Abstract: Background and Objective: Magnoliaceae is one of the arguable botanist families in classification by conventional taxonomic methods. DNA barcoding has been proposed as a powerful tool for identifying, confirming species and genetic relationship among species. Using DNA barcoding can overcome problems of morphological-based species identification. The objective of this study was to isolate, analyze and determine the DNA barcode regions for species identification and analysis of genetic relationship among some species of Magnolia family. Materials and Methods: The leaf specimens were collected and dried in situ with silica gel and then stored in -20°C freezer. Total DNA extraction and PCR amplification were performed with standard chemicals and Kit and these DNA fragments were sequenced by Sanger method. The data analysis was supported by software programs such as Mega6, BioEdit, GeneDoc. Results: In this study, four candidate DNA barcode sequences including matK, rbcL, trnH-psbA and ycf 1b were isolated and sequence analysed. The result showed that their size were 714, 553, 509-514 and 909-962 bp, respectively. The result of genetic distance analysis showed that only trnH-psbA fragment is best discrimination between four studied plant species. The combinations of ycf 1b+trnH-psbA, matK+trnH-psbA or matK+ycf 1b+trnH-psbA also were potential choices to be used as DNA barcode for these four species. The phylogenetic trees were built based on trnH-psbA sequences and the combinations above revealed the genetic relationship of these four species, in which Magnolia chevalieri (Dandy) V.S.Kumar was closer with Michelia braianensis Gagnep. than Michelia tonkinensis A.Chev and Michelia baillonii (Pierre) Finet and Gagnep. Conclusion: Among 4 candidate DNA barcodes in this study, the psbA-trnH region is the most efficient DNA barcode sequence. Furthermore the combinations (matK+trnH-psbA, rbcL+trnH-psbA, ycf 1b+trnH-psbA, matK+ycf 1b+trnH-psbA, matK+rbcL+ycf1b+trnH-psbA) could be used as the DNA barcode for species identification and analysis of genetic relationship among some species of Magnolia family.
INTRODUCTION
Magnoliaceae is a family in order Magnoliales, which has a large number of species. Some of them are valuable economic timber trees1. In this study, four plant species of Magnolia family were chosen including Magnolia chevalieri (Dandy) V.S.Kumar, Michelia tonkinensis A. Chev., Michelia baillonii (Pierre) Finet and Gagnep and Michelia braianensis Gagnep. The wood of these species hass good quality to be used in house furniture, interior decoration and handicraft production2. However, they are found very little in natural forest and these species are on the list of plants that need to be protected and conserved. Therefore, the study on classification, conservation and development of their genetic resources is essential. Previously, identification and classification of plants were mainly based on morphological methods3-6. Although there have been improvements in the application process, there are still many difficulties as in cases that the samples have been distorted (nature, colour changes), processed samples can not be identified. Recently, the development of DNA barcode technology application, which is an advanced method using the sequence of short DNA strands featured in genome of organisms to identify and distinguish species, brings high efficiency in a short time, contributes to improve the drawbacks of the previous method7-10.
In fact, the identification of some species belong to Magnolia family based on morphological indicator sometimes faces difficulties and get confusedness because of the similarity in their features. Therefore, the new classification methods need to applied to overcome these limitations11-15. In this study, conducted a selection of 4 candidate DNA barcode regions (matK, rbcL, trnH-psbA and ycf1b). They are all located in the chloroplast genome, in which the sequences have high conservation, suitable to be DNA barcode in plant16-19. Hence, the study on creation of DNA barcode sequence data used as indicative standard DNA molecule method in order to serve the identification species and the study of genetic relationship is necessary, contributes to natural resources management, national precious gene sources conservation and development.
MATERIALS AND METHODS
The study was conducted at Vietnam National University of Forestry (VNUF) and completed on July, 2017.
Materials
Plant materials: The leaf specimens were collected from 4 plant species from different populations. Three specimens were collected from 3 different individuals of each species. The all specimens were labeled with full information, dried in situ with silica gel and then stored in -20°C freezer until DNA extraction to avoid DNA degradation. Symbols of these samples were taken in accordance with their abbreviations and scientific names of species.
Chemical materials: The chemicals used to isolate the total DNA from leaf samples: Plant DNA Isolation Kit of Norgen, Canada, Chemicals for PCR cloning fragments of DNA barcode: Master mix of iNtRON Biotechnology, Korea, PCR Purification Kit of Norgen, Canada, Chemicals for electrophoresis on Agarose gel: Agarose, 1 kb DNA ladder, Redsafe nucleic acid staining solution provided by Norgen, Canada. The primers were designed for amplification of DNA barcode sequences as in Table 1.
Methods
Total DNA isolation: Total DNA was isolated from dried leaf samples of 4 plant species under the guidance of Plant DNA Isolation Kit, Norgen, Canada. Concentration, purity and integrity levels of total DNA are determined by spectrophotometric and electrophoresis methods on 1.0% agarose gel.
| Table 1: | List of primers to amplify the DNA barcode regions |
PCR amplification of candidate DNA barcodes: The 4 candidate DNA barcode fragments (matK, rbcL, trnH-psbA and ycf1b) were amplified by PCR technique on PCR 9700 Thermal Cycler Applied Biosystems (USA), each PCR reaction was performed in a total volume of 20 μL, including: H2O deion (7 μL), 2x PCR Master mix Solution (10 μL), 10 pmol μL–1 of forward primers (1.0 μL), 10 pmol μL–1 of reverse primers (1.0 μL) and 50 ng μL–1 of DNA template (1 μL). The PCR reaction program: 94°C in 3 min (94°C: 30 sec, 59°C: 30 sec, 72°C: 1 min) repeating 40 cycles, 72°C in 5 min, PCR product preservation was at 4°C. Primers annealing temperature in PCR reactions were different depended on primers. Each PCR reaction was repeated 3 times for each sample. The PCR results were tested by electrophoresis on 1.2% agarose gel and observed under ultraviolet light (UV). The PCR products were purified according to instructions of PCR Purification Kit of Norgen, Canada.
Sequencing of candidate DNA barcode regions: The amplified and purificated PCR products were sequenced by Sanger’s method, using Kit BigDye® Terminator v3.1 Cycle Sequencing. These sequences were processed and analyzed using specialized softwares such as Mega6, BioEdit. The completed sequences of these 4 candidate DNA barcode loci would be deposited on the Vietnam DNA Database Bank.
Genetic distance and phylogenetic analysis: After sequencing, the DNA barcode sequences would be aligned using the MUSCLE algorithm. The whole 4 DNA barcode sequences and their combinations were evaluated using 2 different methods, genetic distance and phylogenetic tree. The genetic distance was caculated based on Kimura 2-parameter method in MEGA6 software. Phylogenetic trees were constructed following Neighbor Joining estimation, Kimura 2-parameter model, boostrap value 1000.
RESULTS
Total DNA extraction: Total DNA extracted from dried leaves of Magnolia chevalieri (Dandy) V.S.Kumar, Michelia tonkinensis A. Chev., Michelia baillonii (Pierre) Finet and Gagnep. and Michelia braianensis Gagnep were tested by electrophoresis on 1.0% agarose gel to preliminarily evaluate the content and quality. The result of electrophoresis test showed that the all DNA bands were relatively sharp, with little breakage, it proves that the total DNA is quite intact (Fig. 1). The concentration and purity of the DNA solution were determined by spectrophotometric method at A260 nm and A280 nm wavelengths. The result indicated that the total DNA solution extracted from leaves of these 4 species were ready to amplify 4 DNA fragments.
PCR amplification: Total DNA products after extracting and diluting to appropriate concentration would be used directly as templates for amplification of matK, rbcL, trnH-psbA and ycf1b fragments using specific pairs of primers. The composition and reaction cycle are described in the method section. Each PCR reaction was repeated 3 times on each sample. PCR products were examined by electrophoresis on 1.2% agarose gel, using DNA marker 1 kb ladder (Fig. 2). Electrophoresis result showed that the corresponding DNA bands had the size as expected: 750, 600, 500 and 950 bp for matK, rbcL, trnH-psbA and ycf1b, respectively. The DNA bands were bold, sharp, with no byproducts proving that PCR products were specific amplified, could be purified and directly used in subsequent research steps.
DNA sequencing: The purified PCR product of 4 candidate DNA barcode fragments were proceeded to determine the nucleotide sequence. The result of sequencing analysis showed that matK, rbcL, trnH-psbA and ycf 1b fragments had the lengths of 714, 553, 509-514 and 909-962 bp, respectively.
| Fig. 1: | Electrophoresis of total DNA from 12 specimens of 4 species |
| The using high ranger 1 kb DNA ladder (Cat. 11900, Norgen, Canada) | |
| Fig. 2: | PCR result of cadidate DNA barcode regions |
The using high ranger 1 kb DNA ladder (Cat. 11900, Norgen, Canada) |
|
| Table 2: | Evaluation of four DNA barcode regions |
This result indicated that the size of these 4 DNA fragments were suitable to expected size and the bands size on electrophoresis result. Comparing DNA sequences in all 3 iterations, found no significantly difference with all 4 fragments. Though trnH-psbA was the shortest sequence, it had the most variable sites, followed by ycf1b, matK and rbcL (Table 2). The number of infomative sites was proportional to number of variable sites, with 4 in both trnH-psbA and ycf1b while none in matK and rbcL. The completed sequences after analysis were deposited on the Vietnam DNA Database Bank.
Genetic distances: The pairwise distances were calculated to estimate the discrimination of 4 DNA barcode loci and their combinations. Because there was no difference among 3 samples for each fragment in the same species, interspecific distance was measured instead of both intra-and inter-specific distance. According to evaluation result, the pairwise distance in these 4 loci ranged from 0-3.2%, the mean ranged from 0.09% (rbcL) to 2.23% (trnH-psbA). The fragment had the highest genetic distance was trnH-psbA, followed by ycf1b, matK and rbcL (Fig. 3, 4). This result was suitable with the number of variable sites in each sequence had shown above. In all 4 sequences, it was realized that trnH-psbA was the only one could discriminate all four species, which had minimum of interspecific distance was 0.79% while three others was 0%. It showed that trnH-psbA was a potential DNA locus to discriminate species in Magnolia family. We continued calculating the pairwise distance of these four fragments combinations and found out several possible candidates had the minimum interspecific distance greater than 0% such as matK+trnH-psbA (0.33%), rbcL+trnH-psbA (0.38%), ycf1b+trnH-psbA ( 0.28%) for combination of 2 sequences, or matK+ycf1b+trnH-psbA (0.19%) for 3 sequences (Table 3).
Phylogenetic tree analysis: To construct phylogenetic tree analysis, author had downloaded several sequences of other species in Magnoliaceae from GenBank (Table 4). Four phylogentic trees were built correspond to four DNA loci as matK, rbcL, trnH-psbA, ycf1b (Fig. 5). In general, these 4 species had the closed relationship with some other species like Magnolia officinalis, Magnolia tripetala, Magnolia yunnanensis. The result indicated that Magnolia chevalieri (Dandy) V.S. Kumar and Michelia braianensis Gagnep. were closed while Michelia tonkinensis A. Chev. and Michelia baillonii (Pierre) Finet and Gagnep. were further.
| Fig. 3: | Comparison of sequence characteristics |
| Fig. 4: | Genetic distance of four DNA barcode loci |
| Table 3: | Genetic distance of four DNA barcode loci and their combinations |
It will also be observed in phylogenetic trees built based on the potential combinations has shown above: matK+trnH-psbA, rbcL+trnH-psbA, ycf1b+trnH-psbA, matK+ycf1b+trnH-psbA and matK+rbcL+ ycf1b+trnH-psbA (Fig. 6).
DISCUSSION
The study was conducted on 4 species in Magnoliaceae family, including Magnolia chevalieri (Dandy) V.S. Kumar, Michelia tonkinensis A. Chev., Michelia baillonii (Pierre) Finet and Gagnep. and Michelia braianensis Gagnep. with 4 candidate DNA fragments matK, rbcL, trnH-psbA and ycf1b19-20. These are the valuable in classification and identification that used extensively in plant. The matK is one of the most potential segments to be DNA barcode for plant, which are useful in constructing the system of higher taxa like order or family. However, amplifying matK segments has got trouble because there are no available pair of primers can be used for all species21.
| Fig. 5(a-d): | Phylogenetic trees constructed based on sequence of four DNA barcode loci and sequences from GenBank, (a) matK, (b) rbcL, (c) trnH-psbA and (d) ycf1b |
To overcome this disadvantage, researchers recommended to use the core combination of rbcL+matK. rbcL is a fragment in chloroplast genome with high amplification and sequencing. Though its unremarkable discriminating power, rbcL is regarded as a standard segment in phylogenetic construction, especially at family and genus level. Along with rbcL, trnH-psbA also exhibits high PCR success reaches to over 90% by standard primers. This non-coding region is appreciated even when be used as single locus or combinated with others22. Apart from three fragments above, Dong’s19 research has proposed another powerful locus with 75% discrimination possibility which can be used to identification lower taxa. As the 2nd largest gene in chloroplast genome, this locus contains 2 region ycf1a and ycf1b, is the most variable plastid genome region and can be the core barcode for land plant19.
In this study, amplification using the specific pair of primer correspond to each candidate loci and sequencing got 100% successfully. However, the result of sequence analysis revealed that not all fragments were suitable to be DNA barcode for 4 species Magnolia chevalieri (Dandy) V.S. Kumar, Michelia tonkinensis A.Chev., Michelia baillonii (Pierre) Finet and Gagnep. and Michelia braianensis Gagnep. rbcL was the lowest discrimination power locus with only one variable site and the maximum genetic distance was 0.18%. Followed rbcL by matK with 9 variable sites, but this difference did not afford to discriminate four species. With low divergence in sequence, the combination of these two candidates as CBOL proposal did not bring a satisfactory result. Different from rbcL and matK, ycf1b and trnH-psbA showed the potential discrimination with the maximum genetic distance obtained 1.56 and 3.2%, respectively.
| Fig. 6(a-f): | Phylogenetic trees constructed based on several potential combinations of four DNA barcode loci, (a) trnH-psbA, (b) matK+trnH-psbA, (c) ycf1b+trnH-psbA, (d) rbcL+trnH-psbA ,(e) matK+ycf1b+trnH-psbA and (f) matK+rbcL+ycf1b+trnH-psbA |
| Table 4: | Reference DNA sequences from NCBI GenBank |
The Ycf 1b-the longest fragment in this study had seventeen variable sites and mean of genetic distance was 1.11% but the minimum was 0%, mean that it was disability to distinguish these species. Meanwhile, trnH-psbA the shortest fragment had the maximum divergence with genetic distance from 0.79-3.2%. This result revealed that not only discriminated successfully all four species in this study, trnH-psbA fragment but also performed the promised ability to identify other species in Magnoliaceae family by the safe distance in sequence. In fact, many researches have shown the similar result. In 1999, the study conducted by Azuma et al.11 with 25 Magnolia, two Michelia and two Liriodendron taxa, analyzed three DNA sequences as trnK intron (including matK coding region), trnH-psbA and atpB-rbcL indicated that trnH-psbA had the shortest fragment but exhibited the highest sequence divergence12. Although, at this period of time, the definition "DNA barcode" have not been proposed, the idea of using short DNA sequence was applied extensively. In Magnolia family, there were several other outstanding researches using this effective tool such as study of Azuma et al.11 indicated that trnH-psbA got the biggest rate of variable and informative charaters among three candidate matK, trnH-psbA and atpB-rbcL, Kim et al.13 research provided the new look in classification at genus level in Magnoliaceae based on ndhF sequence13. Another study of Kim has carried out with ten chloroplast regions from 48 taxa to address phylogenetic question in Magnolia family in 2013, including ndhF, rbcL, matK, ORF350, trnL intron, trnL-trnF, trnH-psbA, rbcL-atpB, trnK 5’ intron and trnK 3’ intron. In 2014, Yu Hua et al.15 published their study of Magnoliaceae species identification using seven candidate DNA region psbA-trnH, matK, rbcL, ITS, ITS2, rpoB and rpoC1. These researches have obtained several results, but they actually were not sufficient to ascertain the relationship within Magnolia family, required further studies to construct the most expedient phylogenesis of this family.
Research in classification and identification of Magnoliaceae were executed in the early of the 20th century, firstly by morphological method with research of Dandy et al. proposed 10 genara into 2 tribes1. The classification based on morphological marker had got lots of argument3-6, it was necessary to have another means to resolve this question. Therefore, method using molecular marker in general and DNA barcode in particular has attracted many researchers not only for family Magnolia family but also all organisms system. This study for 4 species Magnolia chevalieri (Dandy) V.S. Kumar Michelia braianensis Gagnep., Michelia tonkinensis A. Chev. and Michelia baillonii (Pierre) Finet and Gagnep. species that have not received much attention in previous researches has contributed in the campaign of building classification system for Magnolia family by DNA barcode. With outstanding advantage compare to traditional method that not require standard specimen, this technique is a powerful tool complement the morphological method in classification and identification, especially for complicated and controversal clusters.
CONCLUSION
Four candidate DNA barcodes (matK, rbcL, trnH-psbA and ycf1b) were successfully amplified and sequenced from 4 plant species of Magnolia family. The result of sequencing indicated that the lengths of matK, rbcL, trnH-psbA and ycf1b fragments were 714, 553, 509-514 and 909-962 bp, respectively. These nucleotide sequences have been registered on the Vietnam DNA Data Bank. The study result showed that among four candidates DNA barcodes were studied, the psbA-trnH region is the most efficient DNA barcode sequence with maximum genetic distance reached 3.2%. Furthermore, the combinations including matK+trnH-psbA, rbcL+trnH-psbA, ycf1b+trnH-psbA, matK+ycf1b+trnH-psbA, matK+rbcL+ycf1b+trnH-psbA could be used as the DNA barcode for species identification and analysis of genetic relationship among some species of Magnolia family.
SIGNIFICANCE STATEMENTS
This study develops DNA barcodes for selected species within Magnolia family. Results of this study are not only supported species identification and analysis of genetic relationship among some species of Magnolia family but also contributed to the conservation and commercialization of these important forest tree species.
ACKNOWLEDGMENT
This study was supported financially by the project "Creating the database of DNA barcode for some economic-value large timber trees, non-timber plant forest products" belonging to the agricultural biotechnology program of the Ministry of Agricultural and Rural Development, Vietnam. (Grant number: QD5515). The study was conducted in Vietnam National University of Forestry (VNUF), 2015-2017 period.