Simple cDNA normalization using kamchatka crab duplex-specific nuclease

Abstract

We developed a novel simple cDNA normalization method [termed duplex-specific nuclease (DSN) normalization] that may be effectively used for samples enriched with full-length cDNA sequences. DSN normalization involves the denaturation-reassociation of cDNA, degradation of the double-stranded (ds) fraction formed by abundant transcripts and PCR amplification of the equalized single-stranded (ss) DNA fraction. The key element of this method is the degradation of the ds fraction formed during reassociation of cDNA using the kamchatka crab DSN, as described recently. This thermostable enzyme displays a strong preference for cleaving ds DNA and DNA in DNA-RNA hybrid duplexes compared with ss DNA and RNA, irrespective of sequence length. We developed normalization protocols for both first-strand cDNA [when poly(A)+ RNA is available] and amplified cDNA (when only total RNA can be obtained). Both protocols were evaluated in model experiments using human skeletal muscle cDNA. We also employed DSN normalization to normalize cDNA from nervous tissues of the marine mollusc Aplysia californica (a popular model organism in neuroscience) to illustrate further the efficiency of the normalization technique.

Figure 1

Scheme of DSN normalization of SMART™-prepared first-strand cDNA using the RNA ‘driver’. Black line, abundant transcript; grey line, intermediate transcript; dotted line, rare transcript.

Figure 2

Agarose gel electrophoresis of non-normalized and normalized amplified SMART™-prepared cDNA from skeletal muscle. Lane 1, non-normalized cDNA; lane 2, normalized first-strand cDNA (FN); lane 3, normalized amplified cDNA (AN); lane M, 1 kb ladder (Gibco-BRL).

Figure 3

Virtual northern blot analysis of normalized and non-normalized skeletal muscle cDNA. Lane 1, normalized first-strand cDNA; lane 2, normalized amplified cDNA; lane 3, non-normalized amplified cDNA. DDBJ/EMBL/GenBank accession nos of marker genes employed are listed in table 1.

Figure 4

Insert size distribution in the non-normalized (white bars) and normalized FN (gray bars) and AN (black bars) human skeletal muscle cDNA libraries.