Diagnosis and characterization of the ribosomal DNA-ITS of potato rot nematode (Ditylenchus destructor) populations from Chinese medicinal herbs

NI Chun-hui ,HAN Bian ,LIU Yong-gang ,Maria MUNAWAR ,LIU Shi-ming ,LI Wen-hao ,SHI Ming-ming,LI Hui-xia#,PENG De-liang#

1 College of Plant Protection, Gansu Agricultural University/Biocontrol Engineering Laboratory of Crop Diseases and Pests of Gansu Province, Lanzhou 730070, P.R.China

2 Institute of Plant Protection, Gansu Academy of Agricultural Sciences, Lanzhou 730070, P.R.China

3 Department of Biological Sciences, University of Lethbridge, Lethbridge 4401, Canada

4 State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing 100193, P.R.China

Abstract The potato rot nematode (Ditylenchus destructor) is a very economically important nematode in agronomic and horticultural plants worldwide.In this study,43 populations of D.destructor were collected from different hosts across China,including 37 populations from Chinese herbal medicine plants.Obtained sequences of ITS-rDNA and D2–D3 of 28S-rDNA genes of D.destructor were compared and analyzed.Nine types of significant length variations in ITS sequences were observed among all populations.The differences in ITS1 length were mainly caused by the presence of repetitive elements with substantial base substitutions.Reconstructions of ITS1 secondary structures showed that the minisatellites formed a stem structure.Ten haplotypes were observed in all populations based on mutations and variations of helix H9.Among them,3 known haplotypes (A–C) were found in 7 populations isolated from potato,sweet potato,and Codonopsis pilosula,and 7 unique haplotypes were found in other 36 populations collected from C.pilosula and Angelica sinensis compared with 7 haplotypes (A–G) according to Subbotin’ system.These unique haplotypes were different from haplotypes A–G,and we named them as haplotypes H–N.The present results showed that a total of 14 haplotypes (A–N) of ITS-rDNA have been found in D.destructor.Phylogenetic analyses of ITSrDNA and D2–D3 showed that all populations of D.destructor were clustered into two major clades: one clade only containing haplotype A from sweet potato and the other containing haplotypes B–N from other plants.For further verification,PCR-ITS-RFLP profiles were conducted on 7 new haplotypes.Collectively,our study suggests that D.destructor populations on Chinese medicinal materials are very different from those on other hosts and this work provides a paradigm for relevant researches.

Keywords: Ditylenchus destructor,minisatellites,ITS-RFLP,phylogeny,RNA secondary structure

1.Introduction

The potato rot nematode,DitylenchusdestructorThorne,1945,is one of the serious pests in potato (Solanum tuberosumL.) and gained significant importance due to their parasitic potentials and quarantine regulations(Thorne 1945;Goodey 1962;Vovlaset al.2016).Biologically,the principal host ofD.destructoris potato,however,it has been reported in association with a wide variety of hosts such as bulbous iris,sweet potato,sugar beet,and several other crops (Bassonet al.1993;Zhaoet al.2021).In addition,D.destructorcan feed and reproduce on mycelia of approximately 70 fungal species such asAlternaria,Botrytis,Fusarium,andPenicilliumgenera (Wuet al.1960;Faulkner and Darling 1961).In China,D.destructorwas first reported in 1983,causing severe damage to sweet potato production in Shandong Province (Yi and Zhang 1983),afterwards,it was also found in potato,sweet potato,and angelica in different regions (Yi and Zhang 1983;Wanget al.1990;Qiet al.2008;Liet al.2016;Ouet al.2017;Zhanget al.2019).Particularly,D.destructorwas found implicating and reducing angelica production in Gansu,Qinghai,and Yunnan provinces,China (Xuet al.2009),where Chinese herbal medicines are mainly cultured.Recently,D.destructorwas reported inCodonopsispilosula(a wellknown traditional Chinese herbal medicine) from Gansu(Niet al.2020).

Several reports showed thatD.destructorhad intraspecific variation in pathogenicity,salt and cold tolerance,and drug resistance among different populations from different hosts or regions (De Waeleet al.1989;Subbotinet al.2005;Wanget al.2011b).Subsequently,molecular characterization ofD.destructorrevealed significant variations in length and composition of ITS-rDNA by sequence analysis(Wanet al.2008;Wanget al.2011a;Jeszkeet al.2014).PCR-ITS-RFLP analyses showed that different isolates ofD.destructorcan be distinguished using several restriction enzymes (Jiet al.2006;Liuet al.2007;Mareket al.2010;Mahmoudiet al.2020).Wanet al.(2008) found that ITS-rDNA sequences ofD.destructorwere split into two types: the short (940 bp) and the long(1 100 bp) sequences among 21 populations collected from sweet potato in China and one population from potato in Korea.However,the D2–D3 28S-rDNA was identical in length (ca.780 bp),but classified into two branches in phylogenetic analysis conducted using 21 populations obtained from sweet potato and garlic (Yuet al.2009).Consequently,Zhang and Zhang (2008)showed that the variations of ITS-rDNA sequences from differentD.destructorpopulations were mainly localized in the ITS1 region.

Moreover,several short and long repeated DNA sequences were described in the ITS1 region ofD.destructorby Mareket al.(2010).The tandem repeats(12 and 13 nucleotides) were found in ITS-rDNA gene sequences obtained from sweet potato,potato,andAstragalusmembranaceusacross the world.Based on mutations and helix H9 length,the ITS-rDNA sequences can be classified into 7 haplotype groups (namely A,B,C,D,E,F,and G) (Subbotinet al.2011).Recent studies have indicated the presence of additional haplotypes that are different from haplotypes A–G (Niet al.2021).The earlier data on the presence ofD.destructorin Chinese herbal medicines was not adequately addressed,and there was a lack of molecular sequencing data in GenBank.

In view of limited information and poor documentation,the objectives of the present study were to understand the variations of ITS-rDNA and 28S-rDNA gene sequences from different populations ofD.destructorfrom Chinese herbal medicines,namely angelica andC.pilosula;to study the phylogeny ofD.destructorpopulations from different hosts and geographical areas;and to verify variations of ITS-rDNA by ITS-RFLP and PCR using species-specific primers.

2.Materials and methods

2.1.Nematodes

Symptomatic plants (tubers and roots) of angelica,C.pilosula,sweet potato and potato were collected from Gansu,Heilongjiang,Inner Mongolia,Qinghai,and Shaanxi,China.Nematodes were extracted using a modified Baermann funnel (Hooper 1990).To obtain pure cultures ofD.destructor,individual nematodes were collected from the mixture and transferred to fungal cultures (Fusariumsolani) for propagation.Detailed information of location and the associated hosts are listed in Table 1.

Table 1 Ditylenchus destructor collection information and sequence BLAST results

2.2.DNA extraction,PCR,and sequencing

Single female nematodes were picked from each

population to extract DNA.The single female was transferred into a 0.2-mL Eppendorf tube containing 10 μL of worm lysis buffer (50 mmol L–1KCl,10 mmol L–1Tris(pH=8.3),2.5 mmol L–1MgCl2,0.45% Nonidet P-40 and 0.45% Tween 20).The tubes were transferred to liquid nitrogen container for 1 min,followed by 2 min incubation at 80°C in a water bath.After cooling down,the tubes were briefly spun followed by the addition of 1 μL of proteinase K (1 mg mL–1) (Sangon Biotech,Shanghai,China) and incubation at 65°C for 1.5 h with an extension of 95°C for 10 min,and finally centrifuged at 14 000 r min–1for 2 min.

The PCR reaction was conducted in a final volume of 25 μL,containing 12.5 μL 2× San Taq Fast PCR Master Mix (with Blue Dye,Sangon Biotech,Shang,China),1 μL of each primer (10 μmol L–1),3 μL DNA,and double distilled water to 25 μL.Two sets of primers(synthesized by Tsingke Biotech Co.Ltd.,Xi’an,China)were used to amplify the ITS-rDNA region.The first set was TW81 (5′-GTTTCCGTAGGTGAACCTGC-3′)and AB28 (5′-ATATGCTTAAGTTCAGCGGGT-3′) as described by Subbotinet al.(2001).The second was the combination of 18S (5′-TTGATTACGTCCCTGCCCTTT-3′)and 26S (5′-TTTCACTCGCCGTTACTAAGG-3′)as described by Vrainet al.(1992).To amplify the region of 28S-rDNA D2–D3,the primers D2A(5′-ACAAGTACCGTGAGGGAAAGTTG-3′) and D3B(5′-TCGGAAGGAACCAGCTACTA-3′) (Subbotinet al.2006) were used.The PCR amplification program consisted of 4 min at 94°C,38 cycles of 10 s at 94°C,20 s at 55°C,and 15 s at 72°C.After amplification,5 μL of the products were separated in 1.5% TAE buffered agarose gel and visualized by staining with GelRed (Tsingke Biotech Co.Ltd.,Xi’an,China).Cloning and sequencing of PCR products were carried out at Tsingke Biotech.The newly acquired sequences were deposited in the GenBank database to obtain the accession numbers (Table 1).

2.3.PCR with species-specific primers and ITSRFLP

The species-specific primers of ITS-rDNA region DITuni F (5′-CTGTAGGTGAACCTGC-3′) and DITdes R(5′-GTTTTTCGCCCACAAATTAGC-3′) (Jeszkeet al.2015) were used for PCR amplification.PCR mixtures were prepared as described above.PCR program consisted of initial denaturation of 4 min at 94°C,38 cycles of 10 s at 94°C,20 s at 52°C,and 15 s at 72°C.Amplified PCR products were resolved by electrophoresis in 3% agarose gels and visualized by staining with GelRed.PCR products containing amplified DNA fragments of interest were sequenced at Tsingke Biotech.

PCR-RFLP was performed as described by Subbotinet al.(2011).Briefly,the ITS purified products were digested successively by restriction enzymesDdeI,HinfI,Tru9I (MseI),andSduI in the 200-μL Eppendorf tube.Each digestion reaction was set for 20 μL and contained 2 μL of 10× reaction buffer,8 μL of ITS purified products,and 1 μL of restriction enzyme (10 U μL–1),and 9 μL of distilled water.The tube was incubated at 37°C for 1.5 h.The digested products were resolved by electrophoresis in 1% agarose gels and visualized by staining with GelRed (Tsingke Biotech).WebCutter 2.0 (http://www.firstmarket.com/cutter/cut2.html) was applied to obtain each restriction fragment’s exact length from the PCR products by virtual digestion of the sequence.

2.4.Sequence alignment and secondary structure reconstruction of ITS-rDNA

The exact length of ITS-rDNA sequences amplified using primers TW81/AB28 were obtained using software BioEdit,and sequence alignment of ITS1 region was conducted referring to the sequence partition of ITS-rDNA fromD.destructor(Subbotinet al.2011) in GenBank.The newly obtained ITS-rDNA and 28S D2–D3 sequences ofD.destructorwere aligned with otherD.destructorpopulation sequences present in the GenBank using BioEdit v7.0.9 (Hallet al.2012).The secondary structures for ITS1 and ITS2 were predicted by applying the UNFold Web Server (Mfold: RNA Folding Form,http://www.unafold.org/mfold/applications/rna-folding-form.php)(Zuker 2003) using the energy minimization approach.The structures were visualized using Varna (Dartyet al.2009) and drawn with Photoshop CC 2018.Finally,sequence differences of the secondary structures for helices H9 were aligned using BioEdit.

2.5.Sequence and phylogenetic analyses

The Tandem Repeats version 4.09 (https://tandem.bu.edu/trf/trf.submit.options.html) (Benson 1999) was used to analyze tandem repeat sequences and their copies within the ITS1 for each sequence.ITS-rDNA sequences of haplotypes A to G and 28S D2–D3 sequences ofD.destructordescribed by Subbotinet al.(2011) were used for phylogenetic analysis.The sequences ofD.myceliophagus(DQ151458,AM232236) were used as outgroup taxon for analysis of ITS-rDNA,andMesoanguina millefolii,Eutylenchusexcretorius(EU915490,DQ328722)were used as outgroup taxon for analysis of 28S D2–D3.The sequence datasets were analyzed using Bayesian inference (BI) as described by Subbotinet al.(2011),and Mafft v7.149b was used for sequence alignment.The best fit model of DNA evolution for BI was obtained using MrModeltest2.3 (Nylander 2004) according to the Akaike Information Criterion (AIC) in conjunction with PAUP＊.BI analysis using MrBayes 3.1.2 (Huelsenbeck and Ronquist 2001) under the best fit model was initiated starting tree and run with four Markov chains Monte Carlo (MCMC,three heated and one cold) for 1 000 000 generations.The MCMC were sampled at intervals of 100 generations and the burn-in value was 25%.Two runs were performed for each analysis.The log-likelihood values of the sample points stabilized after approximately 1 000 generations.After discarding burn-in samples,the remained samples were used to generate a 50% majority-rule consensus tree.Posterior probabilities (PP) were given on appropriate clades.The phylogenetic consensus trees were visualized using FigTree v.1.4.3 (Rambaut 2016).

3.Results

3.1.Sequences variations of ITS-rDNA and 28S D2–D3 of D.destructor

In the present study,a total of 43D.destructorpopulations were collected,including 37 from Chinese herbal medicines,5 from potato,and 1 from sweet potato (Table 1).The size of the fragment of ITS-rDNA sequences amplified using primers TW81/AB28 was 727 to 969 bp,and 6 sequence from potatoes and sweet potatoes showed the same length withD.destructorpopulations deposited in GenBank with sequence identity of 99.34 to 100% (with 0–5 nucleotides differences) (Table 1).We also noted that other 12 populations collected from Chinese herbal medicines showed comparatively short sequences (727 bp),however,these sequences have 98.06% with a 14 nucleotides difference (Table 1),similar to that of theD.destructorpopulation reported from Beijing,China (EF418002).Interestingly,the sequences (786 to 969 bp) of other 25 populations from Chinese herbal medicines were different(92.11–98.53% identity with 13–69 nucleotides differences)from the sequences deposited in GenBank.An exception was population TCC1 fromC.pilosulain Tanchang County,with the difference of only one nucleotide from haplotype B sequence (FJ911551,identity of 99.89%) (Table 1).Alignment of ITS1 sequences showed that lengths of ITS1 varied from 285 to 527 bp while lengths of 5.8S and ITS2 of all populations were 155 and 208 bp,respectively.The inserts of 59 to 242 bp were observed in the ITS1 region in someD.destructorpopulations.Therefore,we anticipate that these inserts caused length variation of ITS1.

Regarding,D2–D3 of 28S,731-bp long fragments were obtained from 16 populations ofD.destructor(GenBank accession nos.MZ664407–MZ664422).The nBLAST results showed that our sequences were 99% similar to that ofD.destructor.The sequence alignment showed 4 types of base differences among all samples: (1) the type of LXC1,TCC1,ZXC1,MXC1,DXP1,DXP2,HLJP1,SXP1,DTA1,HZA3,TCA4,and ZXA2 sharing exactly the same sequence;(2) the type of MXA8 and HZA2 with a 3-base difference from the type (1);(3) the type of TCC2 with a one-base difference from the type (1);and (4) the type of SXSP1 with an 8-base difference from the type (1).

3.2.Characterization of tandem repeat sequences in ITS-rDNA of D.destructor

Two categories of tandem repeat sequences (13 nucleotides (nt) and 12 nt,Fig.1) were detected in our 29 populations ofD.destructor(the ITS-rDNA sequence length of 888 to 969 bp).One category contained 13 nt with 4 types: (GTTGCGCTGTGTR)n (LXC1),(TTGCGCTGTGTRC)n (MXA2,3,6,8;HLJP;HZA1,2;DXP2;NMP1;SXP1;and WYA2),(GCTGTGTGCTTGY)n(DTA1;DXP1;LXC2;MXA1,10;MXC1;TCA1,2,3;TCC1;WYA1,4,7,10,12;and HZA3),and(TGCGCTGTGTGCC)n (WYA11) with a consensus sequence: GCTGTGTR (Fig.1-A).The other category contained 12-nt with 2 types: (CTTGCATTTGAG)n(DXP2;HLJP1;NMP1;SXP1) and (GCTTGCATTTGH)n(other 25 populations) with a consensus sequence:CTTGCATTTGM (Fig.1-C).In the ITS1 region,these core sequences were present 4,5,6 or 8 times (Fig.1-B),and these minisatellites were unique forD.destructorconfirmed by nBLAST analysis.Secondary structure prediction of the ITS1 sequences showed that this tandem repeat sequence fragment was folded in a stem-like structure named as helix H9 (Figs.1-A,2 and 3).The 12-nt minisatellite was a reversed complement of the 13-nt minisatellites,and all of them were able to form a stable secondary structure in the helix H9 (Figs.1-A and 3).

Fig.1 The tandem repeat sequences in helix H9 for Ditylenchus destructoer (LXC1,MW522936).A,consensus motif of 13-nt tandem repeat sequences.B,helix H9 structure for D.destructor (LXC1,MW522936).Nucleotides with red background are 13-nt tandem repeat sequences and blue background are 12-nt tandem repeat sequence in helix H9.C,consensus motif of 12-nt tandem repeat sequences.

3.3.Inferred secondary structures of ITS1 and ITS2 sequences of D.destructor

The optimal secondary structures were predicted by Mfold for 3′-end of 18S rDNA-ITS1 and the 3′-end of 5.8S-ITS2-5′-end of 28S for all populations.The ITS1 secondary structures (the first 15 nucleotides in 5′-end belonging to 18S) showed them with 3 domains forD.destructor: (1) helix H3,(2) helices H5–H9 and(3) helices H10–H14 (Figs.2-B–F and 4-A),and with 4 domain structures forD.myceliphagus: (1) helix H3,(2)helix H4,(3) helices H5–H8 and (4) helices H10–H14(Figs.2-A and 4-A).The ITS1 secondary structures were conservative,and the main differences among sequences ofD.destructorwere whether they have helix H9,and lengths of helix H9,helices H8 and H11 (Figs.2-B–F,3,4-B and C,and 5).In ITS1 secondary structures,the helix H9 structure and interior loop I3 were lacked in the populations having short ITS-rDNA sequences (727 bp)(Fig.2-E and F),and ITS-rDNA sequences (786–969 bp)of other populations had helix H9 (Fig.2-B–D).Two types (short/long) were observed in helix H8 of ITS1 secondary structures (Fig.4-B and C),the long H8 helix(Figs.2-C and D,and 4-C) was found in 8 populations:TCA1,TCA2,TCA3,WYA7,WYA10,DTA1,HZA1,and HZA3.There were interior loops I4 and I5 in long H8 helix,but not in short H8 helix (Fig.4-B).The helix H11 showed 2 types: the terminal loop T8 with 6 nucleotides in SXSP1 and 8 nucleotides in other populations,these led to differences in H11 length (Fig.4-A).We also found that the short sequences of ITS-rDNA were different in helices H7/H8 and H13 between SXSP1 and other populations from Chinese herbal medicines.The ITS1 secondary structures of 86 to 88 bp and 100 to 102 bp formed an interior loop in helix H7/H8 in short sequence populations of Chinese herbal medicines,but SXSP1 did not have this structure.SXSP1 had an internal loop in the nearby interior loop I7 of helix 13,but another short length sequence from Chinese herbal medicines did not have this structure (Fig.2-E and F).Helix H4 was only found inD.myceliophagus(Figs.2-A and 4-A).The 17 locations with nucleotide differences were found in the ITS1 secondary structures without helix H9 among all samples ofD.destructor(Fig.4-A).Eight different types of helices H9 were found among all populations by comparing secondary structures and sequences (Figs.3 and 5).

Fig.2 Putative secondary structures of the ITS1 of Ditylenchus destructor and D.myceliophagus.A,D.myceliophagus (DQ151458).B,D.destructor (HQ235689).C,D.destructor (DQ471335).D,D.destructor (DTA1;MW048797).E,D.destructor (ZXA1;MT895792).F,D.destructor (EF418002).Free energy levels (dG in kcal mol–1) are given for six ITS1 structures.A,B and C are consistent with structures by Subbotin et al.(2011),other structures are mainly similar with those given by Subbotin et al.(2011)except for helix H9.

Fig.3 Types of H9 helix of ITS1 for Ditylenchus destructor.Type I to VI reference to Subbotin et al.(2011).

Fig.4 Putative secondary structures of the ITS1 and ITS2 of Ditylenchus destructor and D. myceliophagus.A,consensus ITS1 structures of D.destructor and D.myceliophagus.Helix H4 and the terminal loop T8b are given for D.myceliophagus.B,consensus fragments of ITS1 of D.destructor having helix H9 and showing helices H7 and H8 with interior loop I3.C,fragments of ITS1 of partial D.destructoer populations showing helix H8 with an insertion.D,consensus ITS2 structures of D.destructor and D.myceliophagus.The first 15 nucleotides in the 5′-end of ITS1 are the 18S rRNA,the first 13 nucleotides in the 5′-end of ITS2 are the 5.8S rRNA,and the last 11 nucleotides in the 3′-end of ITS2 are the 28S rRNA.H=helix;I=interior loop;T=terminal loop.Arrows indicate point mutations observed in at least two studied sequences of D.destructor.Nucleotides with a black background are point mutations of D.myceliophagus.

Fig.5 Alignment of helix H9 sequence for Ditylenchus destructor.Type I to VI reference to Subbotin et al.(2011).

Among all samples ofD.destructorobtained in the present study,the putative secondary structures of ITS2(the first 13 nucleotides in 5′-end belonged to 5.8S and the last 11 nucleotides in 3′-end belonged to 28S) showed them with 3 domains: (1) helices H1 and H2;(2) helices H3 to H5;and (3) helices H6 to H11,and 22 locations with nucleotides were found (Fig.4-D).

3.4.Classification and diagnostics of ITS haplotypes of D.destructor

At least 10 distinct haplotypes of ITS-rDNA sequences were observed in this study.Compared with haplotypes A–G from Subbotinet al.(2011),the sequence of SXSP1 belonged to haplotype A;the sequences of DXP1 and TCC1 were the same as haplotype B,the sequences ofDXP2,SXP1,NMP1,and HLJP1 belonged to haplotype C.However,other populations did not belong to any haplotypes A–G.The short sequences (727 bp) of ITSrDNA from Chinese herbal medicines were different from haplotype A in both sequences and secondary structures of ITS1.The constructed phylogenetic tree also showed that these two type sequences were classified into 2 clades (Fig.6).So,we named them as haplotype H.Six additional haplotypes included haplotypes I,J,K,L,M,and N.Based on the present knowledge,haplotypes J,L,M,and N were only found in angelica,haplotype I was only found inC.pilosula,haplotypes H and K were found in both angelica andC.pilosula,while haplotypes A,B,and C were found in both potato and sweet potato.In addition,we found that one population TCC1 fromC.pilosulabelonged to haplotype B.The geographical distribution of haplotypes was as follows: haplotypes A and B were found in Shaanxi Province,haplotype C was in Inner Mongolia,Heilongjiang and Gansu provinces(autonomous region),haplotypes L,M,and N were in Qinghai Province,and haplotypes B,C,H,I,J,K,L,and N were observed together in Gansu Province.Haplotype I was only found in Longxi Country,Gansu Province,and haplotype M was only found in Huzhu County,Qinghai Province.

Fig.6 The 50% majority rule tree from Bayesian analysis generated from the ITS-rDNA gene sequence dataset for Ditylenchus destructor using the GTR+I+G.The newly obtained sequence is indicated in bold.

The species-specific primers developed by Jeszkeet al.(2015) were used to detect all haplotypes ofD.destructorin this study (Table 2;Fig.7).The lengths of species-specific fragments ranged from 127 to 389 bp,and 9 length types of the species-specific fragments were found.Haplotypes A and H were the same length with differences of 3 nt.Haplotypes B and C were the same length with 24 nt differences.Haplotype K had 2 types of length fragments: 362 and 389 bp.Sequence alignment showed that these length differences of the species-specific fragment were the same as ITS1 sequences among all haplotypes.Therefore,using the species-specific primers combination: DITuni F and DITdes R proposed by Jeszkeet al.(2015) can quickly identify each haplotype forD.destructorin the present study.

Fig.7 PCR amplification results using ITS1-specific primers.M,marker;WYA5 and TCC2,haplotype H;WYA8,haplotype J;WYA2 and MXA2,haplotype N;DXP1,haplotype B;DXP2,haplotype C;HZA1,haplotype M;WYA10 and DTA1,haplotype L;WYA4 and WYA11,haplotype K;LXC1,haplotype I;WYA7 and TCA3,haplotype K.

Table 2 PCR amplification results using Ditylenchus destructor ITS1-specific primers

The digestion of the ITS-rDNA by four restrictive enzymes for 10 haplotypes ofD.destructorwithD.arachisas the control was shown in Fig.8,and the exact size of the digestion fragments were given in Table 3.ITS-RFLP results showed that a unique profile was found in each haplotype.The ITS haplotypes B,C,I,J,K,L,M and N,could be distinguished from each other by combination of four restriction enzymes,DdeI,HinfI,Tru9I (MseI),andSduI.However,the haplotypes A and H were the same in profiles and sequence fragments,and four types of profiles were observed in haplotype K.

Fig.8 PCR-RFLP of the ITS haplotypes for Ditylenchus destructor (A–L) andDitylenchus arachs (M).A,haplotypes A and H.B,haplotype J.C,haplotype N.D,haplotype B.E,haplotype C.F,haplotype M.G,haplotype L.H,haplotype K.I,haplotype K.J,haplotype I.K,haplotype K.L,haplotype K.M=marker;U=unrestricted PCR product,1=DdeI;2=HinfI;3=Tru9I (MseI);4=SduI.

Table 3 Approximate sizes of restriction fragments generated by virtual digestion by some diagnostic restriction enzymes for PCR products of the ITS-rDNA regions amplified for Ditylenchus destructor and Ditylenchus arachs

3.5.Phylogenetic relationships among D.destructor populations

The ITS-rDNA phylogenetic tree was constructed using Bayesian inference (BI) under the GTR+I+G evolutionary model (Fig.6).All seventy sequences of ITS-rDNA were clustered into two main clades: (1) all sequences of haplotype A clustered together into Clade a with maximal support (PP=100),without helix H9 in ITS1 sequences;(2) haplotypes B to N sequences clustered together into Clade b with moderate support (PP=87),with H9 in ITS1 except haplotype H.Clade b can be subclustered into 8 subclades: haplotypes F,I,M,H and N forming separate branches,haplotypes C,D,E and J clustered together,haplotypes B and G clustered together,and haplotypes L and K clustered together.

The D2–D3 of 28S rDNA phylogenetic tree was constructed using BI analysis under the GTR+I+G evolutionary model (Fig.9).Two main clades were observed on the D2–D3 tree: (1) the population SXSP1(ITS-rDNA haplotype A) and some sequences from sweet potato deposited in the GenBank clustered together(Clade a;PP=63);(2) the other populations and some sequences from potato and sweet potato which were clustered together (Clade b,PP=67).This clade can be subdivided into two subclades: one subclade was formed by MXA8 and HZA2,and the other was formed by LXC1,TCC1,ZXC1,MXC1,DXP1,DXP2,HLJP1,SXP1,DTA1,HZA3,TCA4 and ZXA2.The sequence alignment results showed that Clade a was different from Clade b withca.8 nucleotides,and 2 subclade sequences in Clade b were different with 3 nucleotides.These results seemed to correspond with 2 ITS types: haplotype A (Clade a) and other haplotypes (Clade b).

Fig.9 The 50% majority rule tree from Bayesian analysis generated from the 28S gene sequence dataset for Ditylenchus destructor using the GTR+G.The newly obtained sequence is indicated in bold.

4.Discussion

The extensive length variations of ITS-rDNA sequences ofD.destructorwere previously observed in several studies (Wanet al.2008;Mareket al.2010).The 6 length types of ITS-rDNA sequences using primers TW81/AB28 were reported by Subbotinet al.(2011).In the present study,we found 9 length types of ITS-rDNA sequences using primers TW81/AB28 and 7 of them were found in Chinese herbal medicine populations.Mareket al.(2010) found repetitive elements of 3 to 11 nt generally interspersed in ITS1 and pure tandem,and the longest repeated element (GCATTTGTGCT) was present with two copies situated in close mutual proximity.Subbotinet al.(2011) found that repetitive elements had consensus sequences: KCTRTGTRCYTGC and GCTYKYATTWGH,and the 12-nt element was inversely complemented to the 13-nt,forming a stable stem structure in the ITS1 secondary structure model.In this study,we found that the 12-and 13-nt elements also existed in the stable stem structure formed in ITS1,which is consistent with the report of Subbotinet al.(2011).We found that the 13-nt element had 4 motif types with a consensus sequence:GCTGTGT,the 12-nt element had 2 motif types with a consensus sequence: CTTGCATTTG in pure tandem.Minisatellites appeared only in ITS1 of several populations ofD.destructor;and this likely was a characteristic of the evolution of the ITS1 in some organisms,which made it as a good PCR marker for population genetic and biogeographical studies of certain species such asD.destructor(Mareket al.2010;Subbotinet al.2011).The repeat motifs of the ITS1 were found in several organisms such as dipterans and coleopterans (Paskewitzet al.1994),volvocales (Colemanet al.1998),ladybird beetles (von der Schulenburget al.2001),pinyon pines(Gernandtet al.2001),several fungi (Den Bakkeret al.2004),trematodes (Warberget al.2005),and mosquitoes(Boweret al.2009).The internal repeats constituted most of the sequenced length variations,and some results showed that these repeat motifs form stem-loop secondary structure in ITS as well (Subbotinet al.2011).Although,there were no evidences for any biological roles of the repetitive elements in the rRNA gene sequences,several studies showed that intra-individual variations were often due to internal repletion,the form of microsatellites,or longer repeated sequences,these repeat elements could be a highlight to the contribution to adaptive evolution (Biémont and Vieira 2006).However,determination was needed to the subsequent evolution of these repeated sequences within a species (Boweret al.2009).

Analysis of RNA secondary structures is an increasing application in assisting phylogenetic analysis through finding the best homology between nucleotides during alignment procedures and generation of optimal alignment (Subbotinet al.2005).Subbotinet al.(2011)found that the main variations were in helices H9 and H8 by using the secondary structures of ITS1 to improve alignment forD.destructor,and 6 types of H9 helices were found.In the present work,the same results were obtained.Also,both secondary structures of ITS1 and sequence variations showed that at least 6 new type helix H9 structures and 7 new haplotypes were observed in populations from Chinese herbal medicines.So far,12 helices H9 structures (Figs.3 and 5) and 14 have been discovered inD.destructor.Liet al.(2022) newly reported there were 14 haplotypes inD.destructorfrom different hosts in China and haplotype A was predominant haplotype.Interestingly,2 types of H8 helices (short and long) were observed in ITS1 secondary structures of haplotype K among different population sequences.

The results of ITS-RFLP and PCR amplification with specific primers also showed that these sequences had a significant variation,but this population had the same H9 structure.However,almost identical sequences were observed within other different population sequences of individual haplotypes.This is found inD.destructorfor the first time.

The putative ITS2 secondary structures reconstructed in this work agreed with those proposed forD.destructorby Subbotinet al.(2011),but there was a difference in the number of different sites.The ITS2 secondary structure constructed here was similar to those proposed forD.dipsaciby Subbotinet al.(2005) and forD.destructorby Mareket al.(2010).Although ITS2 sequences showed length and sequence variations in the evolution of eukaryotes,the secondary structure of ITS2 exhibits several standard features.Four domain models of the ITS2 have been observed in green algae (Mai and Coleman 1997),monogeneans (Morgan and Blair 1998),trichostongyloids (Chiltonet al.1998),mammals (Michotet al.1999),fungi (Josephet al.1999),Heteroderaspp.(Maet al.2008),and birch (Singewaret al.2020).However,the four domain structures of ITS2 were not all models appearing in eukaryotes (Wolfet al.2005).The report of Wolfet al.(2005) showed that onlyca.17.9%of ITS2 sequences had the 4-domain structure among 140 000 ITS2 sequences in GenBank.The research presented by Subbotinet al.(2005,2011) and our study showed that the ITS2 secondary structure was organized around a preserved central core from which 3 domains were formed in bothD.destructorandD.myceliophagus.However,the 5-domain structures were found in some nematodes of the genusContracaecumparasitizing fish and fish-eating birds (Maet al.2008).This result showed that the secondary structures of ITS2 sequence for nematode were more diverse than other organisms studied.It was known that there were 3 to 9 domains in ITS2 in different eukaryotes (Sunet al.2010).The ITS1 sequences were less conserved than those of ITS2,so the putative secondary structures of ITS1 were less consistent than those of ITS2,and an open-loop with several helices in ITS1 was observed in most eukaryotes(Sunet al.2010).Both research of Subbotinet al.(2011)and our study showed that three domains were observed in the ITS1 transcripts ofD.destructorand 4 domains inD.myceliophagus.

In this study,several complementary base changes(CBCs) and hemi-CBCs appeared in the models of ITS1 and ITS2 secondary structures,maintaining stability of structure.However,CBCs were usually observed in reproductively isolated species (Mülleret al.2007;Coleman 2009;Tarieievet al.2021).Hybridizing experiments with certain species from different hosts or geographic areas showed that only one-sided crosses were possible,but crosses failed in reciprocal combination between haplotypes A and B populations from sweet potato,the hybrid was possible between different populations within the same haplotype (Wanget al.2019),most hybrid offspring were weaker pathogenicity (Wanget al.2011b),ITS1 sequences and secondary structures of some offspring were different from their parents (Niet al.2021).This partial reproductive isolation could reveal thatD.destructormay presently be in the state of speciation.

The trends of increasing length and higher G+C contents were observed in the evolution of eukaryotic ITS regions (Goldmanet al.1983).In the present study,higher variations in both the length (727 to 969 bp) and G+C contents (46.48 to 47.16%) in ITS sequences were observed (Appendix A),but no correlation seemed to exist between the length and the G+C contents.The ITS inserts lead to the changes in sequence length and G+C contents,however,it is still unclear whether these changes affect the evolution.In previous studies,D.destructormight represent a complex of species or subspecies(Wanet al.2008;Wanget al.2019).Two main groups of ITS haplotypes (A and B–G) were observed in species separation experiments inD.destructor(Subbotinet al.2011;Wanget al.2011a).Subbotinet al.(2011)combined ITS secondary structures,ITS and D2–D3 tree among populations ofD.destructorand inferred two main groups: (1) with H9 helix in ITS1 sequences (haplotypes B–G);(2) without H9 helix in ITS1 sequences (haplotype A).In the present study,we also found two main groups:haplotypes B–N and haplotype A.Also,the populations(haplotype H) without H9 helix in ITS1 sequences were observed in the first group,which was different from previous studies.These results indicated that haplotype A as a unique race was different from other populations.Moreover,Huanget al.(2009) found differences in measurement of values c,V,V’ and tail lengths between haplotypes A and B.Furthermore,Xuet al.(2009) did not find any differences in morphological or morphometrical data between haplotypes A,B and F from different hosts.In addition,the research of Wanget al.(2009) showed that there were no correlations between the malate dehydrogenase phenotypes and the pathogenicity groups in ITS haplotypes (A and B).In the present study,the populations ofD.destructorfrom Chinese medicines were different from other populations from potato,and sweet potato regarding ITS sequence length and its secondary structures.Ditylenchusdestructorhas a wide host range and a global distribution.Although variations in sequences of ITS-rDNA and 28S-rDNA,pathogenicity and some biological characteristics among populations from different sources were distinct,the classification of the nematode species is still unclear.Further analysis of comprehensive molecular,morphological,pathogenicity and biological characteristics of populations from different sources should be conducted to define the classification boundaries belowD.destructorlevels.

5.Conclusion

In the present study,9 types of significant length variations in ITS sequences were observed among all collected populations ofD.destructor.The differences in ITS1 length were mainly caused by the presence of repetitive elements and can be characterized by a relatively high rate of base substitutions.Seven unique haplotypes were found in 36 populations collected from Chinese herbal medicinal plants compared with 7 haplotypes (A–G) of known system.These unique haplotypes were different from haplotypes A–G,and we named them as haplotypes H–N.The present results showed that 14 haplotypes(A–N) of ITS-rDNA have been found inD.destructor.Phylogenetic analyses of ITS-rDNA and D2–D3 showed that all populations were clustered into two major clades:one clade only containing haplotype A from sweet potato and the other containing haplotypes B–N from other plants.Taken together,D.destructoris rich in diversity in terms of molecules,especially in the populations of Chinese medicinal materials.These results will be promising for relevant research in nematode disease caused byD.destructorin Chinese medicinal materials.

Ackonwledgements

This work was supported by the National Natural Science Foundation of China (31760507) and the National Key R&D Program of China (2018YFC1706301).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendixassociated with this paper is available on http://https://doi.org/10.1016/j.jia.2022.08.126