Molecular-grade glycogen is widely used to recover nanogram or picogram quantities of DNA and RNA across molecular biology applications in the life sciences. As a result, its purity is critical to obtain reliable results. Using agarose gel electrophoresis, we detected pg/µL (DNA) to ng/µL (RNA) concentrations of nucleic acid in two of the nine glycogen samples obtained from commercial suppliers. Denaturing gradient gel electrophoresis of 16S rRNA gene PCR-amplified products indicated that an additional two samples contained detectable contamination. We also tested a synthetic polymer co-precipitant, linear polyacrylamide (LPA); none of the four samples tested with LPA were detectably contaminated. The partial 16S rRNA gene sequence associated with the contaminated samples of the shellfish-derived glycogen was nearly identical to the sequence of Actinobacteria lwoffii, which has been isolated from mussels previously. By testing the recovery of low-nanogram amounts of DNA with multiple precipitants and simulated experimental conditions, we demonstrated that LPA was a preferable co-precipitant for sensitive protocols.

Introduction

Glycogen is routinely used to aid in the precipitation and recovery of nucleic acids in many molecular biology applications. These include gene expression analysis (1,2,3), RNA precipitation following tissue microdissection (4), DNA methylation studies (5), DNA-protein interactions (6), cancer research (7,8), as well as the detection of other human diseases, such as HIV provirus in host DNA (9). Glycogen is also applied extensively for microbial community analyses, including the analysis of nucleic acids from the human microbiome (10), microbial communities in low-biomass environmental samples coupled with whole-community genome amplification (11), and stable-isotope probing (SIP) protocols focusing on the detection of functional groups within a microbial community (12,13). A recently published procedure for environmental genomic DNA sequencing with next-generation sequencing technologies (e.g., 454 pyrosequencing) includes glycogen as a carrier for precipitation (www.454.com/downloads/protocols/NatureMethods_ metagenomics.pdf). Because of the small quantities of nucleic acid being manipulated (e.g., nanogram or picogram quantities) and the sensitive nature of the downstream analyses in many studies, it is imperative that co-precipitants (other than yeast-derived tRNA, for example) be free from nucleic acid.

Despite the critical importance of co-precipitant reagent purity for the manipulation of small amounts of nucleic acid, commercially available supplies of molecular-grade glycogen have never been rigorously assessed for nucleic acid contamination. Because the biological sources of glycogen for commercial distribution are typically mussels and oysters, the synthetic nature of alternative co-precipitants [e.g., linear polyacrylamide (LPA)] would suggest higher purity. LPA has been demonstrated as a suitable co-precipitant for ethanol precipitations recovering DNA fragments greater than 20 base pairs (14). Here, we used PCR and denaturing gradient gel electrophoresis (DGGE) to test the hypothesis that co-precipitants from biological origin are associated with nucleic acid contamination, and test both glycogen and LPA for the ability to recover small amounts of DNA in isopropanol and polyethylene glycol (PEG) precipitations, with and without the addition of cesium chloride (CsCl) to the precipitation. The latter condition simulates precipitations commonly associated with RNA-SIP and DNA-SIP protocols (12,13).

Materials and methods

Co-precipitant samples

We obtained a total of nine glycogen samples and four LPA samples (Table 1) in 2008 and 2009 to assess possible nucleic acid contamination. We assigned a letter designation randomly to each sample to maintain anonymity (A-I for glycogen samples; J-M for LPA samples). For one sample (sample A), we obtained multiple production lots from the same supplier (A1 and A2).

Table 1. Glycogen and LPA samples used in this study (Click to enlarge)

PCR and denaturing gradient gel electrophoresis (DGGE)

We amplified a ~500-bp fragment of bacterial 16S rRNA (63f-GC and 518r; Invitrogen, Carlsbad, CA, USA) to detect DNA contamination of co-precipitants. PCR was conducted in 50-µL reactions, each containing 5 µL 10× PCR buffer (New England BioLabs, Ipswich, MA, USA), 200 µM dNTPs (New England BioLabs), 25 pmol each primer (Invitrogen), 5 µg BSA (Sigma-Aldrich, St. Louis, MO, USA) and 1.25 U Taq DNA polymerase (New England BioLabs). A total of 20 µg each glycogen sample (samples A-I) and approximately 30 µg LPA (samples J-M) were used as template. PCR was carried out in a DNA Engine thermocycler (Bio-Rad, Hercules, CA, USA) and the protocol consisted of 95°C for 4 min; 30 cycles of 95°C for 60 s, 55°C for 60 s, and 72°C for 60 s; followed by a final extension at 72°C for 10 min. Aliquots (4 µL) of each PCR product were run on a 1% agarose gel, which was subsequently stained with a 1 µg/mL ethidium bromide solution. All 14 PCR products were also run on a 6% DGGE polyacryl-amide gel consisting of a denaturing gradient of 40% to 70% on a CBS DGGE system (CBS Scientific, Solana Beach, CA, USA) at 85V for 14 h at 60°C, as described previously (15). Following electrophoresis, gels were stained with SYBR green I nucleic acid stain (Invitrogen) and imaged on a Typhoon 9400 imager (Amersham, Pittburgh, PA, USA).

Direct agarose gel electrophoresis of co-precipitant

Glycogen samples A1/A2 and D (200 µg each) were run on a 1% agarose gel containing 1:10,000 concentration of Gel Red nucleic acid stain (Biotium, Hayward, CA, USA), and visualized on an Alpha DigiDoc RT (Alpha Innotech, San Leandro, CA, USA).