In earlier chapters, we saw how genes produce the enzymes that control specific biochemical reactions. Normal development depends on these information coding and regulatory systems working properly. But DNA replication is not perfect. Biochemical mistakes happen. Most replication errors are corrected by repair enzymes, but those that are missed become new mutations. In the broadest sense, then, a mutation is a heritable genetic change passed from one cell to another. For that reason, the biochemical correction mechanisms that work in parallel with replication are important for biological continuity.
Unfortunately, repair systems themselves can mutate, as in a patient with xeroderma pigmentosum (Figure 7-1). These patients have an increased mutation rate, as seen for example in higher rates of skin cancer, because of an inability to repair genetic damage caused by ultraviolet radiation. Other kinds of mutation repair deficiency are also known. Clearly, mutation rate is not a mathematical constant. Mutation rates can change.
Figure 7-1. An individual showing the effects of xeroderma pigmentosum, a defect in one of the genes involved in the process of nucleotide excision repair. Individuals with this condition are unable to repair UV-induced mutations, which gives them a predisposition to skin cancer and related problems. (Reprinted with permission from Rünger TM, DiGiovanna JJ, Kraemer KH: Chapter 139. Hereditary Disorders of Genome Instability and DNA Repair. In: Goldsmith LA, Katz SI, Gilchrest BA, Paller AS, Leffell DJ, Dallas NA, eds. Fitzpatrick’s Dermatology in General Medicine. 8th ed. New York: McGraw-Hill; 2012.)
Consider the following question. Would it be good if the process of mutation could be reduced? Would it be better if no new genetic changes ever occurred again? At first thought, the logical answer would probably be “yes.” Eliminating mutation would be good. When you consider the many who must deal with developmental or physiological disabilities due to harmful mutations, it is easy to see the negative side of the process. But the environment in which we live is not constant. We continue to face new biologic challenges. Exposure to novel disease pathogens is only one obvious example. Physiological processes allow us to respond to changes in the environment. But genetic diversity adds another mechanism of response. In theory, therefore, the genetic diversity created by mutation may be fundamentally good—at least for the long-term survival of the species.
One way to think about this question is to consider biochemical pathways controlled by proteins under allosteric regulation. Binding with cofactors can change protein conformation within limits affected by environmental variables like temperature. A heterozygote for a key regulatory step can produce alternate protein forms with slightly different temperature optima. For that reason, a heterozygote is better able to handle the range of environmental conditions it naturally encounters. This can lead to the establishment of a polymorphism (literally “poly,” many or multiple; “morph,” forms). In such situations, there is not just one single form that is truly the “normal” or “wild type,” since different forms may be best for different geographical conditions or seasons. Without mutation to generate new variation, the diversity of a gene pool will deteriorate. In fact, many species are on the verge of extinction, because their population size is so small that rare mutation can no longer replace the genetic diversity eroded by random processes.
Mutation is fundamentally intrinsic to the imperfect mechanism of DNA replication. But can the spontaneous mutation rate change? It might surprise you to learn that the answer is “yes,” an answer that would have surprised most geneticists even a short time ago. The biochemistry of DNA replication is complex. Errors occur. We have already mentioned the important role that genetic repair enzymes play in correcting some of these errors. But in addition to loss of function by mutation, as in xeroderma pigmentosum, our repair mechanisms themselves can vary in efficiency so mutation rates change. Experimental evidence comes from many sources, including the improved efficiency in DNA repair found in organisms that live in highly mutagenic environments like the radioactive waste near uranium mines. Thus, mutation rate can be modified to some degree by influencing repair efficiency.
Clearly, there is a long-term benefit to the population from producing a low background level of genetic variation by mutation. But typically our medical focus is on the harmful consequence of mutations for a patient. In this chapter we will explore some of the common molecular mechanisms that lead to heritable genetic change.
Part 1: Background and Systems Integration
Types of Mutation
Mutations are heritable genetic changes. They occur both in the nuclear genome and in the DNA of organelles like mitochondria. Mutations of chromosome structure have already been discussed in Chapter 5, so our focus here will be on point mutations, that is changes at the base pair level within a gene or its control regions (Table 7-1). Such mutations can be viewed from different perspectives, like the type of nucleotide change or the effect of that change on coding outcomes.
Table 7-1. Types of Point Mutations
There are several ways to change a nucleotide pair (Figure 7-2), so that one base is substituted for another. A transition mutation occurs when one purine is replaced by the other purine (A ↔ G), or one pyrimidine is replaced by the other pyrimidine (T ↔ C). In a transversion mutation, on the other hand, a purine is replaced with a pyrimidine, or vice versa. Because of the degeneracy of the genetic code we discussed earlier, a transition mutation at the third position may often have no effect on the amino acid that is incorporated into the final protein. Such a mutation can be called a silent mutation. In contrast, transversion mutations usually cause a change in the amino acid. Since any change is more likely to be bad than it is to be good, transversion mutations are less likely to remain long in the population. A base change that translates as a change in an amino acid is called a missense mutation. They change the encoded information. A base change that results in a stop codon (UAA, UAG, or UGA) is called a nonsense mutation.
Figure 7-2. Transition mutations involve substituting one purine for the other, or one pyrimide for the other. Transversion mutations occur when a purine is substituted for a pyrimidine or vice versa.
A base deletion or base addition involves the loss or gain, respectively, of one or more nucleotides. If the affected number of nucleotides is not divisible by three (in other words, if it does not involve one or more complete codons), the triplet reading frame is shifted for the remainder of gene translation. Such a frameshift mutation can dramatically alter the amino acid makeup of the protein. It also often results in a stop triplet (a nonsense mutation) soon after the point of the frameshift and leads to early termination of translation.
Mutations that change an amino acid in a protein can differ in the impact they have on development. At the least severe end of the spectrum, the amino acid change does not alter the protein’s function in any important way. A change from one small hydrophilic amino acid to another in a non-active part of the protein is a real change, but it is unlikely to have much impact. That can be described as a neutral mutation. Sometimes the change is expressed under some conditions, but not others; its expression is conditional. The altered protein might change in the way it works in different temperatures or different chemical environments. The mutations that have medical significance, however, generally alter biological function in critical ways. They are classed as deleterious mutations. At the most extreme are those that affect such a critical aspect of biological function that life cannot continue. Lethal mutations cause death at a point in development where their function becomes crucial. Many act very early in development, but some act fairly late or even into adulthood.
As we saw in our earlier discussions of gene expression, a developmental step often involves the interaction of more than one protein. Occasionally a second mutation can balance, and thus essentially “repair,” the effect of the first. Compensating, or suppressor, mutations can even occur within the same gene. The combination of potential changes is almost endless. For that reason, one must keep an open mind to the events that might explain a particular case. The relationship between genes and phenotypes is complex.
Frequency versus Rate
The terms “mutation frequency” and “mutation rate” are often used interchangeably. That is a mistake. They refer to two distinctly different things. Mutation frequency refers to the proportion of alleles of a given type in the gene pool. One might say, for example, that the frequency of melanism (black color) in Malasian leopards is almost 50%. Depending on the context, a statement like this could refer either to the proportion of alleles of that type in the gene pool or how often the trait is seen in members of the population. Typically, the reference will be clear from the context. Mutation rate, on the other hand, refers to the rate at which new mutant alleles are formed, and it is reported at different scales. One can talk about the mutation rate per nucleotide, per gene, per gamete, per generation, or some other molecular level or developmental scale. Thus, mutation frequency is a measure of existing genetic diversity and is generally easier to measure than mutation rate. The effects of selection pressure, population size, magnitude of developmental effect, and other factors mean that mutation frequency and mutation rate are not necessarily even correlated.
The first direct measure of a human mutation rate was based on DNA sequence comparisons from the human Y chromosome: one mutation per 30 million base pairs (3.3 × 10-8 mutations/bp). This is in the same range as estimates from more traditional surveys of mutant phenotypes. Although different genes have slightly different mutation rates that can range in magnitude because of factors like gene size (number of base pairs that can change), a typical average rate is about 1 × 10-5 to 1 × 10-6 mutations per gene per generation. In addition to gene size, the measurement of mutation rate can be affected by factors like mutational hotspots, which are areas of increased mutational activity. But even these cases do not paint a complete picture of the difficulties encountered in mutation detection. Silent, neutral, and conditional mutations, by definition, are not clinically obvious. Mutation tests are not done on people who meet the definition of being normal or healthy. Thus, the detection of mutant phenotypes generally uncovers only a portion of the real underlying genetic diversity of the gene pool.
Sample estimates of mutation rate in humans include achondroplasia, 4.2 to 14.3 × 10-5; aniridia, 0.5 × 10-5; and retinoblastoma, 1.2 to 2.3 × 10-5. In contrast, the rate for the dystrophin gene, which is mutated in Duchenne and Becker muscular dystrophies, is as high as about 1 × 10-4. This high rate is not surprising when one realizes this is one of the largest genes known.
Several factors complicate the estimation of mutation rates. For example, retinoblastoma has a penetrance of 80% or less, so there is a problem of ascertainment involved in its detection. When a mutation is not expressed in some individuals, it may only be identified in pedigrees. Other altered phenotypes, indeed perhaps most of them, are due to several different loci with similar phenotypic expressions. For that reason, data about which specific gene has mutated may be unreliable without molecular support. But the range of rates cited here will at least give us a working basis on which to discuss predictive models.
There is another interesting factor that can affect mutation. Its rate may not be the same in both sexes. Indeed, this is known to be true in experimental organisms like Drosophila. But in humans, it is difficult to measure mutation rate accurately in most parts of the genome. Mutations are rare and, especially if they are recessive, they may be hard to detect. Yet, careful studies of the pseudoautosomal region shared by the human X and Y chromosomes show that mutation rate may be higher in males than in females.
Before passing from this general discussion of mutation rate and frequency into more specific aspects of genetic instability, we should take another look at the concept of mutation rate. Superficially, this is a straightforward measurement of the occurrence of a new mutation. But this seemingly simple concept is actually much more complex. One can begin to appreciate this by considering the genetic events that can occur between the establishment of the germ-line cells in embryonic testes and ovaries and the eventual gametes produced in the sexually mature adult (Figure 7-3). If we think for a moment about mutation as a genetic error, there are actually three different phases to consider. The literature sometimes confuses things by equating them. But they are definitely not the same.
Figure 7-3. Three different levels at which one can assess mutation: the “genetic damage rate” during DNA replication has many errors, but most are repaired; the unrepaired changes are the measured as the “mutational event rate”; and the changes that actually enter the gene pool are represented in the classical “mutation rate,” which can include multiple identical copies of an earlier genetic change, a cluster, due to mitotic replication in the germ cell line. (Reprinted from Thompson et al: Environ. Molec. Mutagenesis 1998;32:292-300).
First, the genetic damage rate is the rate at which errors occur during the process of DNA replication. Mispairing of nucleotides and single- and double-strand DNA breaks are very common. But efficient repair mechanisms identify and correct most of the errors. The mutational event rate measures what is left after repair has taken place. Many people incorrectly assume that this is the rate of new mutation that is estimated experimentally from population data and pedigrees. But in fact, a new mutation occurring early in the growth of testis and ovary tissue (i.e., long before meiosis begins) can be replicated during cell division so that many cells in the growing gonad carry the same new mutation. Then, when meiosis finally begins, many eggs or sperm may carry a copy of the same original new mutation (gonadal mosaicism). Mutation rate is typically measured by counting the number of new members of a population that carry a particular genetic change. The measured mutation rate will, therefore, be higher if a single early premeiotic mutation event has been duplicated into a number of gametes yielding a cluster of related individuals carrying the same new mutation. Clusters of mutation have now been documented in humans and in many other organisms. Thus, the genetic damage rate, the mutational event rate, and the traditional mutation rate can be very different. They vary independently as a function of mutation-generating factors and repair efficiency.
What Is a Polymorphism?
We highlight this question, because there is an important difference in the way the concept of polymorphism is interpreted in specialty areas of genetics. Being aware of this can help avoid serious confusion. The term “polymorphism” was first used to describe genetic diversity studied by population geneticists. In that field, by definition, a polymorphism occurs when the most common allele is at a frequency of less than 99%. That cut-off value is important. Mathematically, it is unlikely for a mutant allele to have a frequency of 1% or more unless it has some advantage under some environmental conditions. When a population geneticist finds a polymorphism, the next question is “What advantage keeps the rare form so common?”
On the other hand, in a clinical setting, the term is used in a more general way. There, “polymorphism” can refer to any gene or nucleotide in which more than one form is found in the population. Frequency is not important. In a clinical laboratory report, an identifiable genetic change from the accepted standard sequence will be defined as a “benign polymorphism” having no clinical relevance, a “mutation,” or a “change of unknown clinical significance.” The focus is on individual patients, rather than on the population or gene pool. The takeaway lesson is, therefore, to be aware of the context of usage and not be confused if different sources use the term in ways that appear inconsistent.
Measurement of Mutation Rate in Model Organisms
It is difficult to measure mutation rate directly in humans. Preexisting genetic diversity in our gene pool and the inability to carry out definitive experimental matings are limiting factors, as is the length of our life cycle. Obviously, one cannot argue such an experimental approach seriously. It is not surprisingly, therefore, that most of our understanding of the mutation process comes from studies in model organisms. To illustrate the approaches that are possible in the large field of mutagenesis, we will explore the logic behind three classical methods: the Ames test with bacteria, the X-linked lethal assay in Drosophila that led to a Nobel Prize for H.J. Muller, and the seven-locus test in mice. The key to an effective assay system is, of course, that a lot of data can be collected efficiently and inexpensively. Even enhanced mutation rates are rare phenomena.
The Ames test (Figure 7-4) uses back-mutation in bacteria to assess the mutagenic effect of a chemical treatment. A test chemical is added to a culture of bacteria that have a nutritional deficiency, such as the inability to synthesize histidine. The purpose of the screen is to measure back-mutation from the abnormal to a normal gene function. How well can treated mutant bacterial cells grow on medium that lacks their nutritional requirement? Given the large number of bacterial cells that can be tested for reversion, this is a very effective assay for potential mutagenic agents.
Figure 7-4. The Ames test uses a strain of the bacterium Salmonella typhimurium to test chemicals for their mutagenicity. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
Muller was awarded the Nobel Prize in 1946 for demonstrating with Drosophila that X-rays are mutagenic. Drosophila males with their one X chromosome are mated to genetically marked females, such as Basc, with X chromosomes carrying a dominant eye shape mutation and inversions that reduce crossing over. Each first-generation female is then mated to genetically marked males, and their offspring are screened for survival. If a new lethal mutation occurs on the parental male’s X chromosome during gamete formation, it will be passed to his non-Basc second generation male off-spring. A significant increase in the absence of this genetic class is confirmation that a new lethal mutation was produced by the treatment. This assay continues to be used to screen for hazardous mutagenic chemicals from industrial by-products and for other experimental questions.
The mouse seven-locus test begins with an inbred strain that carries seven easily identifiable recessive mutations. These are mated with mice treated with a potential mutagen or control condition. If a new mutation occurs at one of these seven representative genes, the offspring will be homozygous for the recessive and show the mutant trait. Although the conclusions from this mammalian assay may be most closely associated with human biology, the cost for experiments is much higher than for work with bacteria or Drosophila. Insights into the mutagenic effect of suspected chemicals are, therefore, aided by conclusions drawn from all experimental systems.
Mechanisms for Spontaneous Mutation
Several different types of chemical change can occur in DNA spontaneously. The most common is a depurination (Figure 7-5), in which either an adenine or a guanine base is removed from a nucleotide, leaving the sugar-phosphate backbone intact, and the partner nucleotide temporarily unpaired. When the DNA then replicates after such a change, any nucleotide can be inserted opposite this vacated position. Only a quarter of the times, therefore, will the originally correct nucleotide be incorporated, so depurination can lead to a mutational substitution 75% of the time.
Figure 7-5. Depurination. (a) The guanine base is released from the deoxyribose backbone leaving a site with no purine, i.e., an apurinic site. (b) In the absence of a pairing partner during DNA replication, any of the four possible nucleotides can become incorporated into the new chain. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
Another common chemical change occurs when an amino group is removed from cytosine. This deamination (Figure 7-6) produces uracil, which will then pair with adenine during DNA replication and lead to a replacement of a C-G pair with a T-A pair after a complete cycle. Often the presence of uracil in DNA will be recognized as an error and be corrected by repair enzymes. If not repaired, however, a base pair substitution mutation results. In a similar process, methylation of cytosine occurs commonly. If 5-methylcytosine is deaminated, a thymine nucleotide will result. This may be corrected less efficiently, since it is a normal base in DNA.
Figure 7-6. Spontaneous deamination. (a) Removal of the NH2–amino group yields uracil. (b) Removal of the amino group from 5-methylcytosine yields thymine. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
Tautomeric shifts are also common sources of spontaneous mutation. A tautomeric shift (Figure 7-7) involves a reversible change in hydrogen that alters the nucleotide base to a different isomer. The base pairing we assume when predicting DNA structure is altered if a base is in its rare tautomeric form. Tautomeric shifts can, therefore, cause atypical pairing of nucleotides (such as T with G, or C with A) that lead to base pair substitution mutations after one further cycle of replication.
Damage by Mutagens
Many chemicals can cause mutations by altering the structure or base pairing of DNA. The Environmental Protection Agency (EPA) screens chemical by-products of industry and other sources of potentially harmful chemicals to reduce their passage into our environment. To illustrate this large and important area of environmental health, we will look at some examples of common mutagenic agents.
Both chemical and physical agents can cause mutations. They elevate the rate of mutation above the baseline that results from normal chemical events like those described in the previous section. Furthermore, many chemicals that cause mutation are also potent carcinogens. A couple of examples will illustrate the ways these agents can work.
One way that mutagens work is by modifying the nucleotide base so it mispairs, similar to the natural mispairings that we saw can occur because of tautomeric shifts. For example, nitrous acid removes the amino group from adenine or cytosine (Figure 7-8), causing them to pair incorrectly at replication. Base analogues like 5-bromouracil (5BU) are chemicals that can become incorporated in the new DNA strand, but then undergo tautomeric shifts and mispair. 5BU, for example, undergoes a tautomeric shift at a fairly high rate so it mispairs with guanine instead of adenine (Figure 7-9). After a second round of replication, the original A-T base pair has been replaced by a G-C base pair.
Figure 7-7. The mutagen nitrous acid replaces amino groups with keto groups, thus converting cytosine into uracil or adenine into hypoxanthine. The pairing of these modified bases leads to a mutation when the strand replicates. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
Figure 7-8. The action of 5-bromouracil (5BU), an analog of thymine, as a mutagen. (a) In its keto form, 5BU pairs with adenine and no change occurs. But in its enol form, 5BU pairs with guanine. (b) When it pairs with guanine, a mutation can result after the next round of replication. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
Additions or deletions can be caused by chemicals that insert (i.e., intercalate) into a strand and distort the normal pairing between bases. This leads to a frameshift mutation that affects the product of translation. Acridine dyes cause mutations this way.
Thymine dimers were discussed earlier. They form when two adjacent thymine bases absorb increased energy from ultraviolet light and bond each other, rather than binding the matching adenines on the complementary strand (Figure 7-10). At replication these do not pair with new nucleotides and the new strand is shortened by two bases, thus shifting the reading frame.
Figure 7-9. In a thymine dimer, the energy absorbed by adjacent thymine nucleotides causes them to bind to each other, rather than to the adenine nucleotides of the complementary strand. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
Figure 7-10. In a thymine dimer, the energy absorbed by adjacent thymine nucleotides causes them to bind to each other, rather than to the adenine nucleotides of the complementary strand. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
DNA Repair
More than 100 genes are involved in DNA repair, one of the only biological molecules that routinely undergoes such a process instead of simply being replaced. Without highly efficient repair, the information content of DNA would be so degraded that the existence of complex biological systems is probably not possible. DNA repair mechanisms will be illustrated by several examples.
In some cases, an error can be repaired directly, such as having a thymine dimer repaired by breaking the bond between adjacent thymine bases. A more common mechanism is base excision repair, in which the incorrect base is removed and then replaced (Figure 7-11). For example, in many species an enzyme called N-glycosylase removes the incorrect base leaving an apurinic or apyrimidinic site. In a process called nick translation, a second enzyme, AP endonuclease, recognizes the abnormal site and makes a nick on the 5′ side. DNA polymerase then removes it and replaces it with a new nucleotide. Ligase completes the bond.
Figure 7-11. Excision repair occurs when the enzyme N-glycosylase recognizes an incorrect base and removes it from its sugar. In this example, the uracil is removed leaving an apurinic site. AP endonuclease acts on this site by breaking the 5′ connection of the sugar-phosphate DNA backbone. DNA polymerase removes nucleotides in the region and replaces them, followed by the formation of the final bond by DNA ligase. (Reprinted with permission from Brooker RJ: Genetics: Analysis & Principles, 3rd ed. New York: McGraw-Hill, 2008.)
Nucleotide excision repair can correct many different kinds of genetic damage by removing and replacing a short segment of the DNA molecule. Enzymes recognize and remove a series of nucleotides on the mismatched strand, and correct nucleotides are then inserted using the complementary sequence. These and other ways of identifying and correcting errors in the DNA molecular help minimize the number of inherited changes transmitted to the next generation. Clearly, they do not correct all errors, but their efficiency is a critical element of the stability of the DNA molecule from one generation to the next.
Transposable Elements and Trinucleotide Repeats
When recognized as a common element of the genome, transposable elements were nicknamed “jumping genes.” But mobile genetic elements were an unsettling addition to the events that influence genome organization. One of the earliest to identify this unusual genetic activity was Barbara McClintock, who received the Nobel Prize for her work on transposable elements in maize in 1983. The significance of her work went unrecognized for many years. Her story should resonate with all who pursue new knowledge. Do not be intimidated into silence by the current authorities in your field. Be open to new ideas. Be rigorous in employing good scientific standards. Transposable elements changed our understanding about genome stability. Human elements like the Alu sequence were discussed in Chapter 4. Active transposable elements are important mutation-generating factors in many species. We bring them back into our discussion here to emphasize the role that mobile genetic elements can play in generating new mutations.
Although unrelated to transposable elements, trinucleotide repeats offer another example of how elements of genome structure can affect genetic stability. Several human diseases, such as Huntington disease (CAG repeat, more than about 27 times), fragile X syndrome (CGG or GCC repeats, more than about 200 times), and myotonic muscular dystrophy (CTG repeat, more than about 200 times), have been traced to variations in repeat number. In many cases, the repeated sequence increases the number of glutamine amino acids in the translated protein, which causes the proteins to accumulate. In other cases, like the CpG islands generated in fragile X syndrome, methylation of the islands can lead to reductions in transcription.
Germinal versus Somatic Mutation
The focus of this chapter has been on heritable changes, germinal (or germ-line)mutations, passed in eggs or sperm to affect development of an offspring. But the mutational mechanisms we have discussed can also cause mutations in cells within the body well after birth. Somatic mutations may be limited to a small cell population, but they can have a life-changing effect on the carrier. Many cancer tumors, for example, show genetic differences from other cells of the body. Other examples include patches of tissue, like a spot of white hair or of differently pigmented skin, with a phenotype distinct from the rest of the body.
If the somatic mutation occurs early enough in embryonic development, the cells it produces can include germ-line tissue. In that case, a mutation that originated in early somatic cell lineages can be passed to offspring and may be indistinguishable from a normally inherited germ-line mutation. This will be discussed further in Chapter 12, Atypical Inheritance.
Paternal Age Effect
Since mutations accumulate in somatic cells, it is clear that the nuclear genetic content can gradually change as an individual ages. One special example is the accumulation of mutations in the premeiotic cell lineage of germ tissue, like the testes of human males. As mentioned in the discussion of parental age effects in a previous chapter, premeiotic mutations do not accumulate in the human female germ line, because the cells are arrested in prophase I during fetal development. But in males there are many cycles of mitotic division during which new mutations can occur before the final meiosis.
Another interesting possibility is that the phenotypic similarity between fathers and offspring may actually increase as a function of the father’s age. This may be indirectly due to reduced developmental homeostasis. The logic is a bit involved, but essentially it is proposed that, as compensating mechanisms of homeostasis are degraded by these new mutations, an extreme genotype can be expressed more accurately. They are less effectively buffered toward normal. Thus, more extreme phenotypes may appear in the offspring of older male parents. Without effective phenotypic compensation from homeostasis, these later offspring may express extreme phenotypes shared with their father that would otherwise have been at least partly masked. Thus, heritability as measured by the similarity of father and offspring increases. Heritability will be discussed in more detail in Chapter 10 on gene × environment interactions.
Genetic Variability: The Role of Mutation in Replenishing Genetic Resources
DNA biochemistry is a complex topic. Spontaneous errors in DNA replication will occur because of normal changes in the structure or bonding of nucleotide bases. Most such errors are corrected by repair enzymes that work during the G2 phase prior to nuclear division. But the efficiency of repair enzymes can vary, and some environmental conditions might activate even more accurate or sensitive repair enzyme alternatives. Whatever the mechanisms at work in a given situation, it is clear that mutation is a normal background event in all living systems, and mutation rates vary among genes and from one biological system to another.
On top of that, of course, is the increased mutation that occurs from the action of mutation-creating agents in our environment. Environmental mutagens are a potent source of health problems like cancer. Actions that we take as a society to recognize and reduce exposure to mutagens are repaid by safer, longer, healthier lives. Still, it is an awareness we can never take for granted. Chemicals in our environment are changing constantly. Furthermore, we are living longer than we did 100 years ago, so the risk of mutation in a lifetime is higher.
On the other hand, mutation, at least in producing minor allelic or quantitative variation, is also a prime contributor to developmental mechanisms like allosteric flexibility at rate-controlling biochemical steps. An ability to survive successfully in a cycling, changing environment depends in part on well- understood mechanisms like this. The success of life in a changing environment stands on a balance between the good and the bad, between the subtle and the severe. Mutation has consequences that are both harmful and beneficial. Its effects are often sad for individuals, but we cannot live without it as a species. This multifaceted view of such a central process of all biology should not surprise anyone who recognizes the dynamic nature of life.
Part 2: Medical Genetics
By far, the most common application of genetics in the practice of medicine is in the realm of diagnostics. Patients will have specific medical signs and symptoms or a significant family history. The question posed is “Why?” “What is the cause?” “What are the implications for the family?” The diagnostic approach in genetics may involve a special type of physical examination, the dysmorphologicexam, to make a specific diagnosis based on physical characteristics. Therein can be found the root of clinical genetics. But with the frequent advances now seen in readily available molecular techniques, genetic testing is increasingly becoming a mainstay of genetic diagnostics. The crux of clinical genetic testing is the identification of specific mutations that can be linked to medical problems, confirming or making the diagnosis. It is thus crucial that all physicians have a solid understanding of mutations—what they are, what they do, how they occur. This is particularly important in accurately interpreting the results of clinical genetic testing. The practitioner must understand the results and interpretations of the tests and be able to decipher the nomenclature in which they are provided.
What Is a mutation?—The Medical Perspective
As we saw in the first section of this chapter, a mutation in the purest sense is simply a heritable change (somatic or germ-line) in the genomic sequence that causes it to vary from the “normal” or “wild type” sequence. Polymorphism refers to having multiple forms of a gene. In population genetics, this term is reserved for those mutations that occur with a high enough frequency (in at least 1% of the general population).
For medicine, the central question regarding mutations is “what are the clinical consequences?” Just because a genetic change occurs, it is not necessarily true that an abnormality will result. Once again, the concepts of genotype and phenotype must be emphasized. The genotype refers to the genetic code. A mutation as noted earlier is simply a genotypic change—as compared to the reference (normal) sequence. The phenotype is an observable feature in the individual. The phenotype may be defined at any of many different levels. A phenotype can be a biochemical measurement, a physiologic state, an antigenic presence, or a physical feature. An abnormal phenotype, then, is any deviation from that typically seen in the general population. It is also important to note that caution must be used in designating a phenotypic change as abnormal; consideration has to be given to normal population variation.
So to return to our earlier discussion, all changes in the DNA code that differ from the wild type (normal) sequence are abnormal and would be termed mutations. The most important subsequent question is “what effect does the mutation have on the individual?” Many changes in gene sequences produce phenotypic changes by changing the function of the gene. These are simply pathogenic mutations.
However, some genetic changes produce no discernible clinical effects. So how does this happen? Why don’t all mutations produce clinical problems? There are several possible explanations for this:
1. Certain nucleotide changes will not alter the amino acid sequence in the translated protein because of the degenerate nature of the genetic code. The presumption is that this would not have any effect on the structure or function of that protein.
2. Changes in the DNA sequence may occur in a non-coding region of the gene, i.e., within an intronic section of the gene. In theory such changes would not be predicted to cause problems. (Recent studies, however, have clearly shown that this is not always the case. Certain intronic changes, while not altering the coding sequence, may have other secondary consequences, such as altering splice sites, and thus still have adverse phenotypic effects.)
3. A specific mutation may change the gene function somewhat, but not enough to exceed the biologic threshold needed to tip the balance toward clinically apparent problems.
4. Along these same lines, a problem may not be readily apparent at one point in time. There may still be an effect that is yet to come and be identified as a “late-onset” disorder.
Regardless of the mechanism, the common outcome is a mutation that does not produce a phenotypic change. These are collectively referred to as benign variants.
It is important that these concepts are understood when reviewing reports from genetic tests. As certain genes are repeatedly sequenced, labs will develop a library or inventory of identified changes. If a variant is identified on a clinical test, the laboratory will report the specific change and provide an interpretation based on experience and the existing medical literature. An identified variant will typically be reported as:
1. Normal sequence
2. Variant present. If present, it will be classified as:
• Known pathogenic (disease causing) mutation
• Known benign variant
• Variant of unknown/uncertain significance
In the vernacular, the term “mutation” carries a negative connotation. But in reality, a mutation is just an observable change in the genomic sequence. Some changes may actually be beneficial. In fact, in evolutionary theory these must occur. In the brief history of medical practice, however, relatively few beneficial mutations have been reported. Clinically, most polymorphisms identified in a person have no effect or are deleterious. Still, there are notable exceptions. A 32-base pair deletion in the chemokine receptor type 5 (CCR5) confers resistance to HIV infections in homozygotes and delays progression of an HIV infection to clinical AIDS in heterozygotes. Likewise, some mutations may be deleterious in homozygotes affected with a recessive disorder but actually be advantageous in heterozygotes (see Chapter 6 on Mendelian inheritance).
To prevent confusion some have advocated not even using the terms mutation and polymorphism, but instead neutral terms like “sequence variant,” “alteration,” and “allelic variant”. For us, “mutation” is a valid term with strong historical and practical use. For the purposes of this book, we will typically limit the term “mutation” to disease-causing variants.
Frequency of Mutations
Most people intuitively think of mutations as being rare events. This is largely due to the historical observations of dramatic, easily observable consequences of mutations. The genetic bases of conditions like congenital anomalies, genetic syndromes, and neuro-developmental disorders were among the first to be defined. Thus, the association of rare observable conditions with an identifiable correlating mutation led to the assumption that mutations were rare events. In reality, nothing could be further from the truth. It is true that a specific mutation event will be rare. But cumulative evidence over the past four decades has led to the clear conclusion that mutations as a group occur with an amazingly high frequency—if one simply knows where to look.
Mutation rates vary greatly across species. The measured rates of spontaneous mutation per genome under research settings are remarkably similar within broad groups of organisms, but they differ strikingly among the groups. Since unprotected DNA has a higher mutation rate, the more complex organisms have structural modifications of their DNA, such as bundling into chromosomes with further protection by chromatin, histone proteins, and other factors, which buffer against mutations. In humans, mitochondrial DNA is not protected like chromosomal DNA and, thus, has a much higher mutation rate. This is discussed further in Chapter 13, Disorders of Organelles.
A large body of literature exists on the frequency of mutations in humans. In order to get a better grasp on the genetic diversity among people, one must look at these other lines of documented evidence. We will summarize these to put the answer to the question about human mutation load into a clinically relevant context.
1. Changes in DNA transmitted to the next generation. Current estimates for a human genome is that there are 2.5 × 10−8 mutations per base per generation and 1 × 10-6 per gene per generation. Every time human DNA is passed from one generation to the next, it accumulates 100 to 200 new mutations, according to a DNA-sequencing analysis of the Y chromosome. Given a current estimate of 22,000 genes in the human genome, this suggests at least 1 in 20 persons is expected to inherit a clinically significant gene change (germinal mutation) from one or the other parent.
2. Prevalence of identifiable polymorphisms in an individual. For 20 million coding base pairs, there are an estimated 60,000 single nucleotide polymorphisms (SNPs) in humans. It is estimated that everyone has 5 to 15 genes in our approximately 22,000 gene makeup that are functionally “abnormal,” i.e., possess recessive mutations. James Watson, of Watson and Crick DNA structure fame, allowed his genome to be sequenced. He had 12 identified recessive mutations, which is consistent with estimates of the mutation load of a typical individual. As the genetic basis of common disorders is worked out, none of us may be “free” from genetic disease.
3. Somatic mutations. As an individual lives out his/her life, the natural history of an individual includes the accumulation of mutations. Somatic mutations occur with a frequency of “certainty” in every individual. They will occur. The accumulations of these mutations are then responsible for much of the commonality of the human experience including aging, adult onset disorders, and effects on longevity.
Simply put, mutations are common events. A closer look at mutations and at how and when they occur reveals several important observations about the nature of mutations in general. New mutations are not evenly distributed across the genome. Distinct mutation “hot spots” exist. Certain areas of the genome clearly have a higher mutation rate than the observed baseline. Examples of these are provided in Table 7-2. There are several possible reasons why one particular gene (locus) may show a higher mutation rate. In general, it is not surprising that larger genes have a higher mutation than smaller ones. The presumption is that if there are more nucleotides, there is a greater chance of a change occurring in one of them. Research observation confirms this assumption.
Table 7-2. Mutation Rates for Selected ‘Hotspots’ in the Human Genome*
Many of the other known explanations for the occurrence of mutation hot spots are sequence specific. For example, a particular sequence may be close to a transposable element, be a site recognized for viral incorporation, or be prone to recombinant loops. In the realm of immunogenetics, a unique mutational phenomenon is observed. In the genes that code for immunoglobins and their associated proteins, somatic hyper-mutation is noted in the immune response. In order to generate as much diversity in antibody production as possible, the immune-responsive genes have an inherent mechanism for generating genetic changes.
A few loci are even known to have a lower than average mutation rate. For instance, von Hippel-Lindau (vHL) syndrome is a familial cancer syndrome characterized by predisposition to several malignant and benign tumors, most frequently in the eyes, CNS, and abdomen. The estimated mutation rate for vHL is about 1 × 10-7. Finally, some loci may show sex differences in mutation rates. Hemophilia A is an X-linked disorder of factor VIII of the clotting cascade. For this locus, the mutation rate in males is significantly higher than in females, estimated at a ratio of 9:1.
Types of Mutations
There are several ways to alter genetic material. Small-scale changes can be classified by various schemes. One method is to classify them by the effect of the change on gene structure. This is reviewed in the first section of this chapter and summarized in Table 7-1. Alternative schemes include grouping by the effect on function, i.e., on pathogenesis (further discussed in Chapter 16), or in populations by the effect on fitness.
Genetic changes can be complex and can occur in a variety of ways. For that reason there is a clear need for convention in describing the changes. As such, standard nomenclature has been developed to denote specific changes uniformly. It is important to note that the description can be at multiple levels. The first parameter is to define what biochemical is being described. A single letter is used to designate which compound is being referenced:
• “c.” for a coding DNA sequence (cDNA, complementary to a mature mRNA molecule)
• “g.” for a genomic sequence
• “m.” or “mt.” for a mitochondrial sequence
• “r.” for an RNA sequence
• “p.” for a protein sequence
For DNA changes, further descriptors include:
1. Which of the nucleotides is changed [adenine (A), guanine (G), thymine (T), and cytosine (C)]. The first letter represents the wild-type nucleotide, and the second letter represents altered nucleotide (for example, A > G).
2. The position or location within the gene is numbered, starting with the first nucleotide at the 5′ end of the sequence as number 1.
3. Specific symbols to designate the type of change [for example ‘>’ for a nucleotide substitution or ‘Δ’ (delta) for a deletion].
So, for example, c.76A > T would describe a nucleotide substitution in which the adenine at the 76th position is replaced by a thymine in the cDNA. This is read as “seventy-six A to T.” Sometimes there may be confusion over the different nomenclatures. For instance, there are times when description of the change in the protein sequence is preferable to describing that of the nucleotide change. In those situations, a similar but different convention is used.
1. The 20 amino acids are designated by an alphabetical letter. Table 7-3 lists the assigned letters for each amino acid.
Table 7-3. Listing of Codes Used in Description of Protein Sequence Variants
2. The position number begins with methionine (the translation initiator) numbered as + 1.
3. Symbols are used to describe the different type of changes:
• > for substitution
• Del or Δ for deletion
• Dup for duplication
• Ins for insertion
The most common mutation seen in cystic fibrosis is a 3- nucleotide deletion that results in a missing phenylalanine at amino acid position 508. This would be designated p.ΔF508.
Details of the rules for generally accepted sequence variant nomenclature may be found on the Human Genome Variant Society website (http://www.hgvs.org/mutnomen/).
The clinically focused physician might ask why so much attention is paid to the type of mutation. Isn’t it enough simply to know that there is a mutation in the gene of interest? As will be detailed in the Clinical Correlation section of this chapter, knowing the type of mutation and the mechanism by which it exerts its effect is critical for developing potential therapies and interventions.
Causes of Mutations
Traditionally, the causes of mutations are felt to be either spontaneous or induced. A spontaneous mutation is one that arises “naturally,” rather than being the result of exposure to mutagens. These are hypothesized to be random mistakes usually due to errors in the normal functioning of cellular mechanisms like DNA replication, i.e., the system malfunctions without external influence. These could be errors anywhere along the normal course of mitosis, meiosis, or DNA replication and repair. The assumption is that over time, with enough replications, a mistake will eventually occur for no apparent reason. This phenomenon can be referred to as molecular decay. An intriguing question remains. Is there really such a thing as a spontaneous mutation or are these simply occurrences in which the causative agent or event has not been discovered?
Induced mutations occur when an eliciting agent is directly responsible for observed DNA changes. These agents are referred to as mutagens. As the name would imply, mutagens are any agent that can induce or increase the overall rate of new mutations. Here, it is worth making a couple of definition distinctions. A carcinogen is an agent that can induce or increase the risk of cancer. While many mutagens also act as carcinogens, the two lists are not identical. Some mutagens do not lead to cancer and some carcinogens are not mutagens. Likewise, teratogens are external agents that can result in congenital anomalies if a developing fetus receives a significant exposure. While exposures to mutagens during pregnancy can readily cause birth defects, many of the clinically important teratogenic agents produce effects by mechanisms other than the generation of mutations.
The best understood mutagens are:
1. Electromagnetic radiation
• Increased mutation rates are associated with shorter wave lengths. Longer waves (light, radio, microwave) are less mutagenic. But, the association of the ultraviolet waves from sunlight exposure and skin cancer is strong.
• X-rays and gamma rays are highly mutagenic.
• In clinical medicine, it is extremely important to protect patients from as much radiation as possible.
2. Chemicals
• The first identified chemical mutagen was nitrogen mustard, which was used as a toxic weapon in World War II.
• Chemicals may be mutagenic by affecting DNA reassembling, altering DNA replication at various stages, or by altering DNA structure by causing breaks or abnormal cross links.
• Oxidative stress
3. Viral agents
• Not surprisingly, double-stranded DNA viruses that insert their own genome into the human host DNA are the most mutagenic.
4. Transposable elements
DNA Repair
The human genome possesses multiple complex mechanisms for DNA repair. The fact that mutations are such common events essentially makes these mechanisms necessary for the survival of complex organisms. Early clues to the existence of these mechanisms were found in a group of rare genetic syndromes. A list of some of these syndromes is provided in Table 7-4. These conditions are linked by common clinical features that include predisposition to cancer, skin rashes or sensitivity to sunlight, immune deficiencies, premature aging, brittle hair, dysmorphic facies, structural congenital anomalies, bone marrow depression, short stature, cognitive dysfunction, and neurosensory abnormalities. The individual conditions exhibit their own specific combination of a subset of these features as well as other unique features that help define them as separate entities (Figures 7-12 and 7-13). The common feature of these conditions is genomic instability. The association of clinical findings such as genomic instability in the context of elevated cancer risks, premature aging, bone marrow suppression, and immune deficiencies are highly suggestive of abnormalities in DNA repair. Ultimately, the discovery of the gene(s) associated with these conditions has provided great insight into the mechanisms of normal DNA repair.
Table 7-4. Rare Genetic Syndromes Caused by Mutations in Genes Involved in DNA Repair
Figure 7-12. Young girl with ataxia-telangiectasia. (a) Note the telangiectasias (small, dilated blood vessels) on the sclera. (b) Note cerebellar atrophy on the MRI. This is the source of the ataxia. This condition is caused by a mutation in the ATM gene, a major component of the mechanism for double-stranded DNA repair.
Figure 7-13. Young girl with Bloom syndrome. This condition is characterized by short stature, sun-sensitivity, pigmentary skin changes, chromosome instability, and an increased propensity to malignancies. This child had spontaneous chromosome breakage noted on a karyotype. Subsequent testing showed compound heterozygosity for mutations in the DNA helicase RecQ protein-like 3 gene.
Current understanding of DNA repair mechanisms shows that different types of repair exist to fix different types of damage. The most common forms of damage actually have devoted repair enzymes, i.e., there can be direct repair of a specific change. O6-methylguanine is the major mutagenic lesion in DNA induced by alkylating agents. The methylguanine-DNA methyltransferase (MGMT) enzyme can remove the methyl group from O6-methylguanine, directly reversing the abnormal modification. More complex repair mechanisms exist for more complex abnormalities. Excision repair mechanisms will correct missing or altered bases with changes on only one strand. Mismatch repair systems correct the situation where both bases are “normal” but the combination of the two is not. Finally, interstrand cross-link or double-stranded break repair mechanisms are needed when both strands are damaged. This is a particularly tricky “fix,” because if both strands are damaged, there is no template for repair. Specific disorders have now been associated with malfunction of these different systems. This information is also included in Table 7-4.
Impact of Mutations
Political and public health analyses of health care often talk about cost estimates of a specific condition. It is helpful to know the magnitude of the impact of a problem on the overall health system. In actuality, however, it is practically impossible to make an accurate estimate of the scope of the impact of mutations on human health. Given the way in which they might be classified, mutations potentially have a role in all human medical conditions. Their impact goes well beyond the standard parameters of illness, lost productivity, and quality of life. On the other hand, as we discussed earlier, the complete absence of mutation is likely to be a detriment to the long-term health and well-being of a population. The ratio of benefits to harm of genetic change in populations versus individuals is not the same. For now, however, the study of medical genetics continues to focus on the occurrence of harmful mutations and their role in human disease—as best it can be grasped.
Part 3: Clinical Correlation
As mentioned earlier, knowledge about the underlying nature of a mutation responsible for a specific disorder is likely to have direct implications for therapy. Different types of mutations affect their changes in different ways. This, then, may lead to differential responses to therapy.
For most common disorders, the gene(s) responsible for the disorder have been identified. Study of the mutations found in these conditions usually identifies significant genetic heterogeneity. For a given condition, patients may be seen who have mutations of any conceivable type. Two common conditions [cystic fibrosis (CF) and Duchenne muscular dystrophy (DMD)] exhibit these characteristics. Cystic fibrosis (Figure 4-21) is an autosomal recessive disorder characterized by clinical symptoms secondary to the obstruction of exocrine glands by inspissated (i.e., having a thickened or dried consistency) mucous excretion. It has been shown that the primary defect is in a chloride transport gene called the cystic fibrosis transmembrane receptor (CFTR) that causes increased chloride concentrations in the exocrine secretions. A hyper-viscous mucous results from high chloride concentrations. Obstruction of the exocrine glands results in pancreatic insufficiency and chronic progressive obstructive lung disease as the primary symptoms. Currently, patients with CF have an average life expectancy in the 30s and continuing to increase. Duchenne muscular dystrophy (Figure 6-31) is an X-linked muscle disorder caused by mutations in a gene called dystrophin. Dystrophin functions as a biologic “shock absorber” for muscle contractions by anchoring to the sarcolemmal membranes. Missing or poorly functioning dystrophin leads to a progressive muscle disease due to an underlying mechanical destruction of the muscle fibers. Clinically, patients with DMD exhibit progressive weakness, cardiomyopthy, and ultimately death by the late teens or early 20s.
Nonsense mutations change the genetic code by producing a premature stop codon. This results in a shortened (truncated) protein. This shortened protein typically is partly or completely dysfunctional. Genetic testing for both CF and DMD has indeed shown marked genetic heterogeneity. Salient to this discussion, 10% to 15% of patients with DMD and 10% of patients with CF are found to have disease caused by nonsense mutations. Sequence analysis (not just deletion/duplication analysis alone) is needed to identify mutations that lead to premature stop signals in these patients.
In the study of antibiotics and their effects on microbes, it was discovered that the antibiotic gentamycin could induce “read-throughs” of nonsense mutations. In other words, it would allow the transcription mechanism to skip over the ectopically placed stop codon. The use of gentamycin as a drug to treat genetic mutations is not feasible, since the concentrations required to have a reasonable effect proved to be too toxic. Using this information, a drug was subsequently developed that had the same effects on nonsense mutations as gentamycin, but without the toxicity. This drug was originally called PTC124 during its time as an early investigational drug (Post-Transcriptional Control = PTC of the regulatory processes that occur after mRNA molecule has been made). Later the trade name Atrulen was given to this drug. This medication is odorless and tasteless. It is taken orally as a powder that can be dissolved in water or milk. The mechanism of action is to allow ribosomes to read through premature stop codons. In doing so, the stop codon is replaced with a random amino acid, not necessarily the correct one. A point of caution is that a drug will not generally be gene-specific. It also has the potential to turn on something it should not.
This drug is well advanced into clinical trials. It is being studied as a potential therapy for nonsense mutation mediated genetic disease, with the first clinical trials being conducted with patients with DMD and CF. It is not yet known how effective these trials may be. Still, this type of approach is clearly marking the current era as being on the threshold of true genetic therapies—actually correcting genetic mistakes. In addition, this example highlights the importance of knowing which type of mutation is present. In the case of Atrulen, it specifically identifies which subgroup of patients would benefit from this type of therapy. If this fails to excite you, maybe you are not ready for 21st-century medicine!
Board-Format Practice Questions
1. A common clinical feature seen in persons affected with a disorder of DNA repair is
A. albinism.
B. predisposition to cancer.
C. excess stability of the genome.
D. hyperproduction of immune globulins.
E. delayed aging.
2. A patient is suspected of having a specific medical condition. A DNA sequencing test is performed to confirm the clinical diagnosis. In the report, the laboratory says that a change in the DNA sequence was found. The report also states that this particular change has been seen in several other individuals, none of whom have had the suspected condition. The best interpretation of this is that the genetic change is a
A. normal sequence.
B. known pathogenic mutation.
C. known benign polymorphism.
D. polymorphism of unknown significance.
E. codon malalignment.
3. Which is the correct statement regarding mutations?
A. Mutations are rare events.
B. Mutations occur at a rate evenly distributed across the genome.
C. Spontaneous mutations are typically more severe than induced mutations.
D. Because mutations are set in the DNA, therapy is not possible.
E. Mutations may occur in either germ-line or somatic cells.
4. Agents that are known to induce mutations (mutagens):
A. often are also carcinogenic (i.e., induce cancer).
B. cannot be teratogenic (i.e., induce birth defects).
C. are mostly man-made (i.e., are not naturally occurring).
D. are typically unavoidable.
E. typically produce a mut0 phenotype.
5. Mutations
A. occur less often in unprotected DNA.
B. may be induced by bacterial infections.
C. may occur in the movement of transposable elements.
D. designated by the Δ sign are more likely to occur around rivers.
E. have a large impact on the health of the population that is quantified by the National Cost Estimates survey.