Periodontology 2000, Vol. 39, 2005, 91117 Printed in the UK.

All rights reserved

Copyright Blackwell Munksgaard 2005

PERIODONTOLOGY 2000

The genetic basis of periodontitis

D E N I S F. K I N A N E , H I D E K I S H I B A & T H O M A S C. H A R T
It is now evident that there is a genetic basis for most diseases including periodontitis. Elucidation of the genetic basis of periodontitis should permit a better understanding of disease etiology, allowing improved classication, diagnosis, and treatment of periodontal diseases. Thus there has been enormous effort expended so far to identify gene polymorphisms associated with the risk for the periodontal diseases. Reports of genetic polymorphisms associated with periodontal disease are increasing, yet the limitations of these reports are not widely recognized. To appreciate these limitations it is necessary to understand the genetic basis of complex diseases. Thus this review has two sections, the rst aimed at introducing modern genetic principles and the second focusing on the utility of genetics in periodontics. It is increasingly evident that genetic variance is a major determinant of the differential risk for many human diseases (24, 49, 92). While microbial and other environmental factors initiate and modulate periodontal disease, individuals are known to respond differently to common environmental challenges, and this differential response is inuenced by the individuals genetic prole. Genes clearly play a role in the predisposition to and progression of periodontal diseases (62, 63, 73, 78, 122, 156). Support for the idea that genetic factors are important determinants of periodontitis susceptibility and progression comes from studies of humans and animals which indicate that genetic factors which impair inammatory and immune responses in general, affect periodontitis experience specically. It is hoped that identication of the specic genes causing periodontal disease susceptibility may have diagnostic and therapeutic value. To understand the potential clinical relevance of genetic variability on periodontitis it is necessary to understand how different genes can contribute to disease. There are estimated to be 20,00025,000 different genes in the human genome. Genes can exist in different forms or states. Geneticists refer to the different forms of a gene as allelic variants, or alleles. Genetic alterations could also change the transcript level of the protein, that is, the polymorphism may occur not in the protein coding region but in the gene promoter region, and thus inuence the amount of protein produced by the gene. Allelic variants of a gene differ in their nucleotide sequence. Different allelic forms of a gene can produce an identical protein or different isoforms of a protein. In the former case, the genetic polymorphism does not change the amino acid and the genetic change is said to be silent. In the latter case, a nucleotide change alters the amino acid composition of a protein. The results can range along a continuum of functional consequences, from no observable change in protein function, to a minor change in function, to a dramatic change or obliteration of function. Individual genetic changes that are causal of disease are typically the result of a genetic alteration that dramatically alters a proteins function. Such genetic changes are typically termed mutations and are rare on a population level, typically being present in less than 0.1% of the population. When a specic allele occurs in at least 1% of the population, it is said to be a genetic polymorphism. In contrast to mutations, these more common genetic polymorphisms are usually considered normal variants in the population. Common genetic polymorphisms may change the function of a protein, but usually the change is relatively minor. Consequently, the specic protein products of different alleles may function differently. These differences in physiological functioning of different proteins can be enhanced by certain environmental exposures, e.g. diet, smoking, or microbial factors. If the affected protein functions in a biological process, e.g. inammatory response to a specic microbial agent, certain polymorphisms may increase or decrease a persons risk for a disease phenotype. Whether a particular gene variant contributes to a disease phenotype depends upon the magnitude of the effect in contributing to the development of disease.

91

Kinane et al.

To appreciate the contribution of a genetic variant to a disease it is critical to understand how genes contribute to genetic diseases (Fig. 1). The contribution of an allelic variant to a disease can vary from being deterministic to having only a minor effect on the etiology. The manner and extent to which genetic factors play a role in disease have important implications for identifying the genetic basis of etiology and for utilizing this information for diagnosis and treatment. Geneticists have traditionally divided genetic diseases into two broad groups: simple Mendelian diseases and complex diseases. The distinction of these broad groups is based on the pattern of transmission of the disease, which reects the manner in which genes contribute to each disease.

Simple Mendelian diseases

Diseases that follow predictable and generally simple patterns of transmission have been called Mendelian conditions. The name reects the fact that these diseases occur in simple patterns in families and in most cases genetic alterations at a

single gene locus are the major determinant of the clinical disease phenotype. These diseases follow a classic Mendelian mode of inheritance (autosomal dominant, autosomal recessive or X-linked). The disease phenotype usually manifests over a broad range of environments, and although environmental factors and other genes can modify the clinical presentation, in most cases the mutation will manifest in a remarkably similar phenotype. The population prevalence of individual Mendelian diseases is rare (typically much less than 0.1%), with the exception of some unique populations that have been isolated from other human populations. Examples of Mendelian type diseases include amelogenesis imperfecta, Crouzon syn` drome, cleidocranio dysplasia, and PapillonLefevre syndrome (Table 1). When the gene responsible for a Mendelian disease has been identied, it is often possible to develop a diagnostic test to identify individuals who carry a disease-causing mutation in the responsible gene. Depending upon the mode of transmission, it is also possible to make fairly specic determinations of the probability of the

Phenotype

Environment

Genotype*

Biological Interactions*

Fig. 1. The classic relationship of phenotype (disease absence or presence), environment, and genotype. Genotype represents strictly genetic effects, environment simply environmental effects and Biological Interactions*

includes gene gene (epigenetic interactions) and gene environment interactions. For the periodontal disease phenotype, environmental risk factors include smoking status and plaque control.

Table 1. Examples of syndromic forms of periodontitis in which inheritance is Mendelian and due to a genetic alteration at a single gene locus
Condition ` PapillonLefevre syndrome HaimMunk syndrome EhlersDanlos syndrome type IV EhlersDanlos syndrome 8 Cyclic neutropenia Chronic familial neutropenia ChediakHigashi syndrome Congenital disorder of glycosylation type IIc (Leukocyte adhesion deciency type 2) Leukocyte adhesion deciency type I Biochemical tissue defect Cathepsin C Cathepsin C Collagen Collagen Neutrophil elastase Defect unknown Lysosomal trafcking regulator gene Glucose diphosphatefucose transporter-1 Leukocyte chain adhesion molecule CD18 Inheritance Autosomal recessive Autosomal recessive Autosomal dominant Autosomal dominant Autosomal dominant Autosomal dominant Autosomal recessive Autosomal recessive OMIM 245000 245100 130050 130080 162800 162700 214500 266265

Autosomal recessive

116920

92

Periodontal disease genetics

mutant gene being passed to a child and often it is possible to predict the course of clinical disease.

Complex genetic diseases

Genetically complex diseases differ from simple Mendelian diseases in several important ways. Genetically complex diseases do not follow a simple pattern of familial distribution or transmission. In contrast to simple Mendelian traits, which are caused by a single gene, these complex traits are the result of the interaction of alleles at multiple different gene loci. Environmental factors are usually etiologically important, and often necessary, in the development of complex diseases. On a population level, complex genetic diseases are much more common than simple Mendelian diseases, and many occur with a population prevalence of greater than 1%. Unlike rare mutations responsible for simple genetic diseases, the allelic variants important for many complex genetic diseases are common in the population. In contrast to mutations that can eliminate a gene product or change the protein product of a gene so signicantly that it acts to disrupt other biological processes, the individual genetic variants that are important in complex diseases are much less disruptive, and usually function within the normal range. Many disease-associated genetic polymorphisms are common in the population and can be present at allele frequencies of > 20%, with some disease-associated alleles reported in > 50% of populations studied (38). The hypothesis that most of the genetic risk for common, complex diseases is due to common polymorphisms at disease loci is known as the Common DiseaseCommon Variant hypothesis (142). Recently, the idea concept has emerged that at least a portion of susceptibility variants for common diseases are less common (510%) but more penetrative (141). The fact that these genetic alterations are not sufcient alone to cause disease has important implications. Disease-associated alleles may be present in a signicant proportion of the unaffected general population. Knowledge of the presence of one disease-associated allele in an individual does not provide enough information to make a clinical diagnosis. In contrast to genetic mutations that are often diagnostic for simple Mendelian conditions, the presence of a polymorphism in a complex trait can be difcult to interpret, and must be assessed together with other information. It is important to have information about the allele frequency in the population tested, and also to be able to quantify in a meaningful way the magnitude of effect a

disease-associated allele has on the disease processes. For this reason some measure of the specicity and sensitivity of a disease-associated allele to predicting disease is desirable. Currently, considerable attention is being focused on the clinical validity and clinical utility of genetic polymorphisms that have been reported to be associated with a disease.

Polymorphism vs. mutation

A major difference in the genetic basis for simple Mendelian diseases vs. complex genetic diseases is the number of genes involved and the contribution of each gene to the overall disease phenotype. In Mendelian disease, alteration of a single gene locus can result in disruption of a protein product that has a major physiological impact and therefore may be considered to be deterministic of the disease. The fact that the genetic alteration is predictably associated with a disease phenotype indicates that there is no redundancy or compensation in the particular biological system that can overcome the effect of the underlying genetic defect. In contrast, the genetic alterations that contribute to complex diseases are individually responsible for much more subtle perturbations of protein function. Consequently, it is difcult to establish causal links with genetic polymorphisms that are associated with complex diseases. There is not a one-to-one correlation of the presence of a specic genetic allele and the occurrence of a complex disease, but rather specic alleles are reported to be found more frequently in diseased individuals than in nonaffected controls. Often initial support for the link is the statistical association of an allele with a disease state. It is also important to understand that disease alleles reported to be associated with a complex disease are also found in unaffected individuals. Because multiple (perhaps 58) different gene alleles can contribute to a complex disease state, an individual with a complex disease need not have all of the alleles reported to be associated with that disease. Thus, the presence of a disease-associated allele in an individual is not diagnostic for a disease. Environmental factors are also critical to the etiology of most common diseases. Most chronic diseases are of adult onset, and therefore take many years to develop. Genes involved in the innate, inammatory, and immune responses are often involved. These physiological and biological pathways have many compensatory and redundant aspects, and therefore it is extremely difcult to quantify the effect of any one genetic variant to the disease state. Consequently, if an individual is found

93

Kinane et al.

to have a disease-associated genetic polymorphism, it is often not clear what the clinical signicance of this is.

been many clinical reports suggesting a familial aggregation of periodontitis, but until recently the research tools to pursue these reports were lacking (16, 66, 73).

Single nucleotide polymorphisms

The most common type of genetic polymorphism is a Single Nucleotide Polymorphism (SNP). There are an estimated 10 million SNPs in the human genome. SNPs occur throughout the genome, in regions encoding genes as well as in regions without identied open reading frames. SNPs in genes may occur in protein coding regions (exons) and noncoding regions (introns and regulatory regions). Many SNPs that occur in genes have no effect on the encoded protein, but a large number of SNPs do have an effect on the gene product. Whether this change contributes to a disease phenotype depends on the specic consequence of the genetic variant on the function of the gene and its protein product. An SNP is further dened as occurring in at least 1% of the population, and indeed several disease-related polymorphisms occur at much higher frequencies (2050%).

Twin studies
Through the phenomenon of twins, in particular monozygous twins, who arise from one fertilized egg, nature has provided a wonderful tool for the examination of genetic inuences in disease and for partitioning the relative contribution of genes and environment to a trait. Monozygous twins are genetically identical. Dizygous twins are only as genetically similar as brothers and sisters would be, on average they share  50% of their genes in common (dizygous twins are from two different eggs and two different sperm). Discordance or differences in the presence of disease between monozygous twins must be due to environmental factors. Disease discordance between dizygous twins could arise from both environmental and genetic differences. The difference in concordance between monozygous and dizygous twins for a particular phenotype can be used to estimate the relative contribution of genes (heredity) and environmental factors to a disease and studying disease presentation in twins is often a valuable rst step in this process.

Methods of genetic analyses

Clinical and scientic data from a variety of sources suggest that genetic variants are major determinants of syndromic and nonsyndromic periodontitis. To evaluate the quality of supporting studies requires an understanding of the formal genetic analytical methods that have been used. Geneticists use a variety of techniques to demonstrate the genetic basis of disease. Some methods are general, whereas others permit precise identication of genetic variants that cause or contribute to disease. The methods summarized below have been important in the evaluation of genetic diseases including periodontitis.

Segregation analysis
Genes are passed from parents to children in a predictable manner, and usually segregate in families as predicted by Mendels laws (127). Geneticists can study the pattern of disease transmission in families using a method called segregation analysis. Segregation analysis evaluates the relative support for different transmission models to determine which model can account for the observed segregation of a trait through families. By sequentially comparing models to each other, segregation analysis identies the model that best accounts for the observed transmission of a trait in a given population. Geneticists generally apply segregation analyses to determine whether a trait transmission appears to t a Mendelian or another mode of genetic transmission. When comparing genetic models of transmission, genetic characteristics including mode of transmission (e.g. autosomal, X-linked, dominant, recessive, complex, multilocus, or random environmental), penetrance, phenocopy rates and frequencies for disease and nondisease alleles are some of the characteristics included in the different models evaluated.

Familial aggregation
Familial aggregation of a trait or disease can suggest genetic etiology. However, families also share many aspects of a common environment, including diet and nutrition, exposures to pollutants, and behaviors such as smoking (active and passive). Certain infectious agents may cluster in families. Thus, familial aggregation may result from shared genes, shared environmental exposures and similar socioeconomic inuences. To determine the evidence for genetic factors in familial aggregation of a trait, more formal genetic studies are required. There have

94

Periodontal disease genetics

But segregation analysis does not necessarily provide the true model. Since it is a comparison of two models, segregation analyses are only as good as the models tested. If important assumptions of the model tested are incorrect, this will limit the results. This limitation of segregation analysis must be realized, as it has resulted in inaccurate conclusions for the transmission of at least one form of early onset periodontitis (110). Segregation analysis tests alternative models to try to develop the best characterization of transmission characteristics within a set of data. As such, it is most appropriately applied to data sets of many families to determine the best tting model. Segregation analysis does not nd or aim to nd a specic gene responsible for a trait.

genetic traits have not been as successful for a variety of reasons (52, 171). A limiting factor in the traditional application of linkage to complex diseases is that complex diseases are due the combined effect of multiple genes of minor effect. When multiple genes each contribute a small amount to the disease phenotype, traditional parametric linkage studies are much less powerful. Fortunately, newer adaptations of the linkage approach and the availability of Association testing approaches offer a practical alternative (106, 113, 186).

Association studies
Genes contributing to common, complex diseases such as periodontitis have proven more difcult to isolate. When multiple, perhaps many, genes act with environmental factors to contribute to disease liability, it is difcult to formulate disease models. In the absence of specic genetic models, the etiology of complex diseases is often conceptualized as due to multiple factors, i.e. several genetic loci interacting with each other to produce an underlying susceptibility, which in turn interacts with additional environmental factors to produce an actual disease state. For complex traits, such as bipolar disorder (11), diabetes (61), obesity (21), and oralfacial clefting (18, 128), traditional parametric linkage analysis has produced either negative results or a plethora of weak, positive results not easily replicated. Theoretical research suggests several reasons for the ambiguity of the linkage results in these cases. First, if a disease gene is neither necessary nor sufcient to cause a disease, but rather is a modier gene that elevates a nonzero baseline risk, conventional parametric linkage analysis may not detect the gene (57). Second, if the relative contribution of a gene to a disease phenotype is small, i.e. the disease susceptibility allele raises the risk by a factor of < 2, linkage analysis using affected sibling pairs will not be powerful enough to detect the gene, given realistic sample sizes (143). Thus, linkage analyses may not be a useful strategy to detect modier genes or genes that exert small effects precisely those genes which might be operating in chronic periodontitis and many other complex disorders. Consequently, attention has shifted away from model-dependent parametric linkage analysis to model-free, nonparametric association analysis as an alternative means of locating disease susceptibility genes, especially since association studies can sometimes detect weaker effects than can linkage analysis (77).

Linkage analysis
Linkage analysis is a technique used to localize the gene for a trait to a specic chromosomal location. Genetic linkage studies are based on the fact that alleles at syntenic gene loci in close proximity on the same chromosome tend to be passed together from generation to generation (i.e. segregate), as a unit. Such genes are said to be linked, and violate Mendels law of independent assortment. Geneticists can apply quantitative analyses to detect this lack of independent assortment of genetic loci, and can use it to map (localize) genes to specic chromosome locations. Over the past 15 years, genetic maps have been developed that show the position of millions of polymorphic genetic loci spanning the human genome (58) (http://www.ncbi.nlm.nih.gov/genome/ guide/human/). Scientists can follow a specic trait as it segregates through families of interest and determine whether the trait appears to segregate with a known genetic polymorphism that has been localized to a specic chromosomal location. In this manner, scientists can test whether a trait appears to segregate in a manner consistent with linkage to a known genetic marker. Because the precise chromosomal location of the genetic marker is known, when linkage is detected, the gene responsible for the trait can be placed in the vicinity of the linked genetic polymorphism. Linkage can therefore prove the genetic basis of disease. Linkage is often used as a rst step to determine the approximate location of a gene of interest, permitting subsequent studies to identify the mutation responsible for a disease trait. Linkage studies have been particularly effective in identifying the genetic basis of simple Mendelian traits (OMIM 2004), where mutation of a single gene can cause a disease. Linkage studies of complex

95

Kinane et al.

Two types of association analysis are commonly employed in genetic studies: population-based and family-based (76). The population-based approach utilizes a standard case-control design, in which marker allele frequencies are compared between cases (affected individuals) and controls (either unaffected individuals or individuals randomly chosen from the population). When a positive association is found, several interpretations are possible: the associated allele itself is the disease-predisposing allele; the associated allele is in linkage disequilibrium with the actual disease-predisposing locus; the association is due to population stratication; the association is a sampling, or statistical, artifact. The rst two interpretations represent the alternative hypotheses of interest in a gene mapping context. In the rst case, the marker itself is the disease-susceptibility locus. This outcome is the rationale behind candidate gene studies, in which alleles of the genes being tested have some a priori expectation of being directly involved in the disease process. Evidence of a positive association can be followed up by investigations to establish a functional role. In the second case, the associated allelic polymorphism itself does not play a functional role in causing disease, but rather the polymorphism is in close physical proximity to the gene that does contribute to susceptibility. A classic example is the human leukocyte antigen (HLA) system, in which various HLA haplotypes are associated with a number of diseases, including insulin dependent diabetes mellitus, rheumatoid arthritis, and ankylosing spondylitis (169). There is currently considerable attention being directed towards the clinical use of disease-associated genetic polymorphisms for genetic testing. However, most initial reports of these polymorphisms have not been replicated (75), reinforcing the need to develop acceptable criteria to determine the clinical validity of such reports (52). Fortunately, new approaches hold promise to identify signicant disease associations that may be important for understanding susceptibility for complex diseases. There is currently great interest in comprehensively characterizing human SNPs to facilitate evaluation of their role in common diseases.

to determine the genetic location of each SNP and characterize how these genetic variants are distributed in several different population groups. The project is at present studying DNA samples representative of African, Asian, and European ancestry. By identifying most of the 10 million SNPs that occur in humans, the HapMap project will identify the DNA variants responsible for most of the genetic diversity in humans. The project is not intended to identify specic disease-associated polymorphisms. Instead, it is hoped that by providing such a catalog of common genetic variants, clinicians and scientists can work together to identify SNPs with important disease associations to understand disease etiologies and develop new diagnostic and treatment strategies.

Evidence for the role of genetic variants in periodontitis

The risk for many diseases, including periodontal diseases, is not borne equally by all individuals in a population (88, 89). A variety of microbial, environmental, behavioral, and systemic disease factors are reported to inuence the risk for periodontitis (132). An individuals genetic makeup is a crucial factor inuencing their systemic or host response-related risk. There are reports in the literature on familial aggregation of periodontal diseases, but it is difcult to compare them. Although periodontal disease terminology has changed many times over the years, most familial reports of early onset forms of periodontitis are now referred to as aggressive periodontitis (8, 9, 14, 15, 17, 23, 46, 47, 90, 99, 110, 112, 121, 129, 147, 158, 161, 173). Reports of the familial nature of chronic forms of periodontitis are less frequent but the aggregation within families is consistent with a genetic predisposition. We should remember, however, that familial aggregation of periodontal disease may also reect exposure to common environmental factors. Shared environmental factors include education, socioeconomic grouping, oral hygiene, shared transmission of bacteria, diseases such as diabetes and environmental features such as passive smoking, sanitation, etc. Some aspects of behavior are determined in part by genetics, which may inuence potential modifying factors such as education, lifestyle and, of dental importance, oral hygiene. Complex interactions between genes and environment are difcult to quantify, but are likely to be important when considering the familial risk for the periodontal diseases.

International HapMap project

The international HapMap project is being conducted to identify and catalog the common genetic variants that occur in human beings (http:// www.hapmap.org). The project will describe each of the common SNPs in the human genome. The goal is

96

Periodontal disease genetics

In chronic periodontitis, clinical disease characteristics do not present until the third decade of life, whereas in the aggressive forms of periodontal disease, the presentation can occur much earlier, in the early teens or younger (38). This variability in presentation makes diagnosis difcult, not only in declaring disease but also in detecting patients who are free of the disease and in differentiating between chronic and aggressive forms of periodontitis. The problems associated with the clinical diagnosis of periodontal disease are not uncommon in medical genetics as similar problems arise when studying other delayed onset genetics (16, 139). The problems of genetic model testing in aggressive periodontitis have been highlighted by Boughman et al. (16) who noted that aggressive periodontitis has a variable age of onset and is often not recognized until after puberty. Diagnosis of aggressive periodontitis in older individuals is problematic due to the difculties of distinguishing between aggressive periodontitis and chronic adult onset forms of periodontitis on the basis of the clinical signs. Similarly, diagnosis in older edentulous individuals is also problematic. The effects of the environment, for example plaque accumulation and smoking, also have major long-term inuences on disease experience (Fig. 1) and these confound the diagnosis of aggressive periodontitis. Diagnostic quandaries create signicant problems for genetic studies of periodontal disease, limiting attempts to correlate cellular, functional, and immune response variables with early onset periodontitis phenotypes in families. While there is evidence for a gene of major effect in aggressive periodontitis, early onset forms of periodontitis appear to be etiologically complex and heterogeneous (6, 15, 139, 140, 158). Although bacterial transmission between subjects has been suggested to explain aggressive periodontitis clustering within families, this observation alone is insufcient to account for familial clustering (15). While the heterogeneity paradigm discussed by Potter (139) is borne out in subsequent familial studies of what is now classied as aggressive periodontitis, the striking familial aggregation of the trait is consistent with a signicant genetic etiology. Characterization of the specic genetic components in the etiology of this disease requires more formal genetic analyses.

Twin studies
Twin studies have been invaluable in studying the genetic basis of simple and complex traits. Large,

worldwide registers of data on twins and their relatives have been established (13). Such twin studies registers offer unique opportunities for selected sampling of quantitative trait loci linkage and association studies. Twin studies of periodontitis, however, have generally been limited in scope and in subject numbers. However, studies of concordance for periodontitis and for clinical indices related to periodontal health and disease generally support a signicant heritable component for periodontitis. Most twin studies have studied the more prevalent form of periodontitis, which is chronic periodontitis, as well as chronic gingivitis. Corey et al. (26) studied self-reported periodontal health in 4908 twin pairs and found that 9% of subjects, consisting of 116 identical and 233 nonidentical twin pairs, reported a history of periodontitis. The concordance rate, or level of similarity in disease experience, ranged from 0.23 to 0.38 for monozygous twins, and was much lower (0.080.16) for dizygous twins. These ndings suggest that heritable factors are important in the reported periodontitis experience. Unfortunately, environmental factors such as smoking status were not factored into this analysis and could introduce a bias towards nding a correlation between twins. Michalowicz et al. (124) studied dizygous twins reared apart (dizygous-A) and reared together (dizygous-T) and monozygous twins reared apart (monozygous-A) and reared together (monozygous-T). The mean probing depth and clinical attachment level scores were found to vary less for monozygousT than for dizygous-T twin pairs, further supporting the role of genetics in this disease. Michalowicz et al. (123) investigated alveolar bone height and showed signicant variations related to genotype. The twin groups had similar smoking histories and oral hygiene practices. It was concluded that genetics plays a role in susceptibility to periodontal disease. In a subsequent study of 117 adult twin pairs, Michalowicz and coworkers estimated genetic and environmental variances and heritability for gingivitis and chronic periodontitis (125). Genetic and environmental variances and heritability were estimated using models with maximum likelihood estimation techniques. Monozygous twins were found to be more similar than dizygous twins for all clinical measures. Statistically signicant genetic variance was found for both the severity and the extent of disease. Chronic periodontitis was estimated to have approximately 50% heritability, and was unaltered after adjusting for behavioral variables, including smoking. Monozygous twins were also more similar

97

Kinane et al.

than dizygous twins for gingivitis scores but there was no evidence of heritability for gingivitis after behavioral covariates such as utilization of dental care and smoking were incorporated. These results conrm previous studies and indicate that approximately half of the variance for chronic periodontitis is attributable to genetic variance. The basis for the heritability of periodontitis appears to be biological and not behavioral (125).

Segregation analysis
While familial aggregation suggests a heritable component in the etiology of aggressive periodontitis, and twin studies support a genetic component in chronic periodontitis, neither is appropriate to identify the genetic model or specic gene loci involved in these periodontal diseases. Segregation analyses can evaluate the relative support for different models to identify the one that most closely represents the clinical data. Segregation analysis is very dependent upon the assumptions of the analyses and there have been few rigorous segregation analyses of aggressive periodontitis. Many are merely studies of one or more families and are statistically underpowered. Genetic segregation analysis needs accurate clinical identication of affected individuals and familial relationships as well as genetic assumptions of the analysis. If inaccurate assumptions or data are used, the outcomes will reect this. Early studies of aggressive forms of periodontitis were hampered by diagnostic classication issues and by an overrepresentation of affected females (67, 147), falsely supporting X-linked transmission (68). Linkage reports of aggressive periodontitis in an extended kindred from Maryland supported an autosomal dominant transmission (14). A segregation analysis of North American families was performed by Marazita and coworkers, who studied more than 100 families segregating aggressive forms of periodontitis; their results supported an autosomal dominant transmission (112). They concluded that autosomal dominant inheritance with approximately 70% penetrance occurred for both Blacks and nonBlacks. The currently held theory on the genetics of aggressive periodontitis is that prepubertal periodontitis, localized aggressive periodontitis, and generalized aggressive periodontitis are probably due to a major gene locus transmitted in an autosomal manner with reduced penetrance; there is evidence for both autosomal recessive (147) and autosomal dominant forms (14, 112). It is likely that

these aggressive forms of periodontitis are genetically heterogeneous, meaning that while the mutated gene responsible for the condition is likely to be the same in any given family, there are probable several different genetic loci that, if mutated, can cause aggressive periodontitis. The expression reduced penetrance means that some subjects with the genotype may not actually express the phenotype, i.e. the clinical manifestations of aggressive periodontitis, whereas others may express it fully. Environmental factors (such as smoking and plaque control) as well as epigenetic interactions of multiple susceptibility genes may play a large role in whether the phenotype is expressed clinically. Clinical evidence as well as laboratory studies of periodontitis patients suggest genetic heterogeneity (6, 69). Consequently, the suggestion that aggressive periodontitis may be transmitted in different ways in different families is not a surprise. Some reviews have reported this as problematic, but this is not necessarily the case. There is a common precedent in genetics for a heritable pathologic condition to show different modes of inheritance in different families. Often, once the genetic basis of the condition is understood, it is found that mutations in different genes can cause a clinically similar group of conditions, as occurs in the EhlersDanlos syndromes, a heterogeneous group of heritable connective tissue disorders. These conditions involve aberrations of collagen, and several are known to have periodontal disease manifestation. More than 15 forms of EhlersDanlos syndromes are known, and these show different inheritance patterns including autosomal dominant, recessive, and X-linked forms. These ndings reect that different genetic loci are capable of causing the disease in dominant and recessive ways. Some of the genes responsible are autosomal and others are X-linked, accounting for the different observed modes of transmission. For aggressive forms of periodontitis, the preponderance of the evidence supports autosomal dominant transmission in North America and autosomal recessive transmission in certain European populations (112, 146, 147). These different modes of transmission may reect genetic heterogeneity such as that seen with EhlersDanlos syndromes.

Linkage studies in aggressive periodontitis

Three linkage studies have been performed to date on families with localized aggressive periodontitis.

98

Periodontal disease genetics

Boughman et al. (14) identied an autosomal dominant form of localized aggressive periodontitis in an extended family from Southern Maryland. In this family, type III dentinogenesis imperfecta and a localized form of aggressive periodontitis were segregating as dominant traits. Since the gene for dentinogenesis imperfecta-III had been previously localized to chromosome 4, they performed a linkage analysis on this chromosome and demonstrated a relatively close linkage with the suspected locus for aggressive periodontitis (14). This was an important study because it supported autosomal dominant inheritance of a single major gene locus, clearly indicating a major genetic component to the aggressive periodontitis disease etiology. Hart et al. (69) evaluated support for linkage to this region of chromosome 4 in a different population of families (14 African American and 4 Caucasian). Their ndings supported genetic locus heterogeneity of aggressive periodontitis, as they excluded a chromosome 4 major gene locus for aggressive periodontitis in the families they studied. Thus, this Brandywine population appears to have a different form of periodontal disease with a different gene being responsible compared with the AfricanAmerican and Caucasian families studied by Hart and coworkers. These ndings support genetic heterogeneity, with at least one gene locus responsible for aggressive periodontitis located on chromosome 4. Recently, Li and coworkers (107) reported evidence of a gene responsible for localized aggressive periodontitis located on chromosome 1q25. To date, a gene of major effect for aggressive periodontitis has not been identied.

Because altered proteins function in different structural and immune pathways, genetic modulation of a variety of different genes can affect a variety of different physiological and cellular pathways (63). These conditions illustrate that the genetic contribution to periodontitis susceptibility is multifaceted, and may potentially involve many different gene loci. However, in contrast to nonsyndromic forms of periodontitis, these conditions have periodontal disease manifestations as part of a collection of syndromic manifestations. In most cases of aggressive periodontitis, individuals present with clinical manifestations of periodontitis, but do not appear to have any other clinical disease manifestations. This is not inconsistent with a genetic disease etiology. Expression of genes can vary in different tissues, and mutations of a ubiquitously expressed gene can result in a tissue-specic condition. Recently, mutation of the SOS1 gene has been identied in individuals with hereditary gingival bromatosis (72). SOS1 is important in determining whether cells grow, divide or differentiate and is ubiquitously expressed. However, the only clinical manifestation of this gene defect appears to be enlargement of the periodontium, an instance of the tissue-specic nature of many diseases. A similar tissue-specic manifestation of a gene defect may occur in nonsyndromic aggressive periodontitis. Neutrophil functional disorders Molecular biology has highlighted the important role of several receptors on the polymorphonuclear leukocyte surface in adhesion, and emphasized that defects in the number of these receptors may lead to increased susceptibility to infectious disease (2). Adhesion is crucial to the proper function of the polymorphonuclear leukocyte because it affects phagocytosis and chemotaxis which, if decient, might predispose to severe periodontal destruction. Page et al. (131) proposed that the generalized form of prepubertal periodontitis (this disease has generalized and localized forms) is a localized oral manifestation of the leukocyte adhesion deciency syndromes. More recent studies have indicated that generalized and localized prepubertal periodontitis occur in otherwise healthy children (17, 151). Leukocyte adhesion deciency occurs in two forms, leukocyte adhesion deciency syndrome type 1 and leukocyte adhesion deciency syndrome type 2, both of which are autosomal recessive traits. Circulating leukocytes have reduced or defective surface receptors and do not adhere to vascular

Syndromic forms of periodontitis

Signicant, irrefutable clinical and laboratory data indicate that genetic variants do predispose to disease states in humans. Severe periodontitis presents as part of the clinical manifestations of a number of monogenic syndromes and the gene mutation and biochemical defect is known for many of these conditions. A commonality of these conditions is that they are inherited as simple Mendelian traits due to genetic alterations of a single gene locus. Examples of these monogenic conditions are given in Table 1. The signicance of these conditions is that they clearly demonstrate that a genetic mutation at a single locus can impart susceptibility to periodontitis. Additionally, these conditions illustrate that this genetic susceptibility may segregate by different transmission patterns.

99

Kinane et al.

endothelial cells; thus they do not accumulate in sites of inammation where they are needed. Reports of leukocyte adhesion deciency syndromes indicate that although the blood vessels are full of neutrophils, the disease sites lack sufcient leukocytes to combat the microbial challenge and thus infections ensue rapidly in these patients. Affected homozygotes suffer from acute recurrent infections that are commonly fatal in infancy. Those surviving will develop severe periodontitis, which will begin as the primary dentition erupts (174). Other disorders of neutrophil function are associated with severe forms of periodontal destruction. The ChediakHigashi syndrome is a rare disease transmitted as an autosomal recessive trait. Those affected are very susceptible to bacterial infections due to alterations in the functional capacity of the polymorphonuclear leukocyte. Humans (59) and other animals (105) with ChediakHigashi syndrome exhibit generalized, severe gingivitis and extensive loss of alveolar bone and premature loss of teeth (168). The polymorphonuclear leukocyte chemotactic and bactericidal functions are thought to be abnormal in these patients. These diseases related to anomalies of leukocyte and polymorphonuclear leukocyte function are excellent examples of how monogenic defects can cause periodontitis through a clearly attributable mechanism. The host response is made up of a vast number of processes, all of which are under genetic control and all of which are feasible candidates for irregularities that may result in variability in the clinical presentation of periodontal disease. Deciency in neutrophil numbers (neutropenias) A further neutrophil deciency is found in infantile genetic agranulocytosis, a rare autosomal recessive disease where polymorphonuclear leukocyte numbers are very low and which has been associated with aggressive periodontitis (144). Cohens syndrome is another autosomal recessive syndrome and is characterized by mental retardation, obesity, dysmorphia, and neutropenia. Individuals with Cohens syndrome show more frequent and extensive alveolar bone loss than do age-, sex-, and mental ability-matched controls (1). Not all neutropenias result in periodontal disease. Familial benign chronic neutropenia has variable expressivity and although several individuals within a family may be neutropenic, not all are affected by recurrent infections or periodontal disease (34). These ndings might be explained by the variable

genetic expression of the disorder or by the variable effects of the environment (such as plaque or smoking) on these patients. Genetic defects of structural components ` PapillonLefevre Syndrome is a condition in which the cardinal clinical features are severe periodontitis and great variation in the severity and extent of palmar plantar hyperkeratosis (56, 60, 70). Genetic ` linkage studies narrowed the PapillonLefevre gene locus to chromosome 11 and subsequent mutational analyses permitted identication of mutations in the ` cathepsin C gene in patients with PapillonLefevre syndrome (64, 170). Subsequent studies have identied more than 40 different cathepsin C mutations in individuals from many different ethnic groups (65). This is an excellent example of the success of genetic studies in contributing to the identication of a gene defect of periodontal importance. Genetic linkage studies permitted localization of the gene defect to a specic chromosome, permitting focused mutational analyses on genes within an area of the chromosome. These focused analyses uncovered gene mutations in the cathepsin C gene. Mutations of this gene are associated with the loss of protease activity of the cathepsin C protein. Additional work has demon` strated that PapillonLefevre syndrome and Haim Munk syndrome (a slightly different clinical variant ` within the PapillonLefevre syndrome group of disorders) are allelic variants of cathepsin C gene mutations, as predicted by Gorlin et al. (56, 65, 182, 183). EhlersDanlos syndrome refers to a collection of connective tissue disorders characterized by defective collagen synthesis. EhlersDanlos types IV and VIII are related to an increased susceptibility to periodontitis (71, 108) and are inherited in an autosomal dominant manner. Clinical characteristics of type VIII EhlersDanlos syndrome include fragility of the oral mucosa and blood vessels, and a severe form of aggressive periodontitis (3). Other genetic conditions related to defects in structural components that maintain a healthy periodontium include the very rare WearyKindler syndrome and hypophosphatasia. Aggressive periodontitis has been reported in WearyKindler syndrome where abnormalities of the epidermal keratinocytes occur (176). Patients with hypophosphatasia have a decreased serum alkaline phosphatase and the presence of phosphoethanolamine in the urine (48). In these patients, there is severe loss of alveolar bone and premature loss of the primary teeth (12, 19),

100

Periodontal disease genetics

particularly anteriorly (7). There is histologic evidence of enlarged pulp chambers and a disturbance in cementogenesis, the cementum being either absent or hypoplastic. Gingival broblasts are also decient in alkaline phosphatase. The lack of connective tissue attachment between the tooth and bone accounts for the early spontaneous exfoliation of the primary teeth. Baab et al. (7) described a family where all three children manifested premature exfoliation of the primary teeth similar to that seen in prepubertal periodontitis (133). These children were assigned a diagnosis of hypophosphatasia on the basis of alkaline phosphatase and phosphoethanolamine levels. Baab et al. (7) noted that their data suggest an autosomal dominant mode of transmission and suggested that hypophosphatasia might be considered in the etiology of some forms of aggressive periodontitis.

the appropriateness of incorporating these studies into rigorous analyses such as the meta-analysis techniques used in Cochrane-style systematic reviews.

Gene polymorphisms of host response elements and periodontitis

The genetic basis of many aspects of the periodontal host response has been discussed in reference to genetic disorders predisposing to periodontal disease. The aim of this section is to generally summarize the potential inuence of innate, inammatory, and immunological genetic variations and to consider where the most promising candidates lie from the viewpoint of a genetic diagnostic approach to periodontitis (Tables 25). There is a genetic basis to many aspects of the periodontal host response. Mutations of genes have been identied in Mendelian inherited syndromes that have signicant periodontitis ndings (Table 1), clearly indicating that single gene defects can predispose to periodontitis. Data from human and animal studies indicate that genetic variance can inuence the innate, inammatory, and immunological response to microbial infection. The availability of the annotated human genome and technological advances in genotyping have made it possible to genotype allelic variants and to test for association in casecontrol studies. This has facilitated studies to evaluate the support for against the association of an array of genetic polymorphisms with periodontitis periodontitis (Tables 25). It is important to realize that most of these studies are underpowered to draw denitive conclusions. Several features of the hosts innate immune system which may contribute to genetic susceptibility to aggressive periodontitis have been outlined already and include epithelial, connective tissue, and broblast defects. Functional defects or a decient number of polymorphonuclear leukocytes, as discussed previously, have profound effects on the hosts susceptibility to periodontitis. Another aspect of the host inammatory response, proinammatory cytokines, has attracted much attention as potentially crucial variants inuencing the host response in periodontitis. This review will not attempt to summarize the plethora of SNP studies in periodontal disease but will summarize tabularly (Tables 25) the areas of current research and focus on the interleukin-1 polymorphism by way of example and illustration of the SNP periodontal literature.

Polymorphism studies in periodontitis

In complex diseases, genetic variants at multiple gene loci contribute to overall disease susceptibility. As such, a simple cause and effect relationship between a particular genetic allele and a disease is not possible. Support for an etiologic role for genes (Table 1), complex genetic diseases which are not diachronic periodontitis, is based on the statistically signicant association of specic genetic alleles in individuals with disease (cases) compared to individuals without disease (controls). This mathematical association is not necessarily biological or physiological. Studies reporting such associations vary in design and rigor. Association studies ideally should evaluate large numbers in population-based studies and have the power to detect a signicant association. The issues of allele frequency in the population studied, case control design, and population stratication are very important but are unfortunately often omitted from dental studies. This is not acceptable, and studies performed without sufcient rigor need to be interpreted in this light (52, 75, 109). The overstating of results has become commonplace such that rigorous, scientically principled approaches are needed to guard against unfounded and erroneous conclusions. Tables 25 report a multitude of studies evaluating associations of genetic polymorphisms with periodontitis. The variation in design and power of these studies raises questions regarding

101

Kinane et al.

Table 2. Cytokine polymorphism studies in periodontal disease

Gene Interleukin (IL) IL-2 ()330) IL-4 (PP and IP) IL-4 ()590) IL-4 ()590) IL-4 ()590) IL-4 (PP and IP) IL-6 ()572) IL-6 ()174) IL-10 (ATA haploytpe) IL-10 ()1087) IL-10 ()597, )824) IL-10 (gene promoter region) IL-10 (R and G microsatellites) Tumor necrosis factor-a (TNF-a) TNF-a ()308) TNF-a ()308) TNF-a ()308) and Lymphotoxin (+252) TNF-a () 1031, )863, )857) TNF-a ()376, )308, )238, +489) TNF-a ()1031, )863, )857, )308, )238) TNF-a ()308) TNF-a (microsatellite polymorphisms) TNF-a ()238, )308, +252) TNF receptor 2 (+587) ChP ChP ChP ChP ChP AgP AgP AgP ChP ChP No assoc. No assoc. Assoc. Assoc. No assoc. No assoc. No assoc. No assoc. No assoc. Assoc. Folwaczny et al. (45) Fassmann et al. (41) Fassmann et al. (41) Soga et al. (157) Craandijk et al. (27) Endo et al. (39) Shapira et al. (152) Kinane et al. (94) Galbraith et al. (50) Shimada et al. (153) ChP AgP ChP ChP ChP AgP ChP ChP ChP ChP AgP, ChP ChP AgP Assoc. No assoc. No assoc. No assoc. No assoc. Assoc. Assoc. (resist.) Assoc. Assoc. Assoc. No assoc. No assoc. No assoc. Scarel-Caminaga et al. (149) Gonzales et al. (53) Pontes et al. (138) Kang et al. (91) Scarel-Caminaga et al. (150) Michel et al. (126) Holla et al. (85) Trevilatto et al. (172) Scarel-Carminga et al. (148) Berglundh et al. (10) Gonzales et al. (54) Yamazaki et al. (178) Kinane et al. (94) Disease Association Authors (Ref.)

ChP: chronic periodontitis. AgP: aggressive periodontitis. Assoc.: related to susceptibility to periodontitis. No assoc.: not related to susceptibility to periodontitis. Assoc. (resist.): associated with resistance to periodontitis.

Interleukin-1 gene polymorphisms in periodontal disease

The interleukin-1 gene polymorphisms associated with periodontitis are a useful example for considering the strengths and limitations of using gene polymorphisms in disease association studies in the periodontal diseases. In 1997, Kornman et al. (100) found an association between polymorphisms in the genes encoding for interleukin-1a () 889) and interleukin-1b (+ 3953) (termed the composite genotype) and an increased severity of periodontitis. This initial study spawned numerous publications and has been

highly inuential in creating interest in gene polymorphisms and periodontal disease. The specic genotype of the polymorphic interleukin-1 gene cluster (periodontitis susceptibility trait, PST or composite genotype) was only associated with severity of periodontitis in nonsmokers, and distinguished individuals with severe periodontitis from those with mild disease (odds ratio 18.9 for ages 4060 years, but wide condence intervals of 1.04343.05). Functionally, the specic periodontitisassociated interleukin-1 genotype constitutes a variant in the interleukin-1b gene that is associated with high levels of interleukin-1 production (136, 137).

102

Periodontal disease genetics

Table 3. Receptor polymorphism studies in periodontal disease

Gene TLR-2 (Arg753Gln, Arg677Trp) TLR-4 (Asp299Gly, Thr399Ile) FccRIIa (HH) FccRIIa (RR) FccRIIa(131R) Gm (23) allotypes FccRIIa (HH), FccRIIIa(158 V) FccRIIb (695) FccRIIb(646184, intron 4) FccRIIIb (NA2), FccRIIIa(158 V) FccRIIIa (VV genotype) Disease ChP ChP ChP smokers ChP AgP, ChP ChP AgP, ChP AgP Chp ChP ChP (Severity of bone loss) ChP AgP ChP ChP AgP, Chp ChP AgP ChP ChP ChP ChP AgP AgP, ChP AgP AgP AgP Association No assoc. No assoc. Assoc. Assoc. No assoc. Assoc. Assoc. Assoc. Assoc. Assoc. Assoc. Authors (Ref.) Folwaczny et al. (43) Folwaczny et al. (43) Yamamoto et al. (177) Tang et al. (167) Chung et al. (22) Chung et al. (22) Loos et al. (111) Yasuda et al. (180) Yasuda et al. (180) Kobayashi et al. (97) Meisel et al. (118)

FccRIIIb (NA1) FccRIIIb (NA2), FccRIIIa (158F) FccRIIa, FccRIIIb (Haplotypes), Gm (23) allotype FccRIIIb (NA2) Estrogen receptor-a (PvuII ER) Estrogen receptor-a (XbaI ER) FMLP receptor (329,378) CCR5 (D32) RAGE(1704) VDR (TaqI, BsmI) VDR (TaqI) VDR (TaqI) VDR (TaqI) VDR (Bb) VDR (Bb) and FccRIIIb (NA1NA2) VDR (TaqI)

Assoc. (resist.) Assoc. No assoc. Assoc. No assoc. Assoc. Assoc. No assoc. Assoc. Assoc. Assoc. Assoc. Assoc. No assoc. Assoc. No assoc.

Sugita et al. (159) Kobayashi et al. 2000 (95) Colombo et al. 1998 (25) Kobayashi et al. (96) Zhang et al. (184) Zhang et al. (184) Zhang et al. (185) Folwaczny et al. (42) Holla et al. (84) de Brito et al. (30) Tachi et al. (163) Sun et al. (160) Tachi et al. (164) Yoshihara et al. (181) Yoshihara et al. (181) Hennig et al. (74)

ChP: chronic periodontitis. AgP: aggressive periodontitis. Assoc.: related to susceptibility to periodontitis. No assoc.: not related to susceptibility to periodontitis. Assoc. (resist.): associated with resistance to periodontitis.

Kornman et al. (100) found that 86.0% of the severe periodontitis in patients was accounted for by either smoking or the interleukin-1 genotype. Similar results were reported by McDevitt et al. (115) and from smaller studies by McGuire et al. (117), Laine et al. (101), and Gore et al. (55) (who suggested that the two polymorphisms within the composite genotype may be in linkage disequilibrium). Other

contradictory reports such as Meisel et al. (119) stated that the composite genotype showed a strong interaction with smoking, an established risk factor for periodontitis (93), whereas nonsmokers, even when genotype positive, were not at any increased risk. A similarly contradictory study (134) of 132 periodontitis patients who were age- and sex-matched with controls, did not show any association between

103

Kinane et al.

Table 4. Miscellaneous host response polymorphisms

Gene CD14 ()1359) CD14 ()159) NOD2CARD15(3020insC) NOD2CARD15(3020insC, C2104T) HLA-DQb1 (BamHI) HLA-DRb1(1501) HLA-DQb1(0602) HLA-DQb1(0503, 0602) HLA-DQb (BamHI) MMP-1 ()1607, 1G) MMP-1()1607), MMP-3 ()1171) MMP-1 ()1607, 2G) TGF-b1 ()509) TGF-b1 ()800, )509, L10P, R25P) Fibrinogen (H1H2, H2H2) Plasminogen activator inhibitor-1(4G) NAT2 (two mutated alleles) Disease ChP ChP ChP ChP AgP AgP AgP AgP ChP nonsmokers AgP, ChP ChP nonsmokers ChP ChP ChP ChP ChP smokers Association Assoc. No assoc. No assocn No assoc. No assoc. Assoc. Assoc. Assoc. Assoc. No assoc. Assoc. Assoc. No assoc. Assoc. Assoc. Assoc. Authors (Ref.) Holla et al. (82) Yamazaki et al. (179) Folwaczny et al. (44) Laine et al. (102) Hodge et al. (79) Takashiba et al. (166) Ohyama et al. (130) Takashiba et al. (165) Holla et al. (85) Itagaki et al. (86) de Souza et al. (32) de Souza et al. (33) Holla et al. (83) Sahingur et al. (145) Izakovicova et al. (87) Kocher et al. (98)

TGF: transforming growth factor. ChP: chronic periodontitis. AgP: aggressive periodontitis. Assoc.: related to susceptibility to periodontitis. No assoc.: not related to susceptibility to periodontitis. Assoc. (resist.): associated with resistance to periodontitis.

Table 5. Multiple gene polymorphism assessments

Gene IL-1a ()889), TNF-a ()308), IL-6 ()174) Comprehensive 125 Genes 310 SNPs TNF-b (NcoI bi), ACE (I D), Endothelin-1 (TaqI) Disease ChP ChP and AgP ChP Association Assoc. with serum IL-6 Assoc. noted Assoc. Authors DAuito et al. (29) Suzuki et al. (162) Holla et al. (84)

IL, interleukin. TNF, tumor necrosis factor. ACE, angiotensin converting enzyme. ChP: chronic periodontitis. AgP: aggressive periodontitis. Assoc.: related to susceptibility to periodontitis.

the composite genotype and periodontitis. The prognostic utility of the interleukin-1 genotype on chronic periodontitis progression following nonsurgical therapy was performed by Ehmke et al. (37). Of the 33 patients studied, 16 had the susceptible composite genotype reported by Kornman et al. (100). Following 2 years of periodontal maintenance care, no differences in tooth or attachment loss were detected between those with or without the genotype. Equivocal studies such as those of Cullinan et al. (28) demonstrated an interaction of the interleukin-1 positive genotype with age, smoking, and P. gingivalis, which suggests that interleukin-1 genotype is a

contributory but nonessential risk factor for periodontal disease progression in this population. Cattabriga et al. (20) reported no signicant differences in tooth loss in patients with the interleukin-1 genotype after 10 years in a nonsmoking, well-maintained periodontal population. De Sanctis et al. (31) demonstrated that genotype expression did not affect guided tissue regeneration treatment response at 1 year, but had a great impact on long-term stability (year 4). In a 3-year period, patients with a positive interleukin-1 genotype lost about 50% of the clinical attachment level gained in the rst year and were about 10 times more likely to experience 2 mm

104

Periodontal disease genetics

clinical attachment loss when compared to oral hygiene-matched genotype-negative patients. Indeed, numerous studies, as depicted in Tables 25, show associations and lack of associations across different and similar populations for various forms of chronic and aggressive periodontitis. The polymorphisms in the interleukin-1 gene cluster linked with periodontitis (100) are found in approximately 30% of the European population. However, the prevalences are dramatically lower in Chinese (2.3%) and thus the usefulness of the composite genotype of allele 2 of both interleukin-1a (+ 4845) and interleukin-1b (+ 3954) for determining susceptibility in Chinese patients is dubious (5).

occurred at a higher frequency in aggressive periodontitis smokers (odds ratio 4.9) than in control smokers. These ndings of Parkhill et al. (135), in contrast to those of Hodge et al. (81), found that an interleukin-1b genotype in combination with smoking is associated with aggressive periodontitis. Similar negative ndings for this composite genotype and both chronic and aggressive periodontitis populations from different racial and ethnic backgrounds have been demonstrated and thus the diagnostic utility of the composite genotype may be restricted to specic populations, i.e. the results do not appear to be applicable globally and across ethnic populations, and certainly not for aggressive periodontitis.

Interleukin-1 polymorphism in aggressive periodontitis

Hodge et al. (80) examined interleukin-1a and interleukin-1b genetic polymorphisms in unrelated European white Caucasian patients with generalized early onset periodontitis and found no signicant differences between patients and controls for any of the composite genotypes described by Kornman et al. (100). No signicant differences were found between patients and controls whether smoking was included as a covariate or not. It was concluded that there was a lack of association between the interleukin-1 polymorphisms and aggressive periodontitis, which questions the utility of these candidate genes as markers of susceptibility. This was a relatively homogeneous Scottish population and the results, although negative, merely reect the lack of utility of this composite genotype test in this population. Other studies on the composite genotype reported by Kornman et al. (100) and aggressive periodontitis have had similarly mixed results. For example, the studies by Diehl et al. (35, 36) actually found that allele 1 rather than allele 2 of the interleukin-1b + 3953 exhibited polymorphism. Furthermore, Parkhill et al. (135) investigated the frequency of polymorphisms in the genes encoding interleukin-1b in Caucasians with aggressive periodontitis compared to controls. The frequency of interleukin-1b genotypes homozygous for allele 1 of the interleukin1b + 3953 SNP was found to be signicantly increased in aggressive periodontitis patients (P 0.025). Upon stratication for smoking status, a signicant difference was found in the interleukin-1b genotype distribution between aggressive periodontitis smokers and control smokers (F-exact test, P 0.02), but not between aggressive periodontitis nonsmokers and control nonsmokers. The interleukin-1b 1 1 genotype

Biological plausibility for the composite interleukin-1 polymorphisms

Many investigators have suggested a role for interleukin-1 in the initiation and progression of periodontitis and have quoted in vitro and in vivo studies showing that interleukin-1 activates the degradation of the extracellular matrix and bone of the periodontal tissues. Elevated tissue or gingival uid levels of interleukin-1b have been associated with periodontitis. Kornman et al. (100) quotes abstracts of in vitro studies from Pociot et al. (137) and others (104) which claim that the interleukin-1 polymorphism associated with severe periodontitis in the Kornman study is also known to correlate with a two- to fourfold increase in interleukin-1b production. A problem with this line of reasoning is that interleukin-1 is a proinammatory cytokine intimately involved in all inammatory reactions as well as immune and reparative or healing responses and any perturbation of its level so as not to be homeostatically controlled could have widespread consequences not limited to periodontal disease. In addition, interleukin-1 is one of many proinammatory cytokines (interleukin-6, tumor necrosis factor-a, etc.) with overlapping activities and thus some redundancy exists in the cytokine system. Furthermore, interleukin-1 has many controlling mechanisms which include inhibition of transcription, release controls, and receptor antagonists, and thus is highly regulated such that any polymorphism coding for increased production of this molecule could readily be controlled by the elaborate positive and negative feedback loops associated with its regulation. The claimed association of severe periodontitis with smoking and the interleukin-1 genotype (in

105

Kinane et al.

that in smokers the composite interleukin-1 genotype did not inuence susceptibility) poses further problems. Do smoking and the overproduction of interleukin-1 work along the same pathogenic pathway and thus is the action of both factors not additive but renders the other redundant, or is the overall effect of smoking is so overriding that the composite genotype has little or no effect? These possibilities, while feasible, require much more mechanistic knowledge for both risk factors, but especially for the genotype, given that the association with periodontitis is not as established as that found in the literature on smoking (93). Socransky et al. (155) investigated the association between the composite genotype and carriage of periodontal species. They found the mean counts of specic species were higher in general in interleukin-1 genotype positive than in negative subjects. The species detected at higher levels were those frequently associated with measures of periodontal inammation. A further study aimed at studying the composite genotype and inammation was performed by Lang et al. (103). Genotype-negative subjects had a signicantly lower percentage of bleeding on probing (P 0.0097) and it was concluded that the increased bleeding on probing prevalence and incidence observed in interleukin-1 genotype-positive subjects indicates that some individuals have a genetically determined hyperinammatory response that is expressed in the clinical response of the periodontal tissues. Shirodaria et al. (154) have taken the research further by attempting an assessment of the functional effect of the composite genotype in terms of the quantity of interleukin-1a protein in gingival crevice uid of severe chronic periodontitis patients. These researchers found that allele 2 at position ) 889 of the interleukin-1a gene (one of the alleles linked with susceptibility to periodontitis by Kornman et al. (100)) was associated with a fourfold increase in interleukin-1a as determined by enzyme linked immunoassay. This technique does not demonstrate activity but merely protein presence or absence and would not differentiate protein bound to receptors or inhibitors. Furthermore, it is feasible that inhibitors of proinammatory cytokines may concomitantly be produced to dampen this effect. The authors noted reduced levels of interleukin-1a protein in heavy smokers regardless of genotype but this may be related to the reduced gingival crevice uid noted in smokers (93). This is a useful study given that it addresses the in vivo effects of the polymorphism on interleukin-1 protein quantities,

However, differences in local gingival crevice uid production among patients, sites, smokers, and across gender add considerable variance to such a study, and these factors have to be considered in the interpretation of the data. Engebretson et al. (40) also found elevated levels of interleukin-1b in the gingival crevice uid in shallow sites of patients who were positive for the composite genotype reported by Kornman et al. (100). Smoking was not considered in the study and no statistically signicant differences were noted for deeper pockets. Interestingly, of the 22 chronic periodontitis patients examined, only seven were positive for the susceptible genotype. Mark et al. (114) studied peripheral blood monocytes from composite genotype positive and negative patients to examine whether the interleukin-1b polymorphism was correlated with increased interleukin-1b expression by monocytes in response to periodontal bacterial stimulus. Contrary to previous reports, these workers found no signicant differences in interleukin-1b production in response to any stimulant tested. They went on to report marked interindividual variation in the production of interleukin-1b in both the genotype positive and negative patient groups. Clearly, either the genotype is not important in monocyte production of interleukin-1 or other genetic loci may determine the monocyte interleukin-1 responses.

Summary of the ndings on the interleukin-1 composite genotype in periodontitis

It appears that the interleukin-1 composite genotype has an equivocal ability to detect susceptibility to periodontitis and may at best be limited in its utility to only specic populations. It would appear from the mixed reports on this composite genotype that: it appears irrelevant in aggressive periodontitis; it may be in linkage disequilibrium with the gene contributing susceptibility to chronic periodontitis; the composite polymorphisms may be part of several involved in the genetic risk for chronic periodontitis; the polymorphism is only a useful marker in dened populations (5, 175); conrmation of the functional signicance of this gene polymorphism remains to be conrmed; clinical utilization of the composite polymorphisms for risk assessment and prognostic determination is premature.

106

Periodontal disease genetics

General considerations
Clinical diagnoses
Many confounding factors have been recognized in population association studies of periodontal disease. Only recently have more carefully dened classications of periodontal disease appeared (4). Prior to the American Academy of Periodontology Workshop in 1999, there was much variation in the literature with respect to disease denition and numerous differences in what we dene as a case and indeed what constitutes a control in periodontitis studies. These variable denitions make comparisons across studies virtually impossible. Other complications arise related to the variable presentation over time of periodontal diseases, notably the aggressive forms of periodontitis, which generally appear prior to the age of 35 years of age but which are not conned to this age. In many cases the aggressive forms will appear very early in the teenage years or quite late in the early 30s. Chronic periodontitis is similar in that its presentation and its categorization are quite complex and very much dependent on environmental exposures of the patient over their lifetime. Exposures such as smoking, microbial plaque deposition (oral hygiene) and other factors such as systemic disease modiers (e.g. diabetes) or general factors (e.g. socioeconomic status) have a large inuence on the expression of the phenotype. An additional criticism arises in the suggested utility of these population association studies in that there are often methodologic problems within the reported studies. These include underreporting, where multiple polymorphisms may be tested but few are reported and where, for those few reported, there is a lack of understanding of the importance of multiple range testing, i.e. the reduced statistical inference that can be made in such exploratory studies. On many occasions the results are reduced such that only the positive ones are reported, with obvious statistical implications. The bias against publishing negative results is one that has a tremendous impact on our literature and recognition of the importance of such results is needed. Many do not report the sensitivity and specicity of tests or the many associated environmental aspects which need to be considered. Most studies have insufcient numbers of cases and controls to permit robust pronouncements on the association between polymorphisms and disease. A precise account of the racial and ethnic background of

both cases and controls is necessary as such biases within populations would render the inference questionable.

Environmental variables
Numerous assumptions have been made regarding the populations researched in periodontal disease association studies. Are the controls truly resistant to the disease? Or are they performing high standards of oral hygiene such that environmental factors required to interact with the genotype to express the phenotype are missing? In addition, these subjects may have excellent access to health care and health education and may not smoke, thus avoiding further known risk factors which may be needed to reveal the genotype. In this situation, misplacing the subject(s) in the control group dilutes the chances of seeing the real disease association. Similarly, it is accepted that periodontal destruction is incremental and thus the disease experience and the clinical features increase with age. Thus, the age of the cases and controls have to be considered and matched. Gender and, more importantly in genetic studies, race have to be considered and documented carefully. Thus, the importance of environmental factors in revealing the genotype can not be overstated in the categorization of cases and controls, and knowledge of the racial or genetic heterogeneity is needed. A further difculty is knowing which environmental factors will exert an inuence. Twin studies can be particularly useful here in sorting out genetic and environmental factors (discussed in detail later).

Smoking and susceptibility genes

An important environmental risk factor is smoking. The interplay between smoking and genes in the risk for periodontitis has been rightly discussed in most studies on genetic risk. Interestingly, the attributable risk for periodontitis from smoking may be inuenced by the polymorphism of the N-acetyltransferase (NAT2) gene, which may change the subjects metabolism to produce more damaging smokingderived xenobiotics. Meisel et al. (120) hypothesized that an NAT2 genotype could be a risk factor for periodontal disease. Severe chronic periodontitis patients were predominantly slow acetylators (OR 2.385.02). Thus, the slow acetylator phenotype may be associated with a higher risk for periodontitis. This is an example of a strong gene environment interaction.

107

Kinane et al.

Biologic plausibility
Researchers should always consider the biological plausibility of the effect of the gene polymorphism in the periodontal disease process. There are, however, a great many unknown aspects which provide considerable leeway for investigators in this area. The pathogenesis of periodontal disease is to some extent known. It is a chronic inammatory condition which progresses to the destruction of bone and supporting structures and ultimately the loss of teeth, but the reason that some patients are more susceptible than others is not understood. Any of the innate, inammatory, or immune processes may be involved in the etiology of periodontal disease and, in addition, it is feasible that structural aspects and aspects which may inuence the healing process may be similarly involved. This leaves an extremely wide range of factors, which could be considered fortuitous from a research justication viewpoint but frustrating in the search for the true etiology of the periodontal diseases.

Criticisms of the lack of logic in association studies

A hypothetical example of the logic in play when studying gene polymorphism associations with diseases might be the following. In periodontitis (condition A), bone is lost (abnormality B), and since bone is removed by osteoclasts, which are inuenced by a specic gene (gene C), we then study the association of a known polymorphism (polymorphism D) of gene C with periodontal disease. Obvious problems in this logic are apparent even supercially. Whether the osteoclast has a pivotal role in the disease process is unknown and to assume it does would be purely speculation. Nevertheless, the link between A and B typically results in the search for a gene C and a polymorphism D, which happens to be readily investigatable. Linking gene C with periodontal disease is a task for molecular geneticists, who would employ linkage analysis techniques and positional cloning together with robust statistical approaches including a precise genetic model and knowledge of the population frequency, penetrance of the disease, mutational searches, and laboratory pathogenesis investigations. Gambaro et al. (51) have stressed the crucial importance of demonstrating that allelic variants or polymorphisms are actually functional prior to engaging in population polymorphism association studies. But it is unusual for investigators to investigate a

particular polymorphism after having concluded that it is functionally active. Researchers are more likely to choose polymorphisms of candidate genes (of which there are an extremely wide range in periodontal disease due to our limited understanding of the etiology) on an arbitrary or convenience basis. A highly illustrative example exists in the medical literature on how the I D polymorphism associated with serum angiotensin converting enzyme concentrations was researched. Gambaro et al. (51) recount that in 1989 a polymorphism had been related to serum concentrations of angiotensin converting enzyme, an enzyme possibly involved in atherosclerosis. In 1992, this polymorphism was clearly associated with an atherosclerosis-dependent clinical event. For the next 5 years, hundreds of researchers investigated thousands of patients to look at new associations with various diseases based on a simplistic view of the pathophysiology of the way in which this polymorphism (ID) might inuence the disease process. Many studies were spawned to reconrm other researchers ndings and it was not until 1997 that basic ndings appeared to support the pathogenic and pathophysiological relationship between the polymorphism and the disease. The disease in question turned out to be renal disease, which although not cardiovascular disease, does inuence hypertension and thus cardiovascular disease indirectly. Thus, it was only in 1997 that the rationale fortuitously appeared to support studies which had previously been published. The ID polymorphism story had a happy ending in that the link between hypertension and renal disease was related to the gene polymorphism but the original population association studies were never based on a clear demonstration that angiotensin converting enzyme was linked with atherosclerosis and that the polymorphism (ID) was relevant to the gene in a functional way. This polymorphism was and is extensively studied and is an example of a useful polymorphism emerging from a weak research foundation but most other polymorphisms with similarly weak foundations have not been so fortunate (51) and thus much research effort has been wasted.

Reporting bias
A further very real concern with polymorphism population studies that are not based on rm assumptions is that once they are published, they take on a veracity of their own. This is largely due to a conscious or subconscious reporting bias. Simply put, authors tend not to believe data, even their own,

108

Periodontal disease genetics

which do not agree with already published ndings. The rare exception is when they are established researchers with a track record which commands respect or when they have other hypotheses which they are pursuing (note the publication bias described previously). In addition, to disprove a hypothesis, a much larger study may be needed than the original study that was signicant at the 5% or even the 1% level (approximately four times the n number of the original study may be required).

Interpretation of results
An interpretative problem with the logic sequence that disease A is related to aberration B, which is inuenced by gene C, which has a polymorphism D, is the possible oversimplication by clinicians to gene C aberration B disease A or, worse, polymorphism D aberration B disease A or, innitely worse, that polymorphism D disease A (Fig. 2). A commonly overlooked but consistent mistake is to interpret studies which do not show an association between polymorphism D and disease A as meaning that gene C is not involved in the pathogenesis or is not a risk factor for disease A, when in fact only one of possibly thousands of candidate polymorphisms related to the gene in question may have been ruled out. Linkage disequilibrium could also confuse the researcher into considering that because polymorphism D is found to be associated with disease A, then the gene C, which contains polymorphism D, is involved in the etiology of the disease. It may be that polymorphism D is in linkage disequilibrium with another allele (polymorphism E) that is crucial to the proper function of gene C or it may in fact be part of another unrelated gene X. Another issue which arises is that if the association is found in one population, it has to be tested in other populations in order to avoid population stratication biases. If polymorphism D is in linkage disequilibrium with polymorphism E but polymorphism D is the actual susceptibility allele, it should exist in every population. Conversely, if the susceptibility locus is polymorphism E, the association between D and E may only occur in the original homogeneous population due to a phenomenon

termed the founder-effect, that is, the ancestors have the linkage disequilibrium which is passed on to offspring but the same linkage is not seen in completely distinct populations. Two points are worth considering here. Firstly, if polymorphism D is in linkage disequilibrium with E, polymorphism D is still a useful marker for the homogeneous population. Secondly, testing polymorphism D in other populations will truly test whether this polymorphism has a crucial inuence on gene C and also the disease in question. Of course, the biological plausibility of the polymorphism or marker, if checked initially, would have been crucial evidence in clarifying the association.

Proving the association

Recommendations emanating from population polymorphism association studies are invariably overstated in the periodontal and medical literature and are not based on a proper interpretation (conservative or ambitious) of the ndings. If researchers are lucky and nd an association between a polymorphism and periodontal disease, their immediate priority should be to determine whether the polymorphism actually does anything, i.e. to address whether the polymorphism is non-neutral. A nonneutral nding might be that the polymorphism results in the production of a protein with enhanced activity, as has been claimed for the interleukin-1 polymorphisms (131, 136), which Kornman and colleagues (100) have associated with periodontal disease. A further approach might be to consider what other genes the polymorphism is close to on the chromosome and whether these could plausibly have a role in the etiology of periodontal disease (Table 6). Other important considerations should be population biases, casecontrol selection, possible confounders, and whether their probability values are strong enough to discount false-positive associations if allowance has not been made for multiple comparisons. The typical next stage is to consider the association in larger and diverse populations, which would be independently assessed and the results reported in the literature (whether they conrm or deny the association). Associations of

caused Disease A by Aberration B

encoded Gene C by

which Polymorphism D contains

Fig. 2. Steps in determining a link between a polymorphism and a disease. Does Polymorphism D inuence Gene C functionally? Does Gene C uniquely result in Aberration or defect B? Does Aberration B have mechanistic plausibility for Disease A?

109

Kinane et al.

Table 6. Requirements in proving a diseasepolymorphism association

1) The polymorphism should inuence the gene product. 2) Biases in study population should be recognized, controlled and reported. 3) Confounders such as smoking and socioeconomic class must be factored into the analyses. 4) The affected gene product should be part of the disease etiopathology or there should be a plausible biological mechanistic explanation for its involvement.

polymorphisms with periodontal disease studies are at best hypotheses-generating exercises and clinicians should be clear about the extreme limitations of these approaches when trying to develop robust associations that can be used as clinically relevant risk evaluators for patients. In the vast majority of cases, associations will not relate to the function of the gene. Proving a functional relationship requires a considerable multidisciplinary effort in population, pathogenesis, and molecular genetics research.

most useful actions we can perform to ensure the early diagnosis of this disease. By careful clinical diagnostic procedures we may detect susceptible patients early and instigate therapy to prevent signicant disease occurring. In the pursuit of better genetic diagnostic tests for chronic and aggressive periodontitis we must plan our research using plausible biological arguments and carefully avoid bias and misinterpretation of genetic associations with disease.

Susceptibility proles
An interesting approach which is emerging in other elds of medicine is the development of the susceptibility prole concept in diseases such as Alzheimers. The risk of Alzheimers has been considered to be substantially inuenced by 10 genetic polymorphisms of inammation-related molecules including proinammatory cytokines (interleukin-1a and interleukin-1b and interleukin-6, tumor necrosis factor-a) and the protease inhibitors (a2-macroglobulin, a1-antitrypsin). McGeer et al. (116) consider that several of these relatively common high risk polymorphisms may be inherited by an individual, giving them a susceptibility prole which reects the combined inuence of the high risk polymorphisms. A similar high risk prole may yet emerge for periodontal disease but clearly much testing across different populations will be needed. If such information was available, therapeutic intervention strategies could be envisaged, aimed at preventing the development of periodontal disease.

Conclusions
We have reviewed the fundamental differences between simple Mendelian and complex genetic diseases, and highlighted the signicance of these differences for genetic testing. Simple or Mendelian genetic traits generally occur when a single gene defect disrupts the normal function of a protein sufciently to cause a disease or syndrome. Complex genetic diseases occur when allelic variants of multiple different genes act synergistically with environmental factors to increase or decrease the likelihood of developing a disease. While there have been dramatic successes in identication of mutations responsible for rare syndromic forms of periodontitis, few genetic polymorphisms reported for more complex genetic forms of periodontitis have been demonstrated to be clinically valid or to have clinical utility. A review of genetic associations reported for more common forms of periodontitis reveals that we are still some way from determining the genetic basis of either aggressive or chronic periodontitis. We have, however, gained some insight into the hereditary patterns for aggressive periodontitis, which in the US population is typically inherited as an autosomal dominant trait with reduced penetrance. The clinical message here is that the risk for offspring and siblings of patients affected with aggressive periodontitis approaches 50%.

Genetic screening for periodontitis risk

The current practical clinical utility of genetic knowledge in periodontics is limited. However, performing clinical periodontal assessments of siblings of aggressive periodontitis probands is one of the

110

Periodontal disease genetics

Advances in the annotation of the human genome and technological advances in high throughput genotyping of common genetic polymorphisms such as single nucleotide polymorphisms (SNPs) offers the possibility to test for signicant associations with common forms of periodontitis. A recent effort is the international HAPMAP project, which will identify each of the common SNPs in the human genome and characterize their allelic frequencies in different ethnic groups. The ultimate goal of the project is to determine the genetic location of each SNP and characterize how these genetic variants are distributed across the genomes of different ethnic population groups. The HAPMAP project will help in our search to identify genetic polymorphisms that can be used as diagnostic, prognostic, and even therapeutic clinical biomarkers. This will be a rst step in developing clinically valid and useful genetic tests. Determination of the biological basis of these functional polymorphisms will require appreciation of the genetic complexity of these conditions and careful avoidance of biases will be necessary to avoid misinterpretation of genetic associations. As the number of reports describing genetic polymorphisms associated with periodontal disease increase exponentially, the limitations of such studies need to be appreciated. There is often a large disparity in the signicance placed on genes and genetic polymorphism associations between geneticists and clinicians. A strong plea in the development of better genetic diagnostic tests for chronic and aggressive periodontitis is that we plan our research using plausible biological arguments and carefully avoid bias and misinterpretation of genetic associations with disease. Many studies fail to quantify meaningfully the magnitude of contribution of a particular diseaseassociated allele to disease risk. This failure to consider the sensitivity and specicity of these tests precludes determination of the clinical utility of the association. Additionally, environmental confounders are often not adequately addressed. Underpowered studies have insufcient numbers of cases and controls to make robust pronouncements on associations. Rarely is a precise account of race and ethnicity for cases and controls included. Biases can occur in multiracial areas of different socioeconomic classes, making the inference from such studies questionable. Ultimately, development of a susceptibility prole concept for complex diseases may emerge as a clinically valid approach to using genetic information for periodontal diseases. For example, a number of

relatively common SNPs could be demonstrated to contribute to a high susceptibility prole. It is likely that environmental factors (smoking, microbes, dietary components) would be included in the prole assessment. To prove such a high risk prole in periodontal disease requires much testing across different populations. Such genetic information would be invaluable in therapeutic intervention strategies aimed at preventing the development of periodontal disease.

References
1. Alaluusua S, Asikainen S, Lai CH. Intrafamilial transmission of Actinobacillus actinomycetemcomitans. J Periodontol 1991: 62: 207210. 2. Anderson DC, Springer TA. Leukocyte adhesion deciency: an inherited defect in the Mac-1, LFA-1, and p150,95 glycoproteins. Annu Rev Med 1987: 38: 175194. 3. Apaydin A. EhlersDanlos syndrome (type VIII). J Nihon University Sch Dent 1995: 37: 214217. 4. Armitage GC. Development of a classication system for periodontal diseases and conditions. Ann Periodontol 1999: 4: 16. 5. Armitage GC, Wu Y, Wang HY, Sorrell J, di Giovine FS, Duff GW. Low prevalence of a periodontitis-associated interleukin-1 composite genotype in individuals of Chinese heritage. J Periodontol 2000: 71: 164171. 6. Astemborski JA, Boughman JA, Myrick PO, Goodman SB, Wooten RK, Agarwal S, Vincent JW, Suzuki JB. Clinical and laboratory characterization of early onset periodontitis. J Periodontol 1989: 60: 557563. 7. Baab DA, Page RC, Ebersole JL, Williams BL, Scott CR. Laboratory studies of a family manifesting premature exfoliation of deciduous teeth. J Clin Periodontol 1986: 13: 677683. 8. Beaty TH, Boughman JA, Yang P, Astemborski JA, Suzuki JB. Genetic analysis of juvenile periodontitis in families ascertained through an affected proband. Am J Hum Genet 1987: 40: 443452. 9. Benjamin SD, Baer PN. Familial patterns of advanced alveolar bone loss in adolescence (periodontosis). Periodontics 1967: 5: 8288. 10. Berglundh T, Donati M, Hahn-Zoric M, Hanson LA, Padyukov L. Association of the -1087 IL 10 gene polymorphism with severe chronic periodontitis in Swedish Caucasians. J Clin Periodontol 2003: 30: 249254. 11. Berrettini WH. Genetics of psychiatric disease. Annu Rev Med 2000: 51: 465479. 12. Beumer J 3rd, Trowbridge HO, Silverman S Jr, Eisenberg E. Childhood hypophosphatasia and the premature loss of teeth. A clinical and laboratory study of seven cases. Oral Surg Oral Med Oral Pathol 1973: 35: 631640. 13. Boomsma D, Busjahn A, Peltonen L. Classical twin studies and beyond. Nat Rev Genet 2002: 3: 872882. 14. Boughman JA, Halloran SL, Roulston D, Schwartz S, Suzuki JB, Weitkamp LR, Wenk RE, Wooten R, Cohen MM. An autosomal-dominant form of juvenile periodontitis: its localization to chromosome 4 and linkage to

111

Kinane et al.
dentinogenesis imperfecta and Gc. J Craniofac Genet Dev Biol 1986: 6: 341350. Boughman JA, Astemborski JA, Suzuki JB. Phenotypic assessment of early onset periodontitis in sibships. J Clin Periodontol 1992: 19: 233239. Boughman JA, Beaty TH, Yang P, Goodman SB, Wooten RK, Suzuki JB. Problems of genetic model testing in early onset periodontitis. J Periodontol 1988: 59: 332337. Butler JH. A familial pattern of juvenile periodontitis (periodontosis). J Periodontol 1969: 40: 115118. Carinci F, Pezzetti F, Scapoli L, Martinelli M, Carinci P, Tognon M. Genetics of nonsyndromic cleft lip and palate: a review of international studies and data regarding the Italian population. Cleft Palate Craniofac J 2000: 37: 3340. Casson MH. Oral manifestations of primary hypophosphatasia. A case report. Br Dent J 1969: 127: 561566. Cattabriga M, Rotundo R, Muzzi L, Nieri M, Verrocchi G, Cairo F, Pini Prato G. Retrospective evaluation of the inuence of the interleukin-1 genotype on radiographic bone levels in treated periodontal patients over 10 years. J Periodontol 2001: 72: 767773. Chagnon YC, Perusse L, Bouchard C. The human obesity gene map: the 1997 update. Obes Res 1998: 6: 7692. Chung HY, Lu HC, Chen WL, Lu CT, Yang YH, Tsai CC. Gm (23) allotypes and Fcgamma receptor genotypes as risk factors for various forms of periodontitis. J Clin Periodontol 2003: 30: 954960. Cohen DW, Goldman HM. Clinical observations on the modication of human oral tissue metabolism by local intraoral factors. Ann N Y Acad Sci 1960: 85: 6895. Collins FS, McKusick VA. Implications of the Human Genome Project for medical science. JAMA 2001: 285: 540544. Colombo AP, Haffajee AD, Dewhirst FE, Paster BJ, Smith CM, Cugini MA, Socransky SS. Clinical and microbiological features of refractory periodontitis subjects. J Clin Periodontol 1998: 25: 169180. Corey LA, Nance WE, Hofstede P, Schenkein HA. Selfreported periodontal disease in a Virginia twin population. J Periodontol 1993: 64: 12051208. Craandijk J, van Krugten MV, Verweij CL, van der Velden U, Loos BG. Tumor necrosis factor-alpha gene polymorphisms in relation to periodontitis. J Clin Periodontol 2002: 29: 2834. Cullinan MP, Westerman B, Hamlet SM, Palmer JE, Faddy MJ, Lang NP, Seymour GJ. A longitudinal study of interleukin-1 gene polymorphisms and periodontal disease in a general adult population. J Clin Periodontol 2001: 28: 1137 1144. DAiuto F, Parkar M, Brett PM, Ready D, Tonetti MS. Gene polymorphisms in pro-inammatory cytokines are associated with systemic inammation in patients with severe periodontal infections. Cytokine 2004: 28: 2934. de Brito Junior RB, Scarel-Caminaga RM, Trevilatto PC, de Souza AP, Barros SP. Polymorphisms in the vitamin D receptor gene are associated with periodontal disease. J Periodontol 2004: 75: 10901095. De Sanctis M, Zucchelli G. Interleukin-1 gene polymorphisms and long-term stability following guided tissue regeneration therapy. J Periodontol 2000: 71: 606613. 32. de Souza AP, Trevilatto PC, Scarel-Caminaga RM, Brito RB, Line SR. MMP-1 promoter polymorphism: association with chronic periodontitis severity in a Brazilian population. J Clin Periodontol 2003: 30: 154158. 33. de Souza AP, Trevilatto PC, Scarel-Caminaga RM, de Brito RB, Line SR. Analysis of the TGF-beta1 promoter polymorphism (C-509T) in patients with chronic periodontitis. J Clin Periodontol 2003: 30: 519523. 34. Deasy MJ, Vogel RI, Macedo-Sobrinho B, Gertzman G, Simon B. Familial benign chronic neutropenia associated with periodontal disease. A case report. J Periodontol 1980: 51: 206210. 35. Diehl SR, Wang Y-F, Beck JD, Gao Y, Khanna A, Gu X, et al. Interleukin-1 genotypes and risk of early onset periodontitis a family study of linkage disequilibrium. J Dent Res 1998: 77: 717 (Abstract). 36. Diehl SR, Wang Y, Brooks CN, Burmeister JA, Califano JV, Wang S, Schenkein HA. Linkage disequilibrium of interleukin-1 genetic polymorphisms with early-onset periodontitis. J Periodontol 1999: 70: 418430. 37. Ehmke B, Kress W, Karch H, Grimm T, Klaiber B, Flemmig TF. Interleukin-1 haplotype and periodontal disease progression following therapy. J Clin Periodontol 1999: 26: 810813. 38. Eichner JE, Dunn ST, Perveen G, Thompson DM, Stewart KE, Stroehla BC. Apolipoprotein E polymorphism and cardiovascular disease: a HuGE review. Am J Epidemiol 2002: 155: 487495. 39. Endo M, Tai H, Tabeta K, Kobayashi T, Yamazaki K, Yoshie H. Analysis of single nucleotide polymorphisms in the 5-anking region of tumor necrosis factor-alpha gene in Japanese patients with early-onset periodontitis. J Periodontol 2001: 72: 15541559. 40. Engebretson SP, Lamster IB, Herrera-Abreu M, Celenti RS, Timms JM, Chaudhary AG, di Giovine FS, Kornman KS. The inuence of interleukin gene polymorphism on expression of interleukin-1beta and tumor necrosis factor-alpha in periodontal tissue and gingival crevicular uid. J Periodontol 1999: 70: 567573. 41. Fassmann A, Holla LI, Buckova D, Vasku A, Znojil V, Vanek J. Polymorphisms in the +252 (AG) lymphotoxinalpha and the -308 (AG) tumor necrosis factor-alpha genes and susceptibility to chronic periodontitis in a Czech population. J Periodontal Res 2003: 38: 394399. 42. Folwaczny M, Glas J, Torok HP, Fricke K, Folwaczny C. Prevalence of the chemokine receptor CCR5-Delta32 gene mutation in periodontal disease. Clin Immunol 2003: 109: 325329. 43. Folwaczny M, Glas J, Torok HP, Limbersky O, Folwaczny C. Toll-like receptor (TLR) 2 and 4 mutations in periodontal disease. Clin Exp Immunol 2004: 135: 330335. 44. Folwaczny M, Glas J, Torok HP, Mauermann D, Folwaczny C. The 3020insC mutation of the NOD2CARD15 gene in patients with periodontal disease. Eur J Oral Sci 2004: 112: 316319. 45. Folwaczny M, Glas J, Torok HP, Mende M, Folwaczny C. Lack of association between the TNF alpha G-308 A promoter polymorphism and periodontal disease. J Clin Periodontol 2004: 31: 449453. 46. Fourel J. Periodontosis: a periodontal syndrome. J Periodontol 1972: 43: 240255.

15.

16.

17. 18.

19. 20.

21. 22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

112

Periodontal disease genetics

47. Fourel J. Periodontosis, juvenile periodontitis or Gottlieb syndrome? Report of 4 cases. J Periodontol 1974: 45: 234 237. 48. Frazer D. Hypophosphatasia. Am J Med 1957: 22: 730 746. 49. Friedrich MJ. Relating genomic research to patient care. JAMA 2000: 284: 25812582. 50. Galbraith GM, Steed RB, Sanders JJ, Pandey JP. Tumor necrosis factor alpha production by oral leukocytes: inuence of tumor necrosis factor genotype. J Periodontol 1998: 69: 428433. 51. Gambaro B, Angiani F, DAngelo A. Association studies of genetic polymorphisms and complex disease. Lancet 2000: 35: 308311. 52. Glazier AM, Nadeau JH, Aitman TJ. Finding genes that underlie complex traits. Science 2002: 298: 23452349. 53. Gonzales JR, Kobayashi T, Michel J, Mann M, Yoshie H, Meyle J. Interleukin-4 gene polymorphisms in Japanese and Caucasian patients with aggressive periodontitis. J Clin Periodontol 2004: 31: 384389. 54. Gonzales JR, Michel J, Diete A, Herrmann JM, Bodeker RH, Meyle J. Analysis of genetic polymorphisms at the interleukin-10 loci in aggressive and chronic periodontitis. J Clin Periodontol 2002: 29: 816822. 55. Gore EA, Sanders JJ, Pandey JP, Palesch Y, Galbraith GM. Interleukin-1beta+3953 allele 2: association with disease status in adult periodontitis. J Clin Periodontol 1998: 25: 781785. 56. Gorlin RJ, Cohen MM, Hennekam RC. Syndromes of the head and neck. Oxford: Oxford University Press, 2001. 57. Greenberg DA. Linkage analysis of necessary disease loci versus susceptibility loci. Am J Hum Genet 1993: 52: 135 143. 58. Guttmacher AE, Collins FS. Genomic medicine a primer. N Engl J Med 2002: 347: 15121520. 59. Hamilton RE Jr, Giansanti JS. The ChediakHigashi syndrome. Report of a case and review of the literature. Oral Surg Oral Med Oral Pathol 1974: 37: 754761. ` 60. Haneke E. The PapillonLefevre syndrome: keratosis palmoplantaris with periodontopathy. Report of a case and review of the cases in the literature. Hum Genet 1979: 51: 135. 61. Hansen L, Pedersen O. Genetics of type 2 diabetes mellitus: status and perspectives. Diabetes Obes Metab 2005: 7: 122135. 62. Hart TC. Genetic considerations of risk in human periodontal disease. Curr Opin Periodontol 1994: 311. 63. Hart TC. Genetic risk-factors for early-onset periodontitis. J Periodontol 1996: 67: 355366. 64. Hart TC, Hart PS, Bowden DW, Michalec MD, Callison SA, Walker SJ, Zhang Y, Firatli E. Mutations of the cathepsin C ` gene are responsible for PapillonLefevre syndrome. J Med Genet 1999: 36: 881887. 65. Hart TC, Hart PS, Michalec MD, Zhang Y, Firatli E, Van Dyke TE, Stabholz A, Zlotogorski A, Shapira L, Soskolne ` WA. HaimMunk syndrome and PapillonLefevre syndrome are allelic mutations in cathepsin C. J Med Genet 2000: 37: 8894. 66. Hart TC, Kornman KS. Genetic factors in the pathogenesis of periodontitis. Periodontol 2000 1997: 14: 202215. 67. Hart TC, Marazita ML, Schenkein HA, Brooks CN, Gunsolley JG, Diehl SR. No female preponderance in juvenile periodontitis after correction for ascertainment bias. J Periodontol 1991: 62: 745749. 68. Hart TC, Marazita ML, Schenkein HA, Diehl SR. Re-interpretation of the evidence for X-linked dominant inheritance of juvenile periodontitis. J Periodontol 1992: 63: 169173. 69. Hart TC, Marazita ML, McCanna KM, Schenkein HA, Diehl SR. Reevaluation of the chromosome 4q candidate region for early onset periodontitis. Hum Genet 1993: 91: 416422. ` 70. Hart TC, Shapira L. PapillonLefevre syndrome. Periodontol 2000 1994: 6: 88100. 71. Hart TC, Stabholz A, Meyle J, Shapira L, Van Dyke TE, Cutler CW, Soskolne WA. Genetic studies of syndromes with severe periodontitis and palmoplantar hyperkeratosis. J Periodontal Res 1997: 32: 8189. 72. Hart TC, Zhang Y, Gorry MC, Hart PS, Cooper M, Marazita ML, Marks JM, Cortelli JR, Pallos D. A mutation in the SOS1 gene causes hereditary gingival bromatosis type 1. Am J Hum Genet 2002: 70: 943954. 73. Hassell TM, Harris EL. Genetic inuences in caries and periodontal diseases. Crit Rev Oral Biol Med 1995: 6: 319 342. 74. Hennig BJ, Parkhill JM, Chapple IL, Heasman PA, Taylor JJ. Association of a vitamin D receptor gene polymorphism with localized early-onset periodontal diseases. J Periodontol 1999: 70: 10321038. 75. Hirschhorn JN, Lohmueller K, Byrne E, Hirschhorn K. A comprehensive review of genetic association studies. Genet Med 2002: 4: 4561. 76. Hodge SE. Linkage analysis versus association analysis: distinguishing between two models that explain diseasemarker associations. Am J Hum Genet 1993: 53: 367 384. 77. Hodge SE. What association analysis can and cannot tell us about the genetics of complex disease. Am J Med Genet 1994: 54: 318323. 78. Hodge P, Michalowicz B. Genetic predisposition to periodontitis in children and young adults. Periodontol 2000 2001: 26: 113134. 79. Hodge PJ, Riggio MP, Kinane DF. No association with HLA-DQB1 in European Caucasians with early-onset periodontitis. Tissue Antigens 1999: 54: 205207. 80. Hodge PJ, Riggio MP, Kinane DF. Failure to detect an association with IL1 genotypes in European Caucasians with generalised early onset periodontitis. J Clin Periodontol 2001: 28: 430436. 81. Hodge PJ, Teague PW, Wright AF, Kinane DF. Clinical and genetic analysis of a large North European Caucasian family affected by early-onset periodontitis. J Dent Res 2000: 79: 857863. 82. Holla LI, Buckova D, Fassmann A, Halabala T, Vasku A, Vacha J. Promoter polymorphisms in the CD14 receptor gene and their potential association with the severity of chronic periodontitis. J Med Genet 2002: 39: 844848. 83. Holla LI, Fassmann A, Benes P, Halabala T, Znojil V. 5 polymorphisms in the transforming growth factor-beta 1 gene (TGF-beta 1) in adult periodontitis. J Clin Periodontol 2002: 29: 336341. 84. Holla LI, Fassmann A, Vasku A, Znojil V, Vanek J, Vacha J. Interactions of lymphotoxin alpha (TNF-beta), angiotensin-converting enzyme (ACE), and endothelin-1 (ET-1)

113

Kinane et al.
gene polymorphisms in adult periodontitis. J Periodontol 2001: 72: 8589. Holla LI, Jurajda M, Fassmann A, Dvorakova N, Znojil V, Vacha J. Genetic variations in the matrix metalloproteinase-1 promoter and risk of susceptibility andor severity of chronic periodontitis in the Czech population. J Clin Periodontol 2004: 31: 685690. Itagaki M, Kubota T, Tai H, Shimada Y, Morozumi T, Yamazaki K. Matrix metalloproteinase-1 and -3 gene promoter polymorphisms in Japanese patients with periodontitis. J Clin Periodontol 2004: 31: 764769. Izakovicova Holla L, Buckova D, Fassmann A, Benes P, Znojil V. Plasminogen-activator-inhibitor-1 promoter polymorphism as a risk factor for adult periodontitis in non-smokers. Genes Immun 2002: 3: 292294. Jenkins WM, Kinane DF. The high risk group in periodontitis. Br Dent J 1989: 167: 168171. Johnson NW, Grifths GS, Wilton JM, Maiden MF, Curtis MA, Gillett IR, Wilson DT, Sterne JA. Detection of highrisk groups and individuals for periodontal diseases. Evidence for the existence of high-risk groups and individuals and approaches to their detection. J Clin Periodontol 1988: 15: 276282. Jorgenson RJ, Levin LS, Hutcherson ST, Salinas CF. Periodontosis in sibs. Oral Surg Oral Med Oral Pathol 1975: 39: 396402. Kang BY, Choi YK, Choi WH, Kim KT, Choi SS, Kim K, Ha NJ. Two polymorphisms of interleukin-4 gene in Korean adult periodontitis. Arch Pharm Res 2003: 26: 482 486. Khoury MJ, McCabe LL, McCabe ER. Population screening in the age of genomic medicine. N Engl J Med 2003: 348: 5058. Kinane DF, Chestnutt IG. Smoking and periodontal disease. Crit Rev Oral Biol Med 2000: 11: 356365. Kinane DF, Hodge P, Eskdale J, Ellis R, Gallagher G. Analysis of genetic polymorphisms at the interleukin-10 and tumour necrosis factor loci in early-onset periodontitis. J Periodontal Res 1999: 34: 379386. Kobayashi T, Sugita N, van der Pol WL, Nunokawa Y, Westerdaal NA, Yamamoto K, van de Winkel JG, Yoshie H. The Fcgamma receptor genotype as a risk factor for generalized early-onset periodontitis in Japanese patients. J Periodontol 2000: 71: 14251432. Kobayashi T, Westerdaal NA, Miyazaki A, van der Pol WL, Suzuki T, Yoshie H, van de Winkel JG, Hara K. Relevance of immunoglobulin G Fc receptor polymorphism to recurrence of adult periodontitis in Japanese patients. Infect Immun 1997: 65: 35563560. Kobayashi T, Yamamoto K, Sugita N, van der Pol WL, Yasuda K, Kaneko S, van de Winkel JG, Yoshie H. The Fc gamma receptor genotype as a severity factor for chronic periodontitis in Japanese patients. J Periodontol 2001: 72: 13241331. Kocher T, Sawaf H, Fanghanel J, Timm R, Meisel P. Association between bone loss in periodontal disease and polymorphism of N-acetyltransferase (NAT2). J Clin Periodontol 2002: 29: 2127. Korkhaus G. Hereditary disposition of paradentosis [in German]. Dtsch Zahnarztl Z 1952: 7: 441448. Kornman KS, Crane A, Wang HY, di Giovine FS, Newman MG, Pirk FW, Wilson TG Jr, Higginbottom FL, Duff GW. The interleukin-1 genotype as a severity factor in adult periodontal disease. J Clin Periodontol 1997: 24: 7277. Laine ML, Farre MA, Gonzalez G, van Dijk LJ, Ham AJ, Winkel EG, Crusius JB, Vandenbroucke JP, van Winkelhoff AJ, Pena AS. Polymorphisms of the interleukin-1 gene family, oral microbial pathogens, and smoking in adult periodontitis. J Dent Res 2001: 80: 16951699. Laine ML, Murillo LS, Morre SA, Winkel EG, Pena AS, van Winkelhoff AJ. CARD15 gene mutations in periodontitis. J Clin Periodontol 2004: 31: 890893. Lang NP, Tonetti MS, Suter J, Sorrell J, Duff GW, Kornman KS. Effect of interleukin-1 gene polymorphisms on gingival inammation assessed by bleeding on probing in a periodontal maintenance population. J Periodontal Res 2000: 35: 102107. Larsen PM, Fey SJ, Larsen MR, Nawrocki A, Andersen HU, Kahler H, Heilmann C, Voss MC, Roepstorff P, Pociot F, Karlsen AE, Nerup J. Proteome analysis of interleukin1beta-induced changes in protein expression in rat islets of Langerhans. Diabetes 2001: 50: 10561063. Lavine WS, Page RC, Padgett GA. Host response in chronic periodontal disease. V. The dental and periodontal status of mink and mice affected by ChediakHigashi syndrome. J Periodontol 1976: 47: 621635. Li X, Rao S, Elston RC, Olson JM, Moser KL, Zhang T, Guo Z. Locating the genes underlying a simulated complex disease by discriminant analysis. Genet Epidemiol 2001: 21 (Suppl. 1): S516S521. Li Y, Xu L, Hasturk H, Kantarci A, DePalma SR, Van Dyke TE. Localized aggressive periodontitis is linked to human chromosome 1q25. Hum Genet 2004: 114: 291297. Linch DC, Acton CH. EhlersDanlos syndrome presenting with juvenile destructive periodontitis. Br Dent J 1979: 147: 9596. Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN. Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet 2003: 33: 177182. Long JC, Nance WE, Waring P, Burmeister JA, Ranney RR. Early onset periodontitis: a comparison and evaluation of two proposed modes of inheritance. Genet Epidemiol 1987: 4: 1324. Loos BG, Leppers-Van de Straat FG, Van de Winkel JG, Van der Velden U. Fcgamma receptor polymorphisms in relation to periodontitis. J Clin Periodontol 2003: 30: 595 602. Marazita ML, Burmeister JA, Gunsolley JC, Koertge TE, Lake K, Schenkein HA. Evidence for autosomal dominant inheritance and race-specic heterogeneity in early-onset periodontitis. J Periodontol 1994: 65: 623630. Marazita ML, Neiswanger K. Association analysis. In: Wysznyski D, editor. Cleft lip and palate: from origin to treatment, Ch 2. Oxford: University Press, 2002. Mark LL, Haffajee AD, Socransky SS, Kent RL Jr, Guerrero D, Kornman K, Newman M, Stashenko P. Effect of the interleukin-1 genotype on monocyte IL-1beta expression in subjects with adult periodontitis. J Periodontal Res 2000: 35: 172177. McDevitt MJ, Wang HY, Knobelman C, Newman MG, di Giovine FS, Timms J, Duff GW, Kornman KS. Interleukin)1 genetic association with periodontitis in clinical practice. J Periodontol 2000: 71: 156163.

85.

101.

86.

102.

103.

87.

88. 89.

104.

105.

90.

106.

91.

107.

92.

108.

93. 94.

109.

110.

95.

111.

96.

112.

97.

113.

114.

98.

99. 100.

115.

114

Periodontal disease genetics

116. McGeer PL, McGeer EG. Polymorphisms in inammatory genes and the risk of Alzheimer disease. Arch Neurol 2001: 58: 17901792. 117. McGuire MK, Nunn ME. Prognosis versus actual outcome. IV. The effectiveness of clinical parameters and IL-1 genotype in accurately predicting prognoses and tooth survival. J Periodontol 1999: 70: 4956. 118. Meisel P, Carlsson LE, Sawaf H, Fanghaenel J, Greinacher A, Kocher T. Polymorphisms of Fc gamma-receptors RIIa, RIIIa, and RIIIb in patients with adult periodontal diseases. Genes Immun 2001: 2: 258262. 119. Meisel P, Siegemund A, Dombrowa S, Sawaf H, Fanghaenel J, Kocher T. Smoking and polymorphisms of the interleukin-1 gene cluster (IL-1alpha, IL-1beta, and IL-1RN) in patients with periodontal disease. J Periodontol 2002: 73: 2732. 120. Meisel P, Timm R, Sawaf H, Fanghanel J, Siegmund W, Kocher T. Polymorphism of the N-acetyltransferase (NAT2), smoking and the potential risk of periodontal disease. Arch Toxicol 2000: 74: 343348. 121. Melnick M, Shields ED, Bixler D. Periodontosis: a phenotypic and genetic analysis. Oral Surg Oral Med Oral Pathol 1976: 42: 3241. 122. Michalowicz BS. Genetic and heritable risk factors in periodontal disease. J Periodontol 1994: 65: 479488. 123. Michalowicz BS, Aeppli DP, Kuba RK, Bereuter JE, Conry JP, Segal NL, Bouchard TJ Jr, Pihlstrom BL. A twin study of genetic variation in proportional radiographic alveolar bone height. J Dent Res 1991: 70: 14311435. 124. Michalowicz BS, Aeppli D, Virag JG, Klump DG, Hinrichs JE, Segal NL, Bouchard TJ Jr, Pihlstrom BL. Periodontal ndings in adult twins. J Periodontol 1991: 62: 293299. 125. Michalowicz BS, Diehl SR, Gunsolley JC, Sparks BS, Brooks CN, Koertge TE, Califano JV, Burmeister JA, Schenkein HA.Evidence of a substantial genetic basis for risk of adult periodontitis. J Periodontol 2000: 71: 16991707. 126. Michel J, Gonzales JR, Wunderlich D, Diete A, Herrmann JM, Meyle J. Interleukin-4 polymorphisms in early onset periodontitis. J Clin Periodontol 2001: 28: 483488. 127. Monaghan F, Corcos A. On the origins of the Mendelian laws. J Hered 1984: 75: 6769. 128. Murray JC. Face facts. Genes, environment, and clefts. Am J Hum Genet 1995: 57: 227232. 129. Ohtonen S, Kontturi-Narhi V, Markkanen H, Syrjanen S. Juvenile periodontitis a clinical and radiological familial study. J Pedod 1983: 8: 2833. 130. Ohyama H, Takashiba S, Oyaizu K, Nagai A, Naruse T, Inoko H, Kurihara H, Murayama Y. HLA Class II genotypes associated with early-onset periodontitis: DQB1 molecule primarily confers susceptibility to the disease. J Periodontol 1996: 67: 888894. 131. Page RC, Beatty P, Waldrop TC. Molecular basis for the functional abnormality in neutrophils from patients with generalized prepubertal periodontitis. J Periodontal Res 1987: 22: 182183. 132. Page RC, Beck JD. Risk assessment for periodontal diseases. Int Dent J 1997: 47: 6187. 133. Page RC, Bowen T, Altman L, Vandesteen E, Ochs H, Mackenzie P, Osterberg S, Engel LD, Williams BL. Prepubertal periodontitis. I. Denition of a clinical disease entity. J Periodontol 1983: 54: 257271. 134. Papapanou PN, Neiderud AM, Sandros J, Dahlen G. Interleukin-1 gene polymorphism and periodontal status. A case-control study. J Clin Periodontol 2001: 28: 389396. 135. Parkhill JM, Hennig BJ, Chapple IL, Heasman PA, Taylor JJ. Association of interleukin-1 gene polymorphisms with early-onset periodontitis. J Clin Periodontol 2000: 27: 682 689. 136. Pociot F, Briant L, Jongeneel CV, Molvig J, Worsaae H, Abbal M, Thomsen M, Nerup J, Cambon-Thomsen A. Association of tumor necrosis factor (TNF) and class II major histocompatibility complex alleles with the secretion of TNF-alpha and TNF-beta by human mononuclear cells: a possible link to insulin-dependent diabetes mellitus. Eur J Immunol 1993: 23: 224231. 137. Pociot F, Molvig J, Wogensen L, Worsaae H, Nerup J. A TaqI polymorphism in the human interleukin-1 beta (IL-1 beta) gene correlates with IL-1 beta secretion in vitro. Eur J Clin Invest 1992: 22: 396402. 138. Pontes CC, Gonzales JR, Novaes AB Jr, Junior MT, Grisi MF, Michel J, Meyle J, de Souza SL. Interleukin-4 gene polymorphism and its relation to periodontal disease in a Brazilian population of African heritage. J Dent 2004: 32: 241246. 139. Potter RH. Etiology of periodontitis: the heterogeneity paradigm. J Periodontol 1989: 60: 593597. 140. Potter RH. Genetic studies of juvenile periodontitis. J Dent Res 1990: 69: 9495. 141. Pritchard JK, Cox NJ. The allelic architecture of human disease genes: common disease-common variantor not? Hum Mol Genet 2002: 11: 24172423. 142. Reich DE, Lander ES. On the allelic spectrum of human disease. Trends Genet 2001: 17: 502510. 143. Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science 1996: 273: 1516 1517. 144. Saglam F, Atamer T, Onan U, Soydinc M, Kirac K. Infantile genetic agranulocytosis (Kostmann type): a case report. J Periodontol 1995: 66: 808810. 145. Sahingur SE, Sharma A, Genco RJ, De Nardin E. Association of increased levels of brinogen and the -455GA brinogen gene polymorphism with chronic periodontitis. J Periodontol 2003: 74: 329337. 146. Saxen L. Heredity of juvenile periodontitis. J Clin Periodontol 1980: 7: 276288. 147. Saxen L, Nevanlinna HR. Autosomal recessive inheritance of juvenile periodontitis: test of a hypothesis. Clin Genet 1984: 25: 332335. 148. Scarel-Caminaga RM, Trevilatto PC, Souza AP, Brito RB, Camargo LE, Line SR. Interleukin 10 gene promoter polymorphisms are associated with chronic periodontitis. J Clin Periodontol 2004: 31: 443448. 149. Scarel-Caminaga RM, Trevilatto PC, Souza AP, Brito RB, Line SR. Investigation of an IL-2 polymorphism in patients with different levels of chronic periodontitis. J Clin Periodontol 2002: 29: 587591. 150. Scarel-Caminaga RM, Trevilatto PC, Souza AP, Brito RB Jr, Line SR. Investigation of IL4 gene polymorphism in individuals with different levels of chronic periodontitis in a Brazilian population. J Clin Periodontol 2003: 30: 341345.

115

Kinane et al.
151. Shapira L, Schlesinger M, Bimstein E. Possible autosomaldominant inheritance of prepubertal periodontitis in an extended kindred. J Clin Periodontol 1997: 24: 388393. 152. Shapira L, Stabholz A, Rieckmann P, Kruse N. Genetic polymorphism of the tumor necrosis factor (TNF)-alpha promoter region in families with localized early-onset periodontitis. J Periodontal Res 2001: 36: 183186. 153. Shimada Y, Tai H, Endo M, Kobayashi T, Akazawa K, Yamazaki K. Association of tumor necrosis factor receptor type 2 +587 gene polymorphism with severe chronic periodontitis. J Clin Periodontol 2004: 31: 463469. 154. Shirodaria S, Smith J, McKay IJ, Kennett CN, Hughes FJ. Polymorphisms in the IL-1A gene are correlated with levels of interleukin-1alpha protein in gingival crevicular uid of teeth with severe periodontal disease. J Dent Res 2000: 79: 18641869. 155. Socransky SS, Haffajee AD, Smith C, Duff GW. Microbiological parameters associated with IL-1 gene polymorphisms in periodontitis patients. J Clin Periodontol 2000: 27: 810818. 156. Sofaer JA. Genetic approaches in the study of periodontal diseases. J Clin Periodontol 1990: 17: 401408. 157. Soga Y, Nishimura F, Ohyama H, Maeda H, Takashiba S, Murayama Y. Tumor necrosis factor-alpha gene (TNFalpha) -1031-863-857 single-nucleotide polymorphisms (SNPs) are associated with severe adult periodontitis in Japanese. J Clin Periodontol 2003: 30: 524531. 158. Stabholz A, Mann J, Agmon S, Soskolne WA. The description of a unique population with a very high prevalence of localized juvenile periodontitis. J Clin Periodontol 1998: 25: 872878. 159. Sugita N, Kobayashi T, Ando Y, Yoshihara A, Yamamoto K, van de Winkel JG, Miyazaki H, Yoshie H. Increased frequency of FcgammaRIIIb-NA1 allele in periodontitisresistant subjects in an elderly Japanese population. J Dent Res 2001: 80: 914918. 160. Sun JL, Meng HX, Cao CF, Tachi Y, Shinohara M, Ueda M, Imai H, Ohura K. Relationship between vitamin D receptor gene polymorphism and periodontitis. J Periodontal Res 2002: 37: 263267. 161. Sussman HI, Baer PN. Three generations of periodontosis: case report. Ann Dent 1978: 37: 811. 162. Suzuki A, Ji G, Numabe Y, Muramatsu M, Gomi K, Kanazashi M, Ogata Y, Shimizu E, Shibukawa Y, Ito A, Ito T, Sugaya A, Arai T, Yamada S, Deguchi S, Kamoi K. Single nucleotide polymorphisms associated with aggressive periodontitis and severe chronic periodontitis in Japanese. Biochem Biophys Res Commun 2004: 317: 887892. 163. Tachi Y, Shimpuku H, Nosaka Y, Kawamura T, Shinohara M, Ueda M, Imai H, Ohura K. Vitamin D receptor gene polymorphism is associated with chronic periodontitis. Life Sci 2003: 73: 33133321. 164. Tachi Y, Shimpuku H, Nosaka Y, Kawamura T, Shinohara M, Ueda M, Imai H, Ohura K, Sun J, Meng H, Cao C. Association of vitamin D receptor gene polymorphism with periodontal diseases in Japanese and Chinese. Nucleic Acids Res Suppl 2001: 111112. 165. Takashiba S, Noji S, Nishimura F, Ohyama H, Kurihara H, Nomura Y, Taniguchi S, Murayama Y. Unique intronic variations of HLA-DQ beta gene in early-onset periodontitis. J Periodontol 1994: 65: 379386. 166. Takashiba S, Ohyama H, Oyaizu K, Kogoe-Kato N, Murayama Y. HLA genetics for diagnosis of susceptibility to early-onset periodontitis. J Periodontal Res 1999: 34: 374378. 167. Tang Y, Zhang JC, Zhang WH, Pang RY. The association between Fc gamma receptor IIA gene polymorphism and susceptibility to chronic periodontitis in Chinese Han nationality [in Chinese]. Hua Xi Kou Qiang Yi Xue Za Zhi 2004: 22: 158161. 168. Temple TR, Kimball HR, Kakehashi S, Amen CR. Host factors in periodontal disease: periodontal manifestations of ChediakHigashi syndrome. J Periodontal Res 1972: 7 (Suppl.): 2627. 169. Thomson G. HLA disease associations: models for insulin dependent diabetes mellitus and the study of complex human genetic disorders. Annu Rev Genet 1988: 22: 3150. 170. Toomes C, James J, Wood AJ, Wu CL, McCormick D, Lench N, Hewitt C, Moynihan L, Roberts E, Woods CG, Markham A, Wong M, Widmer R, Ghaffar KA, Pemberton M, Hussein IR, Temtamy SA, Davies R, Read AP, Sloan P, Dixon MJ, Thakker NS. Loss-of-function mutations in the cathepsin C gene result in periodontal disease and palmoplantar keratosis. Nat Genet 1999: 23: 421424. 171. Townsend GC, Aldred MJ, Bartold PM. Genetic aspects of dental disorders. Aust Dent J 1998: 43: 269286. 172. Trevilatto PC, Scarel-Caminaga RM, de Brito RB Jr, de Souza AP, Line SR. Polymorphism at position -174 of IL-6 gene is associated with susceptibility to chronic periodontitis in a Caucasian Brazilian population. J Clin Periodontol 2003: 30: 438442. 173. Van Dyke TE, Schweinebraten M, Cianciola LJ, Offenbacher S, Genco RJ. Neutrophil chemotaxis in families with localized juvenile periodontitis. J Periodontal Res 1985: 20: 503514. 174. Waldrop TC, Anderson DC, Hallmon WW, Schmalstieg FC, Jacobs RL. Periodontal manifestations of the heritable Mac-1, LFA)1, deciency syndrome. Clinical, histopathologic and molecular characteristics. J Periodontol 1987: 58: 400416. 175. Walker SJ, Van Dyke TE, Rich S, Kornman KS, di Giovine FS, Hart TC. Genetic polymorphisms of the IL-1alpha and IL-1beta genes in African-American LJP patients and an African-American control population. J Periodontol 2000: 71: 723728. 176. Wiebe CB, Silver JG, Larjava HS. Early-onset periodontitis associated with WearyKindler syndrome: a case report. J Periodontol 1996: 67: 10041010. 177. Yamamoto K, Kobayashi T, Grossi S, Ho AW, Genco RJ, Yoshie H, De Nardin E. Association of Fcgamma receptor IIa genotype with chronic periodontitis in Caucasians. J Periodontol 2004: 75: 517522. 178. Yamazaki K, Tabeta K, Nakajima T, Ohsawa Y, Ueki K, Itoh H, Yoshie H. Interleukin-10 gene promoter polymorphism in Japanese patients with adult and early-onset periodontitis. J Clin Periodontol 2001: 28: 828832. 179. Yamazaki K, Ueki-Maruyama K, Oda T, Tabeta K, Shimada Y, Tai H, Nakajima T, Yoshie H, Herawati D, Seymour GJ. Single-nucleotide polymorphism in the CD14 promoter and periodontal disease expression in a Japanese population. J Dent Res 2003: 82: 612616.

116

Periodontal disease genetics

180. Yasuda K, Sugita N, Kobayashi T, Yamamoto K, Yoshie H. FcgammaRIIB gene polymorphisms in Japanese periodontitis patients. Genes Immun 2003: 4: 541546. 181. Yoshihara A, Sugita N, Yamamoto K, Kobayashi T, Miyazaki H, Yoshi H. Analysis of vitamin D and Fcgamma receptor polymorphisms in Japanese patients with generalized early-onset periodontitis. J Dent Res 2001: 80: 20512054. 182. Zhang Y, Hart PS, Moretti AJ, Bouwsma OJ, Fisher EM, Dudlicek L, Pettenati MJ, Hart TC. Biochemical and mutational analyses of the cathepsin c gene (CTSC) in ` three North American families with PapillonLefevre syndrome. Hum Mutat 2002: 20: 75. 183. Zhang Y, Lundgren T, Renvert S, Tatakis DN, Firatli E, Uygur C, Hart PS, Gorry MC, Marks JJ, Hart TC. Evidence of a founder effect for four cathepsin C gene mutations in ` PapillonLefevre syndrome patients. J Med Genet 2001: 38: 96101. 184. Zhang L, Meng H, Zhao H, Li Q, Xu L, Chen Z, Shi D, Feng X. Estrogen receptor-alpha gene polymorphisms in patients with periodontitis. J Periodontal Res 2004: 39: 362366. 185. Zhang Y, Syed R, Uygar C, Pallos D, Gorry MC, Firatli E, Cortelli JR, VanDyke TE, Hart PS, Feingold E, Hart TC. Evaluation of human leukocyte N-formylpeptide receptor (FPR1) SNPs in aggressive periodontitis patients. Genes Immun 2003: 4: 2229. 186. Zhao H. Family-based association studies. Stat Methods Med Res 2000: 9: 563587.

117