Expression of nitric oxide synthase and effect of substrate manipulation of the nitric oxide pathway in mouse ovarian follicles

Leila M. Mitchell1, C.Richard Kennedy and Geraldine M. Hartshorne2

Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK

1 Present address: Bourn Hall, Bourn, Cambridge CB3 7TR, UK

2 To whom correspondence should be addressed. e-mail: geraldine.hartshorne{at}warwick.ac.uk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Nitric oxide (NO) is a cell messenger with multiple actions in different biological systems, implicated in the control of follicle and oocyte function. NO is formed from L-arginine by isoforms of nitric oxide synthase (NOS) via NG-hydroxy-L-arginine, with L-citrulline as a byproduct. This study aimed to show how modulation of NO by manipulating NOS substrates would affect mouse follicle growth and ovulation in vitro, where vascular effects of NO are attenuated. METHODS: Immunohistochemistry [endothelial (eNOS) and inducible (iNOS)] and in situ hybridization (iNOS) were applied on mouse ovaries. Cultured follicles were also stained for iNOS by immunohistochemistry. For follicles cultured in the presence or absence of L-arginine, the ability of L-citrulline or NG-hydroxy-L-arginine to substitute for L-arginine was assessed in terms of follicle growth and ovulation. RESULTS: iNOS and eNOS were localized in oocytes and theca, with some staining in granulosa. iNOS mRNA occurred predominantly in granulosa and oocyte. Omission of L-arginine significantly reduced follicle survival and ovulation. Partial compensation for L-arginine withdrawal was achieved with L-citrulline and NG-hydroxy-L- arginine. Specific abnormalities of follicle growth were noted. CONCLUSIONS: NOS is present in mouse follicles, and its action is necessary at a local level for normal follicle development in vitro. Reduced growth, persistent basement membranes and reduced ovulation were associated with in vitro disruption of NO.

Key words: mouse follicles/nitric oxide/nitric oxide synthase/oocytes/ovulation


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The smallest known bioactive product of mammalian cells is nitric oxide (NO), a free radical gas with a half-life of <5 s (Nathan, 1992Go). NO, first identified as endothelial-derived relaxing factor (EDRF; Ignarro et al., 1987Go), causes vascular smooth muscle relaxation and mediates normal physiological processes in many systems including the immune, inflammatory and cardiovascular systems (Bredt and Snyder, 1994Go; Xie and Nathan, 1994Go). NO is formed from L-arginine (L-arg) in a reaction catalysed by nitric oxide synthase (NOS). NOS exists as three isoforms, inducible (iNOS), endothelial (eNOS) and neuronal (nNOS) (Palmer et al., 1988Go; Forstermann et al., 1994Go; Nathan and Xie, 1994aGo). NOS oxidizes L-arg to produce L-citrulline (L-cit) and NO via the intermediate NG-hydroxy-L-arginine (NG-OH-L-arg; Stuehr et al., 1991Go). Only these compounds can support the complete reaction (Griffith and Stuehr, 1995Go).

NO is implicated in various aspects of reproductive function, including uterine contractility, penile erection, ovulation and fertilization (Burnett et al., 1996Go; Kuo et al., 2000Go). NO might affect follicle growth and development via a vasodilatory action influencing blood flow to the follicle (Jansen, 1975Go; Bonello et al., 1996Go), or via a direct effect on local cellular interactions, like other reactive oxygen species. For example, NO might be involved in follicle wall breakdown at ovulation (Jablonka-Shariff and Olson, 1997Go) or have effects on apoptotic processes required for tissue remodelling (Chun et al., 1995Go).

The role of free radicals such as NO in the reproductive system is complex, as they may influence cellular processes via a variety of mechanisms and with both positive and negative effects. Free radicals must be closely controlled to avoid unplanned tissue destruction; however, little is known about how this may occur in an integrated system such as the follicle. Transferrin is one factor found in follicles that may suppress the generation of reactive oxygen species. It is often added to culture media as a supplement, usually in combination with insulin and selenium. Although beneficial effects of the combination have been observed in human and bovine follicle cultures (Katska and Rynska, 1998Go; Wright et al., 1999Go), in mouse follicle cultures, insulin was detrimental (Eppig et al., 1998Go). The role of transferrin alone in follicle culture has not been examined previously, so we took this opportunity to perform a pilot study examining the possible interaction of transferrin with the NO free radical pathway in follicle culture.

There is uncertainty as to which NOS isoform(s) may be involved in NO production during follicle development in mammalian ovaries in vivo. In cultured follicles, a functional vascular system is lacking, although some carry-over of endothelial cells resident in the nascent thecal layers may occur. This model offers an opportunity to study any local effects of the NO pathway on follicle growth. If NO is involved, then significant changes in at least one of the NOS isoforms might occur during follicle growth and ovulation, which seems to be the case with iNOS and eNOS, though not with nNOS (Srivastava et al., 1997Go).

We set out to study the location and role of NO in follicle development using an in vitro model of mouse follicle growth and ovulation (Cortvrindt et al., 1996Go; Mitchell et al., 2002Go). In this system, manually dissected individual follicles remain associated with thecal cells, but lack a blood supply and innervation as would be present in vivo. However, the flattened follicular architecture enables clear observations of follicle development, and results in viable oocytes (Cortvrindt et al., 1996Go). We assessed the presence of iNOS and eNOS in unstimulated immature mouse ovaries, and in cultured follicles from the same source, using immunocytochemistry. The NO pathway was then manipulated in cultured mouse follicles to determine if the precursor (L-arg), intermediate (NG-OH-L-arg) and end product (L-cit) of NOS action affected mouse follicle development, survival and ovulation in vitro. A preliminary assessment of the possible interaction of apo-transferrin with the NO system is also presented.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
B6CBF1 mice were kept at the University of Warwick with a light:dark cycle of 12:12 h and food and water ad libitum. The mice were monitored daily and females aged between 24 and 30 days were used in these experiments.

The methods of follicle culture were as described and extensively characterized previously (Mitchell et al., 2002Go). Briefly, small preantral follicles were dissected manually from quartered ovaries using 29 gauge needles. Follicles selected for culture were spherical, had an intact basement membrane and a centrally located oocyte. The average diameter of follicles at the start of culture was ~135 µm. Each mouse yielded ~30–60 follicles. Individual follicles were cultured in microdrops under oil in minimum essential medium (MEM) {alpha} (Gibco, UK) containing 5% fetal calf serum (FCS; Gibco), penicillin (50 µg/ml), streptomycin (50 µg/ml) and FSH (100 mIU/ml; hpMetrodin; Serono, UK). The medium was refreshed every second day. Follicles were considered to have survived when the oocyte was retained, they appeared round in shape, and the follicle cells were not overtly degenerate. These criteria had been previously validated using fluorescent vital dyes (Mitchell et al., 2002Go). hCG (1.5 IU/ml; Profasi; Serono) was added on day 10 and ovulation assessed 16, 24, 40 and 48 h later. The response to hCG, the promptness of ovulation, and the appearance and expansion of cumulus of the ovulated oocyte were used as indicators of follicular health, as previously described (Mitchell et al., 2002Go). Normal ovulation was considered to be ovulation within 16 h after hCG of a spherical translucent oocyte surrounded by mucified cumulus cells. The total ovulation rate was calculated as the number of follicles which ovulated as a proportion of the total number originally cultured. The timely ovulation rate was calculated as the number of follicles ovulating within 16 h of the hCG dose, as a proportion of the total number originally cultured.

To facilitate histological processing, some follicles were cultured individually in 8-well chamber slides (Nunc; Gibco) in 0.5 ml medium without oil. The morphological appearance of follicles in these vessels was similar to that in standard culture.

For immunocytochemistry, isolated ovaries were fixed at room temperature in neutral buffered formalin for 24–72 h, dehydrated in ethanol and cleared in xylene before embedding in paraffin wax. Sections of ~6 µm were mounted on Vectabond-coated slides (Vector Laboratories, UK).

Cultured follicles in chamber slides were fixed in 150 µl of neutral buffered formalin for 16 h at room temperature. The fixative was removed and 10% molten agar at ~65°C was added and left to harden for 30 min at room temperature. The chambers were then removed and the agar block containing the follicle was gently dislodged from the slide before trimming and processing for histology as above. Due to difficulties with cutting and processing sections from these follicles which became flattened in culture, only iNOS immunocytochemistry was attempted.

NOS immunocytochemistry
iNOS detection was performed using one of two primary antibodies: (i) polyclonal rabbit anti-mouse iNOS (PA3-030; Affinity Bioreagents, UK) and (ii) polyclonal rabbit anti-rat iNOS (N52920; Transduction Laboratories, USA). Dilutions of 1/400 were found to be optimal. eNOS was detected using polyclonal rabbit anti-bovine eNOS (Affinity Bioreagents) using an optimal concentration of 1/300. This antibody was reported by the manufacturer to cross-react with mouse eNOS.

Deparaffinized, rehydrated slides were incubated in 1.5% normal blocking serum in phosphate-buffered saline (PBS) for 20 min. Blotted slides were incubated in primary antibody for 1 h at room temperature. Detection was performed using the anti-rabbit IgG Vectastain Elite ABC kit, according to the manufacturer’s instructions, producing a red–brown colour in regions of primary antibody binding. Harris haematoxylin provided a blue counterstain.

For each experiment, the negative control comprised sections incubated without primary antibody. In some experiments, iNOS knockout mouse sections of lung and uterus provided a second negative control. Samples from iNOS knockout mice were kindly provided by Dr Andrew Thomson of Glasgow Royal Infirmary, Scotland.

In situ hybridization for iNOS mRNA
In situ hybridization was carried out using a variation of the methods described by Heidaran et al. (1988Go) and Shih and Kleene (1992Go). A digoxigenin (DIG)–nucleic acid detection system was used according to the manufacturer’s instructions.

In situ hybridization was performed using an anti-sense iNOS RNA probe. The corresponding sense probe was used as a negative control. The RNA probes were produced from a partial cDNA clone for rat iNOS (95% homologous to mouse iNOS), generously provided by Dr Bruce Kone, University of Texas, Houston, USA. The probes produced were ~250–300 base pairs in length and free from contamination (data not shown).

Whole mouse ovaries were fixed immediately after dissection by overnight incubation in 4% paraformaldehyde, pH7, at 4°C. Ovaries were then equilibrated in fresh 0.5 mol/l sucrose in PBS at room temperature for 1 h before placing in 0.85% (w/v) NaCl for 15 min at 4°C. Ovaries were then dehydrated in an ethanol series, cleared in xylene and embedded in paraffin wax for sectioning at 6 µm thickness.

DIG-UTP-labelled probes
The partial cDNA clone of rat iNOS was cloned 5' to 3' into the EcoRI and BamHI sites of pBluescript KS+. The sample was centrifuged for 15 min at 13 000 g, the supernatant was removed and the pellet was washed with 100 µl 75% ethanol. The ethanol was removed and discarded, and remaining ethanol was allowed to evaporate from the DNA pellet at room temperature for 20 min. DNA was resuspended in 10 µl of dH2O and the concentration determined spectrophotometrically.

Restriction enzyme digests were performed according to the manufacturer’s instructions. Plasmid DNA (4 µg) was digested with 1 µl BamHI (Gibco, 10 IU/µl) or 1 µl EcoRI (Gibco, 10 IU/µl) with 1 µl reaction buffer (Gibco) and 4 µl distilled water. This was mixed thoroughly and left for 60 min at 37°C. Linearized DNA was cleaned using QIA Quick Gel Extraction kit (Qiagen) and run in 1xTBE (Tris, borate, EDTA buffer) on a 1% agarose gel with {lambda} standards to determine the amount of DNA present.

Plasmids were then transcribed in a 20 µl reaction which contained ~1 µg DNA, 50 IU of T3 or T7 (Boehringer, UK), 1xtranscription buffer (Boehringer), 10 IU RNase inhibitor (Gibco) and 0.5 mmol/l of DIG-UTP labelled nucleotides (Boehringer). The solution was thoroughly mixed and incubated for 2 h at 37°C. The reaction was stopped at 65°C for 10–15 min and then DNase (RNase-free; Boehringer) was added at 1 IU/µg of DNA and incubated at 37°C for 15 min. This reaction was stopped by the addition of 2 µl of 0.2 mol/l EDTA (BDH, UK) pH 8.0. RNA was precipitated at –70°C for ≥30 min with 0.5 volumes of 10 mol/l ammonium acetate (BDH) and 2.5 volumes of 100% ethanol. RNA was recovered by centrifugation at 3000 g for 20 min at 4°C, washed in 100 µl 70% ethanol and centrifuged again for 5 min. The ethanol was then removed carefully and the pellet resuspended in 20 µl dH2O and placed on ice. One microlitre of RNA preparation was run on a 1% agarose gel (Gibco), together with known amounts of {lambda} DNA and photographed under UV light. The RNA probe was stored at –70°C until needed.

Hybridization and detection of probe
Sections were deparaffinized in xylene and hydrated in an ethanol series, placed into PBS and finally into 2xsaline sodium citrate (SSC). Sections were digested with 10 µg/ml proteinase K (Sigma) in distilled water with 0.1 mol/l Tris–HCl (BDH, pH 8) and 50 mmol/l EDTA for 5, 10 and 30 min at 37°C (Inderdeo et al., 1996Go; Berruti et al., 1998Go). Sections were treated with 0.25% (v/v) acetic anhydride (Sigma) in 0.1 mol/l triethanolamide (BDH) for 10 min and dehydrated in an ethanol series. Sections were then prehybridized in hybridization buffer for 60 min at 40°C. Hybridization buffer comprised 5 ml 50% formamide, 100 µl 1 mol/l Tris–HCl pH 8, 20 µl 0.5 mol/l EDTA, 200 µl 50xDenhart’s solution, 500 µl 10 µg/ml tRNA (Sigma), 4.18 ml dH2O. Dextran sulphate (0.5 g; Sigma) was added to 5 ml hybridization buffer, then 5 ng/µl probe was added to a sufficient volume of hybridization buffer resulting in 20 µl probe solution/slide. This mixture was heated to 80°C, chilled on ice and applied to the sections. A parafilm covering was applied and the slides were heated to 80°C for 10 min and then incubated for 16 h at 40°C in a humidity box. Sections were washed in 2xSSC at room temperature for 30 min and then treated with 100 µg/ml RNase A (Boehringer) in 2xSSC for 60 min at 37°C. The sections were finally washed in 0.5xSSC for 30 min at 50°C. Sections were rinsed in Tris-buffered saline (TBS) and blocking solution applied [8 ml TBS, 0.3 g bovine serum albumin, 10 µl Triton X-100 (BDH), and 2 ml normal sheep serum (Sigma)] for 30 min at room temperature. Sections were rinsed in TBS, then 1/500 dilution of anti-DIG alkaline phosphatase conjugate (Boehringer) in TBS was applied to each section and incubated for 3 h at room temperature. Slides were rinsed twice in TBS and then equilibrated in substrate buffer (18 ml distilled water, 2 ml Tris pH 9, 0.22 g MgCl2 and 0.12 g NaCl) for 5 min at room temperature. Alkaline phosphatase substrate (Boehringer, 10 ml substrate buffer, 80 µg/ml nitro-blue tetrazolium and 175 µg/ml of 5-bromo-4-chloro-3-indolyl-phosphate) was applied to the sections, covered with parafilm and left for 3 days in the dark. Slides were rinsed under running water for 15 min, dehydrated in an ethanol series, cleared in xylene and mounted with DPX.

Control sections were incubated without probe or with sense probe. In addition, a section of lung from an iNOS knockout mouse was incubated with anti-sense probe.

Modulation of the NO pathway
In order to modulate the NO pathway, various concentrations of NOS substrates were applied in vitro. Follicles were cultured in L-arg-free MEM{alpha} (special order, Gibco) with other culture supplements unaltered and using standardized culture and analysis techniques that we have previously validated for this system (Mitchell et al., 2002Go). Each experiment was repeated a minimum of six times with ≥30 follicles per experiment and the results of the experiments collated to produce the results tables. Follicle survival and ovulation rates were compared with control cultures in complete MEM{alpha} (L-arg concentration 600 µmol/l) or L-arg-free medium with L-arg (Sigma) added back at 600 µmol/l. The influence of apo-transferrin (10 µg/ml) on follicle development was assessed in the presence and absence of L-arg.

To determine the relative effects of different intermediates of the NO pathway, L-cit (0, 6, 60, 600 µmol/l; Sigma) or NG-hydroxy-L-arg (Sigma, 0, 6, 60, 600 µmol/l) were added as supplements to L-arg-free MEM{alpha}. Ovulation and survival rates in vitro were compared with follicles cultured with or without L-arg.

Prior to use, L-cit and NG-hydroxy-L-arg were made up as x100 stock concentrations in media lacking L-arg and FCS. These stock reagents were added in volumes of 1% to the culture medium under test. This avoided the possibility of carried-over L-arg affecting the results.

Experiments modulating the NO pathway were carried out in the absence of apo-transferrin.

Statistics
Survival and ovulation rates were assessed using 2x2 contingency tables, with {chi}2-test. P < 0.05 was considered significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Detection of iNOS
In sections of ovary, anti-iNOS antibody PA3-030 produced positive staining in the theca and ooplasm with weak staining in the granulosa of follicles of all sizes from preantral to antral. The second anti-iNOS antibody (N52920) showed a similar distribution with more intense staining (Figure 1a). Negative control sections lacking primary antibody remained unstained (Figure 1b). Sections from an iNOS knockout mouse which were incubated according to the full protocol showed no detectable staining.



View larger version (90K):
[in this window]
[in a new window]
 
Figure 1. Localization of iNOS, iNOS mRNA and eNOS in mouse ovary. Four week old mouse ovary stained with (a) anti-iNOS antibody (N52920, 1/400) detecting iNOS (brown) in oocytes and thecal cells and faint staining in granulosa cells. The germinal vesicle nucleus, visible in one oocyte is negative for iNOS. (b) Negative control. (c and d) Sections of mouse follicles cultured for 10 days. Anti-iNOS antibody (PA3-030, 1/400) shows positive brown staining in the oocytes and in the outer granulosa cells. Although such peripheral staining may be a technical artefact, a few inner granulosa cells are also positive for iNOS. Four week old mouse ovary with in situ hybridization for iNOS mRNA. (e) Anti-sense probe shows positive signal (purple) in the oocyte and surrounding granulosa cells. (f) Anti-sense probe stains positive in the oocyte and granulosa cells and weakly in thecal cells. The nucleus within the oocyte is not stained. (g) Negative control, buffer only. (h) Negative control, sense iNOS mRNA probe. Four week old mouse ovary stained with (i) anti-eNOS antibody (1/300) detecting eNOS (brown) in oocyte, thecal cells and in the granulosa cells of some follicles. Blood vessels were also stained. (j) Negative control. Bar = 100 µm.

 
Mouse follicles embedded after culture for 10 days had flattened structures, so only granulosa cells and sometimes the oocyte were present in horizontal sections. Antibody PA3-030 against iNOS stained the oocyte and some of the outer follicular cells (Figure 1c, d). Negative controls lacking primary antibody remained unstained. The sectioning and handling of these specimens was difficult and so limited data could be obtained.

Detection of iNOS mRNA
A positive signal for iNOS mRNA was detected in granulosa cells and oocytes (Figure 1e). Granulosa cells and oocytes in follicles of various sizes from preantral to antral were stained. There was weak, irregular staining of thecal cells (Figure 1f). No signal was produced in the negative controls; hybridization buffer only (with no probe, Figure 1g), iNOS sense probe (Figure 1h) or using anti-sense probe on sections of liver from an iNOS knockout mouse (not shown).

Detection of eNOS
Positive staining for eNOS was detected in blood vessels, as expected, and also in the oocyte and thecal cells. Staining was also evident in the granulosa cells of some follicles of various sizes from preantral to antral, as shown in Figure 1i. Slight non-specific staining occurred in negative control sections incubated without primary antibody (Figure 1j).

Follicle culture
Effect of absence of L-arg
Figure 2 shows that omission of L-arg significantly reduced follicle survival and ovulation. Follicles cultured in standard MEM{alpha} had a 77% survival rate with 71% ovulating and in the absence of L-arg, 36% survived, 15% ovulated (P < 0.001). When 600 µmol/l L-arg was added back to the L-arg-free medium, the survival rate of 88% was significantly greater than in complete medium (P < 0.05) and medium lacking L-arg (P < 0.001). In medium with L-arg added back, 75% of follicles ovulated which was not different to standard MEM{alpha}, but was significantly greater than in the absence of L-arg (P < 0.001). Eighty-five per cent of follicles that ovulated normally, with L-arg present, released oocytes with cumulus cells attached, compared with 47% of those in L-arg-free medium (P < 0.001).



View larger version (38K):
[in this window]
[in a new window]
 
Figure 2. Effect of L-arginine on mouse follicle survival and ovulation in vitro. Follicles cultured in medium lacking L-arginine (Arg–) had a significantly (P < 0.001) lower survival rate than those cultured in standard medium (Arg+), or Arg– medium to which 600 µmol/l L-arg had been added back. The ovulation rate was similar in Arg+ and medium to which L-arg had been added back; both were significantly greater than in Arg– (P < 0.001).

 
The morphological development of follicles cultured with or without L-arg was markedly different from about day 5 of culture (see Figure 3). By day 3, follicles cultured both with and without L-arg had attached to the bottom of the dish. In complete medium, by day 5 cells had broken out of their normal spherical structure, presumably by breaching of the basement membrane, and were spreading over the dish; antral cavities formed by about day 8 and follicles appeared healthy with translucent cells. Follicles cultured without L-arg developed in a variety of ways from day 5 onwards (see Table I). Some retained an intact structure with antral cavities, while others degenerated. Some follicles in these conditions prematurely released a nude oocyte by day 8.



View larger version (116K):
[in this window]
[in a new window]
 
Figure 3. Follicles cultured with and without L-arginine (all original magnification x100). (a) Day 3, Arg+. Central oocyte, thecal cells attached to dish, follicular cells appear healthy and basement membrane is intact. This follicle ovulated normally in response to hCG on day 11, indicating its healthy status. (b) Day 3, Arg–. No apparent difference from a. This follicle formed an antrum on day 5 with the basement membrane remaining intact until the end of culture. No ovulation in response to hCG. (c) Day 5, Arg+. Cells have breached the basement membrane and migrated, oocyte is central and follicular cells appear healthy. This follicle formed an antrum on day 6 but ovulated late (>16 h post hCG). (d) Day 5, Arg–. Thecal cells attached to the dish, granulosa cells healthy in appearance and an antral cavity has begun. This follicle appears to be developing normally except that the basement membrane is still intact. This follicle retained an intact basement membrane until the end of culture and did not respond to hCG. (e) Day 8, Arg+. Antral cavity evident with centrally located oocyte surrounded by cumulus cells. This follicle ovulated normally in response to hCG. (f) Day 8, Arg–. Antral cavity with basement membrane remaining intact until the end of culture. Granulosa cells became dark and there was no response to hCG. (g) Day 8, Arg–. Follicular cells attached to the dish but dark and degenerate in appearance, oval oocyte, no response to hCG. (h) Day 8, Arg–. Premature release of a nude oocyte by day 8. Bar = 100 µm.

 

View this table:
[in this window]
[in a new window]
 
Table I. Comparison of follicle growth and ovulation in culture conditions where L-arginine (600 µmol/l) and/or apo-transferrin (10 µg/ml) are present
 
Effect of apo-transferrin on follicles cultured with and without L-arg
Table I shows that all follicles cultured without L-arg had low survival and ovulation rates. In the absence of L-arg, follicles cultured with apo-transferrin had a significantly higher survival rate (55%) than those cultured without (27%) (P < 0.001); however, a variety of abnormal features was observed in these follicles, including dark or unhealthy-appearing granulosa cells, which may indicate impending degeneration. The ovulation rates were similarly low at 9 and 14% respectively.

The majority of follicles considered to have survived using our previously established criteria showed features of abnormal development and deterioration (Table I). There was an unexpected and highly significant difference (P < 0.001) in the proportions of follicles in which the granulosa structure, normally contained within the basement membrane, remained intact at the end of culture (68% with apo-transferrin versus 12% without). Interestingly, 3% of follicles cultured without L-arg or apo-transferrin spontaneously released oocytes surrounded by cumulus cells, without exposure to hCG. A total of eight follicles released their oocytes in this way (6 on day 8, 1 on day 9 and 1 on day 10) in five separate experiments, but only when media lacked both L-arg and apo-transferrin. Such oocyte release was visibly indistinguishable from that considered to be ‘ovulation’ in vitro when it occurred in response to hCG.

In contrast to the results when L-arg was absent, follicles cultured in medium containing L-arg but lacking apo-transferrin had significantly higher total (86 versus 75%, P < 0.01) and timely (<16 h) ovulation rates (78 versus 59%, P < 0.001), although survival rates were not significantly different, compared with those containing both L-arg and apo-transferrin (Table I). The majority of follicles cultured with L-arg were healthy in appearance, regardless of whether apo-transferrin was present. Apo-transferrin therefore had a positive effect on follicle survival when L-arg was absent, but a detrimental effect upon ovulation when L-arg was present.

Addition of L-cit to L-arg-free medium
Table II shows the response of follicles to culture in conditions where L-arg was replaced with L-cit at concentrations between 0 and 600 µmol/l, compared with complete medium containing 600 µmol/l L-arg, all in the absence of apo-transferrin. Follicle survival and ovulation rates were significantly lower in the absence of L-arg compared with complete medium (P < 0.001), confirming previous results. The inclusion of progressively greater concentrations of L-cit resulted in improvements in follicle survival and ovulation rates, as well as the appearance of the cumulus at ovulation, although the maximum concentration of 600 µmol/l did not fully compensate for the absence of L-arg.


View this table:
[in this window]
[in a new window]
 
Table II. Comparison of follicle growth and ovulation in culture conditions where L-arginine (600 µmol/l) was replaced with 0–600 µmol/l L-citrulline
 
Follicles cultured without L-arg or L-cit developed in a manner similar to those in control media until day 5 of culture. Follicular cells attached to the bottom of the dish and occasionally a small antral cavity could be seen (Figure 4a, b). After this time, follicles without L-cit or L-arg began to degenerate into dark masses with cells sparsely attached (Figure 4g) and by day 10 were clearly unhealthy (Figure 4l). Follicles cultured with 6 µmol/l L-arg were similar at day 4 (Figure 4c) and by day 8 some had formed small antral cavities remaining similar in appearance (Figure 4h) to the control (Figure 4f). However, by day 10, follicular cells were dark and sparsely attached to the dish without an antral cavity present (Figure 4m). Surviving and non-degenerative follicles cultured in 60 µmol/l L-cit developed in a similar manner to the control follicles until day 10 (Figure 4d, i). By day 10 some follicles in this culture group still appeared healthy (Figure 4n) but others had degenerated with cells more sparsely attached. Follicles cultured in 600 µmol/l L-cit grew similarly to controls (Figure 4e, j, o).



View larger version (160K):
[in this window]
[in a new window]
 
Figure 4. Effect of L-citrulline in replacing L-arginine requirement in cultured mouse follicles (all original magnifications x100). A progressive tendency towards normal development (shown in top row) for longer in vitro is observed with increasing concentrations of L-citrulline. (a) Cells have breached the basement membrane and attached to the dish. This follicle ovulated in response to hCG. (b) Cells have attached to the dish and appear healthy. On day 8, this follicle spontaneously released an oocyte surrounded by dark mucified cumulus cells. (c) Cells attached to the dish. This follicle survived until day 10 and ovulated after hCG addition but was dark in appearance and released a nude oocyte. (d) Follicle appearing healthy on day 4 but only survived until day 7 when it prematurely released a nude oocyte. (e) Cells attached to the dish. This follicle formed an antral cavity by day 8, remained healthy until day 10 and ovulated normally in response to hCG. (f) Follicle with large antral cavity. This follicle ovulated normally in response to hCG. (g) Cells are attached sparsely and a nude oocyte was released prematurely on day 10. (h) Follicle with an antral cavity. This follicle survived but was dark in appearance. It ovulated promptly in response to hCG but dark cumulus cells surrounded the oocyte. (i) Follicle with an antral cavity. This follicle survived until day 10 and ovulated normally. (j) Follicle with a large antral cavity. The cells are healthy in appearance and this follicle ovulated normally. (k) Normally developing follicle with antral cavity, central oocyte surrounded by cumulus. This follicle ovulated normally. (l) Cells are dark and sparsely attached. This follicle ovulated in response to hCG but released a nude oocyte. (m) Cells are attached to the dish but have spread out and an antral cavity has not been formed. Some cells appear dark. This follicle ovulated promptly in response to hCG but the oocyte was devoid of cumulus. (n) Cells are detaching from the dish and appear dark but an antral cavity is present. Despite its appearance, this follicle ovulated normally. (o) Cells are attached to the dish with an antral cavity forming. This follicle ovulated normally. Bar = 100 µm.

 
Addition of NG-hydroxy-L-arg to medium lacking L-arg
Table III shows the effects of supplementing L-arg-free medium with NG-hydroxy-L-arginine on mouse follicle survival and ovulation. Similarly to L-cit supplementation, increasing concentrations of NG-hydroxy-L-arg allowed a dose-dependent partial recovery of follicle survival and ovulation, although full compensation for withdrawal of L-arg was not achieved at equimolar concentrations.


View this table:
[in this window]
[in a new window]
 
Table III. Comparison of follicle growth and ovulation in culture conditions where L-arginine (600 µmol/l) was replaced with 0–600 µmol/l NG-hydroxy-L-arginine
 
The relative effectiveness of L-cit and NG-hydroxy-L-arg in replacing L-arg, in terms of follicle survival, total ovulation and timely ovulation, is presented in Figure 5. The log dose–response curves are different, the concentration resulting in a 50% response being approximately one order of magnitude greater for L-cit than for NG-hydroxy-L-arg.



View larger version (22K):
[in this window]
[in a new window]
 
Figure 5. Log dose–response curves showing the relative efficiency of L-citrulline and NG-hydroxy-L-arginine in supporting survival and ovulation of L-arginine-deprived mouse follicles in culture. {diamondsuit} = L-arginine; {square} = L-citrulline; {Delta} = NG-hydroxy-L-arginine.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The results presented demonstrate the presence of NOS in the ovary, indicating the potential for NO production as an active participant in female follicular development. Manipulation of L-arg is a standard method of affecting NO production, used because direct measurements of NO are technically difficult because of its very short half-life. In this study, manipulating elements of the NO pathway had detrimental effects upon follicle function, measured using sensitive indicators of in vitro development and ovulation efficiency, which were previously characterized in our laboratory (Mitchell et al., 2002Go). These results suggest that NO may affect follicle growth via a direct mechanism even in the absence of systemic influences.

In this study, eNOS protein was detected in oocytes, theca and in the granulosa cells of some follicles in immature mouse ovaries. eNOS has been previously reported in mature oocytes of rats and mice (Nishikimi et al., 2001Go), and its presence in the granulosa cell layer increases during stimulated follicle growth (Jablonka-Shariff and Olson, 1997Go). Others have found eNOS located on the surface of pre-ovulatory and ovulated oocytes and in the theca and stroma of healthy growing follicles, while staining in granulosa cells was punctate (Powers et al., 1996Go; Jablonka-Shariff and Olson, 1998Go).

In sections of ovary and cultured follicles, iNOS protein and mRNA, and eNOS protein were localized most clearly to the ooplasm. The presence of iNOS in immature oocytes in growing and small follicles had not been previously demonstrated, although NOS involvement in oocyte maturation has been inferred from experiments using NOS inhibitors (Sengoku et al., 2001Go). iNOS is known to increase substantially after fertilization in ovulated mouse oocytes and embryos (Nishikimi et al., 2001Go) and its presence is stimulated by hCG in rats (Jablonka-Shariff and Olson, 1997Go).

It is interesting that iNOS protein was located mainly in the oocytes and theca of immature mouse ovaries, while iNOS mRNA was detected in granulosa cells and the oocyte. Inter-communication between the cell types of the follicle may partly explain this finding. Ribonucleotides are transferred into the oocyte via gap junctions with the granulosa cells (Brower and Schulz, 1982Go) and increased oocyte growth and RNA polymerase activity are among the first signs of follicle growth (Gosden et al., 1993Go). Others, working in rats, have shown cycle-dependent variation in mRNA and protein levels. In the immature rat ovary before ovulation, high levels of iNOS mRNA (Van Voorhis et al., 1995Go; Srivastava et al., 1997Go) but low levels of iNOS protein (Zackrisson et al., 1996Go) occur in granulosa cells. After ovulation, iNOS mRNA decreases, and iNOS protein increases (Jablonka-Shariff and Olson, 1997Go; Srivastava et al., 1997Go). Various transcription factors and stimuli, including NO itself, can regulate iNOS gene expression transcriptionally, post-transcriptionally and translationally (Nathan and Xie, 1994bGo; Lysiak et al., 1995Go). Many factors present in follicles have the capacity to influence iNOS mRNA levels, such as interleukin (IL)-4, IL-10 and transforming growth factor-{beta} (Roy and Kole, 1998Go). Therefore, the lack of coincidence of iNOS mRNA and protein detected in our study seems to be in keeping with what is known of iNOS in other situations, and suggests a dissociation between transcription and translation.

The presence of iNOS protein in cultured follicles provided the possibility of studying NO actions on follicle development in vitro by substrate manipulation. Evidence for NO involvement in rodent ovulation arises from many in vivo or perfused ovary studies (e.g. Hesla et al., 1997Go; Mitsube et al., 1999Go). NO has also been found to inhibit steroidogenesis of cultured rat granulosa cells (Ahsan et al., 1997Go; Dave et al., 1997Go); however, no previous studies have examined the role of NO in individual follicle cultures.

The depletion and reintroduction of L-arg is widely used as a method of studying the role of NOS and hence NO on cellular systems in vitro. This assumes that there is no other effect of L-arg withdrawal apart from those mediated by the NO pathway. The literature does not provide any evidence of other mechanisms being involved, but this cannot currently be ruled out as a possible contributory factor in the observations that we present. Information on physiological levels of NOS substrates is available only for endothelial cells. Intracellular L-arg concentrations up to 2 mmol/l may occur in freshly isolated endothelial cells (Hecker et al., 1990bGo), whereas in cultured cells, levels range from 100 to 800 µmol/l (Gold et al., 1989Go; Baydoun et al., 1990Go; Hecker et al., 1990bGo) and 200–400 µmol/l after 24 h of culture without L-arg (Baydoun et al., 1990Go; Hecker et al., 1990aGo). L-cit levels of 150 µmol/l were found in cultured bovine aortic endothelial cells, decreasing to 50 µmol/l after 24 h of L-arg depletion (Baydoun et al., 1990Go). Omission of L-arg led rapidly to a decrease in the total intracellular amino acid pool, but amino acid levels began to normalize within 24 h, except for L-arg and L-cit.

If intracellular L-arg is regulated similarly in cultured follicular cells, then follicles without L-arg could maintain similar development to control follicles by utilizing intracellular L-arg until exhaustion. Follicles cultured without L-arg appeared healthy until day 5, which probably indicates the time taken to reach critically low levels of L-arg. Our dose–response data suggest that L-cit may be recycled into L-arg, if L-arg was lacking, as occurs in endothelial cells and macrophages (Hecker et al., 1990aGo; Nussler et al., 1994Go). Follicles cultured with L-cit may produce L-arg from it via enzymes with activities similar to argininosuccinate lyase and argininosuccinate synthetase. The synthesis of L-arg in this way in low L-arg conditions could play a significant role in regulating NO production (Nussler et al., 1994Go); however, the efficiency of this process is likely to be low unless the prevailing concentrations of L-cit are ≥60 µmol/l, as shown in Figure 5. NG-Hydroxy- L-arg, if present, would be a more effective substrate, but serum levels in rats were reported as 3.7 µmol/l in basal and 15.8 µmol/l in NOS-stimulated conditions (Hecker et al., 1995Go). While follicular levels of NG-hydroxy-L-arg are unknown, the levels used in our experiments are supraphysiological and the availability of NG-hydroxy- L-arg may be limited in the physiological situation.

A curious observation in this study was the apparent maintenance of an intact basement membrane in more than half the follicles cultured in medium where L-arg was lacking but apo-transferrin was supplemented. Transferrin readily enters follicles from the circulation and is active in iron transport, although it may have local actions also (Tilly, 1998Go). For example, transferrin may suppress the generation of reactive oxygen species and may have specific functions in granulosa cells, although little is known of its role in follicle development (Briggs et al., 1999Go). Transferrin is expressed in mouse granulosa cells, where its receptor is also present (Briggs et al., 1999Go). This may be particularly relevant for the follicle culture system, where extrafollicular transferrin is unavailable unless added to culture media. Omitting L-arg has been shown to increase free radicals by formation of superoxide and NO which forms peroxynitrite, causing cellular injury (Xia et al., 1996Go). Transferrin may protect cells by acting as a chelator of contaminating toxic metals, as a competitive substrate for proteolytic enzymes (Orly et al., 1980Go), as well as in reducing oxidative stress from free radicals (Cairo et al., 2002Go). Our observation may therefore provide some support for the idea that basement membrane breakdown involves free radicals (Luck et al., 1995Go). The addition of apo-transferrin would tend to reduce free radicals which may have been increased by the omission of L-arg (Xia et al., 1996Go), possibly contributing to follicular demise. Alternatively, it is possible that transferrin may exert an inhibitory effect upon cell proliferation, reducing the tendency for the basement membrane to rupture under the culture conditions applied. We do not have evidence on the mechanism by which this effect is occurring, but this observation warrants further examination, since the inhibition of ovulation would be potentially useful clinically, and embryos and gametes can be adversely affected by an excess of free radicals (Dinara et al., 2001Go). Experiments in this series will be continued using specific inhibitors of NOS, to assess their effects upon follicle growth in vitro.

In conclusion, these results demonstrate the importance and relevance of L-arg, the precursor of NO, in normal follicle development in vitro. Without L-arg in the culture medium, follicle survival and ovulation were inhibited. The addition of other NO pathway components (NG-hydroxy-L-arginine, L-citrulline and NO) did not fully replace L-arg as survival and ovulation rates did not reach control levels.


    Acknowledgements
 
The authors wish to acknowledge the invaluable contributions of those who assisted with this work. Dr Bruce Kone generously provided a partial cDNA clone of rat iNOS. Dr Andrew Thomson and staff at the Royal Infirmary in Glasgow assisted with immunocytochemistry and provided tissues from NOS knockout mice. Dr Rob Catalano helped with in situ hybridization and the technical staff of the Department of Biological Sciences are warmly thanked for their support. Funding was provided from the Reproductive Medicine Research Fund, University Hospitals Coventry and Warwickshire NHS Trust.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Ahsan S, Lacey M and Whitehead SA (1997) Interactions between interleukin-1{beta}, nitric oxide and prostaglandin E2 in the rat ovary: effects on steroidogenesis. Eur J Endocrinol 137,293–300.[ISI][Medline]

Baydoun AR, Emery PW, Pearson JD and Mann GE (1990) Substrate-dependent regulation of intracellular amino acid concentrations in cultured bovine aortic endothelial cells. Biochem Biophys Res Commun 173,940–948.[ISI][Medline]

Berruti G, Perego L, Borgonovo B and Mantegani E (1998) MSJ-1, a new member of the DNAJ family of proteins, is a male germ cell-specific gene product. Exp Cell Res 239,430–441.[CrossRef][ISI][Medline]

Bonello N, McKie K, Jasper M, Andrew L, Ross N, Braybon E, Brannstrom M and Norman RJ (1996) Inhibition of nitric oxide: effects on interleukin-1{beta}-enhanced ovulation rate, steroid hormones, and ovarian leukocyte distribution at ovulation in the rat. Biol Reprod 54,436–445.[Abstract]

Bredt DS and Snyder SH (1994) Nitric oxide: a physiologic messenger molecule. Annu Rev Biochem 63,175–195.[CrossRef][ISI][Medline]

Briggs DA, Sharp DJ, Miller D and Gosden RG (1999) Transferrin in the developing ovarian follicle: evidence of de-novo expression by granulosa cells. Mol Hum Reprod 5,1107–1114.[Abstract/Free Full Text]

Brower PT and Schulz RM (1982) Intercellular communication between granulosa cells and mouse oocytes: existence and possible nutritional role during oocyte growth. Dev Biol 90,144–153.[ISI][Medline]

Burnett AL, Nelson RJ, Calvin DC, Liu JX, Demas GE, Klein SL, Kriegsfield LJ, Dawson VL, Dawson RM and Snyder SH (1996) Nitric oxide-dependent penile erection in mice lacking neuronal nitric oxide synthase. Mol Med 2,288–296.[ISI][Medline]

Cairo G, Recalcati S, Pietrangelo A and Minotti G (2002) The iron regulatory proteins: targets and modulators of free radical reactions and oxidative damage. Free Radic Biol Med 32,1237–1243.[CrossRef][ISI][Medline]

Chun S-Y, Eisenhauer KM, Kubo M and Hsueh AJW (1995) Interleukin-1{beta} suppresses apoptosis in rat ovarian follicles by increasing nitric oxide production. Endocrinology 136,3120–3127.[Abstract]

Cortvrindt R, Smitz J and Van Steirteghem AC (1996) In-vitro maturation, fertilization and embryo development of immature oocytes from early preantral follicles from prepuberal mice in a simplified culture system. Hum Reprod 11,2656–2666.[Abstract]

Dave S, Farrance DP and Whitehead SA (1997) Evidence that nitric oxide inhibits steroidogenesis in cultured rat granulosa cells. Clin Sci (Lond.) 92,277–284.[Medline]

Dinara S, Sengoku K, Tamate K, Horikawa M and Ishikawa M (2001) Effects of supplementation with free radical scavengers on the survival and fertilization rates of mouse cryopreserved oocytes. Hum Reprod 16,1976–1981.[Abstract/Free Full Text]

Eppig JJ, O’Brien MJ, Pendola FL and Watanabe S (1998) Factors affecting the developmental competence of mouse oocytes grown in vitro: follicle-stimulating hormone and insulin. Biol Reprod 59,1445–1453.[Abstract/Free Full Text]

Forstermann U, Closs EI, Pollock JS, Nakane M, Schwarz P, Gath I and Kleinart H (1994) Nitric oxide synthase isozymes. Characterisation, purification, molecular cloning and functions. Hypertension 23,1121–1131.[Abstract]

Gold ME, Bush PA and Ignarro LJ (1989) Depletion of arterial L-arginine causes reversible tolerance to endothelium-dependent relaxation. Biochem Biophys Res Commun 164,714–721.[ISI][Medline]

Gosden RG, Boland NI, Spears N, Murray AA, Chapman M, Wade JC, Zohdy NI and Brown N (1993) The biology and technology of follicular oocyte development in vitro. Reprod Med Rev 2,129–152.

Griffith OW and Stuehr DJ (1995) Nitric oxide synthases: properties and catalytic mechanism. Annu Rev Physiol 57,707–736.[CrossRef][ISI][Medline]

Hecker M, Sessa WC, Harris HJ, Anggard EE and Vane JR (1990a) The metabolism of L-arginine and its significance for the biosynthesis of endothelium-derived relaxing factor: cultured endothelial cells recycle L-citrulline to L-arginine. Proc Natl Acad Sci USA 87,8612–8616.[Abstract]

Hecker M, Mitchell JA, Harris HJ, Katsura M, Thiemermann C and Vane JR (1990b) Endothelial cells metabolise NG-monomethyl-L-arginine to L-citrulline and subsequently to L-arginine. Biochem Biophys Res Commun 167,1037–1043.[ISI][Medline]

Hecker M, Schott C, Bucher B, Busse R and Stoclet JC (1995) Increase in serum NG-hydroxy-L-arginine in rats treated with bacterial lipopolysaccharide. Eur J Pharmacol 24,275,R1–3.[CrossRef]

Heidaran MA, Showman RM and Kistler WS (1988) A cytochemical study of the transcriptional and translational regulation of nuclear transition protein 1 (TP1), a major chromosomal protein of mammalian spermatid. J Cell Biol 106,1427–1433.[Abstract]

Hesla JS, Preutthipan S, Maguire MP, Chang TSK, Wallach EE and Dharmarajan AM (1997) Nitric oxide modulates human chorionic gonadotrophin-induced ovulation in the rat. Fertil Steril 67,548–552.[ISI][Medline]

Ignarro LJ, Buga GM, Wood KS, Byrns RE and Chaudhuri G (1987) Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci USA 84,9265–9269.[Abstract]

Inderdeo DS, Edwards DR, Han VKM and Khokha R (1996) Temporal and spatial expression of tissue inhibitors of metalloproteinases during the natural ovulatory cycle of the mouse. Biol Reprod 55,498–508.[Abstract]

Jablonka-Shariff A and Olson LM (1997) Hormonal regulation of nitric oxide synthases and their cell-specific expression during follicular development in the rat ovary. Endocrinology 138,460–468.[Abstract/Free Full Text]

Jablonka-Shariff A and Olson LM (1998) The role of nitric oxide in oocyte meiotic maturation and ovulation: meiotic abnormalities of endothelial nitric oxide synthase knock-out mouse oocytes. Endocrinology 139,2944–2954.[Abstract/Free Full Text]

Jansen PO (1975) Effects of luteinising hormone on blood flow in the follicular rabbit ovary, as measured by radioactive microspheres. Acta Endocrinol (Copenh) 79,122–133.[Medline]

Katska L and Rynska B (1998) The isolation and in vitro culture of bovine preantral and early antral follicles of different size classes. Theriogenology 50,213–222.[CrossRef][ISI][Medline]

Kuo RC, Baxter GT, Thompson SH, Stricker SA, Patton C, Bonaventura J and Epel D (2000) NO is necessary and sufficient for egg activation at fertilisation. Nature 406,633–636.[CrossRef][ISI][Medline]

Luck MR, Jeyaseelan I and Scholes RA (1995) Ascorbic acid and fertility. Biol Reprod 52,262–266.[Abstract]

Lysiak JL, Hussaini IM, Webb DJ, Glass WF, Allietta M and Gonias SL (1995) {alpha}2-Macroglobulin functions as a cytokine carrier to induce nitric oxide synthesis and cause nitric oxide-dependent cytoxicity in the RAW 264.7 macrophage cell line. J Biol Chem 270,21919–21927.[Abstract/Free Full Text]

Mitchell LM, Kennedy CR and Hartshorne GM (2002) Effects of varying gonadotrophin dose and timing on antrum formation and ovulation efficiency of mouse follicles in vitro. Hum Reprod 17,1181–1188.[Abstract/Free Full Text]

Mitsube, K, Mikuni, M, Matousek, M and Brannstrom, M (1999) Effects of a nitric oxide donor and nitric oxide synthase inhibitors on luteinising hormone-induced ovulation in the ex-vivo perfused rat ovary. Hum Reprod 14,2537–2543.[Abstract/Free Full Text]

Nathan C (1992) Nitric oxide as a secretory product of mammalian cells. FASEB J 6,3052–3064.

Nathan C and Xie Q-W (1994a) Regulation of biosynthesis of nitric oxide. J Biol Chem 269,13725–13728.[Free Full Text]

Nathan C and Xie Q-W (1994b) Nitric oxide synthases: roles, tolls and controls. Cell 78,915–918.[ISI][Medline]

Nishikimi A, Matsukawa T, Hoshino K, Ikeda S, Kira Y, Sato EF, Inoue M and Yamada M (2001) Localisation of nitric oxide synthase activity in unfertilised oocytes and fertilised embryos during preimplantation development in mice. Reproduction 122,957–963.[Abstract/Free Full Text]

Nussler AK, Billiar TR, Liu Z and Morris SM (1994) Coinduction of nitric oxide synthase and argininosuccinate synthetase in a murine macrophage cell line. J Biol Chem 269,1257–1261.[Abstract/Free Full Text]

Orly J, Sato G and Erickson GF (1980) Serum suppresses the expression of hormonally induced function in cultured granulosa cells. Cell 20,817–827.[ISI][Medline]

Palmer RMJ, Ashton DS and Moncada S (1988) Vascular endothelial cells synthesise nitric oxide from L-arginine. Nature 333,664–666.[CrossRef][ISI][Medline]

Powers RW, Chambers C and Larsen WJ (1996) Diabetes-mediated decreases in ovarian superoxide dismutase are related to blood-follicle barrier and ovulation defects. Endocrinology 137,3101–3110.[Abstract]

Roy SK and Kole AR (1998) Ovarian transforming growth factor-{beta} (TGF-{beta}) receptors: in vitro effects of follicle stimulating hormone, epidermal growth factors and TGF-{beta} on receptor expression in human preantral follicles. Mol Hum Reprod 4,207–214.[Abstract]

Sengoku K, Takuma N, Horikawa M, Tsuchiya K, Komori H, Sharifa D, Tamate K and Ishikawa M (2001) Requirement of nitric oxide for murine oocyte maturation embryo development and trophoblast outgrowth in vitro. Mol Reprod Dev 58,262–268.[CrossRef][ISI][Medline]

Shih, DM and Kleene, KC (1992) A study by in situ hybridisation of the stage and appearance and disappearance of the mRNAs encoding transition protein 2 and the mitochondrial capsule seleno-protein during spermatogenesis in the mouse. Mol Reprod Dev 33,222–227.[ISI][Medline]

Srivastava V, Jones BJ, Dookah H, Hiney JK and Dees LW (1997) Ovarian nitric oxide synthase (NOS) gene expression during peripubertal development. Life Sci 61,1507–1516.[CrossRef][ISI][Medline]

Stuehr DJ, Kwon NS, Nathan CF, Friffiths OW, Feldman PL and Wiseman J (1991) NG-hydroxy-L-arginine is an intermediate in the biosynthesis of nitric oxide from L-arginine. J Biol Chem 266,6259–6263.[Abstract/Free Full Text]

Tilly JL (1998) Molecular and genetic basis of normal and toxicant-induced apoptosis in female germ cells. Toxicol Lett 103,497–501.[CrossRef][ISI]

VanVoorhis BJ, Moore K, Strijbos PJ, Nelson S, Baylis SA, Grzybicki D and Weiner CP (1995) Expression and localisation of inducible and endothelial nitric oxide synthase in the rat ovary. J Clin Invest 96,2719–2726.[ISI][Medline]

Wright CS, Hovatta O, Margara R, Trew G, Winston RM, Franks S and Hardy K (1999) Effects of follicle-stimulating hormone and serum substitution on the in-vitro growth of human ovarian follicles. Hum Reprod 14,1555–1562.[Abstract/Free Full Text]

Xia Y, Dawson VL, Dawson TM, Snyder SH and Zweier JL (1996) Nitric oxide synthase generates superoxide and nitric oxide in arginine-depleted cells leading to peroxynitrite-mediated cellular injury. Proc Natl Acad Sci USA 93,6770–6774.[Abstract/Free Full Text]

Xie Q-W and Nathan C (1994) The high-output nitric oxide pathway: role and regulation. J Leuk Biol 56,576–582.[Abstract]

Zackrisson U, Mikuni M, Wallin A, Delbro D, Hedin L and Brannstrom M (1996) Cell-specific localisation of nitric oxide synthases (NOS) in the rat ovary during follicular development, ovulation and luteal formation. Hum Reprod 11,2667–2673.[Abstract]

Submitted on January 22, 2003; resubmitted on June 6, 2003; accepted on September 24, 2003.