madman
Super Moderator
Michael Zitzmann (2009)
Variable phenotypes of androgen insensitivity exist in humans, mainly owing to defective, mutated androgen receptors. A More subtle modulation of androgen effects is related to the CAG repeat polymorphism ([CAG]n) in exon 1 of the androgen receptor gene, in vitro, transcription of androgen-dependent target genes is attenuated with the increasing length of triplets. As a clinical entity, the CAG repeat polymorphism can relate to variations of androgenicity in (apparently) eugonadal men in various tissues and psychological traits, the longer the (CAG)n, the less prominent the androgen effect when individuals with similar testosterone concentrations are compared. A strictly defined threshold to hypogonadism is likely to be replaced by a continuum spanned by genetics as well as symptom specificity. In addition, the effects of externally applied testosterone can be markedly influenced by the (CAG)n and respective pharmacogenetic implications are likely to influence indications as well as modalities of testosterone treatment of hypogonadal men.
*Physiology of AR
The AR is intracellularly located and structurally related to other steroid hormone receptors. The transcriptional regulation mediated by the AR is a process involving ligand (i.e., androgen) binding and interaction of cofactor, as well as conformational changes of the AR protein, receptor phosphorylation, and nuclear trafficking and binding to DNA target regions, and, finally, transcription activation. The human AR gene is located on the X-chromosome (Xq11–12) (e.g., [6]), spanning approximately 90 kb, and comprising eight exons.
The AR has a modular composition of three major functional domains: an N-terminal domain precedes the DNA-binding domain, followed by the C-terminal ligand-binding region (for review see [7]). Upon entering target cells, androgens will interact with the ligand-binding pocket of the AR, thus initiating an activation cascade that includes a change of conformation and nuclear translocation. Prior to binding at target DNA regions, homodimerization of two AR proteins occurs. The resulting homodimer binds to hormone-responsive elements, usually consisting of two palindromic sequences within the promoter regions of androgen-regulated genes. Facilitated by chromatin remodeling, direct interaction with other transcription factors and specific coactivators, repressors, and specific modulation of the assembly of the preinitiation complex is achieved: this results in specific activation or repression of target gene transcription [8].
Not only does testosterone bind to the AR, but its 5-a-reductase metabolite, dihydrotestosterone also binds with a much stronger relationship. Estrogenic hormones, progesterone, and dehydroepiandrosterone are also able to bind to and activate the AR, but only on a clinically nonrelevant basis [8].
*Clinical implications: mutations & polymorphisms of the androgen receptor
Men presenting with features of hypogonadism may exhibit normal testosterone levels but CAG repeat lengths above the normal average (Europe: 21, Africa: 17, Asia: 23): hence, a CAG repeat length longer than 25 is still considered to be within the normal range, but can already be associated with reduced androgen action and accompanying clinical features, suggesting classical hypogonadism in the case of still normal testosterone concentrations [4].
In Klinefelter patients, who have two AR gene alleles due to their 47, XXY karyotype, the shorter CAG repeat allele is preferentially inactive. In this group of patients, CAG repeat length is positively associated with body height, while bone density and the relation of arm span to body height are inversely related to CAG repeat length [37]. The presence of long CAG repeats was seen as predictive of gynecomastia, while shorter CAG repeats were associated with a stable partnership and professions requiring higher standards of education [37]. Also, aspects of puberty and masculinization of younger Klinefelter patients seem to be influenced by this polymorphism, boys who have longer CAG repeats exhibit mitigated androgen effects [38,39]. In the rare syndrome of 46, XX males, the inactivation patterns of AR gene alleles in XX males were found to be significantly more skewed than in Klinefelter patients and women. A total of seven of ten heterozygous XX male patients displayed an extreme skewing of more than 80% with no preference toward the shorter or longer AR allele. The length of the AR gene CAG repeat polymorphism was positively related to traits of hypogonadism [40]
Studies examining the effects of the AR gene (CAG)n polymorphism have to consider some aspects for proper description of clinical relevance, for example, the involvement of hypogonadal men will distort results owing to lack of sufficient ligand binding, hence, activation of the AR. Similarly, not considering testosterone levels, possibly compensating for the effects of the (CAG) n polymorphism, takes an incomplete approach. When examining the effects of androgens, regression models are required to include both testosterone concentrations and the length of the AR gene CAG repeat polymorphism. In this light, a recent study in a large cohort of aging men did not demonstrate a relationship of the AR gene CAG repeat polymorphism to the risk of heart disease, possibly due to the fact that serum testosterone concentrations were not considered [41]
The weaker androgen action induced by longer (CAG)n will also affect the feedback mechanism of the hypothalamic-pituitary-gonadal (HPG) axis. In healthy men, longer (CAG)n will usually provoke higher luteinizing hormone (LH) secretion [33,34,42]. This can, in persons with intact and fully responsive Leydig Cell capacity, result in higher concentrations of testosterone and, hence, compensation of weaker androgen action. As a result of higher testosterone concentrations, its aromatization product estradiol will also be present in higher amounts [42]. Such higher estradiol concentrations in these men with intact HPG regulation or feedback mechanisms can even lead to enhanced effects in estrogen-dependent tissues, such as bones [43].
However, the investigation of men with an intact HPG axis does not focus on the clinically relevant clientele including men with symptoms of androgen deficiency and disturbances of the HPG axis. Such disorders can be of milder nature, as seen in late-onset hypogonadism [44] or subjects with metabolic syndrome [45], conditions in which both pituitary function and Leydig cell capacity are impaired. In case of longer (CAG)n, these men will present with inadequate concentrations of LH within the lower normal range and/or testosterone levels that are in the lower normal range, but, nevertheless, will require higher concentrations of testosterone to compensate for their attenuated androgen action. Thus, they are likely to present with features of hypogonadism in the presence of normal testosterone levels [4] (see Figure 2). This is the patient group most likely missed for investigation, diagnostics, and putative treatment in clinical andrology to date.
In the case of ‘classical’ hypogonadism, for example, the (almost) complete breakdown of the HPG axis due to primary or secondary origin, testosterone levels are low and patients require substitution. The findings in men with intact HPG axes demonstrate that persons with longer (CAG)n require higher testosterone concentrations for normal androgen action in comparison with men with shorter (CAG)n [42]. Thus, hypogonadal persons with longer (CAG)n will then, as do their healthy counterparts, need higher testosterone levels (and, hence, higher testosterone doses) to compensate for mitigated androgen action. Correspondingly, persons with rather short (CAG)n will need lower doses of testosterone substitution in case of hypogonadism. This is exactly the subject of pharmacogenetic tailoring of testosterone substitution that will be elucidated below.
*Pharmacogenetics of testosterone therapy
*Discussion: a hypothetical model of androgen action
Testosterone concentrations within the normal range more or less saturate ARs and it has been shown that androgen effects reach a plateau at certain levels, which are probably tissue- or symptom-specific [52]. Just recently, a saturation model for prostate tissue in relation to testosterone concentrations has been propagated [53]. Marked increments of testosterone-caused effects are only seen beyond the normal range when highly supraphysiological levels, such as in doping, are reached. Hence, it can be argued that within the range of normal eugonadal testosterone concentrations, which are desired to be achieved by substitution therapy of hypogonadal patients, genetically determined functional differences in AR activity can be observed and will be of significance, while in a condition of hypogonadism, androgenicity will rather depend on androgen levels as testosterone binds to ARs and will augment androgen effects until a saturation level is reached (Figure 3).
Upon testosterone substitution of hypogonadal men, effects of the androgen are likely to be induced by the increment of androgen levels from the hypogonadal into the eugonadal range, as well as through modulation mediated by the AR gene CAG repeat polymorphism, the latter mainly within the eugonadal range, which is achieved during sufficient therapeutic approaches; the AR needs a normal amount of substrate (testosterone) to exhibit the effects of the (CAG)n polymorphism.
Conclusion
In summary, a confined and universal threshold for testosterone levels to hypogonadism does not exist, but, instead, individual thresholds of testosterone levels according to the length of the CAG repeat polymorphism are likely to be of clinical relevance. The model of Figure 3 explains why androgen effects are found when comparing hypogonadal and eugonadal men, but can often not be described for various testosterone levels within the eugonadal range. Also, tissue or symptom-specific thresholds of testosterone concentrations as well as testosterone levels themselves contribute to span a continuum between eugonadal and hypogonadism respecting individuality rather than a universal threshold.
Especially in regard to late-onset hypogonadism or the metabolic syndrome, clinical entities that are related to the advancing age of men or obesity causing a deterioration of the HPG axis and, hence, disturbance of feedback mechanisms that can compensate for long (CAG)n, these aspects might play a role in the decision whether to start testosterone replacement therapy or not.
In the future, substitution therapy could be individually tailored by the AR gene (CAG)n polymorphism: men with shorter repeats may require lower doses of testosterone while men with longer repeats could be in need of higher doses of androgens. Correspondingly, men with longer repeat tracts may need substitution therapy at baseline testosterone concentrations, which are still considered normal for the total population. Future pharmacogenetic studies will be required to find clinically robust answers.
It should be mentioned that many other factors are most likely to be influencing testosterone action, be it the endogenous hormone or exogenously administered androgens. Enough genetic information is not available to substantially review the possible effects of other androgen receptor polymorphisms, enzymes involved in androgen metabolism (5-a-reductase, aromatase), protein binding, and elimination (conjugation), as well as differences in co-activator and co-repressor activities, most likely tissue-specific. The CAG repeat polymorphism cannot be the only determining factor in androgen sensitivity.
Nonlinear calculation models according to Figure 3 involving testosterone concentrations and the length of the AR (CAG)n may be useful to assess the individually needed testosterone dose to achieve a certain desired androgen-related action or avoid an adverse side effect, such as elevated hematocrit (see Figure 4 for a summary). These insights require further research within the frameset of prospective studies as to this time point only semiquantitative statements can be made in terms of testosterone doses needed in hypogonadal men with different CAG repeat lengths [5]. Future calculations might be possible to state the required dose of testosterone in relation to the AR polyglutamine stretch. The different doses that are assumed according to the cross-sectional literature are most likely to be of clinical relevance and not marginal.
The level of knowledge required to include the AR gene CAG repeat polymorphism into routine andrological assessments is not sufficient, but it has become a clinically and scientifically worthy concept to determine the CAG repeat length in subjects with normal testosterone levels and unexplained features of hypoandrogenism. Concerning the dose adaption of testosterone to the polymorphism in androgen substitution, independent verification of the above-mentioned findings is still required before giving general recommendations.
Future perspective
This is a speculative viewpoint stating that determination of the AR gene (CAG)n length will play a clinically robust role in diagnosis and treatment of hypogonadal men within 5–10 years, both for adaption of testosterone levels to the clinical features as well as the dose required for proper androgen substitution.
Variable phenotypes of androgen insensitivity exist in humans, mainly owing to defective, mutated androgen receptors. A More subtle modulation of androgen effects is related to the CAG repeat polymorphism ([CAG]n) in exon 1 of the androgen receptor gene, in vitro, transcription of androgen-dependent target genes is attenuated with the increasing length of triplets. As a clinical entity, the CAG repeat polymorphism can relate to variations of androgenicity in (apparently) eugonadal men in various tissues and psychological traits, the longer the (CAG)n, the less prominent the androgen effect when individuals with similar testosterone concentrations are compared. A strictly defined threshold to hypogonadism is likely to be replaced by a continuum spanned by genetics as well as symptom specificity. In addition, the effects of externally applied testosterone can be markedly influenced by the (CAG)n and respective pharmacogenetic implications are likely to influence indications as well as modalities of testosterone treatment of hypogonadal men.
*Physiology of AR
The AR is intracellularly located and structurally related to other steroid hormone receptors. The transcriptional regulation mediated by the AR is a process involving ligand (i.e., androgen) binding and interaction of cofactor, as well as conformational changes of the AR protein, receptor phosphorylation, and nuclear trafficking and binding to DNA target regions, and, finally, transcription activation. The human AR gene is located on the X-chromosome (Xq11–12) (e.g., [6]), spanning approximately 90 kb, and comprising eight exons.
The AR has a modular composition of three major functional domains: an N-terminal domain precedes the DNA-binding domain, followed by the C-terminal ligand-binding region (for review see [7]). Upon entering target cells, androgens will interact with the ligand-binding pocket of the AR, thus initiating an activation cascade that includes a change of conformation and nuclear translocation. Prior to binding at target DNA regions, homodimerization of two AR proteins occurs. The resulting homodimer binds to hormone-responsive elements, usually consisting of two palindromic sequences within the promoter regions of androgen-regulated genes. Facilitated by chromatin remodeling, direct interaction with other transcription factors and specific coactivators, repressors, and specific modulation of the assembly of the preinitiation complex is achieved: this results in specific activation or repression of target gene transcription [8].
Not only does testosterone bind to the AR, but its 5-a-reductase metabolite, dihydrotestosterone also binds with a much stronger relationship. Estrogenic hormones, progesterone, and dehydroepiandrosterone are also able to bind to and activate the AR, but only on a clinically nonrelevant basis [8].
*Clinical implications: mutations & polymorphisms of the androgen receptor
Men presenting with features of hypogonadism may exhibit normal testosterone levels but CAG repeat lengths above the normal average (Europe: 21, Africa: 17, Asia: 23): hence, a CAG repeat length longer than 25 is still considered to be within the normal range, but can already be associated with reduced androgen action and accompanying clinical features, suggesting classical hypogonadism in the case of still normal testosterone concentrations [4].
In Klinefelter patients, who have two AR gene alleles due to their 47, XXY karyotype, the shorter CAG repeat allele is preferentially inactive. In this group of patients, CAG repeat length is positively associated with body height, while bone density and the relation of arm span to body height are inversely related to CAG repeat length [37]. The presence of long CAG repeats was seen as predictive of gynecomastia, while shorter CAG repeats were associated with a stable partnership and professions requiring higher standards of education [37]. Also, aspects of puberty and masculinization of younger Klinefelter patients seem to be influenced by this polymorphism, boys who have longer CAG repeats exhibit mitigated androgen effects [38,39]. In the rare syndrome of 46, XX males, the inactivation patterns of AR gene alleles in XX males were found to be significantly more skewed than in Klinefelter patients and women. A total of seven of ten heterozygous XX male patients displayed an extreme skewing of more than 80% with no preference toward the shorter or longer AR allele. The length of the AR gene CAG repeat polymorphism was positively related to traits of hypogonadism [40]
Studies examining the effects of the AR gene (CAG)n polymorphism have to consider some aspects for proper description of clinical relevance, for example, the involvement of hypogonadal men will distort results owing to lack of sufficient ligand binding, hence, activation of the AR. Similarly, not considering testosterone levels, possibly compensating for the effects of the (CAG) n polymorphism, takes an incomplete approach. When examining the effects of androgens, regression models are required to include both testosterone concentrations and the length of the AR gene CAG repeat polymorphism. In this light, a recent study in a large cohort of aging men did not demonstrate a relationship of the AR gene CAG repeat polymorphism to the risk of heart disease, possibly due to the fact that serum testosterone concentrations were not considered [41]
The weaker androgen action induced by longer (CAG)n will also affect the feedback mechanism of the hypothalamic-pituitary-gonadal (HPG) axis. In healthy men, longer (CAG)n will usually provoke higher luteinizing hormone (LH) secretion [33,34,42]. This can, in persons with intact and fully responsive Leydig Cell capacity, result in higher concentrations of testosterone and, hence, compensation of weaker androgen action. As a result of higher testosterone concentrations, its aromatization product estradiol will also be present in higher amounts [42]. Such higher estradiol concentrations in these men with intact HPG regulation or feedback mechanisms can even lead to enhanced effects in estrogen-dependent tissues, such as bones [43].
However, the investigation of men with an intact HPG axis does not focus on the clinically relevant clientele including men with symptoms of androgen deficiency and disturbances of the HPG axis. Such disorders can be of milder nature, as seen in late-onset hypogonadism [44] or subjects with metabolic syndrome [45], conditions in which both pituitary function and Leydig cell capacity are impaired. In case of longer (CAG)n, these men will present with inadequate concentrations of LH within the lower normal range and/or testosterone levels that are in the lower normal range, but, nevertheless, will require higher concentrations of testosterone to compensate for their attenuated androgen action. Thus, they are likely to present with features of hypogonadism in the presence of normal testosterone levels [4] (see Figure 2). This is the patient group most likely missed for investigation, diagnostics, and putative treatment in clinical andrology to date.
In the case of ‘classical’ hypogonadism, for example, the (almost) complete breakdown of the HPG axis due to primary or secondary origin, testosterone levels are low and patients require substitution. The findings in men with intact HPG axes demonstrate that persons with longer (CAG)n require higher testosterone concentrations for normal androgen action in comparison with men with shorter (CAG)n [42]. Thus, hypogonadal persons with longer (CAG)n will then, as do their healthy counterparts, need higher testosterone levels (and, hence, higher testosterone doses) to compensate for mitigated androgen action. Correspondingly, persons with rather short (CAG)n will need lower doses of testosterone substitution in case of hypogonadism. This is exactly the subject of pharmacogenetic tailoring of testosterone substitution that will be elucidated below.
*Pharmacogenetics of testosterone therapy
*Discussion: a hypothetical model of androgen action
Testosterone concentrations within the normal range more or less saturate ARs and it has been shown that androgen effects reach a plateau at certain levels, which are probably tissue- or symptom-specific [52]. Just recently, a saturation model for prostate tissue in relation to testosterone concentrations has been propagated [53]. Marked increments of testosterone-caused effects are only seen beyond the normal range when highly supraphysiological levels, such as in doping, are reached. Hence, it can be argued that within the range of normal eugonadal testosterone concentrations, which are desired to be achieved by substitution therapy of hypogonadal patients, genetically determined functional differences in AR activity can be observed and will be of significance, while in a condition of hypogonadism, androgenicity will rather depend on androgen levels as testosterone binds to ARs and will augment androgen effects until a saturation level is reached (Figure 3).
Upon testosterone substitution of hypogonadal men, effects of the androgen are likely to be induced by the increment of androgen levels from the hypogonadal into the eugonadal range, as well as through modulation mediated by the AR gene CAG repeat polymorphism, the latter mainly within the eugonadal range, which is achieved during sufficient therapeutic approaches; the AR needs a normal amount of substrate (testosterone) to exhibit the effects of the (CAG)n polymorphism.
Conclusion
In summary, a confined and universal threshold for testosterone levels to hypogonadism does not exist, but, instead, individual thresholds of testosterone levels according to the length of the CAG repeat polymorphism are likely to be of clinical relevance. The model of Figure 3 explains why androgen effects are found when comparing hypogonadal and eugonadal men, but can often not be described for various testosterone levels within the eugonadal range. Also, tissue or symptom-specific thresholds of testosterone concentrations as well as testosterone levels themselves contribute to span a continuum between eugonadal and hypogonadism respecting individuality rather than a universal threshold.
Especially in regard to late-onset hypogonadism or the metabolic syndrome, clinical entities that are related to the advancing age of men or obesity causing a deterioration of the HPG axis and, hence, disturbance of feedback mechanisms that can compensate for long (CAG)n, these aspects might play a role in the decision whether to start testosterone replacement therapy or not.
In the future, substitution therapy could be individually tailored by the AR gene (CAG)n polymorphism: men with shorter repeats may require lower doses of testosterone while men with longer repeats could be in need of higher doses of androgens. Correspondingly, men with longer repeat tracts may need substitution therapy at baseline testosterone concentrations, which are still considered normal for the total population. Future pharmacogenetic studies will be required to find clinically robust answers.
It should be mentioned that many other factors are most likely to be influencing testosterone action, be it the endogenous hormone or exogenously administered androgens. Enough genetic information is not available to substantially review the possible effects of other androgen receptor polymorphisms, enzymes involved in androgen metabolism (5-a-reductase, aromatase), protein binding, and elimination (conjugation), as well as differences in co-activator and co-repressor activities, most likely tissue-specific. The CAG repeat polymorphism cannot be the only determining factor in androgen sensitivity.
Nonlinear calculation models according to Figure 3 involving testosterone concentrations and the length of the AR (CAG)n may be useful to assess the individually needed testosterone dose to achieve a certain desired androgen-related action or avoid an adverse side effect, such as elevated hematocrit (see Figure 4 for a summary). These insights require further research within the frameset of prospective studies as to this time point only semiquantitative statements can be made in terms of testosterone doses needed in hypogonadal men with different CAG repeat lengths [5]. Future calculations might be possible to state the required dose of testosterone in relation to the AR polyglutamine stretch. The different doses that are assumed according to the cross-sectional literature are most likely to be of clinical relevance and not marginal.
The level of knowledge required to include the AR gene CAG repeat polymorphism into routine andrological assessments is not sufficient, but it has become a clinically and scientifically worthy concept to determine the CAG repeat length in subjects with normal testosterone levels and unexplained features of hypoandrogenism. Concerning the dose adaption of testosterone to the polymorphism in androgen substitution, independent verification of the above-mentioned findings is still required before giving general recommendations.
Future perspective
This is a speculative viewpoint stating that determination of the AR gene (CAG)n length will play a clinically robust role in diagnosis and treatment of hypogonadal men within 5–10 years, both for adaption of testosterone levels to the clinical features as well as the dose required for proper androgen substitution.
