SEARCH

SEARCH BY CITATION

Keywords:

  • varicocele;
  • nitric oxide;
  • nitric oxide synthase

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. CONFLICT OF INTEREST
  9. REFERENCES

OBJECTIVE

To determine the concentration of nitric oxide (NO) and the location and change in the expression of NO synthase (NOS) isoforms in the testes of subfertile men with varicocele, and to compare the NO concentration or NOS expression with clinical variables, to determine the role of NO on the pathophysiology of varicocele.

PATIENTS AND METHODS

In all, 27 men who had a left varicocelectomy and five with ‘normal’ spermatogenesis (controls) who had scrotal surgery for other reasons were enrolled. Intratesticular fluid was taken from the men and the NO concentration determined colorimetrically. The expression and location of NOS isoforms were examined by Western blotting and immunohistochemistry, using testicular biopsy specimens, and NADPH-diaphorase (NADPH-d) staining used to identify NO-producing cells. The relationship between the NO concentration and the expression of NOS isoforms or clinicopathological variables was investigated.

RESULTS

In testes with grade 2 and 3 varicoceles there were significant increases in the concentration of NO or the expression of inducible NOS. There was no change in the expressions of endothelial NOS, which is located in vascular endothelial cells, while NADPH-d activity was mainly located in these cells. The concentration of NO was significantly correlated with the maximum and total vein diameter (both P < 0.01). In patients aged >35 years, the concentration of NO significantly correlated with a deterioration in total motile sperm count (P < 0.05).

CONCLUSIONS

Increased production of NO in the testis is involved in the enlargement of varicocele and indirectly deteriorates spermatogenesis.


Abbreviations
NO(S)

nitric oxide (synthase)

ITF

intratesticular fluid

TM

total motile sperm

4-HNE

anti-4-hydroxy-2-nonenal

(e)(n)(i)NOS

(endothelial) (neuronal) (inducible) NOS

NADPH-d

NADPH-diaphorase.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. CONFLICT OF INTEREST
  9. REFERENCES

Oxidative stress causes a deterioration in spermatogenesis and sperm function in men with a varicocele [1]. In the human testis with varicocele, evidence of oxidative stress was shown as increased levels of malondialdehyde [2] and 4-hydroxy-2-nonenal (4-HNE)-modified proteins [3], which represent the generation of lipid peroxidation products. To provide strategies for the diagnosis and treatment of oxidative stress in testes with varicocele, it is necessary to identify specific molecules that are involved in oxidative stress. Nitric oxide (NO) is a short-lived free radical and is one of the well-known factors that causes oxidative stress in many diseases. NO has been implicated in various reproductive functions and the location of NO synthase (NOS) isoforms in the testis has been investigated [4–7].

In men with varicocele there is excessive release of NO into the spermatic vein [8–10], indicating the production of NO in the testis. Recent studies using a rat experimental varicocele showed increased production of NO in testes with varicocele [11,12], although there is no information as to whether the increased amount of NO is involved in oxidative stress and has harmful effects on spermatogenesis. Santoro et al. [13] reported that increased inducible NOS (iNOS) protein in Leydig cells might cause a deterioration in spermatogenesis and cause sperm dysfunction in varicocele. The evaluation of NOS activity (i.e. the production of NO) is crucial for investigating functional aspects.

The aim of the present study was to determine the concentration of NO in the testes of subfertile men with varicocele, by measuring the nitrite (a stable metabolite of NO) concentration in intratesticular fluid (ITF) using the Griess method. Furthermore, the location and changes in the expression of NOS protein isoforms were examined using testicular biopsy specimens. Last, we compared the NO concentration or NOS expression with clinical variables, to determine the role of NO in the pathophysiology of varicocele.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. CONFLICT OF INTEREST
  9. REFERENCES

In all, 27 patients with varicocele in the left testis (subclinical in four; grade 1 in nine; grade 2 in nine; grade 3 in five) were enrolled in this study (mean age 33.2 years, sd 6.1, range 24–49). Varicocele was diagnosed after a physical examination and duplex and colour Doppler ultrasonography. The testicular volume was determined using a ‘punched-out’ orchidometer. Semen samples were analysed at least three times before varicocelectomy. The total motile sperm count (TM) was calculated by multiplying sperm concentration (× 106/mL) and semen volume (mL). The mean TM was calculated for each patient. Microsurgical varicocelectomy was performed through a sub-inguinal approach under spinal anaesthesia. Before ligating each isolated vein, its diameter and total number of veins were recorded. At the end of surgery, ITF and biopsy samples were collected from the left testes via a scrotal incision. Five control testicular biopsy specimens were also taken from men with ‘normal’ spermatogenesis who had scrotal surgery for other reasons. All the physical examinations, microsurgical sub-inguinal varicocelectomy and biopsy were done by one of the authors (K.S.). After explaining the purpose of this study, informed consent from all patients, including those providing control samples, and institutional ethical approval were obtained.

The ITF was obtained using the method reported by Jarow et al. [14] with some modifications. Before incising the tunica albuginea, an 18-G butterfly needle attached to a 10-mL syringe with stopcock was placed into the testis; suction was applied intermittently until fluid appeared in the tube. The ITF was transferred to a microcentrifuge tube and centrifuged at 300 g, at 0 °C. The supernatant was stored at −80 °C until assayed. The nitrite concentration in the ITF was measured using the Griess method [15] with a commercially available kit (Promega, Madison, WI, USA); 50 µL of the diluted samples, which were typically dilute two- or five-fold, were added to a 96-well plate. The nitrite concentration (µmol/L) was measured by recording the absorbance at 540 nm after 10 min incubation at 37 °C in the dark, in duplicate.

Frozen specimens were homogenized in a buffer with protease inhibitors. Equal amounts of protein (15 µg) were electrophoresed on 7.5% gels and transferred to polyvinylidene difluoride membranes. After blocking, the membranes were incubated with rabbit polyclonal endothelial NOS (eNOS) antibody (at a dilution of 1 : 1000, Santa Cruz Biotech., CA, USA), neuronal NOS (nNOS) antibody (1 : 500, Santa Cruz), iNOS antibody (1 : 1000, Santa Cruz) or mouse monoclonal anti-4-HNE-modified protein (1 : 1000; Japan Institute for the Control of Ageing, Shizuoka, Japan) in 1% BSA overnight at room temperature. The membranes were then reacted with secondary antibody at room temperature for 1 h. The antigens were visualized by chemiluminescence using an ECL Western blotting detection kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA) and quantified using an image analyser (Densitograph AE-6900M; Atto Co., Tokyo, Japan). For the 4-HNE-modified protein, the sum of total bands per lane between 247 and 43 kDa were calculated. The data were calculated as a percentage of the control where control values were assigned a value of 100%. Data were expressed as the mean (sem).

After fixation in Bouin’s solution for 30 min, 4-µm paraffin wax sections were mounted on silane-coated glass slides (Dako Japan, Kyoto, Japan). After inhibiting endogenous peroxidases in H2O2, antigen was retrieved by heating in citrate buffer (pH 6) at 98 °C. Overnight incubation was carried out using the antibodies for eNOS (at a dilution of 1 : 200), nNOS (1 : 100) or iNOS (1 : 100) at 4 °C. Goat anti-rabbit immunoglobulin was used as a secondary antibody (1 : 100, Dako) for 1 h. Sections were counterstained with haematoxylin.

The location of NADPH-diaphorase (NADPH-d) activity, which was confirmed by the co-location of NOS activity, was detected histochemically according to methods described previously [16], with modifications; 6-µm cryostat sections were placed in 4% formaldehyde/PBS, pH 7.4 for 1 h at 4 °C. The slides were washed three times in 0.1 m Tris-HCl buffer (pH 8.0) and then incubated in 1 mg/mL NADPH, 0.25 mg/mL nitroblue tetrazolium and 0.2% Triton X-100 in 0.1 m Tris-HCl buffer in the dark for 1 h at room temperature. Subsequently, the sections were rinsed with 0.1 m Tris-HCl buffer and mounted. In the negative controls β-NADH was used instead of β-NADPH.

The differences in nitrite concentration or expression of the NOS isoforms between each group were evaluated by the Kruskal–Wallis test. To examine the relationship between nitrite concentration and the expression of NOS isoforms or clinicopathological variables, Spearman’s rank correlation coefficients were used; P < 0.05 was considered to indicate statistical significance.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. CONFLICT OF INTEREST
  9. REFERENCES

The mean (sem, range) volume of the ITF taken was 31 (6, 23–55) µL. All the samples were clear-yellow with no blood or turbidity. The mean (sem) concentration of nitrite in the control, subclinical + grade 1 and grade 2 + 3 were 26.0 (2.1), 29.6 (2.0) and 48.2 (6.3) µm, respectively. In testes with grade 2 or 3 varicocele, there was a significant increase in nitrite concentration (P < 0.05, Fig. 1). All the NOS isoforms appeared as a single band with a molecular mass of ≈130 kDa (Fig. 2). Membranes incubated in the presence of non-immune rabbit IgG in the absence of primary antibody showed no immunoreactivity. There was no change in the expressions of eNOS and nNOS proteins among the controls and different grades of varicocele. There was significantly more iNOS protein (1.5 times) in testes with grade 2–3 varicocele than in the controls or lower grades of varicocele (P < 0.05). In testes with grade 2–3 varicocele, expression of iNOS protein was 0.7–7 times that of the controls, resulting in a large sem. There was a significant relationship between nitrite concentration and the expression of iNOS, but not eNOS or nNOS (P < 0.05, Table 1). As we reported previously, many proteins are detected using the anti-4-HNE-modified protein antibody [3], and the sum of the band intensities was significantly higher in testes with grade 2 or 3 varicocele (data not shown). There was no significant correlation between nitrite concentration and generation of 4-HNE-modified proteins (Table 1), indicating that NO is not directly associated lipid peroxidation.

image

Figure 1. The ITF nitrite concentration measured by the Griess method in control and varicocele testes. Data are expressed as the mean (sem). *P < 0.05 vs control or subclinical + grade 1. sub, subclinical; G1, grade 1; G2, grade 2; G3, grade 3.

Download figure to PowerPoint

image

Figure 2. Representative immunoblots for eNOS (A), nNOS (B) and iNOS (C) and their quantitative data from controls and patients with varicocele. Data are expressed as the mean (sem) multiple of increase above the control. *P < 0.05 vs control or subclinical + G1. sub, subclinical; G1, grade 1; G2, grade 2; G3, grade 3.

Download figure to PowerPoint

Table 1.  The correlations between nitrite concentration in ITF and NOS expressions in testis
VariableRP
  1. R, Spearman’s rank correlation coefficient; NS, not significant.

eNOS expression0.014NS
nNOS expression0.191NS
iNOS expression0.162<0.05
Generation of 4-HNE modified proteins0.316NS

eNOS was primarily located in the cytoplasm of vascular endothelial cells (Fig. 3A) and there was weak staining in the cytoplasm of Leydig cells. Sertoli and germ cells were not positive for eNOS. There was no difference in these staining patterns of eNOS protein between the control and varicocele testes. There was staining for nNOS in Leydig cells; representative nNOS-positive Leydig cells are shown in Fig. 3B. Some Leydig cells had low or negative immunoreactivity and some Sertoli cells were also weakly stained. There was no difference in these staining patterns of nNOS protein between the control and varicocele testes. There was no iNOS immunoreactivity in control testes (Fig. 3C). The expression of iNOS in the varicocele testes varied in each case. Figure 3D shows a representative section positive for iNOS. Eight of the 14 cases with grade 2 or 3 varicocele had immunoreactivity for iNOS in Leydig cells. Also, nine of these 14 cases showed immunoreactivity for iNOS in Sertoli cells, yet there were variations in the intensity and the number of Sertoli cells stained. Sections incubated with IgG in the absence of primary antibody were completely negative. NADPH-d staining was mainly localized to vascular endothelium and the staining intensity was higher in the testis with varicocele (Fig. 3E,F). There was weak NADPH-d activity in interstitial Leydig cells and Sertoli cells (Fig. 3E). There was no NADPH-d activity in the negative control sections.

image

Figure 3. A, a section immunostained for eNOS (× 400). Vascular endothelium was strongly positive for eNOS (arrowhead). Immunoreactivity of eNOS was weakly positive around the nucleus of Leydig cells (arrow). B, a section immunostained for nNOS (× 200). Leydig cells, but not all, were positive for nNOS (arrowhead). Immunoreactivity of nNOS was weakly positive in the cytosol of Sertoli cells (arrow). There was no iNOS-positive cells in control testis (C, ×400). Some specimens from testes with varicocele showed immunoreactivity for iNOS in Leydig cells (arrowhead) and Sertoli cells (arrow) (D, ×400). Two sections stained for NADPH-d activity from a control (E, ×200) and varicocele testis (F, ×400). Arrowheads indicate an intense staining in vascular endothelial cells. There was weak granular staining in Sertoli cells (arrow) and Leydig cells (dashed arrow).

Download figure to PowerPoint

The correlations between nitrite concentration, expression of NOS isoforms and several clinicopathological variables are shown in Table 2. The increase in nitrite concentration strongly correlated with maximum and total vein diameter (P < 0.01). The expression of iNOS was significantly correlated with Johnsen score count and total vein diameter (P < 0.05). There was no significant correlation between eNOS or nNOS expression and these variables. To examine the effect of nitrite concentration in combination with age, the patients were subdivided into two groups (Fig. 4). In patients aged >35 years, there was a significant inverse correlation between nitrite concentration and TM (P < 0.05). Also, the maximum and total vein diameter significantly correlated with the decline in TM in patients aged >35 years (P < 0.05).

Table 2.  The correlations between nitrite concentration in ITF, expressions of NOS in testis and clinicopathological variables
VariableSpearman’s rank correlation coefficient (P)
NitriteeNOSnNOSiNOS
  1. Differences were not significant unless shown otherwise.

Age0.1070.1710.0990.116
Serum:
 LH0.0880.1980.1680.201
 FSH0.0810.2800.2370.014
 testosterone0.0820.0190.0530.041
Testicular volume0.1670.1070.2040.282
TM0.2030.0880.3150.017
Johnsen score0.5420.3020.3530.381 (<0.05)
Vein diameter:
 maximum0.449 (<0.01)0.2360.0590.311
 total0.571 (<0.01)0.2320.1280.393 (<0.05)
image

Figure 4. The correlations between nitrite concentration and TM. Patients were subdivided into those aged <35 years (A) and ≥35 years (B).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. CONFLICT OF INTEREST
  9. REFERENCES

In the testis, NO regulates various functions, including inhibition of Leydig cell steroidogenesis [17], germ cell apoptosis [5], Sertoli cell tight-junction dynamics [7] and regulation of testicular blood flow [18]. In the present study, there were strong correlations between nitrite concentration and internal spermatic vein number and diameters (Table 2). iNOS is considered to be a major source of NO because of a parallel between iNOS expression and nitrite concentration. eNOS in testicular endothelial cells was also shown to be a source of NO because of intense NADPH-d activity. These observations imply that locally synthesized NO might be involved in the control of vascular tone, and is drained into internal spermatic veins, after their dilatation. There were no apparent correlations between nitrite concentration (determined by the Griess method using ITF; Table 2) and either spermatogenic variables or hormonal data. Also, there was no correlation between nitrite concentration and generation of 4-HNE-modified proteins (Table 1), indicating that NO is not involved in oxidative stress and has no direct role in the disturbance of spermatogenesis and endocrine function in testes with varicocele. Given that nitrite concentration correlated with TM in patients aged >35 years, but not in younger patients (Fig. 4), NO might be considered to have some time-dependent harmful effects on spermatogenesis.

Previous reports showed elevation of NO in dilated spermatic veins [8–10]. In agreement with the results reported by Ozbek et al. [9], showing a 1.9-fold increase in NO levels in blood taken from the internal spermatic veins of men with varicocele (determined using the Griess method), the nitrite concentration in the ITF of varicocele testis was 1.9 times greater than that of the controls (Fig. 1). Unlike testosterone [14], NO diffuses freely into the veins and is drained from testis, resulting in an almost identical concentration in the spermatic veins [9].

Although changes in nitrite concentration and iNOS expression were comparable (Figs 1 and 2, Table 1), only iNOS correlated with the deterioration in spermatogenic variables (Johnsen’s score, Table 2). Conditions inducing iNOS expression, e.g. abundant interferons, tumour necrosis factors and lipopolysaccharides [7], but not NO itself, might be considered to impair spermatogenesis. Another possible explanation of this discrepancy is that the nitrite concentration obtained from ITF represents total testicular conditions, while biopsy specimens give local information in the testis. Unlike in ischaemia-reperfusion, iNOS expression gave only a 1.8-fold increase in nitrite concentration; this NO concentration is not sufficient to cause cytotoxic effects [7]. Interactions with other molecules, such as superoxide anions and peroxynitrite anions, might have a causative role in amplifying NO toxicity.

The co-location of eNOS protein and NADPH-d activity in vascular endothelial cells also supports the idea that eNOS also participates in the production of NO. Leydig, Sertoli and endothelial cells of human testis show strong NADPH-d activity [4,5]. In normal rat testis, NADPH-d activity was reported in endothelial cells, but not in Leydig or Sertoli cells [19]. Also, myofibroblasts of the peritubular lamina propria of human were shown to be NO-producing cells [6]. We expected a strong NADPH-d activity in Leydig cells because iNOS is mainly expressed in the Leydig cells (Fig. 2C) [13], but there was no intense NADPH-d activity in Leydig cells in the testes with varicocele. In agreement with the results reported by Lissbrant et al. [18] and Burnett et al. [19], NADPH-d activity was mainly localized to endothelial cells, as these are the only cells that express intense NADPH-d activity (Fig. 3F). NADPH-d activity is correlated with the level of NOS activity [16], while the degree of NOS expression does not always correlate with enzyme activity. In the varicocele testis, there could be some changes in the regulation of eNOS activity without enhancing its expression. The mechanisms for activating this NO signalling pathway are under investigation.

The strong correlations between nitrite concentration and the internal spermatic vein diameters (Table 2) are important when considering the pathophysiology of varicocele. Basically, NO is a strong vasodilator, determining vasodilatation and contributing to blood stasis, resulting in the generation and progression of varicocele. Local hypoxia caused by stagnation of blood in the microcirculatory vessels within testis was reported [20]. Increased NOS activity in varicocele testes is thought to result from a compensatory mechanism against hypoxia by increasing testicular blood flow [18]. However, we could not exclude the involvement of, e.g. the retrograde flow of adrenal and renal metabolites, elevation of scrotal temperature, etc., which have been considered as the pathophysiology of varicocele.

Although, there are no reports of direct cytotoxic effects of NO on spermatogenesis, NO might have a causative role in abnormal spermatogenesis through enlargement of the varicocele. Effects of NOS inhibition on spermatic vein diameter and spermatogenesis, using a rat model, are under investigation, and NOS inhibition might be one treatment option to prevent further development of varicocele.

ACKNOWLEDGEMENT

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. CONFLICT OF INTEREST
  9. REFERENCES

We thank Drs Hiroshi Takihara (Onoda City Hospital) and Yoriaki Kamiryo (Saiseikai Shimonoseki General Hospital) for providing the opportunity to collect ITF and testicular biopsy samples.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGEMENT
  8. CONFLICT OF INTEREST
  9. REFERENCES