Colorectal Neoplasia Differentially Expressed (CRNDE), a Novel Gene with Elevated Expression in Colorectal Adenomas and Adenocarcinomas

Abstract

An uncharacterized gene locus (Chr16:hCG_1815491), now named colorectal neoplasia differentially expressed (gene symbol CRNDE), is activated early in colorectal neoplasia. The locus is unrelated to any known protein-coding gene. Microarray analysis of 454 tissue specimens (discovery) and 68 previously untested specimens (validation) showed elevated expression of CRNDE in >90% of colorectal adenomas and adenocarcinomas. These findings were confirmed and extended by exon microarray studies and RT-PCR assays. CRNDE transcription start sites were identified in CaCo2 and HCT116 cells by 5'-RACE. The major transcript isoforms in colorectal cancer (CRC) cell lines and colorectal tissue are CRNDE-a, -b, -d, -e, -f, -h, and -j. Except for CRNDE-d, the known CRNDE splice variants are upregulated in neoplastic colorectal tissue; expression levels for CRNDE-h alone demonstrate a sensitivity of 95% and specificity of 96% for adenoma versus normal tissue. A quantitative RT-PCR assay measuring CRNDE-h RNA levels in plasma was (with a threshold of 2(-ΔCt) = 2.8) positive for 13 of 15 CRC patients (87%) but only 1 of 15 healthy individuals (7%). We conclude that individual CRNDE transcripts show promise as tissue and plasma biomarkers, potentially exhibiting high sensitivity and specificity for colorectal adenomas and cancers.

Keywords: CRNDE; RNA biomarker; colorectal adenoma; colorectal cancer; colorectal neoplasia.

Figure 1.

Genomic locus for CRNDE. Exons (wide bars with colored fill) are shown approximately to scale; introns (black lines) are shown likewise but on a much smaller scale. Nucleotide numbering is for GRCh37/hg19. The region of greatest sequence conservation, here called gVC for “genomic Vertebrate Conserved” (stippled box), and transcribed extensions to exons (narrow bars with black fill) are depicted using the intron scale. The alternative exons E1A and E1B overlap by 9 nt. The hypothetical exons upstream of E1A and downstream of E6 correspond to speculative probe sets in the Affymetrix exon arrays. Numerical exon boundaries are provided in Supplementary Table S3. Transcript identification letters are based on those for hCG_1815491 in the Apr07 release of AceView, augmented by our own findings (e.g., In1T transcripts). While our nomenclature system would naturally identify transcript CRNDE-g and -h as CRNDE-g^1B and -g^1A, respectively, we have retained the former designations to maintain parity with AceView. Wherever the presence of a partial exon sequence in an AceView isoform is considered merely to reflect incomplete recovery of the actual sequence, the complete exon is shown. While the 2 major 3′ boundary options for E6 are indicated, others are known but not shown (Suppl. Table S3). The current NCBI Reference Sequence (RefSeq) isoforms for CRNDE are identified on the left, while the original RefSeq for hCG_1815491 is shown at the bottom. Primer pairs (red arrows) are as listed in Supplementary Table S1B; any primer that spans an exon splice junction is shown as a bipartite arrow with a dotted line spanning the absent intronic sequence.

Figure 2.

Affymetrix expression microarray data for colorectal tissue. (A, B) Discovery data for CRNDE probe sets 238021_s_at and 238022_at, respectively, in Affymetrix HG U133 A/B arrays. Both are scatter plots of RMA-normalized fluorescence intensity values, expressed against an arbitrary index to provide a horizontal offset for each array. Sample categories are normal tissue (N) (n = 222), inflammatory bowel disease (I) (colitis) (n = 42), adenoma (A) (n = 29), and colorectal cancer (CRC) (n = 161). (C) Box and whisker plots of RMA-normalized fluorescence intensity for CRNDE probe sets in exon microarrays hybridized with cDNA from colorectal tissue. Sample categories are normal tissue (N) (n = 5), adenoma (A) (n = 5), and colorectal cancer (CRC) (n = 5). For each probe set, the change in expression signal for neoplastic relative to normal tissue is reflected qualitatively by the shading of the disc at the probe set location, as follows: little or no change in white, modest increase in gray (pink in online color art), and large increase in black (red in online color art). Larger versions of the probe set box plots are presented in Supplementary Figure S1C.

Figure 3.

CRNDE RT-PCR on colorectal tissue and plasma RNA. Primer pairs given in Supplementary Table S1B. (A) Endpoint PCR with generic E2-E6 primers (n = 6 for each sample category). (B) qPCR with E2-E6 primers with cDNA from normal tissue (n = 30), adenoma (n = 21), and CRC (n = 20). Plots show 2^–ΔΔCt relative to normal tissue C40_96N (reference sample). The asterisk identifies anomalous sample TB_152_00, which had been classified as normal but which behaved as a cancer specimen in a range of assays including conventional tests unrelated to CRNDE expression. (C) Box and whisker plots of qPCR data from transcript-specific primers (n = 3 for each assay), with the CRNDE splice variant letter shown at the top left of each panel. Data for CRNDE-d could not be plotted on the otherwise common vertical axis range of 0 to 4 and therefore are not shown here. N = normal tissue (n = 28); A = adenoma (n = 21); CRC = colorectal cancer (n = 20). Plots show 2^–ΔΔCt relative to cancer tissue TB_163_97 (reference sample). The symbol ^ denotes one or more values beyond the scale of the graph. (D) Box and whisker plots and ROC plots of qPCR data for CRNDE-h in RNA extracted from plasma. Fold change is plotted as 2^–ΔCt, where ΔCt = Ct_patient – (Ct_Normals)_mean and Ct denotes a mean value (n = 3 for each assay) normalized as described in Materials and Methods (Suppl. Text S1). AUC = area under ROC curve; FC> = fold change (i.e., 2^–ΔCt) threshold above which the test outcome was deemed to be positive, here 2^–ΔCt = 2.8. Full scatter plots for C (including CRNDE-d data) and D are shown in Supplementary Figure S3A and S3B.

Figure 4.

Transcription start sites in CaCo2 cells. (A) Major start sites, from intense amplicon bands obtained in 5′-RACE-1. (B) Minor start sites, from faint amplicon bands obtained in 5′-RACE-1. (C) Transcriptional start sites (major and minor combined) within the E1A (black uppercase, boxed with light fill, yellow fill in online color art) and E1B (dark gray uppercase italics, blue in online color art) regions are underlined, with initiation frequency reflected by nucleotide font size. Nucleotide numbering (horizontal axis in A and B) is for GRCh37/hg19, with nucleotides 54,962,872 to 54,962,336 shown in C (where intron 1 nucleotides are shown in gray lowercase).

Figure 5.

Relative abundance of transcripts within an individual cell line or tissue sample. Plots show 2^–ΔCt, with CRNDE isoforms identified by letter along the horizontal axis. (A) Transcript-specific qPCR data for CRC cell lines (upper histogram) and non-CRC cell lines (lower histogram; noncancer cell lines are marked with an asterisk). The number of independent experiments for each cell line (n = number of bars) is given in the figure key. For CRC cell lines, values for the second HT29 and first LS174T bars have been scaled (divided by 3 and multiplied by 5, respectively) to facilitate co-plotting of data from different experiments, and the first and last CaCo2 bars show the mean values for repeated assays of 2 cDNA preparations (n = 8 and 6 assays, respectively); n = 3 for each assay. For non-CRC cell lines, values for some samples have again been scaled (the third PrEC bar divided by 20, the last 2 MCF10A bars divided by 10, and the last MRC5 bar multiplied by 5) to facilitate co-plotting. (B) Transcript-specific qPCR data for normal colorectal tissue (upper histogram; n = 28), colorectal adenoma (middle histogram; n = 21), and CRC (lower histogram; n = 20) samples; n = 3 for each assay. The transcripts detected within an individual RNA preparation are all shown using bars of the same color. Bars continuing beyond the displayed vertical scale are truncated with a horizontal line and their numerical value shown. In the upper histogram, asterisks identify conspicuous contributions by anomalous sample TB_152_00, which behaved as a cancer specimen in a range of assays (see legend to Fig. 3B). Primer pairs for both panels were as indicated in Supplementary Table S1B. Because qPCR data from non-DNAse–treated cell/tissue RNA preparations had identical profiles to their DNAse-treated counterparts, the figure incorporates both. For a discussion of this issue, see Supplementary Text S3.