Heritable (versus environmental) contribution to determine normal thyroid function pattern

TOPIC: Genetics of hypothalamic-pituitary axis

Title: Heritability of serum TSH, free T4 and free T2 concentrations: a study of a large UK twin cohort.

Authors: Panicker V, Wilson SG, Spector TD, Brown SJ, Falchi M, Richards JB, Surdulescu GL, Lim EM, Flectcher SJ, & Walsh JP.

Reference: Clinical Endocrinology 68: 652-659, 2008



Thyroid hormone action influences many metabolic and synthetic processes, but the degree of regulation attributed to genes and environmental factors affecting normal variation remains controversial.


The authors investigated the magnitude of the genetic and environmental determination of serum concentrations of free (f) T3, fT4, TSH and the fT4'TSH product and their variation, in a large cohort of twin pairs. Female dizygotic and monozygotic twins (respectively, 849 & 213 pairs) from the 'Twins-UK-Registry' (mean age 45.5; range 18'80 years) were studied.


Comparison of thyroid parameters within various groups showed no differences between smoking categories, and higher serum TSH and lower fT3 in subjects with positive thyroid antibodies. Using structural equation modelling, we estimated the heritable contribution to serum thyroid parameters (with 95% confidence intervals) to be 65% (58%'71%) for TSH, 65% (58%'71%) for the fT4'TSH product, 39% (20%'55%) for fT4 and 23% (3%'41%) for fT3.


Genetic regulation is a particularly important determinant of TSH and the fT4xTSH product, and is a less important determinant of fT4 and fT3 concentrations in Caucasian women. These data from a large well-characterized cohort suggest that while there is a strong heritable contribution to serum TSH, variation in fT4 and fT3 concentrations may be less explained by genetic factors and more driven by environmental effects than previously thought.


The concentration of circulating thyroxine (T4), triiodothyronine (T3) and TSH in euthyroid healthy subjects shows wider inter-individual than intra-individual variations. As a consequence, a 'normal' serum level, according to reference population-based ranges, may not necessarily be normal for a given individual. Mechanisms involved in the determination of the individual thyroid function in healthy subjects have not been elucidated so far, but are believed to be the consequence of a specific genetic background interacting with environmental factors.

In this interesting study, carried out on an impressive large cohort of female UK dizygotic and monozygotic twins, the authors provide compelling evidence that up to 65% of both serum TSH and the FT4xTSH product derives from heritable contribution. The genetic component appeared less important for serum FT4 & FT3 concentration, with an estimated contribution reaching 39% & 23%, respectively. The different genetic control of the two main components of thyroid axis could be explained (at least, according to the authors' suggestion) by a more pronounced effect of environmental factors on thyroid hormone synthesis & secretion. These data are in substantial agreement with others reported earlier, although the genetic contribution to FT4 & FT3 was found to be higher in 2 other studies (Hansen et al. JCEM 89:1181, 2004; and Samollow et al. 2004) and the genetic contribution to serum TSH lower in 1 study (Samollow et al. 2004). Differences in study design, gender (all subjects of the present investigation were females), ethnicity, environmental factors (such as iodine intake, other nutrients, smoking attitude') may explain such discrepancies. However, the important message remains that the genetic contribution to individual setting of thyroid function appears globally to be no less than 40-50%.

A further step in our understanding of the setting of normal thyroid function parameters would be to look at the specific genes involved, which are probably multiple and related to several receptors and post-receptor pathways along the hypothalamo-pituitary-thyroid axis. In their study, Panicker et al. underlined that an identical (65%) genetic contribution was found for both serum TSH and the FT4xTSH product, a parameter which has been proposed (on the basis of studies in patients with thyroid hormone resistance) to be a measure of the pituitary thyrotroph cell sensitivity to the negative feedback of circulating T4. However, the same authors correctly stated that this evidence is very indirect and that it may be premature to draw definitive conclusions.

A more direct approach to the understanding of the genes specifically involved in control of thyroid function is represented by the study of polymorphism of candidate genes. Single nucleotide polymorphisms (SNPs) in the TSH receptor (TSHR) and type 1 deiodinase genes are reported to be associated with differences in serum TSH & FT3 (see Peeters et al. in 2006). In particular, the 'Asp727Glu' TSHR polymorphism was found to be associated with lower circulating TSH in the absence of significant changes for serum FT3 & FT4 levels, suggesting a slightly higher sensitivity to TSH of the polymorphic TRSR when compared to the wild type protein. Although the TRSR plays a fundamental role in the setting of thyroid function, the effect of 'Asp727Glu' TSHR polymorphism is small, accounting for 0.91% of the total phenotypic variance of serum TSH levels. On the other hand, limiting genetic evaluation to known candidate genes could also be misleading, thereby preventing the identification of other important genes so far unrecognized. Thus, studies focused on full genomic screening may provide alternative clues. Recently, a wide multi-centric and multi-ethnic study (Lopez et al. in press, Am J Hum Gen, 2008) has been carried out by genotyping 362.129 SNPs in over 8.000 individuals from three genetically distant cohorts. This approach allowed the identification of a SNPs polymorphism located in intron 1 of the PDE8B, encoding a high affinity cAMP-specific phosphodiesterase, and responsible for 2.3% of total serum TSH variation with a very high statistical significance. The mechanism involved could be related to different amounts of intrathyroidal cAMP leading to different efficiencies of signal transduction. Interestingly, associations (although at a lower statistical level) were found by the same genome scan between serum TSH and the genes of other phosphodiesterases ( PDE10A, PDE4A ) and of some candidate genes including those encoding TSHR and thyroid hormone receptor beta.

In conclusion, recent data are unravelling the complex genetic control of the hypothalamo-pituitary-thyroid axis. Further studies are needed to ascertain whether and to what extent 'new' and 'old' genes involved in the regulation of physiologic thyroid function may also contribute to the phenotypic expression of thyroid disease and related pathological conditions.

Summary and commentary prepared by Stefano Mariotti (Related to Chapters 2 & 4 of TDM)