How is melanoma formed
Genotyping of individual skin melanocytes revealed the ways of their malignant transformation
Vyacheslav Kalinin, "Elements"
Fig. 1. The scheme of the work under discussion. On the left – the human epidermis consists of several types of cells, but the vast majority of them are keratinocytes of varying degrees of "maturity". Tree-like melanocytes are mainly distributed singly in the basal layer of the epidermis. With the help of their branched processes, they distribute the pigment melanin to the surrounding keratinocytes to protect against solar ultraviolet radiation. On the right, scientists isolated individual melanocytes from donor skin samples, grew them in a culture medium until small colonies were formed, and then performed amplification and complete sequencing of genomes. In parallel, RNA amplification and sequencing were performed. The results were compared, and only those that were found in both DNA and RNA were recognized as true mutations. Analysis of these mutations allowed us to determine which of them lead to malignant transformation of melanocytes. A drawing from the popular synopsis to the article under discussion in Nature.
Melanomas are among the most aggressive and deadly forms of cancer. At the same time, scientists still know very little about the early stages of melanoma formation. One way to deal with this is to monitor how mutations accumulate in individual melanocytes (cells of the epidermis responsible for skin pigmentation). American scientists have developed a new method for this, which made it possible to compare the genomes of individual melanocytes from different skin areas of several donors with a minimum of errors. As expected, more mutations accumulated in phenotypically normal melanocytes regularly exposed to the mutagenic action of sunlight (for example, from the face or neck) than in melanocytes taken from almost always sun-protected areas of the body. But unexpectedly it turned out that melanocytes from the arms, thighs or back (areas of the skin that receive an intermediate dose of solar ultraviolet radiation during their lifetime) had even more mutations. It has been established that there are many mutations in normal melanocytes that are considered oncogenic, but among them there are no "strong" oncogenic mutations characteristic of melanomas. The scientists also presented a general scheme of ways of additional mutagenesis, which turns normal melanocytes carrying "weak" mutations into malignant ones, from which melanomas subsequently develop.
The epidermis is the outer and thinnest (its thickness is only about 0.1 mm) layer of human skin. Basically (about 90%) it consists of keratinocytes. These cells appear in the deepest – basal – layer of the epidermis as a result of the differentiation of stem cells (thanks to which the skin is constantly renewed). The main functions of keratinocytes are protective and structural. During his life (which lasts 1.5–2 months, see K. M. Halprin, 1972. Epidermal "turnover time" – a re-examination) the keratinocyte shifts to the outer border of the epidermis. At the same time, it gradually flattens and, to put it very briefly, "hardens". Having reached the surface of the skin, the keratinocyte is already just a horny scale, which is quickly exfoliated, giving way to the following.
Another important type of epidermis cells are melanocytes. They produce the pigment melanin, which is responsible for the color of our skin. With the help of melanosomes, melanin is distributed to the surrounding keratinocytes and stored in them, fulfilling its role – to protect skin cells from ultraviolet light (primarily solar). Normally, melanocytes live in the basal layer of the epidermis.
During life, random mutations arise and accumulate in the somatic cells of the body. As a result, a unique set of mutations is formed in each cell. The absolute majority of these mutations are neutral and do not have any significant effect on the quality of the cell, but some may affect the key genes that control the behavior of the cell and direct it along the path of malignant degeneration - uncontrolled division, which results in cancerous tumors.
The main factor provoking skin cancers is the ultraviolet region of sunlight. The cells of the epidermis, designed to protect the body from UV radiation, also suffer from it. In part, this effect is offset by the constant renewal of keratinocytes. But with age, mutations still accumulate, and the risk of developing cancer increases. And if cancerous tumors originating from keratinocytes (for example, basal cell carcinomas and squamous cell carcinomas) are treated very successfully, then the situation is much worse with melanomas arising from melanocytes. Melanomas are very aggressive, tend to form metastases early in distant organs and tissues, and therefore often end in death. Early diagnosis of melanoma is complicated by the fact that at the initial stages of tumor development, it looks like harmless moles formed by the local proliferation of normal melanocytes. Melanomas are also usually caused by solar ultraviolet radiation, but this is not the only reason. For example, some moles can degenerate into a tumor if certain mutations occur in them that are not related to UV radiation. According to the degree of accumulation of damage associated with solar irradiation (cumulative sun damage, CSD) WHO divides melanomas into two main subtypes: low-CSD (low-CSD) and high-CSD (high-CSD). Both are very aggressive and often deadly.
Keratinocytes have a special program that depends on the p53 protein. After irradiation with a large dose of ultraviolet light, it triggers apoptosis, which leads to peeling of the epidermis after, for example, a burn with solar ultraviolet light. The TP53 gene encoding this protein can also be damaged due to mutations. Clones of keratinocytes with mutant TP53, despite the constant renewal of the skin, can remain in its structure and potentially develop into cancer. This is the general outline of the development of tumors from keratinocytes.
At the same time, the mechanisms leading to the onset of malignant transformation of melanocytes into melanomas are not known even to this extent. Deciphering exactly where mutations occur and at what rate they accumulate in precancerous melanocytes of normal skin can provide important information about the early stages of transformation – before the neoplastic proliferation and transformation of normal cells into cancer cells and their uncontrolled reproduction becomes apparent. This is exactly the task that American scientists under the leadership of Hunter Shain (A. Hunter Shain) from the University of California at San Francisco.
Previously, oncogenic mutations have been repeatedly studied on small fragments of various organs and tissues, including skin (see, for example, I. Martincorena et al., 2015. High burden and pervasive positive selection of somatic mutations in normal human skin). For melanocytes, such an approach is impossible, since they are represented by rare inclusions among keratinocytes, and genotyping on individual cells is required. But direct sequencing of the genome of individual cells requires preliminary amplification of cellular DNA, and in this case large fragments of the genome are usually lost and many errors accumulate, which are inevitable during amplification even with high-precision DNA polymerases. An alternative approach involves the preliminary reproduction of single cells in vitro in order to increase the amount of DNA analyzed. But at the same time there are mutations that were not in the original cells. The disadvantages of these approaches were not allowed to be used in the work under discussion, because scientists wanted to determine as accurately as possible the difference in mutagenesis between individual melanocytes. As a result, scientists combined both approaches to take the best out of each.
The experiments used 19 small fragments of normal skin morphology from six deceased donors of European origin aged 63 to 85 years. Two of them had skin cancer, four did not. A total of 133 melanocytes were examined. The fragments were taken from different skin areas with different degrees of exposure to solar radiation. Areas with a high level of exposure included, for example, the face, neck and scalp of bald people; areas with an average level of exposure – the skin of the back and thighs; areas with a low level – the skin from those parts of the body that are almost always hidden under clothing in most people. The epidermis cells obtained from the skin were slightly grown in a culture medium, individual melanocytes were isolated from them and cultured only for a short time, minimizing the possibility of additional mutations. When cultured, 38% of the isolated melanocytes formed small colonies (from 2 to 3000 cells, on average – 184). Most of them were of sufficient size for further analysis. Both DNA and RNA were isolated, amplified, and sequenced from melanocyte colonies. This made it possible to obtain and compare both the genotype (DNA) and the manifested phenotype of melanocytes (RNA), excluding amplification errors. An additional criterion for the elimination of errors was the presence of normal polymorphisms of the DNA structure (neutral variations of the nucleotide sequence) near the suspected mutations (Fig. 2).
Fig. 2. Methods by which scientists separated true mutations from amplification artifacts. At first, only those mutations that were detected in both DNA and RNA were isolated (left). Then, to further verify the truth of the mutations, the scientists determined which single nucleotide polymorphisms (SNPs) they are associated with (right) – only those mutations that uniquely corresponded to specific variants of polymorphism were tested. A drawing from the discussed article in Nature.
According to the results of DNA and RNA sequencing for each clone, the results were compared for 509 genes most commonly found in various malignant tumors, considering them oncogenes that may be associated with melanoma. On average, the mutation rate was 7.9 per megabase (MB), but this value varied greatly: the spread was from less than 0.82 to 32.3 mutations per MB.
As expected, melanocytes from the most sun-protected areas of the skin had significantly fewer mutations than those from the "most irradiated". But unexpectedly it turned out that melanocytes with an "intermediate" radiation dose have even more mutations (Fig. 3, left).
Fig. 3. On the left – melanocytes taken from skin areas that are irregularly exposed to solar ultraviolet (for example, the skin of the hips or back; intermittently sun exposed) had more mutations than melanocytes from both the most exposed skin areas (face, neck; chronically sun exposed) and the most protected (groin area; sun shielded). On the right – some melanocytes from areas of healthy skin near melanoma could have more mutations than the cells of the melanoma itself. Graphs from the discussed article in Nature.
There is no reliable explanation for this observation yet. It is possible that it is associated with different rates of mutation in the genomes of melanocytes of different parts of the body or with their accelerated death and renewal on exposed parts of the body in the sun. But it is in good agreement with the well-known fact that melanomas on parts of the body sometimes exposed to the sun occur disproportionately more often compared to other forms of skin cancer.
The frequency of mutations in different donors varied greatly. Moreover, the following was observed in samples from donors with skin cancer: in some melanocytes from healthy skin areas adjacent to melanoma, the mutation rate was noticeably higher than in melanoma as such (Fig. 3, right). This situation cannot be called typical. For example, in colorectal cancers, the frequency of mutations in tumors is much higher than in adjacent tissues.
Another surprise was the large variation in the frequency of mutations in individual melanocytes isolated from one small (approximately 3 cm2) area of the skin. To find the reasons for these differences, scientists compared the expression levels of a number of genes in such melanocytes. The most convincing correlation was obtained for the MDM2 gene, the product of which causes degradation of the main antitumor and antimutation protein p53. Significantly more mutations were observed in melanocytes with enhanced MDM2 expression than in their neighbors with normal levels. But the possible reasons for these differences are not limited to different levels of MDM2 expression, and further research is required to uncover these reasons.
Comparison of a set of mutations in melanomas and melanocytes from areas of normal skin adjacent to tumors revealed their significant differences. Consequently, the researchers concluded, tumor cells do not "jump" into the areas of the surrounding skin and vice versa.
Further, the authors analyzed the occurrence of mutations in melanocytes, for which their role as drivers of malignant cell degeneration in other cancers was previously shown. Most of these mutations were associated with the signaling pathway of MARK, which controls survival, activation of mobility and, most importantly, activation of cell division. But mostly it was the so-called "weak" oncogenic mutations. So, among them there was no "strong" mutation BRAFV600E, which is very common in melanomas – the main driver of the development of low-CSD melanomas. Mutations found in normal melanocytes are relatively weak activators of the MARK pathway and cannot be active drivers of oncogenesis. For the formation of high-CSD melanomas, for which they are characteristic, additional driver mutations are needed.
Fig. 4. Diagram of possible ways of melanoma occurrence. A drawing from the discussed article in Nature.
Mutations were also found in genes controlling the cell cycle and in genes controlling chromatin modifications. As such, these mutations cannot independently induce the neoplastic process, but they can accelerate it in cells with mutations in the genes of the MARK pathway. It is noteworthy that no mutations were found in the promoter of the TERT gene (reverse transcriptase, a component of telomerase) in melanocytes. Despite the fact that such a mutation is often found in melanomas, it hardly somehow stimulates the process of neoplastic transformation of melanocytes.
The results obtained represent extremely detailed data on the structure of the genomes of normal melanocytes. They also made it possible to obtain detailed information on mutagenesis caused by solar ultraviolet radiation, and on the ways of occurrence of melanomas. The developed methodological approach for analyzing the genomes of individual melanocytes can be used to analyze the genomes of individual cells of other organs and tissues.
A source: Tang et al., The genomic landscapes of individual melanocytes from human skin // Nature. 2020.
Portal "Eternal youth" http://vechnayamolodost.ru