Mutation or Polymorphism?
Genetic changes are frequent. New alleles arise at an estimated rate of approximately 70 per diploid human genome per generation. Most of these genetic changes are functionally irrelevant, but some may have minor effects on the phenotype or, occasionally, may be of such serious consequence that the conception will result in a miscarriage rather than a live birth.
Complex Problems—Simple Answers?
The variants in the human genetic material are as manifold as their consequences. One of the big challenges in molecular genetic diagnostics is the correct interpretation of novel variants in a patient’s genetic sequence, especially in cases of point mutations where a different amino acid is predicted to occur in the coded protein (“coding nonsynonymous”), or where there is (seemingly) no influence on the protein sequence. Is it a serious “mutation” or a silent variant?
The human genome is too complex to permit simple answers. Mutations in the promoter regions, in splice sites, within introns, or in the untranslated regions may (or may not) lead to changes in gene expression or the posttranslational modification of gene products. In comparing a random piece of human DNA, one finds that two individuals (or rather the homologous chromosomes of the diploid genome) differ in their sequence at every 1,000th base.
Therefore, a sequence of 3 billion base pairs would have 3 million base-pair differences between two individuals! When limiting the comparison to coding sequences, 1 in 2,500 bases varies between two individuals (due to the fact that these variants are more likely to have functional effects and to be subject to selection).
A Question of Frequency?
In determining the functional effects of a particular genetic variant, it is useful to consider its frequency within a population. A rare variant found in a person (or several persons) with a specific clinical abnormality but not in control individuals is more likely to be disease causing than a variant that is frequently found in the general population. Nevertheless, a novel variant identified in a patient may be clinically irrelevant even if it has never been observed in hundreds of other persons.
Determining the effect of “unclassified variants”, also called variants of uncertain significance or varicents of unknown significance (VUS) is a particular challenge in the dominantly inherited cancer predisposition syndromes such as inherited breast and ovarian cancer (Chapter 32.3). More than 1,000 disease-causing mutations in the BRCA1 and BRCA2 genes have been identified so far, but there are also many VUSs whose pathogenetic relevance is uncertain.
This can present a challenge in genetic counseling since it cannot be ruled out that they are disease causing. Nevertheless, predictive mutation analysis would not be offered to a family based on the detection of such a VUS.
Attempting a Definition
Mutation or polymorphism? These two terms are used inconsistently even by geneticists. Mutation derives from the Latin mutatio, which means “alteration.” It can mean both the occurrence of a change in the genetic information and the resultant change itself. The term is primarily unrelated to the functional effect of a genetic alteration, even though in general practice it is mostly used for rare disease-causing variants. Sequence alterations without functional effects are often called polymorphisms, but again this definition, based on functional criteria, may be problematic in some instances.
How can one be sure that a specific sequence variant has no functional effect? What about frequent variants that do have functional relevance yet rarely or never trigger a disease or serve as common risk factors for a disease? The blood groups of the ABO system, for example, are due to “mutations,” but most people would not use this term to describe carriers of blood group O.
Most geneticists recommend that the term polymorphism be defined according to frequency.
Polymorphisms are genetic variants where the rarer allele in a population occurs with a frequency of at least 1%, independent of the functional or pathogenetic relevance of this alteration. There is nothing wrong with using the term mutation colloquially for rare DNA alterations that cause disease, but one should be aware of the gray zones of such a definition. When in doubt, it is advisable to use the neutral term variant for all deviations from the wild-type sequence (the standard “normal sequence”).
A genetic variant where the rarer allele in a population occurs with a frequency of at least 1%, independent of the functional or pathogenetic relevance of this alteration.
Single nucleotide polymorphism (SNP):
A polymorphism where the alleles vary within one single nucleotide base.
The occurrence of a change in the genomic sequence, or the resultant change itself. The term is generally used for disease-causing genetic variants.
One of two or more variants of a gene; the term may be used in connection with various denominators such as a particular DNA sequence (e.g., polymorphism), a combination of genetic markers (haplotype), functional characteristics (wild type vs. disease causing), or parental origin (maternal or paternal).
Allele is another term that is often used imprecisely. It derives from the Greek word allelon, which means “belonging together.” It designates a specific version or variant of a gene or a chromosome segment that is different from another allele (or several other alleles). This can be in reference to a mere DNA sequence at the same locus, but the term can also refer to a combination of genetic markers (haplotype) or to functional aspects (normal allele vs. disease allele).
Homologous chromosomes contain corresponding gene copies. If a person is homozygous for a particular gene sequence in all its characteristics and base positions, then this person has two copies of the same allele of this gene. Stated differently, this person carries the same allele on both chromosomes, and therefore is homozygous for that allele.
At the same time, it is okay to refer to the alleles as maternal and paternal in this situation, since the origin of the chromosomal strand may be a valid distinguishing feature even if it is the same allele with regard to the DNA sequence. The use of the term allele thus depends on the distinguishing feature that one likes to emphasize.
Single Nucleotide Polymorphisms
More than 90% of all variations in human DNA involve exchanges of single nucleotide bases. These are called single nucleotide polymorphisms (SNPs). By definition, these are loci of the human genome where the alleles vary within one single nucleotide base, and where the rarer allele occurs with a frequency of at least 1% in the population (as stated previously).
SNPs are found in coding as well as noncoding regions of the human genome. In the coding regions they are called cSNPs. In the noncoding regions, as a rule, one such polymorphism occurs at every 1,000th nucleotide of the human genome. Most of the known SNPs differentiate humans from primates and are found worldwide in people of all populations.
In the course of the Human Genome Project, SNPs became a central tool for genetic analyses, ranging from the genetics of complex diseases to pharmacogenetics and population genetics. Because of their low mutation rate (compared to other polymorphisms), great stability, and high frequency in the entire genome, SNPs are especially suited for association analyses in the identification of complex diseases. They can also easily be used for familial segregation analyses.
Types of Mutations
Genetic disorders can be caused by numerous types of mutations. Generally speaking, there are three groups:
- Numerical chromosome abnormalities
- Structural chromosome abnormalities
- Gene mutations that alter the function of individual genes
Chromosome abnormalities occur as congenital abnormalities in all or most body cells, or as somatically acquired clonal abnormalities that play an important pathogenetic role (e.g., in the formation or progression of various tumors). Our perspective on what we consider “chromosomal abnormalities” has changed considerably over the past 5 to 10 years. Originally, chromosomal anomalies were those that could be detected by light microscopy.
The development of molecular cytogenetic techniques enabled the detection of submicroscopic deletions that encompass many genes and have typical characteristics of “chromosomal disorders,” even though specific features may be due to haploinsufficiency of specific genes. The recent advent of array technology (Chapter 15.3) allows for the detection of smaller and smaller copy number variants (CNVs), thereby bridging and blurring the gap between “chromosomal” and “monogenic” conditions.
We now know that most individuals (65% to 85% of the population) harbor a CNV of at least 100 kb of DNA. Much larger (exceeding 500 kb) variants occur in 5% to 10% of individuals, while at least 1% of the population carries a CNV exceeding 1 Mb. While some of these CNVs are clearly pathogenic, others may represent risk modifiers for common diseases or may be completely benign.
Numerical chromosome abnormalities stem from errors that occur during the segregation of intact chromosomes in meiosis or mitosis. They are always clinically evident, and as a rule, they are de novo (i.e., not inherited). By contrast, structural chromosome abnormalities can be balanced (without loss or gain of genetic material), in which case they are typically not evident clinically.
Structural chromosome abnormalities can be passed on from generation to generation. Among structural chromosome abnormalities are duplications, deletions, inversions, and translocations. Larger chromosome abnormalities are demonstrable by light microscopic chromosome analysis (cytogenetically, typical resolution is down to approximately 4 Mb); smaller deletions and duplications are demonstrable by molecular cytogenetic or DNA analytical methods. Gene mutations can occur in many different forms and are identified through a multitude of molecular genetic methods.
Gene mutations occur within the process of DNA replication, either spontaneously or through specific physical or chemical agents (mutagens). The vast majority of gene mutations are identified and corrected by efficient mechanisms for DNA repair. Constitutional mutations are found in all cells of the body and are present in the zygote, while somatic mutations are present in only a proportion of cells and arise after fertilization.
Frequency of Individual Mutation Types
Numerical chromosome abnormalities represent some of the most frequent genetic disorders. Overall, 10% of all spermatozoa and 25% of all oocytes are estimated to be aneuploid. The average frequency of numerical chromosome abnormalities in newborns is 1 in 400. This, however, is only the tip of the iceberg, since most chromosomal abnormalities affect fetal development so severely that they result in spontaneous miscarriage.
Approximately one-sixth of clinically recognized pregnancies are spontaneously aborted. Approximately 50% of all spontaneous abortions in the first trimester are caused by chromosomal anomalies. These figures do not include the large number of pregnancies that went unnoticed because the loss of the embryo occurred soon after conception. Chromosome abnormalities occur in somatic cells as a result of faulty mitotic cell division. Especially frequent are aneuploid cells in malignant tumors.
Structural chromosome abnormalities are less frequent than numerical abnormalities; they are estimated to occur at a rate of 1 chromosomal rearrangement per 1,700 cell divisions. Among structural abnormalities, deletions are the most frequently observed, followed by insertions and duplications.
This is partly due to the fact that deletions used to be more readily identifiable by molecular cytogenetics (fluorescence in situ hybridization [FISH] analysis) than duplications; also, the clinical consequences of a missing chromosome region (monosomy) are usually more severe than those of a duplication (trisomy), which means that deletion patients are more likely to come to our clinical attention as compared to duplication patients.
As far as gene mutations are concerned, we are also just seeing the tip of the iceberg. The vast majority of replication errors are promptly recognized by the human cell’s excellent DNA repair mechanism, excised from the DNA, and repaired. DNA polymerase has this proofreading function. It initially recognizes the faulty strand of the newly synthesized DNA double helix, excises the wrong base, replaces it with the correct complementary nucleotide base, and takes care of re-ligating the fragments.
DNA polymerase works so efficiently that despite a speed of 20 base pairs per second, only 1 in 10 million bases is inserted incorrectly. Of the initial replication errors, 99.9% are identified and repaired, so that the total mutation rate due to replication errors amounts to 1 in 1010 base pairs. In theory, this means that less than 1 replication error remains as a de novo mutation for the 6 × 109 base pairs of the diploid human genome per cell division.
To have a significant impact, this mutation would have to be clinically relevant, either as a somatic mutation (e.g., in cancer cells) or as a germline mutation with implications for the following generation. The effect of exogenous agents, such as cigarette smoking, is not part of this calculation. Among the known gene mutations, missense mutations are the most frequent, followed by nonsense mutations and splice mutations.
Numerical Chromosome Abnormalities
Changes in the total number of chromosomes (aneuploidy) are called numerical chromosome abnormalities. They arise from faulty segregation (nondisjunction) of the chromosomes in meiosis or mitosis. Meiotic failure of a pair of chromosomes to disjoin occurs either in the first meiotic division (i.e., in 75% of cases two homologous chromosomes move together to one cell pole instead of to opposing cell poles) or in the second meiotic division (i.e., in 25% of cases two sister chromatids of a homologous chromosome reach the same cell pole).
Examinations of aneuploid fetuses have shown that in greater than 90% of cases the aneuploidy resulted from meiotic nondisjunction in the mother. The risk of numerical chromosome abnormalities is largely dependent on maternal age. While the risk of giving birth to a child with an abnormal chromosome count is less than 0.2% for mothers up to age 30 years, it increases to greater than 0.5% at 35 years, and to 1% at 38 years.
At the time of delivery, a 45-year-old mother has a risk of 1 in 20 for the newborn to have a chromosomal abnormality—which also translates to a 95% chance of the child having a normal chromosomal complement. It should be remembered that more chromosome abnormalities are diagnosed at the time a chorionic villus biopsy or amniocentesis is performed, since many aneuploidies result in miscarriages or fetal death as late as the second or even third trimester.
Faulty segregation during mitosis only occurs in a portion of the cells of an organism. This creates a mosaic of normal cells and cells with chromosome abnormalities, called a somatic mosaic if the germ line is not affected. Nondisjunction of chromosomes during the generation of germ cells or in meiosis, however, results in chromosome abnormalities that affect all cells of the offspring.
Not only individual chromosomes but also entire chromosome complements can occur in abnormal numbers. This is called polyploidy. Triploid (3n) and tetraploid (4n) chromosome complements have been observed during prenatal chromosome analyses. Some children with triploidy are born alive but usually die within the first few days. The most frequent cause of triploidy is fertilization of an oocyte by two spermatozoa (dispermy).
Less frequent are oocytes or spermatozoa with a diploid chromosome complement. Tetraploidies always have the karyotype 92,XXXX or 92,XXYY, which suggests that they originate in normal zygotes that subsequently went through an incomplete division during the first meiotic division.
Only three numerical chromosome abnormalities involving autosomes are viable after birth: trisomies 13 (see p. 187), 18 (see p. 185), and 21 (see p. 180). Exceptions are partial trisomies or mosaics of trisomic and normal cell lines. Monosomies of autosomal chromosomes result in a very early spontaneous miscarriage, often before the pregnancy is identified. Numerical abnormalities of the human sex chromosomes cause relatively mild phenotypes, unlike numerical abnormalities of the autosomes (see pp. 189).
The human cell utilizes mostly the genetic information of one X chromosome and, therefore, can largely compensate for the loss of a Y chromosome or the second X chromosome, as well as additional X chromosomes (by way of X inactivation, as described in Chapter 2.8). As a result, numerical sex chromosome abnormalities in newborns rarely lead to severe physical changes and, in most instances, do not cause substantial intellectual disability. Monosomy X is the only viable monosomy in humans.
A normal diploid chromosome complement (2n) of 46 chromosomes (46,XX or 46,XY).
The number of chromosomes varies from that of a normal diploid chromosome sequence (e.g., 47,XY,+21).
Multiples of the haploid chromosome complement of greater than 2n.
Triple chromosome complement (3n) (e.g., 69,XXX).
- Digynic triploidy:
The origin of the additional set of chromosomes is maternal. This results in a partial mole.
- – Diandric triploidy:
The origin of the additional set of chromosomes is paternal.
A quadruple chromosome complement.
Three copies of a chromosome.
One copy of a chromosome.