Unexpected DNA loss mediated by the DNA binding activity of ribonuclease A

Abstract

Ribonuclease A (RNase A) is widely used in molecular biology research both for analytical assays and for nucleic acid preparation. The catalytic mechanism of RNase A is well understood and absolutely precludes activity on DNA; however anecdotal reports of DNA degradation by RNase A are not uncommon. Here we describe a mechanism by which RNase A treatment can lead to apparent DNA degradation. This results from the surprising finding that RNase A remains functional in a phenol:chloroform mixture, to our knowledge the only enzyme that survives this highly denaturing solvent environment. Although RNase A does not cleave the DNA backbone it is capable of binding to DNA, forming stable RNase A-DNA complexes that partition to the interphase or organic phase during phenol:chloroform purification. The unexpected survival of the RNase A DNA-binding activity in phenol means that these complexes are not dissolved and a substantial amount of RNase A-bound DNA is permanently removed from the aqueous phase and lost on phase separation. This effect will impact DNA recovery from multiple procedures and is likely to represent a source of sequence bias in genome-wide studies. Our results also indicate that the results of analytical studies performed using RNase A must be considered with care.

Figure 1. RNase and DNase analysis of major satellite species.

Total RNA from NIH/3T3 cells was treated with indicated enzymes for 30 min at 37°, phenol:chloroform extracted, ethanol precipitated and glyoxylated prior to electrophoresis, then blotted and probed for major satellite sequences, ethidium staining of ribosomal RNA is shown as a control. A: Samples treated in water without and with RNase A. B: Samples treated in NEBuffer 3 with indicated ribonucleases. C: Samples treated in RQ1 DNase buffer without and with DNase I. For quantification, error bars represent ±1 s.d., y-axes in arbitrary units, p values calculated using a one-way ANOVA (n = 4) for B and Student’s t-test (n = 3) for C.

Figure 2. RNase A activity on DNA.

A, B: 0.5 ng major satellite PCR product was mixed with 1 µg NIH/3T3 total RNA and analysed as in Fig. 1. A: Treatment with RNase A from different manufacturers. B: Treatment with RNase A in different buffers. C: 50 ng DNA molecular weight marker was mixed with 1 µg NIH/3T3 total RNA and treated without or with RNase A, purified as in Fig. 1 and separated on a non-denaturing 1xTBE gel before blotting and probing for the molecular weight marker. D: ³²P-labelled 50 bp ladder mixed with 1 µg RNA was treated without or with RNase A for 30 min at 37°, directly separated on an 8% PAGE gel and exposed to a phosphorimaging screen. For quantification, error bars represent ±1 s.d., y-axes in arbitrary units, p values were calculated by one-way ANOVA (A, B) or Student’s t-test (C, D), n = 3 in all cases.

Figure 3. Inhibition of DNA removal by RNase A.

A: 0.5 ng major satellite PCR product was mixed with 1 µg NIH/3T3 RNA (lanes 1, 2) or with 1 µg DNA molecular weight marker (lanes 3, 4), and treated without or with RNase A followed by phenol:chloroform extraction and analysis as in Fig. 1. B: Identical to A, except that samples were incubated with 20 µg proteinase K for 15 min at 37° after RNase A treatment and phenol:chloroform extraction was omitted. For quantification, error bars represent ±1 s.d., y-axes in arbitrary units, analysis by one-way ANOVA (n = 3), differences in B are not significant.

Figure 4. Re-partitioning of RNase-treated DNA.

A, B: 5 ng ³²P-labelled major satellite PCR product mixed with 1 µg of NIH/3T3 RNA was incubated without or with RNase A for 30 min at 37°. Reactions were then diluted to 100 µl with water and extracted with 100 µl phenol:chloroform. A: 10 µl of the aqueous phase was analysed by electrophoresis in a non-denaturing 1xTBE gel which was dried and exposed to a phosphorimaging screen. B: Radioactivity present in the aqueous and phenol phases quantitated using a Geiger counter. C: DNA loss after RNase A treatment with phenol:chloroform extraction at pH 7 or pH 8. D: Activity of RNase A in phenol:chloroform. 1 µl 1 mg/ml RNase A was added to 100 µl phenol:chloroform and vortexed. 1 µl 1 mg/ml NIH/3T3 RNA was added, vortexed and reactions incubated at 37° for 30 min, followed by extraction with 100 µl water, precipitation and gel electrophoresis. For quantification, error bars represent ±1 s.d, y-axes in arbitrary units for A, D or Bequerels in B, analysis by Student’s t-test, n = 3.