# Full Text: TwoLineages

> Extracted from `2017_TwoLineages.pdf`

---

## Page 1

NEWS AND VIEWS
PERSPECTIVE
Two lineages that need each other
D. M. GORDON
and D. A. FRIEDMAN
Department of Biology, Stanford University, Stanford, CA, USA
Received 28 November 2016; revision received 14 December
2016; accepted 14 December 2016
In ants as in bees, a diploid female is either a reproductive
or worker. In honeybees, female larvae fed a high protein
substance known as ‘royal jelly’ become reproductives,
while those not fed the necessary nutrients become work-
ers (Linksvayer et al. 2011). Feeding experiments in ants
suggested that like honeybees, the fate of a diploid female
egg is usually determined as a larva by food supply (Brian
1951). It was thus unexpected to discover about 15 years
ago (Volny & Gordon 2002; Cahan & Keller 2003), using
microsatellite markers, that in some populations in the har-
vester ant genus Pogonomyrmex, there is an association
between genotype and reproductive status. There are two
interdependent lineages. Matings between a reproductive
and a male of the same lineage produce daughter repro-
ductives, while matings between a female reproductive
and a male of the other lineage produce daughter workers
(Fig. 1). The haploid males are produced from unfertilized
eggs. The two lineages need each other because a colony
cannot produce offspring colonies without reproductives,
and it cannot raise and maintain reproductives without
workers. In this issue of Molecular Ecology, Romiguier
et al. use RNA sequencing to demonstrate a similar system
in Messor, another harvester ant genus.
The dependent-lineage system raises intriguing evolu-
tionary questions. One is how this process is related to
hybridization. The first, and original sense of ‘hybridogene-
sis’, described in frogs, is a hemiclonal mode of sex deter-
mination that produces two distinct sympatric genomes. In
ants, however, diploids are female, and the developmental
fate of a diploid egg is not due purely to genetic factors. In
Pogonomyrmex populations with the dependent-lineage sys-
tem, mated queens produce worker and reproductive eggs
in the ratio in which they mated with the same and other
lineages (Schwander et al. 2006). But colonies do not pro-
duce adult reproductives until the colony has reached a
mature size, at about 5 years in P. barbatus (Gordon 1995).
Thus, the decision whether to produce reproductives is in
part made by workers, who apparently eat the reproduc-
tive eggs while the colony is small and young. It is possible
that
the
process
that
leads
to
the
development
of
reproductives requires some kind of special feeding from
the workers, as in honeybees and other species of ants. All
of this contributes to the determination of the reproductive
status of females.
‘Hybridogenesis’ in ants sometimes refers to the hypoth-
esis that the dependent-lineage system is the result of
hybridization
between
related
sympatric
species.
This
hypothesis for Pogonomyrmex harvester ants is controversial
and has not been supported by phylogenetic evidence
(Anderson et al. 2008; Sirvi€o et al. 2011; Mott et al. 2015). It
seems that there may have been hybridization in the past,
but not in the present, and that even within populations,
different systems for producing workers and reproductives
may be at work. When we have a broader set of examples
of species with dependent-lineage systems, it will be inter-
esting to consider whether the origins of such systems are
in fact related to hybridization among related species.
Another use of ‘hybrid’ is that workers are produced by
two independent lineages. They are ‘hybrids’ in the sense
that they are the product of two lineages without gene flow
between them. But those two lineages are both contributing
to the production of colonies, with parent colonies of the
same species, in a single species that cannot persist without
both lineages. Thus, in the dependent-lineage system,
female reproductives and workers are not hybrids in the
usual sense of having parents that come from different spe-
cies that can each reproduce independently.
The dependent-lineage system also raises interesting eco-
logical questions. In dependent-lineage systems, a queen
must mate more than once, because a queen must mate with
at least one male of each lineage to produce a viable colony
with both reproductives and workers. This means, as
Romiguier et al. point out, that species with the dependent-
lineage system are likely to be among the thousands of ant
species with large, population-wide mating aggregations.
Are there other ecological factors associated with depen-
dent-lineage mating systems? This study by Romiguier et al.,
using RNA sequencing, and the recent study by Norman
et al. (2016), now demonstrate a dependent-lineage mating
system in a second genus of seed-eating ants. An intriguing
feature of the dependent-lineage system in Pogonomyrmex
harvester ants is that the ratio of colonies of the two lineages
(colonies with a founding queen of a particular lineage)
tends to be strongly asymmetrical. For the same reasons that
sex ratios ought to be 1:1 – the two lineages need each other
– so should the lineage ratios, but they are not. In a long-
term study of a Pogonomyrmex population, with a persis-
tently asymmetrical ratio of colonies of the two lineages, we
found no ecological differences between colonies of the two
lineages in survival, mortality, nest mound size or foraging
behaviour (Gordon et al. 2013). This suggests that the depen-
dent-lineage system may be tenuously maintained, not by
Correspondence: D. M. Gordon; Email: dmgordon@stanford.edu
© 2017 John Wiley & Sons Ltd
Molecular Ecology (2017) 26, 975–976

## Page 2

ecological forces, but merely by differences between lineages
in male production.
The two harvester ant genera in which such systems have
been discovered, Messor and Pogonomyrmex, are within the
same subfamily. But this may be a coincidence. It appears
that two-lineage systems are widespread among the more
than 14 000 species and about 20 subfamilies of ants (Nor-
man et al. 2016). It seems likely that the more we look for
these systems in other genera and subfamilies, the more we
will find them. The methods introduced by Romiguier et al.,
using high-throughput sequencing, will contribute greatly
to this effort. Until we investigate the ants more broadly, it
is difficult to interpret the significance of the discovery of
similar systems in two harvester ant genera.
An intriguing question is why it is apparently so easy
for such diverse systems to evolve in ants. In the Pogono-
myrmex system, for the males of one lineage, matings with
the other lineage are a dead end, because their offspring
are sterile workers. Yet each lineage needs the other, as a
colony cannot raise reproductives without workers. One
feature of the Messor system suggested here is that there
may be occasional gene flow across lineages, if workers
can produce males carrying alleles from both lineages; this
may also occur in Pogonomyrmex, although population
genetic data show this has not happened recently (Curry
et al. 2010). How exactly such gene flow would contribute
to the progeny of those males remains an open question.
The genetic systems of ants, like everything else about
them, are likely to be extremely diverse.
References
Anderson KE, Linksvayer TA, Smith CR (2008) The causes and
consequences of genetic caste determination in ants (Hymenop-
tera: Formicidae). Myrmecological News, 11, 119–132.
Brian
MV
(1951)
Caste
determination
in
a
Myrmicine
ants.
Experientia, 7, 182.
Cahan SH, Keller L (2003) Complex hybrid origin of genetic caste
determination in harvester ants. Nature, 424, 306–309.
Curry MM, Wheeler DE, Yang K, Anderson KE (2010) The poten-
tial for gene flow in a dependent lineage system of a harvester
ant: fair meiosis in the F1 generation. Journal of Heredity, 101,
378–384.
Gordon DM (1995) The development of an ant colony’s foraging
range. Animal Behaviour, 49, 649–659.
Gordon DM, Pilko A, De Bortoli N, Ingram KK (2013) Does an eco-
logical advantage produce the asymmetric lineage ratio in a har-
vester ant population? Oecologia, 173, 849–857.
Linksvayer TA, Kaftanoglu O, Akyol E et al. (2011) Larval and
nurse worker control of developmental plasticity and the evolu-
tion of honey bee queen-worker dimorphism. Journal of Evolu-
tionary Biology, 24, 1939–1948.
Mott BM, Gadau J, Anderson KE (2015) Phylogeography of Pogono-
myrmex barbatus and P. rugosus harvester ants with genetic and
environmental caste determination. Ecology and Evolution, 5,
2798–2826.
Norman V, Darras H, Tranter C, Aron S, Hughes WOH (2016)
Cryptic
lineages
hybridize
for
worker
production
in
the
harvester ant Messor barbarus. Biology Letters, 12, 20160542.
Romiguier J, Fournier A, Yek SH, Keller L (2016) Convergent evo-
lution of social hybridogenesis in Messor harvester ants. Molecu-
lar Ecology, 26, 1108–1117.
Schwander T, Cahan SH, Keller L (2006) Genetic caste determina-
tion in Pogonomyrmex harvester ants imposes costs during colony
founding. Journal of Evolutionary Biology, 19, 402–409.
Sirvi€o A, Pamilo P, Johnson RA et al. (2011) Origin and evolution
of the dependent lineages in the genetic caste determination sys-
tem of Pogonomyrmex ants. Evolution, 65, 869–884.
Volny VP, Gordon DM (2002) Genetic basis for queen–worker
dimorphism in a social insect. Proceedings of the National Academy
of Sciences of the USA, 99, 6108–6111.
Fig. 1 The dependent-lineage system in
harvester ants. Matings between a repro-
ductive and a male of the same lineage
produce daughter reproductives, while
matings between a female reproductive
and a male of the other lineage produce
daughter workers.
© 2017 John Wiley & Sons Ltd
976 NEWS AND VIEWS: PERSPECTIVE


---
*Extraction method: pymupdf*