Non-obstructive azoospermia is evaluated by the use of conventional histopathological methods. In some cases focal spermatogenesis is present in NOA testes which are almost undetectable by histopathological evaluations. Therefore, application of a molecular method can increase the probability to find sperm in NOA testis. Detection of germ cell-specific transcripts in semen can predict the patient's spermatogenesis state. This study was conducted to compare the molecular and histopathological methods for the evaluation of spermatogenesis in azoospermic men.
Semen samples were collected from 203 azoospermic men attending Avicenna Infertility Clinic in Tehran, Iran during 2007-2008. Total ribonucleic acid (RNA) was extracted from the semen precipitates. First-strand complementary deoxyribonucleic acid (cDNA) was synthesized by reverse transcriptase polymerase chain reaction (RT-PCR). Polymerase chain reactions were carried out using sequence specific primers for stage-specific genes (DAZ, AKAP4, PRM1 and PRM2). Biopsies of testicular tissue were prepared for histopathological examination. Serum LH, FSH and testosterone levels were measured by chemiluminescence method. Molecular results for DAZ and PRM2 genes in semen were compatible with the presence of spermatogonia and spermatid in testicular tissues, respectively. Absence of AKAP4 and PRM1 transcripts in semen due to germ cell maturation arrest confirmed the histopathological results. Most NOA men had elevated FSH levels in serum; however, measurement of gonadotropin levels cannot precisely predict minute foci of spermatogenesis in NOA patients. Using PRM1, PRM2, AKAP4 and DAZ transcripts in semen as non-invasive molecular diagnostic tools, can prevent a large number of NOA men from undergoing unnecessary testicular biopsy.
INTRODUCTION
Infertility affects 10-15% of all populations world-wide, half of which is due to male infertility (Matzuk and Lamb 2002). Non-obstructive azoospermia (NOA) which is due to impaired spermatogenesis in the testis, occurs in 5% of infertile male subjects (Marcelli et al. 2008; Van Peperstraten et al. 2008). Assisted reproductive technology (ART) has helped azoospermic men to conceive a child from their own gametes, according to the fact that focal spermatogenesis might still be present in small parts of the testes (Silber et al. 1997).
Approximately, 30-70% of NOA patients could successfully retrieve mature sperm applicable for intracytoplasmic sperm injection (ICSI) following microdissection testicular sperm extraction (mTESE) or testicular sperm aspiration (Devroey et al. 1995; Tournaye et al. 1995; Jezek et al. 1998; Sobek et al. 1998; Su et al. 1999; Houwen et al. 2008).
At present, different methods such as endocrine profiles (FSH, LH, Inhibin and AMH), histological examinations and testicular volume and consistency are used to predict the presence of mature sperm in testis. Yet, the predictive values of these methods are relatively low (Sobek et al. 1998; Goulis et al. 2008). In addition, testicular biopsy is an invasive procedure with a small coverage of the entire testis (Schlegel and Su 1997).
Regarding biopsy limitations and technical problems affecting the evaluation of spermatogenesis, more accurate methods are required to predict the presence of mature sperm in the entire testes of NOA men.
The presence of immature germ cells in semen at different stages of spermatogenesis, has been previously documented (Bassol et al. 1990; Huang and Agarwal 2004; Yeung et al. 2007). Presence of different stages of spermatogenic cells in the testis can be confirmed by detecting specific messenger RNAs, exclusively expressed at a relevant maturation level (Matzuk and Lamb 2008).
Recently, highly specific cellular and molecular methods such as flow cytometry (Koscinski et al. 2005) and Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) (Patrizio et al. 2000; Song et al. 2000; Bauer and Patzelt 2003; Ferras et al. 2004) have been used to analyze the presence of germ cells in the testis. However, molecular methods are mainly based on detecting testis specific transcripts in testicular biopsies rather than seminal fluid.
Identification of molecular markers in semen samples as non-invasive diagnostic tools, gives us a thorough image of the events going on in the testis (Yatsenko et al. 2006). This would yield prediction of even minute foci of spermatogenesis and small number of spermatozoa in NOA testis. In addition, this method will prevent patients with complete absence of spermatozoa (germ cell aplasia) to undergo the invasive procedure of testicular biopsy.
In this study, the presence of stage-specific gene transcripts in semen samples, histopathological findings and endocrine profiles were used to evaluate spermatogenesis status in testis of NOA subjects.
Results
Expression of testis-specific genes in semen precipitates was used as an indicator for the presence of different spermatogenic cells in the testis. The presence of testis-specific gene transcripts in semen was compared with testicular histopathology of NOA patients. In addition, Endocrine profiles of azoospermic men were compared with molecular finding on their semen precipitates. 93 semen samples were discarded due to patients' refusal to undergo testicular biopsy and consequently absence of histopathological results.
The mean age of patients was 35.51 ± 7.1 (Ranging from 23 to 63) years. The mean duration of infertility was 8.8 ± 6.9 years.
All semen samples were analyzed for the presence of β-actin, DAZ, AKAP4, PRM1 and PRM2 transcripts by RT-PCR (Figure 1). The expression of the aforementioned genes was proven in the testicular tissue with normal spermatogenesis as positive control.
Histological and histopathological evaluation of the testis from 110 azoospermic men were as follow: 65 men with sertoli cell only-syndrome (Germ cell aplasia), 4 men with spermatogonial maturation arrest (MA), 20 men with spermatocyte MA, 6 men with spermatid MA and finally 15 men with hypospermatogenesis.
The RT-PCR results for the four groups are summarized in figure 2.
According to the Kappa measure of agreement, there was a significant agreement between the expression of DAZ (Kappa value=0.197, p=0.035) and PRM2 (Kappa value=0.305, P=0.002) genes and the histopathological results.
Presence of PRM2 transcripts in semen had 70% sensitivity and a 77.5% specificity to predict the presence of mature spermatozoa in testis. Presence of AKAP4 and PRM1 transcripts in semen were not compatible with histopathological findings of testis in NOA men. However, the specificity for both genes was high enough to confirm the absence of spermatid / spermatozoa in testis (85.5% and 81.6%, respectively).
LH, FSH and testosterone concentrations in serum of NOA men were summarized in table 2. The probable relationship between the presence of gene transcripts in semen and hormonal results was monitored using independent-samples T-test. However, no significant results were found between the molecular findings and hormonal variations in NOA men.
DISCUSSION
At present, treatment of many NOA patients is possible by the use of spermatozoa obtained from TESE or mTESE followed by ICSI (Devroey et al. 1995; Silber et al. 1995b). Retrieving mature sperm with normal morphology from the testis of NOA patients plays an important role in these treatment approaches (Silber et al. 1995a; Tournaye et al. 1995).
Currently, determining the presence or absence of appropriate mature sperm in the testis of NOA patients is based on the traditional histopathological methods, which make use of testicular biopsy. Due to limitations of these methods, application of seminal based methods as non-invasive diagnostic tools are receiving more attention (Yatsenko et al., 2006). Immature germ cells in semen (Bassol et al. 1990; Huang and Agarwal 2004) could provide valuable information about spermatogenesis status in the testis. Koscinski used flow cytometry to differentially detect small amounts of immature germ cells in the semen of NOA patients (Koscinski et al. 2005). More specifically, the pattern of germ cell-specific gene transcripts in semen can be used as a molecular marker for depicting spermatogenesis status and may be a more suitable alternative for invasive diagnostic methods, such as testicular biopsy. Song has introduced PRM2 transcript as a molecular marker to predict the presence of spermatids / mature sperms in the testis of NOA patients (Lee et al. 1998; Song et al. 2000). However this method was based on invasive testicular biopsy with limited diagnostic values.
In the present study, more than 50% of the subjects were negative for germ cells in histopathological evaluations of their testicular biopsies (Germ cell aplasia). Hence, the establishment of a none-invasive and accurate screening test with high predictive values could prevent more than 50% of these patients from undergoing testicular biopsy.
A major association was observed between the presence of spermatogonia and spermatid/spermatozoa and the presence of DAZ and PRM2 transcripts in the semen precipitates of NOA patients, respectively. Expression of DAZ gene starts from premeiotic phase and continues up to the final stages of spermatogenesis (Menke et al. 1997; Lee et al. 1998; Szczerba et al. 2004). Consequently, DAZ might be a useful germ cell-specific internal control that confirms the presence of immature germ cells in semen; however, absence of DAZ transcripts and presence of β-actin transcripts in semen demonstrate germ cell aplasia in the testis; another internal control or proof for the aforementioned molecular procedures.
Song showed that deletion of DAZ gene in NOA patients that express PRM2 could cause production of spermatozoa with poor morphology (Song et al. 2000). Such genes can provide useful information about the pathophysiology of spermatogenesis failures (Lee et al. 1998), and provide more understanding on the molecular basis of spermatogenesis and causes of male infertility.
AKAP4 expression results were not compatible with histopathological findings, but its expression pattern specificity was high enough to confirm the absence of spermatids and mature spermatozoa in semen and subsequently in the testis. Since histology and histhopathological results were used as the golden standard to evaluate molecular results in this study; significant relation may be found between AKAP4 transcripts in semen with the presence of spermatid in testis using more precise evaluation method as golden standard.
An insignificant agreement was seen between the molecular and histopathological results, regarding the results of PRM1 transcripts in semen. Three reasons might justify this finding; first, although PRM1 and PRM2 genes are both located on chromosome 16q13.3 (Gene Bank, Z46940) (Domenjoud et al. 1991), their expression is totally independent (Viguie et al. 1990; Wykes et al. 1995). Second, translational control of PRM1 mRNA in spermatids is accomplished under the formation of ribonucleoproteins (RNPs) (Kleene 1996; Steger et al. 2000; Kleene 2003). Consequently, PRM1 transcripts, which are bound to repressor proteins, are not accessible by reverse transcriptase enzyme and subsequently remain undetected in the RT-PCR process. Finally, Steger has demonstrated that PRM2 gene is expressed twice as much as PRM1 gene in human testis using radioactive in-situ hybridization and Northern blot analysis (Steger et al. 2000).
Therefore, digestion of RNPs is recommended prior to evaluate expression of PRM1 and also PRM1 and PRM2 should be evaluated independently.
We also sought the association between the presence of PRM2 transcripts and presence of all other previously mentioned genes in semen. Interestingly, AKAP4 and PRM1 correlated significantly with PRM2 transcripts. In view of that, absence of AKAP4, PRM1 and PRM2 transcripts in semen depicts absence of mature sperm in the testis, ruling out the need for testicular biopsy.
This study proved the previously stated increases of gonadotropins, especially FSH, in most NOA patients, but testosterone concentrations were normal due to the normal function of Leydig cells and the interstitial tissue. Therefore, one may argue that sex hormone concentrations (especially FSH) could be helpful in diagnosing and differentiating NOA subjects from infertile patients with obstructive azoospermia (OA), but it cannot be used as a prognostic tool to predict the presence of sparse focal spermatogenesis in the testis.
As argued previously, presence of germ cell-specific DAZ and PRM2 transcripts in the semen of NOA patients can be used as specific non-invasive markers to predict the presence of mature spermatid / sperm in the testis. This method appears specific enough to help a large number of NOA patients avoid unnecessary diagnostic testicular biopsies.
In conclusion, to raise the possibility of detecting minute foci of spermatogenesis in patients with non-obstructive azoospermia application of high precision and none-invasive methods are needed. Combination of highly specific molecular tools using different stages testis specific genes (PRM1, PRM2, AKAP4, DAZ …) and highly sensitive flow cytometry methods can help to choose the best treatment approach in NOA Men. The negative molecular results for the above mentioned genes may help us prevent most of NOA men from undergoing invasive testicular biopsy. In future studies analysis of other germ-cell specific genes in larger populations would further confirm the results obtained in this study.
MATERIALS AND METHODS
Semen Sample Collection
Semen samples were collected from 203 azoospermic men attending Avicenna Infertility Clinic for infertility treatment. The semen samples were analyzed in accordance with the WHO guidelines (1999). As some couples refused to follow their infertility treatment, testicular biopsy was done only on 110 azoospermic men and their semen samples were chosen for further study. Written informed consent was taken from all participants for the use of their semen in the study. The Avicenna Research Institute's (ARI) medical ethics committee approved this study. Testicular tissues of azoospermic men with all stages of spermatogenesis including mature spermatozoa were used as positive control.
Semen Sample Preparation
Following liquefaction of semen at 37°C, each sample was centrifuged at 200 g for 15 minutes. The pellets were washed 3-4 times by sterile phosphate buffered saline (PBS, 0.5 M) at 300 g for 5 min. Final pellets were kept in liquid nitrogen (-196°C) prior to RNA extraction.
RNA Extraction
Frozen samples were thawed and washed twice by sterile PBS. The total RNA was extracted using RNA Bee kit (Biosite, Sweden). Lysis buffer (600 µl/106 sperm) was added to the samples. The lysates were homogenised with a pellet pestle (Sigma, USA) in a phenol-chloroform-isoamylic acid solution. Isopropanol was added to isolate and precipitate RNA from the aqueous phase. The isolated RNA was washed twice by ethanol (75%) and air dried. The pellet was dissolved in diethylpyrocarbonate (DEPC)-treated water and stored at -70°C (Chomczynski and Sacchi 2006; Goodrich R 2007; Aarabi et al. 2008). The purity of extracted RNA was checked spectrophotometrically (Biophotometer, Eppendrof, Germany) at 260 and 280 nm.
RT-PCR Amplification
First-strand cDNA was synthesized by using 150 µg of the total RNA. The cDNA reactions were performed at 42ËšC in 60 minutes. The first-strand cDNA synthesis reaction mixture was prepared by using 10 µl of the isolated mRNA, 2 µl of 20 mM random hexamer (Roche, Germany), 2 µl of 10 mM dNTPs (Roche, Germany), 4 µl of 5X M-MLV reverse transcriptase buffer (Fermentase, EU) and 200 IU M-MLV-RT (Fermentase, EU) in a final volume of 20 µl.
PCR amplifications were carried out in a 25 µl reaction mixture, composed of 1 µl of the prepared cDNA, 1 µl of each primer (10 pM) (Bioneer, South Korea), 1 µl of 10 mM dNTPs (Roche, Germany), 1 IU Taq polymerase (Roche, Germany), 3 µl of 25 mM MgCl2 (Roche, Germany) and 2.5 µl of 10X PCR buffer (Roche, Germany).
PCR Amplification
Testis-specific genes were selected according to their expression stage. Deleted in azoospermia (DAZ) gene expression starts at spermatogonial level (Menke et al. 1997; Lee et al. 1998; Warchol et al. 2001; Szczerba et al. 2006; Huang et al. 2008). A-kinase anchoring protein-4 (AKAP4) gene is expressed only in spermatid stage (Brown et al. 2003). PRM1 and PRM2 expressions were regarded as markers indicative of the presence of late spermatid and mature sperm (Matzuk and Lamb 2008). β-actin was used as a housekeeping gene.
All primers were chosen from different exons to eliminate possible contamination by genomic DNA (Table 1).
For positive control reactions, cDNA was synthesized from testicular tissues of the azoospermic men containing mature sperm and all stages of spermatogenic cells. Positive and negative (Water) controls were included in all PCR amplifications. PCR products were run on 1.5% agarose gel, subsequently stained by ethidium bromide and visualized by UV transillumination (Richter et al. 1999; Lambard et al. 2004).
Histopathological Examination
Biopsied testicular tissues of azoospermic men were prepared for histopathological examination. Paraffin-embedded sections of testicular tissues were dewaxed, rehydrated and stained with hematoxylin and eosin (Song et al. 2000; Aarabi et al. 2006).
Results from histopathological evaluation fell into five categories: (1) sertoli cell only-syndrome (Germ cell aplasia), (2) spermatogonial maturation arrest (MA), (3) spermatocyte MA, (4) spermatid MA and (5) hypospermatogenesis.
Hormonal Analysis
All NOA patients underwent hormonal (LH, FSH, testosterone) assessment using a two-site immunoluminometric method (LIAISON, Italy) according to the manufacturer's manual. The hormonal levels in patients were categorized into high, normal and low levels according to hormone's reference ranges.
Statistical Analysis
The data was statistically analyzed, using SPSS v.13.0. Kappa coefficient was used for agreement analysis between molecular and pathological results. Sensitivity and specificity for each molecular marker (gene transcript) was determined according to corresponding pathological results (Gold standard test). Correlation of molecular and hormonal results was analyzed in each sample using the independent-samples T test.
