Reduction of Tooth Replacement Disproportionately Affects the Evolution of Enamel Matrix Proteins

Abstract

In most vertebrates, teeth are continuously shed and replaced throughout life, while mammals and several lineages of reptiles have reduced replacement to only one or two generations. In contrast to the vast majority of their living relatives, members of the lizard families Chamaeleonidae and Agamidae have dispensed with lifelong tooth replacement, instead developing acrodont dentition that fuses to the jawbone to be used for the lifetime of the animal. Though, the loss of tooth replacement has not come without a cost. In order to mitigate the consequences that come with tooth replacement loss, mammals and acrodont lizards have evolved adaptations that strengthen enamel structure and minimize wear and tear experienced during the life of the animal. While these physical adaptations are well documented, the effect that loss of tooth replacement has had on the molecular components of teeth has not received significant attention. Here, we analyze the coding and amino acid sequences of six tooth proteins (AMBN, AMEL, AMTN, ACP4, ENAM, and MMP20) from acrodont lizards, pleurodont lizards that replace teeth, and mammals. We show that the reduction of tooth generations has disproportionately affected the evolutionary trajectory of proteins associated with enamel structure, with a particularly magnified effect on the evolution of AMEL.

Keywords: Acrodont; Agamidae; Chameleon; Enamel matrix proteins; Pleurodont; Tooth replacement.

Fig. 1

Amino acid sequence identity for (a) amelogenin (AMEL); (b) ameloblastin (AMBN); (c) enamelin (ENAM); (d) amelotin (AMTN); (e) acid phosphatase 4 (ACPT4); (f) matrix metalloproteinase-20 (MMP20). Percentage identity calculations with MAFFT alignment of full‐length amino acid sequences. Genera with a number next to the name represent an average percent identity of that number of species. Acr, Acrodonta; Pl, Pleurodonta

Fig. 2

Branch-specific ω estimates. (a) Estimated ω ratios from three-ratios analysis: labeling Acrodonta (Ac), Pleurodonta (Pl), and Mammalia (Ma). (b) Estimated ω ratios from four-ratios analysis: labeling chameleons (Ch), agamids (Ag), pleurodonts (Pl), and mammals (Ma)

Fig. 3

Cluster-based functional distance analysis. The diagrams illustrate the degree of the divergence of each cluster based on site-specific shifts in evolutionary rates after divergence. (a) Schematic representation of functional distance between mammals, pleurodonts, and acrodonts from a theoretical inferred ancestor (white circle). (b) Functional distance between chameleons, agamids, and pleurodonts from a theoretical inferred ancestor (white circle). Branch lengths are proportional to depicted b-values for all analyses

Fig. 4

Convergent evolution analysis using the PCOC toolkit. (a) Graphical representation of the proportion of amino acid sequence for each protein identified as being convergent with pp cutoffs of ≥ 0.9 and ≥ 0.95 (according to the Profile Change with One Change [PCOC] model) depicted as percentage of the average amino acid sequence length of all sequences used in the analysis. (b–d) Sites with PCOC pp values ≥ 0.95 for EMPs. Posterior probabilities for different models are shown in different colors at the bottom of each column. Agamidae (Ag); Chamaeleonidae (Ch), pleurodonts (Pl), and mammals (Ma)

Fig. 5

Substitution rate analysis. Estimated substitution rates from four-rates analysis. *ACP4 four-rates model was not significantly different than three-rates model. Error bars indicate standard error