Axis Inhibition Protein 2 (AXIN2) Polymorphisms and Tooth Agenesis

Abstract

Tooth agenesis is a common congenital disorder that affects almost 20 percent of the world’s population. A number of different genes have been shown to be associated with cases of tooth agenesis including AXIN2, IRF6, FGFR1, MSX1, PAX9, and TGFA. Of particular interest is AXIN2, which was linked to two families segregating oligodontia and colorectal cancer. We studied two collections of families affected with tooth agenesis and tested them for association with AXIN2. Significant association between tooth agenesis and AXIN2 was found (p = 0.02) in cases with at least one missing incisor. Our work further supports a role of AXIN2 in human tooth agenesis and for the first time suggests AXIN2 is involved in sporadic forms of common incisor agenesis. Future studies should identify which specific tooth agenesis subphenotypes are consequence of AXIN2 genetic variations. A subset of these cases could have an increased susceptibility for colon cancer or other types of tumors and this knowledge would have significant clinical implications.

Introduction

Dental abnormalities can provide some of the earliest warnings signs of some systemic disorders, such as bulimia (Bartlett, 2005) and consequences of antineoplastic treatments (Lopes et al., 2006). The dentist could be the first professional to notice these symptoms, which in turn can lead to early detection, faster treatment, and a higher survival rate. There is compelling evidence that we can soon add colorectal carcinomas to this list of diseases.

Tooth agenesis affects approximately twenty percent of the world’s population. If third molars are excluded, still over five percent of the population is afflicted (Vastardis 2000). People with missing teeth have many problems with esthetics, phonetics, and mastication.

Studies in a Finnish family with severe familial oligodontia who did not have mutations in MSX1 (muscle segment homeobox 1) or PAX9 (paired box 9), the two genes that when mutated can lead to oligodontia, disclosed the presence of a mutation in the axis inhibition protein 2 (AXIN2) gene. While studying the family’s medical records, a history of familial adenomatous polyposis was discovered (Lammi et al., 2004). The possibility of the identification of a dental clinical marker for cancer brings additional interest to the studies of the genetic regulation of tooth agenesis.

The association with AXIN2 and tooth agenesis was independently investigated in a relatively small case-control cohort from Poland and an association was found between tooth agenesis and markers in the gene (Mostowska et al 2006). On the other hand, germline AXIN2 mutations were found to be rare in patients with multiple polyposis (Lejeune et al., 2006).

AXIN2 is a negative regulator of the Wnt signaling pathway. Axin2 is expressed during mice odontogenesis in the dental mescenchyme, enamel knot, dental papilla mescenchyme, and in mesenchymal odontoblasts. There is also extensive evidence of the expression of AXIN2 in colorectal tissues leading to carcinomas (Lammi et al. 2004).

The present study expands the investigations of AXIN2 in tooth agenesis, and further supports a role of this gene in human tooth agenesis.

MATERIALS AND METHODS

Our study group consisted of 167 patients with tooth agenesis and their parents. The patients were from two different cohorts, 116 were from Rio de Janeiro, Brazil, which is an admixed population of Europeans (from Portugal) and Africans, with a very small percentage of Native South Americans. This population contains 71 sporadic cases and 45 familial cases. Seventy-four were females and 42 were males. The second cohort consisted of 51 trios from Turkey. Twenty six were females and 25 were males. All Turkish cases were of sporadic origin. Tooth agenesis was the sole disorder affecting these patients and second premolars were the teeth most commonly absent, followed by lateral incisors. None of the families reported history for clefts. Complete details about these two populations are presented elsewhere (Vieira et al., 2004; Vieira et al., 2008). The study was approved by the University of Pittsburgh Institutional Review Board (IRB), as well as the appropriate ethical committee at the Federal University of Rio de Janeiro and Istanbul University, and/or appropriate informed consent was obtained from human subjects. After informed consent was obtained, cheek swabs, saliva, or whole blood were collected from each individual. Clinical analysis, sample collection, and DNA extraction were performed using a consolidated protocol described elsewhere (Jezewski et al., 2003; Vieira et al., 2004; Vieira et al., 2008). Cases were analyzed not only as a single group, but also in three subgroups: cases with positive family history for tooth agenesis, cases with at least one missing incisor, and cases with at least one missing premolar (Table 1).

Table 1.

Demographic data for both cohorts.

Family Characteristic	Brazilian N (%)	Turkish N (%)
Gender distribution
Males	42 (36)	24 (47)
Females	74 (64)	27 (53)
Number of teeth missing
1	44 (38)	7 (14)
2	46 (40)	17 (33)
3 or more	26 (22)	27 (53)
Type of teeth more often missing (total teeth missing)	274	207
Second premolar	104 (37.9)	75 (36.2)
Lateral incisor	82 (29.9)	54 (26)
First premolar	26 (9.7)	8 (3.8)
Second molar	24 (8.7)	11 (5.3)
Central incisor	16 (5.8)	32 (15.4)
First molar	11 (4)	9 (4.3)
Canines	11 (4)	18 (8.6)
Associated dental anomalies	16 (13.8)	-
Number of cases missing incisors	55 (47.4)	29 (56.8)
Number of cases missing premolars	63 (54.3)	44 (86.2)
Number of cases missing molars	14 (12)	8 (15.6)
Number of cases missing canines	7 (6)	10 (19.6)
Positive family history	41 (35.3)	-

Three intragenic AXIN2 markers were studied (rs7591, rs11867417, and rs2240308; Table 2). These markers were chosen based on the information on the gene structure and linkage disequilibrium blocks available at the International HapMap Project website (http://www.hapmap.org/). Genotypes were obtained using an ABI PRISM 7900 Sequence Detection System (Valencia, CA, USA) and TaqMan chemistry. Reagents and SNP genotyping assays were supplied by Applied Biosystems (Valencia, CA, USA). Pairwise calculations of linkage disequilibrium were computed with the Graphical Overview of Linkage Disequilibrium (GOLD) software for both the squared correlation coefficient (r2) and Lewontin’s standardized disequilibrium coefficient (D′) (Abecasis and Cookson, 2000). Alleles at each marker were tested for association with tooth agenesis with the use of the Family Based Association Test (FBAT) software (Horvath et al., 2001; Horvath et al., 2004). The hbat function of FBAT was used to calculate haplotype frequencies and associations between AXIN2 haplotypes and tooth agenesis.

Table 2.

AXIN2 SNPs studied.

Variants	Alleles	Average Heterozygosity ± Standard Error	Location in AXIN2	mRNA Position
rs7591	A	0.466 ± 0.126	3′UTR	3858
	T
rs11867417	C	0.486 ± 0.081	Intron 7	-
	T
rs2240308	A	0.417 ± 0.186	Exon 1 (P50S)	462
	G

RESULTS

Table 3 provides the results for single marker analysis. Also, it presents the results for the subgroups of the cases with positive family history, cases with at least one missing incisor, and cases with at least one missing premolar. Table 4 provides a summary of the most significant haplotype results in the two populations. The pairwise linkage disequilibrium analysis confirms that the marker set we selected is for the most part in weak linkage disequilibrium (Table 5), which provides almost non-overlapping information about the gene.

Table 3.

FBAT Results for combined cohorts of Brazilian and Turkish families.

Variants	Overtransmitted Allele	Brazil		Turkey		Combined
All Trios		n	p	n	p	n	p
rs7591	-	56	0.742	30	0.286	86	0.373
rs11867417	-	56	0.359	34	0.199	90	0.128
rs2240308	-	56	0.154	32	0.334	88	0.086
Trios with Probands Missing at Least One Incisor		n	p	n	p	n	p
rs7591	-	26	0.612	17	0.22	43	0.241
rs11867417	-	26	0.086	21	0.857	47	0.172
rs2240308	A	28	0.037	17	0.587	45	0.055
Trios with Probands Missing at Least One Premolar		n	p	n	p	n	p
rs7591	-	35	0.500	25	1.000	60	0.600
rs11867417	-	37	0.674	28	0.262	65	0.292
rs2240308	-	32	0.886	28	0.374	60	0.471
Trios with Positive Family History		n	p	n	p	n	p
rs7591	-	19	0.317	-	-	-	-
rs11867417	-	19	0.852	-	-	-	-
rs2240308	-	19	0.480	-	-	-	-

Notes: n = number of informative families; p = p-value; bold indicates p-values lower than 0.05.

Table 4.

Summary results of haplotypes analysis for combined cohorts of Brazilian and Turkish families.

Haplotype	Allele	Brazil		Turkey		Combined
All Trios		n	p	n	p	n	p
rs7591 rs11867417	A-T-
rs2240308	A	14	0.077	7	0.959	21	0.156
rs7591 rs11867417	T-C	47	0.162	26	0.231	73	0.069
Trios with Probands Missing at Least One Incisor		n	p	n	p	n	p
rs7591 rs11867417	A-T-
rs2240308	G	15	0.037	14	0.131	29	0.096
	T-C-
	A	17	0.099	13	0.956	30	0.021
	T-T-
	A	3	0.083	2	0.784	5	0.046
rs7591 rs11867417	T-C	20	0.027	15	0.235	35	0.057
	A-C	15	0.654	10	0.091	25	0.519
rs11867417 rs2240308	C-A	23	0.048	16	0.594	39	0.065
Trios with Probands Missing at Least One Premolar		n	p	n	p	n	p
rs7591 rs11867417	T-T-
rs2240308	G	3	0.083	-	-	3	0.083
	T-C-
	G	18	0.041	17	0.33	35	0.465
rs7591 rs11867417	A-C	18	0.088	14	0.467	32	0.427
rs11867417 rs2240308	C-G	25	0.041	20	0.434	45	0.271
Trios with Positive Family History		n	p	n	p	n	p
rs7591 rs11867417	T-C-
rs2240308	G	9	0.052	-	-	-	-

Notes: n = number of informative families; p = p-value; bold indicates p-values lower than 0.05.

Table 5.

Linkage Disequilibrium analysis of the AXIN2 markers studied.

Marker	rs7591	rs11867417	rs2240308
rs7591	-----	0.384	0.106
rs11867417	0.679	-----	0.143
rs2240308	0.377	0.488	-----

^ar² is above the diagonal; D′ is below the diagonal.

The most significant association was between the AXIN2 rs2240308 marker and cases with at least one missing incisor (p=0.037). This trend could also be seen in the haplotype analysis, in which most of the significant signals were for cases with at least one missing incisor.

Since the marker rs2240308 is a coding change in the gene (P50S: proline to serine at position 50), we used the software Polyphen (Sunyaev et al., 2000; Sunyaev et al., 2001; Ramensky et al., 2002) to predict the possible impact of the amino acid substitution on the structure and function of the AXIN2 protein. The change was predicted to be benign. We also used the ESE Finder 3.0 software to predict if the change would alter exonic splicing enhancers, which are thought to serve as binding sites for specific serine/arginine-rich proteins that help determine splicing (Cartegni et al., 2003; Smith et al., 2006). The A to G nucleotide change did not affect any predicted exonic splicing enhancer site.

DISCUSSION

Our work provides further evidence that AXIN2 contributes to tooth agenesis. The marker rs2240308 (P50S) was significant by itself in Brazilian cases with at least one missing incisor (p=0.037), as well as was part of an associated haplotype in the combined Brazilian-Turkish dataset of cases with at least one missing incisor (p=0.021). The P50S change does not appear to be etiologic based on our in silico experiments. It is possible this variant is in linkage disequilibrium with the functional variant that contributes to tooth agenesis.

While concerned about multiple testing, we did not apply the strict Bonferroni correction as it would increase type II errors and a major focus of this study was to identify putative associations between AXIN2 and tooth agenesis for further studies. For example, under the Bonferroni correction, we would have lowered the alpha to 0.0056 (0.05/9), and the possible association between AXIN2 and cases with at least one missing incisor would have been ignored. Therefore we report here all results with p-values below 0.05. However, our data must be carefully interpreted since it is expected that some of the p-values below 0.05 can be due to chance.

The suggestive association with cases with at least one incisor missing raises interesting questions. The inactivating mutations described in AXIN2 that lead to oligodontia and higher susceptibility to colon cancer affected incisor development in 11 out of 12 cases (Lammi et al., 2004). The study from Poland that suggested an association between AXIN2 variation and tooth agenesis (Mostowska et al., 2006) was also enriched by cases missing the upper lateral incisors (36 cases in a total of 55). Interestingly, this report did not show an association between the AXIN2 P50S change and tooth agenesis, and the significant results were found for another two AXIN2 variants in strong linkage disequilibrium with each other, the silent mutation L688L and the intronic variant c.956+16A>G). According to the authors, these variants were not in linkage disequilibrium with P50S. In combination with our findings, these results may suggest that multiple AXIN2 variants could contribute to tooth agenesis in humans. The L688L was predicted to disrupt exonic splicing enhancer sequences, and although it is a synonymous change, this could contribute to the tooth agenesis phenotype. The c.956+16A>G was also suggested as possibly exerting a weak effect on splicing as it creates an additional donor-splicing site within the sequence of exon 2.

Finish, Polish, Turkish, and the studied Brazilian dataset, which is comprised by mainly Portuguese descendents, can be assumed to be mainly European subgroups. The current evidence regarding AXIN2 and tooth agenesis suggests that this gene may play a role in congenitally missing teeth of groups of European ancestry.

Previously, we have suggested that different types of teeth are regulated by independent gene expression (Vieira, 2003). A number of Wnt genes are expressed in the developing teeth and changes in their expression may be one of the factors determining tooth agenesis (Zhang et al., 2005). There is evidence that genes related to the Wnt pathway are differentially expressed in molars versus incisors. Studies in mice targeting Lef1 function, a transcription factor that can be activated by Wnt proteins, showed that transgene expression of human LEF1 (lymphoid enhancer-binding factor 1) promoter can be seen where the incisor and molar teeth develop at E12.5. Transgene molar expression of the LEF1 promoter during molar development was observed only in E12.5 embryos and disappeared in E13.5 embryos. Conversely, LEF1 promoter expression during incisor tooth development persisted from E12.5 to 17.5 (Amen et al., 2007). Is it reasonable to hypothesize that whereas AXIN2 loss-of-function affects both molar and incisor development (Lammi et al., 2004), hypomorphic AXIN2 alleles would affect incisors more often due to the persistent expression of LEF1 in these developing teeth, when compared to molars. In addition, the interactions between LEF1 and PITX2 (paired-like homeodomain transcription factor 2) could not only play important roles in dental development, but also in certain types of cancer (Amen et al., 2007).

In summary, we provide further evidence suggesting a role of AXIN2 in tooth agenesis. Future studies should identify which specific tooth agenesis subphenotypes are consequence of AXIN2 genetic variation. A subset of these cases could have an increased susceptibility for colon cancer or other types of tumors. It would be an impressive development if a patient could walk into a dental office with a congenitally missing tooth, something a medical doctor would most likely overlook, and just by taking a spit sample, the dentist was able to tell if the patient was at high risk for colorectal cancer.