Page 1 - Scani_et_al1995

Boll. Zool. 62: 59-70 (1995)
Downloaded by [31.185.101.124] at 08:47 23 March 2016Mate recognition and gamete cytology
features allow hybrid species
production and evolution in Bacillus
stick insects
VALERIO SCALI
FAUSTO TINTI
BARBARA MANTOVANI
OMBRETTA MARESCALCHI
Università di Bologna, Dipartimento di Biologia Evoluzionistica
Sperimentale, Sede Zoologia,
via S. Giacomo 9, I-40126 Bologna (Italy)
ABSTRACT
The increasing number of recognized hybrid unisexual complexes
among invertebrate and vertebrate animals has promoted in-
vestigations about their composition and origin. Morphological,
karyological and genetic (protein and DNA) analyses clearly show
that, owing to their persistence and incomplete reproductive
isolation from ancestors, several all-female complexes are much
more diversified than generally assumed and that they may also have
an evolutionary role. Here the case of the stick-insects of the genus
Bacillus is reported in some detail. This holomediterranean genus
comprises three well differentiated species that in Sicily have
hybridized repeatedly. The Bacillus mate-recognition system has not
followed the species-specific differentiation of the allozyme-coding
loci, allowing interspecific crosses to occur in areas of species sym-
patry with the production of two hybridogens, a corresponding
allodiploid parthenogen and a trihybrid triploid parthenogenetic
species. Hybridogenetic females eliminate the paternal haploset
(grandii) while passing the unassorted rossius hemiclone to off-
spring, which will be again of F, hybrid structure through a real fer- teraction with the environment establish the actual
tilization by host male sperm. The polyspermic eggs of the
hybridogens can also produce full-paternal fertile progeny of both
sexes (androgenetics), when mixis occurs between two sperm heads.
The parthenogenetic mechanism of the corresponding hybrid B.
whitei is very similar to the hybridogenetic one, excepting the
automictic re-use of the segregated grandii haploset; therefore B.
whitei offspring clonally maintain the maternal hybrid structure. The
trihybrid B. lynceorum produces clonal descendants through an
apomictic mechanism undergoing two seemingly normal meiotic
divisions. Each Bacillus hybrid actually realizes a different egg
maturation process; however, the three share one important feature:
an intrameiotic DNA extra-doubling, leading to the formation of
four-stranded chromosomes, and enabling the meiotic system to
produce balanced gametes even under different ploidy level and
hybrid structure. The extra-round of DNA synthesis seems to be
triggered by the hybrid condition impairing the synaptic process.
Also the parthenogenetic B. whitei produces androgenetics and it is
even capable of incorporating a third genome into its automictic but
clonal eggs, following fertilization by B. grandii or B. rossius males 1993).
with the production of fertile «synthetic» triploids. These findings
are evidence of clonal unisexuals reproductively interacting with
related bisexuals and also suggest that evolutionary pathways have
been undertaken by Bacillus. Also other unisexual complexes seem
to have undergone similar microevolutionary steps and their
reproductive traits and persistence, longer than commonly assumed,
make increasingly difficult to think of the whole of unisexuals as
«dead ends» or «blind alleys».
KEY WORDS: Androgenesis - Clonal species - Hybrid
gametogenesis - Hybridogenesis - Hybrid/ancestor interactions.
ACKNOWLEDGEMENTS
This research has been supported by grants from the Italian
Ministero dell'Università e della Ricerca Scientifica e Tecnológica
(MURST), and Consiglio Nazionale delle Ricerche (CNR).
Invited paper at the Symposium «Speciazione: tempi e modi», 55th
Congress of the Unione Zoológica Italiana, Torino, September
27-October 2, 1993.
INTRODUCTION
The origin of thelytokous species through interspecific
hybridization has occurred several times in both inver-
tebrate and vertebrate animals. Among the main con-
straints that hybrid speciation has to face, are the
widespread occurrence of species-specific mate
recognition systems, effectively limiting heterospecific
matings in the field, and hybrid meiotic disruption. In
several instances, perhaps related to habitat changes, pre-
mating isolation is overcome however, and, when viable
hybrids are produced, it often occurs that only the males
are sterile, whereas the females are able to produce balan-
ced gametes (Coyne, 1985, 1992; Moritz et al, 1989;
Vrijenhoek, 1989). These features appear responsible for
the rise of all-female hybrid complexes. Hybrid species
have been widely acknowledged and analyzed in ver-
tebrates, among which, at present, they represent about
0.1 of species. Clonal fish and amphibian hybrids are
gynogenetic unisexuals, whereas the reptilian ones are
parthenogens (Vrijenhoek et al., 1989 for a check-list).
This clear-cut interclass difference shows that in bony
fishes and urodeles clonal reproduction is strictly sperm-
dependent, while in hybrid lizards it can be achieved
without insemination (Dawley, 1989; Mantovani & Scali,
1992). Among invertebrates a patchy occurrence of
mostly parthenogenetic unisexual hybrids has been
reported and a few new instances have been described
recently (Bell, 1982; Bullini & Nascetti, 1990; Mantovani
& Scali, 1990; Mantovani et al, 1991a, b; Scali et al,
1991; Claridge & de Vrijer, 1993).
The genetic structure of all-female hybrids is mainly af-
fected by the differentiation level achieved by hybrid-
originating ancestors and by their meiotic system
alterations; polyploidy as well usually adds to hybrid
variability. The combination of these factors and their in-
genetic structure and diversification of the hybrid species.
Nonetheless, unisexuals are commonly assumed to be
genetically invariant and destined to last shortly on an
evolutionary timescale (Maynard Smith, 1978; White,
1978). These assumptions hold for some unisexuals, but
they cannot be uncritically applied to all of them. We
acknowledge that in most cases the assessment of an-
tiquity and of the evolutionary rate of unisexuals is pure
guessing, because adequate analyses are still lacking;
however, mitochondrial DNA studies have suggested that
some vertebrate hybrids might have persisted much
longer than supposed, even for millions of years and cer-
tainly for hundreds of thousands. In these cases, the
mutation rate becomes increasingly important since
thelytokous strains pass through hundreds of thousands
of generations (Hedges et al, 1992; Moritz et al. 1992;
Quattro et al, 1992a; Spolsky et al, 1992; Vrijenhoek,
Besides mutation, it seems that still another source of
variability has played a role: unisexuals, including par-
thenogens, are not reproductively isolated from related