Follicular fluid responds endothermically to aqueous dilution

Martin R. Luck1,4, Stephen Griffiths1, Keith Gregson1, Elaine Watson2, Margaret Nutley3 and Alan Cooper3

1 University of Nottingham, Division of Animal Physiology, School of Biosciences, Sutton Bonington Campus, Loughborough LE12 5RD, 2 University of Edinburgh, Department of Veterinary Clinical Studies, Easter Bush, Midlothian and 3 University of Glasgow, Department of Chemistry, Glasgow G12 8QQ, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Studies suggest that ovarian follicles are cooler than their surrounding tissues. The mechanism of this remarkable phenomenon is unclear. We postulate that endothermic reactions accompany the growth-associated hydration of follicular fluid. METHODS: We performed two types of experiment, using human and animal follicular fluids. In the first, saline (50 µl) was injected into follicular fluid (500 µl) held in an equilibrated incubator, with monitoring of sample temperature. In the second, an adiabatic microcalorimeter recorded thermal shifts after injection of buffer (10 µl) into previously dialysed samples (1.4 ml). The relevance of changes observed was assessed by mathematical modelling. RESULTS: In the incubator study, 9/17 bovine and 6/12 human fluids showed a temperature fall (0.05–0.2°C). Cooling was delayed by up to 2 min but sustained for 7–25 min. Remaining fluids showed no change. In the microcalorimeter, 4/9 human, 4/6 bovine, 5/5 porcine and 1/4 equine samples showed an endothermic response. Remaining samples showed either no response (bovine) or exothermy (human, equine). Pre-concentration of human follicular fluid amplified the endothermy or reversed the exothermy. Modelling indicated that the incubator-type response was of appropriate magnitude to explain follicular hypothermy. CONCLUSION: Follicular fluid responds endothermically to aqueous dilution and may contribute to follicular cooling during growth.

Key words: endothermy/follicle/hydration/microcalorimetry/temperature


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The ovarian follicles of a number of species have been observed to be cooler than both their surrounding ovarian stroma and other abdominal viscera; human: (Grinsted et al., 1985Go); rabbit: (Grinsted et al., 1980Go); bovine: (Greve et al., 1996Go; Grøndahl et al., 1996Go); pig: (Hunter et al., 1995Go); sheep: (Benoit et al., 1976Go). Using laparotomy and infra-red detection, pig and rabbit follicles were found to be as much as 2°C cooler than adjacent tissue (Grinsted et al., 1980Go; Hunter et al., 1997Go). The temperature difference increased with follicle size and depended on the follicles being metabolically active. Further studies using an artificially heated operating theatre and an infra-red laparoscope (to minimize body environment temperature gradients and evaporative heat losses), as well as infra-red endoscopy, have confirmed that the gradient exists at the follicle–stroma level (Hunter et al., 2000Go).

These observations are counter-intuitive: all tissues within an enclosed, thermally-conductive environment such as the abdomen would be expected to be of a similar temperature, with any local variations produced by transiently different rates of metabolism equilibrating rapidly. The existence of significant hypothermy, in particular tissues or regions of tissue, therefore requires a specific mechanistic explanation. Two possible mechanisms are apparent: counter-current heat exchange and biochemically-generated endothermy (Hunter et al., 1997Go). The counter-current hypothesis, though an attractive explanation for the regional heterothermy of the oviduct (David et al., 1972Go; Hunter and Nichol, 1986Go) or for temperature gradients between ovary and other viscera (Einer-Jensen, 1988Go), lacks sufficient physiological evidence when applied to the individual follicle. Fundamentally, it would require the presence of an anatomically discrete heat sink within close range of the ovary, but no such site is apparent. Furthermore, although follicles have an external network of fine capillaries and lymphatics (Andersen, 1926Go), it is difficult to envisage how the peri-follicular microcirculation could selectively maintain a significant local coolant effect on an avascular body of fluid whose volume may be very large (several millilitres in the maturing follicles of human and domestic species).

The aim of the present study was to examine the possibility that follicular hypothermy is biochemically generated. Our working hypothesis was that follicular fluid contains a component(s) which reacts endothermically to hydration. It is envisaged that such a component would be secreted continually within the follicle as part of the growth process, undergoing hydration by water derived from the circulation. We have tested this hypothesis in follicular fluids from four species, using two different laboratory approaches, and observed measurable enthalpic responses to aqueous dilution. We have also attempted to model mathematically the heat balance of a cool follicle so as to establish whether the measured amounts of enthalpy could account for the observed follicular phenomenon.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Experimental material
Human follicular fluid was collected from patients undergoing oocyte recovery during conventional IVF treatment at the NURTURE unit of Queen's Medical Centre, Nottingham, UK. Prior written consent was obtained from each patient. Oocyte recovery took place 36 h after administration of human chorionic gonadotrophin and followed a period of ovarian stimulation by human menopausal gonadotrophin in the presence of pituitary down-regulation by gonadotrophin-releasing hormone (GnRH) agonist. The fluid of the first aspirated follicle, <=17 mm diameter, was obtained from each patient and stored at –20°C. Samples containing red blood cells or contaminated with flushing medium were rejected.

Bovine and porcine fluids were obtained from slaughterhouse material. Ovaries with medium to large healthy follicles but without active corpora lutea were selected according to standard visual criteria (Luck et al., 2000Go). Fluid was aspirated with a syringe and 25 G needle and either used immediately or stored frozen (–20°C). Bovine follicles were treated individually but, for reasons of limited volume, pig samples consisted of the pooled fluids from all large follicles on individual ovaries. Size details are given in Results.

Equine follicular fluid was obtained from two horse mares undergoing oocyte recovery prior to gamete intra-Fallopian tube transfer (GIFT). Follicular fluid was aspirated during oestrus from two pre-ovulatory follicles (29–35 mm diameter) in each mare. The follicles were accessed by flank laporotomy after administration of appropriate sedation and analgesia. The fluid was stored at –20°C.

Incubator studies
A sample of follicular fluid (500 µl) was placed in a central well of a 24 well polystyrene culture plate (Nunc, Denmark), held on the middle shelf of a humidified water-jacketed, fan-circulating laboratory incubator (size one; Gallenkamp, London, UK) maintained at 37 ± 0.3°C. The sample well was covered with a plastic test tube cap into which two holes had been bored. One hole facilitated the positioning of a fine copper-constantan thermistor junction (10 mV/°C; precision ± 0.05°C) within the fluid sample; the thermistor wire passed out of the incubator at the door seal and was attached via an amplifier to a pen recorder. The other hole allowed the suspension of a fine-bore plastic tube (internal radius 0.785 mm) just above the surface of the sample. This tube was filled with saline (0.9% w/v) and connected to a Hamilton syringe held outside the incubator. The length of tube inside the incubator was at least 0.5 m. The sample was stirred continuously by means of a short (~2 mm) home-made glass-covered stirring bar, maintained in motion by a water-driven magnetic stirrer positioned about 5 cm below the sample chamber. (Electrically driven stirrers were found to cause an unacceptable sine wave distortion of the thermistor trace.) The stirrer was driven by a warm (37 ± 1°C) water supply whose pipes entered and exited the incubator through a lagged access point some distance from the sample chamber.

All experiments were preceded by a long (at least 1 h) equilibration period during which both the indicated temperature of the incubator and the record trace produced by the thermistor were stable. Experiments were initiated by briefly opening the outer incubator door (the inner glass door remaining sealed) and causing a 50 µl aliquot of saline to drop into the sample by means of the externally held Hamilton syringe. The outer door was then kept shut for the remainder of the data collection period. Preliminary runs and control experiments (samples of saline instead of follicular fluid) demonstrated that this procedure caused no detectable change in incubator temperature or deviation in the thermistor trace. Prior to each day's experiments, the thermistor trace was calibrated over the 38–35°C range against a mercury thermometer held in a beaker of water. The pen deflection was 22 mm/°C and occurred as a virtually instantaneous response.

Microcalorimeter studies
Absolute heats of dilution were determined at 25°C using sensitive isothermal titration calorimeters [Microcal Omega and VP-ITC systems; Microcal Inc., Northampton, MA, USA; (Wiseman et al. 1989Go). These computer-controlled instruments consist of an adiabatic chamber containing sample and reference cells, the former having a combined micro-injector and stirrer. Programmed aliquots of diluent are injected with constant stirring into the sample at 3 min time intervals. Any temperature difference between the sample and reference cells is constantly monitored and automatically corrected by internal Peltier feedback heaters to the appropriate cell. The enthalpy of the dilution reaction is recorded as the amount of heat adjustment required to maintain a zero temperature difference between the cells.

Follicular fluid samples for these experiments were prepared by overnight dialysis (Slide-A-Lyzer cassette®; Pierce, Rockford, IL, USA; Mr cut-off 3500) against phosphate buffered saline (PBS) (0.1 mol/l, pH 7.4) at 4°C. Preliminary experiments had shown that undialysed samples produced a large, artefactual exothermic response to PBS, probably due to salt equilibration. Experiments were performed on a sample volume of 1.4 ml (the volume of the ITC sample chamber), with repeated injections of 10 µl aliquots of the spent dialysis medium.

A number of human follicular fluid samples and a single bovine sample were subjected to additional treatment before ITC analysis. Half of each dialysed sample was concentrated by placing in a Slide-A-Lyzer® cassette and covering with Dialysorb® gel (Sigma, Poole, Dorset, UK) for 30 min. This reduced the volume of the sample by between 30 and 50%. A quantity of spent dialysis medium was treated in the same way for use as a control.

The specific heat of bovine follicular fluid was estimated using a Microcal® VP-DSC scanning microcalorimeter (Plotnikov et al., 1997Go).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Incubator and microcalorimeter studies
Half (6/12) of the human and just over half (9/17) of the bovine fluid samples tested by the incubator method showed an endothermic response to dilution (Table IGo; Figure 1Go). The remaining samples, together with the porcine samples and saline controls, showed no detectable response. In no case was there a detectable exothermic response to dilution. Each of the endothermic responses was preceded by a lag period, lasting between 10 s and 2 min. The temperature depression in the responsive samples was up to 0.2°C and was sustained for between 9 and 26 min.


View this table:
[in this window]
[in a new window]
 
Table I. Incubator experiments: temperature changes in 500 µl aliquots of follicular fluid when diluted with 50 µl saline under controlled conditions as described in the text. Control samples were saline. Values are mean ± SEM (range)
 


View larger version (29K):
[in this window]
[in a new window]
 
Figure 1. Example of a temperature trace obtained using the incubator method. Saline was injected remotely, at the indicated time point, into a stirred sample of bovine follicular fluid, previously held at a constant temperature for at least 2 h. Sample temperature was monitored using a thermistor as described in the text. Paper moves from left to right. Lag: time between injection of saline and first detectable change in temperature. Duration: time between end of lag period and resumption of baseline temperature. Temperature fall was estimated from a calibration as described in the text. The slight upward drift of the baseline was linear for at least the periods either side of the response; this occurred in control and treatment traces and is unconnected with the experiment.

 
Responses in the microcalorimeter studies were recorded as a change in heat flux rather than temperature (Table IIGo; Figure 2Go) and were carried out on samples previously dialysed against PBS. In contrast to the incubator responses, all microcalorimeter effects were of rapid onset, were corrected by transient inputs of external heat energy and were capable of repetition. All porcine (5/5) and most bovine (4/6) samples showed an endothermic response. Three equine samples responded exothermically and one responded endothermically. Of the human samples, four responded endothermically and five responded exothermically. Reducing the volume of dialysed human fluid by about a third, using Dialysorb® gel, reversed the exothermic response of two samples and amplified the endothermy of three others by approximately four-fold (Table IIIGo). A single bovine sample, previously unresponsive, exhibited an endothermic response after being treated in this way.


View this table:
[in this window]
[in a new window]
 
Table II. ITC microcalorimeter experiments on follicular fluids from four species. Enthalpies (categorized as exothermic or endothermic) recorded during repeated 10 µl injections of spent phosphate buffered saline (PBS) dialysate into 1.4 ml of dialysed fluid. Individual enthalpy values were averages obtained from 6–25 injections at 3 min intervals; each enthalpic deviation recovered to baseline before the next injection. Cell temperature was 37°C. A control experiment (PBS injected into PBS) produced no response. Values are mean ± SEM (range) for n follicular fluid samples.
 


View larger version (28K):
[in this window]
[in a new window]
 
Figure 2. Example of microcalorimeter output, obtained on a sample of dialysed bovine follicular fluid. Each peak follows an injection (10 µl) of spent dialysis fluid into the sample (1.4 ml). The first small peak results from a syringe-clearance injection of minimal volume. Vertical axis shows rate of adiabatic heat adjustment by the calorimeter, with a positive deflection indicating heat input in response to an endothermic change in the sample. Area under peak indicates the total heat absorbed or liberated per injection.

 

View this table:
[in this window]
[in a new window]
 
Table III. ITC microcalorimeter studies (details as for Table IIGo) on follicular fluid before and after Dialysorb® concentration. All samples were initially dialysed against PBS. Dialysorb® concentrated samples were injected with Dialysorb® treated PBS
 
Bovine follicular fluid had a specific heat of 4.1 J/°C/ml (~1% less than water).

Mathematical modelling
The thermal conditions of a growing follicle can be modelled using simple equations in order to determine the rates of enthalpy which would be required to produce a hypothermia of the magnitude observed in vivo. A cool follicle growing within a warm tissue environment will gain heat in two ways: (i) as a result of fluid accretion (assuming that the fluid taken up during growth is at ambient circulatory temperature), and (ii) by conduction from surrounding stromal tissue. The former will occur at the rate of increase in follicle volume. The latter will occur at a rate predicted by Newton's law (proportional to the temperature gradient) and will increase with the surface area of the growing follicle. The absence of capillary circulation within the follicle means that the follicular surface is the only site of heat entry and that heat can move within the antral fluid only by convection or conduction. The bulk of the avascular cellular components of an antral follicle (granulosa layer) are located just beneath its surface membrane; thus any metabolic energy produced by these cells can be conveniently included in the excess heat of the surrounding stromal tissue. The energy production of the cells of the cumulus–oocyte complex is taken to be negligible. In the absence of any other metabolic effects, the temperature at the centre of the body of follicular fluid cannot be greater than that of any other region of fluid or that of the follicle surface.

The rate of heat entry (in Js–1) by water accretion,

where V is the volume of the follicle, SW is the specific heat of water, TS and TF are the temperatures (°K) of the stroma and follicle respectively. Note that Qw is the rate of excess heat entry (that is, the rate of entry of heat due to the excess temperature of the incoming water).

The rate of heat entry by conduction,

where r is the radius of the follicle, K is a conductivity constant for the site of heat entry and L is the thickness of the heat-conducting surface. Assuming the follicle wall to be a wet membrane with no significant insulation properties, we take K to be the thermal conductivity of water at 310°K [0.625 Js–1m–1°K–1; (Ramirez et al., 1995Go)]. For stromally-embedded regions of the follicle, heat will be transferred over the thickness of the follicle wall between the edge of the thecal capillary bed and the granulosa cell-fluid boundary. Histological evidence from cow and pig ovaries (Conley et al., 1995Go; Irving-Rodgers and Rodgers, 2000Go) suggests that this distance (L) has a minimum value of about 50 µm.

An initial simplification can be achieved by comparing the magnitudes of Qc and Qw. Given that

, where ris the radius of the growing follicle,

A representative bovine pre-ovulatory follicle of 15 mm diameter grows at a rate of 5 mm diameter d–1 (Savio et al., 1988Go) which is equivalent to a radial increase of 0.03 µm s–1 or a volume increase of 0.03 µl s–1. Taking the specific heats of water and follicular fluid to be identical, equation (3Go) gives a value of 1.05x105 for the magnitude ratio of the heat sources. Thus, heat entry by fluid accretion is quantitatively negligible and can be ignored. Evaluating Qc at an instantaneous follicle radius of 7.5 mm (1.8 ml volume), in the presence of a 2°K temperature gradient, gives a rate of heat uptake by conduction of 17.7 J.s–1. This result can be compared with the dilution-induced heat uptake detected in incubator and microcalorimetric experiments, as follows.

In the incubator studies, increasing the volume of the 500 µl fluid sample by 50 µl produced an average fall in temperature of 0.11°C, sustained for an average of 935 s (bovine values). This represents a total heat absorption of 0.23 J (sample volumexSwxtemperature difference). In the microcalorimeter studies, each injection of 10 µl into 1.4 ml gave instantaneous endothermic heat pulses approaching 100 µJ per injection. Using a heat capacity of 4.2 J degK–1 ml–1, this corresponds to a temperature decrease of 1.7x10–5°C per 10 µl injection. By proportion, the corresponding temperature change for injection of 50 µl into 500 µl would be approximately –2.4x10–4°C, which is significantly less than in the incubator experiments. The sustained nature of the incubator effect allows a rough estimation of a rate of heat uptake: assuming that heat was transferred to the cooling sample by conduction from the incubator air and that this took place predominantly at the open surface of the culture plate well (area 2 cm2), an adaptation of equation (2Go) provides a value of 0.24 J/s–1. This is equivalent to a total energy uptake of 224 J over 935 s. (For convenience, this calculation assumes as previously that the heat transfer surface is a wet membrane of 50 µm thickness. Note also that by ignoring any heat transferred to the sample through the plastic walls of the plate well, the rate of energy absorption will tend to be underestimated). This value for heat transfer rate approaches an order of magnitude similar to that calculated for the representative follicle, given that the temperature gradient in the incubator was lower than that assumed to occur in vivo and that the sample volume was smaller.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
These experiments show that follicular fluid is capable of responding endothermically to aqueous dilution. Not all fluids were found to be responsive and there were differences between species, but a significant number showed evidence of effects which might provide a basis for understanding the physiological and biochemical origin of follicular hypothermy. We have detected dilution-induced endothermy by two different methods but the magnitudes of the effects suggest that more than one biochemical mechanism may be operating. The result of the modelling exercise suggests that the incubator-detected mechanism of heat uptake may be sufficient to account for the in-vivo phenomenon. The microcalorimeter-detected mechanism is unlikely to do so. Whatever the mechanism, it is evident that a cool follicle will rapidly absorb heat by conduction and that a robust and continuous process of energy absorption must be available if the temperature gradient is to be maintained. The crudity of the incubator experiments appears to have revealed such a mechanism within some follicular fluids.

Overall, the incubator method produced either a fall in temperature (half the human and bovine samples) or no response (porcine samples); exothermic effects were never seen. The format of the response profiles (a lag followed by sustained suppression of temperature), together with the lack of response in controls, indicates that the effects were responses to dilution and not procedural artefacts of the relatively crude equipment (for example, due to opening the external incubator door at the time of saline injection). In contrast, using the more sensitive microcalorimeter, all porcine and most bovine and samples exhibited endothermy. Equine and human samples showed both types of response.

The two types of experimental approach produced different types of response. The incubator studies, which used undialysed samples diluted by about 11% with saline, typically produced a delayed reduction in temperature which could be maintained for up to 25 min against the thermal gradient imposed by the circulating incubator air. In contrast, the microcalorimeter method, using dialysed samples hydrated with spent dialysis medium, produced near instantaneous falls in temperature (rapidly equilibrated by the machine) in response to repeated volume changes of about 0.6%. Thus, although both methods produced evidence of negative enthalpy, differences in response profile and magnitude suggest that more than one kind of biochemical response is being observed. This is also illustrated by the porcine samples, which showed no response in the incubator but uniform endothermy in the microcalorimeter. It is possible that the relative insensitivity of the incubator method hid more rapid, transient changes in heat balance occurring immediately after addition of diluent, or that the different basal temperatures of the two experimental approaches revealed different characteristics of the mechanism. Further experiments are needed to resolve these possibilities.

The bovine and porcine samples came from animals whose reproductive cycle stage was unknown, whereas the human and equine samples were from follicles in the final stages of stimulated (human) or natural (equine) maturity. All follicles were large enough to be considered as being either in a phase of active volumetric growth or at a stage of immediate pre-ovulatory maturity (when fluid accretion probably ceases and follicles exhibit flaccidity; Lenz, 1985Go; Espey and Lipner, 1994Go; Hanna et al., 1994Go; Fisch et al., 1996Go; Kerban et al., 1999Go). Using our hypothesis that hydration-induced endothermy is a growth-associated phenomenon, unrecorded differences in the exact stages of maturity may well explain the presence or absence of an endothermic response amongst the follicles tested. These conclusions are supported by the response to Dialysorb® treatment of the human and bovine fluids. The effect of post-dialysis volume reduction was either to amplify existing endothermy or to reverse the previous exothermy. This is consistent with there being a saturable hydration-dependent effect which is masked either by the completion of volumetric follicle growth under the IVF stimulation protocol, or by the experimental requirement for pre-calorimetric dialysis. We hypothesize that concentrating the fluid with Dialysorb® reverses the previous hydrational equilibrium of the sample, allowing a greater endothermic response to be seen.

The biochemical basis for the effect(s) we have observed remains to be elucidated. The follicular membrane is quite `leaky' to fluids and electrolytes and little or no hydrostatic or osmotic pressure can be detected between sampled fluid and blood (Gosden et al., 1988Go; Salustri et al., 1999Go). We must therefore conclude that the molecules responsible are large and, in the case of the microcalorimetrically-detectable effect, larger than 3000 Mr. On this basis, the effects we have seen represent a residual amount of passive hydration (incubator experiments) or a forced over-hydration of the fluid (microcalorimeter experiments: samples previously equilibrated by dialysis). The viscosity of follicular fluid, though positionally and temporally variable (Kerban et al., 1999Go), does not appear to change as a simple function of growth (Luck et al., 2000Go; human, bovine, porcine), suggesting that high molecular weight components are secreted continuously as the follicle expands. The present data are consistent with the hypothesis that a large, hydratable molecule, such as a mucopolysaccharide (Zachariae, 1959Go), is secreted within the follicle as part of the growth process and that this undergoes a rapid hydration with endothermic consequences. The continuous secretion of such a material during follicle growth would result in a constant uptake of heat energy from the surrounding tissue. Secretion at an increasing rate, to keep pace with the increasing rate of volume expansion associated with maturation (van Wezel and Rodgers, 1996Go; Irving-Rodgers and Rodgers, 2000Go), would be consistent with the greater temperature depression of larger follicles (Hunter et al., 1997Go, 2000Go).

Although the nature of the molecules responsible for this phenomenon have yet to be determined, it is known that the hydration of protein/proteoglycan complexes, as for example secreted by tracheal cells (Dodd et al., 1998Go) or in the cornea (Doughty, 2001), takes several seconds or minutes to complete. Such phenomena are multifactorial (Hatakeyama and Hatakeyama, 1998Go) but the time course of the incubator responses in the present study (including the presence of a significant lag period) suggests that similar molecules may be involved.

In conclusion, we have evidence that follicular fluid from humans and domestic species is capable of responding endothermically to dilution. The mechanism(s) of the response is unknown but may provide a basis for understanding the hypothermy of large follicles observed in vivo. We propose that follicle growth involves the continuous secretion of a large molecule whose hydration results in the endothermic cooling of the fluid.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
We thank H.Wahabi, H.Almislimani and clinical staff at NURTURE for collection of human follicular fluids, J.Corbet and V.Brankin for collection of bovine and porcine follicles, C.Deuchar for constructing the thermocouple, and R.H.F.Hunter and S.E.Harding for helpful discussion. The Biological Microcalorimetry Facility (Glasgow) is funded jointly by BBSRC and EPSRC.


    Notes
 
4 To whom correspondence should be addressed. E-mail: martin.luck{at}nottingham.ac.uk Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Andersen, D.H. (1926) Lymphatics and blood-vessels of the ovary of the sow. Contrib. Embryol., 17, 107–123.

Benoit, H.J., Borth, R., Ellicott, A.R. et al .(1976) Periovulatory changes in ovarian temperature in ewes. Am. J. Obstet. Gynecol., 124, 356–360.[ISI][Medline]

Conley, A.J., Kaminski, M.A. and Dubowsky, S.A. et al .(1995) Immunolocalization of 3ß-hydroxysteroid dehydrogenase and P450 17a-hydroxylase during follicular and luteal development in pigs, sheep, and cows. Biol. Reprod., 52, 1081–1094.[Abstract]

David, A., Vilensky, A. and Nathan, H. (1972) Temperature changes in different parts of the rabbit's oviduct. Int. J. Gynaecol. Obstet., 10, 52–56.

Dodd, S., Place, G.A., Hall, R.L. et al .(1998) Hydrodynamic properties of mucins secreted by primary cultures of guinea-pig tracheal epithelial cells: determination of diffusion coefficients by analytical ultracentrifugation and kinetic analysis of mucus gel hydration and dissolution. Eur. Biophys. J., 28, 38–47.[ISI]

Doughty, M.J. (2001) Changes in hydration, protein and proteoglycan composition of the collagen-keratocyte matrix of the bovine corneal stroma ex vivo in a bicarbonate-mixed salts solution, compared to other solutions. Bioch. Biophys. Acta, 1525, 97–107.[ISI][Medline]

Einer-Jensen, N. (1988) Countercurrent transfer in the ovarian pedicle and its physiological implications. Oxf. Rev. Reprod. Physiol., 10, 348–381.

Espey, L.L. and Lipner, H. (1994) Ovulation. In Knobil, E. and Neill, J.D. (eds) The Physiology of Reproduction, 2nd edn. Raven Press, New York, pp. 725–767.

Fisch, B., Harel, L., Amit, S. et al .(1996) Viscosity and refractive index of follicular fluid in relation to in vitro fertilization. J. Assist. Reprod. Genet., 13, 468–471.[ISI][Medline]

Gosden, R.G., Hunter, R.H.F., Telfer, E. et al .(1988) Physiological factors underlying the formation of follicular fluid. J. Reprod. Fertil., 82, 813.[Medline]

Greve, T., Grøndahl, C., Schmidt, M. et al .(1996) Bovine preovulatory follicular temperature: Implications for in vitro production of embryos. Arch. Tierz. Dummersdorf, 39, 7–14.

Grinsted, J., Blendstrup, K., Andreasen, M.P. et al .(1980) Temperature measurements of rabbit antral follicles. J. Reprod. Fertil., 60, 149–155.[Abstract]

Grinsted, J., Kjer, J.J., Blendstrup, K. et al .(1985) Is low temperature of the follicular fluid necessary for normal oocyte development? Fertil. Steril., 43, 34–39.[ISI][Medline]

Grøndahl, C., Greve, T., Schmidt, M. et al .(1996) Bovine preovulatory follicles are cooler than ovarian stroma and deep rectal temperature. Theriogenology, 45, 289.

Hanna, M.D., Chizen, D.R. and Pierson, R.A. (1994) Characteristics of follicular evacuation during human ovulation. Ultrasound Obstet. Gynecol., 4, 488–493.[ISI][Medline]

Hatakeyama, H. and Hatakeyama, T. (1998) Interaction between water and hydrophilic polymers. Thermochim. Acta, 308, 3–22.[ISI]

Hunter, R.H.F. and Nichol, R. (1986) A preovulatory temperature gradient between the isthmus and ampulla of pig oviducts during the phase of sperm storage. J. Reprod. Fertil., 77, 599–606.[Abstract]

Hunter, R.H.F., Grøndahl, C., Schmidt, M. et al .(1995) Graafian follicles are cooler than ovarian stroma. J. Reprod. Fertil. (Abstr. Ser.), 16, 8–9.

Hunter, R.H.F., Grøndahl, C., Greve, T. et al .(1997) Graafian follicles are cooler than neighbouring ovarian tissues and deep rectal temperatures. Hum. Reprod., 12, 95–100.[Abstract]

Hunter, R.H.F., Bøgh, I.B., Einer-Jensen, N. et al .(2000) Pre-ovulatory Graafian follicles are cooler than neighbouring stroma in pig ovaries. Hum. Reprod., 12, 95–100.[Abstract]

Irving-Rodgers, H.F. and Rodgers, R.J. (2000) Ultrastructure of the basal lamina of bovine ovarian follicles and its relationship to the membrana granulosa. J. Reprod. Fertil., 118, 221–228.[Abstract/Free Full Text]

Kerban, A., Doré, M. and Sirois, J. (1999) Characterization of cellular and vascular changes in equine follicles during hCG-induced ovulation. J. Reprod. Fertil., 117, 115–123.[Abstract]

Lenz, S. (1985) Ultrasonic study of follicular maturation, ovulation and development of corpus luteum during normal menstrual cycles. Acta Obstet. Gynecol. Scand., 64, 15–19.[ISI][Medline]

Luck, M.R., Ye, J., Almislimani, H. et al .(2000) Follicular fluid rheology and the duration of the ovulatory process. J. Reprod. Fertil., 120, 411–421.[Abstract/Free Full Text]

Plotnikov, V.V., Brandts, J.M., Lin, L.N. et al .(1997). A new ultrasensitive scanning calorimeter. Anal. Biochem., 250, 237–244.[ISI][Medline]

Ramirez, M.L.V., Nieto de Castro, C.A., Nagasaka, Y. et al .(1995) Standard reference data for the thermal conductivity of water. J. Phys. Chem. Ref. Data, 24, 1377–1381.[ISI]

Salustri, A., Camaioni, A., Di Giacomo, M. et al .(1999) Hyaluronan and proteoglycans in ovarian follicles. Hum. Reprod. Update, 5, 293–301.[Abstract/Free Full Text]

Savio, J.D., Keenan, L., Boland, M.P. et al .(1988) Pattern of growth of dominant follicles during the oestrous cycle of heifers. J. Reprod. Fertil., 83, 663–671.[Abstract]

Van Wezel, I.L. and Rodgers, R.J. (1996) Morphological characterization of bovine primordial follicles and their environment in vivo. Biol. Reprod., 55, 1003–1011.[Abstract]

Wiseman, T., Williston, S., Brandts, J.F. et al .(1989) Rapid measurement of binding constants and heats of binding using a new titration calorimeter. Anal. Biochem., 179, 131–137.[ISI][Medline]

Zachariae, F. (1959) Acid mucopolysaccharides in the female genital system and their role in the mechanism of ovulation. Acta Endocrinol., 33 (Suppl. 47), 11–16.

Submitted on May 18, 2001; accepted on August 31, 2001.