Silent mutations: a way to conquer cancer?

On the road of scientific exploration, it is often necessary to break long-standing concepts in order to create new theories.

Written by | XZ

We know that gene mutations can cause cancer. A mutation in a base pair in a DNA fragment will change the amino acid sequence encoded by the gene. This type of mutation is often judged to be the main factor causing the disease, known as a "non-synonymous mutation", and is also the main focus of cancer research.

Correspondingly, there is a "silent" mutation, also known as a "synonymous mutation" or "neutral mutation", which does not change the type of amino acid encoded.

Synonymous mutations do not affect the amino acid sequence of proteins, so they have always been considered harmless. They are "silent" because they are "useless" and can only be described as a strange biological phenomenon. For a long time, synonymous mutations have never received attention in cancer research.

But things are changing, and the turning point will eventually happen.

Mutations that silence genes may also cause disease

More and more studies have shown that synonymous mutations are not "harmless" and may still cause disease. Lu Xin, a cancer biologist at the University of Notre Dame, realized this a few years ago.

One day, Lu Xin received a patient with Von Hippel–Lindau disease (VHL syndrome) and analyzed his condition. VHL syndrome is a rare autosomal dominant genetic disease caused by a mutation in the VHL tumor suppressor gene (Von Hippel–Lindau tumor suppressor gene) located on chromosome 3P25.3. Functional VHL protein plays an important role in cell proliferation and vascular regulation. When the VHL tumor suppressor gene mutates, cells cannot express functional VHL protein, which will manifest as multi-organ tumor syndrome. In layman's terms, tumors may grow anywhere in the body.

The VHL gene is inherited in an autosomal dominant manner, so only when both VHL alleles on the chromosome (genes located at the same position on a pair of homologous chromosomes that control different forms of the same trait) are mutated, the cells will be unable to express VHL protein, leading to tumor formation.

During the conversation, Lu Xin learned that the patient's two daughters also inherited VHL syndrome. The two seven-year-old twin girls had unresectable tumors growing on their retinas. However, after checking the genome sequencing data of the three people, Lu Xin was surprised to find that there was no protein-changing mutation in the VHL gene of the three people, but only a synonymous mutation, that is, the sequence CCA mutated into the sequence CCG. The two sequences encode the same amino acid (both proline). In theory, such a mutation will not produce any biological effect and should not cause disease, but the three people do have VHL syndrome!

Lu Xin was very confused. To find out what was going on, he collected skin samples from the family, isolated cell lines from them, and conducted a series of analyses [1]. He found that synonymous mutations in the VHL tumor suppressor gene affected the process of mRNA splicing. The so-called mRNA splicing refers to the process of removing introns from the initial transcription product transcribed from the DNA template chain and connecting exons to form a continuous RNA molecule (see "20 years after gene sequencing, we finally figured out what junk DNA does"). In this process, synonymous mutations destroyed the binding of splicing proteins to specific nucleotide sequences, resulting in a section of mRNA being cut off, producing a "simplified version" of mRNA. Therefore, these three patients did not have complete VHL protein in specific tissues, which led to the symptoms of VHL syndrome.

In fact, there is growing evidence that synonymous mutations are not always harmless, and Lu Xin's discovery is just the tip of the iceberg. In the process of life evolution, external pressure will change the direction of evolution. Under such selective pressure, synonymous mutations often appear. Synonymous mutations can affect gene expression in a variety of ways and are associated with diseases such as cystic fibrosis, autism, and cancer [2]. Despite this, since there is currently no systematic analysis of synonymous mutations, whenever the topic of cancer is discussed, the impact of synonymous mutations is basically ignored by researchers, and non-synonymous mutations are the "core" of the discussion.

Some scientists are now trying to reverse the disdain for synonymous mutations. It is estimated that certain synonymous mutations in DNA may account for 5% to 8% of all cancer-causing mutations[3]. These scientists are working to decipher the impact of these mutations on cancer through systematic studies and understand the extent of their impact on the entire genome. Such studies could not only identify disease-driving mechanisms in some patients, but also open up new treatment avenues.

How do synonymous mutations create "noise"?

Synonymous codons have different nucleotide sequences but encode the same amino acids, so they are called "synonymous". However, research by Lu Xin and others shows that things are not as simple as "synonymous interchange" - synonymous codons are not completely interchangeable. In fact, codons in a specific sequence can affect the binding of gene expression regulatory proteins (such as transcription factors) to DNA regulatory sites, and can also affect the interaction between splicing proteins and mRNA precursors. For example, some mRNA precursors originally express tumor suppressor genes, but some synonymous mutations disrupt their splicing, which results in the promotion of tumor growth [4].

Synonymous mutations may also change the effects of non-synonymous mutations, that is, they may promote the expression of non-synonymous mutant genes by affecting the splicing of mRNA[5]. For example, researchers used gene editing technology to design various mutations into human cancer cell lines and found that due to the change in the type of bases, the original splicing site was destroyed, and the non-synonymous mutation produced a new splicing site, which effectively removed part of the coding sequence of the mutant protein from the mRNA, making it impossible for the mutant protein to be fully expressed. However, synonymous mutations can destroy the splicing site produced by non-synonymous mutations, ultimately leading to the complete expression of the mutant protein[6]. In other words, some non-synonymous mutations need to occur together with certain synonymous mutations in order to express the mutant protein together. This strange setting surprised and puzzled scientists.

Different synonymous codons also affect the two-dimensional structure of mRNA, thereby affecting the survival time of mRNA in the cytoplasm and its binding to ribosomes. For example, synonymous mutations in the gene encoding the human dopamine receptor D2 cause mRNA to fold into a less stable conformation, resulting in insufficient mRNA translated into protein[7].

Codons also affect the translation efficiency of mRNA. Different species, and even different tissues and genes within an organism, will preferentially use certain synonymous codons to encode specific amino acids in order to optimize their specific translation mechanisms. A rarely used synonymous codon often takes longer to translate than a frequently used codon because there are fewer corresponding transfer RNAs (tRNAs) during the translation process. In human cells, synonymous mutations have been observed to slow down translation by replacing frequently used codons with rare codons, or to speed up translation by converting rare codons to common codons, ultimately changing the structure or concentration of proteins [8, 9].

So far, the effects of synonymous mutations on gene expression can be summarized as follows:

1. Change the binding site between regulatory proteins and DNA;

2. Destroy splicing sites;

3. Affect the stability of mRNA;

4. Affect the translation of mRNA.

A deeper search

In the past, most synonymous mutations that affect cancer were discovered by chance, and there was a lack of systematic analysis. It was not until a study in 2019 that some progress was made. In this study, researchers analyzed a database containing nearly 3 million mutations from more than 18,000 tissue samples from 88 types of tumors. Based on this, they created a new database containing more than 650,000 synonymous mutations. The data showed that, like non-synonymous mutations, synonymous mutations also tend to cluster in genes related to cancer [10].

Based on this conclusion, the research team selected the famous KRAS oncogene, found synonymous mutations in it, and continued to conduct further detailed research. One by one, they used plasmids (see "What is a plasmid? From biological weapons to genetically modified foods, it is related to it") to express mutant genes in human cell lines. As expected, some synonymous mutations will lead to increased levels of KRAS protein, which means an increased risk of cancer in the organism.

As mentioned earlier, synonymous mutations may affect the way mRNA folds, thereby affecting gene expression. To verify this hypothesis, the research team tested the chemical interaction between mRNA and nucleotide-containing solutions to determine the actual structure of mRNA. The results showed that synonymous mutations do affect the folding of mRNA. This is the first evidence that synonymous mutations can change the structure of cancer gene mRNA [11].

However, this study is far from perfect, because the plasmid introduced in the experiment is already spliced ​​genetic material, which is not enough to evaluate the impact of mutations on splicing. If synonymous mutations can be systematically introduced into the nuclear genome of cells, the data obtained will be more convincing; of course, this also poses a greater challenge to the technology.

At the same time, other scientists are also looking for new methods to predict the effects of synonymous mutations on gene expression, mainly protein translation. Tamir Tuller, a computational and synthetic biologist at Tel Aviv University in Israel, is building a computational model to predict the effects of synonymous mutations on protein translation initiation and translation speed.

Helena Persson, a molecular geneticist at Lund University in Sweden, uses the frequency of common codons mutating into rare codons as a new indicator to reflect the impact of synonymous mutations on protein translation, because frequency changes can reflect the speed of translation. Persson's team introduced a synonymous mutation in the estrogen receptor (a transcription factor associated with breast cancer) to convert common codons into rare codons. They found that the translation speed of the protein was reduced, and the translation speed is the key to the correct folding of the protein, which may cause the protein to fail to fold correctly. However, strangely, this phenomenon only has an effect when multiple synonymous mutations occur together, indicating that there may be some connection between different synonymous mutations. This connection may help explain why some patients with specific mutations do not respond to drugs designed for them.

In general, studying how synonymous mutations interact is not only the key to understanding tumor generation, but also the key to understanding tumor sensitivity and resistance. Currently, Persson's team is studying combinations of synonymous mutations to explain this connection.

Distance from “tumor treatment”

In addition to deepening people's understanding of cancer, studying synonymous mutations may also bring some good news to patients. For example, it can be used to improve the screening of familial cancers such as breast cancer [12]. Currently, about 5-10% of breast cancer cases are hereditary, but scientists have only discovered a small number of pathogenic mutations. Other pathogenic mutations may be found in synonymous mutations. In the future, we may be able to include these mutations in the scope of cancer screening, thereby improving cancer screening indicators.

In addition, the study of synonymous mutations also has a role in developing more general cancer diagnostic and prognostic tools. In a 2021 study [13], researchers trained machine learning algorithms using a dataset of synonymous mutations found in human cancer samples to predict cancer type and the likelihood that a patient would survive 10 years after initial diagnosis. When they tested the algorithms on a new subset of data, they found that the accuracy of these algorithms was as good as those trained only on non-synonymous mutations. This predictive power of synonymous mutations does not necessarily point to a disease-driving effect, but it may simply reflect the mutational landscape in the cancer genome. This provides scientists with a strong rationale for incorporating synonymous mutations as biomarkers into tools for tumor classification and determining appropriate treatments.

All research on cancer is ultimately aimed at treatment, and cancers caused by synonymous mutations seem to be more difficult to treat than cancers caused by non-synonymous mutations. Most therapies for cancers caused by non-synonymous mutations are designed to inhibit the pathogenic proteins produced by the mutated gene, while most synonymous mutations may not affect the structure of the final protein at all. But then again, even if synonymous mutations do not affect protein structure but affect other processes of gene expression, they can be treated with the same treatment methods as non-synonymous mutations. For example, mutations that cause changes in mRNA structure can be blocked by an antisense oligonucleotide (antisense oligonucleotides bind to the RNA of the target gene and can affect the expression of the target gene through various mechanisms). Currently, antisense oligonucleotides have been approved for the treatment of some diseases, such as correcting the defective gene splicing that causes Duchenne muscular dystrophy [14].

In recent decades, cancer genomics has made significant progress in the study of mechanisms and treatments of non-synonymous mutations, but research on synonymous mutations has just started. In recent years, there has been new evidence that they have an impact on cancer; although synonymous mutations have been ignored in the past, scientists should now regain their attention to synonymous mutations, perhaps opening up a new research route that will benefit mankind.

References

[1]Liu, F., Calhoun, B., Alam, MS et al. Case report: a synonymous VHL mutation (c.414A > G, p.Pro138Pro) causes pathogenic familial hemangioblastoma through dysregulated splicing. BMC Med Genet 21, 42 (2020). https://doi.org/10.1186/s12881-020-0976-7

[2]https://www.spectrumnews.org/news/synonymous-mosaic-mutations-may-autism-risk/

[3]https://www.cell.com/cell/pdfExtended/S0092-8674(14)00145-7

[4]https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3746936/

[5]https://pubmed.ncbi.nlm.nih.gov/30321177/

[6]https://www.science.org/doi/10.1126/scisignal.abp8972

[7]https://pubmed.ncbi.nlm.nih.gov/12554675/

[8]https://www.science.org/doi/10.1126/science.1135308?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed

[9]https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0163272

[10]https://www.nature.com/articles/s41467-019-10489-2

[11]https://www.nature.com/articles/nprot.2006.249

[12]https://www.createhealth.lth.se/canfaster/finished-canfaster-projects/functional-analysis-of-silent-mutations-in-familiar-breast-cancer/

[13]https://www.nature.com/articles/s41525-021-00229-1

[14]https://practicalneurology.com/news/fda-approves-antisense-oligonucleotide-therapy-for-duchenne-muscular-dystrophy-subtype

This article is supported by the Science Popularization China Starry Sky Project

Produced by: China Association for Science and Technology Department of Science Popularization

Producer: China Science and Technology Press Co., Ltd., Beijing Zhongke Xinghe Culture Media Co., Ltd.