**3. Results and discussion**

#### **3.1 DNA isolation methodology**

Commercial DNA isolation kits are widely used for their single-step methods and the relatively short amount of time required (usually about 1-2 h). These kits have also proven effective for isolating DNA from common plants such as rice, barley and Arabidopsis. We first attempted to isolate DNA from the leaves of *R. soongorica* using three commercial DNA isolation kit: the Plant Genomic DNA Kit (TIANGEN Biotech Co., Ltd., Beijing) which is designed specifically to extract DNA from plant tissues rich in secondary metabolites. In our hands, this kit was not able to isolate any DNA from the leaves of *R. soongorica* (data not shown). Then, we carried out the improved kit method according to the suggestion of this company' technical assistance employee, which yielded a small amount of DNA, but it was seriously contaminated (Fig. 3, lanes 1,2). The failure of the kit may be explained by the DNA likely formed a sticky, a glue-like gel in complex with these secondary metabolites and this could not be properly separated into two phases by centrifugation.

During the course of the RNA isolation, none of the kits were able to isolate any RNA from the leaves of *R. soongorica* (Wang et al. 2011), so we did not try more other commercial DNA kits to isolate DNA, but attempted to use a improved CTAB method to extract DNA from

(Beijing) Co. Ltd.) with 3 μl of genomic DNA at room temperature for 1 min. Then the sample was loaded on 0.8% agarose formaldehyde denaturing gels stained with ethidium bromide (EtBr) (Biotechnology grade, Amresco), and run on gels in the 1× formaldehyde

Two microliters of each genomic DNA sample was diluted into 200 μl of sterilized ultra pure water (pH 7.0). The absorbance of each diluted genomic DNA sample was evaluated at 260 and 280 nm using a ND-2000C (Thermo, America). The yield of genomic DNA was calculated according to the formula: DNA yield = 50 × OD260 × dilution factor × volume of sample in milliliters/material weight (g). Measured the values at the wavelengths of 260,

The DNAs isolation from different *R. soongorica* populations by our promoted CTAB-B protocol were used as template for inter simple repeat sequence primers (ISSR) amplification (Gajera et al., 2010). ISSR amplification reactions were performed in 20 μl reaction volume containing 1 μl gDNA template, 0.25 mmol/L of each dNTPs, 2.5 mmol/L MgCl2, 1×PCR buffer (10mmol/L Tris-HCl pH8.3, 5mmol/L KCl), 1 U Taq DNA polymerase and 0.5 mmol/L of UBC-807 primer (AGA GAG AGA GAG AGA GT). The amplification reaction were carried out on a thermocycler (Biometra) and programmed for an initial pre-denaturing at 94°C for 5 min, followed by 35 cycles of 1 min at 94°C (denaturation), 1 min at 48°C (annealing temperature), and 1.5 min at 72°C (extension) followed by a final extension step at 72°C for 10 min. Amplification products (5 μl) were electrophoresed in 1.5% agarose in 1× TBE buffer and stained with ethidium bromide.

Commercial DNA isolation kits are widely used for their single-step methods and the relatively short amount of time required (usually about 1-2 h). These kits have also proven effective for isolating DNA from common plants such as rice, barley and Arabidopsis. We first attempted to isolate DNA from the leaves of *R. soongorica* using three commercial DNA isolation kit: the Plant Genomic DNA Kit (TIANGEN Biotech Co., Ltd., Beijing) which is designed specifically to extract DNA from plant tissues rich in secondary metabolites. In our hands, this kit was not able to isolate any DNA from the leaves of *R. soongorica* (data not shown). Then, we carried out the improved kit method according to the suggestion of this company' technical assistance employee, which yielded a small amount of DNA, but it was seriously contaminated (Fig. 3, lanes 1,2). The failure of the kit may be explained by the DNA likely formed a sticky, a glue-like gel in complex with these secondary metabolites and

During the course of the RNA isolation, none of the kits were able to isolate any RNA from the leaves of *R. soongorica* (Wang et al. 2011), so we did not try more other commercial DNA kits to isolate DNA, but attempted to use a improved CTAB method to extract DNA from

this could not be properly separated into two phases by centrifugation.

280 nm and 230 nm and calculated the ratios of A260/A280 and the A260/A230.

**2.6 Assessment of the purity and the yield of the genomic DNA** 

electrophoresis buffer at 5-7 V/cm.

**2.7 ISSR amplification** 

**3. Results and discussion 3.1 DNA isolation methodology**  the leaves of *R. soongorica*. When we carried out the modified CTAB-A method based on the classical Doyle and Doyle (1987) method, which consistently resulted in significant RNA contamination of the DNA samples (Fig. 3, lanes 3,4 ). To remove RNA contaminants, additional purification steps must be performed, which not only reduce DNA yield but also increase the time required for DNA extraction.

Fig. 3. Electrophoretic analysis of *R. soongorica* genomic DNA isolated using various extraction methods. The genomic DNA (3 μl) of each sample was loaded into the different well, and then was run on a 0.8% agarose gel stained with ethidium bromide in 1×TAE buffer. Marker indicates the Molecular weight marker – 1 kb DNA ladder. Lanes 1 and 2, the modified TianGen Plant Genomic DNA Kit; lanes 3 and 4, the modified CTAB-A method; lanes 5 and 6, our promoted CTAB-B method. As the TianGen Plant Genomic DNA Kit was able to isolate any DNA from the leaves, the photos were not shown in Fig. 3.

The present study was motivated by the need for better methods of extracting sufficient quantities of high-quality DNA from plant tissue rich in secondary metabolites for use in molecular marker assays. The promoted CTAB-B protocol described here efficiently eliminates most of the interfering molecules, including polyphenols, polysaccharides, and proteins, and it yields translucent and water-soluble DNA pellets without RNA contamination. The main protocols made in this method were grounding PVPP together with the plant material, an increase in the volume of high salt extraction buffer, adding 3 mol/L sodium acetate (NaAc) in extraction buffer, dissolving the crude pellet in 1 M NaCl followed by RNase treatment, the purification of acid phenol extraction (phenol: chloroform: isoamylal alcohol (PCI)=25:24:1) and the use of pre-cooled ethanol and sodium acetate in precipitation and all these modifications helped to remove the interference of secondary metabolites in the DNA isolation. PVPP was sprinkled directly onto the frozen fresh leaf tissue in the mortar and vigorously ground with the leaf tissue in the presence of liquid nitrogen, which can avoid the oxidation of released polyphenols into quinines, which in turn bind to nucleic acids and hinder the isolation of high quality DNA. Increase in the volume of extraction buffer can completely break down the cell walls and make more nucleotide acid released resulted in increasing the yield of DNA isolation. Krizman et al. (2006) postulated that the plant tissue amount per volume of extraction buffer has an effect on DNA quality and yield. Since the extraction buffer is responsible for the lysis of membranes and liberation of DNA from cellular organelles (Weising et al., 2005), the smaller the quantity of plant tissue per unit volume, the more optimal the lysis process. Striking the correct balance between plant tissue amount and extraction buffer volume would reduce the probability of co-precipitation of contaminants with the DNA pellet as the saturation concentration during precipitation is less likely reached or exceeded (Krizman et al., 2006). During the extraction, 3 mol/L sodium acetate (NaAc) added combined with chloroform/isoamyl alcohol extraction can reduce markedly the coprecipitation of polysaccharides with the nucleic acids and remove most proteins, polysaccharides, polyphenols and other impurities for the first time. The crude nucleic pellet was dissolved in 1 ml of 1 M NaCl instead of dissolving it in Tris-EDTA (TE), which ensured further reduction of viscosity of the mucilaginous substances (Chen and Chen, 2004; Ghosh et al., 2009). DNase-free RNase was added to crude DNA samples dissolved in1 ml of 1 M NaCl to completely clear residual RNA. After RNase treatment, the DNA solution requires purification with an acid-phenol: chloroform : isoamyl alcohol (25: 24: 1) extraction because small amounts of protein in DNA pellets and salts in the RNase reaction buffer and stop solution both influence downstream molecular procedures such as restriction endonuclease digestion, ISSR-PCR amplification and full the genomic sequencing. Thus, we used acidphenol to remove residual protein and the remaining salts after the RNase treatment. Finally, it is necessary to precipitate DNA simultaneously with sodium acetate (pH 5.2) and absolute ethanol which can completely remove the residual polysaccharides from the DNA sample resulted in increasing the yield of DNA isolated.

#### **3.2 Assessment of the quality and quantity of the total DNA**

The success of an DNA isolation protocol may be judged by the quality and quantity of DNA recovered. The quality of DNA was assessed by gel electrphoresis, spectrophotometry, restriction endonuclease digestion and PCR amplification. The mean yield of DNA extracted by our promoted method was approximately 60.29 ± 20.16 μg/100mg of fresh leaves, which was higher than that of the modified CTAB-A method (35.72 ± 15.41 μg/100mg) and the modified TianGen Plant Genomic DNA Kit (20.54 ± 8.43 μg/100mg) (Table 1). The DNA isolated by our promoted method also exhibited good purity. DNA absorbs UV light maximally at 260 nm, whereas protein absorbs at 280 nm and other contaminants including carbohydrates, phenol, and aromatic compounds generally absorb around 230 nm. Therefore, the A260/A280 and the A260/A230 ratios are often used as indicators of DNA sample purity. Generally, ratio values of A260/A280 in the range of 1.8–2.0 indicate highpurity DNA; the ratio values of A260/A280 less than 1.8 indicate protein contamination in DNA samples; the ratio values of A260/A280 more than 2.0 indicate much RNA or many DNA fragments in DNA samples. With our method, the A260/A280 and A260/A230 ratios were 1.86 ± 0.16 and 1.92 ± 0.13, respectively, indicating that the DNA was free of protein and polysaccharides/polyphenol contamination (Table 1). In addition, there were no other bands visible in the bands (Fig. 1, lanes 5,6), indicating that the DNA was free of genomic RNA contamination. In contrast, the DNA from the CTAB-A method had poor purity as assessed by A260/A280 (2.12 ± 0.18) and the A260/A230 ratios (1.55 ± 0.36 μg/100mg), and

quinines, which in turn bind to nucleic acids and hinder the isolation of high quality DNA. Increase in the volume of extraction buffer can completely break down the cell walls and make more nucleotide acid released resulted in increasing the yield of DNA isolation. Krizman et al. (2006) postulated that the plant tissue amount per volume of extraction buffer has an effect on DNA quality and yield. Since the extraction buffer is responsible for the lysis of membranes and liberation of DNA from cellular organelles (Weising et al., 2005), the smaller the quantity of plant tissue per unit volume, the more optimal the lysis process. Striking the correct balance between plant tissue amount and extraction buffer volume would reduce the probability of co-precipitation of contaminants with the DNA pellet as the saturation concentration during precipitation is less likely reached or exceeded (Krizman et al., 2006). During the extraction, 3 mol/L sodium acetate (NaAc) added combined with chloroform/isoamyl alcohol extraction can reduce markedly the coprecipitation of polysaccharides with the nucleic acids and remove most proteins, polysaccharides, polyphenols and other impurities for the first time. The crude nucleic pellet was dissolved in 1 ml of 1 M NaCl instead of dissolving it in Tris-EDTA (TE), which ensured further reduction of viscosity of the mucilaginous substances (Chen and Chen, 2004; Ghosh et al., 2009). DNase-free RNase was added to crude DNA samples dissolved in1 ml of 1 M NaCl to completely clear residual RNA. After RNase treatment, the DNA solution requires purification with an acid-phenol: chloroform : isoamyl alcohol (25: 24: 1) extraction because small amounts of protein in DNA pellets and salts in the RNase reaction buffer and stop solution both influence downstream molecular procedures such as restriction endonuclease digestion, ISSR-PCR amplification and full the genomic sequencing. Thus, we used acidphenol to remove residual protein and the remaining salts after the RNase treatment. Finally, it is necessary to precipitate DNA simultaneously with sodium acetate (pH 5.2) and absolute ethanol which can completely remove the residual polysaccharides from the DNA

sample resulted in increasing the yield of DNA isolated.

**3.2 Assessment of the quality and quantity of the total DNA** 

The success of an DNA isolation protocol may be judged by the quality and quantity of DNA recovered. The quality of DNA was assessed by gel electrphoresis, spectrophotometry, restriction endonuclease digestion and PCR amplification. The mean yield of DNA extracted by our promoted method was approximately 60.29 ± 20.16 μg/100mg of fresh leaves, which was higher than that of the modified CTAB-A method (35.72 ± 15.41 μg/100mg) and the modified TianGen Plant Genomic DNA Kit (20.54 ± 8.43 μg/100mg) (Table 1). The DNA isolated by our promoted method also exhibited good purity. DNA absorbs UV light maximally at 260 nm, whereas protein absorbs at 280 nm and other contaminants including carbohydrates, phenol, and aromatic compounds generally absorb around 230 nm. Therefore, the A260/A280 and the A260/A230 ratios are often used as indicators of DNA sample purity. Generally, ratio values of A260/A280 in the range of 1.8–2.0 indicate highpurity DNA; the ratio values of A260/A280 less than 1.8 indicate protein contamination in DNA samples; the ratio values of A260/A280 more than 2.0 indicate much RNA or many DNA fragments in DNA samples. With our method, the A260/A280 and A260/A230 ratios were 1.86 ± 0.16 and 1.92 ± 0.13, respectively, indicating that the DNA was free of protein and polysaccharides/polyphenol contamination (Table 1). In addition, there were no other bands visible in the bands (Fig. 1, lanes 5,6), indicating that the DNA was free of genomic RNA contamination. In contrast, the DNA from the CTAB-A method had poor purity as assessed by A260/A280 (2.12 ± 0.18) and the A260/A230 ratios (1.55 ± 0.36 μg/100mg), and the yield was lower than that obtained with our promoted CTB-B method and the DNA was badly contaminated with RNA because there were two other visible bands on the lane (Fig. 1, lanes 3,4; Table 1). The TianGen Plant Genomic DNA Kit was not able to isolate any DNA from the leaves of *R. soongorica*. Furthermore, even the improved kit method only yielded a small amount of DNA (20.54 ± 8.43 μg/100mg), and it was seriously contaminated (Fig. 1, lanes 1,2 ). The above results show that the improved CTAB-B protocol described herein efficiently eliminates most of the interfering molecules (including polyphenols, polysaccharides, proteins and salts), and it also provides a higher yield of DNA pellets that are translucent, water-soluble and lack RNA contamination, indicating that it is superior to the CTAB-A method and the commercial kits.


Table 1. The genomic DNA purity and yield in *R. soongorica* leaves by different methods

Fig. 4. ISSR-PCR profiles of the genomic DNAs isolation from different *R. soongorica* populations by our promoted CTAB-B protocol using the UBC-807 primer were analyzed on a 1.5% agarose gel stained with ethidium bromide in 1×TAE buffer. "M" represents the Molecular weight marker – DNA marker DL2000.

The suitability of extracted DNA for downstream molecular processes was further verified by molecular markers ISSR-PCR amplification. AS shown in Fig. 4, the genomic DNA of five different *R. soongorica* populations were highly amplifiable by ISSR-PCR as indicated by the amplification products resolved on 1.5% agarose gel. This further confirmed the purity of DNA, free of polysaccharide and polyphenol contamination, which would otherwise inhibit Taq DNA polymerase and restriction endonucleases (Ahmad et al., 2004). Plant molecular applications such as RAPD and AFLP necessitate the successful isolation of high quality DNA (Michiels et al., 2003; Ahmad et al., 2004), devoid of contaminants. Without high quality DNA such downstream molecular manipulations are not feasible (Varma et al., 2007). To confirm the applicability of our method, this DNA extraction method has also been found to be efficient in other desert plants, including *Tamarix ramosissima*, *Nitraria tangutorum* and *Caragana korshinskii* Kom. (data not shown).
