- Email: [email protected]
Nitric oxide inhibition of human sperm motility
Vol. 64, No.2, August 1995
FERTILITY AND STERILITY Copyright
~"
Printed on acid-free paper in U. S. A.
1995 American Society for Reproductive Medicine
Nitric oxide inhibition of human sperm motility
J. Brice Weinberg, M.D.*t:j:§ Elissa Doty, B.S.:j: Joseph Bonaventura, Ph.D.II~ A. F. Haney, M.D.:j: Duke University and Veterans Administration Medical Centers, Durham, North Carolina and Duke University Marine Biomedical Center, Beaufort, North Carolina
Objective: To determine the effect of nitric oxide (NO) on sperm motility in vitro. Design: Normal human sperm separated by centrifugation through a discontinuous Percoll gradient and subsequent swim-up were incubated for up to 24 hours with NO donors, with and without the known NO quencher hemoglobin, as well as with agents that raise intracellular cyclic 3',5'-guanosine monophosphate (cGMP). Sperm respiration was determined by a tetrazolium-formazan spectrophotometric assay. Setting: Andrology laboratory. Main Outcome Measures: Absolute sperm motility and respiration. Results: Sperm incubated with the NO donors 1 mM nitroprusside, 100 to 125 11M 3-morpholinosydnonimine, and 25 to 125 11M pure nitric oxide gas dissolved in buffer were inhibited in motility in a dose-dependent fashion. The inhibition could be reversed by the NO quencher hemoglobin. Agents that raise cellular cGMP (dibutyryl cGMP or 8-bromo-cGMP) did not inhibit motility. Nitric oxide inhibited sperm respiration, as measured by the tetrazolium-formazan assay. Conclusions: Nitric oxide reduces sperm motility, possibly by a mechanism involving inhibition of cellular respiration independent of an elevation of intracellular cGMP. Nitric oxide elaborated in the female or male genital tract in vivo could adversely influence sperm function and fertility. Fertil Steril 1995; 64:408-13 Key Words: Nitric oxide, sperm, macrophage, hemoglobin, infertility
Sperm motility is required for fertilization in vertebrates. In humans, sperm are deposited in the vagina, actively penetrate the cervical mucus, and
Received June 9,1994; revised and accepted February 17,1995. * Supported in part by the Veterans Administration Medical Research Service, Washington, D.C.; National Institutes of Health grant AR39162, Bethesda, Maryland; and the James Swiger Hematology Research Fund, Durham, North Carolina. t Departments of Medicine, Veterans Administration and Duke University Medical Centers. t Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Duke University Medical Center. § Reprint requests: J. Brice Weinberg, M.D., Veterans Administration Medical Center, 508 Fulton Street, Durham, North Carolina 27705 (FAX: 919-286-6891). IIDepartment of Biochemistry, Duke University Marine Biomedical Center. ~Supported in part by National Institutes of Health grant ESO 1908, Bethesda, Maryland. 408
Weinberg et al. Nitric oxide and sperm
progress through the uterine cavity into the distal oviduct where they encounter the oocyte. Sperm require high levels of adenosine triphosphate (ATP) to maintain their motility. Adenosine triphosphate is generated by the glycolytic pathway and the mitochondrial electron transport system (1). Recent studies have noted that nitric oxide (NO) can reduce ATP levels in cells by inhibiting the ATP-generating ability of enzymes in these pathways (e.g., iron-containing proteins of the electron transport system and the tricarboxylic acid cycle and glyceraldehyde phosphate dehydrogenase of the glycolytic pathway [25]). There are several types of cells capable of generating NO in the male and female genital tracts, including mononuclear phagocytes, endothelial cells, smooth muscle cells, and fibroblasts (5, 6). In earlier work, we showed that peritoneal or oviductal macrophages could immobilize and destroy sperm (7,8). In a preliminary communication, TomFertility and Sterility
linson et al. (9) showed that sodium nitroprusside (nitrosoferricyanide) could inhibit sperm motility. It was not clear in that work whether the effect was related to cyanide or whether the inhibition was mediated through effects on guanylyl cyclase or mitochondrial respiratory enzymes. In the work reported here, we show that the inhibition of sperm motility is mediated via NO (not by cyanide-NO delivered as a pure gas in solution or from two different pharmacologic NO donors was active), that purified hemoglobin (a quencher of NO activity) can block this deleterious NO action, and that the inhibitory effect most likely is caused by an inhibition of sperm respiratory function. These results suggest that NO could be a physiological modulator of sperm motility and fertility and, in certain pathologic states accompanied by increased NO production, that NO could cause sperm hypomotility and reduced fertility in humans. Also, inhibitors of NO action (such as hemoglobin) might prove useful in the treatment of infertility in selected clinical circumstances. MATERIALS AND METHODS
Normal human sperm were obtained by a modification of a technique described previously (10). Semen samples from men with proven fertility and with normal semen analyses were collected by masturbation, allowed to liquefy for 30 minutes at 23°C, and then diluted at 1:1 with Ham's F-10 medium supplemented with magnesium sulfate (0.124 gIL; 0.50 ILM), potassium bicarbonate (0.508 gIL; 5.07 ILM), sodium bicarbonate (1.680 gIL; 20.00 ILM) calcium lactate (0.300 gIL; 1.37 ILM), penicillin G (0.5 gIL; 0.13 ILM), streptomycin sulfate (0.050 gIL; 0.03 ILM), and 7.5% pooled heat-inactivated (56°C for 30 min) human umbilical cord serum. This was centrifuged for 30 minutes at 23°C at 326 X g over a discontinuous Percoll gradient (47% and 90%). The sperm pellet was washed twice by centrifugation at 65 X g and overlaid with 2.5 mL of the supplemented medium. Sperm were then allowed to swim-up during incubation at 37°C for 30 minutes, after which the top layer containing the highly motile sperm was removed. These sperm were used in the experiments; they were cultured in the supplemented medium in a humidified chamber with 95% air and 5% CO 2 in 6 mm diameter chambers in 0.2 mL (1 to 4 X 106 sperm per chamber) with the additives as indicated. After the designated times, sperm motility (i.e., sperm motion) was assessed microscopically before and after water immobilization as described before (10, 11). Briefly, an aliquot of medium was placed in a standard clinical hemocytometer, and the total number of nonmotile sperm in the sample was Vol. 64, No.2, August 1995
counted. For the purposes of these experiments, motile sperm were defined as those with any movement, regardless of the character of that movement. A standard volume of water was then added to immobilize all remaining motile sperm, and the counting procedure was repeated to determine the total number of sperm in the sample. The number of motile sperm was calculated by subtracting the number of nonmotile sperm in the initial sample from the total number of sperm in the entire sample. The percent of sperm that remained motile was calculated by dividing the number of motile sperm by the total number of sperm in the aliquot. The percent motility observed in the samples incubated with only the diluent was considered the control value for that time interval. We elected to use this measure of absolute sperm motility as there is no question as to the clinical implications of any observed decrease in percent of sperm with motility. For determination of the effects of NO on sperm cellular respiration, sperm were incubated with NO donors and then assessed for their abilities to reduce the tetrazolium dye, 3-(4,5-dimethylthiazol-2-yl)-5(3 carboxymethoxy-phenyl)-2-H-tetrazolium (MTS) (12, 13). Sperm were cultured with NO, 3-morpholinosydnonimine (SIN-l), or phosphate-buffered saline (PBS) control in media without phenol red for 18 hours. At that time, MTS and phenazine methosulfate were added as described (12). After culture for an additional 2 hours, absorbances of formazan were determined using a plate reader spectrophotometer. Tetrazolium-formazan assays have been used to measure cell viability and proliferation, as well as cell respiratory function (13, 14). Sodium nitroprusside (nitrosoferricyanide) was obtained from Elkins Sinn, Inc. (Cherry Hill, NJ) and the potassium ferricyanide, dibutyryl cyclic guanosine-3'5'-phosphate (cGMP), and 8-bromo-cGMP were obtained from the Sigma Chemical Company (St. Louis, MO). Nitroprusside and SIN-1 were prepared from powder and used immediately. Pure NO gas was obtained from Carolina Specialty Gases (Raleigh, NC). A saturated solution of NO was prepared in PBS (0.1 M phosphate, pH 7.4) by first deoxygenating the PBS by gassing with pure nitrogen gas in a sealed container for 15 minutes followed by gassing with pure NO gas. We estimated the molarity of this solution as 2.5 mM at room temperature. Aliquots of the NO solution were diluted anaerobically directly into culture medium in the wells and used immediately. Chromatographically pure human hemoglobin free of catalase and superoxide dismutase was prepared as described previously (15). The reagents for the MTS assay were obtained from Promega (Madison, WI). Statistical analyses were performed by one-way Weinberg et al.
Nitric oxide and sperm
409
...
100
III
80
E Q) Q.
~
:;::; 0
--- ----0-__
T
- ------0 1
60
E
-...
40
c
----e-
Q)
u
Q)
20
1
--0-- 0
D..
nitroprusside
0 0
4
8 Time
12
16
(hours)
Figure 1 Modulation of sperm motility by nitroprusside. The effect of nitroprusside on sperm motility over time in culture is shown. The difference between the nitroprusside and ~ontrol was significant CP < 0.025).
moglobin (Fig. 2B). The pure gas NO, delivered from a diluted saturated PBS solution, produced a comparable inhibition of sperm motility (Fig. 3). The sperm morphology after incubation with the three different sources of NO or hemoglobin was indistinguishable from that of control sperm by light microscopy in unstained preparations. Nitric oxide is known to activate soluble guanylyl cyclase and elevate levels of cGMP in several cell types (6). To determine whether elevating sperm levels of cGMP could inhibit motility, we cultured sperm with dibutyryl cGMP or 8-bromo-cGMP, two agents that cause elevation of intracellular cGMP. As noted in Figure 4, these two agents did not inhibit sperm motility. Cellular respiration, as measured by the ability to reduce the tetrazolium dye MTS to
2A
100
analysis of variance testing and a P value of <0.05 was considered significant. Each experiment was repeated a minimum of three times with sperm samples from different donors. No disparity in the results of the various experiments was observed between samples and representative experiments are depicted in the figures. The data are displayed as means of triplicate samples ± SD and, because of the small variance in some instances, the SD bars cannot be seen.
...E
Q) Q.
T T . ------------~ .1
80
--- 1.
III
-... ~
60
0
E
40
c
--0-- 0 Hb
Q)
u
20
Q)
--.-
D..
RESULTS
0 0
Sperm isolated by the Percoll-swim-up technique were free of leukocytes, and their motility was approximately 90%; in control cultures this fell to approximately 75% to 80% after incubation for 18 to 24 hours. The addition of 1 mM nitroprusside markedly reduced this motility, with the effect being clearly apparent after 8 to 16 hours of culture (Fig. 1). Nitroprusside liberates NO and cyanide when it is put into solution. Potassium ferricyanide (1 j.l,M), a comparable molecule without the NO, had no inhibitory effect at 20 hours of culture (82.7% ± 4.6%; P > 0.1; mean ± SD) compared with control (88.3% ± 11.5%) in experiments where 1 /LM nitroprusside was suppressive (2.3% ± 4.0%; P < 0.0005). This indicates the inhibitory effect of nitroprusside could not be accounted for by liberated cyanide. The molsidomine derivative SIN-I, another molecule that liberates NO, also potently inhibited sperm motility in a dosedependent fashion, with approximately 100 to 125 /LM causing 50% inhibition (Fig. 2A). This inhibition was blocked by the NO quencher hemoglobin (Fig. 2A and B). The inhibitory effect of hemoglobin was dose dependent, with 50% of the motility inhibition of SIN-1 being blocked by approximately 50 /LM he410
Weinberg et al. Nitric oxide and sperm
200 uM Hb 100
50
SIN-1
(uM)
28
100
...
E Q) Q.
200
150
80
III
.!!
60
+= 0
-... E
40
c
f!
20
- - 0 - SIN-l
Q)
D..
0 0
50
100
200
Hemoglobin (uM)
Figure 2 Modulation of sperm motility by SIN-1 and hemoglobin. CAl, The effect of SIN-1 on sperm motility and the blockage of this effect by 200 J1.M human hemoglobin during 20 hours of coculture is demonstrated. CB), The effect of hemoglobin on the inhibition of sperm motility by 200 J1.M SIN-1 during 20 hours of coculture is shown. In A, SIN-l significantly inhibited sperm motility at doses of 100 and 200 J1.M CP < 0.025), and hemoglobin inhibited these effects CP < 0.01). In B, no significant inhibitory effect ofSIN-l was noted with all doses of hemoglobin. Hb, hemoglobin. Fertility and Sterility
PBS
Nitric oxide (uM) or (equivalent dilution)
Figure 3 Modulation of sperm motility by NO gas. The effect of NO or PBS on sperm motility after 4 or 20 hours of culture is presented. No differences were noted at 4 hours but, at 24 hours, NO at doses of 50 or 250 MM significantly inhibited sperm motility (P < 0.05).
types (5, 6, 16), either by direct action or by interaction with other molecules, such as superoxide with formation of toxic moieties such as peroxynitrite and hydroxyl radical (17, 18). Nitric oxide has been observed to cause smooth muscle relaxation, inhibit platelet aggregation, inhibit cellular proliferation, induce hematopoietic cell differentiation, and cause cytostasis or killing of tumor cells and microbes. The effects on smooth muscle relaxation and platelet aggregation apparently are mediated by activation of the cells' soluble guanylyl cyclase with a resultant increase in cellular cGMP. The growth inhibitory effect of NO is caused by nitrosylation of iron in pro-
...
100
c..
80
-
60
E CD
UJ
formazan, was inhibited significantly (P < 0.05) by 200 ,uM SIN-l (51.0% ± 1.8%) as well as 25 ,uM (70.6% ± 5.2%) and 250 ,uM (20.0% ± 2.8%) of NO compared with the control (100% ± 3.8%).
CD
0
E
-... CD
DISCUSSION
40
c
()
20
CD
Nitric oxide supplied by a variety of donors inhibits human sperm motility in vitro, possibly by inhibiting cellular respiration. The effect is potent, with 80% to 95% inhibition noted at micro molar concentrations of NO. Although the NO quencher hemoglobin alone did not alter sperm motility, hemoglobin blocked the inhibitory effect of NO. The inhibitory effect potentially is mediated by inhibition of sperm cellular respiration. L-arginine is converted to NO endogenously in vivo by the action of the enzyme nitric oxide synthase (5). NO synthase exists as both constitutive and inducible forms. Constitutive NO synthase characteristically is found in neural cells and endothelial cells. It is produced continually and is activated by a variety of signals that raise intracellular levels of calcium, such as bradykinin and acetylcholine. Inducible NO synthase is found in many different cell types, but most characteristically in mononuclear phagocytes, hepatocytes, and smooth muscle cells. Inducible NO synthase is capable of producing much higher levels of NO than is constitutive NO synthase. The activity of this inducible enzyme is regulated transcriptionally, with increases in activity being produced by new messenger RNA transcription and protein synthesis induced by various factors, including endotoxin and certain inflammatory cytokines (4). Nitric oxide affects the function of numerous cell Vol. 64, No.2, August 1995
D.
0 0
50
500
5000
dbcGMP (uM)
...
E
80
CD
c..
UJ
-
60
CD
0
E
-... c
CD
()
40
20
CD
D.
o
o
50
500
8-bromo-cGMP
5000
(uM)
Figure 4 Lack of effect of dibutyryl cGMP or 8-bromo-cGMP on sperm motility. Sperm were cocultured with 8-bromo-cGMP or dibutyryl cGMP for 20 hours and no effect on motility was observed. There were no statistical differences between the groups. Weinberg et a1. Nitric oxide and sperm
411
teins in the mitochondrial electron transport system (complex I and complex II), aconitase, ribonucleotide reductase, and glyceraldehyde phosphate dehydrogenase (2, 5). This nitrosylation results in a depletion of ATP and a consequent inhibition of cellular proliferation. Although sperm cells do not proliferate, they require large amounts of ATP as an energy source to maintain their motility (1). Depletion of sperm ATP by inhibition of mitochondrial electron transport proteins, aconitase, or glyceraldehyde phosphate dehydrogenase should result in a loss of motility as noted here. Our experiments demonstrate that NO may cause an inhibition of sperm cellular respiration. This, coupled with the observation of no alteration of sperm motility by cGMP analogues, suggests that NO likely affects sperm motility by interfering with ATP-generating abilities rather than through cGMP alterations. In earlier work, we and others have demonstrated that certain women with infertility have increased numbers of peritoneal macrophages, with the highest numbers being found among women with endometriosis (7, 8, 19). Also, patients with endometriosis have increased numbers of macro phages in the oviduct, the site of fertilization (8). We previously noted in in vitro studies that normal human sperm are immobilized by and adhere to macro phages and subsequently become engulfed and destroyed (7). Macrophages and smooth muscle cells can express high levels of inducible NO synthase and produce high levels of NO. Thus, NO elaborated by cells (e.g., macrophages or smooth muscle cells) in the uterus or oviduct has the potential to have an adverse effect on fertilization by reducing sperm motility. Macrophages also have been noted in high numbers in the male genital tract, being found in the epididymis, testes, and semen (20). Although most emphasis has been placed on sperm phagocytosis and destruction by the macro phages (spermatophages), it also is possible that macrophage-elaborated NO could adversely influence reproduction. Inflammatory or immune conditions affecting the male genital tract (e.g., prostatitis, infectious orchitis, allergic or granulomatous orchitis, or epididymitis) could cause high levels of NO, which then could inhibit sperm motility. Likewise, some cases of idiopathic sperm hypomotility could be caused by increased levels of genital tract NO. Others have noted that the reactive oxygen species superoxide and hydrogen peroxide can be produced by cells in semen (sperm and leukocytes) and that sperm function may relate to reactive oxygen species-generating abilities (21-23). Depletion of ATP appears to play an important role in the inhibition of sperm motility by reactive oxygen species (24). In a preliminary communication, Tomlinson et al. (9) 412
Weinberg et al. Nitric oxide and sperm
showed that sodium nitroprusside (nitrosoferricyanide) could inhibit sperm motility and that hemoglobin inhibited this effect. They did not demonstrate definitively that the NO (and not cyanide liberated from the nitroprusside) was causing the effect, and they did not investigate the mechanism of inhibition. By using pure NO gas and other NO donors, we demonstrate here that the effect is mediated via NO rather than cyanide. We also show that the inhibitory effect correlates with an inhibition of sperm cellular respiration by NO, independent from the levels of intracellular cGMP. The ability of NO and NO quenchers to modulate sperm motility has therapeutic implications regarding contraception and infertility treatment. Nitric oxide could be administered locally (e.g., intravaginally or in a condom) for use as a contraceptive; however, potential genotoxic effects of NO (25) would necessitate caution in this application. Alternatively, hemoglobin or another NO quencher used locally in the female genital tract (e.g., the vagina, cervix, or uterus) might prove useful to enhance or preserve sperm motility (and possibly to prevent NO-mediated mutagenesis) and thus counter infertility in selected circumstances of NO overproduction. Also, NO quencher agents could be used in vitro to enhance and preserve the motility of sperm during storage (cryopreservation) or in specimens used for assisted reproductive technologies (lUI, IVF, and GIFT) in circumstances where reduced sperm motility contributes to reproductive failure.
Acknowledgment. 3-Morpholinosydnonimine (SIN-I) was a gift from Cassela ( Frankfurt, Germany).
REFERENCES 1. Mitchell JA, Nelson L, Hafez ESE. Motility of spermatozoa.
2.
3.
4.
5. 6.
7.
In: Hafez ESE, editor. Human semen and fertility regulation in men. St. Louis: C.V. Mosby Company, 1976:83-99. Dimmeler S, Lottspeich F, Brline B. Nitric oxide causes ADPribosylation and inhibition of glyceraldehyde-3-phosphate dehydrogenase. J BioI Chern 1992;267:16771-4. Hibbs JB Jr, Taintor RR, Vavrin Z, Granger DL, Drapier J-C, Amber IJ, Lancaster JR Jr. Synthesis of nitric oxide from a terminal guanidino nitrogen atom of L-arginine: a molecular mechanism regulating cellular proliferation that targets intracellular iron. In: Moncada S, Higgs EA, editors. Nitric oxide from L-arginine: a bioregulatory system. New York: Elsevier Science Publishers B.V., 1990:189-223. Moncada S, Higgs-EA. Endogenous nitric oxide: physiology, pathology and clinical relevance. Eur J Clin Invest 1991; 21:361-74. Nathan C. Nitric oxide as a secretory product of mammalian cells. FASEB J 1992;6:3051-64. Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacol Rev 1991;43: 109-42. Muscato JJ, Haney AF, Weinberg JB. Sperm phagocytosis by
Fertility and Sterility
J
8.
9.
10.
11.
12.
13.
14.
15.
16.
human peritoneal macrophages: a possible cause of infertility in endometriosis. Am J Obstet GynecoI1982;144:503-10. Haney AF, Misukonis MA, Weinberg JB. Macrophages and infertility: oviductal macrophages as potential mediators of infertility. Fertil Steril 1983;39:310-5. Tomlinson MJ, East SJ, Barratt CLR, Bolton AE, Cooke ID. Preliminary communication: possible role of reactive nitrogen intermediates in leucocyte-mediated sperm dysfunction. Am J Reprod ImmunoI1992;27:89-92. Haney AF, Hughes SF, Weinberg JB. The lack of effect of tumor necrosis factor-alpha, interleukin-1-alpha and interferon-gamma on human sperm motility in vitro. J Androl 1992; 13:249-53. Yates CA, Thomas C, Kovacs GT, de Kretser DM. Andrology, male factor infertility and IVF. In: Wood C, Trounson A, editors. Clinical in vitro fertilization. 2nd ed. New York: Springer-Verlag, 1989:95-111. CellTiter 96 aqueous non-radioactive cell proliferation assay. In: Promega Technical Bulletin. Vol. 169. Madison: Promega Corporation, 1992:1-7. Buttke TM, McCubrey JA, Owen TC. Use of an aqueous soluble tetrazoliumlformazan assay to measure viability and proliferation of lymphokine-dependent cell lines. J Immunol Methods 1993; 157:233-40. Klostergaard J, Kilbourn RG, Foster WA, Lopez BG. Rapid, quantitative microassay for the monokine respiration inhibitory factor. J Immunol Methods 1987;101:97-108. Bonaventura J, Bonaventura C, Sullivan B, Godette G. Hemoglobin Deer Lodge: consequences of altering the 2,3 diphosphoglycerate binding site. J BioI Chern 1975;250: 9250-5. Magrinat G, Mason SN, Shami PS, WeinbergJB. Nitric oxide modulation of human leukemia cell differentiation and gene expression. Blood 1992;80:1880-4.
Vol. 64, No.2, August 1995
17. Radi R, Beckman JS, Bush KM, Freeman BA. Peroxynitriteinduced membrane lipid peroxidation: the cytotoxic potential of superoxide and nitric oxide. Arch Biochem Biophys 1991;288:481-7. 18. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Nat! Acad Sci USA 1990;87:1620-4. 19. Halme J, Becker S, Hammond MG, Raj S. Pelvic macrophages in normal and infertile women: the role of patent tubes. Am J Obstet Gynecol 1982; 142:890-5. 20. Holstein A-F. Spermatophagy in the seminiferous tubules and excurrent ducts of the testis in rhesus monkey and in man. Andrologica 1978; 10:331-52. 21. Kessopoulou E, Tomlinson MJ, Barratt CLR, Bolton AE, Cooke ID. Origin of reactive oxygen species in human semen: spermatozoa or leucocytes. J Reprod FertiI1992;94:463-70. 22. Aitken RJ, Irvine DS, Wu FC. Prospective analysis of spermoocyte fusion and reactive oxygen species generation as criteria for the diagnosis of infertility. Am J Obstet Gynecol 1991; 164:542-51. 23. Aitken RJ, Buckingham D, West K, Wu FC, Zikopoulos K, Richardson DW. Differential contribution of leucocytes and spermatozoa to the generation of reactive oxygen species in the ejaculates of oligozoospermic patients and fertile donors. J Reprod Fertil1992;94:451-62. 24. de Lamirande E, Gagnon C. Reactive oxygen species and human spermatozoa. II. Depletion of adenosine triphosphate plays an important role in the inhibition of sperm motility. J Androl 1992; 13:379-86. 25. Nguyen T, Brunson D, Crespi CL, Penman BW, Wishnok JS, Tannenbaum SR. DNA damage and mutation in human cells exposed to nitric oxide in vitro. Proc Natl Acad Sci USA 1992; 89:3030-4.
Weinberg et al. Nitric oxide and sperm
413
FERTILITY AND STERILITY Copyright
~"
Printed on acid-free paper in U. S. A.
1995 American Society for Reproductive Medicine
Nitric oxide inhibition of human sperm motility
J. Brice Weinberg, M.D.*t:j:§ Elissa Doty, B.S.:j: Joseph Bonaventura, Ph.D.II~ A. F. Haney, M.D.:j: Duke University and Veterans Administration Medical Centers, Durham, North Carolina and Duke University Marine Biomedical Center, Beaufort, North Carolina
Objective: To determine the effect of nitric oxide (NO) on sperm motility in vitro. Design: Normal human sperm separated by centrifugation through a discontinuous Percoll gradient and subsequent swim-up were incubated for up to 24 hours with NO donors, with and without the known NO quencher hemoglobin, as well as with agents that raise intracellular cyclic 3',5'-guanosine monophosphate (cGMP). Sperm respiration was determined by a tetrazolium-formazan spectrophotometric assay. Setting: Andrology laboratory. Main Outcome Measures: Absolute sperm motility and respiration. Results: Sperm incubated with the NO donors 1 mM nitroprusside, 100 to 125 11M 3-morpholinosydnonimine, and 25 to 125 11M pure nitric oxide gas dissolved in buffer were inhibited in motility in a dose-dependent fashion. The inhibition could be reversed by the NO quencher hemoglobin. Agents that raise cellular cGMP (dibutyryl cGMP or 8-bromo-cGMP) did not inhibit motility. Nitric oxide inhibited sperm respiration, as measured by the tetrazolium-formazan assay. Conclusions: Nitric oxide reduces sperm motility, possibly by a mechanism involving inhibition of cellular respiration independent of an elevation of intracellular cGMP. Nitric oxide elaborated in the female or male genital tract in vivo could adversely influence sperm function and fertility. Fertil Steril 1995; 64:408-13 Key Words: Nitric oxide, sperm, macrophage, hemoglobin, infertility
Sperm motility is required for fertilization in vertebrates. In humans, sperm are deposited in the vagina, actively penetrate the cervical mucus, and
Received June 9,1994; revised and accepted February 17,1995. * Supported in part by the Veterans Administration Medical Research Service, Washington, D.C.; National Institutes of Health grant AR39162, Bethesda, Maryland; and the James Swiger Hematology Research Fund, Durham, North Carolina. t Departments of Medicine, Veterans Administration and Duke University Medical Centers. t Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Duke University Medical Center. § Reprint requests: J. Brice Weinberg, M.D., Veterans Administration Medical Center, 508 Fulton Street, Durham, North Carolina 27705 (FAX: 919-286-6891). IIDepartment of Biochemistry, Duke University Marine Biomedical Center. ~Supported in part by National Institutes of Health grant ESO 1908, Bethesda, Maryland. 408
Weinberg et al. Nitric oxide and sperm
progress through the uterine cavity into the distal oviduct where they encounter the oocyte. Sperm require high levels of adenosine triphosphate (ATP) to maintain their motility. Adenosine triphosphate is generated by the glycolytic pathway and the mitochondrial electron transport system (1). Recent studies have noted that nitric oxide (NO) can reduce ATP levels in cells by inhibiting the ATP-generating ability of enzymes in these pathways (e.g., iron-containing proteins of the electron transport system and the tricarboxylic acid cycle and glyceraldehyde phosphate dehydrogenase of the glycolytic pathway [25]). There are several types of cells capable of generating NO in the male and female genital tracts, including mononuclear phagocytes, endothelial cells, smooth muscle cells, and fibroblasts (5, 6). In earlier work, we showed that peritoneal or oviductal macrophages could immobilize and destroy sperm (7,8). In a preliminary communication, TomFertility and Sterility
linson et al. (9) showed that sodium nitroprusside (nitrosoferricyanide) could inhibit sperm motility. It was not clear in that work whether the effect was related to cyanide or whether the inhibition was mediated through effects on guanylyl cyclase or mitochondrial respiratory enzymes. In the work reported here, we show that the inhibition of sperm motility is mediated via NO (not by cyanide-NO delivered as a pure gas in solution or from two different pharmacologic NO donors was active), that purified hemoglobin (a quencher of NO activity) can block this deleterious NO action, and that the inhibitory effect most likely is caused by an inhibition of sperm respiratory function. These results suggest that NO could be a physiological modulator of sperm motility and fertility and, in certain pathologic states accompanied by increased NO production, that NO could cause sperm hypomotility and reduced fertility in humans. Also, inhibitors of NO action (such as hemoglobin) might prove useful in the treatment of infertility in selected clinical circumstances. MATERIALS AND METHODS
Normal human sperm were obtained by a modification of a technique described previously (10). Semen samples from men with proven fertility and with normal semen analyses were collected by masturbation, allowed to liquefy for 30 minutes at 23°C, and then diluted at 1:1 with Ham's F-10 medium supplemented with magnesium sulfate (0.124 gIL; 0.50 ILM), potassium bicarbonate (0.508 gIL; 5.07 ILM), sodium bicarbonate (1.680 gIL; 20.00 ILM) calcium lactate (0.300 gIL; 1.37 ILM), penicillin G (0.5 gIL; 0.13 ILM), streptomycin sulfate (0.050 gIL; 0.03 ILM), and 7.5% pooled heat-inactivated (56°C for 30 min) human umbilical cord serum. This was centrifuged for 30 minutes at 23°C at 326 X g over a discontinuous Percoll gradient (47% and 90%). The sperm pellet was washed twice by centrifugation at 65 X g and overlaid with 2.5 mL of the supplemented medium. Sperm were then allowed to swim-up during incubation at 37°C for 30 minutes, after which the top layer containing the highly motile sperm was removed. These sperm were used in the experiments; they were cultured in the supplemented medium in a humidified chamber with 95% air and 5% CO 2 in 6 mm diameter chambers in 0.2 mL (1 to 4 X 106 sperm per chamber) with the additives as indicated. After the designated times, sperm motility (i.e., sperm motion) was assessed microscopically before and after water immobilization as described before (10, 11). Briefly, an aliquot of medium was placed in a standard clinical hemocytometer, and the total number of nonmotile sperm in the sample was Vol. 64, No.2, August 1995
counted. For the purposes of these experiments, motile sperm were defined as those with any movement, regardless of the character of that movement. A standard volume of water was then added to immobilize all remaining motile sperm, and the counting procedure was repeated to determine the total number of sperm in the sample. The number of motile sperm was calculated by subtracting the number of nonmotile sperm in the initial sample from the total number of sperm in the entire sample. The percent of sperm that remained motile was calculated by dividing the number of motile sperm by the total number of sperm in the aliquot. The percent motility observed in the samples incubated with only the diluent was considered the control value for that time interval. We elected to use this measure of absolute sperm motility as there is no question as to the clinical implications of any observed decrease in percent of sperm with motility. For determination of the effects of NO on sperm cellular respiration, sperm were incubated with NO donors and then assessed for their abilities to reduce the tetrazolium dye, 3-(4,5-dimethylthiazol-2-yl)-5(3 carboxymethoxy-phenyl)-2-H-tetrazolium (MTS) (12, 13). Sperm were cultured with NO, 3-morpholinosydnonimine (SIN-l), or phosphate-buffered saline (PBS) control in media without phenol red for 18 hours. At that time, MTS and phenazine methosulfate were added as described (12). After culture for an additional 2 hours, absorbances of formazan were determined using a plate reader spectrophotometer. Tetrazolium-formazan assays have been used to measure cell viability and proliferation, as well as cell respiratory function (13, 14). Sodium nitroprusside (nitrosoferricyanide) was obtained from Elkins Sinn, Inc. (Cherry Hill, NJ) and the potassium ferricyanide, dibutyryl cyclic guanosine-3'5'-phosphate (cGMP), and 8-bromo-cGMP were obtained from the Sigma Chemical Company (St. Louis, MO). Nitroprusside and SIN-1 were prepared from powder and used immediately. Pure NO gas was obtained from Carolina Specialty Gases (Raleigh, NC). A saturated solution of NO was prepared in PBS (0.1 M phosphate, pH 7.4) by first deoxygenating the PBS by gassing with pure nitrogen gas in a sealed container for 15 minutes followed by gassing with pure NO gas. We estimated the molarity of this solution as 2.5 mM at room temperature. Aliquots of the NO solution were diluted anaerobically directly into culture medium in the wells and used immediately. Chromatographically pure human hemoglobin free of catalase and superoxide dismutase was prepared as described previously (15). The reagents for the MTS assay were obtained from Promega (Madison, WI). Statistical analyses were performed by one-way Weinberg et al.
Nitric oxide and sperm
409
...
100
III
80
E Q) Q.
~
:;::; 0
--- ----0-__
T
- ------0 1
60
E
-...
40
c
----e-
Q)
u
Q)
20
1
--0-- 0
D..
nitroprusside
0 0
4
8 Time
12
16
(hours)
Figure 1 Modulation of sperm motility by nitroprusside. The effect of nitroprusside on sperm motility over time in culture is shown. The difference between the nitroprusside and ~ontrol was significant CP < 0.025).
moglobin (Fig. 2B). The pure gas NO, delivered from a diluted saturated PBS solution, produced a comparable inhibition of sperm motility (Fig. 3). The sperm morphology after incubation with the three different sources of NO or hemoglobin was indistinguishable from that of control sperm by light microscopy in unstained preparations. Nitric oxide is known to activate soluble guanylyl cyclase and elevate levels of cGMP in several cell types (6). To determine whether elevating sperm levels of cGMP could inhibit motility, we cultured sperm with dibutyryl cGMP or 8-bromo-cGMP, two agents that cause elevation of intracellular cGMP. As noted in Figure 4, these two agents did not inhibit sperm motility. Cellular respiration, as measured by the ability to reduce the tetrazolium dye MTS to
2A
100
analysis of variance testing and a P value of <0.05 was considered significant. Each experiment was repeated a minimum of three times with sperm samples from different donors. No disparity in the results of the various experiments was observed between samples and representative experiments are depicted in the figures. The data are displayed as means of triplicate samples ± SD and, because of the small variance in some instances, the SD bars cannot be seen.
...E
Q) Q.
T T . ------------~ .1
80
--- 1.
III
-... ~
60
0
E
40
c
--0-- 0 Hb
Q)
u
20
Q)
--.-
D..
RESULTS
0 0
Sperm isolated by the Percoll-swim-up technique were free of leukocytes, and their motility was approximately 90%; in control cultures this fell to approximately 75% to 80% after incubation for 18 to 24 hours. The addition of 1 mM nitroprusside markedly reduced this motility, with the effect being clearly apparent after 8 to 16 hours of culture (Fig. 1). Nitroprusside liberates NO and cyanide when it is put into solution. Potassium ferricyanide (1 j.l,M), a comparable molecule without the NO, had no inhibitory effect at 20 hours of culture (82.7% ± 4.6%; P > 0.1; mean ± SD) compared with control (88.3% ± 11.5%) in experiments where 1 /LM nitroprusside was suppressive (2.3% ± 4.0%; P < 0.0005). This indicates the inhibitory effect of nitroprusside could not be accounted for by liberated cyanide. The molsidomine derivative SIN-I, another molecule that liberates NO, also potently inhibited sperm motility in a dosedependent fashion, with approximately 100 to 125 /LM causing 50% inhibition (Fig. 2A). This inhibition was blocked by the NO quencher hemoglobin (Fig. 2A and B). The inhibitory effect of hemoglobin was dose dependent, with 50% of the motility inhibition of SIN-1 being blocked by approximately 50 /LM he410
Weinberg et al. Nitric oxide and sperm
200 uM Hb 100
50
SIN-1
(uM)
28
100
...
E Q) Q.
200
150
80
III
.!!
60
+= 0
-... E
40
c
f!
20
- - 0 - SIN-l
Q)
D..
0 0
50
100
200
Hemoglobin (uM)
Figure 2 Modulation of sperm motility by SIN-1 and hemoglobin. CAl, The effect of SIN-1 on sperm motility and the blockage of this effect by 200 J1.M human hemoglobin during 20 hours of coculture is demonstrated. CB), The effect of hemoglobin on the inhibition of sperm motility by 200 J1.M SIN-1 during 20 hours of coculture is shown. In A, SIN-l significantly inhibited sperm motility at doses of 100 and 200 J1.M CP < 0.025), and hemoglobin inhibited these effects CP < 0.01). In B, no significant inhibitory effect ofSIN-l was noted with all doses of hemoglobin. Hb, hemoglobin. Fertility and Sterility
PBS
Nitric oxide (uM) or (equivalent dilution)
Figure 3 Modulation of sperm motility by NO gas. The effect of NO or PBS on sperm motility after 4 or 20 hours of culture is presented. No differences were noted at 4 hours but, at 24 hours, NO at doses of 50 or 250 MM significantly inhibited sperm motility (P < 0.05).
types (5, 6, 16), either by direct action or by interaction with other molecules, such as superoxide with formation of toxic moieties such as peroxynitrite and hydroxyl radical (17, 18). Nitric oxide has been observed to cause smooth muscle relaxation, inhibit platelet aggregation, inhibit cellular proliferation, induce hematopoietic cell differentiation, and cause cytostasis or killing of tumor cells and microbes. The effects on smooth muscle relaxation and platelet aggregation apparently are mediated by activation of the cells' soluble guanylyl cyclase with a resultant increase in cellular cGMP. The growth inhibitory effect of NO is caused by nitrosylation of iron in pro-
...
100
c..
80
-
60
E CD
UJ
formazan, was inhibited significantly (P < 0.05) by 200 ,uM SIN-l (51.0% ± 1.8%) as well as 25 ,uM (70.6% ± 5.2%) and 250 ,uM (20.0% ± 2.8%) of NO compared with the control (100% ± 3.8%).
CD
0
E
-... CD
DISCUSSION
40
c
()
20
CD
Nitric oxide supplied by a variety of donors inhibits human sperm motility in vitro, possibly by inhibiting cellular respiration. The effect is potent, with 80% to 95% inhibition noted at micro molar concentrations of NO. Although the NO quencher hemoglobin alone did not alter sperm motility, hemoglobin blocked the inhibitory effect of NO. The inhibitory effect potentially is mediated by inhibition of sperm cellular respiration. L-arginine is converted to NO endogenously in vivo by the action of the enzyme nitric oxide synthase (5). NO synthase exists as both constitutive and inducible forms. Constitutive NO synthase characteristically is found in neural cells and endothelial cells. It is produced continually and is activated by a variety of signals that raise intracellular levels of calcium, such as bradykinin and acetylcholine. Inducible NO synthase is found in many different cell types, but most characteristically in mononuclear phagocytes, hepatocytes, and smooth muscle cells. Inducible NO synthase is capable of producing much higher levels of NO than is constitutive NO synthase. The activity of this inducible enzyme is regulated transcriptionally, with increases in activity being produced by new messenger RNA transcription and protein synthesis induced by various factors, including endotoxin and certain inflammatory cytokines (4). Nitric oxide affects the function of numerous cell Vol. 64, No.2, August 1995
D.
0 0
50
500
5000
dbcGMP (uM)
...
E
80
CD
c..
UJ
-
60
CD
0
E
-... c
CD
()
40
20
CD
D.
o
o
50
500
8-bromo-cGMP
5000
(uM)
Figure 4 Lack of effect of dibutyryl cGMP or 8-bromo-cGMP on sperm motility. Sperm were cocultured with 8-bromo-cGMP or dibutyryl cGMP for 20 hours and no effect on motility was observed. There were no statistical differences between the groups. Weinberg et a1. Nitric oxide and sperm
411
teins in the mitochondrial electron transport system (complex I and complex II), aconitase, ribonucleotide reductase, and glyceraldehyde phosphate dehydrogenase (2, 5). This nitrosylation results in a depletion of ATP and a consequent inhibition of cellular proliferation. Although sperm cells do not proliferate, they require large amounts of ATP as an energy source to maintain their motility (1). Depletion of sperm ATP by inhibition of mitochondrial electron transport proteins, aconitase, or glyceraldehyde phosphate dehydrogenase should result in a loss of motility as noted here. Our experiments demonstrate that NO may cause an inhibition of sperm cellular respiration. This, coupled with the observation of no alteration of sperm motility by cGMP analogues, suggests that NO likely affects sperm motility by interfering with ATP-generating abilities rather than through cGMP alterations. In earlier work, we and others have demonstrated that certain women with infertility have increased numbers of peritoneal macrophages, with the highest numbers being found among women with endometriosis (7, 8, 19). Also, patients with endometriosis have increased numbers of macro phages in the oviduct, the site of fertilization (8). We previously noted in in vitro studies that normal human sperm are immobilized by and adhere to macro phages and subsequently become engulfed and destroyed (7). Macrophages and smooth muscle cells can express high levels of inducible NO synthase and produce high levels of NO. Thus, NO elaborated by cells (e.g., macrophages or smooth muscle cells) in the uterus or oviduct has the potential to have an adverse effect on fertilization by reducing sperm motility. Macrophages also have been noted in high numbers in the male genital tract, being found in the epididymis, testes, and semen (20). Although most emphasis has been placed on sperm phagocytosis and destruction by the macro phages (spermatophages), it also is possible that macrophage-elaborated NO could adversely influence reproduction. Inflammatory or immune conditions affecting the male genital tract (e.g., prostatitis, infectious orchitis, allergic or granulomatous orchitis, or epididymitis) could cause high levels of NO, which then could inhibit sperm motility. Likewise, some cases of idiopathic sperm hypomotility could be caused by increased levels of genital tract NO. Others have noted that the reactive oxygen species superoxide and hydrogen peroxide can be produced by cells in semen (sperm and leukocytes) and that sperm function may relate to reactive oxygen species-generating abilities (21-23). Depletion of ATP appears to play an important role in the inhibition of sperm motility by reactive oxygen species (24). In a preliminary communication, Tomlinson et al. (9) 412
Weinberg et al. Nitric oxide and sperm
showed that sodium nitroprusside (nitrosoferricyanide) could inhibit sperm motility and that hemoglobin inhibited this effect. They did not demonstrate definitively that the NO (and not cyanide liberated from the nitroprusside) was causing the effect, and they did not investigate the mechanism of inhibition. By using pure NO gas and other NO donors, we demonstrate here that the effect is mediated via NO rather than cyanide. We also show that the inhibitory effect correlates with an inhibition of sperm cellular respiration by NO, independent from the levels of intracellular cGMP. The ability of NO and NO quenchers to modulate sperm motility has therapeutic implications regarding contraception and infertility treatment. Nitric oxide could be administered locally (e.g., intravaginally or in a condom) for use as a contraceptive; however, potential genotoxic effects of NO (25) would necessitate caution in this application. Alternatively, hemoglobin or another NO quencher used locally in the female genital tract (e.g., the vagina, cervix, or uterus) might prove useful to enhance or preserve sperm motility (and possibly to prevent NO-mediated mutagenesis) and thus counter infertility in selected circumstances of NO overproduction. Also, NO quencher agents could be used in vitro to enhance and preserve the motility of sperm during storage (cryopreservation) or in specimens used for assisted reproductive technologies (lUI, IVF, and GIFT) in circumstances where reduced sperm motility contributes to reproductive failure.
Acknowledgment. 3-Morpholinosydnonimine (SIN-I) was a gift from Cassela ( Frankfurt, Germany).
REFERENCES 1. Mitchell JA, Nelson L, Hafez ESE. Motility of spermatozoa.
2.
3.
4.
5. 6.
7.
In: Hafez ESE, editor. Human semen and fertility regulation in men. St. Louis: C.V. Mosby Company, 1976:83-99. Dimmeler S, Lottspeich F, Brline B. Nitric oxide causes ADPribosylation and inhibition of glyceraldehyde-3-phosphate dehydrogenase. J BioI Chern 1992;267:16771-4. Hibbs JB Jr, Taintor RR, Vavrin Z, Granger DL, Drapier J-C, Amber IJ, Lancaster JR Jr. Synthesis of nitric oxide from a terminal guanidino nitrogen atom of L-arginine: a molecular mechanism regulating cellular proliferation that targets intracellular iron. In: Moncada S, Higgs EA, editors. Nitric oxide from L-arginine: a bioregulatory system. New York: Elsevier Science Publishers B.V., 1990:189-223. Moncada S, Higgs-EA. Endogenous nitric oxide: physiology, pathology and clinical relevance. Eur J Clin Invest 1991; 21:361-74. Nathan C. Nitric oxide as a secretory product of mammalian cells. FASEB J 1992;6:3051-64. Moncada S, Palmer RMJ, Higgs EA. Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacol Rev 1991;43: 109-42. Muscato JJ, Haney AF, Weinberg JB. Sperm phagocytosis by
Fertility and Sterility
J
8.
9.
10.
11.
12.
13.
14.
15.
16.
human peritoneal macrophages: a possible cause of infertility in endometriosis. Am J Obstet GynecoI1982;144:503-10. Haney AF, Misukonis MA, Weinberg JB. Macrophages and infertility: oviductal macrophages as potential mediators of infertility. Fertil Steril 1983;39:310-5. Tomlinson MJ, East SJ, Barratt CLR, Bolton AE, Cooke ID. Preliminary communication: possible role of reactive nitrogen intermediates in leucocyte-mediated sperm dysfunction. Am J Reprod ImmunoI1992;27:89-92. Haney AF, Hughes SF, Weinberg JB. The lack of effect of tumor necrosis factor-alpha, interleukin-1-alpha and interferon-gamma on human sperm motility in vitro. J Androl 1992; 13:249-53. Yates CA, Thomas C, Kovacs GT, de Kretser DM. Andrology, male factor infertility and IVF. In: Wood C, Trounson A, editors. Clinical in vitro fertilization. 2nd ed. New York: Springer-Verlag, 1989:95-111. CellTiter 96 aqueous non-radioactive cell proliferation assay. In: Promega Technical Bulletin. Vol. 169. Madison: Promega Corporation, 1992:1-7. Buttke TM, McCubrey JA, Owen TC. Use of an aqueous soluble tetrazoliumlformazan assay to measure viability and proliferation of lymphokine-dependent cell lines. J Immunol Methods 1993; 157:233-40. Klostergaard J, Kilbourn RG, Foster WA, Lopez BG. Rapid, quantitative microassay for the monokine respiration inhibitory factor. J Immunol Methods 1987;101:97-108. Bonaventura J, Bonaventura C, Sullivan B, Godette G. Hemoglobin Deer Lodge: consequences of altering the 2,3 diphosphoglycerate binding site. J BioI Chern 1975;250: 9250-5. Magrinat G, Mason SN, Shami PS, WeinbergJB. Nitric oxide modulation of human leukemia cell differentiation and gene expression. Blood 1992;80:1880-4.
Vol. 64, No.2, August 1995
17. Radi R, Beckman JS, Bush KM, Freeman BA. Peroxynitriteinduced membrane lipid peroxidation: the cytotoxic potential of superoxide and nitric oxide. Arch Biochem Biophys 1991;288:481-7. 18. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Nat! Acad Sci USA 1990;87:1620-4. 19. Halme J, Becker S, Hammond MG, Raj S. Pelvic macrophages in normal and infertile women: the role of patent tubes. Am J Obstet Gynecol 1982; 142:890-5. 20. Holstein A-F. Spermatophagy in the seminiferous tubules and excurrent ducts of the testis in rhesus monkey and in man. Andrologica 1978; 10:331-52. 21. Kessopoulou E, Tomlinson MJ, Barratt CLR, Bolton AE, Cooke ID. Origin of reactive oxygen species in human semen: spermatozoa or leucocytes. J Reprod FertiI1992;94:463-70. 22. Aitken RJ, Irvine DS, Wu FC. Prospective analysis of spermoocyte fusion and reactive oxygen species generation as criteria for the diagnosis of infertility. Am J Obstet Gynecol 1991; 164:542-51. 23. Aitken RJ, Buckingham D, West K, Wu FC, Zikopoulos K, Richardson DW. Differential contribution of leucocytes and spermatozoa to the generation of reactive oxygen species in the ejaculates of oligozoospermic patients and fertile donors. J Reprod Fertil1992;94:451-62. 24. de Lamirande E, Gagnon C. Reactive oxygen species and human spermatozoa. II. Depletion of adenosine triphosphate plays an important role in the inhibition of sperm motility. J Androl 1992; 13:379-86. 25. Nguyen T, Brunson D, Crespi CL, Penman BW, Wishnok JS, Tannenbaum SR. DNA damage and mutation in human cells exposed to nitric oxide in vitro. Proc Natl Acad Sci USA 1992; 89:3030-4.
Weinberg et al. Nitric oxide and sperm
413