Telling the Evolutionary Time: Molecular Clocks and the Fossil Record

Chapter 6

Episodic evolution of nuclear small subunit

ribosomal RNA gene in the stem-lineage of

Foraminifera

Jan Pawlowski and Cédric Berney

ABSTRACT

Several studies provide evidence that the rates of nucleotide substitution vary

across lineages within evolutionary radiations. Little is known, however, about the

episodic variations of rates in the stem-lineages leading to these radiations.

Phylogenetic analyses of nuclear small subunit ribosomal RNA gene of

Foraminifera, calibrated with fossil data, demonstrate extreme rates variations

between different lineages within the foraminiferan radiation. Our present data also

show a significant increase of substitution rates in the stem-lineage of Foraminifera.

The duration of this stem-lineage was estimated at a maximum of 150–250 million

years. Based on this conservative calibration, we calculated that the rate of

substitution of the foraminiferan stem-lineage averaged 1.0–1.65 substitutions/

1000 sites per 10^6 years. This is more than 30 times faster than the typical rates

observed within the radiation of benthic Foraminifera. Such high rates can be

related to a G/C bias in base composition, which apparently led to a reinforcement

of the secondary structure of the foraminiferan SSU rRNA. Our study shows that

an acceleration of substitution rates in a stem-lineage may lead to an erroneous

inference of the phylogenetic position of a group and to an overestimate of its

divergence time.

Introduction

Accurate prediction of evolutionary time from molecular data depends largely on the
homogeneity of nucleotide substitution rates. However, with increasing DNA sequence
data, there is growing evidence that the substitution rates vary across lineages and over
time. Numerous studies demonstrate the important variations of evolutionary rates
between and within different taxonomic groups (Britten 1986; Li et al. 1987; Bousquet et
al. 1992; Philippe et al. 1994; Sorhannus 1996; Ayala 1997; Friedrich and Tautz 1997;
Hwang et al. 1998; Van de Peer et al. 2000). These variations have been proposed to be
due to evolutionary changes in DNA replication or repair mechanisms (Britten 1986), to
the generation-time effect (Li et al. 1996), or to changes of metabolic rates (Martin and
Palumbi 1993), but none of these hypotheses seems to be sufficiently confirmed by
available data (Li 1993).