A feasible method to extract DNA from the cambium of high-canopy trees: from harvest to assessment

ABSTRACT Many tropical trees have high canopies and their leaves are not accessible. Thus, the use of tissue from a more accessible organ (cambium) for DNA extraction may be an alternative for molecular studies. We adapted a feasible methodology for extracting genomic DNA from cambium tissue harvested in the field for the assessment with PCR. We tested three storage conditions (two buffers and a silica gel) and four periods of time after harvest. We used previously described protocols and tested them on three species that occur in Amazonian forests and other biomes: Anadenanthera peregrina var. peregrina, Cedrela fissilis, and Ceiba speciosa. Our protocol obtained suitable PCR-grade genomic DNA for DNA sequencing and microsatellite genotyping. We recommend the use of silica for long-term storage and the buffer with ascorbic acid for short-term storage.


RESUMO
Muitas árvores tropicais possuem dossel alto e folhas não facilmente acessíveis. O uso de tecido de um órgão mais acessível (câmbio) para extração de DNA pode ser uma alternativa para estudos moleculares. Nós adaptamos uma metodologia viável para extrair DNA genômico de tecido cambial coletado no campo para avaliação com PCR. Testamos três condições de armazenamento (dois tampões e sílica gel) e quatro períodos após a coleta. Utilizamos protocolos descritos anteriormente e os testamos em três espécies encontradas em florestas amazônicas e outros biomas: Anadenanthera peregrina var. peregrina, Cedrela fissilis e Ceiba speciosa. Nosso protocolo foi eficaz na obtenção de DNA adequado para sequenciamento e genotipagem de microssatélites. Recomendamos o uso de sílica para armazenamento de longo prazo e o tampão com ácido ascórbico para curto prazo. PALAVRAS-CHAVE: ácido ascórbico; ditiotrietol; isolamento de DNA Molecular data has been helpful in revealing hidden aspects of the evolutionary history of plants (e.g., Hughes et al. 2013). The number of biodiversity studies using plant molecular data has increased over the last years (Vinson et al. 2018). Most of the traditional methods for DNA extraction from plants were designed to use leaves as the source of genomic DNA (e.g., Doyle 1990;Cota-Sánchez et al. 2006). However, many tropical trees have high canopies and their leaves are not readily accessible. Moreover, the vegetation of seasonally dry forests usually releases its leaves during the driest period (Oliveira-Filho and Fontes 2000). The use of commercial kits allows for the quick extraction of genomic DNA from leaves but adds cost to the analysis.
The use of an effective and inexpensive protocol that uses cambium tissue may be an alternative when leaves are not readily available. There are few studies that have used cambium tissue as a source of genomic DNA, but most of them adopted a time-consuming method using a commercial kit to achieve DNA extraction (e.g., Asif and Cannon 2005;Colpaert et al. 2005;Novaes et al. 2009;Lanes et al. 2013). Herein, we aimed to adapt a feasible methodology for extracting genomic DNA from cambium tissue, from the field sampling to the assessment of the DNA with PCR.
We sampled cambium tissue from three tree species that are native to the Amazon region and other Brazilian biomes: To collect a piece of cambium, we used a puncher (diameter = 1 cm; thickness = 0.2 cm; length = 16 cm) and hammered it into the tree trunk until it reached the wood fibrous layer. The extracted sample was cleaned with a piece of tissue paper and washed with 100% ethanol.
We collected three samples from each tree (nine samples per species). The first sample was kept in a dithiothreitol transport buffer ("DTT" treatment). The second sample was kept in an ascorbic acid transport buffer ("AA" treatment). Both samples were kept in 15-mL Falcon tubes wrapped in aluminum foil. The third sample was kept in an air-sealed plastic bag containing approximately 50 g of silica gel beads ("SIL" treatment). The DTT and AA buffers consisted of 2/3 absolute ethanol and 1/3 1× CTAB buffer (Colpaert et al. 2005). Instead of using DTT (or β-mercaptoethanol) and AA together in the transport buffer, as Colpaert et al. (2005) did, we tested them separately. For the DTT buffer, we added 3 mM dithiothreitol; to the AA buffer, we added 0.3% (w v -1 ) ascorbic acid. We did not use β-mercaptoethanol in these buffers. The DTT and AA buffers were kept at 4 ºC and the SIL was stored at room temperature (approximately 25 ºC).
Another novelty of our study was to test whether the storage conditions decreased the overall quality of the genomic DNA over time. Extractions were performed after four storage periods: "D0" (extraction performed on the same day of harvest); and "D7", "D14", and "D21" (extraction performed 7, 14, and 21 days after harvest, respectively). Extraction from the SIL samples was performed only after 14 and 21 days to allow for tissue dehydration. Prior to DNA extraction, the buffer-stored samples (DTT and AA) were washed with distilled water, sliced, and dried with tissue paper. The SIL samples were only sliced. Approximately 40 to 70 mg of each sample was placed into 2.0 mL microtubes together with two 3.2-mm chrome-steel beads per tube. The samples were then homogenized with a Bead Beater system (Mini-Beadbeater-24, BioSpec, Paulínia, SP, Brazil). We performed the genomic DNA extraction with three replicates following an eight-step protocol, based on Cota-Sánchez et al. (2006) (modified by Riahi et al. 2010), which was here applied for the first time for cambium tissue. The eight steps were applied as follows: 1. We preheated the CTAB buffer (2% CTAB, 100 mM Tris-HCl, pH 7.5, 1.4 M NaCl, 20 mM EDTA, pH 8.0; 4% PVP; adding 1% β-mercaptoethanol immediately prior to use) to 65 °C; 2. 800 μL of the hot CTAB buffer were added to the 2.0 mL microtube with the samples and beads; 3. The samples were pulverized with a bead beater: 3 min at 2,500 oscillations per minute for the silica-dried samples, and 5 min at 3,000 oscillations per minute for the bufferstored samples. This step was repeated if the grinding was not complete after the first round; 4. The tubes were incubated at 65 °C for 15 min with occasional swirling; 5. The samples were cooled on ice for 2 min and 750 μL CIA solution (chloroform:isoamyl alcohol; 24:1) were added to each tube followed by invertion 50 times. The tubes were centrifuged for 15 min at 10,000 rpm (Eppendorf centrifuge 5424) and the supernatant was transfered to a new microtube and the CIA step was repeated; 6. The supernatant was transferred to a new microtube, 0.7-volume ice-cold isopropanol was added and followed by gentle invertion 10 times and centrifugation for 15 min at 10,000 rpm. The supernatant was discarded without disturbing the pellet; 7. The pellet was washed by adding 1 mL 70% ethanol. The microtube was centrifuged for 10 min at 10,000 rpm and the ethanol was discarded. This step was repeated with 1 mL absolute ethanol. The pellets were then dried at room temperature, with care not to overdry them; 8. Each pellet was resuspended in 30 μL TE buffer (10 mM Tris-HCl, pH 8.0, 0.1 mM EDTA, pH 8.0), containing RNAse A (10 mg mL -1 ). The DNA sample was incubated at 37 °C for 30 min and then stored at 4 °C.
DNA quantification and quality control were performed with a NanoDrop Spectrophotometer (Thermo Scientific). We next tested whether the genomic DNA was suitable for PCR. The genomic DNA was amplified (the internal transcribed spacer, ITS) with the primer pair ITS4 (5'-TCCTCCGCTTATTGATATGC-3', White et al. 1990) and ITS Leu (5'-GTCCACTGAACCTTATCATTTAG-3', Baum et al. 1998). For the chloroplast gene, we amplified the intron of trnL (UAA) (CD) with the primer pair C (5'-CGAAATCGGTAGACGCTACG-3') and D (5'-GTTTACTTTTGGGCATGCTTCG-3') (trnL intron region, Taberlet et al. 1991). The PCRs were performed with three replicates, and the amplification (ITS or CD) was considered positive when a single, sharp band of the size expected was visible on a 1% (w v -1 ) agarose gel. We calculated the percentage of positive amplifications from each extraction period (D0, D7, D14, and D21) and each storage condition (DTT, AA, and SIL), for each species. We also used the microsatellite primers Acol15, Acol16 (Feres et al. 2012), Ced54, and Ced65 (Hernández et al. 2008) in order to test the suitability of the DNA. The PCR results were checked on 2% (wv -1 ) agarose gels. ANOVA and Tukey tests, implemented in RBio version 119 (Bhering 2017), were used to analyze the differences in the DNA concentrations and the percentages of positive PCRs among treatments within species.
The two storage conditions (DTT and AA) allowed for PCR amplifications that were not possible previously for frozen bark samples of A. peregrina (Novaes et al. 2009) ( Table  1). The SIL storage at room temperature (approximately 25 ºC instead of 4 ºC for the buffer samples) exhibited some DNA degradation, which was likely due to the temperature difference. However, several positive PCR amplifications were obtained ( Table 2). The SIL storage resulted in a higher amount of DNA compared to previous studies. For samples of A. peregrina var. peregrina stored in SIL, our methodology recovered genomic DNA (496 ng mL -1 ) with a yield higher than previously reported (364 ng mL -1 , Novaes et al. 2009). During field expeditions, silica gel is more amenable to handling than liquid buffers. Additionally, the use of silica gel requires less storage space in the laboratory when considering the size of the plastic bags and the tubes in the refrigerator. Despite some DNA degradation, we recommend SIL for long-term storage.
DNA extraction after different storage periods of cambium tissue has not been previously reported. There was no significant difference in the genomic DNA yield or quality among the four storage periods and none significantly impaired the use of the samples for PCR. This suggests that cambium tissue can be harvested and kept for some weeks prior to DNA extraction, which is frequently convenient in fieldwork schedules. We showed that buffers with either DTT or AA were suitable for extracting the DNA right after sample harvest. Additionally, the AA buffer was easier to prepare and carry into the field, and then be added to the remaining buffer ingredients as needed. The SIL-stored tissues were difficult to process during the first weeks. Nonetheless, the cambium tissues preserved in SIL storage were amenable to PCR four weeks after harvest (data not shown).
Anadenanthera peregrina exhibited higher genomic DNA yield than C. speciosa (Table 1). The high standard deviations might be due to the different sizes of the source tissue samples. Genomic DNA storage in DTT or AA exhibited sharp bands of a high-molecular mass, with little to no smears, suggesting high integrity. The A 260 /A 280 ratio values presented little variation among species. Most of the samples exhibited a A 260 /A 280 ratio greater than 1.8 (Figure 1), suggesting high purity. As a consequence, amplifications via PCR were successful. All tree samples were amplified for the    (Table 2) for the CD and ITS regions, followed by C. fissilis and C. speciosa. Sharp bands characterized most of the amplifications. We checked the PCR products for microsatellites and the presence of bands indicated its suitability for genotyping. To our knowledge, this is the first report on a multiplexing PCR system with microsatellites applied to cambium tissue DNA, with potential for application in studies of the genetic diversity of tree species. We recommend the use of silica for long-term storage and the ascorbic acid transport buffer for short-term storage.