 (bik1), −0.070 to 0.150 (scz2), −0.170 to 0.060 (tir1-1/afb2-3), −0.440 to −0.180 (afb3-4). Note that these confidence intervals from unpaired Wilcoxon rank-sum tests are the estimators of the median of Col-0 minus the mutant for each comparison.

Code availability

ShortStack35 (small-RNA-seq analysis), strucVis (visualization of predicted RNA secondary structures with overlaid small-RNA-seq depths), GSTAr,pl (prediction of miRNA targets) and Shuffler.pl/targetfinder.pl (prediction of miRNA targets controlling for false discovery rate) are all freely available at https://github.com/MikeAxtell. Cutadapt33 is freely available at http://cutadapt.readthedocs.io/en/stable/index.html. The R package DESeq236 is freely available at http://www.bioconductor.org/packages/release/bioc/html/DESeq2.html.

Data availability

Small-RNA-seq data from this work are available at NCBI GEO under accession GSE84955 and NCBI SRA under project PRJNA408115. The draft, preliminary C. campestris genome and transcriptome assemblies used in this study are available at the Parasitic Plant Genome Project website at http://ppgp.huck.psu.edu. C. campestris miRNA loci have been registered with miRBase. Source data availability: Fig. 1b, in Supplementary Data 2 and 3; Figs 1c, 2c, d, 3b, 4b and c, in Supplementary Fig. 1; Fig. 3a, c and Extended Data Fig. 4, included as Source Data; Fig. 4a, in Extended Data Table 1. There are no restrictions on data availability and the corresponding author will provide any data not already included as Supplementary Data or as Source Data upon request.


Primary accessions


Gene Expression Omnibus


  1. 1.

    & Macromolecule exchange in Cuscuta–host plant interactions. Curr. Opin. Plant Biol. 26, 20–25 (2015)

  2. 2.

    , & Phased, secondary, small interfering RNAs in posttranscriptional regulatory networks. Plant Cell 25, 2400–2415 (2013)

  3. 3.

    VIGS, HIGS and FIGS: small RNA silencing in the interactions of viruses or filamentous organisms with their plant hosts. Curr. Opin. Plant Biol. 26, 141–146 (2015)

  4. 4.

    , & Conversations between kingdoms: small RNAs. Curr. Opin. Biotechnol. 32, 207–215 (2015)

  5. 5.

    et al. HIGS: host-induced gene silencing in the obligate biotrophic fungal pathogen Blumeria graminis. Plant Cell 22, 3130–3141 (2010)

  6. 6.

    , , , & Engineering broad root-knot resistance in transgenic plants by RNAi silencing of a conserved and essential root-knot nematode parasitism gene. Proc. Natl Acad. Sci. USA 103, 14302–14306 (2006)

  7. 7.

    et al. Control of coleopteran insect pests through RNA interference. Nat. Biotechnol. 25, 1322–1326 (2007)

  8. 8.

    et al. Interspecific RNA interference of SHOOT MERISTEMLESS-like disrupts Cuscuta pentagona plant parasitism. Plant Cell 24, 3153–3166 (2012)

  9. 9.

    et al. Fungal small RNAs suppress plant immunity by hijacking host RNA interference pathways. Science 342, 118–123 (2013)

  10. 10.

    et al. Bidirectional cross-kingdom RNAi and fungal uptake of external RNAs confer plant protection. Nat. Plants 2, 16151 (2016)

  11. 11.

    et al. Cotton plants export microRNAs to inhibit virulence gene expression in a fungal pathogen. Nat. Plants 2, 16153 (2016)

  12. 12.

    , , , & Plant science. Genomic-scale exchange of mRNA between a parasitic plant and its hosts. Science 345, 808–811 (2014)

  13. 13.

    et al. 22-nucleotide RNAs trigger secondary siRNA biogenesis in plants. Proc. Natl Acad. Sci. USA 107, 15269–15274 (2010)

  14. 14.

    et al. Unique functionality of 22-nt miRNAs in triggering RDR6-dependent siRNA biogenesis from target transcripts in Arabidopsis. Nat. Struct. Mol. Biol. 17, 997–1003 (2010)

  15. 15.

    et al. Plant development is regulated by a family of auxin receptor F box proteins. Dev. Cell 9, 109–119 (2005)

  16. 16.

    et al. The membrane-anchored BOTRYTIS-INDUCED KINASE1 plays distinct roles in Arabidopsis resistance to necrotrophic and biotrophic pathogens. Plant Cell 18, 257–273 (2006)

  17. 17.

    et al. Inverse modulation of plant immune and brassinosteroid signaling pathways by the receptor-like cytoplasmic kinase BIK1. Proc. Natl Acad. Sci. USA 110, 12114–12119 (2013)

  18. 18.

    et al. Phloem ultrastructure and pressure flow: Sieve-element-occlusion-related agglomerations do not affect translocation. Plant Cell 23, 4428–4445 (2011)

  19. 19.

    et al. P-proteins in Arabidopsis are heteromeric structures involved in rapid sieve tube sealing. Front. Plant Sci. 4, 225 (2013)

  20. 20.

    , , & SCHIZORIZA controls an asymmetric cell division and restricts epidermal identity in the Arabidopsis root. Development 129, 4327–4334 (2002)

  21. 21.

    , & SCHIZORIZA controls tissue system complexity in plants. Curr. Biol. 20, 818–823 (2010)

  22. 22.

    et al. SCHIZORIZA encodes a nuclear factor regulating asymmetry of stem cell divisions in the Arabidopsis root. Curr. Biol. 20, 452–457 (2010)

  23. 23.

    et al. miR393 and secondary siRNAs regulate expression of the TIR1/AFB2 auxin receptor clade and auxin-related development of Arabidopsis leaves. Plant Physiol. 157, 683–691 (2011)

  24. 24.

    , , & MicroRNA-directed phasing during trans-acting siRNA biogenesis in plants. Cell 121, 207–221 (2005)

  25. 25.

    , , & A two-hit trigger for siRNA biogenesis in plants. Cell 127, 565–577 (2006)

  26. 26.

    , , & Biology and control of Cuscuta. Rev. Weed Sci. 6, 265–317 (1994)

  27. 27.

    et al. The microRNA miR393 re-directs secondary metabolite biosynthesis away from camalexin and towards glucosinolates. Plant J. 67, 218–231 (2011)

  28. 28.

    et al. A receptor-like cytoplasmic kinase, BIK1, associates with a flagellin receptor complex to initiate plant innate immunity. Proc. Natl Acad. Sci. USA 107, 496–501 (2010)

  29. 29.

    , , & Entangled evolutionary history of Cuscuta pentagona clade: A story involving hybridization and Darwin in the Galapagos. Taxon 64, 1225–1242 (2015)

  30. 30.

    et al. Arabidopsis mutants impaired in cosuppression. Plant Cell 10, 1747–1758 (1998)

  31. 31.

    , , & DICER-LIKE 4 functions in trans-acting small interfering RNA biogenesis and vegetative phase change in Arabidopsis thaliana. Proc. Natl Acad. Sci. USA 102, 12984–12989 (2005)

  32. 32.

    et al. Complex regulation of the TIR1/AFB family of auxin receptors. Proc. Natl Acad. Sci. USA 106, 22540–22545 (2009)

  33. 33.

    Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet. journal 17, 10–12 (2011)

  34. 34.

    , , & Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10, R25 (2009)

  35. 35.

    , , & Improved placement of multi-mapping small RNAs. G3 (Bethesda) 6, 2103–2111 (2016)

  36. 36.

    , & Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014)

  37. 37.

    , & miR156 and miR390 regulate tasiRNA accumulation and developmental timing in Physcomitrella patens. Plant Cell 24, 4837–4849 (2012)

  38. 38.

    & Improved northern blot method for enhanced detection of small RNA. Nat. Protoc. 3, 1077–1084 (2008)

  39. 39.

    et al. A draft genome sequence of Nicotiana benthamiana to enhance molecular plant–microbe biology research. Mol. Plant Microbe Interact. 25, 1523–1530 (2012)

  40. 40.

    , , , & Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol. 139, 5–17 (2005)

Download references


This research was supported in part by awards from the US National Science Foundation (1238057 to J.H.W. and C.W.D.; 1339207 to M.J.A.) and the US National Institute of Food and Agriculture (135997 to J.H.W.).

Author information

Author notes

    • Feng Wang
    •  & Ceyda Coruh

    Present addresses: Department of Biology, Indiana University, Bloomington, Indiana 47405, USA (F.W.); Salk Institute for Biological Studies, La Jolla, California 92037, USA (C.C.).


  1. Huck Institutes of the Life Sciences, The Pennsylvania State University, University Park, Pennsylvania 16802, USA

    • Saima Shahid
    • , Nathan R. Johnson
    • , Feng Wang
    • , Ceyda Coruh
    • , Claude W. dePamphilis
    •  & Michael J. Axtell
  2. Department of Biology, The Pennsylvania State University, University Park, Pennsylvania 16802, USA.

    • Saima Shahid
    • , Nathan R. Johnson
    • , Eric Wafula
    • , Feng Wang
    • , Ceyda Coruh
    • , Claude W. dePamphilis
    •  & Michael J. Axtell
  3. Department of Plant Pathology, Physiology and Weed Science, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061, USA

    • Gunjune Kim
    • , Vivian Bernal-Galeano
    •  & James H. Westwood
  4. Knox College, Galesburg, Illinois 61401, USA

    • Tamia Phifer


  1. Search for Saima Shahid in:

  2. Search for Gunjune Kim in:

  3. Search for Nathan R. Johnson in:

  4. Search for Eric Wafula in:

  5. Search for Feng Wang in:

  6. Search for Ceyda Coruh in:

  7. Search for Vivian Bernal-Galeano in:

  8. Search for Tamia Phifer in:

  9. Search for Claude W. dePamphilis in:

  10. Search for James H. Westwood in:

  11. Search for Michael J. Axtell in:


S.S. and M.J.A. did the bioinformatics analysis. S.S., M.J.A. and N.R.J. prepared figures and tables. G.K., J.H.W., N.R.J., S.S., T.P. and M.J.A. cultivated and harvested plant specimens. E.W., G.K., C.W.D. and J.H.W. performed genome and transcriptome sequencing and assemblies. F.W., S.S. and N.R.J. did RNA blotting. S.S. and M.J.A. performed 5′-RLM-RACE and qRT–PCR. C.C. and T.P. constructed small-RNA-seq libraries. N.R.J. and V.B.-G. performed growth assays. M.J.A. and J.H.W. conceived the project. M.J.A. wrote and revised the manuscript.

Competing interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to Michael J. Axtell.

Reviewer Information Nature thanks M. Albert, F. Tang and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Supplementary information

PDF files

  1. 1.

    Life Sciences Reporting Summary

  2. 2.

    Supplementary Data 4

    This file contains images showing C. campestris hairpins overlaid with color-codes representing total read-depth (all six 'original' C. campestris x A. thaliana small RNA libraries). (pdf, 43 pages).

  3. 3.

    Supplementary Fig. 1

    This file contains uncropped gel figures.

Excel files

  1. 1.

    Supplementary Data 1

    This file contains A. thaliana and C. campestris small RNA loci. Output from ShortStack 3.8.3 (https://github.com/MikeAxtell/ShortStack) showing all small RNA loci identified in this study. (.xlsx format)

  2. 2.

    Supplementary Data 2

    This file contains mature miRNAs and miRNA*s fromC. campestris Interface-induced MIRNA loci. (.xlsx format)

  3. 3.

    Supplementary Data 6

    This file contains oligonucleotide sequences. Excel (.xlsx) format.

Text files

  1. 1.

    Supplementary Data 3

    This file contains details ofC. campestris MIRNA loci: Text-based sequences, predicted secondary structures, and aligned small RNA reads (all six libraries). Lower-case letters indicate small RNA bases that are mismatched to the genomic sequence. Plain-text (ASCII) format.

  2. 2.

    Supplementary Data 5

    This file contains alignments of mature miRNAs and miRNA*s (from Supplementary Data 2 -- the induced C. campestris miRNAs) against the A. thaliana genome. Note that most have three or more mismatches. (SAM format).

About this article

Publication history






Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.