Genetic diversity of human RNase 8

Abstract

Background: Ribonuclease 8 is a member of the RNase A family of secretory ribonucleases; orthologs of this gene have been found only in primate genomes. RNase 8 is a divergent paralog of RNase 7, which is lysine-enriched, highly conserved, has prominent antimicrobial activity, and is expressed in both normal and diseased skin; in contrast, the physiologic function of RNase 8 remains uncertain. Here, we examine the genetic diversity of human RNase 8, a subject of significant interest given the existence of functional pseudogenes (coding sequences that are otherwise intact but with mutations in elements crucial for ribonucleolytic activity) in non-human primate genomes.

Results: RNase 8 expression was detected in adult human lung, spleen and testis tissue by quantitative reverse-transcription PCR. Only two single-nucleotide polymorphisms and four unique alleles were identified within the RNase 8 coding sequence; nucleotide sequence diversity (π = 0.00122 ± 0.00009 per site) was unremarkable for a human nuclear gene. We isolated transcripts encoding RNase 8 via rapid amplification of cDNA ends (RACE) and RT-PCR which included a distal potential translational start site followed by sequence encoding an additional 30 amino acids that are conserved in the genomes of several higher primates. The distal translational start site is functional and promotes RNase 8 synthesis in transfected COS-7 cells.

Conclusions: These results suggest that RNase 8 may diverge considerably from typical RNase A family ribonucleases and may likewise exhibit unique function. This finding prompts a reconsideration of what we have previously termed functional pseudogenes, as RNase 8 may be responding to constraints that promote significant functional divergence from the canonical structure and enzymatic activity characteristic of the RNase A family.

Figure 1

Expression of RNase 8 in adult and fetal human tissues. Copies of RNase 8 transcript per copy of GAPDH (x 10⁶) in (A) adult and (B) fetal tissues. Shown are the mean ± standard deviations of three separate experimental trials performed on pooled cDNAs. Abbreviations include pancr, pancreas; prost, prostate; sm int, small intestine; leukoc, leukocyte; sk muscle, skeletal muscle.

Figure 2

The coding sequence of human RNase 8 includes two non-silent SNPs and a previously unrecognized 5' extended open reading frame. (A) Highlighted in red are SNPs located within the hydrophobic segment previously identified as a signal sequence (in black, in italics) another within the newly-identified coding segment preceding it (in green, in italics). Highlighted in blue are the amino terminus (K) of the RNase domain (at the arrow), the eight cysteines and catalytic histidines (H) and lysine (K); stop codon denoted by asterisk (*) and bold text. (B) Number of independent alleles and percent of total are as shown. GenBank accession numbers for sequences features in this figure are JQ361124-JQ361128.

Figure 3

Identification and characterization of RNase 8 transcript. (A) ~800 and ~1000 bp transcripts encoding RNase 8 were isolated from human spleen cDNA using a nested rapid amplification of cDNA ends (RACE); +, including adapter primer;-without adapter primer. (B) Findings from RACE confirmed by RT-PCR using two unique sources of poly A⁺ RNA; +, including reverse transcriptase;-without reverse transcriptase. (C) RNase 8 on human chromosome 14; black bar, coding sequence originally described [25] green bar, novel amino terminal extension. RACE amplification products are shown above extending 0.8 and 1 kb from primer within the RNase 8 coding sequence; nested RT-PCR amplification product (0.5 kb) is also as shown below. The GenBank accession number for the spleen cDNA RNase 8 transcript sequences is JQ353679.

Figure 4

Predicted amino terminal extensions of the RNase 8 coding sequence are conserved in primate genomes. (A) Shown are coding sequences that incorporate RNase 8 from Build 2.1 of the genome of P. troglodytes and Build 1.1 of M. mulatta and original sequence data generated from RNase 8 isolated from genomic DNA from P. troglodytes (chimpanzee; JQ353681), P. pygmaeus (orangutan; JQ353682), P. hamadryas (babooon' JQ353680), and G. gorilla (gorilla; JQ353683). Amino acids 1-31 are from the extended open reading frame, amino acids 32-55 (separated by hyphen) are those originally predicted as components of a signal sequence; divergence from the H. sapiens (human) sequence is indicated in red. (B) Evaluation of the predicted hydrophobicity of the human RNase 8 amino terminus (amino acids 1-55) via the Kyte-Doolittle algorithm [29].

Figure 5

Evaluation of the distal and proximal translational start sites. (A) COS-7 cells were transfected via the lipofectamine method, and cells expressing carboxy-terminal-FLAG tagged RNase 8 were detected by flow cytometry. Construct CCC3 includes both distal and proximal start sites (see Figure 3); constructs CCC4 and CCC5 include the proximal and distal translational start sites only, respectively, and the final panel is the vector (pcDNA3.1) control alone. IC; rabbit anti-IgG isotype control. (B) Compilation of seven independent experiments; *p < 0.05.