Relevance of genetic and active ingredient content differences in Leonurus japonicus Houtt from different origins

Leonurus japonicus Houtt. (Labiatae), a perennial herb, is used to treat cardiovascular, uterine, and gynecological diseases. In the present study, a phylogenetic tree was constructed based on the ITS + psbA-trnH + rbcL + rpoB concatenation sequence, and partial least squares-discriminant analysis (PLS-DA) was performed based on high-performance liquid chromatography. The phylogenetic tree and PLS-DA were combined to correlate genetic and chemical differences among L. japonicus derived from different origins. The results showed that the concatenation sequence could distinguish among L. japonicus from different origins. Moreover, chemical analysis revealed intergroup differences, but the results were not of su�ciently high quality as that of molecular phylogeny. Furthermore, the results of combined chemical and phylogenetic analyses suggested that differences in metabolites are in�uenced by not only genetic differences but also environmental factors. These results provide valuable information for the arti�cial cultivation of L. japonicus and new ideas for improving its quality.


Introduction
Leonurus japonicus Houtt.(Labiatae) is a perennial herb distributed in China, Japan, Korea, and other countries (Tan et al. 2020).In traditional Chinese medicine, L. japonicus is widely used for treating several gynecological diseases, such as dysmenorrhea, amenorrhea, and postpartum hemorrhage.Modern pharmacological studies have shown that the active compounds of L. japonicus exhibited various bioactivities in the treatment of brosis, cardiovascular diseases, cancers, uterine diseases, and brain injuries (Cheng et al. 2020;Li et al. 2020b).With extensive research on the medicinal value of L. japonicus, market demand for this herb as well as the requirement of high-quality L. japonicus have been increasing (Oliveira et al. 2017).The varying quality of L. japonicus derived from different origins produces varying therapeutic e cacy (Tan et al. 2020).Genetic factors such as intraspeci c differences may affect its quality.Therefore, it is necessary to explore the correlation between differences in the genetic factors and active ingredients of L. japonicus derived from different regions.
DNA markers are an effective method for analyzing phylogenetic relationships and evolutionary processes and the most commonly used DNA markers include ITS, matK, rpoB, rbcL, psbA-trnH, and others (Adolfo et al. 2022; Kurian et al. 2020;Li et al. 2022).As single DNA markers cannot distinguish genetic differences in closely related species, combined sequences (such as ITS + trnL-trnF (Feng et al. 2019), rbcL + matK + ITS2 (Valuyskikh et al. 2020) have been extensively used.However, few studies have used DNA markers in the phylogenetic analysis of L. japonicus.In the present study, the phylogenetic analysis method of the ITS + psbA-trnH + rbcL + rpoB combination sequence was used to analyze the similarities and differences in L. japonicus derived from different habitats.So far, > 280 secondary metabolites, including alkaloids, avonoids, diterpenoids, sesquiterpenoids, and other compounds, have been isolated and identi ed from L. japonicus (Miao et al. 2019).Among them, the main alkaloids are leonurine and stachydrine (Wang et al. 2017;Zhang et al. 2019), which are the measurement indicators in L. japonicus quality control.The content of leonurine and stachydrine was generally determined by highperformance liquid chromatography (HPLC).HPLC is characterized by fast analysis speed, high sensitivity, good separation e ciency, and low sample usage (Barzdina et al. 2022;He et al. 2022;Liu et al. 2021).It has been widely used in the ngerprinting of medicinal plants, content determination of the active ingredients, and species identi cation.
Owing to the complexity and speci city of traditional Chinese medicine, a single chemical analysis is not su ciently accurate to evaluate its quality and explain the in uence of genetic factors on its active ingredient contents.Therefore, the current study combined molecular phylogenetic analysis and HPLC to investigate the correlation between genetic and active ingredient content differences in L. japonicus derived from different origin, thereby producing a scienti c basis for quality evaluation.

Plant materials
A total of 27 fresh L. japonicus samples were collected and analyzed in this study from several different regions.Latitude, longitude and plant height were recorded for each sample (Table 1).The identi ed plants were divided into two parts: one sample was stored in liquid nitrogen for genetic analysis and the other was dried for chemical analysis.Genomic Dna Extraction, Pcr Ampli cation, And Dna Sequencing Genomic DNA was extracted using Trelief™ Hi-Plant Genomic DNA Kit (Tsingke Biotechnology Co., Ltd., Chengdu China).DNA was quanti ed via Spectrophotometer ND5000 (Bioteke Corporation Co., Ltd.) by measuring absorbance at 260 and 280 nm.The relative purity of the extracted DNA was estimated using 1% agarose gel stained with ethidium bromide.Genomic DNA was suspended in TE buffer and stored at − 20℃ until its use.
The PCR products were electrophoresed on a 1% agarose gel to con rm the speci c ampli cation.The successfully ampli ed samples were outsourced to Tsingke Biotechnology Co., Ltd.(Chengdu China) for bidirectional sequencing on ABI3730 Genetic Analyzer (Thermo Fisher Scienti c).

Phylogenetic Analysis
DNA sequences were entered into the NCBI database for initial alignment to determine sequence reliability.Multiple alignments of the nucleotide sequences were obtained using Clustal W in the MEGA 11.0 program, and the alignments of ITS, psbA-trnH, rbcL, and rpoB were analyzed both separately and in concatenation.The low-quality regions were removed, and manual correction and stitching were performed (Zhong et al. 2019).All new sequences obtained in this study were deposited in the NCBI database.All four sequences were used in phylogenetic analysis.A phylogenetic tree was constructed using the neighbor-joining (NJ) method with 1000 bootstrap values (Lee et al. 2022; Zheng et al. 2021), and genetic distances were calculated using the ρ-distance model as implemented in MEGA 11.0.The constructed phylogenetic tree was embellished using the Interactive Tree Of Life software.

Preparation of L. japonicus extracts
The sample powder of L. japonicus was precisely weighed 1 g and placed in an Erlenmeyer ask; 25 mL ethanol (70%) was added to the ask.Then, the samples were accurately weighed and processed via re ux extraction for 2 h.After cooling and adding 70% ethanol to compensate for the decrease in weight, the solution was shaken thoroughly and ltered through a microporous membrane lter (0.45 µm).The leonurine control product was weighed, and 70% ethanol was added to the product to make a 30-µg/mL solution.The stachydrine control product was weighed, and 70% ethanol was added to the product to make a 0.5-mg/mL solution.Prior to HPLC analysis, all samples were stored at 4°C.
The obtained dataset was then imported into the SIMCA 14.1 software (Umetrics, Umea, Sweden) for principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA).

PCR ampli cation and sequencing
A total of 108 DNA sequences were obtained after the ampli cation and sequencing of L. japonicus samples derived from 27 different regions.All samples were successfully ampli ed and sequenced from total DNA.The resulting sequences were compared with known sequences in the NCBI database, and the comparison results showed that the samples were homologous with 96.8-99.9% of the known sequences of L. japonicus in the NCBI database, indicating that the ampli cation and sequencing results were reliable.Sequence analysis showed that the length of different DNA marker sequences ranged from 312 to 698 bp and that the length of the concatenation sequence was 2124 bp.Among the sequences, the psbA-trnH sequence was the shortest at 312 bp and the rbcL sequence was the longest at 698 bp.In a previous study, rbcL and ITS could not be fully ampli ed using the genomic DNA of Nardostachys jatamansi (D.Don) DC (Wen et al. 2020).However, using the same primers and reaction conditions, rbcL and ITS markers were completely (100%) ampli ed in the genomic DNA of L. japonicus, indicating that the differences in ampli cation e ciency may be caused by interspecies differences.

Intraspeci c Genetic Distances And Data Distribution
ITS, psbA-trnH, rbcL, and rpoB had 15, 7, 13, and 3 variant sites, respectively.The genetic distances of ITS, psbA-trnH, and rpoB were mostly distributed in the range of 0.000000-0.003500(Fig. 1); these genetic distances were too small to distinguish the origins.Although genetic differences were most pronounced in rbcL, they did not su ciently cluster in the subsequent analysis.The distribution of genetic distance showed that the concatenation sequence had good genetic divergence.The intraspeci c distance of the concatenation sequence was 0.000000-0.007554.The maximum genetic distance between Guangshui, Hubei, and Yiyang, Jiangxi, was 0.007554, whereas the minimum genetic distances were 0.000000 between Dongchangfu, Shandong, and Shenxian, Shandong; Fushun, Sichuan, and Dujiangyan, Sichuan; and Linqing, Shandong, and Guanxian, Shandong.

Nj Phylogenetic Trees
A phylogenetic analysis of 27 samples of L. japonicus derived from 11 different provinces was performed using four DNA markers (Fig. 2).Based on the ρ-distance model, phylogenetic trees were constructed using the NJ method.The results showed that the phylogenetic trees of different DNA marker methods were different.The psbA-trnH and rpoB markers had low clustering ability and could not distinguish most of the origins, whereas ITS and rpoB had better clustering ability but could not su ciently cluster for Shandong, Guangdong, Henan, and Hebei.Although psbA-trnH (Li et  promiscuous.In the present study, the combination sequence could better distinguish L. japonicus species derived from different origins, and the results showed interspeci c genetic differences in the same species (Fig. 3).The concatenation sequence clustered L. japonicus of the same origin together, validating the method employed in this study.Therefore, it is important to use a concatenation sequence to cluster L. japonicus derived from different origins.
The phylogenetic tree showed that each sample was located on an independent branch, suggesting that the concatenation sequence could distinguish among the samples of different origins with more obvious genetic differences.From the phylogenetic tree analysis, samples from Anhui, Guangdong, Guangxi, Hebei, Henan, Hubei, Hunan, Shandong, and Sichuan were well clustered but those from Fujian and Jiangxi showed a close relation.It was hypothesized that as the Fujian and Jiangxi provinces are adjacent to each other and the habitats of L. japonicus in these regions are similar, genetic and evolutionary similarities occurred.Another possibility is that one of these two different provinces may have been introduced with the herb from the other, resulting in close relationship between the plants from different provinces.Overall, the concatenation sequence could provide a good clustering of L. japonicus derived from different regions.Inconsistencies in the evolutionary direction of species were shown to cause differences in the metabolite content (Korte 2021; Zivkovic et al. 2021).Genetic differences in L. japonicus may lead to differences in the primary active constituents, and further chemical analyses are needed to con rm this result.

Chemical Analysis Of The Main Alkaloids
Leonurine and stachydrine, the main active components of L. japonicus, are commonly used as markers to evaluate the quality (Wen et al. 2019).L. japonicus is widely used in the clinic as well as in daily life for treating gynecological diseases (Zhao et al. 2022b); however, the alkaloid content of L. japonicus derived from different regions varies.Therefore, the contents of leonurine and stachydrine were determined using HPLC to investigate the correlation between genetic and chemical content differences.The retention time and peak area of the standards were used as the criteria for leonurine and stachydrine content detection.
The HPLC chromatograms (Fig. 4) showed that the total run time for detecting leonurine was 25 min, and the compound peak was detected at a retention time of 4.687 min.The total run time for detecting stachydrine was 40 min, and the peak was detected at a retention time of 19.026 min.
PLS-DA, a well-known technique for feature extraction and discriminant analysis in chemometrics, is used to reveal important patterns related to physiological, genetic, and environmental issues.PLS-DA generates visual scatter plots for the qualitative evaluation of variability among multivariate data, and it is now widely used to assess differences among plant species at the metabolome level (Matsuse et al. 2022;Zhao et al. 2022a).The PLS-DA (R 2 X = 0.9998) was used to analyze the HPLC data.PLS-DA and PCA score plots were obtained for all samples, wherein each circle represented an independent sample (Fig. 5).The PCA and PLS-DA score plots have better stereoscopic results than the plane ones.The PCA score plot results show that the 27 samples could be divided into 3 groups with small differences.The PLS-DA results showed that the 27 samples could be divided into 7 groups, indicating differences in the content of active ingredients of L. japonicus derived from different origins.The within-group variation was greatest in Henan, and Sichuan and Henan were clearly separated from other areas.Differences between Guangdong and Anhui were small.The current study results showed that the alkaloid content in L. japonicus derived from different regions varied.It has been suggested that environmental factors and genetic differences can affect the synthesis and accumulation of metabolic components, even leading to different drug effects (Zhan et al. 2022).

Association Of Genetic And Main Active Constituent Analyses
A comparison of genetic and chemical analyses revealed that molecular phylogenetic clustering was not fully consistent with PLS-DA.Molecular phylogenetic analysis could distinguish each provenance, but PLS-DA could not distinguish all provenances.The present study results suggest that molecular phylogenetic analysis may be superior to chemical analysis in identifying and classifying L. japonicus derived from different origins.In molecular phylogenetic analysis, Guangshui, Hubei, and Yiyang, Jiangxi, were phylogenetically the most distantly related regions, but their chemical contents were less different, suggesting the role of environment in this nding.Previous studies have shown that environmental factors can activate different signaling pathways that involve the expression of genes related to the biosynthesis or accumulation of secondary metabolites (Jan et al. 2021).Environmental factors have a signi cant in uence on metabolite composition; metabolites are in uenced not only by genetic differences but also by environmental factors (Li et al. 2020a).The current study results showed that genetic analysis enables the identi cation of speci c taxonomies, but chemical analysis has some limitations in the identi cation of species or variants.In such cases, chemical analysis can provide additional information by identifying speci c metabolites (Zlatic and Stankovic 2017).Thus, a combination of genetic (molecular evolutionary analysis) and chemical analyses is important to ensure reliable results and avoid incorrect taxonomic identi cation, thereby improving the identi cation and quality control of L. japonicus species.

Figures
Figures

Table 1
al. 2019; Philippe et al. 2022) and rpoB (Aydin et al. 2022; Mousavi et al. 2022) are widely used to study traditional Chinese medicine and have good clustering abilities, psbA-trnH and rpoB played a minor role in distinguishing among the different origins of L. japonicus; this may be caused by interspeci c differences.By contrast, other DNA markers could not be clustered effectively and the order in the branches of the phylogenetic tree was