Rapid Evolution of a Primate Sperm Protein: Relaxation of Functional

Constraint or Positive Darwinian Selection?

Alejandro P. Rooney and Jianzhi Zhang
Institute of Molecular Evolutionary Genetics and Department of Biology, Pennsylvania State University

Protamines are arginine-rich proteins that replace histones and bind sperm DNA during spermatogenesis in verte-
brates. Previous studies have shown that protamine exons evolve faster than does the protamine intron. It has been
suggested that this is a result of a relaxation of functional constraint. However, a more likely explanation is that
the evolutionary rate of exons has been accelerated by positive Darwinian selection, because introns are generally
believed to evolve in a neutral fashion. Therefore, we examined the possibility that positive selection has been
acting on the protamine genes of three groups of placental mammals: primates (hominoids and Old World monkeys),
rodents (mice, rats, and guinea pigs), and pecoran ruminants (deer and bovids). We found that the nucleotide
substitution rate at nonsynonymous sites is significantly higher than the rate at synonymous and intron sites for

Downloaded from https://academic.oup.com/mbe/article/16/5/706/2925467 by guest on 12 September 2021

protamine P1 of hominoids and Old World monkeys. This result suggests that positive selection has been operating
on protamine P1 of these species. In contrast, no clear-cut evidence of positive selection was found for protamine
P1 of ruminants and rodents or protamine P2 of primates. The agent of positive selection on primate protamine P1
remains unknown, though sperm competition is a possibility. Further investigations on the function and intraspecific
polymorphism of this protein are needed in order to identify the selection agent.

Introduction
Mammalian sperm DNA is the most densely
packed form of eukaryotic DNA. For example, sperm
DNA is six times more highly condensed than somatic
chromosome DNA in the mouse (Ward and Coffey
1991). This highly condensed form is thought to be
achieved by packaging of DNA molecules in side-by-
side linear arrays, brought about through the action of
protamines (Ballhorn 1982; Ward and Coffey 1991).
Protamines are small arginine-rich proteins that replace
histones during the process of sperm nucleus conden-
sation in spermatogenesis (Oliva and Dixon 1991). A
large number of positively charged arginine residues al-
lows protamine P1 to bind and condense negatively
charged sperm DNA, presumably by inserting into the
minor groove of the DNA helix (Ballhorn 1982; Ward
and Coffey 1991). There are two types of protamines in
mammals: protamine P1 is the major sperm protamine,
whereas protamine P2 is expressed only in primates and
rodents (Retief and Dixon 1993). In primates, protamine
P1 contains 50–53 amino acid residues (Retief et al.
1993), while the number in protamine P2 is between 99
and 102 (Retief and Dixon 1993). In all placental mam-
mals examined to date, protamines have been found to
contain several cysteine residues that form a crystalline-
like structure through the formation of disulfide bridges
in the mature sperm nucleus (Ballhorn, Corzett, and
Mazrimas 1992; Queralt et al. 1995; Retief et al. 1995a).
Thus, the sperm nucleus of placental mammals is much
more stable than the sperm nuclei of marsupials, mono-
tremes, or other vertebrates whose protamines lack cys-
teine residues (Ward and Coffey 1991; Ballhorn, Cor-
Key words: protamine, positive selection, sperm proteins, intron,
sperm competition, primates.
Address for correspondence and reprints: Alejandro P. Rooney,
Institute of Molecular Evolutionary Genetics and Department
of Biology, 311 Mueller Laboratory, The Pennsylvania State Uni-
versity, University Park, Pennsylvania 16802. E-mail:
rooney@imeg.bio.psu.edu.
Mol. Biol. Evol. 16(5):706–710. 1999
q 1999 by the Society for Molecular Biology and Evolution. ISSN: 0737-4038
zett, and Mazrimas 1992; Queralt et al. 1995; Retief et
al. 1995a, 1995b).
Protamines P1 and P2 have been reported to show
an overall rapid rate of sequence divergence among
mammalian species (Retief and Dixon 1993; Retief et
al. 1993, 1995a; Queralt et al. 1995). Retief et al. (1993)
noted that the two exons of the protamine P1 gene in
primates evolve faster than the single intron, a pattern
also observed for the protamine P2 genes of primates
(Retief and Dixon 1993). It has been suggested that the
unusually high variability of protamine P1 exons is in-
dicative of a relaxation of functional constraint (Retief
et al. 1993), although no explanation has been offered
for the observed pattern in the primate protamine P2
gene. However, a significantly higher rate of exon di-
vergence in comparison with that of introns is a feature
of positive selection. Therefore, we reanalyzed prot-
amine P1 and P2 sequences from primates, as well as
protamine P1 sequences from rodents and pecoran ru-
minants, in order to gain a better understanding of the
evolutionary mechanisms underlying protamine evolu-
tion. Specifically, we examined the plausibility of pos-
itive selection in protamines by comparing the rates of
nucleotide substitution at synonymous, nonsynonymous,
and intron sites.
Materials and Methods
Nucleotide sequences of protamines P1 and P2
from primates (human [Homo sapiens], bonobo [Pan
paniscus], common chimpanzee [Pan troglodytes], go-
rilla [Gorilla gorilla], orangutan [Pongo pygmaeus],
common gibbon [Hylobates lar], red guenon [Erythro-
cebus patas; P1 only], rhesus macaque [Macaca mulat-
ta; P2 only], and pig-tailed macaque [Macaca nemes-
trina; P2 only]) and protamine P1 from pecoran rumi-
nants (moose [Alces alces], white-tailed deer [Odoco-
ileus virginianus], elk [Cervus elaphus], European cattle
[Bos taurus], and gazelle [Gazella dorcas]) and rodents
(mouse [Mus musculus], rat [Rattus norvegicus], and
guinea pig [Cavia porcellus]) were obtained either from

706

GenBank or from the literature (Retief et al. 1993; Quer-
alt et al. 1995). We examined the single intron and the
coding sequences of both exons for all species exam-
ined. (Herein, we refer to the coding sequences of exons
1 and 2 combined as, simply, ‘‘exon.’’) These sequences
were aligned by taking into consideration the deduced
amino acid sequences. Because of the relatively large
sequence divergence, we analyzed primate, rodent, and
ruminant sequences separately in order to minimize the
confounding influence of multiple hits and uncertainties
in sequence alignment. Any alignment site containing a
gap was omitted from the analysis. Alignments are
available on request.
We first computed Kimura’s (1980) distances sep-
arately for both exon (dE) and intron (dI) sequences of
all species pairs within each group. The mean values of
dE and dI and their standard errors were also computed.
This was done by using a test version of the software
MEGA2 (S. Kumar et al., unpublished). In addition, the
numbers of synonymous (dS) and nonsynonymous (dN)
substitutions per site were computed for each pair of
sequences within groups by using a modified version of
the Nei and Gojobori method (Nei and Gojobori 1986;
Zhang, Rosenberg, and Nei 1998). For each species
group, the mean values of dS and dN and their standard
errors were computed using MEGA2. In all cases, stan-
dard errors were computed by using the bootstrap meth-
od with 500 replicates.
We also employed the method of Zhang, Kumar,
and Nei (1997) to test for positive selection. In this
method, the numbers of synonymous (s) and nonsynon-
ymous (n) substitutions per sequence are computed for
each branch in a phylogenetic tree. The values of s and
n are then compared with their expected values under
the null hypothesis of neutral evolution. In order to com-
pute s and n, the ancestral sequences of all interior nodes
in the tree must be inferred. Here, we used the distance-
based Bayesian method (Zhang and Nei 1997). As the
species studied were closely related, the inferred ances-
tral sequences were reliable (Zhang and Nei 1997). In
addition, the number of nucleotide substitutions in intron
sequences (i) for each branch in the tree was estimated
in this manner. The numbers of potential synonymous
sites (S) and potential nonsynonymous sites (N) were
estimated using the modified Nei-Gojobori method
(Zhang, Rosenberg, and Nei 1998), while the potential
number of intron sites (I) was set equal to the entire
intron sequence length. Fisher’s exact test was subse-
quently utilized to test the significance of the difference
between n/N, s/S, and i/I, which should be equal under
strict neutrality.
Results and Discussion
Pairwise distances were found to be greater in the
exon sequence of protamine P1 than in the intron for
the primate species examined, and the mean dE was sig-
nificantly higher than the mean dI (t 5 2.87, P , 0.005;
table 1). A greater sequence divergence in the exons
than in the intron was observed only in some ruminant
comparisons involving elk (table 1). Overall, the mean
Positive Selection in Protamine P1 707
Table 1
Percentage Pairwise Distances Between Taxa for Exon
(above diagonal) and Intron (below diagonal) Sequences
of Protamine P1 Using the Kimura Two-Parameter
Model
1 2 3 4 5 6 7
Primates
1. Human . . . . . . . . . . — 6.3 6.3 6.3 7.8 4.9 10.1
2. Bonobo . . . . . . . . . 2.5 — 1.3 5.5 10.0 7.0 12.5
3. Common chimp . . 2.5 0.0 — 5.5 10.0 7.0 12.5
4. Gorilla . . . . . . . . . . 1.2 1.2 1.2 — 10.0 7.0 12.5
5. Orangutan . . . . . . . 1.2 1.2 1.2 0.0 — 8.5 10.9
6. Common
gibbon . . . . . . . . . . . 3.7 3.7 3.7 2.5 2.5 — 10.9
Downloaded from https://academic.oup.com/mbe/article/16/5/706/2925467 by guest on 12 September 2021
7. Red guenon . . . . . . 6.4 6.4 6.4 5.1 5.1 7.7 —
Mean dE 5 8.2 6 1.4; mean dI 5 3.1 6 1.1
Mean dN 5 8.9 6 1.8; mean dS 5 6.4 6 2.2
Ruminants
1. Elk . . . . . . . . . . . . . — 4.1 6.3 6.3 7.0
2. Moose . . . . . . . . . . 4.2 — 3.4 4.8 4.1
3. White-tailed
deer . . . . . . . . . . . . . 3.2 7.6 — 7.0 5.5
4. European cattle . . . 5.4 9.9 6.5 — 3.4
5. Gazelle. . . . . . . . . . 6.5 11.1 10.0 12.5 —
Mean dE 5 5.2 6 1.3; mean dI 5 7.6 6 1.9
Mean dN 5 2.8 6 1.2; mean dS 5 10.9 6 3.5
Rodents
1. Mouse . . . . . . . . . . — 6.3 23.3
2. Rat . . . . . . . . . . . . . 30.9 — 18.9
3. Guinea pig. . . . . . . 35.7 43.6 —
Mean dE 5 15.9 6 3.2; mean dI 5 36.7 6 7.0
Mean dN 5 9.6 6 3.5; mean dS 5 28.8 6 8.5
dE was lower than the mean dI for ruminants (table 1).
For the three species of rodents examined, the intron
sequence divergence was greater than the exon sequence
divergence in pairwise comparisons, and the mean dE
was significantly lower than the mean dI across species
(t 5 2.70, P , 0.005; table 1). Here, positive selection
may be difficult to detect due to a saturation effect in
amino acid substitutions among distantly related se-
quences (e.g., Tanaka and Nei 1989). Alternatively, the
evolutionary forces affecting rodent protamine P1 may
be different than those for primates and ruminants.
Therefore, a study of protamines among closely related
rodent species may be an interesting avenue for future
investigation. However, we did not investigate the ro-
dent sequences any further, because we did not see a
higher rate of substitution in the exon sequence than in
the intron. For primate protamine P2, a greater sequence
divergence in the exons than in the intron was observed,
although the mean dE was not significantly different
from the mean dI (t 5 0.77, P . 0.1; table 2).
Positive selection is usually tested by comparing dN
and dS; a significantly higher value of dN is taken as
strong evidence of positive selection. In some individual
pairwise comparisons of primate protamine P1, we
found that dN is larger than dS. However, the difference
was not statistically significant. Since the numbers of
substitutions were small, we computed the mean values
of dN and dS for all pairwise comparisons. Although the
mean dN was higher than the mean dS, that difference
708 Rooney and Zhang

Table 2
Percentage Pairwise Distances Between Taxa for Exon
(above diagonal) and Intron (below diagonal) Sequences
of Primate Protamine P2 Using the Kimura Two-
Parameter Model
1 2 3 4 5 6 7 8
1. Human . . . . . . . . . . . . — 2.7 3.4 3.0 6.6 5.4 5.4 6.9
2. Bonobo . . . . . . . . . . . 1.9 — 1.3 1.7 6.6 5.4 6.5 8.0
3. Common chimp. . . . . 4.6 4.0 — 1.7 6.2 5.4 6.5 8.0
4. Gorilla . . . . . . . . . . . . 2.6 0.6 3.3 — 6.6 5.1 6.2 7.7 FIG. 1.—Phylogeny used for plotting substitution classes for tests
5. Orangutan . . . . . . . . . 3.9 3.3 0.6 4.0 — 6.2 8.4 9.5 of positive selection in hominoid and Old World monkey protamine
6. Common gibbon . . . . 3.9 3.3 3.3 2.6 4.0 — 6.5 7.3 P1 sequences. The numbers above branches represent n, s, and i, re-
7. Rhesus macaque . . . . 7.4 6.8 7.5 6.8 7.5 5.3 — 2.0 spectively.
8. Pig-tailed macaque . . 7.4 6.8 7.5 6.8 7.5 5.3 0.0 —

Downloaded from https://academic.oup.com/mbe/article/16/5/706/2925467 by guest on 12 September 2021

Note.—Mean dE 5 5.6 6 0.9; mean dI 5 4.6 6 1.1; mean dN 5 5.7 6 1.0;
mean dS 5 5.1 6 1.5.
was not statistically significant either (table 1). An al-
ternative approach to examining the evidence for posi-
tive selection in cases involving small numbers of sub-
stitutions and a well-established phylogeny is to utilize
the phylogenetic approach of Zhang, Kumar, and Nei
(1997). Since the relationships of the primate species
examined in this study are highly corroborated (Ruvolo
1997), we plotted the numbers of substitutions per se-
quence on the species phylogeny shown in figure 1. The
posterior probability of inferred ancestral sequences was
98%–99% for the protamine P1 exons and 99%–100%
for the protamine P1 intron. We observed that the total
numbers of synonymous and nonsynonymous substitu-
tions in all branches of the tree were 9 and 31, respec-
tively, and the numbers of synonymous and nonsynon-
ymous sites were 43 and 107, respectively (fig. 1).
Therefore, the rate of nonsynonymous substitution (31/
107) was 1.4 times that of synonymous substitution (9/
43). However, the nonsynonymous and synonymous
rates were not significantly different (P . 0.1), which
may have been due to the short sequence length (50
codons) of protamine P1 in hominoids and Old World
monkeys. Therefore, to increase the power of the test
for positive selection, we combined synonymous and in-
tron sites to be taken as the neutral sites. A number of
studies have shown that the intron and synonymous sub-
stitution rates are approximately equivalent (Hughes and
Yeager 1997; Li 1997; Smith and Hurst 1998). In ad-
dition, the synonymous substitution rate is not signifi-
cantly different from the intron substitution rate for pri-
mate protamine P1 (P . 0.05). This suggests that intron
sites can generally be regarded as neutral. Li et al.
(1996) computed the distances between humans and an
Old World monkey species for several different intron
sequences. Those authors found the average to be 7.1%.
In the present study, the distance between the single Old
World monkey species (red guenon) and all hominoid
species for the protamine P1 intron was in the range of
5.1%–7.7%, while the distance between humans and Old
World monkeys was 6.4%. These comparisons indicate
that the substitution rate of the protamine P1 intron is
close to the average rate for hominoid and Old World
monkey introns and thus validate its use as a neutral
rate of nucleotide substitution. Consequently, we found
that the substitution rate at nonsynonymous sites was
significantly greater than the rate at intron and synony-
mous sites (P 5 0.003; table 3), suggesting that positive
selection is acting on primate protamine P1.
For protamine P1 of ruminants and protamine P2
of primates, evidence of positive selection was lacking.
First, negative selection was detected in ruminant prot-
amine P1 based on a significantly lower nonsynonymous
rate than the combined synonymous and intron rates (ta-
ble 3). This is also reflected in the mean values of the
numbers of synonymous and nonsynonymous substitu-
tions per site, as dN is smaller than dS (t 5 2.1, P ,
0.05; table 1). For this test, we used the topology ((elk,
(moose, white-tailed deer)), (European cattle, gazelle)),
which is based on the current classification of pecoran
ruminants (Nowak 1991). Second, neutrality could not
be rejected for primate protamine P2 (table 3). The tree
topology used in this test was identical to that shown in
figure 1 except that the two macaque species, sharing a
sister taxon relationship, replaced red guenon. However,
the fact that the nonsynonymous rate is slightly greater
than the synonymous and intron rates (tables 2 and 3)
Table 3
Tests of Neutrality for Protamines P1 and P2 of Species
Groups Examined
Synonymous
Nonsynonymous and Intron
Primate P1
Substitutions . . . . . . . . . . . . . . . 31 17
Sites. . . . . . . . . . . . . . . . . . . . . . 107 126
Substitutions per site . . . . . . . . 0.29 0.13
P value . . . . . . . . . . . . . . . . . . . 0.003
Ruminant P1
Substitutions . . . . . . . . . . . . . . . 8 28
Sites. . . . . . . . . . . . . . . . . . . . . . 104 143
Substitutions per site . . . . . . . . 0.08 0.20
P value . . . . . . . . . . . . . . . . . . . 0.006
Primate P2
Substitutions . . . . . . . . . . . . . . . 38 41
Sites. . . . . . . . . . . . . . . . . . . . . . 218 245
Substitutions per site . . . . . . . . 0.17 0.16
P value . . . . . . . . . . . . . . . . . . . 0.469
Note.—Fisher’s exact test was used to examine homogeneity of substitution
rates at nonsynonymous sites versus combined synonymous plus intron sites (see
Zhang, Kumar, and Nei 1997). The Ts/Tv ratio was 2.0 for all comparisons. The
total numbers of intron sites (I) were 96 for primate protamine P1, 106 for
ruminant protamine P1, and 162 for primate protamine P2.

suggests that positive selection may have operated at a
weak level in primate protamine P2 despite the fact that
a relaxation of functional constraint might also explain
the observed substitution pattern.
Because strict neutrality could not be rejected in
comparisons of dN and dS of primate protamine P1, one
might question the hypothesis of positive selection in
favor of a hypothesis of relaxation of functional con-
straint. Our test of positive Darwinian selection that in-
volves the comparison of n/N with (s 1 i)/(S 1 I) de-
pends on the assumption that intron and synonymous
sites are neutral and not subject to purifying selection.
Previously, we showed that the substitution rate of the
protamine P1 intron of hominoids and Old World mon-
keys is similar to the average intron rate for these spe-
cies groups. Despite this fact, it is possible that different
genomic regions do not have the same mutation rate
(Wolfe, Sharp, and Li 1989; Casane et al. 1997). In cases
in which the mutation rate of an individual intron is
higher than average, the substitution rate of this intron
may still appear to be similar to the average rate if pu-
rifying selection is operating. It has been shown that
regulatory elements exist in the introns of certain genes
(e.g., Tiffany, Handen, and Rosenberg 1996), suggesting
that purifying selection might occur in such instances.
It would be interesting to examine whether such ele-
ments exist in the protamine P1 intron. If so, the number
of potentially variable sites in the intron will be smaller
than the I value used in our analysis. Nonetheless, we
found that our test of Darwinian selection will give a
positive result even when the number of variable sites
in the intron constitutes only 61% of the total intron
length. In reality, this proportion is likely to be substan-
tially higher, which suggests that our test result is robust.
Although our results suggest that positive selection
has been operating on protamine P1 of hominoids and
Old World monkeys, the specific selection agent is un-
known. Positive selection has been found in a number
of genes associated with male reproduction (Lee and
Vacquier 1992; Swanson and Vacquier 1995; Metz and
Palumbi 1996; Tsaur and Wu 1997; Civetta and Singh
1998; Vacquier 1998). Several hypotheses have been put
forth to explain positive selection in these cases, but the
exact reason is unclear. For hominoid and Old World
monkey protamine P1, we suspect that positive selection
may be driven by sperm competition (Harcourt 1991;
Baker and Bellis 1995). In systems in which paternal
investment is important, a diverse array of sperm mor-
phologies in a single male is believed to be beneficial
in sperm competition, in which sperm with different
morphologies act to reduce the fertilization success of
other males (i.e., the Kamikaze Sperm Hypothesis; Bak-
er and Bellis 1995). As heterogeneity in sperm nucleus
shape may arise from heterogeneity in chromatin con-
densation (Curry and Watson 1995), protamine P1 might
play a role in allowing for a diverse array of sperm
morphologies in primates. However, the validity of the
Kamikaze Sperm Hypothesis is still open to debate
(Harcourt 1991). Nonetheless, it would be particularly
helpful to have a study of protamine P1 polymorphism
for at least one hominoid species. If sperm competition
Positive Selection in Protamine P1 709
is the cause of positive selection, one would expect a
high level of polymorphism at nonsynonymous sites.
Until such information has been obtained, the above hy-
pothesis cannot be considered sufficient to explain pos-
itive selection in protamine P1, so the reason for this
phenomenon remains elusive.
Acknowledgments
Helpful comments and discussion were provided by
J. M. Cummins, R. L. Honeycutt, J. Lyons-Weiler, and
W. S. Ward. M. Nei greatly improved this manuscript
by commenting on earlier drafts. We thank Sudhir Ku-
Downloaded from https://academic.oup.com/mbe/article/16/5/706/2925467 by guest on 12 September 2021
mar for providing a prerelease version of MEGA2. This
research was supported by grants from the National Sci-
ence Foundation and the National Institutes of Health to
M. Nei.
LITERATURE CITED
BAKER, R. R., and M. A. BELLIS. 1995. Human sperm com-
petition: copulation, masturbation, and infidelity. Chapman
and Hall, London.
BALLHORN, R. 1982. A model for the structure of chromatin
in mammalian sperm. J. Cell Biol. 93:298–305.
BALLHORN, R., M. CORZETT, and J. A. MAZRIMAS. 1992. For-
mation of intraprotamine disulfides in vitro. Arch. Biochem.
Bipophys. 296:384–393.
CASANE, D., S. BOISSINOT, B. H.-J. CHANG, L. C SHIMMIN,
and W.-H. LI. 1997. Mutation pattern variation among re-
gions of the primate genome. J. Mol. Evol. 45:216–226.
CIVETTA, A., and R. S. SINGH. 1998. Sex-related genes, direc-
tional sexual selection, and speciation. Mol. Biol. Evol. 15:
901–909.
CURRY, M. R., and P. F. WATSON. 1995. Sperm structure and
function. Pp.45–69 in J. G. GRUDZINSKAS and J. L. YOVICH,
eds. Gametes—the spermatozoon. Cambridge University
Press, Cambridge, England.
HARCOURT, A. H. 1991. Sperm competition and the evolution
of nonfertilizing sperm in mammals. Evolution 45:314–328.
HUGHES, A. L., and M. YEAGER. 1997. Comparative evolu-
tionary rates of introns and exons in murine rodents. J. Mol.
Evol. 45:125–130.
KIMURA, M. 1980. A simple method for estimating evolution-
ary rates of base substitutions through comparative studies
of nucleotide sequences. J. Mol. Evol. 16:111–120.
LEE, Y.-H., and V. D. VACQUIER. 1992. The divergence of spe-
cies-specific abalone sperm lysins is promoted by positive
Darwinian selection. Biol. Bull. 182:97–104.
LI, W.-H. 1997. Molecular evolution. Sinauer, Sunderland,
Mass.
LI, W.-H., D. L. ELLSWORTH, J. KRUSHKAL, B. H.-J. CHANG,
and D. HEWETT-EMMETT. 1996. Rates of nucleotide substi-
tution in primates and rodents and the generation-time ef-
fect hypothesis. Mol. Phylogenet. Evol. 5:182–187.
METZ, E. C., and S. R. PALUMBI. 1996. Positive selection and
sequence rearrangements generate extensive polymorphism
in the gamete recognition protein bindin. Mol. Biol. Evol.
13:397–406.
NEI, M., and T. GOJOBORI. 1986. Simple methods for estimat-
ing the numbers of synonymous and nonsynonymous nu-
cleotide substitutions. Mol. Biol. Evol. 3:418–426.
NOWAK, R. M. 1991. Walker’s mammals of the world. Johns
Hopkins University Press, Baltimore, Md.
710 Rooney and Zhang
OLIVA, R., and G. H. DIXON. 1991. Vertebrate protamine genes
and the histone-to-protamine replacement reaction. Prog.
Nucleic Acid Res. Mol. Biol. 40:25–94.
QUERALT, R., R. ADROER, R. OLIVA, R. J. WINKFEIN, J. D.
RETIEF, and G. H. DIXON. 1995. Evolution of protamine P1
genes in mammals. J. Mol. Evol. 40:601–607.
RETIEF, J. D., and G. H. DIXON. 1993. Evolution of pro-prot-
amine P2 genes in primates. Eur. J. Biochem. 214:609–615.
RETIEF, J. D., C. KRAJEWSKI, M. WESTERMAN, and G. H. DIX-
ON. 1995a. The evolution of protamine P1 genes in dasyurid
marsupials. J. Mol. Evol. 41:549–555.
RETIEF, J. D., C. KRAJEWSKI, M. WESTERMAN, R. J. WINKFEIN,
and G. H. DIXON. 1995b. Molecular phylogeny and evolu-
tion of marsupial protamine P1 genes. Proc. R. Soc. Lond.
B Biol. Sci. 259:7–14.
RETIEF, J. D., R. J. WINKFEIN, G. H. DIXON, R. ADROER, R.
QUERALT, J. BALLABRIGA, and R. OLIVA. 1993. Evolution
of protamine P1 genes in Primates. J. Mol. Evol. 37:426–
434.
RUVOLO, M. 1997. Molecular phylogeny of the hominoids: in-
ferences from multiple independent DNA sequence data
sets. Mol. Biol. Evol. 14:248–265.
SMITH, N. G. C., and L. D. HURST. 1998. Sensitivity of patterns
of molecular evolution to alterations in methodology: a cri-
tique of Hughes and Yeager. J. Mol. Evol. 47:493–500.
SWANSON, W. J., and V. D. VACQUIER. 1995. Extraordinary
divergence and positive Darwinian selection in a fusagenic
protein coating the acrosomal process of abalone sperma-
tozoa. Proc. Natl. Acad. Sci. USA 92:4957–4961.
TANAKA, T., and M. NEI. 1989. Positive Darwinian selection
observed at the variable-region genes of immunoglobins.
Mol. Biol. Evol. 6:447–459.
TIFFANY, H. L., J. S. HANDEN, and H. F. ROSENBERG. 1996.
Enhanced expression of the eosinophil-derived neurotoxin
ribonuclease (RNS2) gene requires interaction between the
promoter and intron. J. Biol. Chem. 271:12387–12393.
TSAUR, S.-C., and C.-I WU. 1997. Positive selection and the
molecular evolution of a gene of male reproduction,
Acp26Aa of Drosophila. Mol. Biol. Evol. 14:544–549.
VAQUIER, V. D. 1998. Evolution of gamete recognition pro-
teins. Science 281:1995–1998.
WARD, W. S., and D. S. COFFEY. 1991. DNA packaging and
organization in mammalian spermatozoa: comparison with
somatic cells. Biol. Reprod. 44:569–574.
Downloaded from https://academic.oup.com/mbe/article/16/5/706/2925467 by guest on 12 September 2021
WOLFE, K. H., P. SHARP, and W.-H. LI. 1989. Mutation rates
differ among regions of the mammalian genome. Nature
337:283–285.
ZHANG J., S. KUMAR, and M. NEI. 1997. Small-sample tests of
episodic adaptive evolution: a case study of primate lyso-
zymes. Mol. Biol. Evol. 14:1335–1338.
ZHANG, J., and M. NEI. 1997. Accuracies of ancestral amino
acid sequences inferred by the parsimony, likelihood, and
distance methods. J. Mol. Evol. 44(Suppl. 1):S139–S146.
ZHANG, J., H. F. ROSENBERG, and M. NEI. 1998. Positive Dar-
winian selection after gene duplication in primate ribonu-
clease genes. Proc. Natl. Acad. Sci. USA 95:3708–3713.
SHOZO YOKOYAMA, reviewing editor
Accepted February 8, 1999