American Society of Naturalists

A membership society whose goal is to advance and to diffuse knowledge of organic evolution and other broad biological principles so as to enhance the conceptual unification of the biological sciences.

Topology of Taste: Carotenoid Metabolism Networks and Adaptive Diet Transitions

Posted on by Naama Weksler, edited by Regina Fairbanks and Swapna Subramanian

"Stronger Historical Contingency Facilitates Ecological Specializations: An Example with Avian Carotenoid Networks"

Erin S. Morrison, Caitlin M. Hill, and Alexander V. Badyaev: Read the article

Are specializations evolutionary dead ends? Morrison et al. reveal that in bird carotenoid evolution, continuity and stability are two sides of the same coin

House finches at rest on a saguaro cactus. <br/> Photo credit: Alexander V. Badyaev
House finches at rest on a saguaro cactus.
Photo credit: Alexander V. Badyaev

When I was a child, my family once opened the door to a door-to-door salesman who asked, with a flourish, why flamingos were pink. His sales method hit a roadblock when my brother and I already had an answer for him, that flamingos get their iconic hue from their food: bright pink shrimp. Shrimp are rich in carotenoids, organic pigments which are responsible for much of the yellow, orange, and red colors observed in the natural world, including that of carrots, for which they are named. Flamingos are one of over 200 diverse bird species which absorb carotenoids from their food and use them for plumage color, antioxidant function, and immune support. Like most animals, birds cannot create carotenoids on their own. Instead, they rely on consuming them in their diet before transforming them through chemical reactions into useful forms. However, as any picky eater can attest, reliance on a particular type of food can be a dangerous thing. Such a requirement becomes an issue if the food source in question dwindles or becomes unavailable. Species must be able to rapidly adapt to a different source or face extinction.

Understanding how organisms transition between different adaptive states, such as relying on different food sources, is a major goal of evolutionary biology, but one that is difficult to address due to the transient nature of evolutionary processes. Evolution does not preserve the context in which it happens, only the outcome. We know that flamingoes today eat shrimp, but the bird from which flamingoes evolved no longer exists. How, then, can we make assertions about such ancestral birds, their diets, and their evolution? Evolutionary biologists often use a method called “ancestral reconstruction”, through which researchers use statistics to calculate the most likely state for an ancestral organism that no longer exists. Another approach is to study systems in which species do not modify the context in which they evolve and compare potential and realized evolutionary trajectories. In their new paper, Morrison et al. use both of these approaches to study evolutionary transitions in carotenoid metabolism. In particular, they explore the question of how birds evolve the ability to get carotenoids from different food sources.

To address this question, the authors leverage a large dataset detailing the carotenoid compounds found in 260 species of birds. They map these compounds onto the “universal carotenoid metabolic network” – so called because it describes all known reactions involved in carotenoid metabolism across the tree of life – to visualize networks for each of their study species. Such networks describe how materials get transformed. For example, a network could be used to describe a process of baking: grain, sugarcane, and milk are inputs which might get processed into flour, sugar, and butter, then combined in different ratios to create cookies, cake, or caramel. Bird carotenoid networks are the same – birds ingest carotenoids from their food, then process them into intermediates, and then again into their final forms. As different birds eat and use different carotenoids, these networks can be quite different between species.

The authors analyze this dataset using a variety of network theory metrics. They find diversity in the structure of these networks across the bird tree of life: some birds, especially those which eat only fruit, are very specialized to process only certain types of carotenoids, while other more omnivorous species can process a variety of inputs. The defining factor, it seems, is the presence of “degenerate” carotenoids – intermediates which can be synthesized from multiple inputs and used to create multiple outputs. Networks with more degenerate carotenoids are more resilient than those without them thus enhancing dietary adaptation. Remarkably, this is the same network feature that facilitates transitions between diets.

To establish the link between local adaptation and evolutionary change, the authors use an ancestral reconstruction algorithm to reconstruct carotenoid networks for ancestral species in the bird tree of life, which they then use to probe the role of degenerate carotenoids in evolution. They find that, unlike the starting or ending points of the networks, degenerate intermediate carotenoids tend to be very stable through evolution – if they’re present in an ancestor, they’re very likely to be present in its offspring. Additionally, the authors find that species with specialized food sources are more adaptable when they accumulate more degenerate intermediates in their networks. Imagine only knowing how to make sugar from beets, compared to knowing how to make sugar from beets, sugarcane, or honey. While they both might allow for a sweet cake, the latter network gives much more flexibility if the circumstances change and beets disappear. Thus, the presence of degenerate carotenoids allows birds with specialized food sources to quickly evolve to rely on other sources.

Understanding how species’ evolutionary history affects their adaptation allows scientists to gain greater insight into evolutionary processes. Studying the ways in which species history affects variables such as how fast they evolve and what they evolve into helps us understand the balance between randomness and inevitability in evolution. Morrison et al. make use of a unique example – bird carotenoid processing – to study this question. Their results provide insight into the ways in which the same network structures that affect local adaptability also facilitate evolutionary change while also lending support to the conclusion that species history has a strong role in shaping ecological adaptations. Whether or not the salesman knew it, his flamingo-forward marketing tactic was referencing a fascinating process that can teach us more about the mechanisms by which evolution occurs.


Naama Weksler is a Ph.D. candidate in Biological and Biomedical Sciences at Harvard University. She studies the molecular underpinnings of convergent evolution, aiming to better understand whether the co-option of developmental pathways to create novel morphologies occurs in the same way in independent gain events. Naama also volunteers as a mentor at Cambridge Science Club for Girls and through her PhD program's peer mentor initiative. When not in the lab, she likes to read (sci-fi, autofiction, and science writing), cook, and hang out with her partner and cat.