Occurrence of testicular microlithiasis in androgen insensitive hypogonadal mice

Testicular microlithiasis (TM) is characterised by the presence of microcalcification within the seminiferous tubules. In the normal human population the incidence is between 1.5 and 5.6% [1‐3] and in itself microlithiasis is benign but there may be significant association with malignant conditions such as testicular germ cell tumors (TGCT) [4‐7] and other conditions such as cryptorchidism, varicocele, infertility, and testicular torsion [8, 9]. In addition, it has been suggested that there may be a genetic predisposition to TM which is linked to TGCT formation so that identifying the underlying causes of microlithiasis may help identify susceptibility factors for TGCTs [5]. Development of an animal model of TM would, therefore, not only represent significant progress towards an understanding of the origins and underlying molecular mechanisms of microlithiasis but may help identify associated risk factors for other conditions.

The hypogonadal (hpg) mouse lacks gonadotrophin-releasing hormone (GnRH) and circulating levels of luteinising hormone (LH) and follicle-stimulating hormone (FSH) are undetectable [10]. As part of an ongoing study of the endocrine regulation of spermatogenesis we have crossed these hpg mice with mice lacking androgen receptors (AR) either ubiquitously (ARKO mice) or specifically in the Sertoli cells (SCARKO mice). In this report we describe the unexpected and consistent occurrence of microlithiasis in hpg.ARKO mice.

Microscopic analysis of hpg.ARKO mice showed that in 94% (n = 16) of these animals non-cellular structures, closely resembling the morphology of human testicular microliths [15], were present within the seminiferous tubules (Fig 1). In section the microliths had a round/oval appearance with a maximum diameter of about 45 μm (Fig 1). Most microliths were present singly but in some sections 2 or 3 microliths were present within the same tubule section. Some microliths were closely apposed to Sertoli cells while others had only occasional close cellular associations (Fig 1B&1C).

Examination of hpg and hpg.SCARKO mice showed that occasional, small microliths were present in 36% (n = 11) of hpg testes and 30% (n = 10) of hpg.SCARKO testes (Fig 1D). We did not observe any evidence of the presence of microliths in ARKO mice (n = 4) or in SCARKO mice (n = 7) or in normal cryptorchid mice (n = 4).

The mean number of microliths/testis in hpg.ARKO mice was 81 ± 54 while in hpg and hpg.SCARKO mice there were 2 ± 0.5 and 8 ± 6 microliths per testis respectively. Treatment of hpg, hpg.SCARKO or hpg.ARKO mice with FSH did not alter the number of microliths. Treatment of mice with androgen increased the average number of microliths per testis in hpg.ARKO mice but this was not significant (incidence after T = 234 ± 75, incidence after DHT = 241 ± 75).

The generation of an animal model of TM represents significant progress towards understanding this condition. It is unlikely that the cause of TM in the hpg.ARKO mouse is identical to that in most human cases of TM since TM arises in individuals with normal endocrinology although, interestingly, TM has been shown to be diagnostic of McCune-Albright Syndrome [16] in which testicular symptoms appear to arise most commonly from precocious activation of the Sertoli cells [17]. Irrespective of the cause, importantly we have identified conditions in which there is consistent development of TM and the molecular mechanisms involved in generation of TM in hpg.ARKO mice are likely to be similar to those in the human. This mouse model can be used, therefore, to study the generation and characteristics of TM including identification of changes in gene expression and protein secretion associated with formation of TM. In addition, the hpg.ARKO mouse now provides a chance to trace the origins and physical development of testicular microliths.

In humans TM is normally diagnosed ultrasonically by the presence of multiple 1-3 mm echogenic foci which are clearly far larger than the structures observed microscopically in this study. This is likely to be a reflection of the differences in testis size and tubule diameter between human and hpg.ARKO mouse testes and the morphology of the microliths appears very similar between mouse and human [15]. Nevertheless, it should be noted that the criteria used to define TM in this study is microscopic, rather than ultrasonic, and direct extrapolation to the human condition must be made with caution. Clearly, in this respect, it will be of interest to determine whether there is any development of TGCT in hpg.ARKO mice as they age beyond early adulthood.

The origin of testicular microliths is uncertain. Most studies favour a tubular origin with accumulation of debris from degenerating cells being followed by secretion of glycoproteins and subsequent calcification [18, 19]. Others suggest that microliths originate outside the tubule and then either move into the tubule or become engulfed by the tubule folding around it [15, 20]. Developmental studies of TM in the hpg.ARKO mouse should provide a clear answer to the origin of microliths. It may be of interest, however, that microliths are much more prevalent in hpg.ARKO mice than in hpg.SCARKO mice since one of the differences between these two animal models lies in the presence of ARs in the extratubular space.

It is unlikely that microliths form consistently in the hpg.ARKO mouse simply because there is abundant degeneration of the germ cells. This may be a contributory factor but there is marked degeneration of germ cells in hpg, SCARKO and ARKO mice and only in hpg mice did we see any evidence of occasional microliths in some animals. Similarly, cryptorchidism may be a contributory factor to microlith formation in hpg.ARKO mice but is unlikely to be the principal cause as hpg and ARKO mice are cryptorchid and no evidence of microliths was seen in normal animals rendered cryptorchid by surgical means. In hpg.ARKO mice the epididymides fail to form which means that there will be a block to the flow of fluid through the seminiferous tubules. This may also be a contributory factor to the formation of microliths in the hpg.ARKO mouse but is unlikely to be the major cause since there is a similar failure of epididymal development in ARKO mice. Removal of gonadotrophin stimulation is clearly a contributory factor to microlith formation as, in itself, it appears to be sufficient to induce formation of occasional microliths in the hpg mouse. Overall, therefore, it may be a combination of factors (degenerating cells, cryptorchidism, lack of tubular fluid flow), linked to loss of gonadotrophin- and androgen-stimulation, that leads to consistent TM development in the hpg.ARKO mouse.