Human dentition may adapt functionally to diet and feeding behavior. Bones continue to remodel throughout an individual's life, but tooth enamel does not regenerate once its formation is complete. For these reasons, dental morphology has been studied as an important indicator of both functional adaptations and phylogenetic evolution [1]–[5].

Dental malocclusion is a misalignment of the human dentition, either due to overcrowding within each dental arch or an abnormal relationship between both dental arches. Despite the millions of people around the world undergoing orthodontic treatment, the etiology of dental malocclusion is still not clear [6]. Although little is known about its origin in human populations, dental malocclusion has been often referred to as a “disease of civilization” [7], [8]. Studies examining human [7]–[13] and nonhuman primates [14]–[17] have indicated that the increase in the occurrence of dental malocclusion can be attributed to the availability of a more processed diet and a reduced need for powerful masticatory action. Although there is also some weak evidence suggesting genetic influences on the occurrence of malocclusion in Homo sapiens [18]–[20], these studies have relied on the extrapolation of data from the remains of ancient populations for which we cannot determine the cause of death.

The Arara-Iriri indigenous group lives by the Xingu River in the Amazon region and was first contacted in 1987. A previous anthropological study [21] reports that the people who constitute this village are descendants of a single couple who were expelled about 90 years ago from a larger village. The larger village, named Laranjal, was first contacted in 1983 and is located 200 kilometers away [Fig. 1]. The presence of a single Y chromosome, one mtDNA haplotype, and only one to four alleles at all autosomal loci studied confirmed an extreme case of lineal fission involving related individuals from an ancestral village [22].

Increased tooth wear is considered a phenotypic signal of an adaptive shift to exploit tougher and more abrasive food resources. Tooth wear has been used to describe hominid evolution [1], [2], [23], [24] because it provides direct evidence of what an individual ate in the past [23], [24]. The eating habits of the indigenous tribes inhabiting the Xingu region are predominantly traditional and usually based on cassava, nuts, fish, meat of wild animals, sweet potatoes, yams and wild fruits [25]. All children are breastfed until the birth of the next child, which usually occurs after 1.5–2 years.

Although small founder populations seem to be frequent events in the formation of new tribes among the Amazon Amerindians, the fission process and its consequences on the differentiation of ethnic groups have seldom been recorded. In addition, the influence of founder effects on dentition has never been reported. Given the accessibility of historical [21] and genetic records [22] for these two human populations, the examination of the Arara villages provides a unique opportunity to improve our knowledge about the etiology of dental malocclusion.

While the Arara-Laranjal village was expanded by nonconsanguineous or rare incestuous relations, the initial expansion of the Arara-Iriri group occurred through the mating of closely related people, including parents and their offspring and sibling with their siblings, and later by marriages between relatives somewhat more distant, such as uncle-niece, aunt-nephew and first cousins [22]. The median degree of inbreeding as measured by the inbreeding coefficient (F), for the Arara-Iriri village was 0.25 (IQR = 0.25) and differed significantly (p<0.0001) from the Arara-Laranjal population, whose median is 0 (IQR = 0). This finding confirms the frequent incestuous unions among the Arara-Iriri indigenous group.

Regarding tooth wear evaluation we observed a very similar pattern for both villages (Figure 2), confirming the similarity of the eating habits among the Xingu indigenous groups [25].

In the original village (Laranjal), only one third of the population showed dental malocclusion (33.8%). By contrast, in the resultant village, the rate of malocclusion was nearly doubled (63%; prevalence ratio (PR) = 1.86, p = 0.0005). Normal occlusion was the most common morphological pattern in the original Arara village (Table 1 and Figure 3A), but for the Arara-Iriri indigenous people, the common pattern was a Class III malocclusion in which the lower dentition is more prominent than the upper dentition (32.6%, PR = 6.1, p<0.0001) (Table 1 and Fig. 3B). Class II malocclusion, diagnosed when the upper dentition is more prominent than the lower dentition, was also two times higher in the resultant village. Anterior open bite (PR = 2.64, p = 0.003), anterior crossbite (PR = 2.83, p<0.001), overjet (PR = 3.39, p = 0.03) and posterior crossbite (PR = 4.71, p = 0.02) were also more common in the Arara-Iriri village. Although no subject presented dental crowding or overbite in the resultant village, overbite was evident in 5 individuals (3.9%), and dental crowding was observed in 20 individuals (20.8%, p = 0.005) in the original village. These data confirm a marked divergence between the villages for dental malocclusion prevalence.

Scientifically, the role of environment and genetics on dental malocclusion has been discussed under three models of study designs: the analysis of skull remains of ancient populations [7]–[13], experiments with animal models [14]–[17] and research on human twins [26]–[30]. Most of these investigations have advocated that the increase in the occurrence of dental malocclusion must be attributed to environment changes during human evolution [7]–[17], [27]–[28].

Some studies examining twins have suggested dental malocclusion as a primary consequence of environmental changes [27], [28]. Recently, several criticisms have been attributed to this research model [26]. A major issue of concern in many previous studies of twins has been the accuracy of zygosity determination by comparisons of physical appearance. More recently, the use of highly polymorphic regions of DNA has proved to be more accurate and reliable [30].

We examined the genetic and dietary influences through tooth wear on occlusal variation between two split indigenous villages. Our findings revealed that two indigenous populations belonging to the same ethnicity showed marked differences regarding dental malocclusion prevalence (Table 1). From deciduos to permanent dentition, normal occlusion was the most common morphological pattern in the original Arara village (Table 1 and Figure 3A), but for the Arara-Iriri indigenous people, the common pattern was the presence of dental malocclusion. These findings should not be attributed to dietary consistence, as reports [25] have pointed out similar feeding habits for both villages; these findings were corroborated by similar patterns of tooth wear found in the present investigation (Figure 2). Findings strongly suggest that the marked difference in dental occlusion is related to the genetic distance and molecular variance between the two Arara villages, as previously reported [22].

The Amerindian population exhibits genetic diversity as high as that of any other human population; the large variation between tribes compensates for the low variation within tribes, a feature attributed to the genetic drift acting on small, isolated populations [22]. While a genetic investigation confirmed that the Arara-Iriri village was founded by a single couple who came from the ancestral village [22], the molecular variance between these two contemporary tribes was greater than that observed between them and other Amazonian tribes [22], [31]. The fission process, which was influenced by a marked genetic cohesion and a dramatic founder effect, has produced remarkable genetic differences between the villages. Therefore, despite their common origin, the populations that constitute these villages have different genotypes.

In light of similar tooth wear, an outcome with a pronounced influence of the environment, the striking difference in the dental malocclusion between the Arara indicates that genetics substantially impacts the morphology of the facial bones and dentition. Moreover, our results suggest that a founder effect and genetic inbreeding have exacerbated alterations in dental occlusion in the Arara-Iriri village population. These findings lend support to the notion that variation in continuous quantitative traits is usually polygenic [32], [33] and that we should expect a highly polygenic basis for complex traits such as human craniofacial and dentition morphology and development [33]. Moreover, our findings downplay the widespread influence of tooth wear on dental malocclusion in current human populations [7]–[17].

This study was approved by the National Ethical Committee for Health Sciences of the Brazil registered under number 25000.066559/2010-11. Written Informed consent was obtained prior to data collection.

The total population was composed of 239 individuals living in the Arara-Laranjal village and 80 individuals living in the Arara-Iriri village. The number of subjects not present due to hunting or fishing expeditions during the study period was 11 (4.6%) from the Laranjal tribe and 6 (7.5%) from the Iriri tribe. All individuals present were examined, but only those subjects aged between 2 and 22 years old have been included in this analysis. Therefore, the sample was composed of 130 individuals from the Laranjal tribe and 46 from the Iriri village (Table 2). Tooth wear was examined only for individuals with permanent dentition (n = 81).