Tricky Diet Studies

If you’ve ever thrown up, you can probably attest to the fact that some of the food that you eat is quite easy to identify even after digestion has taken place (e.g., spaghetti); however, some of it takes on a much more ambiguous form and is definitely not readily identifiable. When studying human diets this problem can be remedied by asking individuals what foods they consumed throughout the course of the day, or if you are concerned about accuracy, you could also observe the person in question. If they are eating a piece of cake that readily loses its form in the digestive process, you can see that happening and record it. This process becomes far more difficult, however, when you are attempting to study the diet of organisms that you cannot observe as easily as a human and that consume food resources that are not as easily distinguishable as cake! However, if you happen to study whales, this line of work can actually be quite lucrative.

DCIM999GOPRO
Many fish grazing at one time, as in this photo from Midway Atoll, can make it quite difficult to determine who’s eating what!

Researchers studying fishes have historically relied on visual observation and identification of gut contents to determine their diets. When the fishes consume hard-bodied prey that is easily recognizable, such as crabs or smaller fish, this is not such a difficult task. When the fishes consume smaller prey that are more easily digested the process of identification becomes a bit more difficult, and when the fishes consume algae, which are already quite difficult to identify even before they have been consumed, this process becomes extremely difficult!

turf
Turf algae, as seen in this picture with some macroalgae, can be quite difficult to identify!

One way that researchers have dealt with this problem in the past is to use stable isotopes to determine the diet of an individual over an extended period of time. The prey items consumed by fish contain unique isotopic signatures that are based on their environment and composition which get assimilated into the consumer. These isotopic signatures, determined by ratios of stable isotopes of carbon (13C/12C) and nitrogen (15N/14N), are then used to assess the trophic level of the consumer. Higher nitrogen values suggest a higher trophic position, and carbon values can often be indicative of the type of primary producer consumed. Comparing the isotopic signatures of different species can shed some light on the structure of the food web in complex systems. For example, Dromard et al. (2015) recently found that in the Caribbean Acanthurus coeruleus had a higher 15N value than the other herbivores examined, which they suggested was likely a result of it consuming more macroalgae than the other fishes as well as it being the only fish to consume invertebrates (see graphs below).

Screen Shot 2015-12-01 at 12.30.53 PM
(Figure and caption from Dromard et al. 2015.) Mean (±95% CI) δ13C and δ15N signatures of fish muscles and the potential food sources (closed symbols indicate consumers and open symbols show the potential food sources).

The stable isotope results also suggest that scarids, or parrotfish, may be consuming more macroalgae than is actually seen in visual gut content studies, and the authors suggest that this difference could be due to the physical and chemical breakdown of algae by parrotfishes during consumption and digestion (Dromard et al. 2015). Unfortunately, even in these thorough studies using mixing models to combine the stable isotope diet information with visual inspection it can still be difficult to get very high taxonomic resolution, especially with turf algae which is often left as a broadly identified group (see figure below).

Screen Shot 2015-12-01 at 12.46.37 PM
(Figure and caption from Dromard et al. 2015.) Mean contribution of food sources to the diet of fish, estimated with mixing models. Fleshy macroalgae: Dictyota cf pulchella and Acanthophora spicifera, coral: Orbicella annularis.

Recently, a genetic method known as metabarcoding has also gained popularity as a way of identifying the unknown bits of partially digested materials that are often encountered in diet studies. To break it down, metabarcoding uses a known region of DNA that varies enough between organisms to act as a unique species identifier. This region is then selected for using a universal primer that is able to identify the region, amplified, and sequenced, and this can be done for many unknown species in one sample in order to identify the many components. The sequences from the unknown samples can be compared to a database with sequences from known samples, so matches then unveil the identity of the unknown DNA. To simplify it, imagine your healthy friend hands you a well-blended smoothie, but you don’t know what ingredients were used to make it. Using metabarcoding you could in theory determine that she put kale, avocado, lemon, apples, ginger, and chia seeds in the concoction, but she forgot to add bananas and coconut milk!

(Photo from simplegreensmoothies.com)
(Photo from simplegreensmoothies.com)

In the marine realm a major breakthrough was made in 2013 when Leray et al. published a study that used metabarcoding to examine the gut contents of reef fishes. Leray et al. (2013) were able to identify a universal primer for the mitochondrial Cytochrome c Oxidase subunit I gene, which is known as COI. One problem that remains particularly tricky though is that when you are selecting for this DNA in the prey items in gut contents, you will also be targeting this region in the predator’s DNA. There is likely far more abundant predator DNA than any of the prey DNA, so it may overshadow the DNA that you actually want. There are, however, ways to overcome this problem, such as the use of blocking primers that don’t allow for the selection of predator DNA. The use of metabarcoding to study the diets of herbivorous fishes is particularly exciting, because the algal gut contents can be identified using chloroplast DNA. Given that fish don’t have chloroplasts, this eliminates the problem of amplifying predator DNA. So, while the application of metabarcoding to diet studies is still new, the future appears bright for those of us who study the tricky diets of herbivorous fishes!

 

References:

Dromard et al. 2015. J. of Sea Research. 95: 124-131.

Leray et al. 2013. Frontiers in Zoology. 10: 34.