High-temperature isothermal heating of biominerals has commonly been used to artificially accelerate protein degradation in order to extrapolate kinetic parameters to the lower temperatures experienced in vivo and in the burial environment. It is not easy to test the accuracy of these simulations due to the difficulty of finding samples of known age held at a known temperature. We compare protein degradation in the intra-crystalline organic matrix of heated (80 °C, 110 °C, and 140 °C) massive Porites sp. coral to that directly measured in the skeleton of colonies growing at ∼26 °C and deposited over the last five centuries. This provides the opportunity to critically evaluate the underlying assumption that high-temperature experiments accurately mimic degradation processes and kinetics occurring in a 'naturally aged' biomineral. In all samples the intra-crystalline protein fraction was isolated and the L-and D-concentration of multiple amino acids measured using reverse-phase high-performance liquid chromatography (RP-HPLC). There was no evidence of a failure of the closed system in the high-temperature experiments (assessed by X-ray diffraction, thermogravimetric analyses and determination of leached amino acid concentration). We compared conventional methods for estimation of kinetic parameters with a new 'model-free' approach that makes no assumptions regarding the underlying kinetics of the system and uses numerical optimisation to estimate relative rate differences. The 'model-free' approach generally produced more reliable estimates of the observed rates of racemization in 'naturally aged' coral, although rates of hydrolysis (as estimated from the release of free amino acids) were usually over-estimated. In the amino acids for which we were able to examine both racemization and hydrolysis (aspartic acid/asparagine, glutamic acid/glutamine and alanine), it was clear that hydrolysis was less temperature sensitive than racemization, which may account for the differences in degradation patterns observed between the 'naturally aged' coral and high-temperature data. It is clearly important to estimate the individual temperature dependence of each of the parallel reactions.
Highlights► We compare protein degradation in heated modern, and naturally aged Porites coral. ► Heating to 140 °C does not compromise the integrity of the closed-system. ► We use a new approach that fits relative rates to estimate temperature dependence. ► Protein degradation patterns change over the temperatures investigated (26-140 °C). ► We observe different temperature sensitivities for hydrolysis and racemization.