Evidence for the involvement of blood flow-related mechanisms in the ovulatory process of the rat

Ulf Zackrisson1,4, Masato Mikuni1, Matthew C. Peterson3, Bengt Nilsson2, Per-Olof Janson1 and Mats Brännström1

1 Department of Obstetrics and Gynecology and 2 Department of Surgery, Göteborg University, Sweden


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
To elucidate whether any relationship exists between ovarian blood flow and ovulation rate, the effects on these parameters were examined in equine chorionic gonadotrophin/human chorionic gonadotrophin (eCG/HCG) (15I U/15I U) primed rats after bilateral ligation and severance of either the ovarian branch of the uterine artery and vein (UL), the ovarian artery and vein (OL) or both sites (UL+OL) in comparison to sham operations. Laser Doppler flowmetry demonstrated the presence of microcirculatory vasomotion and a reduction of blood flow after UL, OL and UL+OL performed during the intervals 0–3 h (78, 66 and 19% of pretreatment values respectively) and 6–9 h (68, 57 and 20%) after HCG. Experiments utilizing radioactive microspheres also indicated decreased ovarian blood flow by UL and OL. Ovulation rate was assessed 20 h after HCG in animals where ligations had been performed at 0, 3, 6 and 9 h after HCG. No ovulations were seen after UL+OL and significantly decreased ovulation rates (~50% of sham operated animals) were seen after UL at 0 and 3 h and after OL at 0, 6 and 9 h. Progesterone concentrations in blood 20 h after HCG were reduced by OL but not UL and ovarian weights were unaffected by ligation. It is concluded that acute blood flow reduction during the ovulatory interval reduces ovulation rate in the rat.

Key words: blood flow/laser Doppler/ovary/ovulation/rat


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In rodents and humans, the ovarian blood supply is derived from two major sites. The ovarian artery, which branches from the abdominal aorta or renal artery, and the ovarian branch of the uterine artery anastomose before entry at the hilus of the ovary. The arteries then give rise to tortuous arteries terminating in a capillary network in the theca interna, which surrounds the avascular granulosa cells of the ovarian follicles (Basset, 1943).

In parallel with well described structural changes during the ovulatory process, involving the action of several proteases (Brännström and Janson, 1991Go), dramatic blood flow alterations occur in the ovary. Increased blood flow has been suggested to be of importance for follicular development, rupture, and corpus luteum formation (Janson, 1975Go). The pre-ovulatory luteinizing (LH) surge is known to induce ovarian production or release of several mediators, such as eicosanoids (LeMaire, 1980), histamine (Schmidt, 1988), neuropeptides (Kannisto, 1992) and nitric oxide (Zackrisson et al., 1996Go), all of which may have effects on the vascular system of the pre-ovulatory follicle. It is unknown to what extent the ovulation-promoting actions of these mediators act directly on the extracellular matrix of the follicle wall or via indirect vascular mechanisms.

A number of different techniques, such as venous drainage measurements (Piacsek and Huth, 1971Go; Janson and Albrecht, 1975Go), corrosion cast combined with electron microscopy (Kitai et al., 1985Go), indirect measurements using radioactive isotopes, such as ovarian content of radio-iodinated serum albumin (Ellis, 1961Go), 133Xe clearance (Janson and Jansson, 1997Go), indicator fractionation using 86Ru (Janson and Albrecht, 1975Go) or radioactive microspheres (Janson, 1975Go), thermocouple methods (Makinoda et al., 1988Go), and analysis of haemoglobin content in the ovary (Tanaka et al., 1989Go) have been utilized in attempts to study cyclic changes of ovarian blood flow. These methods have shown various limitations since they are indirect and give only a single recording of blood flow. Thus, these methods may not be optimal for measuring the short and long term variations in blood flow, which undoubtedly occur during the ovulatory process. Recently, it has been shown that laser Doppler flowmetry can be used for recordings of ovarian blood flow in the rat (Brännström et al., 1997Go). With this method and intravital microscopy the existence of regular short term variations of ovarian blood flow, termed vasomotion, were demonstrated. The phenomenon of vasomotion has previously been described in the testis (Damber et al., 1982Go), where laser Doppler flowmetry was used to study blood flow in a longitudinal manner.

Intravital microscopy studies of ovulation in vitro (Löfman et al., 1989Go) and in vivo (Brännström et al., 1997Go) indicate that a sustained high blood flow to the follicle base in combination with increased vascular permeability is of importance during the stages of the ovulatory process immediately preceding the follicular rupture. These vascular changes are postulated to prevent the follicle from collapsing and to ensure the extrusion of the oocyte–cumulus complex from the interior of the follicle. Further support for the importance of high ovarian blood flow is provided by a study utilizing the in-vitro perfused rat ovary model, where a nitric oxide inhibitor decreased the perfusate flow and subsequently suppressed the ovulation rate (Bonello et al., 1996Go). This effect was replicated in perfusions with an intentional reduction of the perfusion pressure, resulting in a decreased number of ovulations. Therefore, a continuous high perfusion pressure appears to be critical to successful ovulation.

It appears that only one study has attempted to investigate a possible causal relationship of blood flow changes and the rate of ovulation in vivo. A reduced ovulation rate in rats was seen when examined more than 10 days after ligation of the uterine artery at different levels (Peppler, 1972Go). The author suggested that blood supply to the ovary via the uterine artery is important for a complete ovulatory response. The results may not necessarily be due to a change in the blood flow, since a possible reduction of the ovarian blood flow can be reversed within this time frame by the remaining ovarian artery (Blasco et al., 1973Go). Furthermore, since counter-current mechanisms between the uterine vein and the ovarian artery have been shown to exist in the human (Bendz, 1977Go) and to affect ovarian function in the sheep (McCracken et al., 1971Go), it is possible that vascular ligations may affect ovulation by similar mechanisms. The aims of the present study were to quantify acute blood flow changes, at the pre-ovulatory stage, after ligation of one or both of the two major arteries supplying the ovary and to determine the effect of these ligations on ovulation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Chemicals and hormones
Equine chorionic gonadotrophin (eCG) and hyaluronidase were purchased from Sigma Chemical Company (St Louis, MO, USA); human chorionic gonadotrophin (HCG) from Serono (Rome, Italy); ketamine from Parke–Davis (Barcelona, Spain); xylazine from Bayer (Leverkusen, Germany).

Animals
The experiments were approved by the animal ethics committee of Göteborg University and were carried out according to the principles and procedures outlined in the NIH Guiding Principles for the Care and Use of Research Animals. Female immature Sprague–Dawley rats were obtained from B-K (Stockholm, Sweden). The animals were kept under controlled conditions with lights on from 05.00 to 19.00 and they had free access to water and pelleted food. At 09.00 on day 28 of age, the rats were given 15 IU eCG s.c. to induce maturation of a first generation of pre-ovulatory follicles. Forty-eight hours later the rats were injected with 15 IU HCG s.c. to induce ovulatory changes with follicular rupture, which was estimated to occur 10–15 h later (Tanaka et al., 1989Go).

Histology and ovarian weights
Initial experiments were performed with bilateral sham operation of ovarian artery/vein (OS) or bilateral ligation and severance of the ovarian artery/vein (OL) at 3 h after HCG and the ovaries were examined about 17 h later with ovulation rate assessed by counting the number of rupture sites as previously described (Bonello et al., 1996Go) and then also by histology. The number of rupture sites, defined as reddish, non-transparent bulging structures, was counted prior to paraffin embedding. Ovaries used for histology were first placed for 24 h in paraformaldehyde (4%), and then embedded in paraffin followed by staining with haematoxylin and eosin. Under a dissecting microscope, the ovaries were examined morphologically by two different persons blind to experimental procedures. Some ovaries were excised 3 h after OL and OS and were weighed after blotting (immediately after excision) and drying (room temperature in a dust-free chamber for 4 weeks).

Experimental design to study ovulation rate
The rats were divided into treatment groups: bilateral sham operation of ovarian artery/vein (OS), bilateral ligation and severance of ovarian artery/vein (OL), bilateral sham operation of ovarian branch of uterine artery/vein (US), bilateral ligation and severance of the ovarian branch of uterine artery (UL), bilateral sham of both the ovarian artery and the ovarian branch of the uterine artery (OS+US), and bilateral ligation and severance of both the ovarian artery/vein and the ovarian branch of the uterine artery (OL+UL). Bilateral operations were performed because preliminary experiments with these animals showed a large variation in ovulation number beween the right and left ovary, with a fairly constant number of ovulations per rat. Each experimental day the same number of rats from age, weight and litter matched rats were allocated to respective sham and ligated groups to minimize variations. Operations were performed immediately prior to (0 h), 3 h after, 6 h after, or 9 h after HCG injection.

The rats were anaesthetized with i.p. injection of a combination of ketamine and xylazine (40 and 7 mg/kg body weight). Prior to laparotomy the fur was carefully washed with 70% ethanol. Paramedial incisions (~20 mm in length) were then performed bilaterally over the ovaries. The ovaries were identified and care was taken not to directly manipulate the ovary, since this procedure has previously been shown to increase ovarian blood flow in the rabbit (Janson 1988Go). The peritoneum was gently opened and the adipose tissues surrounding either the ovarian artery/vein (OS, OL) or ovarian branch of the uterine artery/vein (US, UL) were removed close to their entry into the ovary by gentle traction with dissection forceps. Due to the small size of the arteries supplying the ovary, it was not possible to ligate these vessels separately from the paired veins. Two Autosuture titanium premium surgiclips s-9.0 (Autosuture, Sollentuna, Sweden) were placed to ligate the vessels, after which the vessels were severed with dissection scissors (Figure 1Go). An identical procedure, except for placement of clips and severance, was performed with the sham operated animals. The peritoneum and the skin were then separately sutured with Vicryl 4–0 sutures. After surgery, the rats were kept in separate cages. Twenty hours after HCG injection the animals were anaesthetized a second time, as previously described, and the chest was opened to sample blood by cardiac puncture. Serum was stored at –20°C for later analysis of progesterone concentrations. Ovaries and oviducts were then removed and placed in phosphate-buffered saline (PBS). The oviducts were separated from the ovaries and the ampullary region was identified. The oocyte/cumulus cell masses were identified within the oviduct which was then incised to allow extrusion of the oocyte–cumulus complexes. The adhesive cell mass, containing oocytes and cumulus cells, was placed for 3 min in a drop of PBS containing hyaluronidase (60 IU/ml) to separate the cumulus cells from the oocytes. The oocytes were then counted under a microscope. The observers counting the oocytes were blind to the specific experimental procedures.



View larger version (16K):
[in this window]
[in a new window]
 
Figure 1. Schematic drawing indicating location of procedures for OL (ligation of ovarian vessels) and UL (ligation of ovarian branch of the uterine vessels).

 
Laser Doppler flowmetry
Animals were anaesthetized as described above, which allowed continuous recordings of blood flow up to 60 min in these sets of separate experiments. The carotid artery was exposed and cannulated with a PE 25 polyethylene cannula enabling continuous recording of the systemic blood pressure on a Grass polygraph (Grass Instruments, Quincy, MA, USA) throughout the procedure. Laparotomy incisions were performed paramedially on each side of the abdomen, to allow measurements from both ovaries. To stabilize the ovary, a ligature was gently tied to the periovarian adipose tissue away from the hilus and gentle traction was accomplished by a clamp. A Periflux 4001 Master flowmeter connected to a miniprobe probe (407; fibre separation = 0.25 mm) with an adhesive miniholder (Perimed, Järfälla, Sweden) supported by a micromanipulator, was gently placed on the ovarian surface at a location devoid of larger blood vessels. The skin of the incision was then gently repositioned onto the top of the adhesive miniholder to avoid fluid loss during measurements (Figure 2Go). The signal was continuously recorded by the Perisoft computer program (Perimed, Järfälla, Sweden), as an arbitrary unit (Pfu = perfusion unit) for blood flow. Thus, Pfu is proportional to the number and velocity of moving blood cells in a tissue volume. The mean blood flow over a set time was calculated by measuring the area under the curve. The methodology has been used extensively for estimations of relative changes in blood flow in several organs such as liver (Arvidsson et al., 1988Go) and kidney (Stern et al., 1979Go). Minimum and maximum flow were recorded per time unit. After a stable signal was recorded for at least 5 min one or both of the blood vessels supplying the ovary were ligated (UL, OL or OL+UL) as described above. After this manipulation, the blood flow was recorded until stable blood flow for >5 min (~20 min) had been achieved. The percentage changes from the value before, to immediately after, and during stable conditions after ligation, were calculated. The measurements were performed during two different time periods in relation to HCG (0–3 and 6–9 h post HCG) and animals were randomly allocated to either OL or UL or OL+UL with measurements performed on both ovaries.




View larger version (30K):
[in this window]
[in a new window]
 
Figure 2. Schematic drawing illustrating (upper) locations of the paramedian incisions with the adhesive miniholder/miniprobe and (lower) the adhesive miniholder/miniprobe placed on the ovarian surface.

 
Radioactive microsphere technique
Radioactive 46Sc-labelled microspheres with a diameter of 15.5 ± 0.1 µm, t1/2 of 83.8 days and an energy peak of 889 K eV were purchased from Dupont (Brussels, Belgium). The isotope has a distinct energy maximum which was detected by gamma counting for 3 min at a window (15–1200 K eV) set to cover the energy peak of the isotope. The radioactivity of the samples was expressed as counts per minute.

The experiments were carried out for OL and OS (0 h) as described above and the animals were allowed to recover from this surgery for 3 h after which animals were once more anaesthetized by i.p. injection of a combination of ketamine and xylazine (40 and 7 mg/kg body weight respectively). The right common carotid artery was exposed and cannulated with a PE 25 polyethylene cannula followed by injection of 1 ml of the microsphere suspension over 1 min. To obtain a homogeneous mixture of microspheres, the suspension was mechanically stirred during the injection. Before the intracardiac injection, the microspheres were sonicated for 5 min and then diluted with saline to a concentration of 40 000 to 60 000 spheres per ml to obtain an adequate number of spheres in the tissue, to allow reliable calculation. This was based on previous investigations demonstrating that >384 microspheres must be present in a tissue to permit blood flow to be measured at a 95% confidence level with a precision of 10% (Buckberg et al., 1971Go).

Initial experiments (n = 8) revealed that the right ventricle of the heart was frequently perforated by the carotid artery cannulae. In subsequent experiments, the tip of the cannula was withdrawn at a standardized length from the mandibula. The animals were killed through an air embolus 2 min after the microsphere injection and both ovaries and kidneys were dissected out, weighed, and then counted in a gamma counter. A reference sample was made each experimental day by adhering microspheres to a greased millimetre-squared paper after which the number of adhered spheres was counted under a microscope. The radioactivity of the reference sample was measured, allowing calculation of the exact number of entrapped spheres in the analysed tissue. The weight of the animals varied from 71 to 98 g. Due to the small sizes of the animals and their vessels it was not possible to cannulate the femoral artery to obtain a blood-withdrawal reference sample. Instead, ovarian blood flow was calculated in relation to the kidney blood flow, since it has been shown in rabbits that blood flow to the kidneys is unaffected by LH injection (Janson 1975Go). Ovarian blood flow index was calculated according to the formula:


Although both ovaries were pooled for each animal, a high number of the ovarian pairs contained <384 entrapped spheres. Additionally, the distribution of microspheres between the right and left ovary frequently showed uneven distribution, indicating an incomplete mixture of microspheres with the blood. These methodological difficulties may be caused by variations in the exact positioning of the injection cannula, leading to uneven distribution of spheres in peripheral blood and thus also reflected in the tissue.

Progesterone assay
Twenty hours after HCG injection, serum was sampled after heart puncture during anaesthesia and stored at –20°C until further analysis. Concentrations of progesterone in serum were measured with a Delfia kit (Wallac, Turku, Finland) with an intra-assay variation of <6%. The Delfia kit had been validated and found suitable for the measurements of progesterone in rat serum by serial dilution experiments and in measurements of progesterone concentrations in serum of ovariectomized rats which had been spiked with known concentrations of progesterone. No modifications had to be made.

Statistics
Results are presented as mean % ± SEM and mean ± SEM. For laser Doppler flowmetry experiments, blood flow is presented as the percentage of the mean flow before ligation. Estimations of ovarian blood flow in microsphere experiments are related to kidney blood flow, and are presented as the percentage in relation to kidney flow of the same animals. Statistical differences for laser Doppler flowmetry experiments were calculated by the Friedman test followed by Wilcoxon signed-rank test. The Kruskal–Wallis test followed by the Mann–Whitney U-test were used for comparison of sham groups and ligated groups. These distribution-free tests were used despite the knowledge that blood flow data and the data for number of ovulations were normally distributed. P < 0.05 was considered significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Ovarian blood flow
Initial recordings of ovarian blood flow by laser Doppler flowmetry in animals with no surgical manipulation except laparotomy (n = 5) demonstrated blood flow of >200 arbitrary units, compared to lower flow values found in other organs (liver 90–100, small intestine 110–130, kidney 140–170).

The blood pressure remained stable throughout the experiments (mean arterial pressure 90–95 mmHg). The blood flow demonstrated rhythmical short-term variations (vasomotion) in all preparations, with a frequency between 7 and 12 per min and an amplitude of 27.0 ± 10.8% of the mean blood flow value (Figure 3Go Top). The vasomotion activity remained with unchanged frequency and amplitude after OL or UL. Ligation of the ovarian branch of the uterine artery (UL) performed during the interval 0–3 h after HCG caused a decrease of blood flow with mean values of 83.9 ± 5.4% of initial values which occurred within 15–20 s after ligation (Figure 3Go Middle, Table IGo). A similar decrease was seen after OL (Figure 3Go Lower, Table IGo). Ovarian blood flow continued to decrease until stable values at lower levels were reached 15–20 min after treatment (Table IGo). When UL or OL were performed at later stages of the ovulatory phase (6–9 h after HCG), the reduction was somewhat more pronounced (Table IGo). In preparations with both UL and OL (0–3 h), blood flow was immediately reduced to a stable value of ~1/5 of the value prior to ligation (Table IGo). The same pattern was observed after UL+OL 6–9 h after HCG (Table IGo). In all but one of UL+OL vasomotion was absent.



View larger version (25K):
[in this window]
[in a new window]
 
Figure 3. Representative recordings of ovarian blood flow, by laser Doppler flowmetry, presented as arbitrary perfusion units (Pfu) on the y-axis. Top panel: blood flow prior to ligation, with regular variations of flow within the ovarian tissue. Middle panel: reduction of blood flow after UL (arrow). Lower panel: reduction of blood flow after OL (arrow). Vasomotion activity can be observed before and after ligation. For definition of UL and OL see legend to Figure 1Go.

 

View this table:
[in this window]
[in a new window]
 
Table I. Blood flow measured by laser Doppler flowmetry with values presented as the percentage of the mean ± SEM, compared to values before ligation
 
The radioactive microsphere technique expressed ovarian blood flow related to kidney blood. Ovarian blood flow 3 h after HCG was significantly (P < 0.05) decreased after OL (0 h; n = 7; 73.1 ± 12.4%) compared to OS (0 h; n = 7; 129.4 ± 21.7%). There was a slight, but insignificant, decrease after UL (0 h; n = 6; 95.3 ± 13.0%) compared to US (n = 4; 113.5 ± 7.1%).

Histology and ovarian weights
In initial experiments a reduction of the number of visible corpora lutea per rat was found when OL (3 h; n = 7) and OS (3 h; n = 7) were compared 20 h after HCG injection (17.5 ± 3.6 and 37.0 ± 2.5, respectively). Microscopic examination indicated an increased number of entrapped oocytes in ovaries of the ligated groups (Figure 4Go). There were no other obvious morphological differencies between sham operated and ligated ovaries. The wet weights for OL (3 h; n = 7) and OS (3 h; n = 7) were similar (30.1 ± 0.5 and 30.7 ± 0.7 mg respectively). Dry weights were ~1/5 of the wet weight and did not show any difference between OL and OS (5.4 ± 0.9 and 5.4 ± 1.2 mg respectively).




View larger version (266K):
[in this window]
[in a new window]
 
Figure 4. Rat ovarian tissue after OL (3 h) or OS (3 h), examined 20 h after human chorionic gonadotrophin (haematoxylin/eosin stain). (A) The ovary contains unruptured follicles with entrapped oocytes after OL. (B) Several newly formed corpora lutea are seen in the ovary after OS. OL = ligation and severance of ovarian vessels; OS = sham operation of ovarian vessels. Scale bars = 200 µm.

 
Ovulation rate
The mean number of ovulations (assessed by counting oocytes in the oviduct) per rat in sham operated animals were in the same range at all time points in relation to HCG (Table IIGo). When OL and OS groups were compared, a significantly lower ovulation rate was seen in OL at all time points except at 3 h (Figure 5Go). Ligations of the ovarian branches of the uterine artery (UL) resulted in reduction of ovulation rate when the ligations were performed up to 3 h (Figure 6Go). Ovulation rate was unaffected when UL was performed at 6 and 9 h. In experiments where both UL and OL were performed at 0 h (n = 6) and 9 h (n = 4) no ovulation occurred and the mean number of ovulations in US+OS was within the range of OS and US at 0 and 9 h.


View this table:
[in this window]
[in a new window]
 
Table II. Number of ovulations (mean ± SEM) in sham operation of ovarian vessels (OS) and sham operation of ovarian branch of the uterine vessel (US)
 


View larger version (34K):
[in this window]
[in a new window]
 
Figure 5. Influence on ovulation rate by OL. Numbers of animals are shown in each bar. Significantly decreased ovulation rates compared to OS were found after OL at 0, 6 and 9 h (**P < 0.01). For definition of OS and OL see legend to Figure 4Go. HCG = human chorionic gonadotrophin.

 


View larger version (38K):
[in this window]
[in a new window]
 
Figure 6. Influence on ovulation rate by UL. Numbers of animals are shown in each bar. Significantly reduced ovulation compared to sham operation was found after UL at 0 and 3 h (*P < 0.05, **P < 0.01). UL = ligation and severance of the ovarian branch of the uterine artery; US = sham operation of ovarian branch of uterine vessel. HCG = human chorionic gonadotrophin.

 
Progesterone concentrations
Progesterone concentration in serum 20 h after HCG injection was decreased in all OL groups, although OL 3 and 9 h did not reach significant levels (Table IIIGo). UL did not affect progesterone concentrations (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table III. Progesterone in serum (mean ± SEM) 20 h after human chorionic gonadotrophin (HCG) administration with values presented as the percentage of the mean concentrations of sham operated rats (absolute values (nmol/l) and number of observations (n) given in parentheses
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
During ovulation, distinct haemodynamic and structural changes occur in the ovary. An increase in intrafollicular blood flow (Campbell et al., 1993Go), increased capillary permeability (Okuda et al., 1983Go) and breakdown of the extracellular matrix at the apex of the follicle (Bjersing and Cajander, 1974Go) are major components of the ovulatory process. Ovarian blood flow is comparatively high and increases severalfold during ovulation (Blasco et al., 1975Go; Janson 1975Go; Tanaka et al., 1989Go) with intravital microscopy studies indicating that a sustained high follicle blood flow is of importance during the ovulatory process (Löfman et al., 1989Go; Brännström et al., 1997Go). Recent studies indicate that follicular blood flow in the human, as assessed by transvaginal colour Doppler ultrasonography, can be used to evaluate the oxygenation of the interior of the follicle (Van Blerkom et al., 1997Go) and to predict the outcome for IVF (Chui et al., 1997Go) suggesting a possible role for ovarian blood flow not only for the process of ovulation, but also for implantation and maintainance of pregnancy.

Observations of blood flow with laser Doppler flowmetry showed a rapid and persistent reduction after both OL and UL independent of the stage of the ovulatory process. The blood flow stabilized within 20 min after the procedures at values ~57–78% of the initial blood flow. The anaesthesia only allowed the recording for a time up to ~1 h; we cannot, from these experiments, determine whether these reductions in blood flow remained during the entire ovulatory process. However, the experiments utilizing radioactive microspheres pointed towards a decreased blood flow for at least 3 h after ligation.

The blood flow reduction after OL was slightly more pronounced than after UL and this difference was observed both in the early and late phase of the ovulatory process. Although the difference was not significant, the findings of this study contradict earlier studies that have suggested the uterine artery to be the major contributor of blood flow to the rat ovary (Del Campo and Ginther, 1972Go). In a recent study, using venous outflow measurements, the contribution of ovarian blood in the non-pregnant cycling rat was found to be equal between ovarian and uterine arteries with an increase in uterine contribution during pregnancy (Massa and Bruce, 1997Go).

A low blood flow of ~1/5 of the normal value was observed despite the ligation of both major arterial supplies to the ovary. This demonstrates that there is a small, but significant, collateral circulation in the ovary, where a major part may come through the suspensory ligamental tissue.

In all experiments with laser Doppler flowmetry, regular short term variations of blood flow (vasomotion) were observed with a frequency of 7–12 per min. The functional role of these rhythmical blood flow changes is not clear but it has been speculated that they may facilitate transcapillary fluid exchange since increased flow would promote fluid efflux to the interstitial space, and decreased flow would promote fluid influx to the capillaries (Intaglietta and Gross, 1982Go). This mechanism may be of importance to enhance the transcapillary and interstitial exchange of various hormones and cytokines between the different cell types of the ovary. The fact that vasomotion activity was not recorded after UL+OL procedure, demonstrates that the registration of short term variations are due to an intra-ovarian regulation of blood flow rather than mechanical influence on the laser Doppler probe by direct transmission of uterine/oviductal contractions or intestinal motions.

The significant reduction in ovulation rate when UL or OL were performed during the early ovulatory phase indicates that this stage of the process may be most sensitive to haemodynamic disturbances. It is possible that the ligations at 0 h would reduce the delivery of HCG to the pre-ovulatory follicles and thereby reduce ovulation rate. During later stages of the ovulatory process, the reductions of the ovulation rate in OL were more pronounced compared to UL, indicating an increasing role with time for blood flow through the ovarian artery, in facilitating follicular rupture. The ovarian artery may have a greater capability to compensate for a relative reduction of blood flow, since there was no reduction of the ovulation rate in UL at 6 and 9 h. Another explanation for decreased ovulation rate after OL during late ovulatory phase is based upon the assumption that the hydrostatic pressure gradient over the blood/follicle barrier is required for and also generates the driving force for final follicular rupture and extrusion of the oocyte (Löfman et al., 1989Go). Ligation of the ovarian artery late in the ovulatory process may reduce blood flow and thereby intrafollicular pressure below a value where it is not able to break the partly degraded follicle apex.

Of interest are reports (Peppler 1972Go, 1975Go) of a reduced ovulation rate in the rat, when hysterectomy or ligation of the uterine artery was performed. In these studies, the blood flow was not analysed and the ovulation rate was not measured until 11–16 days after surgery, which theoretically could allow the remaining artery to compensate for a reduction of blood flow (Blasco et al., 1973Go). In a recently published study (Massa and Bruce, 1997Go), ovarian blood flow was examined by venous effluent technique in the rat during dioestrus and pregnancy. In the non-pregnant rat, there was no major difference between the two major ovarian arteries regarding vascular resistance and contribution of ovarian blood flow. However, these variables of circulation tended to be more pronounced for the ovarian artery than for the uterine artery. In the same study, ligation of the individual vessels also indicated that blood flow reduction was larger when ligation was performed on the ovarian artery compared to the uterine artery in line with findings of the present study. Moreover, the blood flow reductions after ligations were in the same range as those observed in our laser Doppler flowmetry experiments during the early ovulatory phase.

Possible effects of the ligations on ovarian nerve supply must also be taken into consideration. The ovarian nerves reach the ovary via two routes. The major portion of the nerve supply forms a plexus parallel to and around the ovarian vascular pedicle and a smaller portion of the ovarian nerve supply reaches the ovary via the ovarian suspensory ligament (Burden and Lawrence, 1978Go; Burden et al., 1983Go). The nerves enter the ovary at the hilus and some are terminated in the theka layer of the follicle (Owman and Sjöberg, 1966Go). The ovarian nerves have been suggested to be important for follicular contractility and vascular regulation (Owman et al., 1979Go) and several transmitters may be functional in the ovulatory process and regulation of ovarian blood flow (Kannisto et al., 1992Go). During OL procedure, it is likely that the ovarian nerves are damaged and this may directly influence ovulation. It was recently demonstrated that intrabursal administration of a noradrenergic neurotoxin reduces the ovulation rate (Goldman et al., 1996Go). It has also been shown that an {alpha}-adrenoreceptor blocker prevents ovulation in the rabbit when injected into the aorta at a site just above the branching of the ovarian arteries (Virutamasen, 1971). On the other hand, transection of the suspenory ligament (Selstam et al., 1985Go) or complete sympathetic denervation in rats due to freezing of the ovarian vascular pedicle and suspensory ligament (Wylie et al., 1985Go) do not affect ovulation rates, indicating that the final stages of follicular growth as well as the mechanism of follicular rupture are unaffected by these treatments.

In the present study, the time-dependent gradual decrease of the ovulation rate after OL suggests mechanisms other than denervation. In these cases of acute denervation, equal inhibition of ovulation would be expected without any relation to the time span from HCG to the denervation. If the damaged nerves were important during the entire ovulatory process, the reduced ovulation rate would be expected to be more pronounced when OL was performed during early stages of the ovulatory process. In the experiments in this study, the suspensory ligament was kept intact and the the ovary would not be totally devoid of nerves. Other factors arguing against denervation as an important factor contributing to the decreased ovulation rate are that ovulation is also achieved in the rat ovarian in-vitro perfusion model (Bonello et al., 1996Go). Furthermore, in the present study a reduction of ovulation rate was also observed after severance of the ovarian branches of the uterine arteries. This treatment should not interfere with the ovarian nerve supply since no ovarian nerves are found in this portion of the vessel (Lawrence and Burden, 1980Go).

The close anatomical position of the artery and vein did not allow separate ligations, prompting the placement of clips over both the corresponding ovarian vein and artery as well as the corresponding vein and artery of the ovarian branch of the uterine vessels. Obstruction of postcapillary vessels with one remaining artery intact could cause obstruction of flow through the remaining vein. This may lead to a pathological interstitial oedema. However, no apparent differences in oedema formation or blood congestion were observed in histological sections or in measurements of ovarian weights comparing OL and OS 3 h, when the ovaries were excised 20 h after HCG injection.

Luteal steroidogenesis is positively related to ovarian blood flow in vivo in rabbits (Janson et al., 1981Go) and linearly related to flow of perfusate in vitro in rat ovaries during vascular perfusion (Sogn et al., 1984Go) indicating that an adequate rate of blood flow is important for an optimal ovarian function. This is also indicated by the present study where reduced serum progesterone was seen after OL. However, after UL no significant changes of serum progesterone concentrations were observed, although ovulation rate was inhibited in UL at 0 and 3 h. This indicates that the ovarian artery has a major role in the regulation of steroid output and the luteinization process. Furthermore, ovarian progesterone production may not only be the result of the numbers of formed corpora lutea but also by the absolute blood flow to the ovary, where a larger blood flow reduction after OL would cause decreased progesterone secretion, which would be unaffected by the smaller blood flow reduction after UL.

In conclusion, the present study demonstrates the existence of vasomotion in the ovarian microvasculature and that the ovulation process is sensitive to disturbances in the blood flow to the ovary.


    Acknowledgments
 
We would like to thank Autosuture (Sollentuna, Sweden) for generously providing the surgiclips and Ms Maria Samuelsson for expert secretarial assistance. This work was supported by grants from the Swedish Medical Research Council (#11607 to M.B.) and the research funds of Hjalmar Svensson, Göteborg Medical Society, Swedish Medical Society, Medical Faculty of Göteborg University, and the University of Utah Department of Obstetrics and Gynecology.


    Notes
 
3 Present address: Department of Obstetrics and Gynecology, University of Utah, 50 North Medical Drive, Salt Lake City, Utah 84132, USA

Back

4 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Göteborg University, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden

Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Arvidsson, D., Svensson, H. and Haglund, U. (1988) Laser-Doppler flowmetry for estimating liver blood flow. Am. J. Physiol., 254, G471–476.[Abstract/Free Full Text]

Basett, D.L. (1943) The changes in the vascular pattern of the ovary of the albino rat during the estrous cycle. Am. J. Anat., 73, 251–292.

Bendz, A. (1977) The anatomical basis for a possible countercurrent exchange mechanism in the human adnex. Prostaglandins, 13, 355–362.[Medline]

Bjersing, L. and Cajander, S. (1974) Ovulation and mechanisms of follicle rupture. Light microscopic changes in rabbit ovarian follicles prior to induced ovulation. Cell Tiss. Res., 149, 287–300.[ISI][Medline]

Blasco, L., Wu, C.H., Ficklinger, G.L. et al. (1973) Blood flow to the reproductive organs of the estrous rabbit. Gynecol. Invest., 4, 270–276.[ISI]

Blasco, L., Wu, C.H., Ficklinger, G.L. et al. (1975) Cardiac output and genital blood flow distribution during the preovulatory period in rabbits. Biol. Reprod., 13, 581–586.[ISI][Medline]

Bonello, N., McKie, K. and Jasper, M. (1996) Inhibition of nitric oxide: effects on IL-1ß-enhanced ovulation rate, steroid hormones, and ovarian leukocyte distribution at ovulation in the rat. Biol. Reprod., 54, 436–455.[Abstract]

Brännström, M. and Janson, P.O. (1991) The biochemistry of ovulation. In Hillier, S.G. (ed.), Ovarian Endocrinology. Blackwell, Oxford, pp. 133–166.

Brännström, M., Löfman, C., Mikuni, M. et al. (1997) Morphological and vascular changes during ovulation in vivo in the rat. Biol. Reprod., 56 (Suppl. 1), 178 (abstract).[ISI]

Buckberg, G.D., Luck, J.C., Payne, B. et al. (1971) Some sources of error in measuring regional blood flow with radioactive microspheres. J. Appl. Physiol., 31, 598–604.[Free Full Text]

Burden, H.W. and Lawrence, I.E. Jr (1978) Experimental studies on the acetylcholinesterase-positive nerves in the ovary of the rat. Anat. Rec., 190, 233–242.[ISI][Medline]

Burden, H.W., Leonard, M., Smith, C.P. et al. (1983) The sensory innervation of the ovary: a horseradish peroxidase study in the rat. Anat. Rec., 207, 623–627.[ISI][Medline]

Campbell, S., Bourne, T., Warwestone, J. et al. (1993) Transvaginal colour blood flow imaging of the periovulatory follicle. Fertil. Steril., 60, 433–438.[ISI][Medline]

Chui, D.K.C., Pugh, N.D., Walker, S.M. et al. (1997) Follicular vascularity — the predictive value of transvaginal power Doppler ultrasonography in an in-vitro fertilization programme: a preliminary study. Hum. Reprod., 12, 191–196.[ISI][Medline]

Damber, J.E., Lindahl, O., Selstam, G. et al. (1982) Testicular blood flow measured by laser Doppler flowmeter: acute effects of cathecholamines. Acta Physiol. Scand., 115, 209–215.[ISI][Medline]

Del Campo, C.H. and Ginther, O.J. (1972) Vascular anatomy of the uterus and the ovaries and the unilateral luteolytic effect of the uterus: guinea pigs, rats, hamsters and rabbits. Am. J. Vet. Res., 33, 2561–2578.[ISI][Medline]

Ellis, S. (1961) Bioassay of luteinizing hormone. Endocrinology, 68, 334–340.[ISI]

Goldman, J.M., Stoker, T.E., Coopler, R.L. et al. (1996) Suppression of oocyte release by local administration of the noradrenergic neurotoxin DSP 4. J. Reprod. Fertil., 106, 275–283.[Abstract]

Intaglietta, M. and Gross, J.F. (1982) Vasomotion, tissue fluid flow and the formation of lymph. Int. J. Microc. Clin. Exp., 1, 55–65

Janson, P.O. (1975) Effects of the luteinizing hormone on blood flow in the follicular rabbit ovary, as measured by radioactive microspheres. Acta Endocrinol., 79, 122–133.[ISI][Medline]

Janson, P.O. (1988) Methodological and functional aspects of ovarian blood flow. In Ichinoe, K., Segal, S.J. and Mastroianni, L. (eds), Preservation of Tuboovarian Function in Gynecologic Benign and Malignant Diseases. Raven Press, New York, pp. 163–176.

Janson, P.O. Albrecht, I. (1975) Methodological aspects of blood flow measurements in ovaries containing corpora lutea. J. Appl. Physiol., 38, 288–293.[Abstract/Free Full Text]

Janson, P.O. and Jansson, I. (1997) The acute effect of hysterectomy on ovarian blood flow. Am. J. Obstet. Gynecol., 127, 349–352.

Janson, P.O., Damber, J.-E. and Ahrén, K. (1981) Luteal blood flow and progesterone secretion in pseudopregnant rabbits. J. Reprod. Fertil., 63, 491–497.[Abstract]

Kannisto, P., Jörgensen, J., Liedberg, F. et al. (1992) Neuropeptides and ovulation. In Sjöberg, N.-O., Hamberger, L., Janson, P.O. et al. (eds), Local Regulation of Ovarian Function. Parthenon, London, pp. 183–200.

Kitai, H., Yoshimura, Y., Wright, K.H. et al. (1985) Microvasculature of preovulatory follicles: Comparison of in vivo and in vitro perfused rabbit ovaries following stimulation of ovulation. Am. J. Obstet. Gynecol., 152, 889–895.[ISI][Medline]

Lawrence, I.E. and Burden, H.W. (1980) The origin of the extrinsic adrenergic innervation to the rat ovary. Anat Rec., 196, 51–59.[ISI][Medline]

Le Maire, W.J. (1980) The role of prostaglandins in the mechanism of ovulation. In Tozzini, R.I., Reeves, G., Pineda, R.L. (eds), Endocrine Physiopathology of the Ovary. Elsevier/North-Holland, Amsterdam, pp. 207–217.

Löfman, C.O., Brännström, M., Holmes, P.V. et al. (1989) Ovulation in the isolated perfused rat ovary as documented by intravital microscopy. Steroids, 5, 481–490.

Makinoda, S., Ichinoe, K. and Koyama, T. (1988) Hemodynamic and histologic studies of ovarian blood flow during the preovulatory period. In Ichinoe, K., Segal, S.J. and Mastroianni, L. (eds), Preservation of Tuboovarian Function in Gynecologic Benign and Malignant Diseases. Raven Press, New York, pp. 177–190.

Massa, H.M. and Bruce, N.W. (1997) Direction of blood blow and changes in resistance of major arteries supplying the ovary of the pregnant rat. Biol. Reprod., 57, 43–48.[Abstract]

McCracken, J.A., Baird, D.T. and Goding, J.R. (1971) Factors affecting the secretion of steroids from the transplanted ovary in the sheep. Recent Progr. Horm. Res., 27, 537–582.[Medline]

Okuda, Y., Okamura, H., Kanzaki, H. et al. (1983) Capillary permeability of rabbit ovarian follicles prior to ovulation. J. Anat., 269, 263–269.

Owman, Ch., Sjöberg, N.-O., Wallach, E.E. et al. (1979) Neuromuscular mechansism of ovulation. In Hafez, E.S.E. (ed.), Human Ovulation Mechanisms. Prediction, Detection and Induction. Pergamon Press, Oxford, pp. 57–100.

Owman, Ch. and Sjöberg, N.-O. (1966) Adrenergic nerves in the female genital tract of the rabbit. With remarks on cholinesterase-containing structures. Z. Zellforsch., 74, 182–197.[ISI][Medline]

Peppler, R.D. (1972) Inhibitory effect of hysterectomy on compensatory ovulation in the unilateral ovariectomized rat. Am. J. Anat., 135, 299–303.[ISI][Medline]

Peppler, R.D. (1975) Effect of uterine artery ligation on ovulation in the rat. Anat. Rec., 184, 183–186.[ISI]

Piacsek, B.E. and Huth, J.F. (1971) Changes in ovarian venous blood flow following cannulation: Effects of luteinizing hormone (LH) and antihistamine. Proc. Soc. Exp. Biol. Med., 138, 1022–1024.

Schmidt, G., Owman, C. and Sjöberg, N.-O. (1988) Cellular localization of ovarian histamine, its cyclic variations, and histaminergic effects on ovulation in rat ovary perfused in vitro. J. Reprod. Fertil., 82, 409–417.[Abstract]

Selstam, G., Norjavaara, E., Tegenfelt, T. et al. (1985) Partial denervation of the ovaries by transection of the suspensory ligament does not inhibit ovulation in rats treated with pregnant mare serum gonadotropin. Anat. Rec., 213, 392–395.[ISI][Medline]

Sogn, J., Abrahamsson, G. and Janson, P.O. (1984) Release of cyclic AMP and progesterone from the isolated perfused luteal ovary of the PMSG treated rat. Acta Endocrinol., 106, 265–270.[ISI][Medline]

Stern, M.D, Boen, P.D., Parma, R. et al. (1979) Measurements of renal cortical and medullary blood flow by laser-Doppler spectroscopy in the rat. Am. J. Physiol., 236, F80–87.[Free Full Text]

Tanaka, N., Espey, L.L. and Okamura, H. (1989) Increase in ovarian blood volume in the gonadotropin-primed rat. Biol. Reprod., 40, 762–768.[Abstract]

Van Blerkom, J., Antczak, M. and Schrader, R. (1997) The developmental potential of the human oocyte is related to the dissolved oxygen content of follicular fuid: association with vascular endothelial growth factor levels and perifollicular blood flow characteristics. Hum. Reprod., 12, 1047–1055.[ISI][Medline]

Virutamasen, P., Hickok, R.L. and Wallach, E.E. (1971) Local ovarian effects of cathecholamines on human chorionic gonadotropin induced ovulation in the rabbit. Fertil. Steril., 22, 235–243.[ISI][Medline]

Wylie, S.N., Roche, P.J. and Gibson, W.R. (1985) Ovulation after sympathetic denervation of the rat ovary produced by freezing its nerve supply. J. Reprod. Fertil., 75, 369–373.[Abstract]

Zackrisson, U., Mikuni, M., Wallin, A. et al. (1996) Cell specific localization of nitric oxide synthases in the rat ovary during follicular development, ovulation and luteal formation. Hum. Reprod., 12, 2667–2673.

Submitted on June 1, 1999; accepted on October 15, 1999.