Genetic diversity and phylogeography of Japanese brome (Bromus japonicus Thunb. ex Murr.) populations in China

Abstract

Japanese brome (Bromus japonicus Thunb. ex Murr.) is a pervasive annual weed with wide distribution in winter wheat fields across the North China Plain. In this study, we researched the genetic diversity and phylogeography of 24 B. japonicus populations in China based on total genomic DNA and chloroplast DNA. The results showed that 106 fragments were scored using 12 inter-simple sequence repeat (ISSR) primers, and 101 fragments (95.28%) were polymorphic. The findings revealed substantial genetic diversity and differentiation among populations from different locations (Ht = 0.2125, Hs = 0.0730, Gst = 0.6562 and Nm = 0.2619). Mean values of Nei’s genetic diversity (H) and Shannon index of diversity (I) were 0.0731 and 0.1068, respectively. In addition, 15 haplotypes were identified based on combined cpDNA regions among the 24 populations exhibiting abundant haplotype (gene) diversity and nucleotide diversity. The AMOVA based on ISSR and cpDNA both showed that genetic variation mainly exists among populations rather than within them. The STRUCTURE analysises of ISSR and cpDNA indicated that geographical location and genetic relationship had no significant correlation. The haplotype network also illustrated that the widespread haplotypes (H1, H2) might represent ancient polymorphism. The results obtained in this study demonstrate the presence of extensive genetic variability among and within the Chinese populations of B. japonicus, which is likely to contribute significantly to its adaptability and infestation as a weed species.

Keywords:

• Bromus japonicus
• inter-simple sequence repeat (ISSR)
• chloroplast DNA
• genetic diversity
• phylogeography

Introduction

Japanese brome, scientifically known as Bromus japonicus, is an annual winter weed that belongs to the Poaceae family. It is indigenous to Eurasia and is frequently encountered along roadsides, floodplain wetlands and agricultural areas such as wheat fields [1]. Typically, seedlings emerge during the months of September and October, flowering initiates in early May and seed dispersal, by early October [2]. This species possesses a strong adaptive capacity to environment [3]. A single B. japonicus plant can produce about 1885 seeds that are easily dispersed by water or wind due to their light weight. These seeds have the ability to germinate over a wide temperature range of 5–30 °C under varying pH levels, even in the absence of light [3]. B. japonicus could heavily infest wheat and may reduce yield by at least 30% in heavily infested fields [4]. Currently, postemergence treatment of herbicides have been widespread adoption for controlling B. japonicus, such as flucarbazone-sodium, pyroxsulam and mesosulfuron-methyl. Unfortunately, due to the extensive and persistent use, B. japonicus has developed a remarkable degree of resistance to these herbicides with target-site resistance in certain regions of China [5]. Owing to the changes in farming systems and the long-term use of herbicides, B. japonicus has become a cause of concern throughout the Huang-Huai-Hai Plain of China in recent years.

According to Fisher’s theory, the rate of spread into new environments by successful invaders is dependent on the level of additive genetic variation present in the invading population [6]. Studies of the genetic diversity in weed populations is crucial as it provides a fundamental understanding for subsequent targeted research, including the identification of dispersal routes, source populations of weeds, their adaptive abilities to varying environments, and the selective pressure exerted by herbicide usage [7–11]. Such background information is indispensable in order to manage existing populations effectively and to elaborate strategies for the prediction and mitigation of new populations in agricultural settings [12–14]. Molecular methods have provided the opportunity to explore not only genetic diversity but also the spread of weeds and have allowed us to theorize about the origin of weed introductions and the formation of hybrids [15,16].

Inter-simple sequence repeat (ISSR) is a dominant and multilocus molecular marker that does not require prior sequence information, thereby offering reliable assessments of genetic diversity. Consequently, it has been extensively utilized in plant population genetics [17,18]. As chloroplast DNA (cpDNA) is maternally inherited and non-recombining in the majority of angiosperms [19], it has found widespread application in phylogeographic studies and population genetics [20–22]. Chloroplast genomes are suitable for investigating the consequences of fragmentation due to limited effective population sizes and restricted seed-mediated gene dispersal. Chloroplast-specific markers, when polymorphism levels are sufficient, are invaluable in identifying genetic bottlenecks and founder effects, as well as assessing genetic drift [23].

The the genetic variability of B. japonicus biotypes has been insufficiently studied and thus, the genetic relations among these are largely unknown. In this study, the genetic diversity and phylogeography were assessed via the ISSR technique and cpDNA in order to understand some regional patterns of genetic variation and the underlying population structure in the Huang-Huai-Hai Plain of China. This is the first report to use ISSR markers and cpDNA in B. japonicus and analyse whether its genetic diversity shows the random or patterned spatial distribution.

Materials and methods

Sampling methods and DNA extraction

Seeds of B. japonicus were collected from eight provinces in China. A total of 24 populations were sampled and each population was separated by at least 100 km. Details of localities are provided in Table 1 and Supplemental Figure S1. For each population, seed was collected from 8 individuals at intervals of at least 20 m and stored in paper bags at room temperature until used in experiments. Fifty seeds were collected per sampled individual.

Table 1. Origins and geographical characteristics of studied B. japonicus populations.

Population	Sample code	Locality	Longitude (E)	Latitude (N)	Altitude (m)
Pop 1	BJ01	Beijing Fangshan	115°58′	39°35′	41
Pop 2	SD01	Shandong Yantai	120°23′	37°37′	23
Pop 3	SD02	Shandong Qingdao	119°47′	35°56′	6
Pop 4	SD08	Shandong Weihai	121°57′	37°10	55
Pop 5	SD09	Shandong Jinan	117°31′	36°44′	51
Pop 6	SD10	Shandong Heze	115°23′	35°10′	54
Pop 7	SD21	Shandong Dezhou	117°30′	37°42′	11
Pop 8	SD23	Shandong Linyi	118°24′	35°5′	70
Pop 9	JS01	Jiangsu Suqian	118°24′	33°42′	18
Pop 10	JS02	Jiangsu Taizhou	120°4′	32°38′	1
Pop 11	HB01	Hebei Xingtai	114°40′	37°0′	46
Pop 12	HB02	Hebei Shijiazhuang	114°39′	38°1′	62
Pop 13	HB04	Hebei Cangzhou	117°7′	37°58′	13
Pop 14	HN01	Henan Xuchang	114°7′	34°5′	64
Pop 15	HN03	Henan Shangqiu	115°51′	34°31′	45
Pop 16	HN04	Henan Luoyang	112°22′	34°45′	343
Pop 17	HN06	Henan Zhumadian	114°37′	33°1′	45
Pop 18	HN08	Henan Nanyang	112°39′	32°25′	114
Pop 19	SS01	Shanxi Xianyang	108°50′	34°31′	404
Pop 20	SX01	Shanxi Linfen	111°40′	36°15′	456
Pop 21	SX02	Shanxi Jinzhong	112°20′	37°20′	766
Pop 22	SX03	Shanxi Yuncheng	110°59′	35°6′	388
Pop 23	GS01	Gansu Tianshui	105°53′	34°34′	1095
Pop 24	GS02	Gansu Longnan	105°36′	33°20′	1300

Similar to the approach used in our previous study [5], the seeds were sown on 9 cm petri dishes lined with two layers of Whatman N°1 filter paper (Cytiva, Marlborough, Massachusetts, USA) moistened with 5 mL distilled water. The dishes were kept in controlled-environment growth chambers at 25 °C and 12 h photoperiod (Model RXZ, Ningbojiangnan Instrument Factory, Ningbo, China). After 4 d, germinated seeds were transplanted into 11 cm × 9 cm plastic pots filled with loam soil. The pots were watered as needed and transferred into a greenhouse with day/night temperatures set at 25 ± 5/20 ± 5 °C with a 12 h photoperiod. Eight plants were selected from eight individuals of each population for DNA extraction. Approximately 100 mg young leaf tissues were harvested from each selected plant at the 3- to 4-leaf stage for total genomic DNA extraction by EasyPure Plant Genomic DNA Kit (TransGen Biotech, Beijing, China).

PCR amplification and sequencing

Initially, a total of 56 ISSR primers were screened: 44 primers from the UBC primer set #9 (University of British Columbia, Canada) and 12 primers from GENEWIZ, China. Ultimately, these were reduced to 12 selected primers that exhibited clear and reproducible banding patterns for the final analysis (Supplemental Table S1). Similar to our previous study [5], ISSR–PCR amplifications were conducted in a total volume of 25 μL per sample, which contained 30 ng template DNA, 1 μL of each primer (10 μmol/L), 2.5 μL of 10 Trans Taq Hifi Buffer (Mg²⁺ Plus, TransGen Biotech, China), 2 μL of dNTP Mixture (2.5 mmol/L, TransGen Biotech, China), 1.25 U of Trans Taq DNA Hifi polymerase High Fidelity and 18.25 μL of ddH₂O. PCR was performed using a thermal cycler (T100, BIO-RAD, Hercules, California, USA) programmed using a program of one cycle of 94 °C for 7 min, followed by 40 cycles of 94 °C for 45 s, 52 to 59 °C for 1 min, 72 °C for 1.5 min, with a final elongation at 72 °C for 10 min. The amplification products were electrophoretic on 1.5% agarose gel buffered with 1 × TAE, stained with ethidium bromide, and digitally photographed under ultraviolet light. The fragment size was estimated by using a 2000 bp marker (TransGen Biotech, China).

Two chloroplast DNA (cpDNA) fragment sequences (trnT-trnL and atpI-atpH) were investigated to explore the phylogeographic patterns of B. japonicus as they showed the most polymorphic sites after a preliminary analysis. These fragments were amplified using the primer pairs in Supplemental Table S2. The two chloroplast regions were successfully amplified and sequenced across all individuals. The PCR was conducted in a 25-μL reaction volume, which comprised of 30 ng of template DNA, 0.5 μL of each primer (10 μmol/L), 2.5 μL of 10x Trans Taq Hifi Buffer (Mg²⁺ Plus, TransGen Biotech, China), 2 μL of dNTP Mixture (2.5 mmol/L, TransGen Biotech, China), 0.6 U of Trans Taq DNA Hifi polymerase High Fidelity, and 18.38 μL of ddH₂O. DNA amplification was programmed using a protocol of one cycle at 95 °C for 5 min, followed by 38 cycles of denaturation at 94 °C for 30 s, annealing at 52–57 °C for 30 s and extension at 72 °C for 1.5 min, with a final elongation step at 72 °C for 5 min. All PCR products were purified using an EasyPure Quick Gel Extraction Kit (TransGen Biotech) and sequenced by GENEWIZ (South Plainfield, New Jersey, USA).

Data analysis

Based on ISSR markers

The ISSR bands were interpreted as being dominant markers and were scored for presence (1) or absence (0) at a particular locus. This study followed the approach used previously [18]. The following parameters were calculated using software POPGENE32 version 1.32 [24] including the percentage of polymorphic loci (P), Nei’s genetic diversity (H), Shannon’s index of diversity (I), coefficient of gene differentiation (Gst) and gene flow (Nm) [25–27]. Analysis of molecular variation (AMOVA) was used to analyze the hierarchical genetic structure using Arlequin 3.5 [28], by which the partitioning of genetic diversity within and among populations was tested. Nei’s genetic distance and genetic identity between all pairs of populations resulting from POPGENE analysis served as a basis to construct the dendrogram by the UPGMA (unweighted pair group method with the arithmetic averaging) method using NTSYS-pc version 2.10. (Exeter Software, NY, USA) [29]. The genetic structure among populations was determined using the Bayesian program STRUCTURE v.2.3.4 software. The analysis involved setting the number of populations (K) ranging from 2 to 7, with a burn-in period of 10,000 steps and 50,000 replicates for each K. The optimal K value was estimated on the basis of ΔK as per Structure Harvester. A Mantel test for correlation of genetic and geographic distances between individuals within populations was then performed using PC-ORD 5.0 [30].

Based on the cpDNA sequences

The DNA sequences of the two cpDNA regions were aligned and combined by CLUSTALX [31]. To assess haplotype diversity (Hd) and nucleotide diversity (π), we utilized DnaSP version 5.1 [32]. A comprehensive haplotype frequency map was generated with the aid of ArcGIS 10.0. Furthermore, we analyzed two indices of population differentiation (Gst and Nst) and two measures of genetic diversity (Hs and Ht) using Permut version 2.0 [33]. Subsequently, we aimed to decipher the phylogenetic relationships among haplotypes of B. japonicus. Toward this end, a maximum parsimony analysis was conducted through the program PAUP 4.0b10 [34]. Genealogical relationships among chlorotypes genotypes were also explored using the median-joining network method as implemented in the program Network 5.0 [35]. We used the program Arlequin 3.5 to carry out ananalysis of molecular variance (AMOVA) [28], and thus to estimate genetic variation within populations, among populations within groups and between groups. A geographic distance matrix based on Euclidian distances was compared with a genetic distance matrix based on FST values between populations by Arlequin. The population genetic structure was analysed as described previously. To test for evidence of range expansion, Tajima’s D and Fu’s Fs statistics were calculated using DnaSP [36,37]. Then, we used a pairwise mismatch distribution to test for population expansion by DnaSP.

Results

Genetic diversity and variation

The number of loci for each ISSR ranged from six (UBC-811) to twelve (A1) with a mean of 8.83. Twelve ISSR primer pairs produced 106 scorable bands across the 192 individual samples from 24 populations. There were 101 polymorphic bands (95.28%) at the species level showing a higher discrimination power (Table S1).

The genetic diversity estimates for the 24 populations (Table 2) showed a percentage of polymorphic loci (P) ranging from 2.83 to 47.17% (mean 18.79%), observed number of alleles (Na) from 1.0283 to 1.4717 (mean 1.1879) and an effective number of alleles (Ne) ranging from 1.0165 to 1.3216 (mean 1.1305). The mean value of Nei’s genetic diversity (H) was 0.0731 and the Shannon index of diversity (I) 0.1068. The HN01 population showed the highest genetic diversity (Na = 1.4717, Ne = 1.3216, p = 47.17%, H = 0.1824, I = 0.2676), the HN06 population showed the second highest (Na = 1.4528, Ne = 1.3081, p = 45.28%, H = 0.1707, I = 0.2495), and the SD23 population presented the lowest (Na = 1.0283, Ne = 1.0165, p = 2.83%, H = 0.0095, I = 0.0143).

Table 2. Genetic diversity parameters among 12 ISSR markers for all 24 B. japonicus populations.

Population	N^a	Na^b	Ne^c	H^d	I^e	P^f
BJ01	8	1.0755	1.0573	0.0318	0.0459	7.55%
SD01	8	1.2358	1.1837	0.1009	0.1452	23.58%
SD02	8	1.0849	1.0578	0.0320	0.0469	8.49%
SD08	8	1.0849	1.0503	0.0289	0.0431	8.49%
SD09	8	1.1226	1.0915	0.0510	0.0740	12.26%
SD10	8	1.1132	1.0805	0.0458	0.0667	11.32%
SD21	8	1.2547	1.1634	0.0933	0.1382	25.47%
SD23	8	1.0283	1.0165	0.0095	0.0143	2.83%
JS01	8	1.0472	1.0314	0.0184	0.0272	4.72%
JS02	8	1.0566	1.0311	0.0191	0.0291	5.66%
HB01	8	1.1604	1.1100	0.0617	0.0904	16.04%
HB02	8	1.0943	1.0545	0.0324	0.0489	9.43%
HB04	8	1.2736	1.2329	0.1257	0.1784	27.36%
HN01	8	1.4717	1.3216	0.1824	0.2676	47.17%
HN03	8	1.1604	1.1161	0.0646	0.0940	16.04%
HN04	8	1.1415	1.0877	0.0508	0.0756	14.15%
HN06	8	1.4528	1.3081	0.1707	0.2495	45.28%
HN08	8	1.4057	1.2243	0.1293	0.1952	40.57%
SS01	8	1.0377	1.0152	0.0105	0.0169	3.77%
SX01	8	1.3679	1.2680	0.1496	0.2170	36.79%
SX02	8	1.3962	1.3400	0.1806	0.2560	39.62%
SX03	8	1.1698	1.1028	0.0586	0.0874	16.98%
GS01	8	1.2264	1.1788	0.0979	0.1404	22.64%
GS02	8	1.0472	1.0096	0.0078	0.0142	4.72%

^aN = number of samples; ^bNa = Observed number of alleles; ^cNe = Effective number of alleles; ^dH = Nei’s gene diversity; ^eI = Shannon’s Information index; ^fP = the percentage of polymorphic loci.

The analysis of the genetic diversity of 24 populations using POPGENE32 revealed a total gene diversity (Ht) of 0.2125. The within-population genetic diversity (Hs) was found to be 0.0730, indicating a coefficient of gene differentiation (Gst) of 0.6562. Furthermore, the gene flow (Nm) was calculated to be 0.2619. The AMOVA analysis using Arlequin determined that 63.72% of genetic variation was distributed among populations, while 36.28% resided within populations (Table 3). These results suggested that the genetic variation occurred mostly among populations.

Table 3. Analysis of molecular variance (AMOVA) of B. japonicus based on ISSR.

Source of variation	d.f.	SSD	MSD	Variance components	Ratio of variance	p-value
Among populations	23	1248.526	54.284	6.334	63.72	<.001
Within populations	168	605.875	3.606	3.606	36.28	<.001

Genetic structure and population relationships

Estimates of Nei’s genetic distance and genetic identity between all pairs of populations ranged from 0.0401 to 0.3524 and 0.7030 to 0.9607 (Supplemental Table S3). The highest genetic distance (0.3524) was observed between the GS01 and GS02 populations, while the least (0.0401) was found in the SD02 in relation to SD08. Although the spatial distance between GS01 and GS02 is short, the genetic distance is highest. The reason may be that the GS02 population is collected in Longnan and this site is in Qinling Mountains surrounded by mountains. The gene flow between them is lower and more difficult than between other population pairs because of the mountain barrier.

Cluster analysis was conducted to group the genotypes into a UPGMA dendrogram based on the genetic distance among populations (Figure 1). The 24 populations formed four major clusters at similarity coefficient levels of 0.155, with the similarity coefficient in the range of 0.13 − 0.35. The largest cluster, Cluster IV, consisted of 19 populations. Cluster I and II comprised one population each (Figure 1(a)). This result demonstrated the adaptability and acceptability of ISSR markers for the genetic diversity analysis among B. japonicus populations.

Figure 1. (a) UPGMA dendrogram of the analyzed populations generated from ISSR data and estimated according to the [26] formula. (b) Population-based cluster analysis using STRUCTURE for K = 4 based on ISSR data. Samples from east of China(cluster IV); samples from middle of China(cluster III); samples from west of China(cluster II); samples from west of China (cluster I). The distribution of the populations to different groups is indicated by color (G1: green, G2: red, G3: yellow, G4: blue).

In this study, the K-values of 2–7 were investigated and the maximum likelihood values showed that the most appropriate number of populations was four (K = 4). Structure cluster analysis (Figure 1(b)) showed that 24 populations were divided into four groups termed G1 (green), G2 (red), G3(yellow) and G4 (blue). The analysis of population structure did not reveal a definitive correlation with geographical distribution. Mantel tests were conducted using PC-ORD software for the correlations between genetic distances and geographic distances were estimated using 10,000 random permutations of matrix elements. A coefficient of r² = 0.2005 (p = .0606) was obtained, suggesting a non-significant correlation between them.

Haplotype variation and geographical distribution

The combined cpDNA sequences of trnT-trnL and atpI-atpH were 1226 bp in length. A total of fifteen haplotypes (H1-15) were identified in the 24 sampled populations, based on seven substitutions in these cpDNA regions (Supplemental Table S4). Haplotype frequencies in each population and geographic distribution are presented in Figure 2. Haplotypes H1 and H2 showed widespread distributions. Haplotype H1 mainly located in northeast with 16 populations and H2 mainly located in the southwest with 15 populations.

Figure 2. Geographical distribution in China of 15 cpDNA haplotypes in B. japonicus (Pie charts shows the different haplotypes with their proportions).

Genetic diversity estimates of cpDNA haplotypes are given in Supplemental Table S5. The haplotype diversity (Hd) ranged from 0 to 0.944 and the nucleotide diversity (π) ranged from 0 to 0.00168. The SD23 population showed the highest diversity (Hd = 0.944, π = 0.00168). Notably, BJ01, JS02, HB01, HN04, HN06 and GS01 populations had no variation and the haplotypes of H8, H9, H10, H11, H13, H14 and H15 occurred in only one population.

Population and phylogeographic analysis

The analysis of molecular variance (AMOVA) based on cpDNA (Table 4) showed higher genetic differentiation among all populations than within populations (Fst = 0.527, p < .001): 52.7% of the variation was among populations, and 47.3% of the variation was within populations. Total genetic diversity (Ht = 0.727) across all populations was higher than the average intrapopulation diversity (Hs = 0.486), and, consequently, the population differentiation was high, with Gst = 0.332 and Nst = 0.517. The permutation test showed that Nst was significantly higher than Gst (p < .001).

Table 4. The analysis of molecular variance (AMOVA) of B. japonicus based on two chloroplast DNA sequences.

Source of variation	SSD	Variance components	Ratio of variance	p-value
Among populations	100.777	0.453	52.721	<.001
Within populations	75.896	0.406	47.279	<.001

Phylogenetic relationships were reconstructed among cpDNA haplotypes of B. japonicus by MP coalescence analysis and three clusters were identified (Figure 3). Cluster I comprised one haplotype (H1); cluster II included seven haplotypes (H2, H3, H8, H9, H10, H11 and H15); cluster III included seven haplotypes (H4, H5, H6, H7, H12, H13 and H14). Based on the median joining network, two main groups were formed (Figure 4). Group A consists of H1, H2, H3, H8, H9, H10, H11 and H15. Group B consists of H4, H5, H6, H7, H12, H13 and H14. The haplotype network was largely consistent with that from the MP tree.

Figure 3. Divergence dating of fifteen cpDNA haplotypes in B. japonicus based on MP coalescence analysis.

Figure 4. Haplotype network of B. japonicus (Size of the circle is the proportional to the relative frequency of the haplotypes).

The STRUCTURE analysis of the cpDNA data revealed that the most likely number of groups was K = 3 (Figure 5). The 24 populations were divided into three groups termed G1, G2 and G3. The results of population structure analysis also did not definitively correlate with geographical distribution. This is consistent with the STRUCTURE analysis of the ISSR dataset.

Figure 5. Population-based cluster analysis using STRUCTURE for K = 3 based on cpDNA data.

The mismatch distributions analysis based on the cpDNA haplotypes data for all populations revealed a double-peak curve, suggestive of the absence of expansion among the entire species’ populations (Figure 6) [38]. Tajima’s D and Fu’s Fs values were also all non-significant, indicating that B. japonicus is a neutral evolution without genetic bottlenecks and natural selection.

Figure 6. Mismatch distribution established of B. japonicus (Thin line represents the observed mismatch distribution; the dotted line represents the expected mismatch distribution).

Discussion

B. japonicus has been one of the most difficult-to-control weeds in wheat fields, due to a strong adaptive capacity to environment [3]. For scientific weed management, the diversity level of weeds should be considered [39]. Here, 24 B. japonicus populations from a broad spatial scale within China were collected and genetic diversity and phylogeography were measured with ISSR technology and cpDNA. The results provide us novel insight into B. japonicus’s genetic diversity and population relationships. In this study, 95.28% of polymorphic sites, 0.717 of haplotype diversity and 0.00137 of nucleotide diversity was detected from 24 populations of B. japonicus. A higher degree of genetic diversity can confer higher fitness to the weeds by developing more resistance to diseases, pests, and herbicides [40–42]. When genetic differences lead to two or more individuals having a heritable differentiation and heterogeneity, strong selective factors like herbicides can select them among plant populations in a short time and over short distances. The tendency of B. japonicus to maintain high genetic diversity may indicate that existing resistant populations have a high potential for being selected by herbicides and become the dominant populations which seriously threatens the effectiveness of herbicides.

The AMOVA based on ISSR and cpDNA both showed that genetic variation mainly exists among populations, rather than within populations. Gst > 0.25 indicates high differentiation among populations [43]. The Gst (0.6562 based on ISSR data and 0.332 based on cpDNA) also showed a high level of genetic differentiation among B. japonicus populations. Gene flow can reflect the degree of genetic differentiation. High gene flow can reduce genetic differentiation and inbreeding decline among populations [44,45]. In this study, we found a limited gene flow among B. japonicus populations. According to Wright, low gene flow often leads to a tendency that the populations are homogenized [43,46]. At this level, gene flow is not strong enough to prevent population genetic differentiation [44]. In general, the species depending on gravity for seed dispersal have lower genetic differentiation among populations than the species depending on wind for seed dispersal [47]. This condition is in accordance with the bioecological characters of B. japonicus.

Both present and historical processes have contributed to the population structure [48]. The results of UPGMA showed that the B. japonicus populations of near geographical location were clustered together. Cluster III comprised the most populations, which could be attributed to the specific targeting of a larger number of repeated sequences within the centromeric region by ISSR markers, thereby significantly influencing the classification pattern [49]. But the STRUCTURE analysis of ISSR and cpDNA indicated that geographical location and genetic relationship had no significant correlation, this was supported by Mantel tests. A significantly higher Nst over Gst usually indicates the existing of phylogeographic structure and the studied populations are strongly differentiated genetically [33]. This result indicated that phylogeographic structure was significant present. Furthermore, it is postulated that populations exhibiting small genetic distances among them exhibit greater gene flow compared to those with substantial genetic distances [50]. Additionally, the haplotype network demonstrated that the prevalent haplotypes (H1, H2) might rather suggest ancient polymorphism than recent gene flow and expansion [51]. Notably, the haplotype network does not exhibit a star-like phylogeny, often indicating population expansion [23,52,53]. Upon conducting mismatch distribution analysis and neutrality tests on the cpDNA dataset, we failed to uncover any indication of population expansion.

Furthermore, the successful biological control is negatively correlated with the target species genetic diversity [54]. High genetic variation in weed species can provide an advantage against coevolving biotic factors and changes in the weed management systems [55]. Therefore, the underlying genetic diversity of the weed should be considered when determining the priorities for biological control [56].

There are a range of factors that collectively affect the structure of B. japonicus populations, for example, herbicide selection, genetic exchange among populations and various other elements, including disparities in agronomic practices such as crop rotations, fertilizer application and herbicide utilization, as suggested in studies on other species [57]. Our earlier studies investigated that B. japonicus has a wider ecological amplitude and its potential distribution is much larger than the area reported now [58]. The ISSR analysis conducted on different populations of B. japonicus revealed consistent genetic diversity, corroborating the findings from phylogenomic analyses. The obtained results shed light on the genetic structure and diversity of B. japonicus populations and may pave the way for further research into the genetic evolution of this weed species.

Conclusions

This study provides insights into the genetic diversity and phylogeography of B. japonicus at the species level, highlighting substantial genetic differentiation among populations. The STRUCTURE analysises of ISSR and cpDNA indicated that geographical location and genetic relationship had no significant correlation. The high level of genetic diversity could confer high evolutionary potential on adaptation to new environments. The obtained data about the phylogeography of B. japonicus could serve as a basis for future studies on the genetic basis of its evolution and adaptation to agricultural environments.

Authors contributions

Q. L. and Y. L. designed research; Q. L., J. Y. and P. B. performed research; Q. L., L. D. and W. Z. analyzed data, Q. L. wrote the paper; L. D. revised the paper; Q. L. and Y. L. acquired the funding. All authors have read and agreed to the published version of the manuscript.

Acknowledgments

The authors thank all the workers (Q. L., J. Y., P. B., W. Z. and Y. L.) from Tianjin Academy of Agricultural Sciences and L. D. from Shandong Peanut Research Institute for their assistance in conducting this research.

Disclosure statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be considered a conflict of interest.

Data availability statement

The data that support the findings of this study are available on request from the corresponding authors.

Additional information

Funding

This work was financially supported by Tianjin Natural Science Foundation (23JCQNJC00450), and Creative Research for Young Scientists of Tianjin Academy of Agricultural Sciences (China) (No. 2021017).