Banco

J Mol Evol (1997) 44:282–288

© Springer-Verlag New York Inc. 1997

Nucleotide Composition Bias Affects Amino Acid Content in Proteins Coded
by Animal Mitochondria
Peter G. Foster, Lars S. Jermiin,* Donal A. Hickey
Department of Biology, University of Ottawa, 30 Marie Curie, Ottawa, Ontario, Canada K1N 6N5
Received: 3 July 1996 / Accepted: 6 November 1996

Abstract. We show that inanimal mitochondria homologous genes that differ in guanine plus cytosine (G +
C) content code for proteins differing in amino acid content in a manner that relates to the G + C content of the
codons. DNA sequences were analyzed using square
plots, a new method that combines graphical visualization and statistical analysis of compositional differences
in both DNA and protein. Square plots dividecodons
into four groups based on first and second position A +
T (adenine plus thymine) and G + C content and indicate
differences in amino acid content when comparing sequences that differ in G + C content. When sequences are
compared using these plots, the amino acid content is
shown to correlate with the nucleotide bias of the genes.
This amino acid effect is shown in all protein-codinggenes in the mitochondrial genome, including cox I, cox
II, and cyt b, mitochondrial genes which are commonly
used for phylogenetic studies. Furthermore, nucleotide
content differences are shown to affect the content of all
amino acids with A + T- and G + C-rich codons. We
speculate that phylogenetic analysis of genes so affected
may tend erroneously to indicate relatedness (or lack
thereof)based only on amino acid content.
Key words: G+C content — Nucleotide bias —
Amino acid bias — Mitochondrial genes — Phylogeny

* Present address: John Curtin School of Medical Research, Australian
National University, GPO Box 334, Canberra, ACT 2601, Australia
Correspondence to: D.A. Hickey; e-mail dhickey@uottawa.ca

Introduction
Whole genomes may differ widely in their average guanineplus cytosine (G + C) content (Lee et al. 1956; Muto
and Osawa 1987). Individual genes from different genomes may also differ in G + C content (Jermiin et al.
1994; Yoshida et al. 1997; Hashimoto et al. 1994). These
differences can occur at both synonymous (silent) and
nonsynonymous (replacement) codon sites (Jukes and
Bhushan 1986). If only synonymous codon sites differ
between genes, thecorresponding protein sequences will
be identical. However, if nonsynonymous codon sites are
affected the corresponding protein sequences will be different.
A relationship between the nucleotide composition
and amino acid content is known from several genomic
sources, including bacterial DNA (Sueoka 1961; Andersson and Sharp 1996), viral RNA (Berkhout and van
Hemert 1994), animalmitochondrial DNA (Jukes and
Bhushan 1986; Jermiin et al. 1994, 1997), and eukaryotic
nuclear DNA (D’Onofrio et al. 1991; Collins and Jukes
1993; Porter 1995; Yoshida et al. 1997). However, this
relationship is not universal (Hashimoto et al. 1994,
1995).
In a phylogenetic context, protein sequences are often
considered more useful than DNA sequences because
they can be immune to base compositionaldifferences
occurring in the latter, and so phylogenetic information
would be unaffected in the protein sequence (Loomis and
Smith 1990; Hasegawa and Hashimoto 1993; Hashimoto
et al. 1995). However, the evidence presented above does
suggest that protein sequences are not always immune to

283

compositional differences in the DNA, and this could
have an effect on the success ofphylogenetic analysis
(Steel et al. 1993, 1995).
In order to delineate the relationship between compositional bias at the DNA and protein levels, and in view
of its potential phylogenetic implications, we have developed an analytical method which constitutes a unifying approach to the analysis of compositional bias. We
introduce the square plot analysis, a joint graphical and
statistical approach...