My approach aims to be innovatively
integrative, spanning ecology to neurophysiology and focusing on
behaviour. The two study systems for which I am best known are phase
change in locusts and the development of a new conceptual and experimental
framework for the study of feeding and nutrition. This Geometric
Framework (which arose from the study of insects) is currently providing
novel insights into the modern human nutritional condition, as well
as being applied to diet optimisation in domestic animals and aquaculture
systems. The following flow diagrams show the relationships and
scope of the work in my lab. The numbered references can be found
in the Research Output page.
1. An integrative analysis
of swarming in locusts
Central to the locust’s status as a
pest is its ability to change phase in response to an increase in
population density, and as a result to form massive swarms. My group
has made fundamental breakthroughs in understanding phase change
- both at the level of its controlling mechanisms and significance
for population dynamics (89). Key discoveries are italicised
below.
Behaviour is the most labile of the
suite of characters that comprises phase change and provides positive
feedbacks that drive the process at a population level. In the ‘solitarious’
phase, locusts avoid each other, but actively aggregate when ‘gregarious’.
The first stage was to produce a simple,
robust and sensitive assay for measuring the behavioural phase-state
of individual locusts, using logistic regression as a means
of encapsulating the multiple behavioural elements involved (89).
The assay was used to show that individuals change behavioural
phase rapidly (in a matter of hours), and also that behavioural
state is transmitted across generations epigenetically, with
females gregarising their developing offspring to an extent which
reflects both maternal and paternal experience of crowding (89).
This maternal influence is by a chemical agent emanating from
the female reproductive accessory glands and introduced into
the foam that surrounds the eggs when they are laid (70). The
female has a memory of when she was last crowded, which is
reflected in the extent to which she adds the gregarising compound
to her eggs at oviposition (89).
In a decisive series of experiments
the gregarising cues from other locusts were identified.
The interactive effects of olfactory, visual, contact chemical
and mechanical stimuli were teased apart (75). The most powerfully
gregarising stimulus was shown to be physical contact among
locusts. Remarkably, mechanosensory inputs from the hind femur
are key (102), a finding that opened the possibility to study
phase change as a model system for neuronal plasticity with
colleagues at the University of Cambridge (117, 126).
We investigated in laboratory and
field experiments and individual-based computer models (80) the
relationship between individual behaviour, population responses,
and the spatial distribution and chemical quality of resources within
the local environment. Whether a local population of solitarious
locust will gregarise, and hence potentially seed larger scale outbreaks,
was shown to depend critically on the fine-scale distribution
and quality of resources. A consequence of forming aggregations
is that local populations of locusts become susceptible to predation
and disease. Newly gregarised locusts feed selectively on poisonous
plants (128), thus conferring anti-predator protection during
the vulnerable, early stages of group formation and also show
elevated immune responses (105), which establishes the desert
locust as a casebook example of 'density-dependent prophylaxis'.
Fundamental insights into population
processes such as these have not been available from previous higher-level
analyses, and have shown why information about the structure and
quality of food resources at fine spatial scales must become a central
component in monitoring and control strategies. They also provide
some of the most compelling examples of the power of individual-based
approaches in ecology.
2. Integrative nutrition
Animals must balance the location, selection, ingestion and use of
numerous nutrients against multiple and changing metabolic requirements.
Several major disciplines (nutrition, animal production, experimental
psychology and behavioural ecology among them) had yet to deal adequately
with the multidimensional nature of feeding and nutrition. To this
end, I and my close colleague David Raubenheimer and our students
and post-docs developed a state-space framework [the ‘Geometric Framework’
(GF)] from extensive studies of insect herbivores (41,68).
The GF builds on an untested idea of McFarland and Sibly (1975). It
unifies within a single model the animal and its multidimensional
nutritional environment and has considerable power as a conceptual
and experimental technology for studying feeding behaviour and post-ingestive
physiology, and placing and interpreting such mechanisms in their
evolutionary, ecological and developmental contexts. The GF has
been used to explore a range of issues in ecology, evolution, agriculture,
human and animal health and welfare. Examples of its ecological
application include nutrient-allelochemical interactions in
herbivore foraging (100, 107), the relationship between nutrient
balancing rules and diet breadth (e.g. 106), the integration
of ecological stoichiometry, optimal foraging theory and resource
exploitation theory (e.g. 123), and the demonstration that predators
balance their nutrient intake, contrary to the central assumption
of current foraging theory that they do not (132). The GF has arisen
from extensive experimental studies by us on insect feeding behaviour
and its controlling mechanisms; studies which have led to the locust
becoming a model system for the study of feeding (86). Such work
has discovered nutritional regulatory mechanisms, such as the taste-feedback
mechanism, whereby nutritional state, as indicated by levels of
blood nutrients, alters the responsiveness of taste receptors and
thus leads to insects making sophisticated nutritional decisions without
requiring complex central neural computation (34), and also learned
nutrient-specific appetites (22). While the GF was derived from
studies of insects, it has provided new insights in vertebrates,
including mammals, birds and fish (e.g. 68). Most recently, we
have applied the GF for the first time to humans (116, 130),
and shown that regulation of protein intake may explain more of the
modern human nutritional condition than has previously been appreciated.
The combination of the fact that protein comprises a minor part of
the total energy budget of humans, yet its intake appears to be strongly
regulated, leads to protein having considerable leverage over food
intake. Protein has the power both to drive the development of
obesity and to assuage it. The public health implications are
considerable.
|
