Background
Non-obstructive azoospermia (NOA) affects approximately 1% of the general population and 10–20% of infertile men worldwide [
1,
2]. A series of factors were associated with NOA, for example, hypogonadotropic hypogonadism (HH), Y microdeletion, chromosomal abnormalities etc. [
3]. The causes and the underlying mechanism of idiopathic NOA still remain unclear. Testicular sperm extraction (TESE) or microdissection testicular sperm extraction (micro-TESE) combined with intracytoplasmic sperm injection (ICSI) was the approach recommended for idiopathic NOA [
4]. However, the total rate of sperm retrieval was only about 50% [
5]. Thus, efficient medical treatment strategies are required.
Hormone replacement therapy would improve the ability of the testis to produce spermatozoa in idiopathic NOA patients [
6,
7]. For example, the improvement of spermatogonial DNA synthesis was demonstrated by Shinjo and coworkers [
8], the elevation of intra-testicular testosterone levels was demonstrated by Kato and coworkers and the hypertrophic change of leydig cells was demonstrated by Oka and coworkers, respectively [
9,
10]. Recently, a multi-institutional prospective study conducted by Shiraishi and coworkers provided a stronger evidence of the efficiency of hormone therapy [
6]. However, the total rate of acquiring sperm was only about 10–20%. A possible explanation of the low success rate was that high plasma gonadotropins in the patients led to dysregulated function of FSH and LH receptors (FSHR, LHR) in Sertoli and leydig cells [
7,
11,
12]. As demonstrated by in vivo and in vitro studies, desensitization and downregulation of FSH signaling in Sertoli cells was induced by the chronic stimulation of FSH [
13‐
15]. Considering the risk of high plasma gonadotropins, a ‘gonadotropin reset’ with leuprolide acetate, a gonadotropin releasing hormone agonist (GnRHα), was proposed to induce a hypogonadotrophic state by Foresta and coworkers [
11]. Thus, the FSHR and LHR in the testis would be ‘released’ and subsequent exogenous hormone stimulation would be beneficial for testis spermatogenesis, as great success has been achieved in the treatment of hypogonadotropic hypogonadism via hormone replacement therapy. Moreover, gonadotropin reset with GnRHα had been demonstrated to improve the function of Sertoli cells and subsequently enhance the sperm concentration in patients with severe oligozoospermia [
11,
16]. However, to our knowledge, there is no data of gonadotropin reset with GnRHα in the NOA patients.
Inhibin B is secreted by Sertoli cells and is involved in the negative feedback of plasma FSH [
17]. The expression of inhibin B was regulated by FSH and plasma inhibin B level was considered as a marker of Sertoli cell function [
18]. Plasma inhibin B level was also closely related with spermatogenesis. Low levels of inhibin B was demonstrated in patients with bad semen quality which may be related with the dysfunction of Sertoli cells [
11,
18]. Moreover, elevated inhibin B levels may indicate improved function of Sertoli cells, reflecting better spermatogenesis environment [
11] .
Cell-cell junction in the seminiferous epithelium played an important role in spermatogenesis including self-renewability and differentiation of spermatogonial stem cells into mature spermatozoa [
19]. Blood testis barrier (BTB) mainly includes tight junctions (TJ) that are present between adjacent Sertoli cells [
19]. Redistribution of the TJs to the cytoplasmic compartment or decreased expression was associated with abnormal spermatogenesis [
20,
21]. NOA patients were also accompanied with dysfunction of TJ proteins, such as occludin 11 etc. [
21]. Zonula occludens-1 (ZO-1) is a membrane protein that distributed peripherally, and interacted together and anchored membrane proteins to the actin cytoskeleton [
22]. However, the expression pattern of ZO-1 has never been reported in the NOA patients.
Hence, in the present study, we tried to suppress the high endogenous gonadotropin levels in the NOA patients with goserelin, another GnRHα to release and restore the receptors’ function and then stimulate them using human menopausal gonadotropin (hMG) and human chorionic gonadotropin (hCG) to improve the ability of testicular spermatogenesis. Inhibin B, ZO-1 and mouse vasa homologue (MVH, a marker of germ cells) were detected to evaluate the response to the intervention.
Discussion
As mentioned previously, TESE or micro-TESE combined with ICSI was the only recommended treatment for NOA patients till date [
1,
2]. However, the patients who failed to acquire sperm in the TESE remained a problem in the whole world. A series of studies have been reported regarding the treatment strategies for NOA patients including empirical medical therapies like testolactone [
27,
28], hormone replacement therapy etc. [
8,
9]. However, the results were not ideal. In 2012, Shiraishi and coworkers tried to trigger ‘gonadotropin reset’ with high-dose hCG stimulation. As a result, the high gonadotropins in NOA patients were decreased and the ability of testicular spermatogenesis was improved [
7]. This indicated that gonadotropin reset is an efficient approach in the NOA patients. However, the dose of hCG was too high in clinical practice. Therefore, we enrolled patients with NOA and explored another way to fulfill gonadotropin reset with GnRHα in the present study.
As we all know, spermatogenesis is regulated via a complex array of paracrine, endocrine and juxtacrine regulatory cross-talk involving Leydig, Sertoli, peritubular, germ cells etc. [
29]. FSH and testosterone were believed to play vital roles in spermatogenesis by stimulating Sertoli cells [
30]. However, the precise mechanism still remain unclear. Studies in animals demonstrated that FSH was involved in the proliferation of early stage germ cells, and testosterone is one of the most important factors for initiating and maintaining spermatogenesis [
31,
32]. As we all know, plasma FSH levels were often elevated in patients with NOA [
7]. Significantly elevated FSH levels had been demonstrated to decrease FSH-FSHR signaling pathway in the testes of NOA patients. The possible explanation may be: (i) the uncoupling reaction of the FSHR from the effector system led to the constant phosphorylation of the C-terminal, intracellular domain of FSHR [
33,
34]; (ii) decreased number of FSHRs was mediated by extensive clustering and internalization of the FSH–FSHR complex and by reduced expression of the receptor as a result of both decreased transcription levels and reduced half-life of mRNA [
15,
35]. In view of the effects of high endogenous FSH, Foresta and coworkers have demonstrated that, leuprolide acetate treatment decreased the high endogenous gonadotropin levels, thus activating the FSH receptors. Followed by stimulation with exogenous FSH, the semen quality of the patients were significantly improved [
11]. Based on these results, we tried to regulate endogenous gonadotropin levels with goserelin in the NOA patients in the present study. As a result, inhibin B was significantly lower than normal range in the NOA patients (Table
1), indicating injuried function of Sertoli cells. Interestingly, inhibin B levels were elevated in 11 of the 25 patients, indicating Sertoli cell function was improved after inhibition of endogenous gonadotropins in the first 4 weeks. The other 14 showed no response. This may be due to the excessive damage of Sertoli cells in the testis as the basic plasma inhibin B levels were significantly lower than the Response group (
p = 0.009, Fig.
4f) which is consistent with the result of Foresta and coworkers who had demonstrated that the function of Sertoli cells was severely injuried and remained irresponsive to hormone treatment while the basic plasma inhibin B was low [
11]. During the following 20 weeks, only 5 patients of the Response group (Response group 2) showed a constant increase of inhibin B, while the other 6 (Response group 1) did not. There was no difference observed in the basic plasma inhibin B levels between Response group 1 and 2, but an increased tendency was shown in Response group 2 (
P = 0.06). This might possibly be due to that the function of Sertoli cells were incomplete in the Response group 1, i.e., the cells were not enough to initiate and maintain spermatogenesis as there was no significant change of the MVH signals after treatment (Fig.
5d and
6d). The inhibin B levels would be higher in presence of germ cells as they are also considered to be the source and involved in the secretion of inhibin B [
36,
37]. Correspondingly, significant increase of positive signals of MVH in Response group 2 indicated the proliferation and meiosis of the germ cells after the treatment (Fig.
6e‑h), therefore, the inhibin B levels were elevated. All of these indicated that plasma inhibin B level may act as a good marker to predict spermatogenesis in the testis and to evaluate the response of the therapies in the NOA patients.
As mentioned previously, BTB played an important role in the spermatogenesis [
19]. Aberrant expressions of TJs were associated with abnormal spermatogenesis [
38]. ZO-1, occludin, and claudin are the TJ proteins identified in the testis [
39]. In the present study, the distribution of ZO-1 was punctuate and discrete in NOA patients as shown in Fig.
6b compared with patients of OA whose ZO-1 distribution were normal and consecutive. The expression and distribution of ZO-1 were improved in Response group 2B with mature spermatozoa were observed in the testis while only expression of ZO-1 was improved in Response group 2A with no mature spermatozoa, although the number of germ cells was increased significantly (Fig.
6e,
f). This indicated that well-distributed ZO-1 played an important role in the sperm maturation changes, which was consistent with other studies [
39,
40]. Under the physiological conditions, the main factor affecting the TJ formation is the endogenous testosterone [
29]. Besides, FSH may also play a key role in the expression and localization of TJ proteins [
29]. The ameliorative expression and distribution of ZO-1 in the Response group 2B also indicated the effectiveness of the protocol in restoring the function of FSHRs and LHRs and the elevation of intra-testicular testosterone levels by subsequent hCG treatment that has been demonstrated by others [
9,
10].
As known previously, HP, MA and SCOS were the three main histological classifications of NOA [
3]. SCOS is excluded here as no tubules contain germ cells in the testis specimens of the patients [
41], hence the possibility of acquiring sperm was very low [
6]. Spermatogenesis of MA was often arrested at the spermatogonial or primary spermatocyte stage. It was mainly related to the presence of a genetic lesion or toxicant exposure [
41]. The process of spermatogenesis in HP was almost intact, but reduced to a certain extent. Hence, we supposed that it was related with terrible testis microenvironment and improve the microenvironment via hormone may be beneficial for spermatogenesis [
7]. HP were also believed to be the ideal objects for medical treatment in the previous studies [
6,
7]. Moreover, the cost of the treatment was high. Therefore, the clinical trial was performed in patients who are more likely to succeed. Besides, there is a deficiency in the present study. The effectiveness of the protocol in the study was lack of strict “control group” who should be given only hCG combined with hMG in the process, that is to say, the role of goserelin in the study was lack of evidence. However, the elevation of inhibin B in the first 4 weeks during which goserelin was given alone indicated the positive effect of it. Moreover, Foresta and coworkers had demonstrated that FSH alone would not enhance plasma inhibin B in patients with high endogenous gonadotropins as the Sertoli cells were irresponsive [
11]. Therefore, the reactivity to FSH was restored via the use of goserelin in the present study as reflected by constant increased inhibin B following hMG treatment was another proof for the role of goserelin, which is also consistent with the result of Foresta and coworkers [
11].