Introduction

Mandibular retrognathism is defined as an abnormally posterior positioned mandible in relation to the anterior skull base [3]. Although the relation of the jaw bases and the craniofacial morphology determine an individual’s malocclusion [7], this craniofacial dysgnathia is often associated with a skeletal class II malocclusion, which occurs in about 23–29% of the population worldwide [8]. Mandibular retrognathism has a polygenic etiological background [1, 2, 14, 26]. Across different populations, some genes have been identified as etiological factors of mandibular retrognathism [9].

Single nucleotide polymorphisms (SNPs) are variations in the DNA sequence that occur, when a single nucleotide varies between members of a biological species or paired chromosomes in an individual. SNPs can influence the expression and/or functions of genes and have been explored in complex traits, including skeletal malocclusions and other dentofacial traits [13]. Previous investigations from different research groups revealed that a variety of SNPs are involved in the mandibular retrognathism phenotype [1, 2, 12, 14, 15, 26]. A recent study in a German sample showed an association between a SNP in the gene encoding the transforming growth factor beta receptor type 2 (TGFBR2) with mandibular retrognathism [12].

Growth factors are mostly proteins or steroid hormones that act as signaling molecules regulating many cellular functions such as cell proliferation, survival, and differentiation. Some growth factors stimulate a cellular response by binding to specific receptors [10]. The epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase that is activated by binding of its ligand, the epidermal growth factor (EGF), resulting in receptor dimerization and autophosphorylation, and activation of signaling pathways promoting proliferation. EGF and EGFR play important roles in skeletal biology [22] and their function is necessary for normal craniofacial development [19]. Resorption, formation, and maintenance of bone are coordinated by the action of several hormones, transcription factors, and growth factors [22]. Since growth factors promote the events of cell growth, the investigation of their potential role as predictive biomarkers for skeletal malocclusions is an exciting approach, which could enable early and individualized diagnostic identification of retrognathic mandibular development by means of genetic screening tests in the future. Therefore, in this study, we investigated whether SNPs in the genes encoding EGF and EGFR are involved in the etiology of mandibular retrognathism of German teenagers.

Materials and methods

This study was approved by the Human Ethics Committee at the University of Regensburg (number 19-1549-101) and conducted according to the ethical principles of the Helsinki Declaration. Informed consent was obtained from all patients and their parents or legal guardians. Furthermore, an age-appropriate assent document was also used for patients younger than 14 years.

The Strengthening the Reporting of Genetic Association study (STREGA) statement checklist [17] was used to design and report this study, and the checklist is presented in supplementary table 1.

Sample size calculation, recruitment and collection of this nested case–control study were previously described by Kirschneck et al. [12]. Briefly, German orthodontic patients were consecutively recruited during orthodontic treatment in 2020 and 2021, and the sample size was determined with a power of 0.80%, α of 0.05, and an effect size of 0.225.

Adults and patients with syndromes, congenital alterations including dental agenesis of permanent tooth/teeth (except for third molar agenesis), patients with cleft lip and/or palate (syndromic or isolated forms of cleft), and patients with facial trauma were excluded. Furthermore, after the cephalometric analysis, patients with mandibular prognathism (SNB > 82°) were also excluded. Only one individual per family was included to avoid genetic bias. In addition, to minimize genetic and phenotypic variance and maximize data interpretation, only patients with Middle European ancestry were included [12].

All patients included were teenagers not biologically related and age ranged from 10–18 years.

Cephalometric analysis

Digital pretreatment lateral cephalograms as part of patients’ orthodontic records with the mandible in maximal intercuspation were used in the cephalometric analysis. Measurements were performed by two trained and calibrated orthodontists who presented good interexaminer and intraexaminer reliability as previously reported in Kirschneck et al. [12].

The radiographs were imported as lossless TIF files into the software ivoris® analyze pro (Computer konkret AG, Falkenstein, Germany, version 8.2.15.110) and calibrated. Cephalometry based on Segner and Hasund [23] was conducted digitally, although only skeletal parameters were considered for analyses. The anatomical landmarks point A, point B, sella (S), and nasion (N) were determined manually using the cephalometric analysis software (ivoris analyze pro), and the angular measurements SNB and ANB were calculated (Fig. 1).

Fig. 1 Abb. 1
figure 1

Determination of mandibular retrognathism and skeletal class using cephalometric variables. S sella, N nasion, A point A, B point B, angle 1 SNB (degree of prognathism of the mandible), angle 2 ANB (skeletal class)

Bestimmung einer mandibulären Retrognathie und der skelettalen Klasse anhand kephalometrischer Parameter. S Sella, N Nasion, A A-Punkt, B B-Punkt, Winkel 1 SNB (Prognathiegrad des Unterkiefers), Winkel 2 ANB (skelettale Klasse)

Full size image

The phenotype definition was as follows: patients with a retrognathic mandible were selected as cases (SNB < 78°), while patients with an orthognathic mandible were selected as controls (SNB = 78–82°). Patients with mandibular prognathism were excluded (SNB > 82°).

Genetic analysis

We selected candidate SNPs at the EGF and EGFR genes (Table 1) mostly based on the minor allele frequency reported in European populations (> 20%), the SNPs function, and based on previous results of studies investigating their association with several phenotypes suggesting clinical relevance of these SNPs (http://www.ncbi.nlm.nih.gov/snp/). SNPs in the promoter, coding, and intronic region were selected. The characteristics and description of the SNPs investigated in this study are presented in Table 1.

Table 1 Tab. 1 Characteristics of studied SNPsEigenschaften der untersuchten SNPs
Full size table

For the genotyping analysis, genomic DNA was isolated from buccal epithelial cells collected using two cytobrushes placed in extraction solution (Tris-HCl 10 mmol/L, pH 7.8; EDTA 5 mmol/L; SDS 0.5%, 1 mL). Briefly, proteinase K (100 ng/mL) was added to each tube. Ammonium acetate was also added to remove nondigested proteins and the solution was then centrifuged. DNA was precipitated with isopropanol and washed with ethanol. The DNA was quantified by spectrophotometry (Nanodrop 1000; Thermo Scientific, Wilmington, DE, USA) [12].

The selected SNPs were blindly genotyped via real-time polymerase chain reaction (PCR) using the Mastercycler® ep realplex‑S thermocycler (Eppendorf AG, Hamburg, Germany). The TaqMan technology was used. A negative control template was included in each reaction plate. In addition, 10% of the samples were randomly selected for repeated analysis and showed 100% concordance. Patients with not enough DNA or DNA samples that failed to be genotyped were excluded from further analyses.

Statistical analysis

The success genotyping rate was calculated for each SNP, and the Hardy–Weinberg equilibrium was obtained by Pearson χ2 test without correction, which was also used to evaluate the distribution of gender between groups. The Mann–Whitney test compared age and SNB medians.

Allele and haplotype frequency comparisons were performed by PLINK version 1.06 (https://zzz.bwh.harvard.edu/plink/ld.shtml). PLINK compares the frequencies between the major allele by Pearson χ2 test without correction and between the expected number of haplotypes by Fisher’s exact test.

The univariate Pearson χ2 without correction or Fisher’s exact test were performed for univariate genotypic analysis. For the multivariate analysis of the genotypes between the orthognathic and retrognathic mandible group a Poisson regression, which was adjusted by age, was used. Furthermore, the prevalence ratio (PR) and the 95% confidence interval (CI) were calculated. Statistical Package for Social Sciences (SPSS) version 25.0 (IBM Corp., Armonk, NY, USA) was employed for these analyses. Bilateral p-values were adopted for all tests, and p < 0.05 indicated a statistically significant difference.

Results

A total of 119 patients were included in this study (57 males and 62 females). Forty-five had an orthognathic mandible, while 74 showed a retrognathic mandible. Table 2 illustrates the characteristics of the sample. The SNB angle was statistically different between the orthognathic mandible and retrognathic mandible groups (p < 0.001).

Table 2 Tab. 2 Characteristics of the studied sampleEigenschaften des untersuchten Kollektivs
Full size table

Table 1 shows the details of the studied SNPs and the Hardy–Weinberg equilibrium values for each SNP in the total sample. All SNPs were within the Hardy–Weinberg equilibrium (p > 0.05).

The minor allele G in rs4444903 (EGF) was statistically more frequent in the orthognathic mandible group compared to the retrognathic group (p = 0.008). The haplotype formed by the mutant alleles for rs4444903|rs2237051 (EGF; G|A) was statistically more frequent in the orthognathic mandible group in comparison with the retrognathic mandible group (p = 0.007; Table 3).

Table 3 Tab. 3 Allele and haplotype distribution between groupsVerteilung der Allele und Haplotypen zwischen den Gruppen
Full size table

Table 4 shows the uni- and multivariate comparison of the genotypes between groups. The rs4444903 and rs2237051 (EGF), and rs2227983 (EGFR) SNPs were statistically associated with a decreasing chance of presenting with a retrognathic mandible. In the codominant model, the heterozygous patients for these SNPs had less chance of exhibiting a retrognathic mandible than the dominant homozygous patients. In the dominant model, heterozygous and recessive homozygous patients had less chance of developing a retrognathic mandible than the dominant homozygous patients (p < 0.05; PR < 1.0).

Table 4 Tab. 4 Univariate and multivariate analysis of genotypes comparison between groupsUnivariate und bivariate Analyse der Genotypen zum Vergleich zwischen den Gruppen
Full size table

Discussion

Mandibular retrognathism is a common maxillofacial alteration that can cause occlusal problems leading to class II malocclusion. The treatment of class II skeletal malocclusion due to mandibular retrognathism is one of the most common challenges in orthodontic practice. Mandibular retrognathism is also associated with esthetic problems and in severe cases with obstructive sleep apnea [11]. Therefore, studies investigating mandibular retrognathism are extremely important in the orthodontic literature and the number of research groups investigating the genetic background of this condition has been increasing in the past decade. In this study, some SNPs in the encoding genes EGF and EGFR were associated with mandibular retrognathism.

Mandibular retrognathism was previously associated with SNPs in MYO1H [1], MATN1 [2], ADAMTS9 [26], BMP2 [14], PTH, VDR, CYP24A1, and CYP27B1 [15] in different populations. Recently, a study indicated that TGFBR2 could be involved in mandibular retrognathism, and this finding was also observed in the sample evaluated in the present study [12]. In another study, four SNPs in transforming growth factor beta 1 were investigated: TGFB1 (rs1800469 and rs4803455) and TGBR2 (rs3087465 and rs764522), which are members of the growth factor family that has numerous key roles in the bone tissue controlling physiological processes [21]. The authors found that the SNP rs3087465 in TGFBR2 was associated with mandibular retrognathism [12]. Thus, we raised the hypothesis that other SNPs in growth factors encoding genes could be involved in the etiology of mandibular retrognathism.

Growth and development of the skeletal system is the main component or driver for postnatal somatic growth. During childhood and adolescence, bone lengthening and acquisition of peak bone mass and its trabecular organization are achieved, involving the production of calcified cartilage and its conversion and modeling into trabecular bone. Mandibular condylar cartilage is known as the center of most growth in the craniofacial complex and is associated with maxillofacial skeletal morphogenesis [20]. Although previous studies demonstrated some important functions of EGF and EGF-like ligands in regulating bone growth and modeling, the expression, roles, and action mechanisms of the EGF family of growth factors and its receptor in bone growth regulation are less explored than for other growth factors [27], especially in craniofacial growth and development.

In our research, the minor allele of the two studied SNPs in EGF (rs4444903 and rs2237051), as well as their haplotype (G|A), were associated with a decreasing risk of mandibular retrognathism. An in vitro experiment observed that EGF negatively regulated chondrogenesis through the inhibition of precartilage condensation and also by modulating signaling [28]. A study with an animal model also showed that defects in bone lengthening were observed in EGF transgenic mice [5]. EGF level can be modulated by the functional selected SNP in EGF at position 61 (A > G; SNP rs4444903), in which the GG genotype has a higher gene expression than the AA genotype [25]. This could explain why the AA genotype was more frequent in patients with a retrognathic mandible. Similar results were observed in Brazilian patients with dentofacial deformities, in which the SNP rs4444903 was involved in mandibular measurements [4].

The rs2237051 SNP in the coding region of the EGF gene is a missense substitution (Met708Ile) and was also associated with mandibular retrognathism in our sample. Although this SNP has never been previously explored in craniofacial growth, it has been explored in dental research in past years. The SNP rs2237051 was associated with generalized aggressive periodontitis [16, 26] and was recently associated with an increased risk of peri-implantitis [6]. In our sample, the GG genotype was more common in patients with a retrognathic mandible than in patients with an orthognathic mandible.

We also found an association between the EGFR and mandibular retrognathism. We observed that the SNP rs2227983 was involved in the risk of developing a retrognathic mandible. The SNP rs2227983 is located in the coding region of the gene and is a missense substitution at codon 497 (Arg497Lys) that leads to an attenuation in ligand binding and growth stimulation [18]. EGFR is expressed in chondroblasts of the developing ossification centers [27]. An animal model study showed that in egfr null mice the growth plate was significantly increased in the region of hypertrophic chondrocytes [24]. Newborn egfr−/− mice presented facial mediolateral defects including narrow, elongated snouts, and an underdeveloped lower jaw [19].

In our study, the three associated SNPs are classified as potentially functional: SNPs that can result in amino acid changes of the corresponding proteins (the missense SNPs), or the SNPs located in the promoter region of the gene and potentially influencing gene expression and EGF levels, which point them as interesting possible biomarkers. Briefly, our research raises potential future research avenues in orthodontic research, since the functional SNPs rs4444903, rs2237051, and rs2227983 could be biomarkers for mandibular retrognathism and should be explored in other populations.

Conclusion

Single nucleotide polymorphisms in the encoding genes EGF and EGFR were associated with mandibular retrognathism in a German sample and could be genetic biomarkers for early and individualized diagnostic identification of retrognathic mandibular development by means of genetic screening tests, which could supplement the cephalometric evaluation in young growing children for individualized orthodontic diagnostics, treatment planning, and prognosis.