Previous Issue | Table Of Contents | Next Issue

Guest Editor: Douglas A. Husmann, MD ANDROGEN RECEPTOR EXPRESSION IS CONTROLLED BY ANDROGENS DURING PHALLIC GROWTH Karen K. Takane, PhD ANDROGENS ARE NOT MAJOR DOWN REGULATORS OF THE ANDROGEN RECEPTOR PROTEIN DURING PENILE GROWTH Nestor F. Gonzalez-Cadvid, PhD Jacob Rajfer, MD CESSATION OF PHALLIC GROWTH: IS IT REALLY DETERMINED BY ANDROGEN RECEPTOR DOWN REGULATION? Douglas A. Husmann, MD	GUEST EDITOR'S NOTES: Whether or not androgens down regulate the androgen receptor (AR) during penile growth is currently one of the major controversies regarding the cessation of penile growth. Indeed, in our own publications we have come down both for13 and against4 the down regulation of AR by androgens. This issue of Dialogues features contributions from representatives of two separate research laboratories that have pioneered investigative endeavors into the regulation of penile growth. We indeed are fortunate and grateful that Dr. Takane and Drs. Gonzalez-Cadvid and Rajfer agreed to submit their dichotomous views on this controversial subject for the benefit of this publication's readers. Perhaps the real question that exists regarding androgen receptor function and the cessation of penile growth is whether or not a cause-and-effect relationship truly exists. In our laboratory endeavors, we have seen down regulation of the rat penile androgen receptor to be present by 10 weeks of age; however, continued penile growth (ie, increase in penile wet weight4) occurs through 16 weeks of age. This finding has also been confirmed by Takane and associates (see her comments and figures). These data suggest one of two possibilities. The first is that androgen actions for penile growth may work through a two-step steroid amplification model.15 In two-step steroid amplification, the steroid hormone activates its receptor which in turn directs the transcription (production of RNA from DNA) of a few specific genes. The eventual proteins produced by the activated genes stimulate the production of a multitude of additional proteins or growth factors that are responsible for the physiologic effect of the hormone, an amplified secondary response. It is important to note that the amplified response,
Editor Richard M. Ehrlich, MD Clinical Professor of Surgery/Urology School of Medicine University of California, Los Angeles Publisher William J. Miller
SUBJECT OF CONTROVERSY: PENILE GROWTH DURING SEXUAL MATURATION due to the delay in its production, may occur in the absence of the activating hormone or at the time that the steroid receptor apparently is down regulated (ie, that continued penile growth may occur after down regulation of the AR). A second possible mechanism is that cessation of penile growth is not even modulated through the AR. Indeed, rather than playing the major role in the cessation of penile growth, AR down regulation may be the consequence of yet undefined cellular interactions that regulate penile size. Douglas A. Husmann, MD	Snochowski, 1979). The role of androgens in driving penis growth and the involvement of the AR in regulating the growth of this organ was examined. Penile growth was found to be associated with a peak of AR binding in the human foreskin during the time of puberty, rising 4.5-fold from 40 fmol/gram of tissue in newborn infants to maximal levels in 16- to 20-year-old subjects (Roehrborn et al, 1987). The cessation of penile growth could also be linked to a loss of AR in the foreskin, falling by age 31 to values similar to those seen in early infancy. These studies raised the possibility that penile growth may be somehow linked to AR content. This phenomenon has been studied in more detail in animal models. In rats, an age-related decrease in the amount of AR in the cytosol of the phallus from immature to mature rats was described by Rajfer et al, 1980. The highest level of AR binding was detected in the cytosol of immature 16-day-old rat penis (approximately 50 fmol/mg protein) and gradually decreased following the onset of pubertal growth to nearly undetectable levels in cytosols from 70-day-ofcJ adult rat penis. We have confirmed that in the intact rat, the penis increases in size from 40 mg at 3 weeks of age to 275 mg at 10 weeks and 340 mg at 16 weeks of age (Takane et al, 1990). This increase in size is accompanied by a decrease in the total AR binding activity in the penis from 730 fmol/gm tissue in 3-week-old prepubertal rats to less than 50 fmol/gm in 10-weekold rat penis (see Figure, next page). The role of androgens in mediating these changes in phallic size and AR levels was demonstrated by studying the effects of androgen withdrawal and replacement. Castration of 3-week-old prepubertal rats resulted in a cessation of penile growth and also halted the age-related decline in AR level, even up to 16 weeks of age. Castrating adult rats, while causing a slight decline in penile size, did not result in any detectable increase in AR levels when measured 6 weeks later. These results indicated that the age-related decline of penile AR levels is androgen-dependent and, once established, is permanent and does not require continued high levels of androgen. Hypophysectomy of 3-week-old male rats had a similar effect to castration in that penile growth was prevented and penile AR levels did not decrease over a course of an additional 3 weeks. Interestingly, daily treatment of hypophysectomized rats for 3 weeks with rat growth hormone caused a 1.7-fold increase in penis weight and a 1.4-fold increase in penile AR levels over saline-treated hypophysectomized rats in the absence of androgens (unpublished data). Using methyltrienolone to measure the level of AR in extracts from the spongy tissue of the penis, Godec et al, 1985 demonstrated very low amounts of androgen binding (about 2 fmol/mg) in the corpus cavernosa of adult males. Similar low levels of the AR have also been measured in tissue extracts prepared from mature porcine, human and equine corpus cavernosum, and from equine corpus spongiosum specimens (Nonomura et al, 1990). Further evidence that the growth of the penis and the fall in AR levels are directly linked to the actions of androgens has been obtained in studies in castrated rats. Administration to rats castrated at 3 weeks of age of pharmacological doses of DHT for 3 weeks led to an acceleration of growth and a
KAREN K. TAKANE, PhD Department of Endocrinology, University of Texas Southwestern Medical Center, Dallas Androgen Receptor Expression is Controlled by Androgens During Phallic Growth Male sexual differentiation during embryogenesis and the growth and maturation of these tissues during puberty is under the control of androgens (George and Wilson, 1994). However, not all tissues which are initially androgen-responsive remain in this state. In some tissues, such as canine and human prostates, androgen-mediated growth continues throughout life. In contrast, the growth of the penis is controlled by androgens only during a defined portion of the lifespan. In man, growth of the penis occurs in two phases. The first period is moderate and extends from birth to about age five and is proportionate to the general increase in body size that occurs during this time. Following a latency period (age 5 until the onset of puberty), little growth occurs. As plasma androgen levels begin to rise with the onset of puberty, penile growth resumes at a rapid rate. This growth is limited, however, and ceases with approaching maturity (Schonfekj and Beebe, 1942) despite the continued presence of high levels of circulating androgens. The actions of the two principal androgens testosterone (T) and 5 a-dihydrotestosterone (DHT) are not equivalent. Although DHT and T exert their effects through the same high affinity nuclear protein, the androgen receptor (AR), (Chang et al, 1988; Lubahn et al, 1988; Trapman et al, 1988; Tilley et al, 1989), a number of lines of evidence suggest that the viriiization of the external genitalia is mediated by DHT which is formed in the tissue from circulating T (Wilson and Walker, 1969; Renfree et al, 1992). Hormone binding then activates the AR by increasing the affinity of the hormone-receptor complex for specific sites on DNA, on which it can act as a regulator of transcription of specific genes (George and Wilson, 1994). The mechanisms controlling the cessation of phallic growth occurring at maturity are poorly understood, and a number of studies have examined the changes in AR expression that occur during puberty. The pubertal phase of penile growth occurs at a time of rising circulating androgens and androgen-binding activity has been detected in human foreskin (Griffin et al, 1976; Evain et al, 1977; Svensson and
Figure. Penile growth and androgen receptor content of a rat penis as a function of age. Each value represents the mean ± SEM (p<0.005 compared with age-matched, intact controls).
prompt decline in AR levels in the penis to levels seen in the adult (above). Similar changes were seen in 3-week-old rats that were treated immediately after castration, and in adult rats that were castrated prepubertally and treated later, implying that the AR in the castrated adult rat is functional and able to respond to androgens, which leads to the fact that it is androgen action and not age which causes phallic growth and the decrease in penile AR. Whether or not the down regulation of penile AR by androgens during puberty is the cause or consequence of terminal differentiation of specific cell types within the penis is difficult to ascertain since the penis is a histologically complex tissue. The many cell types that make up the penis have different growth rates and undergo different types of morphological change during sexual maturation. In theory, the decrease in penile AR content with age could be due to an increase in the number of AR deficient cells or a decrease in AR levels in one or more of the different cell types within the penis. Initial studies designed to measure the content and distribution of the AR in tissues used autoradiography to detect the sites of binding of radiolabeled androgens in tissues. Using this technique to examine the genital tubercles of fetal rats, androgen-binding cells were identified in the precursors of the os penis, corpus cavernosum, preputial mesenchyme, the mesenchymal cells around the urethra, and in urethra epithelial cells (Murakami, 1987-a). Similarly the histological comparison of the development of the penis in normal mice and mice carrying the testicular feminization (Tfm) mutation (Lyon and Hawkes, 1970), which results in	androgen resistance due to a loss of functional AR, demonstrated that the os penis, corpus cavernosum penis, corpus cavernosum glandis, corpus cavemosum urethrae, the urethra, and the spines on the skin of the glans penis in the Tfm mice do not develop normally, findings which support the developmental importance of the androgen receptor (Murakami, 1987-b). The AR has now been localized to specific cells within the developing rat phallus using antibodies raised against two distinct regions of the AR (Husmann et al, 1990). Nuclear staining for AR can be detected in the corpus cavemosum penis, corpus cavernosum urethra, small lateral cavernous bodies, os penis, urethra, and skin epithelium of the 3-week-old rat penis (Takane et al, 1991). In fact, only the tunica fibrosa failed to show nuclear staining. In contrast, only the nuclei of the urethra and skin epithelium stain intensely in the adult rat penis. The corpus cavemosum penis and corpus cavernosum urethra exhibited faint scattered staining for AR in agreement with the androgen binding activity seen in the spongy tissue taken from the penis of adults. In accordance with the androgen binding data, the penis of adult rats castrated prepubertally appeared to be histologically indistinguishable from specimens obtained from 3-week-old intact rats. In addition, the pattern of AR immunoreactivity in the castrated rat did not differ from that seen in the intact, immature rat. Furthermore, castrated immature rats treated for 3 weeks with high doses of DHT had a penis size and morphology similar to that in the intact adult rat. Finally, in agreement with studies using ligand binding, mature rats that were castrated after puberty did not regain any staining for
SUBJECT OF CONTROVERSY: PENILE GROWTH DURING SEXUAL MATURATION
penile AR, indicating that the postpubertal decrease in AR expression could not be reversed. Thus, the decline in penile AR during and following puberty appears to be permanent and is correlated with the loss of AR expression in the corpus cavernosa and os penis. The decrease of AR levels in the spongy tissues and bone is mediated by androgens and is prevented but not reversed by castration. The persistence of the AR in the urethra and epithelia implies that the AR is regulated differently in these compartments of the penis compared with the body of the penis. The factors that mediate the permanent decline in receptor content in the corpus and how these changes are linked to the cessation of penile growth following puberty remain to be resolved. Whether this decline in AR levels is the cause or effect of the cessation of growth remains to be established.	hormone, DNA, and transactivating domains found in other species. It is the product of a 10 kb mRNA transcribed from a large gene located on the X chromosome. The AR in the penis binds dihydrotestosterone (DHT) with about 5-10-fold more affinity than testosterone (T), eliciting its dimerization, phosphorylation, and translocation into the nucleus. This AR activation by ligand binding is crucial for penile growth, since castration in the rat stops this process and androgen supplementation restores it.2910 The majority of ARs in the rat penis are confined to the corpora cavemosa smooth muscle.9 In this species, DHT has been shown to be the active androgen in the embryologic development of the penis,11 is more efficient than T in penile growth,10 and is the main active androgen for maintaining the erectile function.1214 Regulation of Androgen Receptor Levels in the Rat Penis During Sexual Maturation. Age-related changes. The initial demonstration that ARs are down regulated in the penis during sexual maturation1 was based on the measurement of the binding of a synthetic androgen analog (H3-R1881) to the ARs present in the penile cytosol obtained from rats from 2.5 to 10 weeks of age. It was found that during this period the AR concentration decreased 30- to 50-fold in a rapid fashion, to practically negligible levels at adulthood (8-week-old rats). The dramatic loss of penile ARs in the adult rats as compared with the immature rats was confirmed by androgen binding in two other studies,210 and by western blot immunodetection using antibodies against either the amino or the carboxi terminus of the AR protein.15 The reduction, as shown by immunocytochemistry,9 occurs as expected mainly in the corpora cavernosal smooth muscle but not in the urethra or penile skin. The penile AR down regulation that accompanies sexual maturation is not caused by an enhanced degradation of the AR protein. This is inferred from northern blot analysis of penile RNA during this period, revealing a pattern of decrease of the steady state levels of AR mRNA in the corpora cavernosa even more accelerated than that of androgen binding.16-17 The marked difference in AR mRNA content in the penis of adult rats, as compared to immature rats (2.5 weeks old), appears to be due mainly to transcriptional inhibition because the in vitro binding of nuclear proteins to the AR promoter, observed with extracts from the immature penis, disappears in the case of the adult penis.18 The considerable decrease in AR concentration in the rat penis with sexual maturation counteracts, in terms of androgen binding, the steady increase in serum T which in the rat rises approximately by a factor of three from the immature to the adult period.3 This is likely to be the cause of the gradual decline in growth rate of the penis in terms of weight occurring after puberty (4 to 5 weeks of age),15 although apparently the length continues to increase at about the same rate.7 In any case, the penis practically stops growing at adulthood, coinciding with the very low penile AR concentration at that period and despite the fact that body weight continues to increase. This is in marked contrast to the situation in the rat ventral
NESTOR F. GONZALEZ-CADVID, PhD JACOB RAJFER, MD Department of Surgery (Urology), Harbor-UCLA Medical Center, Torrence, CA Androgens Are Not Major Down Regulators of the Androgen Receptor Protein During Penile Growth Introduction. The penis is the only organ in the rat where the concentration of androgen receptors (AR) and to a lesser extent, their total content, are physiologically down regulated.1-2 This process occurs during sexual maturation and provides a mechanism to shut off androgen-dependent pathways in a tissue-specific fashion that contrasts with the increase of ARs in the prostate and the testis during the same period. This differential regulation of the steady state levels of ARs is most likely the major reason why at adulthood the penis stops growing while the prostate continues enlarging during adulthood and senescence.3 Therefore, the mechanism(s) by which AR levels are down regulated in the rat penis may provide the basis for a strategy to manipulate therapeutically androgen-dependent tissue growth not only in the penis but in other tissues as well. The rat penile model of developmental regulation of androgen-responsive growth may contribute to our understanding of microphalus, hypospadias, ambiguous genitalia and other conditions if the defect in these cases is associated with any of the steps in the complex androgen-responsiveness cascade triggered by the AR.4-6 Independently from AR regulation, the mere existence of a model of micropenis in the rat created by androgen blockade7 also supports the usefulness of this laboratory animal model for studying factors other than androgens and ARs that are specifically involved in penile growth, particularly growth factors. Although not immediately relevant to pediatrics, the elucidation of the down regulation of ARs in the growing rat penis may open new therapeutic approaches to halt the abnormal growth of the prostate, ie, benign prostatic hyperplasia and prostatic cancer. The rat AR is well characterized in physiological and molecular terms.8 It is a 110 kDa protein encompassing the
prostate that enlarges as a function of age. Its size increases by a factor of three to four between 4 and 10 weeks of age and continues growing for at least the next four months at about the same rate.2>3 Although the concentration of ARs in this organ declines by about 50% during the puberty/ adulthood period, the total content per ventral prostate rises by a factor of two to four as compared with a reduction greater than 50% in the penile AR content. Thereafter, prostatic AR concentration remains constant at a considerable level when compared with the penis, since it is equivalent to the concentration detected in the penis from 3-week-old rats. The total AR content per ventral prostate increases steadily with age. Another organ in the rat contrasting with the situation in the penis is the rat testis where AR concentration increases during sexual maturation,19 although this appears to occur because of the similar increase in Sertoli and Leydig cell numbers. Therefore, the developmental pattern of AR expression in the penis is rather unique in the rat. Role of androgens. The fact that penile ARs are reduced as serum T increases in the rat may suggest that androgens are responsible for this AR down regulation and that they would act paradoxically as eventual inhibitors of androgen-dependent penile growth. Results obtained in rats castrated before puberty (3 weeks of age) were taken as supporting this hypothesis, since the penile AR concentration remained at immature levels at 7 and 13 weeks after castration, but decreased to the respective age levels if DHT was administered immediately after castration.2 However, it should be noted that the total penile AR content in the castrated animal reaching adulthood remained as low as in the adult animal because the penis grew very little in that period. In contrast to the castration-induced up regulation of penile AR concentration in immature rats, castration of adult animals failed to increase penile AR concentration, thus showing that after a certain stage the down regulation becomes irreversible. These interesting results were confirmed by immunocytochemistry.9 This paradigm was recently reexamined and expanded in immature castrated rats treated with high and medium doses of either T accompanied by finasteride to block DHT synthesis or DHT alone.15 After short (3 days) or longer (10 days) treatments, although castration did increase androgen binding in the penile cytosol, androgen replacement merely normalized AR concentrations to those found in intact animals of the same age and failed to decrease penile ARs below intact levels. Western blot analysis confirmed that the AR protein remained at the normal concentrations in animals receiving T or DHT. Androgens were also unable to decrease penile ARs in intact immature rats or in immature rats receiving a GnRH antagonist and acted as expected on the stimulation of penile growth. As a result, the total penile AR content in intact rats treated with androgens was increased. No major differences were found between the immature penis and prostate in the response of ARs to androgen ablation and replacement. Therefore, it was concluded that androgens are not the main factors in the physiological down regulation of penile ARs during sexual maturation. Estrogens	also did not act in vivo as down regulators despite their effects in vitro. During sexual maturation the reduction of penile ARs is not as marked in terms of total content (2.5-fold) as it is in tissue concentration (30- to 50-fold). Although low, this basal AR concentration in the adult penis assures that some androgen effects continue to be exerted on the penis after it has stopped growing, particularly when combined with high serum T and DHT. This is essential for maintaining the reproductive function of the organ (namely, penile erection) and the adequate levels of nitric oxide synthesis required for a normal erectile response.1214 Therefore, it is most likely that there are different thresholds of androgen-bound ARs for penile growth and for erection. The fact that androgens are not major down regulators of penile ARs does not mean that they do not have an alternative effect on AR regulation. Speculatively, a certain basal level of tissue DHT may actually be required to help maintain the AR steady state content in the penis. This hypothesis is based on results obtained in vitro, where DHT acts as an up regulator of ARs in penile cells. Therefore, the decrease in penile tissue DHT, which has been reported to occur during sexual maturation,20 may be a factor contributing to the AR reduction and penile growth arrest, despite the increase in circulating DHT. A positive rather than a negative role for DHT in the maintenance of penile ARs would be more in agreement with the fact that this compound is essential for penile embryologic development.11 So far, the evidence for the presence of androgen-responsive elements in the rat androgen receptor promoter is weak,18 but this does not rule out an indirect positive role of DHT through other pathways. Role of the hypothalamic/pituitary axis and growth factors. In view of the preceding observations, the question of what compounds participate in the down regulation of penile ARs remains open. This process may occur through the intervention of actual nonandrogenic down regulators or the physiological decrease of up regulators. No gonadotropin-releasing hormone (GnRH), growth hormone (GH), luteinizing hormone (LH), or follicle stimulating hormone (FSH) receptors have been so far reported in the penis. However, some of them have been located in nongonadal tissues,21122 so that it is plausible that the penis may be a direct target for these effectors and that they may modulate penile ARs. In fact, it has been reported that FSH up regulates AR levels in the testis23 and down regulates them in the rat ovary.24 On the other hand, at least one growth factor, transforming growth factor B (TGF-B), has been investigated to determine whether it affects penile AR concentrations15 by giving it directly to the penis of immature rats for one week. Penile ARs were measured by western blot and were shown not to be affected by this treatment. This finding does not rule out that TGF-B at higher doses, or other cytokines presumably involved in penile growth such as IGF1 (see next page), may be able to decrease penile ARs. Pituitary hormones and growth factors are likely to modulate penile growth, irrespective of whether they do or do not control AR levels. These compounds may be part of an
SUBJECT OF CONTROVERSY: PENILE GROWTH DURING SEXUAL MATURATION
androgen-dependent cascade where androgens affect pituitary hormone and growth factor expression, or part of a completely androgen-independent pathway. The demonstration of the latter process would be of considerable relevance for the advanced stages of prostate cancer, characterized by metastatic cells that have become androgen independent. TGF-B has been shown to induce cell death in the rat prostate25 and its receptor has also been found to be down regulated by androgens in this tissue.26 In the immature rat penis, TGF-B1 mRNA is under negative regulation by androgens (Lin, Rajfer, Gonzalez-Cadavid, unpublished). Therefore, it is a likely candidate as a growth inhibitor in the penis whose concentration during sexual maturation would increase as ARs and androgen-mediated effects decline. This appears to be the case, as shown by immunodetermination of TGF-B in the penis from rats of increasing ages and by the significant decrease in the growth of the penile shaft induced by TGF-B administration to the penis of immature rats.15 Marking the difference with the prostate, TGF-B content in this organ was not similarly affected by age. Insulin-like growth factor 1 (IGF-1), basic fibroblast growth factor (BFGF) and GH have been shown to act as stimulators of prostate growth27*28 and are candidates to act similarly on the penis. Speculatively, these compounds may be responsible for the maintenance of a high rate of penile growth in the presence of comparatively low AR concentrations in the penis of 4-5-week-old rats. BFGF in particular is very abundant in the rat penis.29 We are currently testing how the concentration of these compounds varies in the penis as a function of sexual maturation. Regulation of Androgen Receptor Levels in Rat Penile Cells. Reversal of the age-related AR down regulation and role of sex steroids. Cells from the rat penile smooth muscle (RPSMC) have been used to study the regulation of AR expression and of cell proliferation by aging, androgens and growth factors. The paracrine effects occurring in vivo in the whole tissue are eliminated by the process of culturing RPSMC in the absence of the other cell types present in the penis. This is probably the reason why RPSMC cultured from the adult corpora cavernosa express AR mRNA and AR protein at the same level that RPSMC isolated from the immature rat,16 thus suggesting that in vivo other penile cell types down regulate ARs in the corpora cavemosa in a paracrine fashion. In agreement with the results obtained in vivo, DHT does not down regulate AR mRNA levels in RPSMC30 and actually causes a moderate up regulation. Although high doses of T with finasteride do reduce AR mRNA, this effect appears to be due to the conversion of T to estradiol, since the aromatase inhibitor fadrazol blocks the effect.31 Estradiol in turn is a powerful in vitro down regulator, despite its lack of effect in the corpora cavernosa. Therefore, androgens do not reduce penile tissue ARs either in vivo or in vitro. Role of proliferation conditions and growth factors. RPSMC	from adult or immature corpora cavemosa display in vitro the AR down regulation that occurs in vivo with the decline in growth rate. This occurs in highly confluent cells where AR mRNA levels are considerably reduced as compared with those in proliferating cells.30 As in the case of the physiological AR down regulation associated with sexual maturation, the in vitro decrease in ARs induced by growth arrest appears to be due to transcriptional blockade derived from lower binding of nuclear transcription factors to the AR promoter.18 The fow level of AR expression in growth-arrested cells is reduced even further by incubation with TGF-8,32 or stimulated by PDGF.33 The evidence accumulated on the regulation of AR expression in RPSMC is useful to define mechanisms of control operating autocrinely on the smooth muscle and to help to interpret the in vivo results. Although the RPSMC system is insufficient to characterize the complex paracrine interactions that occur in the corpora cavernosa of the intact animal, it allows molecular studies that would be impossible in vivo. Since RPSMC can be transfected with reporter constructs of the rat AR promoter and the AR coding region (Shen, Marquez, Rajfer, Gonzalez-Cadvid, unpublished), it should be possible to detect factors able to regulate AR expression and to determine whether AR hyperexpression stimulates androgen-dependent cell proliferation. Implications of Penile AR Regulation for Clinical Situations. The regulation of AR expression and its relationship with penile growth has scarcely been studied in human penile tissue due to the the difficulty in obtaining samples of the human corpora cavernosa. ARs are very tow in both the human adult corpora cavemosa and corpus spongiosum in comparison to the prostate.34 The only report on the effect of age on penile ARs is on human foreskin35 where the AR increases steadily by a factor of four from one year of age to puberty, and then declines to low levels in the third decade of life. This curve does not correlate with changes in serum T and provides a powerful argument against the involvement of circulating androgens in the postpubertal down regulation of ARs in the human penile skin. The process resembles the steady decline observed in the rat corpora cavernosa from the second week of age, but the marked up regulation of ARs observed in the human foreskin before and during puberty contrasts with what was found during the same periods in the rat. Most of the studies on androgen binding in androgen insensitivity syndromes36 have been carried out in primary cultures of human skin fibroblasts.37 Two different size forms of AR have been detected in these cells and other cells and tissues37 and designated A (87 kDa) and B (110 kDa). They are assumed to differ in the ability to transduce androgen effects in the cell. The ratio of the truncated to the normal form may vary in developmental situations and is worth exploring. The proliferation of genital skin fibroblasts can be stimulated by IGF-1, thus suggesting a possible in vivo role for this growth factor but no effects were observed on AR levels.38 ARs identical to those found in RPSMC have been found in cultures of smooth muscle cells from human penis and clitoris.39
A considerable number of mutations of the human AR gene have been identified which decrease or suppress the expression of AR protein or render it nonfunctional3640 while some of them are responsible for penile growth arrest and the subsequent development of micropenis or ambiguous genitalia. However, no mutations in the open coding region of the AR gene have been found in patients with isolated hypospadias.6 The studies on penile AR expression and growth in the rat model may provide an alternative therapy to the current empirical treatment of micropenis or ambiguous genitalia with T. This ne'w strategy may be based on the modulation of the hypothalamic/pituitary axis, the local administration of pituitary hormones and growth factors, or the stimulation of androgen responsiveness by AR penile gene therapy. None of these approaches would be possible before testing their effectiveness and safety in a suitable animal model, such as the normal growing penis or the induced micropenis in the rat.	androgens and nonandrogenic growth factors or hormones. Cessation of penile growth would therefore be codependent upon the cessation of stimulation by both the androgenic and nonandrogenic growth factors. This hypothesis is supported by the knowledge that approximately 20% of penile growth in the human occurs at a time when the androgen levels are sexually indifferent.78 It is further substantiated by the fact that other hormones, ie, growth hormone and thyroid hormone, have been shown to be partially responsible for penile growth.910 It is therefore hypothetically possible that the continued phallic growth found in the rodent after the down regulation of the AR may be due to continued penile growth from nonandrogen dependent growth factors. The final hypothesis is that cessation of penile growth is not even modulated by the AR. Indeed, rather than playing the major rofe in the cessation of penile growth, AR down regulation may be the consequence of still undefined cellular interactions that regulate penile size. According to this latter hypothesis, cellular interactions limit the size of the penis with down regulation of the AR being the result of these interactions. Rodent as an Animal Model for Phallic Growth. Regarding the cessation of penile growth and down regulation of the androgen receptor, it is extremely important to note that almost all of the data are based upon the findings in the rodent.1"6 Several differences between the rodent and the human penis may result in our inability to extrapolate animal data to human penile growth. In particular, the pattern of penile AR expression is significantly different between the two species, (see Dr. Takane's figures for rodent data and the figure on the following page for human data). The laboratory studies on the rodent AR reveal high levels of expression from the initial time evaluated until down regulation at 9-10 weeks of age.4"6 In the rodent, the down regulation of the penile AR appears to be cell specific with loss of the AR from the corporal bodies and the os penis, while the penile skin and urethra maintain high levels of AR.1 Interestingly, penile growth (as defined by persistent increase in penile weight and stretched penile length) in the rodent continues after down regulation of the AR46 (see Dr. Takane's figure.) In essence, down regulation of the rodent penile AR does not result in immediate cessation of the rodent's penile growth. When comparing the rodent to the human data, rather limited information regarding human phallic androgen receptor expression and cessation of penile growth exists. To be specific, no studies have been performed to see if significant androgen receptor expression differences exist between the corpora cavernosa and the penile skin with maturation. Indeed, the only human study available evaluating penile AR expression in relation to age used penile shaft skin.11 This study reveals two significant differences between the rodent and the human. First, AR expression in the human has a slight bimodel expression with elevations occurring during the newborn period and again at 16 to 20 years of age with down regulation of the AR occurring between 21-30 years of age. When we compare the presence of AR expression in the
DOUGLAS A. HUSMANN, MD Department of Urology, Mayo Clinic, Rochester, MN Cessation of Phallic Growth: Is It Really Determined By Androgen Receptor Down Regulation? Androgen Receptor Down Regulation and Cessation of Penile Growth. Perhaps the major question that exists regarding androgen receptor function and the cessation of penile growth is whether or not a cause-and-effect relationship truly exists. In essence, does down regulation of the androgen receptor cause cessation of penile growth? Indeed, in our laboratory endeavors, we have seen evidence suggesting that the two findings may not be related. To be specific, down regulation of the rat penile androgen receptor is present by 9-10 weeks of age,4"6 however, continued penile growth (ie, increase in penile wet weight and stretched penile length) occurs through 16 weeks of age.4-5 (Please note Dr. Takane's figures). The finding that penile growth continues to occur after the down regulation of the androgen receptor suggests one of three possibilities. First, androgen acts in the production of penile growth through a two-step steroid amplification model.4 In particular, in the two-step steroid amplification hypothesis, a steroid hormone activates its receptor which in turn directs the transcription (production of RNA from DNA) of a few specific genes. The few proteins produced by the androgen activated genes in essence stimulate the production of a multitude of additional proteins or growth factors, the latter of which are responsible for the physiologic effect of the hormone. It is important to note that this amplified response, due to the delayed production of the secondary proteins, may occur in either the relative absence of the activating hormone or alternatively at the time that the steroid receptor is apparently down regulated. It is therefore possible that the continued penile growth occurring after down regulation of the AR is still due to the effect of androgens. An alternative hypothesis is that penile growth is due to both
SUBJECT OF CONTROVERSY: PENILE GROWTH DURING SEXUAL MATURATION
	specific, the rodent has an os penis (a bone within the penis). Of extreme importance is that the os penis has a significant amount of AR present1 and is therefore extremely sensitive to androgen activity similar to the rest of the skeletal anatomy. Since excess androgens given to prepubertal animals are known to cause premature closure of the boney epiphyseal plates and inadequate skeletal growth, we hypothesize that the shortened penis developing from prepubertal androgen therapy for micropenis in the rodent*4 is due to the premature closure of the epiphyseal plate of the os penis. In contrast, the human has no analogous boney structure within their penile anatomy. Indeed, in contrast to the rodent, we have been unable to confirm that excess prepubertal androgens result in a diminutive penis in children with congenital adrenal hyperplasia14 or in children with micropenis secondary to the vanishing testis syndrome treated with prepubertal androgens.15 Because of the substantial difference in penile anatomy, penile androgen receptor expression, and the inability to correlate rodent to human data, we currently believe that the rodent animal model for micropenis may not truly reflect the nature of the human disease process. In summary, the accumulative data present in the human and animal models suggest that cessation of phallic growth is probably not caused by down regulation of the AR, nor by excess prepubertal androgens. Rather, we believe that AR down regulation in the penis occurs as a consequence of still undefined cellular interactions, the latter of which are responsible for the cessation of phallic growth. EDITOR'S COMMENTS: This issue presents a significant research forum that raises many interesting questions. The interaction of androgen receptor function and penile growth cessation needs further clarification. These significant studies, however, are provocative and promise to be a springboard for further research. Our thanks to Doug Husmann and his coauthors for bringing us current to what is going on in the investigation of this complex problem. Much more work is yet to come. Richard M. Ehrlich, MD To obtain a copy of citations for Drs. Takane, Rajfer and Husmann, please write to the publisher at Dialogues in Pediatric Urology, 45 Villa Road, Pearl River, NY 10965. YOUR RESPONSE We'd like your comments on the subject of this and other issues of this publication. Please send them to Richard M. Ehriich, MD, 100 UCLA Medical Plaza, Suite 690, Los Angeles, CA 90024. Opinions expressed in this publication are the sole responsibility of the individuals named and do not necessarily reflect the opinions of the editorial board or the publisher and members of this organization. Copyright © 1996 by William J. Miller Associates, Inc. All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying recording or by any information retrieval system, without written permission from the publisher.
human to cessation of phallic growth, we see that human penile growth ceases between 17-18 years of age1213 (see figure above). In essence, cessation of penile growth in the human is occurring at a time when AR expression is high and precedes the down regulation of the AR. (This latter finding causes this investigator to have significant concerns regarding whether or not our observations in the rodent will be applicable to those in the human.) One additional but significant concern is present regarding the rodent as a model for human penile growth. To be