Available online at www.sciencedirect.com
ScienceDirect
Regulation, development, and evolution of caste ratios
in the hyperdiverse ant genus Pheidole
Angelica Lillico-Ouachour and Ehab Abouheif
Ant colonies are considered complex biological systems major evolutionary transitions, where selection begins to
because many individuals are divided into different castes that act on a higher functional level [6–8]. Examples range
interact to efficiently perform their tasks. Colonies in the from genes to gene networks, from single cells to multi-
hyperdiverse ant genus Pheidole have evolved a worker caste cellular organisms, or from individuals to societies [6–8].
with at least two subcastes: soldiers and minor workers. The
proportion of soldiers and minor workers in a colony has a
Accordingly, a single ant colony is a complex biological
major impact on the colony’s fitness and is tightly regulated.
system with respect to its organization, development, and
Here, we summarize over 100 years of research on the internal,
regulation and can be used as a model to gain insight into
external, and developmental factors that regulate subcaste
other systems. However, ant colonies provide an advan-
production as well as influence subcaste evolution in
tage over other systems because their individual parts are
Pheidole. We hope that summarizing these factors into a
obvious and their colonies can be taken apart and reas-
network of interactions will provide insight into how complex
sembled, what E.O. Wilson has termed as the ‘pseudo-
biological systems regulate, develop, and evolve.
mutant’ technique [9–12 ]. For example, colonies have
Address long been called ‘superorganisms’ and are often compared
Department of Biology, McGill University, 1205 Avenue Docteur
to how multicellular organisms develop and evolve
Penfield, Montre´ al, QC, Canada H3A 1B1
[13 ,14 ,15 ]. Ant colonies have a morphological and
Corresponding author: Abouheif, Ehab ([email protected]) reproductive division of labour within the colony, where
a queen caste primarily functions as the germline, while a
worker caste functions as the soma and performs all other
Current Opinion in Insect Science 2017, 19:43–51
tasks [3 ,10,16,17,18 ]. Like ant colonies, organismal de-
This review comes from a themed issue on Development and
velopment also has a division of labour, which unfolds
regulation
after cells differentiate and acquire their distinct fates,
Edited by Haruhiko Fujiwara and Yoshinori Tomoyasu
generating an incredible degree of cellular diversity
[19,20]. Cellular differentiation is thought to be regulated
by the action of gene regulatory networks that respond to
http://dx.doi.org/10.1016/j.cois.2016.11.003 inputs from both internal and external cues [21–24]. In
ants, the eventual caste fate of eggs and larvae is also
2214-5745/# 2016 Elsevier Inc. All rights reserved.
determined by the action of gene regulatory networks
that respond to hormonal, chemical, social, and environ-
mental cues [25,26 ,27]. The internal and external cues in
ants are equivalent to the signals that a totipotent (undif-
ferentiated) field of cells receive from short-range and
Introduction long-range signalling molecules, such as the Decapenta-
plegic morphogen gradient [28], which induces differen-
Complex systems are ubiquitous, spanning from the
tiation and determines the fate of particular cells within biological (individuals, populations, and ecosystems) to
the embryo [15 ].
the physical (turbulence and weather) to the abstract (free
markets and the Internet) [1,2]. Despite this diversity, all
complex systems are thought to exhibit similar underly- We focus on the hyperdiverse ant genus Pheidole because
ing principles. For a unit to be called a complex system, it their behaviour, physiology, and development has been
must have many interacting components that are respon- studied extensively [25,29–31,32 ,33–40]. Unlike most
sive to feedback and produce an emergent and robust ants, the worker caste is completely sterile and is divided
organization [2]. A fundamental and enduring challenge into at least two morphologically distinct subcastes: ‘mi-
in biology is to uncover how complex systems develop, nor workers’ and ‘soldiers’ (Figure 1). Minor workers
evolve, and regulate themselves. We often refer to the perform most tasks including maintaining the nest, for-
organization of complex systems in biology as a ‘division aging, and caring for brood, while soldiers defend the nest
of labour,’ meaning that different parts within a system and help process food, like cracking large seeds
efficiently perform tasks [3 ,4,5]. This division of labour [35,41,42 ,43,44 ,45]. Soldiers are adapted to this role
plays an important role in the evolution of complex with disproportionately larger heads and mandibles than
systems because increasing specialization between inter- minor workers (Figure 1) [45,46]. The ratio of soldiers:
acting parts is thought to be a driving force behind several minor workers in a colony ranges from 5% soldiers: 95%
www.sciencedirect.com Current Opinion in Insect Science 2017, 19:43–51
44 Development and regulation
Figure 1
The effect of nutrition on the production of soldiers in a
colony was established experimentally by Goetsch [58,60]
(a) (b) in Pheidole pallidula after initial observations made by
Wheeler [61] and Emery [62], and further investigated by
Passera [48 ]. Goetsch [58,60] fed larvae a honey-rich and
sugar water-rich diet (low protein) or a mealworm-rich
diet (high protein) at alternative times during develop-
ment. Only when larvae were fed a protein-rich diet were
soldiers able to be produced [58,60]. This study was the
first to suggest that there is a critical period during larval
development that mediates the soldier to minor worker
switch [58,60]. The confirmation of this critical period and
its connection to JH were investigated later by Wheeler
Current Opinion in Insect Science and Nijhout [59 ], who applied a JH analogue called
methoprene to Pheidole bicarinata at different stages of
Subcastes of Pheidole spadonia. (a) Minor worker and (b) soldier. larval development. During a critical period in the 4th
Images to scale. larval instar, methoprene was able to induce the soldier
developmental pathway [59 ]. Methoprene effectively
delays metamorphosis by several days such that larvae
continue to develop past the point where minor workers
normally pupate and, instead, pupate at a larger terminal
size [51,59 ]. To date, nutrition and JH titres have been
studied independently in Pheidole and associated by the
period in which larvae are sensitive to their effects [60].
minor workers to 25% soldiers: 75% minor workers Research that establishes a direct link between a larva’s
[26 ,29,37,38,47,48 ,49–53]. The ability of Pheidole colo- diet and the JH pathway is needed for a thorough physio-
nies to respond plastically to environmental changes on logical understanding of this developmental decision [60].
shorter time scales and to evolve different subcaste ratios
over longer time scales depends on both genetic and Inhibition of the soldier programme: pheromones
environmental factors. This ability likely played a key Complex systems rely on positive and negative feedback
role in the evolutionary and ecological success of loops for their generation and maintenance [2]. We have
Pheidole. Here, we review the literature on the factors already summarized a key activating influence on soldier
that affect the regulation, development, and evolution of production — nutrition — and now we will consider its
subcaste ratios in Pheidole. complementary inhibitory influence, namely the soldiers
themselves. Evidence that adult soldiers negatively reg-
Internal influences ulate soldier production came first from Gregg [63 ],
Activation of the soldier programme: nutrition and where he perturbed worker subcaste ratios and found
juvenile hormone that they influence the production of soldiers in the next
In Pheidole, individuals undergo a series of developmental generation; increasing the proportion of soldiers decreases
switches that determine queen and worker castes. The soldier production, whereas decreasing the proportion of
first developmental switch is controlled by haplo-diploid soldiers increases soldier production [48 ,63 ]. Further-
sex determination, where unfertilized (haploid) embryos more, Wheeler and Nijhout [26 ] showed that colonies
become males and fertilized (diploid) embryos become with the most soldier contact suppressed the soldier
females [54]. The second developmental switch is medi- programme more effectively than those with the least
ated by Juvenile Hormone (JH) and occurs shortly after contact [26 ]. They treated P. bicarinata larvae with
the first switch, where individuals differentiate into either methoprene to induce the soldier programme and raised
reproductive queens or sterile workers (Figure 2; blue them either in a high soldier contact environment (40 free
lines) [55]. This switch is season-dependent; seasonal soldiers and 40 minor workers) or low soldier contact
cues, such as temperature, are thought to activate JH environment (20 free soldiers and 40 minor workers)
production to induce the queen programme [41,55,56,57]. [26 ]. The high soldier contact environment inhibited
The third developmental switch differentiates worker soldier production more than the low soldier contact
larvae into either soldiers or minor workers (Figure 2; environment [26 ]. Together, these experiments show
blue lines) [27]. This switch is also mediated by JH and is that the higher proportion of soldiers in a colony, the
thought to be largely influenced by nutrition; a protein- greater the degree of inhibition [26 ].
rich diet activates JH production to prolong development
and induce the soldier programme (Figure 2; blue lines) The inhibition of soldier production in response to a
[27,48 ,58,59 ,60]. larger than normal proportion of soldiers in a colony
Current Opinion in Insect Science 2017, 19:43–51 www.sciencedirect.com
The worker caste system in Pheidole ants Lillico-Ouachour and Abouheif 45
Figure 2
competition
resources available inhibitory pheromone
nutrition
queen soldier minor worker
+ –
JH
– +
JH
seasonal colony colony cues ontogeny size
Current Opinion in Insect Science
Network model of the major factors regulating subcaste ratios in Pheidole. Solid lines indicate regulatory mechanisms supported by sufficient
evidence, while dashed lines indicate regulatory mechanisms supported by evidence but requiring further study. Lines with arrowheads indicate
activation, while lines with perpendicular bars indicate repression. Internal influences (blue lines). Two developmental switches are mediated by
JH: an early switch in embryogenesis determines queen or worker fate, while a late switch in larvae determines soldier or minor worker fate.
Nutrition promotes soldier production by activating JH. Soldiers suppress soldier production though the soldier inhibitory pheromone. External
influences (green lines). Resources available in the habitat promote soldier production by providing more nutrition to the colony or effectively
decreasing the numbers of soldiers in the colony by recruiting them to large finds. Competition increases soldier production by promoting the
feeding of larvae by minor workers or increasing the death of adult soldiers and thereby alleviating the effect of the soldier inhibitory pheromone
on larvae. Colony development and life cycle influences (red lines). Seasonal cues increase the production of virgin queens by promoting JH.
Virgin queens reduce the amount of nutrition available to activate the soldier programme, thereby reducing soldier production. As a colony ages
(colony ontogeny), the colony’s size and workforce increases. This effectively promotes soldier production because more minor workers can finally
provide sufficient nutrition to larvae.
can be explained by three alternative hypotheses: first, soldier programme in colonies with minor workers and
soldiers may modulate their behaviour or be inherently caged soldiers, where soldiers could contact minor work-
less effective at caring for and feeding brood, second, ers but could not directly contact brood. Caged soldiers
soldiers may emit a signal (either pheromonal, beha- suppress the soldier programme in developing larvae
vioural or morphological) to the minor workers, inducing which eliminates the first hypothesis (soldier behaviour
them to modulate the way they raise the brood, and third, or ability to care for brood) [26 ]. Sempo and Detrain [30]
soldiers may emit an inhibitory pheromone that directly provide key evidence to distinguish between the second
affects larval development. Wheeler and Nijhout [26 ] (modulation of minor worker behaviour) and third (inhi-
tested the first possibility, whether soldiers adequately bition of larval development by action of a pheromone)
feed larvae and rear them to adulthood, and found that hypotheses; they varied subcaste ratios and observed that
soldiers rear 100% of brood through to adulthood and minor workers do not change their social behaviours at
disseminate food rapidly and effectively. This shows that different subcaste ratios [30]. This shows that modulation
soldiers do not affect the survival of brood or underfeed of minor worker behaviour does not account for soldier
them [26 ]. Furthermore, Wheeler and Nijhout [26 ] inhibition. Therefore, soldiers regulate the numbers of
raised larvae treated with methoprene to induce the future soldiers produced in a colony through the action of
www.sciencedirect.com Current Opinion in Insect Science 2017, 19:43–51
46 Development and regulation
a pheromone that directly inhibits the soldier develop- selection [2,64,65]. Resource availability and competition
mental programme in larvae (Figure 2; blue lines) are two such environmental challenges that stress biolog-
[26 ,63 ]. Wheeler and Nijhout [26 ] propose that the ical systems in nature [66]. Ant colonies are no different; if
inhibitory pheromone decreases the sensitivity of the they cannot buffer against competition, the colony would
larvae to JH, effectively raising the JH threshold for inevitably be eliminated. For example, one way Pheidole
the production of soldiers. colonies can temper this burden is through behavioural
flexibility; if minor workers are lost defending the colony,
There are several exciting directions for future research soldiers will expand their behaviour to include more
on the soldier inhibitory pheromone. One is to elucidate brood care tasks [35,38,67]. In this section, we will discuss
the nature and chemical composition of the pheromone how Pheidole subcaste ratio is generally regulated by
itself, as well as its transmission. Although we know it is external influences as a means of modulating environ-
not necessary for soldiers to be in direct contact with the mental challenges on the colony.
brood for it to be effective, it is unclear whether the
chemical is easily diffusible and transmitted passively or Resource availability
whether the chemical is non-volatile and transmitted Diet is a key activator of the soldier programme at the
through contact. If the pheromone is contact-mediated, scale of individual development. Therefore, resource
it will be important to work out what type of contact is availability at an ecological scale could potentially influ-
required. Once we are able to distinguish between these ence activation of the soldier programme in a dramatic way
alternative possibilities, the pheromone’s physiological and alter the subcaste ratio in colonies [48 ,63 ]. A colony’s
and molecular mode of action on the receiver may be ability to respond to changes in food supply at an ecologi-
resolved, including its relation to the JH pathway. cal scale would be a testament to the robustness of the
superorganism. McGlynn and Owen [68] manipulated the
Finally, both activation and inhibition of the soldier pro- food available to Pheidole flavens colonies in Costa Rica by
gramme contribute to the regulation of subcaste ratios, but providing colonies in the field with clumped or split
it remains unclear how these two factors interact within protein-rich food sources. Food supplementation in-
the social context of the colony. Activation of the soldier creased the number of soldier pupae observed in the
developmental programme by minor workers can occur in colony, especially when food was supplemented in a
at least two different ways: first, minor workers may clumped manner [68]. This treatment may affect subcaste
constantly activate particular larvae by feeding them with ratios through manipulations of nutrition or through
more protein, or second, minor workers may constantly manipulations of soldier presence (Figure 2; green lines).
activate all larvae by feeding them equally. The first If resource availability affects subcaste ratios by way of
possibility implies that activation by minor workers is nutrition, then access to additional protein in the envi-
the primary mechanism for maintaining subcaste ratios ronment is used by the colony to activate the soldier
and the soldier inhibitory pheromone largely fine tunes programme in a greater number of larvae [68]. Alterna-
them. This raises questions about how minor workers tively, if larger and more abundant resources are available,
choose these particular larvae and what role larvae play then soldiers may be recruited for foraging; soldier recruit-
in this choice, that is, do certain larvae beg for food more ment lowers the inhibitory influence of soldier presence
than others? Alternatively, the second possibility implies on larvae in the nest and leads to increased soldier pro-
that minor workers are always activating all larvae and that duction overall [68]. Yet, studies which attempt to connect
the soldier inhibitory pheromone is the primary mecha- caste ratios with ecological correlates, like food abundance
nism for maintaining subcaste ratios. This raises questions or limitation, have not found an association between
about when soldiers produce the soldier inhibitory phero- natural resource availability and the proportion of soldiers
mone and how larvae sense it. Do soldiers constantly in a colony [37,52]. In fact, Yang [69] found that colonies of
produce the soldier inhibitory pheromone but larvae only Pheidole morrisi respond to seasonal changes in food distri-
respond at a certain threshold by inhibiting soldier devel- bution with increased fat stores and behaviourally ‘replete’
opment? Or, do larvae always respond to the soldier soldiers, not with changes in subcaste ratio or body size. It
inhibitory pheromone but soldiers only produce it once is possible that resource availability was not the limiting
they reach a threshold? Formally testing all of these factor for soldier production and other ecological correlates
alternative possibilities is an important future direction. are more important in these populations.
Clearly, we have only scratched the surface in terms of our
understanding of how these internal influences interact to Competition
regulate subcaste ratios in Pheidole. Competition imposes an ecological challenge on any
given colony. Because soldiers work to defend the nest,
External influences regulating soldier production in response to competitive
For a complex biological system to be robust it must be interactions would confer an adaptive advantage for the
able to respond to short-term and long-term environmen- colony. To test this possibility, Passera et al. [70 ] studied
tal challenges or the system will be eliminated by natural the influence of intraspecific competition on subcaste
Current Opinion in Insect Science 2017, 19:43–51 www.sciencedirect.com
The worker caste system in Pheidole ants Lillico-Ouachour and Abouheif 47
ratios. In their experiment, P. pallidula colonies were ly produce more and larger minor workers typical of that
made to perceive, but not directly come into contact found in mature colonies [47]. Similarly, after a minimum
with, a foreign conspecific colony [70 ]. Exposed colonies number of minor workers have been produced, the queen
upregulated the number of soldier pupae and adults over begins to produce nanitic soldiers and then gradually
several weeks [70 ]. Furthermore, studies have also fo- produces more and larger soldiers typical of that found
cused on the effect of interspecific competition on sub- in mature colonies [47]. This has been shown in Pheidole
caste ratios. Yang et al. [44 ] found that P. morrisi colonies obtusospinosa, Pheidole rhea, and Pheidole spadonia, where
in geographic areas exposed to more intense interspecific the distribution of soldier head width increases as colonies
competition from the red fire ant, Solenopsis invicta, are reach their full size [47]. The effect of colony ontogeny on
composed of more soldiers than colonies in areas with less the proportion and physical size of the soldier subcaste
competition. Yang et al. [44 ] was able to then link the may be partially due to nutrition (Figure 2; red lines) [41].
proportion of soldiers in a nest with their success at During early founding stages, the amount of resources
defending against fire ants; having more soldiers available to the colony is limited by a small workforce
decreases the time it takes to kill fire ants. Further [41]. Once established, the large workforce is able to
support comes from Ito and Higashi [50], who found that provide the queen and her brood with a sufficient diet
a larger defense zone (an area in which individuals are for soldier production [41].
likely to encounter competitors) is associated with having
more soldiers in a colony. Competition may regulate If this positive relationship between increasing caste
subcaste ratios, then, by altering subcaste proportions size, proportion of soldiers, colony ontogeny and nutri-
because soldiers are eliminated in combat thereby allevi- tion [74] is a general feature of Pheidole colonies, then
ating the effect of the soldier inhibitory pheromone on colony ontogeny and colony size should be linked. How-
larvae or through altering minor worker behaviour to ever, support for this link has been debated. Kaspari and
promote soldier production via nutrition (Figure 2; green Byrne [52] provide evidence that faster growing colonies
lines). Yet, a study which stressed Pheidole dentata with fire of Neotropical Pheidole invest more in defense by pro-
ants for 19 weeks did not result in changes in subcaste ducing a larger proportion of soldiers, but show that the
proportion when compared to controls stressed with a increased proportion of soldiers is related to colony
‘non-competitive’ species, Tetramorium caespitum [71]. ontogeny and not colony size. However, they could
This lack of subcaste ratio regulation in response to fire not track the ontogeny of individual colonies to confirm
ants could be real and P. dentata may be different from that there is indeed a relationship between size and
other Pheidole species. In support of this explanation, ontogeny because of the general challenges of perform-
some observational field studies also argue that competi- ing this kind of study in the field [52]. Therefore, future
tion does not influence subcaste composition on P. dentata studies should track subcaste composition during colony
[37]. Alternatively, this result may be the consequence of maturation to dissociate the influence of colony size and
the experimental control used–P. dentata colonies may nutrition on colony ontogeny.
perceive any foreign workers as competition and regulate
subcaste ratios accordingly. If so, T. caespitum should not Life cycle and reproductive investment
be classified as ‘non-competitive,’ making it invalid as a During spring or early summer, seasonal cues like in-
control. It therefore remains unclear whether the results creased photoperiod and elevated temperature prompt
of these conflicting studies in P. dentata can be general- mature colonies to invest in reproductive queens and
ized to other Pheidole species. Future research on different males [71]. During this period a ‘developmental trade-
species of Pheidole and of competitors in a controlled off’ has been proposed where queens and males are
environment is necessary to substantiate this relationship produced at the expense of soldiers. The basis of this
mechanistically and at an ecological level. The associa- developmental trade-off is: first, queens and soldiers are
tion between competition and subcaste proportion is similar in size and are energetically costly; and second, the
critical to the notion that the demography of the worker developmental switch for queens and workers is prior to
caste is adaptive, so this is an important area where much that of soldiers and minor workers (Figure 2; red lines)
work is still needed. [27,55,71]. Observations of caste investment in P. dentata
and P. morrisi support the developmental trade-off hy-
Colony development and life cycle pothesis, particularly the observation that queens and
Ontogeny and colony size soldiers are produced largely at different times of year
Subcaste composition is influenced by the development [71,72 ]. However, this inverse relationship between
of the colony as it grows from when the queen founds the queen and soldier investment has not been observed
colony to when it reaches its full size. Winged virgin across all Pheidole species tested and, in some cases, some
queens take part in mating flights in spring or summer studies have reported a positive relationship [50,52]. Ito
[72 ,73], and after they mate, they tear off their wings and and Higashi [50] reported no correlation between pro-
attempt to establish a colony [41]. At first, queens produce duction of queens and subcaste ratio in the Old World
small minor workers called ‘nanitic’ workers and gradual- species, Pheidole fervida [50], and in populations of
www.sciencedirect.com Current Opinion in Insect Science 2017, 19:43–51
48 Development and regulation
Neotropical Pheidole, when there is more queen biomass lead to the evolution of a quantitative increase in subcaste
there is an increase in soldier biomass [52]. ratio through genetic accommodation. They determined
the composition of the worker subcaste in geographically-
How can we account for disagreement between the separated P. morrisi populations along the east coast of the
results of these studies? First, some key methodological USA (New York, North Carolina, and Florida). Colonies
differences exist between the studies supporting the co-existing with an ecologically dominant competitor, the
developmental trade-off hypothesis and those that do fire ant, in Florida have a higher proportion of soldiers
not. Of the two studies that support the hypothesis, compared to colonies in New York and North Carolina
one was experimental in nature and conducted under without fire ants [44 ]. To determine if the relationship
uniform laboratory conditions, while the other was an between subcaste ratios and geography is the conse-
exhaustive field study where numerous whole colonies quence of microevolutionary divergence or phenotypic
were collected via wax-casting [71,72 ], and the studies plasticity, Yang et al. [44 ] performed common-garden
that did not support this hypothesis were both correla- experiments in which whole colonies were supplanted
tional field studies [50,52]. Second, species supporting into artificial nests in the lab. They found that the
this hypothesis have large colony sizes and live in North differences in soldier ratio between geographically-sepa-
America, while those not supporting this hypothesis have rated colonies were maintained when P. morrisi colonies
small colony sizes or located closer to the equator were removed from their natural environment. This
[50,52,71,72 ]. Third, because elevated temperature is shows that observed differences in subcaste proportion
known to increase soldier production in a laboratory was the consequence of microevolutionary divergence. A
setting [71], it is unclear how this would translate to similar study was conducted on P. megacephala; popula-
the field where seasonal changes in sexual production tions of P. megacephala from sites with different competi-
correspond to increased temperature. To identify wheth- tive environments were sampled for subcaste ratio [80].
er temperature and the other factors above contribute to They found that invasion of a soldiers and minor workers
subcaste regulation, it will be important to perform long- were biggest in the highly competitive habitats and minor
term experimental manipulations that identify their workers were smallest in the low competitive habitats, yet
effects on life cycle, subcaste size and subcaste ratio. these results were not correlated with soldier proportion
[80]. Differences in these two studies could stem from
Evolution methodology where Yang et al. [44 ] sampled whole
Pheidole colonies have evolved a remarkable degree of colonies of P. morrisi but Wills et al. [80] could not feasibly
developmental plasticity allowing them to dynamically do so because P. megacephala are polydomous and expan-
activate (through hormones) and inhibit (through phero- sive. Other explanations for the difference between
mones) the soldier developmental programme. On shorter the two studies could stem from resource limitation
time scales, this allows colonies to regulate subcaste ratios [80]. If P. megacephala are not resource limited like
to respond to challenges from their external environment, P. morrisi they may not exhibit developmental trade-offs
as well as changes in colony ontogeny and life cycle. Over due to competition [80].
longer time scales, however, this developmental plasticity
allows colonies to rapidly evolve to adapt to these continual Evidence suggests that the quantitative divergence in
changes and challenges from the environment. Develop- subcaste ratios between Pheidole populations, as shown by
mental plasticity can mediate evolution of both quantita- Yang et al. [44 ], are translated into quantitative diver-
tive and qualitative (novel) changes in subcaste gence in subcaste ratios between species. A study by
composition through evolutionary mechanisms known as McGlynn et al. [49] surveyed many species of Pheidole and
‘genetic assimilation’ and ‘genetic accommodation found that species with smaller body sizes produce
[32 ,75–78]. These mechanisms describe the evolution more soldiers. Species differences in colony composition
of the sensitivity of phenotypes to environmental inputs. and individual size are likely the consequence of envi-
Genetic assimilation occurs when initially plastic pheno- ronmental pressures that drive the evolution of develop-
types evolve to be less responsive to the environment mental plasticity through genetic assimilation or
[75,77]. In contrast, genetic accommodation occurs when accommodation [49].
phenotypes evolve to be more responsive to the environ-
ment [75,77]. These mechanisms can lead to both quanti- Evolution of novel subcastes
tative changes in subcaste ratio, such as from 10% soldier: Complex biological systems can evolve through quanti-
90% minor workers to 15% soldiers: 85% minor workers tatively shifting pre-existing components, but how do
[44 ,49], as well as qualitative changes in subcaste compo- systems evolve novel components? The evolution of an
sition leading to the evolution of novel subcastes [32 ,79]. additional worker subcaste in some Pheidole species is an
excellent example of how novelty in complex biological
Quantitative evolution of subcaste ratios systems arises [32 ]. At least eight Pheidole species have a
A pioneering study by Yang et al. [44 ] provided evidence third worker subcaste called supersoldiers, which are
that populations experiencing intense competition may disproportionately larger in head and body size than
Current Opinion in Insect Science 2017, 19:43–51 www.sciencedirect.com
The worker caste system in Pheidole ants Lillico-Ouachour and Abouheif 49
Figure 3
(a) (b) (c) (d)
Current Opinion in Insect Science
Castes and subcastes of Pheidole obtusospinosa. (a) Queen, (b) supersoldier, (c) soldier and (d) minor worker. Images to scale.
the soldiers [45] (Figure 3). Rajakumar et al. [32 ] dem- we can gain insight into other complex biological systems
onstrated that application of JH on larvae in species and thereby acquire a deeper understanding of the world
without a supersoldier caste was able to environmentally around us.
induce the development of supersoldiers. This result
shows that the supersoldier caste did not evolve de novo
Acknowledgements
in these eight species, but instead evolved once in the
We thank the Abouheif Lab for comments and Dominic Ouellette for
ancestor of all Pheidole and the phenotypic expression of technical assistance on the manuscript. This work was funded by an
NSERC Discovery Grant and the McGill University Tomlinson Science
supersoldiers was subsequently lost in almost all species
Award to E.A. and an NSERC Canada Graduate Scholarship (Master’s)
in the genus [32 ]. However, the ancestral genetic poten-
to A.L-O.
tial was not lost and was retained for 25–47 millions of
years in the genus, most likely because the same physio-
logical pathways regulate the development of soldier and References and recommended reading
Papers of particular interest, published within the period of review,
supersoldier subcastes [32 ,81]. The supersoldier sub-
have been highlighted as:
caste re-evolved at least 4 times in the genus after
of special interest
induction of the ancestral potential for supersoldiers
of outstanding interest
was fixed through genetic accommodation [75,77]. This
work on the supersoldier ants provides evidence that
1. Pagel MD: Encyclopedia of Evolution. Oxford University Press;
phenotypic variation induced by the environment can 2002.
lead to the evolution of new components in biological 2. Ladyman J, Lambert J, Wiesner K: What is a complex system?
Eur J Philos Sci 2013, 3:33-67.
systems and increase their complexity.
3. Anderson C, McShea DW: Individual versus social complexity,
Conclusion with particular reference to ant colonies. Biol Rev 2001,
76:211-237.
From tiny signalling molecules like hormones to major This review discusses complexity within social insects colonies high-
lighting the key features of ‘simple’ and ‘complex’ societies.
ecological stressors like competition, the factors that
influence Pheidole subcaste regulation are multifaceted 4. Bonner JT: The Evolution of Complexity by Means of Natural
Selection. Princeton University Press; 1988.
and intricately linked. Not only do these signals induce
reactive short-term changes in subcaste number, but they 5. Bonner JT: Dividing the labour in cells and societies. Curr Sci
1993, 64:459-466.
also produce the phenotypic variation required for natural
6. Michod RE: Evolution of individuality during the transition from
selection to act upon. We hope to have demonstrated that
unicellular to multicellular life. Proc Natl Acad Sci U S A 2007,
the dynamic relationships between the individuals in a 104:8613-8618.
superorganism and the superorganism with its environ-
7. Michod RE: On the transfer of fitness from the cell to the
ment are rich and complex. Furthermore, studying these multicellular organism. Biol Philos 2005, 20:967-987.
relationships may further our understanding of complex
8. Szathmary E, Smith JM: The major evolutionary transitions.
systems as whole, including the generation of a multicel- Nature 1995, 374:227-232.
lular individual through organismal development with 9. Wilson EO: In Search of Nature. Island Press; 1996.
networks upon networks of regulatory mechanisms gov-
10. Ho¨ lldobler B, Wilson EO: The Superorganism: The Beauty,
erning proliferation and differentiation. More broadly, by Elegance, and Strangeness of Insect Societies. New York, NY:
W.W. Norton & Company Inc.; 2009.
studying the tangible complex systems of eusocial insects
www.sciencedirect.com Current Opinion in Insect Science 2017, 19:43–51
50 Development and regulation
11. Wilson EO: The sociogenesis of insect colonies. Science behavioral plasticity in ants. Behav Ecol Sociobiol 2006,
(Washington) 1985, 228:1489-1495. 60:631-644.
12. Wilson EO: Epigenesis and the evolution of social systems. 32. Rajakumar R, San Mauro D, Dijkstra MB, Huang MH, Wheeler DE,
J Heredity 1981, 72:70-77. Hiou-Tim F, Khila A, Cournoyea M, Abouheif E: Ancestral
Wilson introduces the ‘pseudomutant’ approach and argues that it makes developmental potential facilitates parallel evolution in ants.
social insects an experimental model for the great challenges in biology, Science 2012, 335:79-82.
including developmental biology. This study reveals the significance of ancestral developmental potentials
for the origin of novel castes in ant colonies.
13. Yang AS: Thinking outside the embryo: the superorganism as a
model for EvoDevo studies. Biol Theory 2007, 2:398-408. 33. Patel AD: An unusually broad behavioral repertory for a major
This review highlights developmental processes in social insects and how worker in dimorphic ant species: Pheidole morrisi
they can bridge gaps in evolutionary developmental biology. (Hymenoptera, Formicidae). Psyche (Cambridge, MA) 1990,
97:181-191.
14. Kilfoil ML, Lasko P, Abouheif E: Stochastic variation: from single
cells to superorganisms. HFSP J 2009, 3:379-385. 34. Muscedere ML, Traniello JFA, Gronenberg W: Coming of age in
This review compares stochastic processes and networks at multiple an ant colony: cephalic muscle maturation accompanies
levels of biological organization. behavioral development in Pheidole dentata.
Naturwissenschaften 2011, 98:783-793.
15. Linksvayer TA, Fewell JH, Gadau J, Laubichler MD:
Developmental evolution in social insects: regulatory 35. Mertl AL, Traniello JFA: Behavioral evolution in the major
networks from genes to societies. J Exp Zool B: Mol Dev Evol worker subcaste of twig-nesting Pheidole (Hymenoptera:
2012, 318:159-169. Formicidae): does morphological specialization influence task
This review aims to connect networks, from gene regulatory networks to plasticity? Behav Ecol Sociobiol 2009, 63:1411-1426.
social networks, into a cohesive framework for integrating social evolution
36. Calabi P, Traniello JFA: Behavioral flexibility in age castes of the
and evolutionary developmental biology.
ant Pheidole dentata. J Insect Behav 1989, 2:663-677.
16. Oster GF, Wilson EO: Caste and Ecology in the Social Insects.
37. Calabi P, Traniello JFA: Social organization in the ant Pheidole
Princeton University Press; 1978.
dentata. Behav Ecol Sociobiol 1989, 24:69-78.
17. Bourke AFG: Colony size, social complexity and reproductive
38. Brown JJ, Traniello JFA: Regulation of brood-care behavior in
conflict in social insects. J Evol Biol 1999, 12:245-257.
the dimorphic castes of the ant Pheidole morrisi
18. Wheeler WM: Ants. 1910. (Hymenoptera: Formicidae): effects of caste ratio, colony size,
and colony needs. J Insect Behav 1998, 11:209-219.
This paper is the first to establish the ‘superorganism’ concept.
39. Wheeler DE, Nijhout HF: Imaginal wing discs in larvae of the
19. Valentine JW, Collins AG, Meyer CP: Morphological complexity soldier caste of Pheidole bicarinata vinelandica Forel
increase in metazoans. Paleobiology 1994, 20:131-142. (Hymenoptera: Formicidae). Int J Insect Morphol Embryol 1981,
10:131-139.
20. Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson J: Molecular
Biology of the Cell, Vol 1. New York/London: Garland Publishing 40. Shbailat SJ, Khila A, Abouheif E: Correlations between
Inc.; 1989. spatiotemporal changes in gene expression and apoptosis
underlie wing polyphenism in the ant Pheidole morrisi. Evol
21. Davidson EH: The Regulatory Genome: Gene Regulatory Networks
Dev 2010, 12:580-591.
in Development and Evolution. Academic Press; 2010.
41. Ho¨ lldobler B, Wilson EO: The Ants. Cambridge, MA: Belknap
22. Davidson EH, Levine MS: Properties of developmental gene
Press of Harvard University Press; 1990.
regulatory networks. Proc Natl Acad Sci U S A 2008,
105:20063-20066. 42. Huang MH: Multi-phase defense by the big-headed ant,
Pheidole obtusospinosa, against raiding army ants. J Insect Sci
23. Davidson EH, Erwin DH: Gene regulatory networks and the
2010, 10.
evolution of animal body plans. Science 2006, 311:796-800.
This study follows the development of single colonies in Pheidole species.
24. Davidson EH, Rast JP, Oliveri P, Ransick A, Calestani C, Yuh C-H,
43. Feener DH: Response of Pheidole morrisi to two species of
Minokawa T, Amore G, Hinman V, Arenas-Mena C et al.: A
enemy ants, and a general model of defense behavior in
genomic regulatory network for development. Science 2002,
Pheidole (Hymenoptera: Formicidae). J Kansas Entomol Soc 295:1669-1678.
1987, 60:569-575.
25. Abouheif E, Wray GA: Evolution of the gene network underlying
44. Yang AS, Martin CH, Nijhout HF: Geographic variation of caste
wing polyphenism in ants. Science 2002, 297:249-252.
structure among ant populations. Curr Biol 2004, 14:514-519.
This study presents experimental evidence from the lab and field for
26. Wheeler DE, Nijhout HF: Soldier determination in Pheidole
microevolution of body size and caste ratio of soldiers and minor workers
bicarinata: inhibition by adult soldiers. J Insect Physiol 1984,
in the ant Pheidole morrisi.
30:127-135.
This study reveals the existence of a soldier inhibitory pheromone for the
45. Wilson EO: Pheidole in the New World: A Dominant, Hyperdiverse
repression of soldier development.
Ant Genus, Vol 1. Harvard University Press; 2003.
27. Wheeler DE, Nijhout HF: Soldier determination in Pheidole
46. Pie MR, Traniello JFA: Morphological evolution in a
bicarinata: effect of methoprene on caste and size within
hyperdiverse clade: the ant genus Pheidole. J Zool 2007,
castes. J Insect Physiol 1983, 29:847-854. 271:99-109.
28. Affolter M, Basler K: The decapentaplegic morphogen gradient:
47. Huang MH, Wheeler DE: Colony demographics of rare soldier-
from pattern formation to growth regulation. Nat Rev Genet
polymorphic worker caste systems in Pheidole ants
2007, 8:663-674.
(Hymenoptera, Formicidae). Insectes Sociaux 2011, 58:539-549.
29. Wilson EO: The relation between caste ratios and division of
48. Passera L: Diffe´ renciation des soldats chez la Fourmi Pheidole
labor in the ant genus Pheidole (Hymenoptera: Formicidae).
pallidula Nyl. (Formicidae Myrmicinae). Insectes Sociaux 1974,
Behav Ecol Sociobiol 1984, 16:89-98. 21:71-86.
This study categorizes stages of larval development in Pheidole and
30. Sempo G, Detrain C: Social Task Regulation in the Dimorphic
associates soldier production with environmental cues like temperature
Ant, Pheidole pallidula: The Influence of Caste Ratio, Vol 10.
and nutrition.
2010.
49. McGlynn TP, Diamond SE, Dunn RR: Tradeoffs in the evolution
31. Seid MA, Traniello JFA: Age-related repertoire expansion and
of caste and body size in the hyperdiverse ant genus Pheidole.
division of labor in Pheidole dentata (Hymenoptera:
PLoS ONE 2012, 7:e48202.
Formicidae): a new perspective on temporal polyethism and
Current Opinion in Insect Science 2017, 19:43–51 www.sciencedirect.com
The worker caste system in Pheidole ants Lillico-Ouachour and Abouheif 51
50. Ito F, Higashi S: Tests of four hypotheses on soldier production, 67. Wilson EO: Between-caste aversion as a basis for division of
by using wild colonies of Pheidole fervida F. Smith labor in the ant Pheidole pubiventris (Hymenoptera:
(Hymenoptera: Formicidae). Res Popul Ecol 1990, 32:113-117. Formicidae). Behav Ecol Sociobiol 1985, 17:35-37.
51. Ono S: Effect of juvenile hormone on the caste determination 68. McGlynn TP, Owen JP: Food supplementation alters caste
in the ant, Pheidole fervida SMITH (Hymenoptera: Formicidae). allocation in a natural population of Pheidole flavens, a
Appl Entomol Zool 1982, 17:1-7. dimorphic leaf-litter dwelling ant. Insectes Sociaux 2002,
49:8-14.
52. Kaspari M, Byrne M: Caste allocation in litter Pheidole: lessons
from plant defense theory. Behav Ecol Sociobiol 1995, 69. Yang AS: Seasonality, division of labor, and dynamics of
37:255-263. colony-level nutrient storage in the ant Pheidole morrisi.
Insectes Sociaux 2006, 53:456-462.
53. Sempo G, Detrain C: Between-species differences of
behavioural repertoire of castes in the ant genus Pheidole: a 70. Passera L, Roncin E, Kaufmann B, Keller L: Increased soldier
methodological artefact? Insectes sociaux 2004, 51:48-54. production in ant colonies exposed to intraspecific
competition. Nature 1996, 379:630-631.
54. Crozier RH: Heterozygosity and sex determination in Haplo-
This study demonstrates that Pheidole colonies have the capacity to
Diploidy. Am Nat 1971, 105:399-412.
plastically respond to competition by increasing soldier production.
55. Passera L, Suzzoni J: Le role de la reine de Pheidole pallidula
71. Johnston AB, Wilson EO: Correlates of variation in the major/
(Nyl.)(Hymenoptera, Formicidae) dans la sexualisation du
minor ratio of the ant, Pheidole dentata (Hymenoptera:
couvain apre` s traitement par l’hormone juve´ nile. Insectes
Formicidae). Ann Entomol Soc Am 1985, 78.
sociaux 1979, 26:343-353.
72. Murdock TC, Tschinkel WR: The life history and seasonal cycle
56. Schwander T, Humbert J-Y, Brent CS, Cahan SH, Chapuis L,
of the ant, Pheidole morrisi Forel, as revealed by wax casting.
Renai E, Keller L: Maternal effect on female caste determination
Insectes Sociaux 2015, 62:265-280.
in a social insect. Curr Biol 2008, 18:265-269.
An elegant field study describing various features of Pheidole morrisi
colonies in nature and correlating them with their life history.
57. Fave M-J, Johnson R, Cover S, Handschuh S, Metscher B,
Muller G, Gopalan S, Abouheif E: Past climate change on Sky
73. Judd TM: The effects of water, season, and colony
Islands drives novelty in a core developmental gene network
composition on foraging preferences of Pheidole ceres
and its phenotype. BMC Evol Biol 2015, 15:183.
[Hymenoptera: Formicidae]. J Insect Behav 2005, 18:781-803.
58. Goetsch W: Die Entstehung der ‘Soldaten’ im Ameisenstaat.
74. Tschinkel WR: Colony growth and the ontogeny of worker
Naturwissenschaften 1937, 25:803-808.
polymorphism in the fire ant, Solenopsis invicta. Behav Ecol
Sociobiol 1988, 22:103-115.
59. Wheeler DE, Nijhout HF: Soldier determination in ants: new role
for juvenile hormone. Science 1981, 213:361-363.
75. Suzuki Y, Nijhout HF: Evolution of a polyphenism by genetic
This study lays the foundation for JH as an activator of the soldier
accommodation. Science 2006, 311:650-652.
developmental program.
76. Abouheif E, Fave´ M-J, Ibarrara´ n-Viniegra AS, Lesoway MP,
60. Metzl CW, D.E., Abouheif E. Wilhelm Goetsch (1887–1960) –
Rafiqi AM, Rajakumar R: Eco-evo-devo: the time has come.
Pioneering Studies on the Development and Evolution of the
Ecological Genomics. Springer; 2014:: 107-125.
Soldier Caste in Social Insects. Manuscript in preparation.
77. West-Eberhard MJ: Developmental Plasticity and Evolution.
61. Wheeler WM: A neglected factor in evolution. Science 1902:766-
Oxford University Press; 2003.
774.
78. Moczek AP, Sultan S, Foster S, Ledo´ n-Rettig C, Dworkin I,
62. Emery C: Quels sont les facteurs du polymorphisme du sexe
Nijhout HF, Abouheif E, Pfennig DW: The role of developmental
feminin chez les fourmis. Rev Gen Sci 1921, 32:737-741.
plasticity in evolutionary innovation. Proc R Soc Lond B: Biol Sci
2011, 278:2705-2713.
63. Gregg RE: The origin of castes in ants with special reference to
Pheidole morrisi Forel. Ecology 1942, 23:295-308.
79. Molet M, Wheeler DE, Peeters C: Evolution of novel mosaic
Classic study demonstrating that Pheidole colonies tightly regulate the
castes in ants: modularity, phenotypic plasticity, and colonial
number of soldiers and minor workers to maintain a specific ratio.
buffering. Am Nat 2012, 180:328-341.
64. Kitano H: Computational systems biology. Nature 2002,
80. Wills BD, Moreau CS, Wray BD, Hoffmann BD, Suarez AV: Body
420:206-210.
size variation and caste ratios in geographically distinct
populations of the invasive big-headed ant, Pheidole
65. Buchman TG: The community of the self. Nature 2002,
megacephala (Hymenoptera: Formicidae). Biol J Linnean Soc 420:246-251.
2014, 113:423-438.
66. Krebs CJ: Ecology: The Experimental Analysis of Distribution and
81. Ward PS, Brady SG, Fisher BL, Schultz TR: The evolution of
Abundance. San Francisco, CA: Pearson Benjamin Cummings;
myrmicine ants: phylogeny and biogeography of a
2009.
hyperdiverse ant clade (Hymenoptera: Formicidae). Syst
Entomol 2015, 40:61-81.
www.sciencedirect.com Current Opinion in Insect Science 2017, 19:43–51 本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。
学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，
提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。 图书馆致力于便利、促进学习与科研，提供最强文献下载服务。
图书馆导航：
图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具