Mutagenesis is the driving force behind evolution, and genetic variation is the result of mutations. Any change in genetic information may result in advantageous or disadvantageous phenotypic characteristics that impact an organism's fitness. When a mutation results in a higher fitness, natural selection favors these phenotypes, and these traits are more likely to be passed to offspring.[8]

Mutagenesis occurs as a result of DNA replication errors, DNA damage, and lab techniques. Here we will break mutagenesis down into endogenous and exogenous causes.

Endogenous

1. Errors in DNA replication: Our body possesses high and low-fidelity DNA polymerases. Although the error rate is minimal, high-fidelity polymerases and mismatch repair (MMR) mechanisms still make 1 in 10^6 to 10^8 base substitution per cell per generation. Furthermore, some errors occur due to replication slippage at repetitive sequences, which can lead to insertions and deletions. A mutation will result if these errors are not repaired before the next round of DNA replication.[9][10]
2. Errors in DNA repair mechanisms: The DNA damage response (DDR) is a group of mechanisms that sense DNA damage and promote repair. Errors in repair or mutations affecting the DDR network cause different cancers. There are multiple DNA repair mechanisms, including MMR, base excision repair (BER), nucleotide excision repair (NER), translesion synthesis (TLS), homologous recombination (HR) and non-homologous end-joining (NHEJ) pathways. If any of these repair mechanisms go awry, a cell is predisposed to DNA damage. For example, xeroderma pigmentosum (a rare autosomal recessive skin disorder that makes a person highly prone to developing skin cancer) is caused by a mutation in the NER pathway, resulting in a build-up of UV-associated damage. The TLS repair system and NHEJ are of interest to endogenous mutagenesis. During DNA replication, high-fidelity polymerases have difficulty passing damaged bases (e.g., pyrimidine dimers or crosslinked DNA), which will stall DNA replication. Failure to restart replication can result in double-stranded breaks, chromosomal rearrangements, and cell death. Therefore, it is often beneficial to circumvent these replicative arrests to promote cell survival. One mechanism to accomplish this is the TLS system. TLS involves DNA polymerases with larger active sites that allow them to tolerate and bypass DNA lesions. However, it comes at the expense of lower replication fidelity and high error rates, resulting in a greater likelihood of base substitutions.[10] Otherwise, NHEJ is a repair mechanism for double-stranded DNA breaks. It brings two ends of DNA fragments together and doesn't require homologous sequences. Therefore, this can result in deletions and insertions.[11]
3. Spontaneous base deamination: Base deamination is when a nucleotide base loses an amine group, effectively changing the nucleotide. The following are major deamination reactions: cytosine to uracil (U), adenine to hypoxanthine, guanine to xanthine, and 5-methyl cytosine (5mC) to thymine. If these alterations are not repaired, there can be a change in the DNA sequence. For example, if cytosine deaminates to uracil, in two replication events, there will be a new A:T mutation. This G:C to A:T transition accounts for 33% of the single-site mutations that result in hereditary diseases in humans.[10]
4. Oxidative DNA damage: In normal cell physiology, reactive oxygen species (ROS) are a byproduct of the electron transport chain (ETC) and other cellular processes. ROS serve essential cellular roles, including redox signaling and immune defense. However, in high quantities, ROS becomes damaging to a cell and its DNA. There have been 100 different oxidative base lesions and 2-deoxyribose modifications described. One example is the oxidation of carbon #8 of guanine, forming 8-oxoguanine (8-OG). 8-OG incorrectly pairs with adenine (instead of cytosine), resulting in a G:C to A:T transition. Excess ROS are known to be associated with many diseases, including Alzheimer's disease, cancer, and heart failure.[10]
5. Base methylation: S-adenosylmethionine (SAM) is a molecule used as a methyl donor during physiologic DNA methylation. At a concentration of 4x10^-5M, SAM can generate over 4,000 methylated base changes per cell per day. The significance of this can be depicted in the methylated products O6-methylguanine and O4-methylthymine. These products are highly mutagenic and result in G:C to A:T and T:A to C:G transition mutations.[10][12]
6. Abasic sites (i.e., apurinic and apyrimidinic sites): There are approximately 10,000 abasic sites made every day by spontaneous hydrolysis or DNA glycosylase cleavage. Abasic sites are unstable and commonly removed by endonucleases. In other cases, they are repaired by TLS polymerases. If theses damage sites are not corrected, they may result in mutagenesis.[10][13]

Exogenous

1. Ionizing radiation (IR): IR comes from the soil, radon, medical devices, and cosmic radiation, among others. It can damage DNA directly (e.g., DNA strand breaks) or indirectly (e.g., radiolysis of water molecules producing ROS). Ionizing radiation can generate a range of nucleotide base lesions, similar to that discussed for ROS, resulting in mutagenesis.[10]
2. Ultraviolet (UV) radiation: UV light falls between 100 to 400 nm, with the most harmful radiation at lower wavelengths. UV light damages DNA by direct and indirect (to nearby molecules) energy transfer. The two main products of UV damage are pyrimidine dimers and pyrimidine pyrimidone photoproducts. These byproducts distort the DNA helix, requiring the NER system or the error-prone TLS polymerases to bypass them. Pyrimidine dimers are known to cause C:G to T:A, T:A to C:G, and tandem CC to TT transition mutations.[10]
3. Alkylating agents & Aromatic amines: Alkylating agents (e.g., nitrogen mustard gas, methyl methanesulfonate [MMS], ethyl methanesulfonate [EMS], N-ethyl-N-nitrosourea [ENU]) have a high affinity for nitrogens on nucleotide bases, mainly N3 of adenine and N7 of guanine. MMS, for example, reacts with adenine and guanine to produce N3-methyladenine and N7-methylguanine, respectively. These methyl products are susceptible to N-glycosidic bond cleavage that can create abasic sites. Aromatic amines (e.g., 2-aminofluorene, previously used in insecticides) are metabolized by the CYP450 system and converted into alkylating agents. These products primarily cause lesions to the C8 position of guanine. C8-guanine lesions are known to give rise to base substitutions and frameshift mutations.[10]
4. Polycyclic aromatic hydrocarbon (PAH): PAHs (e.g., dibenzo[a,l]pyrene, naphthalene, anthracene, and pyrene) are carbon compounds with two or more aromatic rings. They are commonly present in tobacco smoke, automobile exhaust, charred food, and combustion products of fossil fuels and organic matter. The CYP450 enzymes convert PAHs into reactive DNA intermediates that intercalate into DNA, ultimately forming a DNA adduct (a segment of DNA bound to a cancer-causing chemical). This results in DNA damage and, thus, can cause mutagenesis.[10]
5. Crosslinking: Crosslinking occurs when two nucleotides form a covalent link. Agents that are commonly associated with crosslinking include cyclophosphamide, cisplatin, and psoralens. Interstrand crosslinking blocks DNA replication, and this requires repair or bypassing. TLS is one of these repair mechanisms and is associated with high substitution rates.[10][14]
6. Insertional mutagenesis: This is the process by which exogenous DNA integrates into host DNA. Insertional mutagenesis can be natural, mediated by transposons or viruses, or accomplished in a laboratory. Given there is an addition of nucleotides, insertional mutagenesis commonly results in frameshift mutations.[15]
7. Other toxins: Aflatoxin is a naturally occurring toxin from Aspergillus. The CYP450 system metabolizes aflatoxin into an active form that adducts with N7 of guanine, which results in depurination. Aflatoxin is a well-established liver carcinogen that is associated with hepatocellular carcinoma. N-nitrosamines are organic compounds commonly found in tobacco smoke, preserved meats, and the environment. They are metabolized by the CYP450 system to form DNA alkylating agents. N-nitrosamines have been implicated in nasopharyngeal, esophageal, and gastric.[10][16]
8. Laboratory techniques: Different laboratory techniques, including PCR, non-PCR, and gene-editing tools, are used to induce mutagenesis. These are discussed further in the "Testing/Methods" section below.

Mutagenesis is a technique used in molecular biology to create mutant genes, proteins, and organisms. Two primary mutagenesis techniques are site-directed mutagenesis (SDM) and random-and-extensive mutagenesis (REM). These methods are largely accomplished by polymerase chain reaction (PCR) and non-polymerase chain reaction (non-PCR). Other methods for accomplishing mutagenasis include CRISPR/Cas9 technology, TALENs (transcription activator-like effector nucleases), and Zinc-Finger Nucleases (ZFN).

Site-directed Mutagenesis

SDM is a technique where DNA can be modified at a specific nucleotide location, causing a predetermined amino acid change. These substitution mutations can result in drastic changes in protein conformation and function. This technique requires (1) a DNA template with a target gene, (2) knowledge of the target gene's nucleotide sequence, and (3) a short primer (commonly 20 to 30 base pairs) complementary to the target sequence that is modified to contain a mismatched nucleotide (typically 1 to 3 base pairs that will cause an amino acid change). The general procedure for SDM is the following:

• Step 1: Separation of the two strands of template DNA. This can be accomplished by heat or an alkali solution.
• Step 2: Addition of the modified DNA primer. Once the DNA primer anneals with a single strand, a DNA polymerase replicates the strand.
• Step 3: The second round of replication yields a mutant DNA strand that can be used to synthesize a modified protein.

Techniques for Performing SDM

PCR and non-PCR techniques are in vitro methods for performing SDM. In vitro synthesis has four essential components: DNA template, modified primers, deoxyribonucleic nucleoside triphosphates (dNTPs [i.e., dATP, dCTP, dGTP, dTTP]), and DNA polymerase (thermostable or thermolabile). The polymerase chain reaction technique utilizes thermostable DNA polymerases (e.g., Taq, Pfu, and Vent), at least two different primers, and multiple heating and cooling cycles (average 30 cycles). Each cycle has three phases: denaturation (approximately 95 degrees Celsius), annealing (approximately 55 degrees Celsius), and extension (approximately 72 degrees Celsius). The denaturation phase is when the template DNA molecule separates into two single-stranded molecules. The annealing phase is when the modified primer base pairs at the sequence of interest. Lastly, the extension phase is when the annealed primers are extended according to the template strand. The advantage of PCR is that it works more favorably with different DNA templates (i.e., single or double-stranded DNA and GC-rich regions) and produces millions of copies of a target gene. The disadvantage is there is less sequence fidelity (i.e., thermostable polymerases are more error-prone in comparison to thermolabile polymerases). The non-polymerase chain reaction technique utilizes thermolabile DNA polymerases, one primer, and constant reaction temperature (e.g., 37 degrees Celsius). In this strategy, the DNA template is denatured with an alkali solution or heat. The template is annealed with the modified primer and DNA synthesis occurs at 37 degrees Celsius. Though single or double-stranded DNA templates can be used, more favorable outcomes occur if a single-stranded DNA template is used (this increases the success of annealing). This process produces a hybrid DNA molecule (i.e., one template strand and one newly synthesized mutant strand), which can be transformed or infected into E. coli. Once in E.coli, the mutant DNA and wild-type DNA can be segregated. Non-PCR has greater sequence fidelity than the PCR method; however, there is only one DNA strand generated (rather than millions).[17]

Screening for Desired Mutants in SDM

There are different methods for selecting and screening for desired mutants. One of the PCR approaches is to use plasmids that are grown in E. coli (these are dam methylated) and contain a gene for antibiotic resistance. As discussed, modified forward and reverse primers are extended around plasmid DNA, creating copies of the template with the inserted mutations. Since the original parental template strand is methylated, a restriction endonuclease Dpn I - specific for methylated and hemimethylated DNA - digests this DNA leaving only the mutant plasmid, which is further amplified. The plasmid is transformed into E.coli, and bacteria are grown on media that contain the appropriate antibiotic for the plasmid. One can then pick single colonies and grow them overnight to isolate the mutant plasmid DNA. Please refer to the article by Bachman for a more detailed review.[18] Since there is the possibility of random mutations in PCR and non-PCR methods, sequencing should be performed to verify the desired mutation was made. DNA sequencing technology (e.g., Sanger sequencing or next-generation sequencing) can be utilized to determine nucleotide sequences.[19]

Random and Extensive Mutagenesis

REM is a useful approach when many mutations are desired; however, there is less control over the resulting modifications. This technique has helped map out critical residues of proteins (e.g., EcoRV restriction endonuclease). REM can be accomplished through different methods:

• Primers randomly produced with mismatched bases. In this technique, a set of mutagenic primers with three mismatched bases at a single base position are synthesized in the same reaction. These sets of primers are then used to synthesize mutant DNA with three different mutations at the same base position. Other than this, a primer can be synthesized using an ambiguous base (e.g., deoxyinosine), which is capable of base-pairing with varying dNTPs. In theory, there is a 75% probability a wrong base will pair with this ambiguous base, which will generate mutations in subsequent rounds of replication.
• Erroneous PCR. Taq DNA polymerase typically has lower-fidelity in comparison to other polymerases (e.g., Pfu and Vent). Furthermore, altering external conditions, such as buffer composition (e.g., high pH or high magnesium concentration), can affect the frequency of errors. Under various combinations of external conditions, the error rate of Taq PCR can be increased to 1 in 150 base pairs, from 1 in 633.
• Use of ambiguous base analogs (e.g., deoxyinosine). As already stated, deoxyinosine triphosphate (dI) can base pair with different dNTPs. In a reaction that contains dI and adjusted dNTPs concentrations (e.g., typical levels of three dNTPs and low levels of the fourth dNTP), a dI is likely to be incorporated into the newly synthesized sequence. Theoretically, there is a 75% chance that a wrong base will pair with a dI during subsequent replication events. This technique can result in 1 in 250 base pair mutation rates.[17]
• Use of mutagenic agents. A classic method for causing widespread random mutations is by exposing a cell or organism to a mutagen (e.g., ENU).[20]

Insertions and Deletions

PCR is effective at producing insertions or deletions. Small insertions or deletions (less than ten base pairs) can occur by utilizing the modified primer design described above. Large deletions can occur by joining two PCR-amplified fragments, which omits a portion of a DNA fragment. Large insertions can occur using megaprimers or overlap-extension PCR.[17]

CRISPR-Cas9

This technology is a genome-editing tool derived from bacterial defenses against viruses and foreign plasmids. It has two vital components: a guide RNA (gRNA) that binds complementary to a target DNA sequence, and an endonuclease (Cas9) that causes a dsDNA break that requires repair. Cas9-induced site-specific dsDNA breaks induce endogenous cell repair mechanisms, which can be exploited to modify DNA. The error-prone nonhomologous end-joining (NHEJ) can rapidly ligate DNA breaks; however, it may generate small insertions and deletions that can cause frameshift mutations. Alternatively, dsDNA breaks can be repaired by homology-directed repair (HDR), which can insert exogenous DNA and introduce precise genome editing.[21]

TALENS vs. Zinc Finger Nucleases

TALENs and ZFN are gene-editing tools with similar principles; however, different underlying designs. They both are proteins that consist of DNA binding domains (i.e., TALE proteins vs. zinc fingers domains) and DNA cleavage domains (e.g., Fok1). DNA specificity resides in the TALE proteins and the zinc finger domains. For cleavage to occur, there must be dimerization of the restriction enzyme; therefore, at least two proteins are needed per target DNA. Once dimerization occurs at a target location, a double-stranded DNA break is made. Similar to CRISPR, this is repaired by NHEJ (with the possibility of insertions and deletions) or homologous recombination (where exogenous DNA can be added). A key difference between TALENs and ZFN is that TALE proteins recognize a single base, while zinc finger domains recognize 3 or 4 bases. Given zinc fingers must recognize multiple bases, it may reduce DNA specificity in comparison to TALE proteins.[22][23]

Mutagenesis has multiple implications in clinical medicine. This article will discuss a few, including carcinogenesis, heritable diseases, microorganism resistance, large scale mutagenic projects, and precision medicine.

Carcinogenesis (i.e., tumorigenesis or oncogenesis) is when a cell begins to divide uncontrollably. Mutations of oncogenes (which promote cell growth), tumor suppressor genes (which inhibit cell growth), or cell-cycle genes (which regulate the cell-cycle) can generate a clonal cell population with highly proliferative properties, leading to cancer.[24] Approximately two-thirds of cancer driver mutations are attributable to spontaneous mutagenesis that occurs during normal DNA replication.[25] Furthermore, multiple agents (such as those described in the "mechanism" section) are known to be associated with carcinogenesis. Tobacco contains DNA methylating agents, alkylating agents, polycyclic aromatic hydrocarbons, and N-nitrosamines, among others. Smoking tobacco is proven to increase the risk of a multitude of cancers, including lung, colorectal, head and neck, and lower urinary tract cancers. Tobacco accounts for 16% of all cancer diagnoses and approximately one-third of all cancer deaths in the US. This highlights why counseling on smoking cessation, and other preventative health measures, becomes so important.[10][24][26] 

Ionizing radiation (IR), as a consequence of atomic radiation (e.g., Chernobyl) or iatrogenic causes, is associated with papillary thyroid carcinoma (PTC), among others. In PTC, IR increases the rate of DNA rearrangements that cause a permanently active promotor region of ret, a tyrosine kinase. This has been associated with the pathogenesis of PTC. Applying this knowledge, anti-promoter agents may one day be a method for future therapy of PTC.[27] Interestingly, some chemotherapeutic drugs used to combat cancers - in particular, cisplatin, cyclophosphamide, and etoposide - are themselves carcinogenic. These drugs have been implicated in the development of tumor resistance and subsequent cancer development. Therefore, care should be taken when choosing treatment regimens, especially in childhood cancers.[28] 

Heritable diseases have their origin in mutagenesis, as these are permanent DNA changes that are passed to offspring. Some examples of hereditary diseases include sickle cell disease, Tay-Sachs disease, cystic fibrosis, Huntington disease, Duchenne muscular dystrophy, and hemophilia, among many others. A well-studied example is sickle cell anemia. This is an autosomal recessive disease resulting from a missense mutation in the B-globin gene (i.e., a point mutation resulting in a glutamic acid to valine substitution). Research shows that this mutation created an adaptive advantage in heterozygote carriers in malaria-endemic areas.[29] However, homozygous inheritance has a poor prognosis due to increased risk for anemia, infection, stroke, and organ damage.[30]

Drug resistance in microorganisms is a result of mutagenesis. Multidrug-resistant (MDR) organisms are associated with increased antibiotic use. MDR organisms have increased mortality rates, and it is estimated that by 2050 there might be 300 million premature deaths worldwide due to MDR organisms. This highlights the need to practice effective antimicrobial stewardship to slow the development of MDR organisms.[31] An interesting example of drug resistance can occur in the treatment of Chagas disease. Benznidazole is a front-line pro-drug used in Chagas treatment. New evidence suggests that Trypanosoma cruzi may process this drug into metabolites that cause whole-genome mutations. Therefore, this may be a reason for treatment failure and the development of drug-resistant parasites. This can even lead to resistance to different classes of drugs, like posaconazole, used in combination therapy of Trypanosoma cruzi. Therefore, vigilance is required when using benznidazole in combination therapy.[32]

Large scale mutagenesis techniques can be used to map functionally important amino acid residues that affect particular phenotypes. For example, a mouse study on the Golgi protein GMAP-210 showed that gene alteration caused a phenotype that resembled human achondrogenesis type 1A. This helped lead to the discovery of a similar mutation found in humans.[33]

Lastly, precision medicine (PM) is a concept where disease treatment is based on knowing an individual's genome abnormalities, which became relevant as whole-genome sequencing became more accessible. PM is dependent on new therapeutic strategies, drug development, and gene-oriented treatment.[34] For example, gene-targeted techniques like CRISPR/Cas9 and TALENs may one day have clinical use. CRISPR is being investigated to treat single-gene mutations (e.g., DMD, cystic fibrosis), HIV, and cancers.[35] There are still multiple challenges to PM, such as validating functional mechanisms of mutated gene expression.[34] However, PM may one day have an essential impact on disease cures.