Investigating Biases Associated With Dietary Starch Incorporation and Retention With an Oral Biofilm Model

Dental calculus has proven to contain a wealth of information on the dietary habits of past populations. These insights have, to a large extent, been obtained by the extraction and identification of starch granules contained within the mineralised dental plaque from a wide range of regions and time periods. The scope of previous studies have been limited to microfossil extraction and identification to reconstruct dietary preferences from the archaeological record, and few studies have attempted to address the biases of starch retention in dental calculus. Those that have considered this problem have been limited to in vivo studies on modern humans and non-human primates. Here, we present a multispecies oral biofilm model, which allows experimental research on starch incorporation and retention to be conducted on in vitro dental calculus in a controlled laboratory setting. The biofilms were exposed to treatment solutions with known quantities of dietary starches (wheat and potato) during the 25 days growth period. After this, the starch granules were extracted from the mature biofilm (by dissolution in EDTA), and counted. We show that the granule counts extracted from the model dental calculus represented a low proportion (ranging from 0.06% to 0.16%) of the total number of granules exposed to the biofilms throughout the experiment. Additionally, we found that the ratios of granule sizes from the extracted starch granules differed from the original treatment solutions, with large granules (>20 mu m) consistently being under-represented. We also found a positive correlation between the absolute granule counts and dry-weight of the biofilm (r = 0.659, 90%CI[0.463, 0.794]), meaning the absolute quantity of starch granules will increase as the size of the calculus deposit increases. A similar, but weaker correlation was found between the concentration (count per mg) of granules and dry-weight (r = 0.3, 90%CI[0.0618, 0.506]). Our results complement and reinforce previous in vivo studies suggesting that dental calculus presents a very small, and partly biased picture of the original dietary intake of starches, with an over-representation of plants producing granules smaller than 20 mu m in size. The experimental model presented here is well-suited to address the need for further validation of methods and biases associated with dietary research on dental calculus.