LOH, in its broadest sense, simply says that a locus for which a cell is heterozygous, meaning that the 2 parental alleles are distinguishable, becomes homozygous. This is often seen in cancer, where a locus that is heterozygous in a normal cell becomes homozygous in a cancer cell derived from this normal cell. The important issue is how does this happen and what does it mean. By far, the most frequent mechanism is a cell cycle error where a chromosome (let us say it is the paternal allele) gets duplicated as a result of this error, leading to transient triploidy, followed by loss of one of the 3 chromosomes. If the chromosome that gets lost is not the one that was duplicated (i.e.: the maternal allele), then you get loss of that allele resulting in loss of heterozygosity. An important point is that there is no net loss of DNA in that process. One parental allele is lost, but the other is duplicated so you end up with a pair of 2 chromosomes, which is what you started with. Other mechanisms include deletions or other chromosome rearrangements, but the one that I explained above is the most important, as evidenced by the fact that loss of heterozygosity is usually not associated with loss of any genetic material. This is an important mechanism of tumor suppressor gene inactivation, as it represents the second hit in Knudsen's two-hit hypothesis for tumor suppressor inactivation.

An loss of heterozygosity is (like you've described) the loss of one allele, making the other allele hemizygote. You are basically losing a stretch of a chromosome and thus rendering the left-over . In hereditary cancer, usually a substitution in a tumor suppressor gene is the "first hit", which means it renders one allele inactive. Now, if the second allele by chance is lost through random mutation, now both alleles are defunctional and the tumor suppressor gene is inactivated totally.

The mutation itself is not the trigger for the LOH. The second allele usually is deleted by chance, however, this may be a growth advantage for the tumor, having a gene which controls growth rendered inactive.

LOH may also be found in sporadic cancer. Here, both mutations are acquired somatically.

Methylation does not lead to the LOH, only the physical loss of DNA defines a LOH.

And a last thing: LOH can also be copy number neutral (sometimes referred to as cnLOH). So, after one allele has been deleted, the remaining allele may be duplicated again. In this case, the copy number of the region looks normal, but the DNA is missing heterozygous mutations. If a damaging mutation was in one gene, you still have no functioning copy in the cnLOH state.

Hope that helps

Methylation at a promoter would not be classified as LOH and would not be detected as such by conventional means, which could be whole exome sequencing or targeted sequencing for a particular locus.   The classic examples of LOH are Wilms' tumor or Retinoblastoma, caused by the complete loss of a tumor suppressor gene.  These can be familial where a germline mutant copy is inherited.  When the wild-type allele is lost or mutated, the tumor forms.  These tumors can be clonal, i.e. all derived form a single LOH event in a single cell.   Yes, it is possible to have LOH in sporadic cancers as well.  We also see LOH in other diseases where a mutant and wild-type allele are present, such as in Polycystic Kidney Disease.   The genetics of these disease are called dominant, because it only takes one mutant allele to get the disease phenotype, even though at a cellular level they are actually recessive in that both alleles are mutated or lost.   

Thank you for your reply, I really appreciate it. So Peter, you say a LOH only happens when there is a physical loss and by consequence, promotor methylation can't be considered like a trigger of LOH nevertheless, it can be considered like 1st or 2nd hit for a tumor suppressor gene is the target (in the case where the methylation causes a silencing in the gene expression), right ? I think I'm confused because I mix a genetic term (LOH) and mechanism term (silencing of the gene by DNA methylation).

And when you say "The mutation itself is not the trigger for the LOH", you want to say if the mutation doesn't trigger a loss of function, there is no LOH but if the mutation triggers a loss of function (if it's second allele which is the target), we can say it's a LOH ? Sorry about all those question but I totally understand the principle but I'm totally confused with different terms

Ok I think I have seen the light in the end of the tunnel! Actually, my problem was just I mixed mutation and LOH, but when you have a mutation in a gene, even it's a loss of function mutation, you just create a new version of the allele. But to have a LOH, you have to delete the entire allele, like with a chromosome part deletion for exemple, and in this case, you lose the allele and not create a new one. Is it right ?

I would say the mutation of the first allele is not the trigger for the loss of the second allele. However, the loss of the second allele might be beneficial for the tumor and thus  it might look liike the trigger, but the mutation itself is just random chance (or "bad luck", if you wish).

Also, LOH has per se nothing to do with loss of function or loss of protein synthesis of a respective gene. It is just a loss of heterozygosity in a region, where parts of a chromosome which intially showed some heterozygous SNPs are now homozygous (or hemizygous to be precise)

ok thank you I think I more understand now ! Have a good day!

LOH, in its broadest sense, simply says that a locus for which a cell is heterozygous, meaning that the 2 parental alleles are distinguishable, becomes homozygous. This is often seen in cancer, where a locus that is heterozygous in a normal cell becomes homozygous in a cancer cell derived from this normal cell. The important issue is how does this happen and what does it mean. By far, the most frequent mechanism is a cell cycle error where a chromosome (let us say it is the paternal allele) gets duplicated as a result of this error, leading to transient triploidy, followed by loss of one of the 3 chromosomes. If the chromosome that gets lost is not the one that was duplicated (i.e.: the maternal allele), then you get loss of that allele resulting in loss of heterozygosity. An important point is that there is no net loss of DNA in that process. One parental allele is lost, but the other is duplicated so you end up with a pair of 2 chromosomes, which is what you started with. Other mechanisms include deletions or other chromosome rearrangements, but the one that I explained above is the most important, as evidenced by the fact that loss of heterozygosity is usually not associated with loss of any genetic material. This is an important mechanism of tumor suppressor gene inactivation, as it represents the second hit in Knudsen's two-hit hypothesis for tumor suppressor inactivation.

Thank you Louis, it's really helpful ! I think thanks to all of you guys, I get it now.

Hi everybody!

Concerning what was discussed in this interesting topic, I have some confused data from NGS and I need some help/advice.

We performed NGS in several members with a protein deficiency and found a mutation that  triggers a loss of function (previously described). We then checked  for genetic markers inherited with the mutation (haplotype) that all carriers share (also previously described). All carriers confirmed the haplotype but one seems to have a small LOH.

The gene is about 20kbp and there is a median of 20-25 SNP distributed within it. The rare case that break the haplotype, inherited only the mutation from a parent (this last has the complete haplotye) and 4 additional SNPs not related with the haplotype. 

Apparently, there is a partial LOH of two parts, in the beginning of the gene (5kb aprox) and in the end (5kb aprox), could that be possible? do you have an explanation? the gene has a small size, recombination might be difficult, right?

Btw, we validated all these date by Sanger sequencing in order to be sure about what we found in NGS. This rare case has a normal methylated CpG islands. We also perform MPLA to detect possible deletions but nothing showed up.

I would greatly appreciate your help!

Best,

Salam

Intresting Study Salam Salloum-Asfar .

This is very very rare even, I thin there is something fishy with the gene. Can you share the gene name??

Loss of heterozygosity (LOH) is a common genetic event in cancer development, and is known to be involved in the somatic loss of wild-type alleles in many inherited cancer syndromes. The wider involvement of LOH in cancer is assumed to relate to unmasking a somatically mutated tumour suppressor gene through loss of the wild type allele.

Abbreviation LOH; also allelic loss; can be experimentally demonstrated in cases in which the two alleles differ. It is the loss of an allele in tumor DNA compared to matched normal DNA from the same individual. LOH is a very frequent somatic genetic change in human tumors. When LOH is found to occur at high frequency in a particular chromosome region, it is generally considered indicative of the location of a tumor suppressor gene whose loss/inactivation occurs by a “two-hit” mechanism (i.e., physical loss of one copy of an allele and mutation or other genetic/epigenetic alteration of the other copy of the same allele; Knudson Hypothesis). LOH can be identified in cancer cells by a number of technical approaches that all make use of the fact that the two alleles on the two homologous parental chromosomes differ in their DNA sequence. Consequently, normal cells can be shown to be heterozygousfor a given chromosomal region. If cancer cells of the same person lack this...