The D1 and D2 deiodinase enzymes are produced in our cells to enable the conversion of FT4 to FT3 to occur within our cells.
Both D1 and D2 enzymes are seleno-proteins, i.e. they have selenium as part of their structure. Hence, if someone has low levels of selenium this can hamper conversion. Many thyroid patients take 100 or 200 mcg of selenium supplement per day to ensure they have sufficient.
The higher the number of these enzymes produced the higher the conversion rate. TSH also has an impact, if TSH goes up the D1 and D2 enzymes are up-regulated, i.e. more are produced in each cell, and the conversion rate of T4 to T3 rises. If TSH goes down, the D1 and D2 enzymes are down-regulated, i.e. fewer are produced and conversion rate lowers (but there is still some conversion occurring).
However, some thyroid patients have gene defects that can mean the quality of the D1 or D2 (or both) enzymes can be impaired. The genes involved in producing the D1 and D2 enzymes are referred to as DIO1 and DIO2. It is possible to have mutations in the genetic material that make up the genes. We receive a good copy of the genetic material from one parent and a mutation from the other parent, in this case, you would be heterozygous for the gene defect. If you have a bad copy from both parents, you are homozygous for the gene defect – which is worse as the quality of the deiodinase enzyme is liable to be worse.
Note: I have heard from one patient whose endocrinologist told her that there is no point in testing these, as if she had them she would not convert T4 to T3 at all (she is very unwell on T4 medication). This is obviously rubbish. The gene defects, even if present with both copies of each DIO1 and DIO2 mutation, impair the quality of the D1 and D2 enzymes, they do not stop them working totally. It just makes conversion worse – it does not totally stop it.
The DIO1 and DIO2 mutations do not always cause obvious conversion problems either. Some people are better at compensating for the defects than others. Geneticists do not fully understand this yet. However, it is this compensation that tends to make the mutations not affect people until their late twenties/early thirties. Perhaps, a higher TSH might help compensate, as more deiodinase enzymes would be produced – but I am speculating. One thing is true though if you have a mutation, it is there genetically in your make-up. It cannot be turned off or removed. It will in some way affect the ability of your cells to make good D1 and D2 enzymes. Over the next ten years, more on all of this will no doubt be discovered.
However, genetic testing will only provide additional information that can back up a suspicion that conversion from T4 to T3 is an issue. Often this testing only gets done when the individual thyroid patient already suspects conversion issues. In some people with the gene defects, there appear to be no conversion problems. In others there appear to be severe conversion issues.
The most important indicator of T4 to T3 conversion issues in a thyroid patient is how that individual responds to treatment. This involves assessing how well they are using symptoms and signs, as well as full thyroid labs that include FT3, FT4, and TSH. This assessment may involve switching from Levothyroxine (T4) medication to T4/T3 medication and perhaps eventually to T3 medication if the thyroid patient does not respond well enough to treatment. Response to treatment is always going to be the most valuable assessment of T4 to T3 conversion.
Having said the above, the DIO1 and DIO2 genetic mutation tests are very useful to do, even if you have to do them privately. Knowing that you have one or both mutations (and whether you have them from one or both parents), will inform you of the potential risk of conversion issues. Both mutations can reduce T4 to T3 conversion, lowering FT3 and raising rT3, thus increasing the chances of on-going hypothyroidism. Having the mutation from both parents make the problem worse, but even from one, it can have an effect.
A DIO1 mutation may affect conversion by the thyroid, liver, and kidneys.
A DIO2 mutation may affect conversion by the brain, pituitary, central nervous system, thyroid, heart, and peripheral tissues (skeletal muscle). Thus, the DIO2 mutation can also cause hypocortisolism, through the impact on the pituitary, i.e. due to HP dysfunction leading to lower ACTH. Note: once you have totally cleared T4 and are reliant on the T3, this conversion issue becomes unimportant, unless it has caused HP dysfunction that is slow to correct.
Both D1 and D2 enzymes are important in the conversion of T4 to T3. More focus has been on the D2 enzymes and DIO2 polymorphisms over recent years. However, DIO1 polymorphisms still have a powerful effect on either increasing or decreasing T3/T4 ratios, so people tend to ignore the importance of DIO1 defects to FT3 levels. The Panicker, Dayan et al study of DIO1 in 2008 showed that this polymorphism affected T3, T4, and RT3 concentrations even in people with healthy thyroids!
Another key insight is that in humans the D1 enzyme has two T3 receptors on it. Neither D2 nor D3 deiodinase enzymes have this. Also, D1 is not as strongly upregulated by TSH as D2 is. Instead, D1 is powerfully upregulated in its ability to convert T4 to T3 by the presence of increasing concentrations of T3.
The D1 deiodinase enzyme is also important in the liver as it involved in the clearance of rT3, so testing for the DIO1 mutation is helpful.
The bottom line is that both DIO1 and DIO2 gene defects are useful to check. I am also confident that significantly more will be learned about the deiodinase enzymes and their potential gene defects (polymorphisms) over the coming years.
If you do a full genome mapping, please make sure the company offers both DIO1 and DIO2 in their raw data.
Note on SNPs: some genes have places in the genetic material that can have a mutation that is known to cause problems with the function of the gene. These locations in the genetic material of a gene that can have these defects are referred to as single nucleotide polymorphisms or SNPs. These SNPs are like ‘hotspots’ in the genetic material of genes where known mutations have been observed to occur and can prevent the genes from doing their proper job. These SNPs are given unique codes. Some genes may have more than one SNP that can cause problems. In the case of the DIO1 and DIO2 genes, if the SNPs have mutations it is possible for any D1 or D2 enzymes produced are compromised in their quality, resulting in lower conversion capability from T4 to T3. Researchers are continuing to look at the DIO1 and DIO2 genes and their polymorphisms. We can expect new information over time (some of which might contradict our current understanding). I intend to update this blog post as any good, new information arises.
How to Interpret the Results
If you have used a company that performs separate DIO1 and DIO2 tests, the results will be obvious in their report.
If you have used a company that does a full genome mapping, you have two choices:
1. Just use the Raw data.
The company may have a raw data browser, or you can look/search through the raw data text file if needed. This involves looking for the genetic marker or SNP associated with the gene defect. These SNPs have unique codes that can be searched for in the raw data. To find them you need to type in the ‘rs’ number associated with the mutation.
DIO1 has two known SNPs that can have mutations and cause problems.
These are referred to as “rs2235544” and “rs11206244”:
- For the DIO1 “rs2235544” SNP: the ‘C’ (or ‘G’) allele can boost the expression of DIO1 and can be associated with higher T3 levels. The ‘T’ (or ‘A’) allele is the risk allele; it can reduce T4 to T3 conversion and raise rT3 when active.
- For the DIO1 “rs11206244” SNP: the normal allele (normal, no defect) is ‘C’ (or ‘G’). The ‘T’ (or ‘A’) allele is the risk allele; it can reduce T4 to T3 conversion and raise rT3 when active.
DIO2 has one known SNP that can have mutations and cause problems.
This is referred to as “rs225014”. The normal/wild-type allele (no defect) is ‘T’ (or ‘A’). The ‘C’ (or ‘G’) allele is the mutation (risk allele); it can reduce T4 to T3 conversion and raise rT3 when active. See Note A below regarding Regenerus Labs.
Important Notes A and B:
A. At least one company (Regenerus Labs) is choosing to use a different notation for the DIO2 mutation – ‘A’ rather than ‘C’. This is their argument, “T92A (this is Threonine is changed at position 92 of the protein into Alanine) you also could write this Thr92Ala. If you look at base level in the DNA (genetic code) the C (=Cytosine) codes for the amino acid Alanine) see below. “A” is not Adenosine but the amino acid alanine in this report. So Ala = “C” in the genetic code (a Cytosine pyrimidine base). So the Ala/Ala or AA type corresponds to the CC genotype. The Thr/Thr or TT type corresponds to the TT (Thymine base, NOT THREONINE, in the DNA).” Their argument may be that there is an amino acid code:
where Alanine is ‘Ala’ but also ‘A’. So, an ‘A’ result for DIO2 from Regenerus labs is the mutation, then it is equivalent to what most of the companies are labelling as ‘C’ for the mutation. I wish they were all consistent.
B. In other genes, the notation for normal vs. abnormal is reversed. This has to do with conventions in designating the change from one base amino acid to another. If looking at other genes, one needs to check each one individually against the main databases. Also one ought to bear in mind that for some other genes, an abnormality may confer health advantages better than the normal un-mutated gene.
As briefly mentioned earlier, the results of testing show both copies of the gene – one from the person’s mother and one from the person’s father. If both alleles show no defect, there is no mutation at all. If one of them is a mutation, the person is said to be heterozygous for the defect, i.e. they have one copy. This is likely to impair the quality of the deiodinase enzymes (but not definitely). If both alleles have the mutation, the person is said to be homozygous for the defect and the chance of the deiodinase enzymes being impaired is higher and the consequence can be more severe.
So, for instance, if you were looking for DIO2, you would search for ‘rs225014’ then look at the results there. ‘TT’ would mean you had no defect inherited from either parent. ‘TC’ or ‘CT’ would mean one parent gave you the defect and one did not (you are heterozygous for the defect). ‘CC’ would mean that you have both defective genes (you are homozygous, but as mentioned some labs may use ‘A’ instead of ‘C’).
For DIO1 you would search for “rs11206244”. ‘CC’ implies there are no inherited defects. ‘CT’ or ‘TC’ means you are heterozygous for DIO1. ‘TT’ means you are homozygous for the DIO1 mutation (you have both copies).
2. Use a special filtering program.
Companies like ‘Genetic Lifehacks’ and ‘Genetic Genie’ offer software that read your genome mapping raw data and give you a simpler analysis of it.
Personally, I think the raw data is easy enough to look through.
3. Laboratories that Do the Testing
Note: it is advisable to always check with the company you are considering using that they still have the DIO1 and DIO2 results in their data, as both these gene defects are important to be aware of. We also know of companies that have removed DIO2 (possibly due to some kind of political pressure).
Ancestry (.com and .co.uk) have both DIO1 and DIO2 in the raw data. All you have to do is to download the raw data file and then load it into a text editor and search for the SNPs outlined above.
Blue Horizon Medicals in the UK have a genetic profile test that as of May 2020 includes the DIO1 and DIO2 SNP RSIDs and allele results:
Note: those of you in the UK can order the Blue Horizon Medicals’ genetic profile using a discount code that they have given me. See this post that explains the discount process:
Regenerus Labs in the UK have a single test for DIO2 (the most important of the two mutations). They do not test DIO1 though. They also deal with international customers, i.e. the USA, rest of Europe:
Regenerus insist that you also pay for a counsellor to give you the result, however, you can try telling them that I know a huge about this area, have written three books on it, and am acting as your practitioner – this may be good enough to avoid this extra charge.
23andMe only offers DIO1 in its raw data now. They do not test DIO2.
Nutrition Genome in the USA had both DIO1 and DIO2 in their raw data in August 2018. But this site only appears to work if you are in the USA.
Fulgent Genetics in the USA also provide DIO1 and DIO2 test results:
Also see SNPedia, a source of information on genes, gene defects etc. You can enter SNPs in the search bar and get information on them: https://www.snpedia.com/index.php/SNPedia
I hope you found this useful. I do believe that the DIO1 and DIO2 tests are useful to do, especially if you can see in your thyroid lab results that you are having poor conversion issues. The presence of one or both of these mutations would at least provide a concrete explanation for why this might be and hopefully persuade your physician to switch the balance of your thyroid medication to more T3 and less T4. Loss of thyroid tissue also loses conversion rate remember though – so this can be another reason for issues. I discuss all of these types of problems in my latest book, The Thyroid Patient’s Manual.
I am sure over time more gene-related issues will be discovered that affect thyroid hormone uptake into cells, the binding to thyroid receptors and much, much more.