Isolation of High Molecular Weight DNA Suitable for BAC Library Construction from Woody Perennial Soft-Fruit Species

Abstract

We have developed a novel nuclei extraction method that allows for the extraction of high molecular weight DNA from leaves of woody perennial soft-fruit species that contain high levels of carbohydrates and polyphenolics. The method utilizes a modified buffer system including 4% (w/v) polyvinylpyrrolidone (PVP)-10 and a combination of nylon filters and Percoll™ gradients to purify nuclei extracts prior to embedding in agarose plugs. The effectiveness of the method was demonstrated on leaves of red raspberry (Rubus idaeus) and blackcurrant (Ribes nigrum), two soft-fruit species that have shown to be recalcitrant to standard genomic DNA extraction methods. Extracted DNA was readily digested by restriction enzymes and, as shown for raspberry, suitable for bacterial artificial chromosome (BAC) library construction.

Introduction

The isolation of high molecular weight DNA (HMW-DNA), either in the form of embedded protoplasts or nuclei, is one of the most challenging steps required for the construction of plant large-insert genomic libraries. Difficulties arise from the presence of plant cell walls, chloroplasts, excess amounts of carbohydrates, and other secondary compounds such as polyphenolics. If not removed or destroyed, isolated HMW-DNA is not readily accessible, is contaminated with organelle DNA, or trapped and covalently bound to oxidizing polyphenolic substances, which render the DNA useless for most molecular applications (1). Whereas numerous buffer systems and HMW-DNA preparation methods have been shown to be applicable to some plant species, biochemical and morphological diversity within the plant kingdom are obstacles to the development of one single universal protocol.

In this study, we have developed a novel method for preparing HMW-DNA, suitable for restriction enzyme digestion and bacterial artificial chromosome (BAC) library construction from woody, fruit-bearing, perennial plant species, such as red raspberry (Rubus idaeus) and blackcurrant (Ribes nigrum). These soft-fruit species contain high levels of carbohydrates, particularly polysaccharides, and polyphenolic compounds and require heavily modified methods for standard genomic DNA isolations (2) and the utilization of activated charcoal in tissue culture to prevent growth inhibition due to excess polyphenolics released into the medium (3). The method presented here is based on a modified buffer system described by Peterson et al. (4) and utilizes a combination of nylon filters and Percoll™ gradients (Amersham Biosciences, Uppsala, Sweden) to purify nuclei extracts prior to embedding in agarose plugs.

Materials and methods

All reagents used were purchased from Sigma (St. Louis, MO, USA) except for Miracloth™ (Calbiochem, La Jolla, CA, USA), Percoll, nylon meshes (Wilson Sieves, Nottingham, UK), λ ladder PFG marker (New England Biolabs, Beverly, MA, USA), and InCert^® agarose (BMA, Rockland, MA, USA). All steps of the nuclei extraction protocol were conducted at 4°C and used sterilized solutions and equipment. Centrifugation steps were carried out at 650× g for 20 min. Nuclei solutions were handled with cut-off, wide-bore pipet tips to avoid shearing of the HMW-DNA.

Tissue Homogenization, Filtering, and Nuclei Collection

Very young, only partially unfolded leaves were harvested and frozen from plants, which were dark-treated for 3 days to reduce the amount of stored polysaccharides. For the procedure, 5–20 g frozen plant material were ground in liquid nitrogen to a very fine powder and transferred into 350 mL MEB-buffer, pH 6.0 [1 M MPD (2-methyl-2,4-pentanediol), 10 mM PIPES KOH, 10 mM MgCl₂, 4% (w/v) polyvinylpyrrolidone (PVP)-10, 10 mM sodium metabisulfite, 0.2% (v/v) β-mercaptoethanol, 0.5% (w/v) sodium diethyldithiocarbamate, 0.5% (v/v) Triton^® X-100, and HCl for pH adjustment]. The homogenate was incubated for 12 min and gently stirred every 2 min before filtering through one layer of Miracloth and then again through two layers of Miracloth. The cleared solution was transferred in small quantities to a stack of one 40-µm nylon mesh on top of one 20-µm nylon mesh and allowed to filter through by gravity and gentle agitation to prevent clotting prior to successive loading of the remaining sample volume. Typically, at this stage, a white precipitate formed on both the 40- and the 20-µm nylon mesh and turned brown within hours, suggesting that it contained carbohydrates and polyphenolics (Figure 1). Nuclei were collected by centrifugation and resuspended in 15 mL MPDB-buffer, pH 7.0 [1 M MPD, 10 mM PIPES KOH, 10 mM MgCl₂, 10 mM sodium metabisulfite, 0.2% (v/v) β-mercaptoethanol, 0.5% (w/v) sodium diethyldithiocarbamate, 0.5% (v/v) Triton X-100, and NaOH for pH adjustment].

Figure 1. Forty-micron nylon membrane (A) before and (B) after filtration of nuclei homogenate.

The brown staining, typically observed after filtration of raspberry or blackcurrant nuclei suspension, most likely consists of withheld polysaccharide and polyphenolics, which are responsible for the coloration.

Nuclei Purification

The isolated nuclei were layered on top of a discontinuous Percoll gradient consisting of 95%, 60%, 45%, 30%, and 15% Percoll diluted with MPDB to give a total volume of 5 mL per gradient step. Following centrifugation, nuclei were harvested from the interphases formed below the 30% Percoll gradient step, pooled, and diluted with 15 mL MPDB solution. The nuclei suspension was loaded on top of 10 mL 85% Percoll diluted with MPDB, centrifuged, and harvested from one single band that formed between the Percoll/MPDB-buffer. After the dilution of the nuclei with 20 mL MPDB, a final Percoll-based centrifugation was implemented by loading the solution on top of 20 mL 37.5% Percoll, diluted with MPDB. The supernatant was discarded, and the nuclei, concentrated at the bottom of the centrifugation tube, were resuspended in a total volume of 25 mL MPDB-buffer. This solution was again centrifuged, and the supernatant was discarded.

Nuclei Embedding and Downstream Applications

To prepare the nuclei for embedding in agarose plugs, a final centrifugation step was carried out that involved resuspending the nuclei in 20 mL MPDB-buffer without β-mercaptoethanol or Triton X-100. The resulting buffer is referred to as MPDB(-). Nuclei were collected by centrifugation, resuspended in 1-2 mL MPDB(-), and embedded in 1.5% InCert agarose prepared with MPDB(-) as a buffer. Nuclei embedding and all downstream treatments of HMW-DNA required for restriction enzyme digest and BAC library construction (data not shown) were performed as described by Chalhoub et al. (5).

Results and discussion

A comparison between the newly developed nuclei extraction method and other methods, such as that described by Zhang et al. (6), is shown in Figure 2. The method from Zhang et al. has been well established in our laboratory and successfully used to extract nuclei and clone HMW-DNA from mutant Arabidopsis thaliana (M.E. Gilroy, G. Loake, P. Birch, and I. Hein, unpublished results) and the barley (Hordeum vulgare) cultivar Golden Promise (I. Hein, H. Liu, and R. Waugh, unpublished results). However, as shown in Figure 2, embedded raspberry nuclei isolated with this method turned dark brown due to the presence of co-precipitated polyphenolics and could not be used for downstream applications. Nevertheless, raspberry nuclei isolated with the method described here showed few signs of oxidation, and HMW-DNA was readily digested with restriction enzymes (Figure 3) and has been used for the construction of the first publicly available red raspberry BAC library. Currently, the library comprises over 15,000 clones with an average insert size of approximately 130 kb and less than 1% contamination with chloroplast and mitochondrial DNA, respectively (Plant and Animal Genome Conference XII, 2004, San Diego, CA, USA. http://www.intl-pag.org/12/abstracts/P2a_PAG12_148.html). Similarly, nuclei extracted from blackcurrant, although more discolored than embedded raspberry nuclei (data not shown), contained HMW-DNA that was easily subjected to restriction enzyme digestion (Figure 3), illustrating the versatility of the technique. This result is remarkable because both raspberry and blackcurrant have proven to be recalcitrant to standard genomic DNA extractions required (e.g., for Southern hybridizations). However, in line with the opaque color development in nuclei extracted from blackcurrant, blackcurrant DNA appears to be more liable to contamination with polysaccharides and polyphenolics than raspberry DNA (2).

Figure 2. Typical example of embedded raspberry nuclei following extraction with (A) standard protocols and (B) our method.

The browning of the agarose plugs is caused by co-precipitated and oxidizing polyphenolics, which covalently bind to proteins and DNA.

Figure 3. Restriction enzyme digests (HindIII) of (A) raspberry HMW-DNA and (B) blackcurrant HMW-DNA.

The λ Ladder PFG marker (M) was run alongside (C) undigested DNA and (D) digested HMW-DNA. The size of marker fragments is shown in kilobases (kb). Raspberry HMW-DNA was subjected to 0.5-80 U HindIII and digested for 20 min at 37°C. Blackcurrant HMW-DNA was digested with 20–120 U HindIII and digested for (D1) 120 min and (D2) 24 h at 37°C. The visualized digested DNA represents a mixture of different amounts of added enzyme. HWA-DNA, high molecular weight DNA.

In conclusion, we have developed a novel nuclei extraction method for leaves of woody perennial soft-fruit species containing excess polysaccharides and polyphenolics. The method utilizes a modified buffer system including 4% (w/v) PVP-10 and a combination of nylon filters and Percoll gradients to purify nuclei extracts prior to embedding in agarose plugs. The isolated HMW-DNA is of high quality, accessible to restriction enzyme digestion, and allows, as demonstrated for raspberry, cloning of DNA fragments with an average size of 130 kb.

Competing Interests Statement

The authors declare no competing interests.

Acknowledgments

We gratefully acknowledge the financial support of the Scottish Executive Environment and Rural Affairs Department (SEERAD). We also thank Mary Woodhead, Craig Simpson, and Gillian Clark for valuable discussions and advice.

Additional information

Funding

We gratefully acknowledge the financial support of the Scottish Executive Environment and Rural Affairs Department (SEERAD). We also thank Mary Woodhead, Craig Simpson, and Gillian Clark for valuable discussions and advice.