The results reveal that under the conditions of the present study, FT, immature ZP (oocytes) are less suitable for use in feline ZBA. Only a very low number of fresh spermatozoa bound to FT ZP, compared with their fresh, in vitro matured counterparts. With FT spermatozoa tested on FT ZP, there was no sperm-zona binding at all. This may partly be due to changes in the capacity of spermatozoa to maintain a normal plasma membrane surface that could bind to the specific receptors present in the ZP.
The spermatozoa used for the ZBA were cleansed prior to exposure to the oocytes. Two different procedures were used, washing and centrifugation or gradient centrifugation through a column of Percoll, processes that are not harmful to spermatozoa [15, 30, 33, 34] but that are used to recover the best spermatozoa, in terms of motility and sperm viability . As expected, the motility of electroejaculated fresh spermatozoa did not differ before and after the procedures, whereas a large variation was seen among the FT spermatozoa. It is not likely that the cleansing procedures, in particular using Percoll gradients, would have influenced the results of the ZBA, considering the very few FT spermatozoa that bound to in vitro matured ZP in comparison with freshly ejaculated spermatozoa.
Both epididymal, electroejaculated spermatozoa and spermatozoa collected by artificial vagina have been used for ZBAs in the domestic cat. A comparison of ZP attachment between epididymal and ejaculated spermatozoa in the domestic cat has revealed that more epididymal compared with ejaculated spermatozoa bind in the first 60 minutes following incubation, without differences in sperm capacitation status between them [10, 36]. By contrast, a study on IVF of hamster ova found no differences regarding penetration rates, time of sperm penetration, and sperm concentration between ejaculated and epididymal spermatozoa . To the authors' knowledge, corresponding work has not previously been performed in the domestic cat, and therefore we do not know how much the sperm source would have influenced the results of the ZBA in the present study. In any case, if epididymal spermatozoa should bind better than ejaculated spermatozoa, it should have influenced the results of FT spermatozoa in a positive way. However, there was no zona binding at all between FT oocytes and FT spermatozoa.
The incubation time for the sperm-oocyte complexes in the present study was based on Goodrowe & Hay's results . These authors revealed that the number of attached sperm/zona and the percentage of zonae with attached spermatozoa reached maximum values after 4 hours of incubation, to decrease thereafter. Under the conditions of the present study, chilled spermatozoa bound to FT oocytes (data not shown) at a rate similar to that of fresh spermatozoa. These results correspond to those of Goodrowe & Hay , who found that chilled spermatozoa could be used for zona-free hamster ova penetration and homologous zona attachment at comparable rates as fresh spermatozoa. Our results clearly show a decreased, but not abolished, capacity of FT spermatozoa to bind to in vitro matured ZP. The viability of the FT spermatozoa used, assessed as progressive motility, was acceptable – albeit lower than that of fresh spermatozoa, even considering the use of cleansing procedures that would have selected for better sperm morphology, viability and motility. Such differences in viability between fresh and FT spermatozoa may explain the differences in ZP binding registered. However, the lack of binding to FT ZP is mainly due to the ZP and not to the spermatozoa, since fresh spermatozoa also had markedly decreased binding to FT ZP. In line with this argumentation, such decreased binding capacity of the ZP may reside in changes of the structure of the ZP that occur during unprotected freezing and/or thawing of the ovaries. The functional ability of the ZP to bind spermatozoa is closely related to its morphological appearance . Oocyte storage can cause structural changes in the ZP, which may affect the number of bound spermatozoa .
However, in the female dog, the freezing of ovaries and zona binding with FT oocytes are possible, as previously indicated in several studies [6–8]. In the present study using SEM on queen oocytes, a clear morphological difference was shown in the ZP outer surface between in vitro matured and FT ZP. The in vitro matured ZP showed a dense surface with few fenestrations in contrast to their FT, immature counterparts, where fenestrations were conspicuously larger. These results agree with those previously reported by Ström Holst et al.  in the female dog. In vitro maturation has been associated with a more porous appearance in other species (e.g. mouse ). The ultrastructural changes in the FT oocytes were probably caused by damage during the freezing-thawing process, causing significantly reduced sperm binding capacity. Moreover, the ultrastructural difference accounted for the observation that few (fresh), or no (FT), spermatozoa bound to FT oocytes in the present study. Storage may affect oocytes from queen cats and female dogs in different ways, and results for female dogs may therefore not be accurate for cats. For instance, it was revealed by Ström Holst et al.  that in dogs, fresh oocytes bind significantly more spermatozoa than salt-stored oocytes do. These results differ from those reported by Andrews et al. , who found no difference in zona binding capacity between fresh, matured oocytes and oocytes that had been salt-stored for 1.5–24 weeks after maturation in the queen. It was shown by Ström Holst et al.  that in dogs, deep-freezing of the ovaries is better than salt storage of the oocytes. Corresponding results have not been shown for the cat. Studies by other authors have indicated that FT queen oocytes could be used for ZBA in cats. For instance, Kashiwazaki et al. , using immature FT oocytes, report binding of epididymal FT spermatozoa to frozen-thawed oocytes, a binding that we were unable to show. Unfortunately, in their study the number of bound spermatozoa per oocyte was not given, and there was no control group with fresh oocytes. The oocytes were, moreover, frozen in a different manner than in the present study, which may explain the differences in results. In our study the ovaries of ovario-hysterectomised queens were simply frozen in NaCl at ~-20°C, and the oocytes retrieved after thawing. By contrast, in the study by Kashiwazaki et al.  the oocytes were recovered before freezing and were frozen in 0.25 mL plastic straws with 1.5 M glycerol, conditions that make the collection and use of these oocytes less practical. Similarly, in the study by Luvoni & Pellizarri  the oocytes were recovered before freezing, and frozen in 0.5 mL straws with cryo-protectant. The freezing procedures described by Luvoni & Pellizarri  and Kashiwazaki et al.  may maintain the structure of the ZP, thus being beneficial for ulterior ZBA. In the present study the ovaries were frozen and the oocytes were not retrieved until after thawing of the ovaries, in order to make collection of material as practical as possible.