Mutations can be either inherited from a parent, in which case they will be present in all cells of the body, or acquired during a person’s lifetime, in which case they could be restricted to a particular part of the body depending on when during the person’s development the mutation was acquired. The former can be passed through families either in an autosomal or a sex-linked mendelian pattern of inheritance (such as dominant or recessive) or by a nonmendelian complex mode of inheritance.

	CLONING OF DISEASE GENES: FUNCTIONAL VERSUS POSITIONAL

TOP ABSTRACT THE BASICS OF HUMAN... CLONING OF DISEASE GENES:... MOLAR HYPODONTIA: AN EXAMPLE... CONCLUSION REFERENCES

 
Until the early 1980s, disease gene identification was accomplished largely by a process called "functional cloning," which identifies the gene based on the known biochemical cause of the disease. In the case of the disease phenylketonuria, the biochemical cause of its etiology was identified from the observed lack of oxidation of phenylalanine in people who had the disease. This, in turn, pointed to the lack of phenylalanine hydroxylase (PAH) activity; researchers then used antibodies to PAH to clone the gene for PAH.²⁵ However, for a vast majority of inherited disorders such as hypodontia, use of this method is not possible, as the disorders have no easily identifiable biochemical clues pointing to an absent protein, thus preventing the genes that encode them from being identified readily.

This, therefore, requires a second identification method called "positional cloning," which can be likened to searching for a single spelling mistake in a 46-volume encyclopedia that has three billion alphabet characters. One first must search for the volume at fault and then narrow down the region to a chapter, then a section, then a paragraph and finally a word within which one will find the spelling mistake. In positional cloning, chromosomes can be thought of as the 46 volumes that are split into two companion sets of 23 volumes that are segregated into discrete chapters by searchable markers, or title pages. The latter are composed of smaller discrete units called "genes" (the sections), which are made up from protein-coding exons (the paragraphs). Identification of a single spelling mistake within a single word, or amino acid codon, would be akin to finding a disease-causing mutation within the three billion base pairs of the human genome.

The positional cloning approach uses one or both of the following strategies:

– identification of a specific chromosomal aberration(s) in a person with the disease;

– analysis of genetic linkage in families segregating the disease in a mendelian pattern.

In an ideal world, all diseases would follow a mendelian pattern of inheritance and in each instance show linkage to a single mutation in a single causative gene.

Chromosome aberrations. Chromosome aberrations—such as deletions, translocations or duplications—offer the greatest boon to the success of positional cloning, as they usually are readily identifiable under the microscope using chromosomal dye banding. By this method, a small deletion was identified on the X chromosome of a boy who had Duchenne’s muscular dystrophy.²⁷ Researchers^28,29 subsequently generated a physical sequence map of the deleted region, which facilitated the identification of the responsible gene.³⁰

Linkage mapping. The vast majority of diseases are caused by more subtle mutations (such as point mutations in gene DNA sequences) rather than chromosomal anomalies. To find the genes of interest in these cases, researchers use a different approach called "linkage mapping" or "positional cloning." Researchers initially localize, or map, the disease gene locus to a specific chromosome, then systematically examine genes in the candidate region for mutations. The initial mapping has a few prerequisites for successful genetic analysis: identification of families segregating the disease phenotype, a distinctive diagnostic criterion to distinguish the affected people, an accurate assessment of the family members for a known mendelian or complex pattern of inheritance, and highly polymorphic DNA markers.

For relatively rare diseases, a single large family with 10 or more affected members or a few large families with five or more affected members are a suitable basis for linkage analysis. For more common diseases in which the pattern of disease inheritance is unclear, a large number ( 10) of families with two or more affected members per family is a good basis for linkage analysis. In all cases, distinctive diagnostic criteria are essential for the correct placement of people into the affected and unaffected groupings. Misdiagnosis of even a single person and his or her placement in the wrong group can hinder the linkage analysis and potentially can mask the correct locus from the researcher. For this reason, recruiting families of suitable size often is difficult, as ambiguities in diagnosis are commonplace. In addition, some mutations have low penetrance—that is, their presence does not always result in a disease phenotype owing to the influence of other "modifying" genes. The presence of phenocopies, who are people within a family with the same phenotype but whose etiology is unrelated to that of all other affected members of the family, can complicate the search. Finally, DNA polymorphisms, which are silent variations present throughout the genome, are the key to positional cloning. Polymorphic DNA markers are points within the genome that exist in two or more forms, or alleles. They enable construction of a "bar code" for the maternal and paternal copy of each chromosome for every person in the family. Linkage mapping follows the movement of each marker through each generation of a family using these bar codes within affected and unaffected people³¹ (Figure 2B). If a particular allele of a marker tends to "travel" with the disease status, then the marker is said to cosegregate with the disease. It is these cosegregating markers for which the geneticist is looking, as they are likely to be close to the responsible gene. Once the geneticist identifies such a marker, he or she creates a map of the region encompassing it to produce a physical map.

View larger version (12K): [in this window] [in a new window]   Figure 2. A. A large kindred segregating autosomal dominant molar hypodontia that allowed the mapping of the hypodontia gene to chromosome 14 and the identification of a single base insertion within the PAX9 gene of affected people.19 B. A small nuclear family with a similar phenotype as the family in A, in which affected family members had a deletion of the entire PAX9 gene,31 identified by absence of the PAX9 717 G/C single nucleotide polymorphism marker in the father (-) and its transmission to his daughter. The pairs of bars below each person represent his or her paternal (blue) and maternal (pink) chromosome 14 homologues. DNA marker names are on the left of the pedigree, and their alleles are shown next to their respective chromosomes.

In an ideal world, all diseases would follow a mendelian pattern of inheritance and in each instance show linkage to a single mutation in a single causative gene ("monogenic"). Unfortunately, diseases such as diabetes,^32,33 coronary heart disease,³⁴ macular degeneration³⁵ and many craniofacial anomalies such as isolated cleft lip/palate³⁶ and temperomandibular disorder³⁷ are determined by changes in more than one gene and are referred to as complex ("polygenic") disorders. While they may cluster in families, they do not follow the same predicted pattern of inheritance seen in autosomal or X-linked dominant and recessive disorders. Sometimes changes in these genes must be in combination with certain environmental factors to cause disease, such as exposure to certain chemicals or medications or maybe even diet. This type of inheritance often is referred to as "multifactorial" because many factors—genetic, environmental or both—are involved. The close relatives of a person with a complex disorder have a higher chance of developing the disorder on exposure to the environmental trigger than do the close relatives of a person who does not have the disorder.

While the identification of cosegregating markers of complex disease traits remains the same as for those of a mendelian disease,³⁸ the association of the disease with multiple loci makes the association of identified mutations with the disease more complicated. That is because each mutation is likely to follow a Gaussian distribution in the population, as it is itself not deleterious. It, therefore, will be present in unaffected people who lack other requisite factors for presentation of the disease. Gene identification for such disorders can use the same family-based linkage approaches that are used for mendelian disorders, albeit requiring a large number of families, but with modifications in the computer algorithms used for the analysis that compensate for the fact that no specific model of inheritance is specified.³⁹ Other approaches include the use of affected sibling pairs⁴⁰ or discordant sibling pairs,⁴¹ trios comprised of parents and the affected subject and the data analyzed by the transmission disequilibrium test,⁴² association studies that compare the frequency of specific alleles or alternative forms of a candidate gene between affected cases and age- and race-matched controls.⁴³ The sample sizes needed for these studies, which can be as many as several hundred, are determined by the recurrence risk ratio of disease in relatives of an index case. The number of disease genes identified by these approaches is much lower than those identified by traditional positional cloning approaches using large families segregating mendelian disorders, largely because of the fact that the effect of individual genes on the phenotype, though important, is small and challenging to detect.

Methodologies before and since the HGP. While the basic methodology of linkage analysis has not changed much between the eras before and since the start of the HGP, the stages taken to achieve the steps have. Figure 3 compares the time investment required at the different stages.

View larger version (54K): [in this window] [in a new window]   Figure 3. Time scales of positional cloning in the eras before and since the start of the U.S. Department of Energy and the National Institutes of Health’s Human Genome Project (HGP). The timeline for the pre-HGP era was generated from the identification of the Friedreich’s ataxia gene collated from tens of publications during that 13-year period. The timeline for the post-HGP era was generated from the identification of the PAX9 gene in a nonsyndromic hypodontia family collated from a single publication after those two months. However, if there are no obvious candidate genes, this process could take much longer. Yellow boxes show methodologies common to both pre- and post-HGP eras, red boxes show those specific to the pre-HGP era and orange boxes show those specific to the post-HGP era.

Physical mapping techniques also have improved. The identification of genes within the physical map has sped up, thanks to the wealth of data produced by the sequencing projects of "model" organisms such as the nematode worm,⁴⁵ fruit fly,⁴⁶ chicken,⁴⁷ mouse,⁴⁸ and the budding and fission yeasts.^49,50 All of these projects offer simpler systems in which to study the function of genes and to allow the application of techniques not permitted for use with humans.

The increasing power of computers and the proliferation of the Internet has allowed computer-based techniques to be developed and implemented from central databases, such as those of the National Center for Biotechnology Information. From there, scientists can access the DNA sequence and the map of the complete human genome with the location of all the known DNA markers. This allows the identification of known genes within the disease-associated region without the time-consuming sequencing and analysis of the pre-HGP era; the only requirement is the identification of linked markers by means of linkage mapping. Their coordinates then can be used to query the database, which will provide a list of genes identified in that region. This leaves only the identification of candidate genes for the scientist, who then can determine the DNA sequences of these genes to identify the causative mutations.

The identification of disease-causing genes and mutations in the post-HGP era, therefore, is much faster than it was before the HGP (Figure 3). However, there still are instances in which the identification of the disease-causing gene and associated mutation can be a time-consuming and costly undertaking. Genes listed in the Internet-based genome sequence databases still are not fully annotated, and sometimes exons are missing from the available sequence. There also are splice variants—in which a single gene encodes multiple forms of a protein that are generated by changes in the way the gene’s blueprint is processed—that, if unknown, can lead the human geneticist to draw incorrect conclusions. Some genes span large distances, making the DNA sequencing required to fully search the coding sequence and promoter region for mutations time-consuming and expensive. Finally, there commonly are dozens of genes located in the region identified by a linkage study. It can take a large investment of time and money to identify the disease-causing gene, making the rate-limiting step in these cases the funding available to the research laboratory.

Despite these problems, the rate-limiting step in modern disease research remains the recruitment of sufficiently large test populations within which to identify disease-causing genes. The dental professionals who have patients with dental anomalies play an important role in this gene discovery process by means of their description and recruitment of subjects for research studies.

	MOLAR HYPODONTIA: AN EXAMPLE OF POSITIONAL CLONING

TOP ABSTRACT THE BASICS OF HUMAN... CLONING OF DISEASE GENES:... MOLAR HYPODONTIA: AN EXAMPLE... CONCLUSION REFERENCES

 
One of the authors (P.I.P.) and collaborators from the University of Texas Dental Branch at Houston studied a large family segregating autosomal dominant oligodontia involving primarily molars.^19,51 The 13-year-old index case first came to attention when he sought correction of a 3-millimeter maxillary central incisor diastema.⁵¹ An astute orthodontics resident noted that the proband and his two brothers, who also were enrolled for orthodontic treatment in the same clinic, all had several developmentally absent permanent teeth. The dentist, working with a geneticist (P.I.P.), drew a family pedigree that demonstrated a clear autosomal dominant mode of inheritance (Figure 2A). They further extended the pedigree with data from a genealogical search that allowed the tracing of predecessors to the year 1645, when the first ancestor migrated to the United States from the United Kingdom.⁵¹ The most severely affected family member lacked 18 permanent teeth, including all maxillary and mandibular first, second and third molars; all maxillary and mandibular second premolars; and all mandibular central incisors. The least affected member lacked all maxillary and mandibular second and third molars, all maxillary first molars and first premolars, and all mandibular central incisors.

A literature search determined that this type of familial hypodontia had not been reported previously. The researchers conducted interviews with the mother and the grandmother of the proband, and they verified information the mother and grandmother provided on the affection status of living relatives via telephone interviews with family members, their dentists or both and via examination of clinical records, when available.

Forty-three members of the family, of whom 19 were affected, enrolled in the study, and each provided a blood sample ( 8 cubic centimeters). The researchers extracted DNA from the leucocytes and used it to identify the gene underlying molar hypodontia. They obtained genotypes or "bar codes" at 350 equally spaced locations in the genome and analyzed the data using computer algorithms that determined which DNA markers were cosegregating with hypodontia in the family members. This exercise determined that the gene localized to an approximately 25 million base-pair interval on the long arm of chromosome 14 (14q21). Database searches determined that the gene PAX9 previously had been mapped to this region, and it also was known that inactivation of this gene in mice resulted in the absence of teeth.⁵² Studies of the expression of the PAX9 gene in the developing mouse had demonstrated that wide expression of PAX9 is seen in the neural crest-derived mesenchyme that develops into craniofacial structures, including teeth.⁵³ Thus, PAX9 was a strong candidate for the phenotype, and so the researchers subjected all protein-coding regions (exons) of the gene to DNA sequencing.

It may be reasonable to assume that insufficient levels of PAX9 or functionally inactive PAX9 result in hypodontia of posterior teeth.

Sequencing of exon 2 of PAX9 revealed an insertion of a G nucleotide at position 218.¹⁹ This insertion resulted in altering the sequence of the protein beyond the point of insertion and truncating the protein by 25 amino acid residues. The researchers found the mutation in all affected members of the family, and screening of 150 unrelated control subjects of the same ethnic background did not reveal the mutation, thus providing strong evidence that the mutation the researchers identified within PAX9 was the cause of the molar hypodontia in the family.

Supporting evidence for the etiologic role of a gene in a disorder is best provided by identification of a mutation within the same gene in an unrelated family with a similar phenotype (Figure 2B). A small nuclear family in which the father and the daughter had severe hypodontia that included primary and permanent molars was referred to the research team.³¹ Genotyping of DNA markers flanking the PAX9 gene suggested that the PAX9 locus could be involved. However, since the family was small, obtaining statistically significant evidence for linkage to PAX9 was not feasible. However, the researchers sequenced the PAX9 gene in both affected people and found evidence suggesting a deletion of the PAX9 gene (Figure 2B). The deletion, which they further confirmed by several molecular approaches, was shown to be greater than 57,000 base pairs.

One of the authors (P.I.P.) and others have reported several other categories of mutations within PAX9 in families with hypodontia, including "missense" mutations with amino acid substitutions,³¹ a "nonsense" mutation that results in a stop codon instead of an amino acid⁵⁴ and a 288–base pair insertion within exon 2 that resulted in a truncated protein.⁵⁵

The mode of action of the submicroscopic deletion involving PAX9 is due to haploinsufficiency for PAX9.³¹ The functional effects of the other categories of mutations are less obvious, but given the phenotypic similarities between the patients with the large deletion and those with the other types of mutations, it may be reasonable to assume that insufficient levels of PAX9 or functionally inactive PAX9 result in hypodontia of posterior teeth. The frame-shift, nonsense and insertion mutations are predicted to result in proteins lacking part or all of the paired domain, which is an important functional domain of the protein, and might, therefore, be expected to cause a loss of function. The missense mutations result in amino acid substitutions in the paired domain. It is possible that these proteins, whose synthesis and stability are likely not affected, fail to recognize the target DNA sequence, which results in functional haploinsufficiency. The PAX9 protein produced from the PAX9 gene with the insertion of a G nucleotide at position 218¹⁹ already has been shown to exhibit reduced DNA-binding efficiency,⁵⁶ which supports functional haploinsufficiency as the cause of this hypodontia, as well as inferring it for others. The other possibility is that these mutant PAX9 proteins may recognize new targets and result in the gain of novel functions. Given the overall phenotypic similarities of these patients to those bearing the other classes of mutations, the former possibility most likely is the mechanism of action of the missense mutations.

While patients with PAX9 mutations typically lack at least six or more molars, there is considerable intrafamilial variability in the identity of the teeth missing among affected members of the families within which various PAX9 mutations are reported to be segregating, and the molecular basis of this is largely unknown. Further research is required to fully elucidate the mechanism by which the disruption of the function of PAX9 results in hypodontia.

	CONCLUSION

TOP ABSTRACT THE BASICS OF HUMAN... CLONING OF DISEASE GENES:... MOLAR HYPODONTIA: AN EXAMPLE... CONCLUSION REFERENCES

 
The HGP has had a massive impact on the identification of the causes of inherited anomalies. Molecular genetic analysis that used to take months or years to complete now can be achieved in weeks or months, which makes the identification, diagnosis and reporting of cases of interest the rate-limiting step in the identification of the underlying genes. Whereas in the past, research was limited to the area near the research institution owing to poor communication and transportation, the modern age has brought electronic communication and rapid transportation that allows researchers to use the whole world in their investigations. This, therefore, presents the opportunity to compare the causes of inherited anomalies in different populations.

To achieve this for dental anomalies, a large number of families with these conditions need to be identified and recruited. The dental professional can facilitate this effort by collaborating with geneticists by referring families with these inherited anomalies for analysis. This will enable geneticists to identify the underlying genetic causes and further our knowledge of human dental development. It also will offer hope for novel therapeutic regiments in the future so that we can move from a mode of diagnosis and treatment to one of prediction and prevention.