-
IP addresses are NOT logged in this forum so there's no point asking. Please note that this forum is full of homophobes, racists, lunatics, schizophrenics & absolute nut jobs with a smattering of geniuses, Chinese chauvinists, Moderate Muslims and last but not least a couple of "know-it-alls" constantly sprouting their dubious wisdom. If you believe that content generated by unsavory characters might cause you offense PLEASE LEAVE NOW! Sammyboy Admin and Staff are not responsible for your hurt feelings should you choose to read any of the content here. The OTHER forum is HERE so please stop asking.
It has been suggested that haplogroup N is associated to good outcome of gastric cancer clinical evolution when compared to haplogroup M.
- Thread starter ginfreely
- Start date
Abstract
Mitochondria are organelles that perform major roles in cellular operation. Thus, alterations in mitochondrial genome (mtGenome) may lead to mitochondrial dysfunction and cellular deregulation, influencing carcinogenesis. Gastric cancer (GC) is one of the most incident and mortal types of cancer in Brazil, particularly in the Amazon region. Here, we sequenced and compared the whole mtGenome extracted from FFPE tissue samples of GC patients (tumor and internal control – IC) and cancer-free individuals (external control – EC) from this region. We found 3-fold more variants and up to 9-fold more heteroplasmic regions in tumor when compared to paired IC samples. Moreover, tumor presented more heteroplasmic variants when compared to EC, while IC and EC showed no significant difference when compared to each other. Tumor also presented substantially more variants in the following regions: MT-RNR1, MT-ND5, MT-ND4, MT-ND2, MT-DLOOP1 and MT-CO1. In addition, our haplogroup results indicate an association of Native American ancestry (particularly haplogroup C) to gastric cancer development. To the best of our knowledge, this is the first study to sequence the whole mtGenome from FFPE samples and to apply mtGenome analysis in association to GC in Brazil.
Introduction
Mitochondria are cytoplasmic organelles that perform major roles in cell operation, including energy generation through oxidative phosphorylation (OXPHOS), cell death, calcium levels control, lipid homeostasis and metabolic cell signaling1. These organelles have their own genome (mtGenome), with 16,569 bp of length and 37 genes, of which 13 are protein-coding genes involved in OXPHOS, 22 are transfer RNA (tRNA) genes and two are ribosomal RNA (rRNA) genes2,3. It also presents a non-coding control region known as displacement loop (D-loop), essential for replication and transcription regulation4. There are many copies of mitochondria in each cell and such copies may present different alleles for the same variant, a state called heteroplasmy.It is well-known that mitochondrial DNA (mtDNA) is more susceptible to alterations in comparison to nuclear DNA5,6,7. These alterations may lead to mitochondrial dysfunction, which in turn may account for cellular deregulation due to DNA repair defects, leading to the development of different diseases, such as cancer8. It is notable that hallmarks of cancer (i.e. abilities acquired by tumor during carcinogenesis to survive and proliferate) include energy deregulation and evasion of cell death, both directly related to mitochondrial function9. In fact, many studies have shown an association of mtDNA instability and heteroplasmy to different types of cancer4,10,11,12,13,14,15.
Last edited:
One of these types of cancer is gastric cancer (GC), which is currently the fifth most incident and the third most lethal type of cancer worldwide16. In Brazil, GC is also one of the most frequent and aggressive types of cancer, being the sixth most incident and the fifth most lethal cancer16. This is even more alarming in the North region of Brazil, where GC is the second most incident type of cancer among men and the fifth among women17, probably due to eating habits and genetic background of the population.
Regarding genetic ancestry, it is important to highlight that, in addition to nuclear DNA, mtDNA also provides such information through haplogroups, which are basically groups of haplotypes. These provide information about women migration and have also been associated to development and outcome of different diseases, including GC18,19,20,21.
Moreover, GC usually presents an unfavorable clinic evolution because of nonspecific symptoms at early stages, leading to late diagnosis and a poor prognosis22. Thus, it is crucial to search for genetic markers that would allow an earlier detection and improve patient outcome.
In the last decade, with the advent of new technologies such as Next-Generation Sequencing (NGS), one focus of oncologic research has been high-throughput analyses of human genome related to cancer development. However, to this date, not many studies have investigated the association of mtGenome alterations to cancer, especially GC. For instance, a recent review has pointed out 16 studies associating mtDNA alterations to gastric cancer, but none involved NGS approaches23.
In this study, we sequenced the whole mtGenome in order to assess and compare variants and their heteroplasmy levels in FFPE samples from gastric cancer patients (paired samples, i.e. samples of both tumor and non-tumoral tissues) and cancer-free individuals from the North region of Brazil. To the best of our knowledge, this is the first study to perform such analysis of gastric cancer in Brazil.
Regarding genetic ancestry, it is important to highlight that, in addition to nuclear DNA, mtDNA also provides such information through haplogroups, which are basically groups of haplotypes. These provide information about women migration and have also been associated to development and outcome of different diseases, including GC18,19,20,21.
Moreover, GC usually presents an unfavorable clinic evolution because of nonspecific symptoms at early stages, leading to late diagnosis and a poor prognosis22. Thus, it is crucial to search for genetic markers that would allow an earlier detection and improve patient outcome.
In the last decade, with the advent of new technologies such as Next-Generation Sequencing (NGS), one focus of oncologic research has been high-throughput analyses of human genome related to cancer development. However, to this date, not many studies have investigated the association of mtGenome alterations to cancer, especially GC. For instance, a recent review has pointed out 16 studies associating mtDNA alterations to gastric cancer, but none involved NGS approaches23.
In this study, we sequenced the whole mtGenome in order to assess and compare variants and their heteroplasmy levels in FFPE samples from gastric cancer patients (paired samples, i.e. samples of both tumor and non-tumoral tissues) and cancer-free individuals from the North region of Brazil. To the best of our knowledge, this is the first study to perform such analysis of gastric cancer in Brazil.
Discussion
In this study, we investigated the association of mitochondrial variants with gastric cancer in individuals from a population of the Brazilian Amazon. This was done through whole mtGenome sequencing of FFPE samples from patients and cancer-free individuals. To the best of our knowledge, this is the first study to successfully sequence mtGenome from FFPE samples.This is particularly important because DNA sequencing from this type of material can be challenging. Although good for long-term storage of tissue samples, the FFPE process may affect the material integrity so that the extracted DNA is generally of low quality, often leading to the loss of samples. However, some studies performing NGS have shown that DNA extracted from FFPE tissue samples can still provide results in concordance to frozen tissues27,28,29. The challenge presented by FFPE samples is probably the reason for the loss of samples and for the limited coverage in our study. In spite of that, we were able to successfully sequence the mtGenome, obtaining interesting results.
When we compared the samples from the same eight gastric cancer patients, i.e. paired tumor and internal control samples, we observed significantly more somatic variants present in tumor samples than in the respective internal control samples (up to 5.5-fold more), a kind of pattern that has been reported in other types of cancer, such as colorectal cancer and penile cancer11,30. In addition, studies comparing mtDNA variants and genomic instability of different types of cancer reported that gastric cancer was among the types with more somatic variants and genomic instability with a tumor-specific pattern10,25.
Regarding the mitochondrial genes in which most of the tumor-exclusive variants were found, we observed that MT-DLOOP2, MT-DLOOP1 and MT-ND5 were among the regions with more variants in at least half of the tumor samples included in this analysis (Table 2). There was no similar pattern observed when we considered the regions with more variants exclusive to internal control. As for shared variants, they were similarly distributed among different genes. Additionally, there was a mean of 2.43-fold increase in the number of genes with exclusive variants when we compared tumor and internal control samples. It is noteworthy that most of these variants are present in more than one sample and are counted as such in this analysis.
Therefore, when we considered single variants only, there were 50 variants present only in tumor samples and most of them was distributed in four regions – MT-DLOOP1 (16%), MT-ND5 (16%), MT-ND4 (10%) and MT-DLOOP2 (10%) –, while there were eight variants present only in internal control samples, of which most was found in MT-ND5 (25%). This suggests that, in tumor, there is an increase not only in the number of variants, but also in the regions in which such variants occur.
Last edited:
This greater presence of variants in tumor samples was reinforced when we grouped and compared the eight pairs. In this analysis, we found 25 tumor-exclusive somatic variants and eight somatic variants that were exclusive to internal control samples, as well as 36 shared variants. While some variants that were considered somatic in the pair-by-pair comparison were then considered germline, tumor samples still showed 3-fold more variants than internal control samples.
Such tumor-exclusive variants were distributed in 13 genes, of which three (MT-ATP8, MT-ND4L and MT-TG) were only observed in this group. Only a few studies have reported variants in MT-ATP831,32and MT-ND4L31,33,34,35 in different types of cancer and none was in gastric cancer. No studies were found associating variants in MT-TGto cancer.
Furthermore, two genes (MT-CO3and MT-RNR1) only presented shared variants, which are probably not involved in tumorigenesis. However, five genes (MT-CO1, MT-CYB, MT-DLOOP1, MT-ND1 and MT-ND5) not only presented germline variants, but also different somatic variants exclusive to tumor and internal control groups, indicating a possible association of such genes and variants with tumor development. It is noteworthy that, among these genes, MT-DLOOP1and MT-ND5 stand out for presenting more germline variants than the others.
In addition to the shared variants, MT-ND5 also presents a high rate of exclusive variants not only in tumor samples but also in internal control samples, suggesting that this gene could also be altered in cells in other adjacent organs of cancer patients, contributing to possible tumor advances. Regardless, MT-ND5encodes a core subunit of Complex I, essential for electron transport from NADH to ubiquinone in the mitochondrial respiratory chain, so that variants in this gene and others related to this process may lead to impairment in OXPHOS and an increase of reactive oxygen species (ROS) generation, which can contribute to cancer proliferation and metastasis36,37. In fact, a high rate of mutations in MT-ND5 has been reported in esophageal cancer38.
Moreover, the regions with more tumor-exclusive variants were the MT-DLOOP1 and MT-DLOOP2 (16% each). This corroborates previous studies that associated an increase of variants in such control regions to the development of different types of cancer, including hepatocellular carcinoma39, brain tumor40, oral squamous cell carcinoma41, colon cancer42 and gastric cancer43. Further, a study have associated five variants in these genes with gastric cancer, including A73G and T16519C44, which we found frequently in our tumor samples but not in internal control samples.
Such tumor-exclusive variants were distributed in 13 genes, of which three (MT-ATP8, MT-ND4L and MT-TG) were only observed in this group. Only a few studies have reported variants in MT-ATP831,32and MT-ND4L31,33,34,35 in different types of cancer and none was in gastric cancer. No studies were found associating variants in MT-TGto cancer.
Furthermore, two genes (MT-CO3and MT-RNR1) only presented shared variants, which are probably not involved in tumorigenesis. However, five genes (MT-CO1, MT-CYB, MT-DLOOP1, MT-ND1 and MT-ND5) not only presented germline variants, but also different somatic variants exclusive to tumor and internal control groups, indicating a possible association of such genes and variants with tumor development. It is noteworthy that, among these genes, MT-DLOOP1and MT-ND5 stand out for presenting more germline variants than the others.
In addition to the shared variants, MT-ND5 also presents a high rate of exclusive variants not only in tumor samples but also in internal control samples, suggesting that this gene could also be altered in cells in other adjacent organs of cancer patients, contributing to possible tumor advances. Regardless, MT-ND5encodes a core subunit of Complex I, essential for electron transport from NADH to ubiquinone in the mitochondrial respiratory chain, so that variants in this gene and others related to this process may lead to impairment in OXPHOS and an increase of reactive oxygen species (ROS) generation, which can contribute to cancer proliferation and metastasis36,37. In fact, a high rate of mutations in MT-ND5 has been reported in esophageal cancer38.
Moreover, the regions with more tumor-exclusive variants were the MT-DLOOP1 and MT-DLOOP2 (16% each). This corroborates previous studies that associated an increase of variants in such control regions to the development of different types of cancer, including hepatocellular carcinoma39, brain tumor40, oral squamous cell carcinoma41, colon cancer42 and gastric cancer43. Further, a study have associated five variants in these genes with gastric cancer, including A73G and T16519C44, which we found frequently in our tumor samples but not in internal control samples.
We also obtained interesting results in the paired analysis of heteroplasmy levels. In most cases, tumor presented notably more heteroplasmic regions than the paired internal control (3 to 9-fold). About these heteroplasmic regions, we highlight that: (i) the four most commonly heteroplasmic genes in tumor samples were not heteroplasmic in all internal control samples; (ii) four genes were heteroplasmic in almost all tumor samples, but not as frequent in internal control genes; (iii) seven genes were heteroplasmic only in tumor samples; and (iv) there were no genes heteroplasmic only in the internal control group. These results are suggestive of an involvement of such heteroplasmic variants in GC development.
This was reinforced in our analyses comparing heteroplasmy distribution between all groups of samples (tumor, internal control and external control). When we compared heteroplasmy level means, the number of variants was different between groups, so we normalized the number of single heteroplasmic variants in all groups and found a statistically significant difference from tumor to internal control (P 0.025) and external control (P 0.006). These control groups did not present such difference between each other (P 0.906).
Therefore, the number of heteroplasmic variants might be as important or even more important to cancer development than heteroplasmy levels per se. In the individual heteroplasmy mean comparison, we observed that individual tumor samples presented almost 3-fold more heteroplasmic variants than internal control samples. A study conducted with patients of hepatocellular carcinoma also reported that tumor samples presented more heteroplasmic variants and a higher heteroplasmy ratio than their matched cancer-free samples45.
Moreover, tumor group presented 39% of all heteroplasmic variants, while internal control group presented 9% and external control group presented 52%. This is especially interesting given that tumor group had about a third of the number of external control samples, but about half of tumor group was composed of samples with a high number of heteroplasmic variants, as seen in Fig. 3. When compared to the control groups, tumor group presented a statistically significant difference regarding this distribution of heteroplasmic variants (P 0.027 for internal control and P 0.008 for external control). There was no significant difference between both controls (P 0.903).
As for location of heteroplasmic variants, out of the 37 mitochondrial genes, 32 presented variants in at least one group, out of which 15 regions carried such variants in all three groups, allowing the comparison shown in Fig. 5. Among these regions, 12 were protein-coding and three were non-protein-coding genes (DLOOP and rRNA only). Further, the higher rate of heteroplasmic variants in tumor groups was statistically significant in six of such regions: MT-RNR1, MT-ND5, MT-ND4, MT-ND2, MT-DLOOP1 and MT-CO1.
This was reinforced in our analyses comparing heteroplasmy distribution between all groups of samples (tumor, internal control and external control). When we compared heteroplasmy level means, the number of variants was different between groups, so we normalized the number of single heteroplasmic variants in all groups and found a statistically significant difference from tumor to internal control (P 0.025) and external control (P 0.006). These control groups did not present such difference between each other (P 0.906).
Therefore, the number of heteroplasmic variants might be as important or even more important to cancer development than heteroplasmy levels per se. In the individual heteroplasmy mean comparison, we observed that individual tumor samples presented almost 3-fold more heteroplasmic variants than internal control samples. A study conducted with patients of hepatocellular carcinoma also reported that tumor samples presented more heteroplasmic variants and a higher heteroplasmy ratio than their matched cancer-free samples45.
Moreover, tumor group presented 39% of all heteroplasmic variants, while internal control group presented 9% and external control group presented 52%. This is especially interesting given that tumor group had about a third of the number of external control samples, but about half of tumor group was composed of samples with a high number of heteroplasmic variants, as seen in Fig. 3. When compared to the control groups, tumor group presented a statistically significant difference regarding this distribution of heteroplasmic variants (P 0.027 for internal control and P 0.008 for external control). There was no significant difference between both controls (P 0.903).
As for location of heteroplasmic variants, out of the 37 mitochondrial genes, 32 presented variants in at least one group, out of which 15 regions carried such variants in all three groups, allowing the comparison shown in Fig. 5. Among these regions, 12 were protein-coding and three were non-protein-coding genes (DLOOP and rRNA only). Further, the higher rate of heteroplasmic variants in tumor groups was statistically significant in six of such regions: MT-RNR1, MT-ND5, MT-ND4, MT-ND2, MT-DLOOP1 and MT-CO1.
In the current specialized literature, not many studies were found associating heteroplasmic variants to the development of gastric cancer, regardless of mitochondrial genomic location, including the six genes above. A study performed showed an association of heteroplasmic variants in DLOOP regions to gastric cancer development in a Chinese population46. Another study have indicated homoplasmic and heteroplasmic variants in mitochondrial 12 S rRNA (encoded by MT-RNR1) to the development of intestinal-type gastric cancer47. No studies were found associating heteroplasmic variants in the four other genes to GC development.
In the mitochondrial haplogroup comparison between case and control, we observed 17 macro-haplogroups, distributed in four main ancestries: Native American (A, B, C and D), European (H, HV, J and V), African (L0, L1, L2, L3 and L3e) and Asian (M, N, R and Y).
Our results showed Native American ancestry with more variants than the other three ancestries in both cases and controls, which could be related to the great number of individuals from this ancestry in each group. Native American haplogroups were the most frequent in both case (36.5%) and control (44.6%). They were followed by European (27.3%), African (18.2%) and Asian (18.2%) in cases and by European and African (25.6% each) and Asian (4.2%) in controls. This ancestry profile based on mtDNA is consistent to the expected for the studied geographic region, considering the admixture process that formed the Brazilian population26,48,49. In addition, it is informative that there was a statistical significance in case group (P 0.047), probably due to the high frequency of Native American haplogroups in this group. This is especially interesting considering that GC is one of the most incident types of cancer in the studied Brazilian region, as previously mentioned.
Moreover, in both case and control groups, the most frequent haplogroup was C (Native American ancestry), but it is important to highlight that this haplogroup accounted for more than 2-fold in case group than in control group (36.4% and 17%, respectively). This suggests that haplogroup C might be related to gastric cancer development, which is reinforced by a study performed with Tibetan patients of gastric cancer, in which haplogroup C was the third most common (accounting for 17%)50.
Considering that the parent haplogroup of C is M51, it is also interesting that this haplogroup is present in our case group, but not in our control group. On the other hand, our study reported sibling haplogroup N in control group, but not in case group. Similarly, it has been suggested that haplogroup N is associated to good outcome of gastric cancer clinical evolution when compared to haplogroup M18. In addition, haplogroup H (European ancestry) was the second most common haplogroup in case group (18.2%). Although this haplogroup has not been previously associated to gastric cancer, it has been indicated that it might influence other types of cancer52.
In the mitochondrial haplogroup comparison between case and control, we observed 17 macro-haplogroups, distributed in four main ancestries: Native American (A, B, C and D), European (H, HV, J and V), African (L0, L1, L2, L3 and L3e) and Asian (M, N, R and Y).
Our results showed Native American ancestry with more variants than the other three ancestries in both cases and controls, which could be related to the great number of individuals from this ancestry in each group. Native American haplogroups were the most frequent in both case (36.5%) and control (44.6%). They were followed by European (27.3%), African (18.2%) and Asian (18.2%) in cases and by European and African (25.6% each) and Asian (4.2%) in controls. This ancestry profile based on mtDNA is consistent to the expected for the studied geographic region, considering the admixture process that formed the Brazilian population26,48,49. In addition, it is informative that there was a statistical significance in case group (P 0.047), probably due to the high frequency of Native American haplogroups in this group. This is especially interesting considering that GC is one of the most incident types of cancer in the studied Brazilian region, as previously mentioned.
Moreover, in both case and control groups, the most frequent haplogroup was C (Native American ancestry), but it is important to highlight that this haplogroup accounted for more than 2-fold in case group than in control group (36.4% and 17%, respectively). This suggests that haplogroup C might be related to gastric cancer development, which is reinforced by a study performed with Tibetan patients of gastric cancer, in which haplogroup C was the third most common (accounting for 17%)50.
Considering that the parent haplogroup of C is M51, it is also interesting that this haplogroup is present in our case group, but not in our control group. On the other hand, our study reported sibling haplogroup N in control group, but not in case group. Similarly, it has been suggested that haplogroup N is associated to good outcome of gastric cancer clinical evolution when compared to haplogroup M18. In addition, haplogroup H (European ancestry) was the second most common haplogroup in case group (18.2%). Although this haplogroup has not been previously associated to gastric cancer, it has been indicated that it might influence other types of cancer52.
Last edited:
Conclusions
In this study, we were able to successfully sequence the whole mitochondrial genome of FFPE samples and associate it to gastric cancer development. To the best of our knowledge, this was the first study to achieve that. We found more mitochondrial variants in tumor group than in controls, suggesting that such variants and mitochondrial heteroplasmy might influence gastric cancer development. In addition, haplogroup C might also be important to the development of this type of cancer. Although limited by sample number, our findings contribute to a greater understanding of mitochondrial influence in gastric cancer. Further, functional studies are recommended to clarify the mechanism of this impact.
Make sense indeed because Brazil people and Hispanic in USA are like Mongolian with high stomach cancer.
Similar threads
