Which type of mutation is most likely to result in a completely nonfunctional protein?

PRNP Nonsense Mutations

PathogenicPRNP nonsense variants (stop codons) are very rare, and only a few families have been reported. Reported pathogenic nonsense variants (stop codons) occur at the C terminal of PrPC at codon 145 or higher and cause premature translational cessation, resulting in a truncated protein (Guerreiro et al., 2014;Minikel et al., 2016). Nonsense variants tend to present quite differently than otherPRNP mutations. They often have relatively early onset (20s–50s) with long disease durations, ranging from 1 year to more than 3 decades. Early cognitive impairment and personality changes are common, and patients with nonsense variants can have phenotypes resembling AD or FTD. Other common early features include chronic diarrhea, other gastrointestinal upset, dysautonomia, and/or sensory neuropathy (Fong et al., 2017;Ghetti et al., 1996;Jayadev et al., 2011;Kim et al., 2018;Mead et al., 2013;Takada et al., 2017). It is not known if these stop codon variants are fully penetrant, but there have been cases in which parent carriers of symptomatic children are asymptomatic (Fong et al., 2017). Based on the available limited data, codon 129 in PRNP does not appear to play a role in age of onset (Kim et al., 2018). Interestingly, on autopsy, most of the reported cases were found to have PrPSc-amyloid plaques, PrPSc amyloid angiopathy, and tau containing neurofibrillary tangles (Finckh et al., 2000a, 2000b;Ghetti et al., 1996;Jansen et al., 2010;Mead et al., 2013;Owen et al., 1989), suggesting a link between prionopathies and tauopathies.

Abstract

Nonsense mutations are involved in about 10% of patients with genetic diseases. These genetic diseases can enter in the rare pathology category, or in the frequent disease class, making nonsense mutation therapies of interest for a significant number of patients. Among the rare pathology category, Duchenne muscular dystrophy (DMD), cystic fibrosis (CF), and spinal muscular atrophy (SMA) will be described. Cancer, metabolic diseases, and neurologic disorder represent three main classes of frequent diseases that will be discussed. The rate of nonsense mutations among all these pathologies is variable, from one gene to another, or from one tissue to another. However, the mechanism of gene silencing occurring as a consequence of a nonsense mutation is shared by all patients harboring a nonsense mutation in the gene responsible for their pathology. Consequently, common therapies can be applied to patients with various diseases, changing the way to see a pathology and an associated treatment.

Nonsense Mutations

Point mutations in a DNA sequence that cause the replacement of the normal codon for an amino acid by one of the three termination (or “stop”) codons are callednonsense mutations. Because translation of messenger RNA (mRNA) ceases when a termination codon is reached (seeChapter 3), a mutation that converts a coding exon into a termination codon causes translation to stop partway through the coding sequence of the mRNA. The consequences of premature termination mutations are twofold. First, the mRNA carrying a premature mutation is often targeted for rapid degradation (through a cellular process known asnonsense-mediated mRNA decay), and no translation is possible. And second, even if the mRNA is stable enough to be translated, the truncated protein is usually so unstable that it is rapidly degraded within the cell (seeChapter 12 for examples).

Whereas some point mutations create a premature termination codon, others may destroy the normal termination codon and thus permit translation to continue until another termination codon in the mRNA is reached further downstream. Such a mutation will lead to an abnormal protein product with additional amino acids at its carboxyl terminus, and may also disrupt regulatory functions normally provided by the 3′ untranslated region downstream from the normal stop codon.

7.4.15 Nonsense Mutations

Nonsense mutations give rise to premature termination of translation and truncated polypeptides. They account for ~11% of all described gene lesions causing human inherited disease and ~20% of disease-associated single base-pair substitutions affecting gene coding regions (492). Pathological nonsense mutations resulting in TGA (38.5%), TAG (40.4%), and TAA (21.1%) occur in different proportions to naturally occurring stop codons (492). Of the 23 different nucleotide substitutions giving rise to nonsense mutations, the most frequent are CGA → TGA (21%; resulting from methylation-mediated deamination) and CAG → TAG (19%) (492). The differing nonsense mutation frequencies are largely explicable in terms of variable nucleotide substitution rates such that it is unnecessary to invoke differential translational termination efficiency or differential codon usage. Nonsense mutations are usually associated with a reduction in the steady-state level of cytoplasmic mRNA (493). This mechanism of “NMD” is responsible for the degradation of mRNAs that contain a premature termination codon at a position at least 50 nt upstream of an exon–exon boundary (494) but it is not universal (495). One or more parameters could be affected: the transcription rate, the efficiency of mRNA processing or transport to the cytoplasm, or mRNA stability.

In the majority of described instances of nonsense mutations, the resulting disorders are recessive in nature as a consequence of the haploinsufficiency resulting from the NMD-induced absence of the truncated proteins (which ensures that such polypeptides do not interfere with the function of the wild-type protein). Nonsense mutations that do not elicit NMD can, however, give rise to a dominant negative condition (e.g. mutations in the SOX10 gene causing Waardenburg Shah syndrome; (496)). Since for NMD to be activated, the nonsense mutation must reside at least 50–55 nt upstream of an exon–exon boundary, it follows that the precise location of the nonsense mutation could be an important factor in predicting the pathogenicity of that lesion. By way of example, nonsense mutations within the last exon of the human β-globin (HBB) gene do not elicit NMD. As a consequence, the truncated β-globin product has near-normal abundance, fails to associate properly with α-globin, and hence gives rise to a dominantly inherited form of α-thalassemia (24). Different nonsense mutations within the same gene may thus be associated with different clinical phenotypes depending on whether or not NMD is activated. Another example of this is provided by a nonsense mutation (Q37X) in the DAX1 gene of an adrenal hypoplasia congenita patient; this lesion is associated with a milder-than-expected clinical phenotype on account of the expression of a partially functional, amino terminal-truncated DAX1 protein synthesized from an alternative in-frame translational start site at Met83 (497). In a recent meta-analysis, the proportion of known disease-causing nonsense mutations predicted to elicit NMD was found to be significantly higher than among nonobserved (potential) nonsense mutations, implying that nonsense mutations that elicit NMD are more likely to come to clinical attention (492). In practical terms, the observation of greatly reduced or absent cytoplasmic mRNA associated with nonsense mutations has important implications for mutation screening. Thus, attempts to obtain mRNA for RT-PCR and mutation detection may result in amplification of nucleic acid from only the non-nonsense mutation-bearing allele. Nonsense mutations in the factor VIII (F8) gene (hemophilia A) and fibrillin (FBN1) gene (Marfan syndrome) have been associated with the skipping of exons containing these mutations (240,420) and this observation has now been extended to other genes; exon skipping is either complete or partial. The mechanism underlying this phenomenon is unknown although a number of intriguing models have been proposed (498).

Some genes are characterized by numerous nonsense mutations but relatively few if any missense mutations (e.g. CHM), whereas other genes exhibit many missense mutations but few if any nonsense mutations (e.g. PSEN1). Genes in the latter category have a tendency to encode proteins characterized by multimer formation (492). Consistent with the operation of a clinical selection bias, genes exhibiting an excess of nonsense mutations are also likely to display an excess of frameshift mutations (492). Recently, an example of the spontaneous read-through of a premature termination codon was reported in a patient who was a compound heterozygote for two nonsense mutations in the LAMA3 gene (R943X/R1159X) (499). The patient, who presented with junctional epidermolysis bullosa, was expected to die as a consequence of harboring these nonsense mutations but was “rescued” by spontaneous read-through of the R943X-bearing allele. This patient’s full-length R943X-bearing LAMA3 mRNA escaped nonsense-mediated decay, thereby ensuring near-normal LAMA3 mRNA and laminin-α3 protein levels. The genetic context of the LAMA3 mutation R943X was found to be close to a hypothetical consensus sequence for optimal premature termination codon read-through.

Hyperkeratosis of the nipple and areola

Hyperkeratosis of the nipple and areola is bilateral and predominantly involves the top of the nipple. Lesions may cause tenderness, pruritus, or discomfort with breastfeeding.1 The occurrence and aggravation around puberty, pregnancy or systemic hormone treatment suggests a hormonal influence.2 Hyperkeratosis of nipple skin has also been described during sorafenib treatment and with the fungus (malassezia).3,4 Nevoid hyperkeratosis of the nipple and areola may either appear isolated or in association with an epidermal nevus and other dermatoses, such as acanthosis nigricans, Darier disease, chronic eczema, chronic mucocutaneous candidiasis, or cutaneous T-cell lymphoma.5–6 Paget disease must be ruled out by histologic examination.

Keratosis of the nipple shows papillomatosis, acanthosis, and hyperkeratosis of the epidermis. The rete ridges are filiform and anastomosing and the basal layer appears hyperpigmented.1 Intraepidermal collections of lymphocytes must not be confused with T-cell lymphoma.7

5.2 Altered Decoding by tRNA May Cause Suppression

Nonsense mutations can be suppressed by alterations in tRNA. As noted earlier, a nonsense mutation occurs when a codon for an amino acid is changed to a stop codon. This results in a truncated and usually nonfunctional protein. Such a defect may be suppressed, at least partially, by changing the anticodon sequence of a tRNA molecule so that it recognizes the stop codon instead. Consider the stop codon UAG. Altering the anticodon of tRNAGln from GUC (reading CAG for Glutamine) to AUC will make it recognize UAG instead. Such an altered tRNA will insert glutamine wherever it finds a UAG stop codon (Fig. 26.40).

Mutant tRNA molecules are known that read stop codons and insert amino acids. This suppresses nonsense mutations.

Such altered tRNA molecules are known as suppressor tRNA. The UAG stop codon is known as amber and the UAA stop codon as ochre. The UGA stop codon has no universally accepted name, but is sometimes called opal. Amber suppressors are mutant tRNAs that read UAG instead of their original codon. Ochre suppressor tRNAs read both UAA and UAG due to wobble. Opal suppressors are rare (Box 26.03).

Box 26.03

Amber, Ochre, and Opal

The stop codons were originally identified by mutations in bacteriophage T4. The first one identified was UAG, the amber codon, which received its name in a curiously convoluted manner. The laboratory of Seymour Benzer at Caltech was looking for a mutation that would allow a certain kind of bacteriophage mutant to grow. Benzer said that whoever identified the mutation would have it named after him. The mutation was eventually isolated by a student named Harris Bernstein. Since “Bernstein” is German for “amber” UAG was named the amber codon. The second stop codon to be found (UAA) was called “ochre” to keep the color theme. The third stop codon (UGA) is less common and so the use of “opal” or less often “umber” is less frequent and not fully settled.

Suppressor tRNA mutations can only occur if a cell has more than one tRNA that reads a particular codon. One tRNA may be mutated while the other must carry out the original function; otherwise, the loss of the original tRNA would be lethal. In practice, cells often have multiple tRNA genes and so suppressor mutations are reasonably common, at least in microorganisms. Bacterial suppressor mutations have been found in tRNA for glutamine, leucine, serine, tyrosine, and tryptophan. The amino acid inserted by the suppressor tRNA may be identical to the original amino acid whose codon mutated to give the stop codon. In this case, the protein made will be fully restored. Alternatively, a different amino acid may be inserted and a partially active protein may be produced.

Remember that stop codons are normally recognized by release factor and have no cognate tRNA. Since suppressor tRNA must compete with release factor, suppression is never complete and typically ranges from 10% to 40%. This may provide enough of the suppressed protein for the cells to survive. However, the suppressor tRNA will also suppress other stop codons in the same cell and so generate longer (and incorrect) versions of many proteins whose genes were never mutated. Not surprisingly, cells with suppressor mutations grow more slowly. Only bacteria and lower eukaryotes (e.g., yeasts, roundworms) can tolerate suppressor mutations. In both insects and mammals, suppressor mutations are lethal.

Frameshift suppressor tRNAs are also occasionally found among bacteria. These mutant tRNA molecules have an enlarged anticodon loop and a four-base anticodon. This enables them to insert a single amino acid by reading four bases in the mRNA. They can suppress the effects of frameshift mutations caused by the insertion of a single extra base. Frameshift suppressor tRNAs with five-base anticodons have been made artificially, but have not been isolated naturally.

Other molecular therapies

Nonsense mutations create a premature stop codon thus producing a truncated, usually non-functioning, protein. It is estimated that nonsense mutations account for 5–15% of disease-causing mutations. It was recognized that aminoglycosides could influence the ability to read through the premature stop codon and so produce a normal length functioning protein. Importantly, this affects only premature, and not normal, stop codons. Subsequent development of small read-through molecules has shown that boosting the affected enzyme by small amounts can significantly reduce disease severity. Clinical trials in cystic fibrosis have been promising and potential trials for patients with IMDs secondary to nonsense mutations are currently being planned.

Exon skipping is another technique to boost functioning protein production, by masking the faulty exon harboring a frameshift mutation, using small pieces of DNA (antisense oligonucleotides). Bypassing the frameshift mutation facilitates reading of the remaining exons. Animal work confirmed the principal in mouse models and now clinical trials have started in Duchenne muscular dystrophy and Huntington disease.

7.20 Nonsense mutations and human disease

Nonsense mutations were reported in Morais et al., a review to account for more that 11% of inherited human diseases. Nonsense mutations included single nucleotide mutations that converted a sense mutation to stop codon. Premature termination of translation was noted to lead to decreased stability and degradation of the partially translated protein and its loss of function. In 5.25% of cases NMD did not occur leading to production of a truncated protein. Importantly, in autosomal dominant disorders the truncated protein was noted to potentially interfere with the function of the normal protein, referred to as dominant negative effects.

Lombard et al. described the use of medications to induce readthrough in cases with stop codon mutations leading to lysosomal storage diseases.

Translation terminating mutations have also been found in some cases with coagulation disorders. Compounds being investigated to suppress NMD decay include aminoglysosides PTC 124, ataluren, geneticin newer compounds gentamycin B1, and amlexanox.

Crispr-Cas technologies are being evaluated for treatment of disorders due to premature termination codons. Borgatti et al. (2020) reviewed screening of compounds to promote readthrough of premature termination codons for treatment of specific forms of thalassemia. They noted that PTC or nonsense mutations are generally associated with a dramatic decrease in gene expression.

Borgatti et al. noted other genetic disorders due to effects of premature termination codons. These include cystic fibrosis, Duchenne muscular dystrophy, spinal muscular atrophy, Usher syndrome with risk for deafness and visual impairment, ataxia telangiectasia. They noted that gene therapy therapeutic studies were primarily at the level of investigation in model organism but that there was some evidence of progression to clinical applications.

6.2 Altered Decoding by tRNA May Cause Suppression

Nonsense mutations can be suppressed by alterations in tRNA. As noted above, a nonsense mutation occurs when a codon for an amino acid is changed to a stop codon. This results in a truncated and usually non-functional protein. Such a defect may be suppressed, at least partially, by changing the anticodon sequence of a tRNA molecule so that it recognizes the stop codon instead. Consider the stop codon UAG. Altering the anticodon of tRNAGln from GUC (reading CAG for Gln) to AUC will make it recognize UAG instead. Such an altered tRNA will insert glutamine wherever it finds a UAG stop codon (Fig. 23.40).

Mutant tRNA molecules are known that read stop codons and insert amino acids. This suppresses nonsense mutations.

Such altered tRNA molecules are known as suppressor tRNAs. The UAG stop codon is known as amber and the UAA stop codon as ochre. The UGA stop codon has no universally accepted name, but is sometimes called opal. Amber suppressors are mutant tRNAs that read UAG instead of their original codon. Ochre suppressor tRNAs read both UAA and UAG due to wobble. Opal suppressors are rare.

Suppressor tRNA mutations can only occur if a cell has more than one tRNA that reads a particular codon. One tRNA may be mutated while the other must carry out the original function; otherwise, the loss of the original tRNA would be lethal. In practice, cells often have multiple tRNA genes and so suppressor mutations are reasonably common, at least in microorganisms. Bacterial suppressor mutations have been found in tRNAs for glutamine, leucine, serine, tyrosine, and tryptophan. The amino acid inserted by the suppressor tRNA may be identical to the original amino acid whose codon mutated to give the stop codon. In this case, the protein made will be fully restored. Alternatively, a different amino acid may be inserted and a partially active protein may be produced.

Remember that stop codons are normally recognized by release factor and have no cognate tRNAs. Since suppressor tRNA competes with release factor, suppression is never complete and typically ranges from 10–40%. This may provide enough of the suppressed protein for the cells to survive. However, the suppressor tRNA will also suppress other stop codons in the same cell and so generate longer (and incorrect) versions of many proteins whose genes were never mutated. Not surprisingly, cells with suppressor mutations grow more slowly. Only bacteria and lower eukaryotes (e.g., yeasts, roundworms) can tolerate suppressor mutations. In both insects and mammals, suppressor mutations are lethal.

Frameshift suppressor tRNAs are also occasionally found among bacteria. These mutant tRNA molecules have an enlarged anticodon loop and a four-base anticodon. This enables them to insert a single amino acid by reading four bases in the mRNA. They can suppress the effects of frameshift mutations caused by the insertion of a single extra base. Frameshift suppressor tRNAs with five-base anticodons have been made artificially, but have not been isolated naturally.

Box 23.1

Amber, Ochre, and Opal

The stop codons were originally identified by mutations in bacteriophage T4. The first one identified was UAG, the amber codon, which received its name in a curiously convoluted manner. The laboratory of Seymour Benzer at Caltech was looking for a mutation that would allow a certain kind of bacteriophage mutant to grow. Benzer said that whoever identified the mutation would have it named after him. The mutation was eventually isolated by a student named Harris Bernstein. Since “Bernstein” is German for “amber” UAG was named the amber codon. The second stop codon to be found (UAA) was called “ochre” to keep the color theme. The third stop codon (UGA) is less common and so the use of “opal” or less often “umber” is less frequent and not fully settled.

gPrD due to nonsense PRNP mutations

Nonsense mutations are very rare and have only been reported in a few kindreds. The clinical presentation is variable, but overall they are characterized by features very atypical for PrD. Some features can include prolonged disease courses (few years to more than a decade), clinical Alzheimer's disease phenotypes, sensory and autonomic peripheral nervous system involvement, chronic gastrointestinal upset (sometimes cyclical), and presence of PrP amyloid plaques and/or PrP cerebral amyloid angiopathy, often combined with tau pathology in the brain (Mead et al., 2013; Guerreiro et al., 2014).

The Q160X mutation (with either cis methionine or valine at codon 129) has been reported in three families, with onset of symptoms ranging from 27 to 59 years of age (Finckh et al., 2000; Jayadev et al., 2011; Guerreiro et al., 2014; Fong et al., 2016). In two of those families, the probands were diagnosed clinically with Alzheimer disease, and in two individuals from one kindred the neuropathologic assessment revealed neurofibrillary tangles and amyloid plaques that immunostained for PrP, but not amyloid β. Chronic diarrhea was also reported two individuals. Other nonsense mutations, such as Q145X-129M, Q163X-129V, Y226X-129V, and a 2bp deletion at codon 178 (Ghetti et al., 1996; Jansen et al., 2010; Matsuzono et al., 2013; Mead et al., 2013), are summarized in Table 29.3.

Table 29.3. PRNP nonsense mutations

AD, Alzheimer-type dementia; Cog, cognitive; c/w, consistent with; D, dementia; Dep, depression; Dysauto, dysautonomia; EEG, electroencephalogram; FHx, family history; JCD, Jakob–Creutzfeldt disease; MRI, magnetic resonance imaging; N/A, not available; NFT, neurofibrillary tangles; Park, parkinsonism; PrP-angio, PrP amyloid angiopathy; PrP-Plqs, PrP-amyloid plaques; PSWC, periodic sharp-wave complexes; Sp, spongiosis; V, vacuolation.

From Kim et al. (2017) with permission from Cold Spring Harbor Laboratory Press.