1
Mutation

A mutation is a stable and potentially heritable change in a DNA sequence. Mutations may occur in the soma of an organism, affecting only a particular cell or lineage of cells, or they may occur in the germline of an organism and be passed to all of that organism’s offspring. Mutations that occur within or near a gene may create a phenotype different from that normally expressed by the wildtype allele of that gene. A number of different types of mutations have been found to cause changes in phenotype. These mutations can be changes in individual base pairs, such as substitutions (e.g. C → T), insertions or deletions of DNA, or they may be chromosomal aberrations such as inversions, translocations, or copy number variants near or within genes. Because this book is fundamentally about mutational analysis, we need to spend some time considering the types of mutations that can occur, both at the molecular level and in terms of the effects they can have. We also need to review the various systems that exist to classify mutations. Such a review is all the more critical because the nomenclature systems that geneticists have developed are keyed to the structure and effects of the mutants they name. Thus, the things themselves – and our names for them – are inextricably intertwined.

1.1 Types of Mutations

Most introductory genetics texts classify mutations simply as recessive or dominant. A mutation (m1) is said to be recessive if m1/m1 organisms display a mutant phenotype, but m1/+ organisms are wildtype. (Note: the symbol “+” denotes the wildtype, or normal, allele of a given gene.) Conversely, a mutation (M2) is said to be dominant if M2/+ organisms display a mutant phenotype while +/+ organisms are normal. Some texts use the term semidominant to describe cases where a dominant mutation, M3, displays a more severe (or extreme) phenotype as a homozygote (M3/M3) than it does as a heterozygote (M3/+), such that the order of phenotypic severity is M3/M3 > M3/+ > +/+. Although such a classification is sufficient for some purposes, it is inadequate to describe the range of mutant types, or phenotypes, that can actually be observed. Accordingly, at least three more‐detailed classification systems have been developed and are discussed in this section.1

Muller’s Classification of Mutants

The first detailed mutant classification scheme was proposed in 1932 by Herman J. Muller (1932), more than two decades before the first statement of the central dogma of molecular biology – that DNA codes for RNA and RNA then codes for protein (Crick 1958, 1970). Muller classified mutants into five basic groups: nullomorphs, hypomorphs, hypermorphs, antimorphs, and neomorphs. The assignment of a mutant to one of these classes was largely based on Muller’s view that mutations can, and should, be described in terms of their effect on activity. A mutation can be assigned to one of these five groups by comparing the phenotypic effects of that mutation in homo‐, hetero‐, and hemizygotes. (A hemizygote is an individual with a single allele at a specific position, instead of two. For example, human males, who are XY, are hemizygous for most X chromosome genes.) Understanding these classifications, and being able to use them, is a critical component of genetic analysis. We will therefore consider each of these types of mutations in some detail. We begin by considering the two classes of loss‐of‐function mutations: nullomorphs and hypomorphs.

Nullomorphs

Also known as amorphs, nullomorphs are mutants with no remaining gene function – they produce no functional gene product but may still create part or all of the protein that the gene encodes. They are often, and far more precisely, called null alleles, and they are the basic mainstay of genetic analysis. Nullomorphic mutations might correspond to internal deletions, frameshift mutations leading to a premature stop codon, or missense mutations that alter a critical site in the protein in such a way as to fully ablate its activity (see Box 1.1). The most characteristic feature of a nullomorph is that it is the equivalent of a full deletion of the gene in terms of its influence on the final phenotype.

Box 1.1 DNA‐Level Terminology

While DNA‐level terminology is covered in more depth in Section 1.3, many of the terms are useful when discussing both early and modern mutant terminology. Here is a brief overview of this DNA‐level vocabulary.

• Single‐nucleotide variant (SNV): replacement of one nucleotide base with another; also referred to as a substitution mutation or point mutation.
• Missense mutation: a type of SNV that changes the amino acid encoded by a codon.
• Indel: a DNA insertion or deletion of less than 50 base pairs.
• Frameshift: an indel that alters the reading frame.
• Deficiency: a large deletion that completely removes an entire gene or region of the genome; this term is used frequently in model organism genetics and less so in human genetics.
• Transposable element (TE): a segment of DNA capable of moving around within the genome; also referred to as a transposon.
• Duplication: a region of the genome that exists in two or more copies. This could be a tandem duplication where the duplicated segments sit next to each other, or the duplicated segment may reside on another chromosome.2
• Inversion: a section of a chromosome that has been reversed.
• Translocation: the transfer of a section of DNA from one chromosome to another.
• Structural variant (SV): a DNA variant greater than 50 base pairs in length; includes insertions, deletions, duplications, inversions, repeat expansions, or translocations.
• Copy number variant (CNV): a type of SV that changes the number of copies of a coding or noncoding genomic region; includes large duplications and deletions/deficiencies.

Null alleles lead to the complete absence of a functional protein product via a variety of defects in gene expression. Using the relevant molecular tools (some of which are discussed in Section 2.2), one can discriminate between transcriptional nulls, protein (or translational) nulls, and mutations that produce completely inactive proteins. In the case of a transcriptional null, no full‐length transcript is produced. Such mutations might reflect, for example, the deletion of crucial elements in the promoter or the insertion of a foreign genetic element (e.g. a transposon) in or near the gene. Transcriptional nulls may be identified by RT‐PCR, by examining chromatin structure, or by RNA sequencing (RNA‐seq) (see Box 1.2), which would show no transcript of your gene of interest. From experience, we can tell you to be careful when calling a mutant a transcriptional null based on the absence of a transcript. Not every gene is expressed at the same time in the same tissue, so make sure you sample from a tissue and time point where and when you know your gene should be expressed. Protein nulls are defined as mutations that fail to produce a protein product at all, as assayed by an antibody specific for that protein. (This classification may include transcriptional nulls in cases where the mutant’s effect on transcription has not been assessed.) Inactivating nulls produce a protein product, but that product exerts no obvious activity. However, the most obvious type of nullomorphic mutation is a deficiency (Df), or a deletion of some or all of the DNA that encompasses the gene in question (Figure 1.1).

Box 1.2 Detecting Gene Expression by RNA‐seq

Gene expression can be detected by several different techniques, including in situ hybridization, northern blot analysis, RNA sequencing (RNA‐seq), and RT‐PCR. In situ hybridization is a technique that detects specific RNA fragments in tissue using labeled complementary DNA fragments, while northern blotting is a way to detect a specific fragment of RNA using a labeled complementary DNA fragment on a gel. Both RT‐PCR and RNA‐seq differ from in situ hybridization and northern blotting in that they typically begin by converting messenger RNA (mRNA) into double‐stranded DNA via reverse transcription. For the purposes of our discussion here, we can differentiate RT‐PCR from RNA‐seq by saying that RT‐PCR uses primers to reverse transcribe the RNA from a specific gene of interest into DNA, while RNA‐seq may target all mRNA transcripts or only some (in the case of poly‐A primers) for reverse transcription to DNA. Note that there are now sequencing techniques that allow direct sequencing of mRNA molecules without a reverse transcription step.

RT‐PCR is more sensitive than RNA‐seq and in the past has been considered more reliable in measuring changes in gene expression between samples (e.g. control compared to treated sample). The advantage of RNA‐seq is that you can amplify a lot more RNA at once, and if you have replicates, the result is likely similar to what you would get with RT‐PCR. The caveat is that you must be careful when designing your experiment because there are several...