Journal of Dental Research

http://jdr.sagepub.com/

Identification of Genetic Risk Factors for Maxillary Lateral Incisor Agenesis

M. Alves-Ferreira, T. Pinho, A. Sousa, J. Sequeiros, C. Lemos and I. Alonso
J DENT RES 2014 93: 452 originally published online 19 February 2014
DOI: 10.1177/0022034514523986

The online version of this article can be found at:

http://jdr.sagepub.com/content/93/5/452

Published by:

http://www.sagepublications.com

On behalf of:
International and American Associations for Dental Research

Additional services and information for Journal of Dental Research can be found at:

Email Alerts: http://jdr.sagepub.com/cgi/alerts

Subscriptions: http://jdr.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - Apr 15, 2014

OnlineFirst Version of Record - Feb 19, 2014

What is This?

Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.

© International & American Associations for Dental Research

research-article2014
JDR 93510.1177/0022034514523986

RESEARCH REPORTS
Clinical

M. Alves-Ferreira1, T. Pinho1,2,
A. Sousa1,3, J. Sequeiros1,3,4, C. Lemos1,3,
Identification of Genetic Risk
and I. Alonso1,3* Factors for Maxillary Lateral
1
UnIGENe, Instituto Biologia Molecular Celular, Universidade
do Porto, Porto, Portugal; 2Centro de Investigação Ciências da
Incisor Agenesis
Saúde, Instituto Superior de Ciências Saúde–Norte / CESPU,
Gandra-PRD, Portugal; 3Instituto Ciências Biomédicas Abel
Salazar, Universidade do Porto, Porto, Portugal; and 4CGPP,
Instituto Biologia Molecular Celular, Universidade do Porto,
Porto, Portugal; *corresponding author, ialonso@ibmc.up.pt

J Dent Res 93(5):452-458, 2014

Abstract
Tooth agenesis affects 20% of the world population,
and maxillary lateral incisors agenesis (MLIA) is one
of the most frequent subtypes, characterized by the
absence of formation of deciduous or permanent lat-
eral incisors. Odontogenesis is a complex mechanism
regulated by sequential and reciprocal epithelial-
mesenchymal interactions, controlled by activators
and inhibitors involved in several pathways.
Disturbances in these signaling cascades can lead to
abnormalities in odontogenesis, resulting in alterations
in the formation of the normal teeth number. Our aim
was to study a large number of genes encoding either
transcription factors or key components in signaling
pathways shown to be involved in tooth odontogene-
sis. We selected 8 genes—MSX1, PAX9, AXIN2, EDA,
SPRY2, TGFA, SPRY4, and WNT10A—and performed
one of the largest case-control studies taking into
account the number of genes and variants assessed,
aiming at the identification of MLIA susceptibility fac-
tors. We show the involvement of PAX9, EDA, SPRY2,
SPRY4, and WNT10A as risk factors for MLIA.
Additionally, we uncovered 3 strong synergistic inter-
actions between MLIA liability and MSX1-TGFA,
AXIN2-TGFA, and SPRY2-SPRY4 gene pairs. We
report the first evidence of the involvement of sprouty
genes in MLIA susceptibility. This large study results
in a better understanding of the genetic components
and mechanisms underlying this trait.
Introduction
M axillary lateral incisors agenesis (MLIA) is characterized by the absence
of formation of deciduous or permanent upper lateral incisors. Tooth
agenesis affects approximately 20% of the world population (Vastardis, 2000),
and MLIA is one of the most frequent forms of hypodontia (agenesis of fewer
than 6 teeth, excluding the third molars) (Arte et al., 2001), with a prevalence
ranging from 1% to 4%. In Portugal, the prevalence of MLIA was estimated at
1.3% (Pinho et al., 2005). We found evidence of familial aggregation of MLIA
in a sample of Portuguese families, suggesting the presence of a strong genetic
component (relative risk =15) for this trait (Pinho et al., 2010a).
Odontogenesis is a complex mechanism regulated by sequential and recip-
rocal epithelial-mesenchymal interactions, controlled by genetic factors, and
is responsible for the determination of the position, number, shape, and size
of teeth (Thesleff, 2003). Several genes have been identified as being
expressed during odontogenesis, and it is known that mutations in MSX1,
PAX9, AXIN2, EDA, SPRY2, TGFA, SPRY4, and WNT10A genes are involved
in several forms of tooth agenesis, including syndromes in which tooth agen-
esis is a regular feature (Nieminen, 2009). These genes are related with all
major signaling pathways and with the transcription factors mediating these
signal transduction cascades. New findings on the function of these genes that
are mutated in particular human syndromes further strengthen the important
role of these genes in tooth agenesis (Nieminen, 2009).
Here, we present a case-control study with the largest number of genes and
single-nucleotide polymorphisms (SNPs) assessed in the same population,
aiming at identifying MLIA susceptibility factors focusing on these genes
involved in tooth odontogenesis.

KEY WORDS: hypodontia, odontogenesis, asso- Material & Methods

ciation studies, candidate genes, polymorphism,
gene-gene interaction.
DOI: 10.1177/0022034514523986
Received July 31, 2013; Last revision January 21, 2014;
Accepted January 24, 2014
A supplemental appendix to this article is published elec-
tronically only at http://jdr.sagepub.com/supplemental.
© International & American Associations for Dental Research
Subjects
A case-control sample of 306 unrelated Portuguese individuals was ascer-
tained: 102 individuals with MLIA and 204 controls (without any kind of
hypodontia) (Table 1). We had a special concern in obtaining a high
case:control ratio, to increase the statistical power of the study. In all cases,
MLIA was confirmed radiographically, and both groups were observed by
experienced clinicians. Informed consent was obtained in all cases, and this
study followed the “Strengthening the Reporting of Observational Studies in
Epidemiology” guidelines (von Elm et al., 2008).

452
Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.

© International & American Associations for Dental Research

J Dent Res 93(5) 2014  453
Maxillary Lateral Incisor Agenesis

Candidate Gene and SNP Selection Table 1. Clinical and Demographic Characteristics of the Study
Sample
Details on SNP selection and genotyping are presented as sup-
plementary material. Study Sample No.

Cases 102
Statistical Analysis   MLIA only 86
The statistical power of our sample was calculated with the    Bilateral (12 and 22) 50
Genetic Power Calculator (http://pngu.mgh.harvard.edu/    Unilateral (12) + microdontia (22) 11
   Unilateral (22) + microdontia (12) 1
~purcell/gpc). We performed these estimates assuming a fre-
  Unilateral (12) 12
quency of 0.1 for the high-risk allele, a relative risk of 2.5 for
  Unilateral (22) 12
the homozygous genotype and 2.2 for the heterozygous geno-
  MLIA + other agenesis 16
type, and a prevalence of 1.3% for MLIA in the Portuguese
   12; 22; 18; 28; 38; 48 1
population, based on α = 0.05 (Pinho et al., 2005).    12; 22; 18 1
Hardy-Weinberg (H-W) equilibrium was tested with SNPator    12; 22; 45 1
software (Morcillo-Suarez et al., 2008). To compare allele fre-    12; 22; 35 1
quencies between cases and controls, a chi-square test was used,    12; 22; 45; 35 3
and odds ratios (ORs) were estimated with 95% confidence    12; 35; 45 1
intervals (CI), also per SNPator.    22; 35; 45 1
A logistic regression was performed (with the commonest    12; 22; 31; 41 2
allele as the reference category) to evaluate the genotypic asso-   12; 22;41 1
ciations. The significance level was set to α = 0.007 (consider-    12; 22; 15; 25 1
ing 7 logistic regressions), based on the Bonferroni test, to    12; 32; 42 1
correct for multiple comparisons. As EDA gene is located in the    12; 41; 45; and microdontia (22) 1
X chromosome, the analyses were performed taking sex into    22; 31; 41; and microdontia (13) 1
account. These analyses were performed with SPSS 18.  Males 29
Haplotype frequencies in cases and controls were compared  Females 73
with Haploview 4.1 (Barrett et al., 2005); haplotype sequences Controls 204
were analyzed from 5′ to 3′. To correct for multiple testing,  Males 85
 Females 119
10,000 permutations were used when allelic and haplotype fre-
quencies were estimated. MLIA, maxillary lateral incisors agenesis
To explore possible gene-gene interactions, we used
Multifactor Dimensionality Reduction software (version 2.0),
which is a nonparametric and genetic model-free approach that
can identify combinations of SNPs involved in disease suscep- PAX9 Genotype as a MLIA Risk Factor
tibility (Ritchie et al., 2001; Moore, 2004). Subsequently, we The A allele of rs8004560 was associated with an increased
used the Permutation Testing Module (version 1.0) of Multifactor MLIA susceptibility (p = .02); however, this result did
Dimensionality Reduction to correct our results for multiple not remain significant after permutation-based correction
testing, based on a 1000-fold permutation test. (Table 2).
Importantly, 2 highly significant genotypic associations were
Results found. The AA genotype of rs7149262 was correlated with
higher trait susceptibility (p < .001). In agreement with the
Here we present a large case-control study composing the analy- allelic results, the AA genotype of rs8004560 was associated
sis of common variation in 8 genes involved in tooth odontogen- with an increased MLIA susceptibility (p = .003). These asso-
esis to identify MLIA susceptibility factors. Our sample had a ciations remained significant after Bonferroni correction
power of 72% to detect a significant association. Genotypic fre- (Table 3). The haplotypic analysis did not reveal any significant
quencies in the control group were in H-W equilibrium (p > .05). association between PAX9 and MLIA susceptibility (data not
In the cases group, H-W equilibrium deviations were found for shown).
some SNPs, although in all these cases, an association with MLIA
susceptibility was confirmed. We were able to show the involve-
ment of PAX9, EDA, SPRY2, SPRY4, and WNT10A as risk factors
Nominal Significant Associations of EDA SNPs With
for MLIA. No significant allelic, genotypic, and haplotypic asso- MLIA Susceptibility
ciations were found regarding AXIN2, TGFA, and MSX1 genes The A allele of rs1160315, the C allele of rs2428151, the T allele
(data not shown). All the analyses, were repeated, excluding of rs2520378, and the A allele of rs2853659 were associated
patients with MLIA and other agenesis (always fewer than 6 with an increase in MLIA susceptibility. None of the results
teeth), and the results were similar (data not shown). We assessed remained significant after multiple-testing correction by permu-
gene-gene interactions and uncovered 3 strong synergistic inter- tation tests (Table 2).
actions between MLIA liability and the following gene pairs: Regarding the genotypic analysis, the GA genotype of
MSX1-TGFA, AXIN2-TGFA, and SPRY2-SPRY4. rs12853659 presented a higher risk for MLIA (p = .021), while the

Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.

© International & American Associations for Dental Research

454  Alves-Ferreira et al. J Dent Res 93(5) 2014

Table 2.  Allelic Association Results for PAX9, EDA, SPRY2, and WNT10A Genes

Alleles (%)

  Cases Controls χ2 Odds Ratio (95% CI) p

PAX9 SNP  
 rs8004560 5.59 1.53 (1.07, 2.17) .02*
  A 81 (39.7) 123 (30.1)  
  G 123 (60.3) 285 (69.9)  
EDA SNP  
 rs1160315 4.30 1.48 (1.02, 2.14) .04*
  A 101 (57.7) 155 (48.0)  
  G 74 (42.3) 168 (52.0)  
 rs2428151 6.49 1.64 (1.12, 2.41) .01*
  C 117 (66.9) 178 (55.1)  
  T 58 (33.1) 145 (44.9)  
 rs2520378 5.16 1.82 (1.08, 3.07) .02*
  T 153 (87.4) 256 (79.3)  
  C 22 (12.6) 67 (20.7)  
 rs12853659 6.32 1.61 (1.11, 2.33) .01*
  A 103 (58.9) 152 (47.1)  
  G 72 (41.1) 171 (52.9)  
SPRY2 SNP  
 rs504122  
  A 89 (43.6) 146 (35.8) 3.54 1.39 (0.99, 1.96) .06
  G 115 (56.4) 262 (64.2)  
WNT10A SNP  
 rs11680244 0.05 1.15 (0.35, 3.78) .82
  A 4 (2.0) 9 (2.2)  
  G 200 (98.0) 391 (97.8)  
 rs2385199 10.25 1.98 (1.30, 3.01) .0014*
  G 51 (25.0) 58 (14.4)  
  A 153 (75.0) 344( 85.6)  
 rs7349332 3.34 1.54 (0.97, 2.46) .068
  T 36 (17.6) 49 (12.2)  
  C 168 (82.4) 353 (87.8)  

CI, confidence interval; EDA, ectodysplasin A; PAX9, paired box gene 9; SNP, single-nucleotide polymorphism; SPRY2, sprouty homolog 2 gene;
WNT10A, wingless-type MMTV integration site family, member 10A.
*p < .05.

GA genotype of rs1160315 (p = .015) and the CT of rs5936523
(p = .038) showed a protective effect. However, the results did not
remain significant after Bonferroni correction (Table 3).
In the haplotypic analysis, a specific EDA haplotype (T-G-T-
A-A-C-T-C-A-A-A-G-G-T-G-T) revealed a significant associa-
tion (p = .032) with increased MLIA susceptibility, although it
did not retain its significance after permutation correction.
SPRY2 SNP and Increased Risk for MLIA
No significant differences were found regarding allelic frequen-
cies of SPRY2 SNPs. However, we observed a trend toward an
increased risk for MLIA conferred by the A allele of rs504122
(p = .06) (Table 2).
Regarding the genotypic analysis, we found that the GA
genotype of rs504122 presented a higher risk for individuals
with MLIA (p = .005; OR = 2.49; 95% CI: 1.31, 4.73), which
remained significant after Bonferroni correction (Table 3). No
significant association was found between SPRY2 haplotypes
and MLIA.
SPRY4 Gene Haplotypes and MLIA Susceptibility
The allelic and genotypic analyses did not reveal any significant
association between this gene and MLIA susceptibility (data not
shown).
Importantly, the T-G-G-A-T-C haplotype revealed a nominal
significant association (p = .016) with an increased MLIA suscep-
tibility (OR = 3.46; 95% CI: 1.19, 10.08); however, this did not
remain significant after permutation-based correction (p = .137).
WNT10A Variants Increase the Risk for MLIA
The G allele of rs2385199 was associated with an increased
MLIA susceptibility (p = .0014), a result that remained signifi-
cant after permutation-based correction (Table 2).

Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.

© International & American Associations for Dental Research

J Dent Res 93(5) 2014  455
Maxillary Lateral Incisor Agenesis

Table 3.  Results from the PAX9, EDA, SPRY2, and WNT10A Multivariable Logistic Regression Analysis

Genotype (%)

  Cases Controls Odds Ratio (95% CI) p

PAX9 SNP  
 rs8004560 .011
  GG (ref) 45 (44.1) 104 (51.0) 1.00 (—)  
  GA 33 (32.4) 77 (37.7) 1.27 (0.72, 2.24) .412
  AA 24 (23.5) 23 (11.3) 3.03 (1.47, 6.28) .003*
 rs7149262 .001
  CC (ref) 66 (64.7) 132 (64.7) 1.00 (—)  
  AC 23 (22.6) 68 (33.3) 0.88 (0.49, 1.58) .662
  AA 13 (12.7) 4 (2.0) 9.39 (2.79, 30.97) <.001*
EDA SNP  
 rs1160315 .019
  GG(ref) 15 (44.1) 30 (51.0) 1.00 (—)  
  GA 33 (32.4) 62 (37.7) 0.05 (0.01, 0.57) .015
  AA 25 (23.5) 27 (11.3) 0.33 (0.02, 5.72) .446
 rs5936523 .003
  CC(ref) 14 (64.7) 11 (64.7) 1.00 (—)  
  CT 23 (22.6) 67 (33.3) 0.46 (0.22, 0.96) .038
  TT 36 (12.7) 41 (2.0) 2.59 (0.82, 8.21) .105
 rs12853659 .038
  GG(ref) 12 (44.1) 31 (51.0) 1.00 (—)  
  GA 38 (32.4) 61 (37.7) 14.43 (1.51, 138.12) .021
  AA 23(23.5) 27 (11.3) 3.24 (0.21, 49.00) .397
SPRY2 SNP  
 rs504122 .02
  GG (ref) 25 (24.5) 81 (39.7) 1.00 (—)  
  AA 12 (11.8) 23 (11.3) 1.97 (0.73, 5.28) .18
  GA 65 (63.7) 100 (49.0) 2.49 (1.31, 4.73) .005*
WNT10A SNP  
 rs2385199 .006*
  GG (ref) 56 (54.9) 147 (73.1) 1.00 (—)  
  GA 41(40.2) 50 (24.9) 2.14 (1.28, 3.58) .004*
  AA 5 (4.9) 4 (2.0) 3.26 (0.84, 12.58) .09

CI, confidence interval; ref, reference category.

After Bonferroni correction, significance level was set at p < .007.
*

Importantly, a highly significant genotypic association was
found for this SNP. Its GA genotype was correlated with higher
MLIA susceptibility (p = .004) that withstood Bonferroni cor-
rection (Table 3). The haplotypic analysis showed a significant
association between MLIA susceptibility and the G-G-T haplo-
type that remained significant after permutation-based correc-
tion (p = .0176).
Functional Impact and Gene-Gene Interactions
We performed an in silico analysis using FuncPred. Particular
attention was paid to rs504122, a nonsynonymous polymor-
phism that results in a missense mutation (p.Pro106Ser), pre-
dicted as benign. We also analyzed SNPs that are in linkage
disequilibrium (LD) with those found to be associated with
MLIA. This analysis showed that SNPs in LD with rs504122
(rs497857 and rs541731) may alter transcription factor binding
sites in this gene. Additionally, this analysis predicts that
rs11911 (which is in LD with rs504122) may affect miRNA
binding sites.
We found a strong synergistic interaction, as shown in the den-
dogram (Figure 1a) for the best model, between the AC and CC
genotypes of rs3775261 in the MSX1 gene and the CT genotype of
rs3771494 of TGFA, with a testing balanced accuracy (TBA) of
0.62, a cross-validation consistency (CVC) of 10/10, and a signifi-
cant p value (p = .015) after a 1000-fold permutation test. The CVC
is the number of times that a model is selected as the best model
among the validation sets indicating the importance of the model
(Heidema et al., 2007). A best model was selected to maximize the
testing accuracy; this is the model most likely to generalize to inde-
pendent data sets (Beretta et al., 2008).
In addition, we have found a significant interaction between
the TT genotype (rs426081) of TGFA and the AC and AA geno-
types of rs3923086 in the AXIN2 gene, with a TBA of 0.63 and
a CVC of 10/10 (Figure 1b). After permutation testing, this
model remained significant (p = .026).

Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.

© International & American Associations for Dental Research

456  Alves-Ferreira et al.
Figure 1. Interaction dendogram among (a) MSX1-TGFA, (b) TGFA-
AXIN2, and (c) SPRY2-SPRY4 in maxillary lateral incisors agenesis
susceptibility. The color of the line indicates the type of interaction.
Black and dark gray suggest a synergistic relationship. Light gray
suggests independence. Shorter length of the lines shows a stronger
interaction.
Potential interaction effects between SPRY2 and SPRY4 were
also examined, and the dendogram (Figure 1c) for the best
model showed a strong synergistic interaction between the AG
genotype (rs504122) of SPRY2 gene and the AA and AC geno-
types (rs12516550) in the SPRY4 gene, with a TBA of 0.64 and
a CVC of 10/10 (p = .012) after permutation testing.
Discussion
MLIA has a high impact in individuals with this trait; therefore,
it is important to understand the underlying causes of tooth
agenesis susceptibility.
This study aimed to identify polymorphisms in a group of
genes underlying multiple pathways related with MLIA through
a large-scale candidate gene approach, thus allowing a better
understanding of the mechanisms involved in MLIA. Our
results show that PAX9, EDA, SPRY2, SPRY4 and WNT10A
genes play a role in MLIA susceptibility as protective factors or
as risk factors, with important interactions between them.
Role of Transcription Factors in Hypodontia
PAX9 is a transcription factor involved in craniofacial and tooth
development. Some mutations in the coding region of PAX9
gene were already identified in individuals with nonsyndromic
oligodontia (Stockton et al., 2000; Lammi et al., 2003;
Jumlongras et al., 2004; Klein et al., 2005; Suda et al., 2011;
Zhu et al., 2012). In a previous study, a polymorphism
(rs4904210) in PAX9 gene was identified by sequencing 12
Portuguese families with MLIA; however, no statistically sig-
nificant association was found (Pinho et al., 2010b). Wang et al.
(2011) also performed a sequencing analysis of this gene in a
Chinese family with oligodontia, and the same SNP was found
in the proband and in the maternal family members; therefore, it
is likely that this polymorphism is involved in oligodontia. By
applying a case-control approach, we found that rs7149262 and
rs8004560 are associated with a high risk of MLIA. The A allele
of both SNPs were found at a significantly higher frequency in
cases when compared with controls; this fact is reinforced by the
Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.
© International & American Associations for Dental Research
J Dent Res 93(5) 2014
high risk for MLIA susceptibility conferred by the AA genotype.
Therefore, this study strengthens the role of PAX9 gene in the
development of the maxillary lateral incisors.
TNF Signaling and MLIA Susceptibility
Although some mutations in the EDA gene have been identified
as the cause of X-linked hypohidrotic ectodermal dysplasia,
they have been pointed out as a cause of nonsyndromic tooth
agenesis that results in a unique hypodontia phenotype pre-
dominantly responsible for incisor agenesis patterns (Tarpey et
al., 2007; Li et al., 2008). Until now, no studies have been per-
formed to associate this gene with isolated tooth agenesis. EDA
encodes ectodysplasin A, a protein that belongs to the TNF-
ligand family. Although our results did not remain significant
after multiple-testing correction, our analysis revealed 3 allelic
associations with an increased MLIA susceptibility. We also
found 3 nominal significant genotypic associations: 1 associa-
tion confers a higher risk for individuals with MLIA and 2
associations, a protective effect. A haplotype of these 16 SNPs
was found to be overrepresented in MLIA individuals when
compared with controls. Therefore, our results provide the first
evidence that polymorphisms in the gene encoding ectodyspla-
sin A might be associated, either as protective factors or as risk
factors, with MLIA in the Portuguese population.
FGF Signaling Inhibitors in Hypodontia
Previous studies provided evidence for the involvement of FGF
signaling in odontogenesis (Jernvall and Thesleff, 2000; Klein
et al., 2006; Klein et al., 2008). Therefore, we selected SPRY2
and SPRY4 as candidate genes owing to their role as inhibitors
of this pathway.
In mice, SPRY2 gene expression possibly prevents SHH
expression in the diastema bud by inhibiting FGF signaling
from the mesenchyme. Another member of the sprouty family is
the SPRY4 gene, which antagonizes the response to FGF signal-
ing from the enamel knot (Klein et al., 2006). Importantly, stud-
ies in mice show that double knockout of Spry2 and Spry4
results in enlarged incisors, with enamel present on both the
lingual and labial sides (Klein et al., 2008). Additionally, in both
Spry2- and Spry4- null mice embryos, a toothless gap known as
diastema develops supernumerary teeth, reinforcing their role in
normal tooth formation.
We found that the GA genotype of rs504122 in the SPRY2
gene leads to an increased risk in individuals with MLIA, point-
ing to an effect of this gene in our sample. Additionally, we
found a SPRY4 haplotype associated with an increased MLIA
risk. Although this result does not stand after permutation test-
ing, we cannot exclude a possible involvement of this haplotype
in MLIA etiology.
WNT Signaling and MLIA
A number of studies aimed to find causal mutations implicated
in teeth agenesis. More recently, WNT10A gene mutations have
been linked to syndromic and isolated severe tooth agenesis.
Additionally, some studies found that this gene could play a role
J Dent Res 93(5) 2014  457
Maxillary Lateral Incisor Agenesis
in nonsyndromic tooth agenesis; however, the frequency of
WNT10A mutations is much higher when more than 4 teeth are
missing, showing that this gene must be involved in severe
forms of this trait (van den Boogaard et al., 2012).
Our results confirm the involvement of WNT10A and the
WNT signaling pathway in this trait susceptibility with the con-
tribution of a common variant.
Gene-Gene Interactions in MLIA Susceptibility
Although we did not find any significant association in the
allelic, genotypic, and haplotypic analyses of AXIN2, TGFA, and
MSX1 and MLIA susceptibility, our analysis revealed 2 strong
significant interactions between TGFA-AXIN2 and MSX1-TGFA.
Some gene-gene interactions have been reported in cases of
tooth agenesis. Vieira et al. (2004) investigated for the first time
the association between human tooth agenesis (MLIA and other
agenesis) and the TGFA gene and did not find any statistically
significant evidence of a TGFA-MSX1 interaction. Conversely, a
significant interaction was found between TGFA and AXIN2;
thus, we can hypothesize the involvement of this WNT pathway
gene in the formation of the maxillary lateral incisors.
We found a strong synergistic interaction between SPRY2
and SPRY4, confirmed by a 1000-fold permutation test. This is
the first study that explored a possible involvement of SPRY2
and SPRY4 in human tooth agenesis. We hypothesize that the
biological changes related to genetic variations in SPRY2-
SPRY4 and their interaction may cause an increased expression
of these genes, preventing SHH expression in the lateral incisor
bud by inhibiting FGF signaling. Therefore, the growth activa-
tion pathway would be blocked, and bud growth would stop
(Peterkova et al., 2009).
Gene-gene interaction analyses are becoming more common
because there is growing evidence that gene-gene interactions
are paramount to determine the susceptibility to common dis-
eases. Our major limitation is the sample size, which may be
considered too small for this type of study. Nevertheless, we
were able to identify common variants as MLIA susceptibility
factors, suggesting that the sample size would be limitative only
for the identification of rare variants. Although the molecular
mechanisms involved in tooth agenesis remain unknown, our
results provide the first evidence of the involvement of sprouty
genes. We also found new polymorphisms that provide prelimi-
nary evidence of a new susceptibility marker for MLIA, leading
to a better understanding of the genetic mechanisms underlying
this trait.
Acknowledgments
We would like to thank all the participants of this study and
Laboratório de Análises Clínicas, Centro Médico da Praça (São
João da Madeira), for collecting control group samples. This
work was supported by national funds though Fundação para a
Ciência e Tecnologia (PEst-C/SAU/LA0002/2011), cofunded by
FEDER through the Programa Operacional Factores de
Competitividade–COMPETE and by a grant from Cooperativa
de Ensino Superior Politécnico e Universitário (01-CI-PI-
CICS-2011; 07/GCD/CICS/09). I.A. is funded by Programa
Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.
© International & American Associations for Dental Research
Ciência, POPH–QREN–Tipologia 4.2–Promoção do Emprego
Científico, cofunded by ESF and national funds by the Ministério
da Ciência e Ensino Superior. The authors declare no potential
conflicts of interest with respect to the authorship and/or publi-
cation of this article.
References
Arte S, Nieminen P, Apajalahti S, Haavikko K, Thesleff I, Pirinen S (2001).
Characteristics of incisor-premolar hypodontia in families. J Dent Res
80:1445-1450.
Barrett JC, Fry B, Maller J, Daly MJ (2005). Haploview: analysis and visu-
alization of LD and haplotype maps. Bioinformatics 21:263-265.
Beretta L, Cappiello F, Moore JH, Scorza R (2008). Interleukin-1 gene
complex single nucleotide polymorphisms in systemic sclerosis: A
further step ahead. Hum Immunol 69(3):187-192.
Heidema AG, Feskens EJ, Doevendans PA, Ruven HJ, van Houwelingen
HC, Mariman EC, et al. (2007). Analysis of multiple SNPs in genetic
association studies: comparison of three multi-locus methods to priori-
tize and select SNPs. Genet Epidemiol 31:910-921.
Jernvall J, Thesleff I (2000). Reiterative signaling and patterning during
mammalian tooth morphogenesis. Mech Dev 92:19-29.
Jumlongras D, Lin JY, Chapra A, Seidman CE, Seidman JG, Maas RL, et al.
(2004). A novel missense mutation in the paired domain of PAX9
causes non-syndromic oligodontia. Hum Genet 114:242-249.
Klein ML, Nieminen P, Lammi L, Niebuhr E, Kreiborg S (2005). Novel
mutation of the initiation codon of PAX9 causes oligodontia. J Dent Res
84:43-47.
Klein OD, Minowada G, Peterkova R, Kangas A, Yu BD, Lesot H, et al.
(2006). Sprouty genes control diastema tooth development via bidirec-
tional antagonism of epithelial-mesenchymal FGF signaling. Dev Cell
11:181-190.
Klein OD, Lyons DB, Balooch G, Marshall GW, Basson MA, Peterka M,
et al. (2008). An FGF signaling loop sustains the generation of differ-
entiated progeny from stem cells in mouse incisors. Development
135:377-385.
Lammi L, Halonen K, Pirinen S, Thesleff I, Arte S, Nieminen P (2003). A
missense mutation in PAX9 in a family with distinct phenotype of oli-
godontia. Eur J Hum Genet 11:866-871.
Li S, Li J, Cheng J, Zhou B, Tong X, Dong X, et al. (2008). Non-syndromic
tooth agenesis in two Chinese families associated with novel missense
mutations in the TNF domain of EDA (ectodysplasin A). PLoS One
3:e2396.
Moore JH (2004). Computational analysis of gene-gene interactions using
multifactor dimensionality reduction. Expert Rev Mol Diagn 4:795-
803.
Morcillo-Suarez C, Alegre J, Sangros R, Gazave E, de Cid R, Milne R, et al.
(2008). SNP analysis to results (SNPator): a web-based environment
oriented to statistical genomics analyses upon SNP data. Bioinformatics
24:1643-1644.
Nieminen P (2009). Genetic basis of tooth agenesis. J Exp Zool B Mol Dev
Evol 312B:320-342.
Peterkova R, Churava S, Lesot H, Rothova M, Prochazka J, Peterka M,
et al. (2009). Revitalization of a diastemal tooth primordium in Spry2
null mice results from increased proliferation and decreased apoptosis.
J Exp Zool B Mol Dev Evol 312B:292-308.
Pinho T, Tavares P, Maciel P, Pollmann C (2005). Developmental absence of
maxillary lateral incisors in the Portuguese population. Eur J Orthod
27:443-449.
Pinho T, Maciel P, Lemos C, Sousa A (2010a). Familial aggregation of max-
illary lateral incisor agenesis. J Dent Res 89:621-625.
Pinho T, Silva-Fernandes A, Bousbaa H, Maciel P (2010b). Mutational
analysis of MSX1 and PAX9 genes in Portuguese families with maxil-
lary lateral incisor agenesis. Eur J Orthod 32:582-588.
Ritchie MD, Hahn LW, Roodi N, Bailey LR, Dupont WD, Parl FF, et al.
(2001). Multifactor-dimensionality reduction reveals high-order inter-
actions among estrogen-metabolism genes in sporadic breast cancer.
Am J Hum Genet 69:138-147.
458  Alves-Ferreira et al. J Dent Res 93(5) 2014

Stockton DW, Das P, Goldenberg M, D’Souza RN, Patel PI (2000).
Mutation of PAX9 is associated with oligodontia. Nat Genet 24:18-
19.
Suda N, Ogawa T, Kojima T, Saito C, Moriyama K (2011). Non-syndromic
oligodontia with a novel mutation of PAX9. J Dent Res 90:382-386.
Tarpey P, Pemberton TJ, Stockton DW, Das P, Ninis V, Edkins S, et al.
(2007). A novel Gln358Glu mutation in ectodysplasin A associated with
X-linked dominant incisor hypodontia. Am J Med Genet A 143:390-
394.
Thesleff I (2003). Epithelial-mesenchymal signaling regulating tooth mor-
phogenesis. J Cell Sci 116(Pt 9):1647-1648.
van den Boogaard MJ, Creton M, Bronkhorst Y, van der Hout A, Hennekam
E, Lindhout D, et al. (2012). Mutations in WNT10A are present in more
than half of isolated hypodontia cases. J Med Genet 49:327-331.
Vastardis H (2000). The genetics of human tooth agenesis: new discoveries
for understanding dental anomalies. Am J Orthod Dentofacial Orthop
117:650-656.
Vieira AR, Meira R, Modesto A, Murray JC (2004). MSX1, PAX9, and
TGFA contribute to tooth agenesis in humans. J Dent Res 83:723-727.
von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke
JP (2008). The Strengthening the Reporting of Observational Studies in
Epidemiology (STROBE) statement: guidelines for reporting observa-
tional studies. J Clin Epidemiol 61:344-349.
Wang J, Jian F, Chen J, Wang H, Lin Y, Yang Z, et al. (2011). Sequence
analysis of PAX9, MSX1 and AXIN2 genes in a Chinese oligodontia
family. Arch Oral Biol 56:1027-1034.
Zhu J, Yang X, Zhang C, Ge L, Zheng S (2012). A novel nonsense mutation in
PAX9 is associated with sporadic hypodontia. Mutagenesis 27:313-317.

Downloaded from jdr.sagepub.com at UNIV OF SOUTH DAKOTA on June 10, 2014 For personal use only. No other uses without permission.

© International & American Associations for Dental Research