Quinone oxidoreductase 2 is involved in haustorium development of the parasitic plant Phtheirospermum japonicum

ABSTRACT

The family Orobanchaceae includes many parasitic plant species. Parasitic plants invade host vascular tissues and form organs called haustoria, which are used to obtain water and nutrients. Haustorium formation is initiated by host-derived chemicals including quinones and flavonoids. Two types of quinone oxidoreductase (QR) are involved in signal transduction leading to haustorium formation; QR1 mediates single-electron transfers and QR2 mediates 2-electron transfers. In the facultative parasite Triphysaria versicolor, QR1 is involved in haustorium induction signaling, while this role is played by QR2 in the model plant Phtheirospermum japonicum. Our results suggest that there is functional diversification in haustorium signaling molecules among different species of the Orobanchaceae.

Parasitic plants in the Orobanchaceae include harmful pests targeting a large range of crops.^1,2 The parasites interact with their hosts via haustorium,³ a unique organ connecting own vasculature to the host conducting system to obtain water, nutrients, and small substances.^4-6 To develop strategies for controlling plant parasites, it is important to understand the intricate molecular mechanisms that control the interactions between host and parasite. Parasitic plants initiate haustorium development upon perception of host-derived haustorium-inducing factors (HIFs). HIFs include quinones and flavonoids, which are broadly distributed in nature.⁷ The quinone 2,6-dimethoxy1,4-benzoquinone (DMBQ), which was originally identified in sorghum root extracts,⁸ is an active HIF for many Orobanchaceae species.^9,10 Recognition by the parasites is associated with the redox potential of DMBQ.¹¹ The quinone oxidoreductase enzymes (_EC 1.6.5) catalyze quinone redox changes via the transfer of one or 2 electrons. The NADPH-dependent quinone oxidoreductase, QR1, catalyzes the transfer of single electrons from quinones to generate the highly reactive free radicals semiquinones, leading to the formation of reactive oxygen species.¹² Another type of quinone oxidoreductase (QR2) catalyzes the divalent reduction of quinones to hydroquinones, possibly working as a scavenger of reactive quinone molecules.¹³ In the facultative parasite Triphysaria versicolor, both QR1 and QR2 transcripts are upregulated during haustorium induction triggered by DMBQ.¹⁴ However, only Tv-QR1 is differentially regulated in response to host contact.¹⁵ The knockdown of QR1 but not QR2 expression in T. versicolor roots significantly reduced the frequency of haustorium formation.¹⁵ These observations suggest that QR1 is involved of in haustorium initiation signaling in T. versicolor.

The facultative parasite Phtheirospermum japonicum also belongs to the Orobanchaceae and has become a model for studies of parasitic plants.^16,17 P. japonicum responds to HIFs in a similar way to other parasitic Orobanchaceae and forms lateral haustoria (i.e., haustoria formed on the lateral side of parasitic roots).¹⁰ To expand our understanding of the molecular events associated with haustorium development, we sequenced the transcriptomes of P. japonicum roots and haustorial tissues and generating a list of genes that are actively expressed during infection.¹⁸ Gene expression patterns were investigated after exposure of parasite roots to DMBQ over a period ranging from 30 minutes to 48 hours. Although most genes in the haustorium transcriptome are similar among parasitic plant species in the Orobanchaceae, there are some differences. For example, our analyses revealed that in P. japonicum the expression patterns of QR1 and QR2 differ from those in T. versicolor. In P. japonicum, Pj-QR1 expression is not altered by contact with host root exudates or DMBQ treatments.¹⁸ In contrast, Pj-QR2 is highly upregulated in response to both treatments.¹⁸ To investigate the expression patterns of QR1 and QR2 in other parasitic species, we analyzed the QR1 and QR2 homologs in S. hermonthica. The homologs Sh-QR1, Sh-QR2, Pj-QR1 and Pj-QR2 were amplified from a cDNA library¹⁹ using primers shown in Table 1, and the full length sequence was confirmed by a combination of RACE® and Sanger-based strategies as described previously.¹⁸ A phylogenetic analysis of the full length putative QR proteins from T. versicolor, P. japonicum, and S. hermonthica clearly assigned the QR1 and QR2 homologs from all 3 species into distinct nodes with high support values (Fig. 1). Transcription profiles for the Sh-QR genes were obtained by mapping the RNA-Seq reads from different developmental stages onto the Sh-QR1 and Sh-QR2 sequences²⁰ (Fig. 2). As a control we included the housekeeping gene β-tubulin 1 (TUB1, Unigene accession StHe1GB1_52449).²¹ Both Sh-QR1 and Sh-QR2 showed basal transcriptional levels at the seedling stage. Similar to Pj-QR2, the transcriptional level of Sh-QR2 was increased during the haustorial development stages, while the Sh-QR1 expression was stabely maintained in both parasitic and non-parasitic vegetative tissues. Both genes showed increased expression with the development of reproductive structures. Our data indicate that although each group of QR homologs in T. versicolor, P. japonicum, and S. hermonthica shares the same origin, their expression patterns differ during haustorial development.

Table 1.

Primers sequences.

Primer Name	sequence	Primer size	Amplicon size	Application
ShQR1-F	CCCAATTGCCAACTTTATTCA	21	1375 bp	Full length
ShQR1-R	AGTAGAACTGATGAGCGGCG	20		Full length
ShQR2-F	CACACTTCACACACCAAATCAA	22	702 bp	Full length
ShQR2-R	TTTCCCGATTCATCAAATAAA	20		Full length
PjQR1-F	CAAACCCTCTACATAACACACAAAGG	26	1204 bp	Full length
PjQR1-R	TTATGTCGTATTTTATATCTTGTTCGATCA	30		Full length
PjQR2-F	CCAACCAACTCATACTAAACCAAA	24	856 bp	Full length
PjQR2-R	GATGCCAATGATTTCTTGC	19		Full length
PjQR2-RNAi-F	GGCAGGTCCCAGAAACTCTG	21	448 bp	RNAi target
PjQR2-RNAi-R	AAAGCTTGTGCGAGTTCGAT	20		RNAi target
PjQR2-qPCR-F	ATGTACATCGCAGGCATCAC	20	73 bp	qRT-PCR
PjQR2-qPCR-R	GGATGCAATTAGCATGATCG	20		qRT-PCR

Figure 1.

Phylogenetic analysis of the QR homologs of T. versicolor, P. japonicum, and S. hermonthica. The tree was generated with the MEGA7 software²³ using the maximum likelihood statistical method. The putative QR proteins were aligned using the CLUSTALW²⁴ algorithm with default settings. The bootstrap percentages of 10000 replicates are shown on the internal nodes. The topology of the tree was also confirmed by the UPGMA and neighbor-joining methods.

Figure 2.

Expression profiles of Sh-QR1 (orange) and Sh-QR2 (gray) at different developmental stages in the S. hermonthica life cycle. Transcript levels were determined as log2 (RPKM +1) values from an RNA-Seq analysis.²⁰ For reference, we also included the expression profile of the S. hermonthica TUB1 housekeeping gene²¹ (blue).

We performed an RNA interference (RNAi) experiment to investigate the relevance of Pj-QR2 in P. japonicum haustorial development. Fragments of Pj-QR2 were amplified with the RNAi primers shown in Table 1 and inserted into the silencing vector pHG8-YFP.¹⁵ The transcript levels of Pj-QR2 were quantified by real-time qPCR in the roots of plants transformed with the silencing construct (pHG8-QR2), and roots harbouring the empty vector (pHG8-YFP) were used as negative controls. In the pHG8-QR2 plants the transcript levels of Pj-QR2 were reduced to about one tenth of the levels in the control plants (Fig. 3A). The total numbers of roots, numbers of lateral roots, and root lengths were similar between the pHG8-QR2 and pHG8-YFP plants (Table 2). However, the percentages of haustoria formed after to exposure to host root exudates were significantly lower in the pHG8-QR2 roots than in the pHG8-YFP roots (Fig. 3B). This result indicated that Pj-QR2 is involved in haustorium formation in P. japonicum.

Figure 3.

Phenotype of the P. japonicum RNAi line targeted to Pj-QR2. (A) Transcript levels of Pj-QR2 in RNAi lines (pHG8-QR2) and empty vector controls (pHG8-YFP) treated with rice root exudates. Transcript levels were determined by quantitative RT-qPCR using the primers (PjQR2-qPCR-F and PjQR2-qPCR-R) shown in Table 1. (B) Rates of haustoria formation in roots of the pHG8-QR2 and pHG8-YFP lines after treatment with rice root exudates. Asterisks (*) represent α = 0.05 by t-Test assuming unequal variances. Values represent average ± SD of 3 to 5 independent experiments with 10 to 40 transformed roots per experiment.

Table 2.

Root morphology in RNAi (pHG8-QR2) and control (pHG8-YFP) P. japonicum lines.

	pHG8-YFP	pHG8-QR2
Number of lateral roots	1.16 ( ± 0.5)	1.0 ( ± 0.2)
Root length (mm)	10.26 ( ± 1.7)	7.6 ( ± 5.1)
Total number of transgenic roots (N)	N = 90	N = 113

Data are means and standard error ( ± ) of 3 to 5 biologic replicates with 5–15 independent transgenic roots per experiment.

In summary, our results suggest that the expression patterns and functions of QR1 and QR2 homologs during haustorium formation are diversified among parasitic plant species. In T. versicolor, Tv-QR1 has important roles in haustorium induction after exposure to HIFs, while in P. japonicum, haustorium formation is mediated by Pj-QR2. The high expression levels of the S. hermonthica homolog Sh-QR2 during the parasitic stages suggest that, like Pj-QR2, Sh-QR2 may also be involved in haustorium development. The biologic function of Tv-QR1 may not be specific for haustorium induction since this gene is recruited for floral development and is responsive to non-HIF quinines.²⁰ These observations suggest that Tv-QR1 may play a role in oxidative stress signaling. Genetics studies of natural populations of T. versicolor from Northern California indicated that the Tv-QR1 alleles retain a high level of molecular diversity at the nucleotide and amino acid levels, with the highest non-synonymous substitution rates in the catalytic domain.²² Such sequence diversity was not found in Tv-Pirin, another gene that is involved in T. versicolor parasitism.²² Such plasticity in QR1 and QR2 may confer evolutionary advantage for parasitic plants that recognize different factors from a wide range of hosts. Thus, the functions of QR1 and QR2 may have been diversified during evolution, depending on the species. Future studies will help to elucidate the diverse roles of these quinone oxidoreductases in plant parasitism.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This work is partially supported by MEXT KAKENHI grants (24228008 and 15H05959 to K.S. and 25114521, 25711019, and 25128716 to S.Y.) and the PhD fellowship programs (MEXT to J.K.I.).