Incorporation of ZP1 into perivitelline membrane after in vivo treatment with exogenous ZP1 in Japanese quail (Coturnix japonica) Mihoko Kinoshita 1 , Kaori Mizui 1 , Tsukasa Ishiguro 1 , Mamoru Ohtsuki 1 , Norio Kansaku 2 , Hiroshi Ogawa 3 , Akira Tsukada 4 , Tsukasa Sato 1 and Tomohiro Sasanami 1 1 Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, Japan 2 Laboratory of Animal Genetics and Breeding, Azabu University, Fuchinobe, Sagamihara, Japan 3 Laboratory of Wild Animals, Tokyo University of Agriculture, Atsugi-shi, Japan 4 Graduate School of Bioagricultural Sciences, Nagoya University, Japan The avian vitelline membrane in laid eggs consists of three layers: the outer layer; the continuous membrane; and the inner layer [1]. The outer layer, which is com- posed of a varying number of sublayers of latticed fine fibrils, is formed in the infundibulum part of the ovi- duct [2]. The continuous membrane, which is a very thin granular membrane, is also formed in the infun- dibulum [2]. The inner layer, a 3D network of coarse fibers, is found between the granulosa cells and the oocyte in follicles before ovulation and is called the perivitelline membrane (PL) [3]. The PL is a homo- logue of the egg envelope in other vertebrates, the zona pellucida in mammals, the vitelline membrane in amphibians and the chorion in teleosts. These egg Keywords Japanese quail; matrix assembly; perivitelline membrane; zona pellucida; ZP1 Correspondence T. Sasanami, Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, 836 Ohya, Shizuoka 422-8529, Japan Fax ⁄ Tel: +81 54 238 4526 E-mail: atsasan@agr.shizuoka.ac.jp (Received 20 March 2008, revised 7 May 2008, accepted 13 May 2008) doi:10.1111/j.1742-4658.2008.06503.x In birds, the egg envelope surrounding the oocyte prior to ovulation is called the perivitelline membrane and it plays important roles in fertiliza- tion. In a previous study we demonstrated that one of the components of the perivitelline membrane, ZP3, which is secreted from the ovarian granu- losa cells, specifically interacts with ZP1, another constituent that is synthe- sized in the liver of Japanese quail. In the present study, we investigated whether ZP1 injected exogenously into the blood possesses the ability to reconstruct the perivitelline membrane of Japanese quail. When ZP1 puri- fied from the serum of laying quail was injected into other female birds, the signal of this exogenous ZP1 was detected in the perivitelline mem- brane. In addition, we revealed, by means of ligand blot analysis, that serum ZP1 interacts with both ZP1 and ZP3 of the perivitelline membrane. By contrast, when ZP1 derived from the perivitelline membrane was administered, it failed to become incorporated into the perivitelline mem- brane. Interestingly, serum ZP1 recovered from other Galliformes, includ- ing chicken and guinea fowl, could be incorporated into the quail perivitelline membrane, but the degree of interaction between quail ZP3 and ZP1 of the vitelline membrane of laid eggs from chicken and guinea fowl appeared to be weak. These results demonstrate that exogenous ZP1 purified from the serum, but not ZP1 from the perivitelline membrane, can become incorporated into the perivitelline membrane upon injection into other types of female birds. To our knowledge, this is the first demon- stration that the egg envelope component, when exogenously administered to animals, can reconstruct the egg envelope in vivo. Abbreviations CBB, Coomassie Brilliant Blue; DIG, digoxigenin; PL, perivitelline membrane; PVDF, poly(vinylidene difluoride); ZP, zona pellucida. 3580 FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS envelopes are mainly constructed of glycoproteins belonging to different subclasses of the zona pellucida (ZP) gene family [4–7]. The components of this matrix include three glycoproteins (ZP1, ZP2 and ZP3) in mouse [4,8] and four glycoproteins (ZP1, ZP2, ZP3 and ZP4) in several other organisms, including humans, bonnet monkeys and rats [9–11]. The ZP gene family proteins are characterized by a highly conserved amino acid sequence called the ZP domain, consisting of about 260 amino acid residues with 8 or 10 con- served Cys residues [12]. Two glycoproteins, ZP3 and ZP1, have been iden- tified as major components of the PL in quail oocytes [13], and each of these glycoproteins is an essential structural component of the extracellular coat. In addition to these major glycoproteins, we recently cloned the cDNA encoding ZPD (GenBank accession number: AB301422) and ZP2 (GenBank accession number: AB295393), which are also expressed in the ovary of laying quail (M. Kinoshita & T. Sasanami, unpublished results). We previously reported that the ZP3 protein was produced in the granulosa cells of developing follicles by the stimu- lation of follicle-stimulating hormone [14,15] and testosterone [16]. On the other hand, the other con- stituent, ZP1, is synthesized in the liver of the laying female, and the expression of ZP1 mRNA was stim- ulated by in vivo treatment with diethylstilbestrol [17]. In a previous study, we demonstrated that the ZP3 secreted from the granulosa cells specifically binds with ZP1, and that this phenomenon might be involved in the formation of insoluble PL fibers in the quail ovary [18]. We also provided evidence dem- onstrating that the C-terminal half of the ZP domain of ZP1 contains a binding site for ZP3 [19]. Similarly, Okumura et al. also demonstrated that ZP1 in the serum of laying hens specifically interacts with ZP3 [20] and that this interaction induces the formation of fibrous aggregates, which are visible under optical microscopy [21]. In the present study, we aimed to clarify whether ZP1, when injected exogenously into the blood, pos- sesses the ability to reconstruct the PL of Japanese quail. To achieve this, we purified ZP1 proteinfrom the serum and from the PL lysate and then injected the proteins intravenously into birds, which were then analyzed to establish the level of ZP1 incorpo- ration in the PL. Moreover, we compared the differ- ences in the incorporation of the ZP1 purified from Japanese quail with the incorporation of ZP1 recov- ered from other Galliformes (chicken and guinea fowl). To our knowledge, this is the first report to demonstrate that exogenous ZP1 derived from the serum, but not from the PL, can be successfully incorporate into the PL. Results Incorporation of exogenously injected ZP1 into quail PL To determine whether ZP1 injected intravenously into a bird was incorporated into the PL of the ovarian fol- licles, we first purified the 97-kDa ZP1 from either the serum or the PL lysate of laying quail and labeled the samples with digoxigenin (DIG). As shown in Fig. 1, presence of the purified serum ZP1 (Fig. 1A, lane 1), as well as of the purified PL ZP1 (Fig. 1B, lane 1), pro- duced a dominant band migrating at a molecular mass of 97 kDa. In addition, both 97-kDa bands reacted strongly with anti-ZP1 serum (Fig. 1A,B, lane 2) as well as with anti-DIG immunoglobulin (Fig. 1A,B, lane 3). These results suggest that each glycoprotein purified by means of the methods in this study was practically pure and was successfully labeled with DIG. The DIG-labeled ZP1 was injected into laying females, and the PL lysates isolated from the largest follicles 6 h after the injection were probed with anti- DIG immunoglobulin. The results are shown in Fig. 2. As shown in Fig. 2A, the PL lysate from the oocytes of birds injected with DIG-labeled PL ZP1 (Fig. 2A, lane 1) or DIG-labeled serum ZP1 (Fig. 2A, lane 2) showed similar banding patterns after staining the gel with Coomassie Brilliant Blue R250 (CBB). When the sample was analysed by western blotting and probed with anti-DIG immunoglobulin (Fig. 2B), immuno- reactive bands migrating at around 175 and 97 kDa, which corresponds to the apparent molecular mass values of dimeric and monomeric ZP1, respectively, were detected in the PL lysate from the animals injected with DIG-labeled serum ZP1 (lane 2). The immunoreactive bands were observed to migrate at a level consistent with a molecular mass of approxi- mately 120 kDa (Fig. 2B, lane 2). This immunoreactive protein might be the 97-kDa ZP1 complex, which was formed through an intermolecular disulfide bond with the ZP1 fragment or with other partner(s), because this band shifted to 97 kDa when the proteins were sepa- rated under reducing conditions [19]. Interestingly, the PL lysate obtained from the DIG-labeled PL ZP1-treated birds contained no immunoreactive mate- rials (Fig. 2B, lane 1). The samples obtained from the birds treated with NaCl ⁄ P i alone also had no immuno- reactive bands (lane 3). These results suggest that the intravenously injected serum ZP1, but not PL ZP1, is M. Kinoshita et al. Incorporation of ZP1 into perivitelline membrane FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS 3581 incorporated into the PL of the quail ovary and the intravenously injected DIG-labeled serum ZP1 could be detected in the endogenous PL 6 h after injection. In subsequent experiments, we investigated the accu- mulation of the intravenously injected ZP1 protein in the PL during follicular development by western blot analysis. As shown in Fig. 3, the F3 lysate contains a small amount of immunoreactive protein (Fig. 3B, lane 1), but the intensity of the band increases in mature follicles (F2: Fig. 3, lane 2; F1: Fig. 3, lane 3). The PL lysate isolated from the control animals, which were injected with NaCl ⁄ P i alone, contained no immuno- reactive bands (Fig. 3A). These results suggest that the exogenously injected serum ZP1 is accumulated in the final stage of follicular growth. To investigate the localization of the injected ZP1 protein in the follicles, we prepared paraffin sections of pre-ovulatory follicles and analyzed them by immunofluorescence microscopy. As shown in Fig. 4, the immunoreactive material recognized by anti-DIG immunoglobulin accumulated in the PL apposed to the apical surface of the granulosa cells of the largest follicle (Fig. 4A). No such intense immunostaining was seen when the specimens prepared from the control animals were stained with anti-DIG immuno- globulin (Fig. 4B). Taken together, these results clearly demonstrate that the serum ZP1 intravenously injected into the animals is transported into the ovary via the blood circulation, and that the protein accumulates in the PL of mature follicles in vivo. 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa M1 2 3 M1 2 3 A B Fig. 1. Purification of ZP1 from the serum or the PL of laying quail. The ZP1 proteins purified from the serum of laying quail (A) or recovered from the PL lysate of the largest follicles (B) were labeled with DIG and separated by SDS-PAGE (1 lg per lane). The proteins were then detected using silver staining (lane 1). The same sample was also western blotted then probed using anti-ZP1 serum (lane 2) or anti-DIG immunoglobulin (lane 3). The molecular mass markers were also separated by SDS-PAGE and detected using silver staining (lane M). Molecular mass values of the mark- ers are given on the left of the respective lanes. Representative results of repeated experiments are shown. 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 21.5 kDa 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa M12 12 3 AB Fig. 2. Detection of exogenously injected ZP1 in the PL lysates. (A) The PL lysates isolated from the largest follicles of the quail administered DIG-labeled PL (lane 1) or serum ZP1 (lane 2) were separated on SDS-PAGE (10 lg per lane) and were stained with CBB. The molecular mass marker was also separated by SDS-PAGE and detected using CBB staining (lane M). (B) The same samples in (A) were probed with anti-DIG immunoglobulin (lanes 1 and 2). The PL lysate of the quail (10 lg per lane) injected with vehicle alone was used as a negative control (lane 3). Representative results of repeated experiments are shown. Incorporation of ZP1 into perivitelline membrane M. Kinoshita et al. 3582 FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS Both ZP1 and ZP3 in the PL interact with serum ZP1 To identify the binding partner of serum ZP1 in the PL, we identified the PL lysate of the largest follicles by means of ligand blotting (Fig. 5). When the sample was probed using DIG-labeled serum ZP1, bands migrating at 175, 97 and 35 kDa, which are the expected molecular mass values of dimeric ZP1, mono- meric ZP1 and ZP3, respectively, were visualized (lane 1). When the same sample was probed using DIG- labeled bovine serum albumin (lane 2) or without ligand (lane 3, blocking buffer alone), no such staining was seen. These results suggest that the serum ZP1 interacts with both ZP1 and ZP3, and that this interac- tion might be responsible for the incorporation of serum ZP1 into the PL. Species-specific interaction of quail ZP3 with ZP1 from various species In our previous study we demonstrated that tritium- labeled ZP3, which is present in the conditioned med- ium of the granulosa cells, binds specifically to ZP1 of the PL in Japanese quail [18,19]. To investigate whether the interaction of ZP3 and ZP1 occurs in other bird species, we analyzed the lysates of vitelline membrane of laid eggs from various species using ligand blotting and probing with radiolabelled ZP3. To achieve this, we first confirmed the presence of ZP3 and ZP1 proteins in the lysate of the vitelline membrane by western blot analysis. As shown in Fig. 6A, our anti-ZP3 serum reacted strongly with the bands of approximately 33 kDa molecular mass in the lysate of the vitelline membrane of Japanese quail (lane 1), blue-breasted quail (lane 2), chicken (lane 3), turkey (lane 4), and guinea fowl (lane 5). Similarly, anti-ZP1 serum also recognized dimeric (around 175 kDa molecular mass) as well as monomeric (around 97 kDa molecular mass) ZP1s in the lysates of all vitelline membranes tested (Fig. 6B). These results suggest that the vitelline membrane of laid 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 12 3 31 kDa 12 3 A B Fig. 3. Western blot analysis of exogenous ZP1 during follicular development. (A) The PL lysate (10 lg per lane) isolated from the third-largest (lane 1), the second-largest (lane 2), or the largest (lane 3) follicles of the control bird (vehicle alone), was transblotted onto a PVDF membrane after separation on SDS-PAGE and then probed with anti-DIG immunoglobulin (A). The samples (10 lg per lane) prepared from the third-largest (lane 1), the second-largest (lane 2), or the largest (lane 3) follicles of the bird receiving intravenous injection of DIG-labeled serum ZP1 were probed with anti-DIG immunoglobulin (B). The blots shown are representative of three independent experiments. A B Fig. 4. Immunohistochemical analysis of exogenous ZP1 in the fol- licular wall. Sections of follicular wall obtained from the largest folli- cles of the animals treated with DIG-labeled serum ZP1 (A) or with vehicle alone (B) were processed for immunohistochemical obser- vation using anti-DIG immunoglobulin (1 : 300 dilution). Shown are the representative results of three experiments. Bar = 50 lm. M. Kinoshita et al. Incorporation of ZP1 into perivitelline membrane FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS 3583 eggs from these birds contains ZP3 and ZP1, like that of Japanese quail. Upon examination of the poly(vinylidene difluoride) (PVDF) membrane that had been used to immobilize the aliquots of SDS-solubilized vitelline membrane proteins of Japanese quail, dimeric and monomeric ZP1s were detected by autoradiography (Fig. 6D, lane 1). This result is consistent with previous reports showing that the dimeric as well as the monomeric ZP1s were visualized by ligand blotting with radio- labelled ZP3 [18,19]. Comparable staining intensity of the ZP1 bands was observed when the lysates from blue-breasted quail (lane 2) and turkey (lane 4) were incubated with quail ZP3. In comparison with that of Japanese quail, the ZP1 bands in the vitelline mem- brane of chicken (lane 3) and guinea fowl (lane 5) were stained weakly. The weaker staining intensity observed in the case of the chicken and guinea fowl was not the result of a lower abundance of ZP1 in the vitelline membrane because comparable amounts of ZP1 were detected in the lysates when the samples were stained with CBB (Fig. 6C). These results indicate that the interaction of ZP3 and ZP1 might occur in a species- specific manner. Incorporation of heterologous ZP1 into quail PL In the next set of experiments, we investigated whether there is species specificity in the incorporation of serum ZP1 into the PL. For this purpose, we purified serum ZP1 from chicken and guinea fowl in addition to Japa- nese quail, and the purified ZP1 of these species was injected intravenously into several laying quail, sever- ally. As shown in Fig. 7A, reruns of the purified quail (lanes 1 and 4), chicken (lanes 2 and 5) and guinea fowl (lanes 3 and 6) ZP1 proteins showed a dominant band after western blotting and probing using anti- ZP1 serum (lanes 1-3) as well as after silver staining (lanes 4-6). As shown in Fig. 7B, the PL lysate of the birds that had been injected with quail (lane 1), chicken (lane 2), or guinea fowl (lane 3) ZP1 contained materials that were immunoreactive with anti-DIG immunoglobulin and migrated close to dimeric and monomeric ZP1. It should be noted that the immunoreactive monomeric ZP1 in the bird injected with serum ZP1 from a differ- ent species showed distinct disparity in mobility on SDS-PAGE. The mobilities of the monomeric ZP1 pro- teins corresponded with those of the purified protein in the silver-stained gel (Fig. 7A, lanes 4-6). Although the reasons are unclear, the composition of the materials immunoreactive with anti-DIG immunoglobulin in the PL lysate isolated from the oocyte of the animals injected with quail ZP1 was different from that shown in Fig. 2B, lane 2, in spite of being the same prepara- tion. On the other hand, no distinct difference in band- ing pattern was observed when the total protein lysate of each sample was visualized using CBB (Fig. 7C). Taken together with the results shown in Fig. 5, these results suggest that the incorporation of serum ZP1 into the PL can occur between different species, and that the specificity of the incorporation might be relatively low in comparison with that of ZP3 incorporation. Discussion As reported previously in our studies using Japanese quail, ZP3 secreted from cultured granulosa cells spe- cifically interacts with ZP1, which is synthesized in the liver, and this interaction appears to be involved in the formation of the insoluble fibers of the PL [18,19]. In the present study, we showed, for the first time, that exogenous ZP1 purified from the serum of laying quail was incorporated into the PL when it was injected into other female birds. To our knowledge, this is the first demonstration that the egg envelope component, when exogenously administered to birds, actually recon- structs the egg envelope in vivo. In accordance with 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 123 Fig. 5. Ligand blot analysis of PL lysate. (A) Aliquots (10 lg per lane) of the SDS-solubilized PL isolated from the largest pre-ovula- tory follicles of Japanese quail were subjected to ligand blot analy- sis using DIG-labeled serum ZP1 (lane 1), DIG-labeled bovine serum albumin (lane 2), or no ligand (lane 3). Representative blots of three independent experiments are shown. Incorporation of ZP1 into perivitelline membrane M. Kinoshita et al. 3584 FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS our data, Okumura et al. recently demonstrated that chicken ZP3 and ZP1 can specifically associate to form a heterocomplex of ZP3 and ZP1, and this interaction induces ZP3–ZP1 fibrous aggregates in vitro ; however, they did not confirm this phenomenon by an in vivo study [20,21]. In the present study we found that when ZP1 puri- fied from the PL of the largest follicles was injected intravenously into birds, it did not become incorpo- rated into the PL. This was not a result of denatur- ation of the protein during the purification process using SDS-PAGE, because the serum ZP1 was also 123 AB 4 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 21.5 kDa 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 21.5 kDa 1234 C 12 3 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 21.5 kDa 45 6 Fig. 7. Cross-species incorporation of serum ZP1 into quail PL. (A) ZP1 was purified from the sera of Japanese quail (lanes 1 and 4), chicken (lanes 2 and 5), or guinea fowl (lanes 3 and 6). They were separated on SDS-PAGE (1 lg per lane) and detected with western blotting using anti-ZP1 immunoglobulin (lanes 1–3) or silver staining (lanes 4–6). (B, C) The PL lysate isolated from the largest follicles of the animals, which were injected with 10 lg of DIG-labeled quail ZP1 (lane 1), chicken ZP1 (lane 2), guinea fowl ZP1 (lane 3), or vehicle alone (lane 4), were separated on SDS-PAGE (1 lgÆlane )1 ) and detected with anti-DIG immunoglobulin (B). The same lysates (20 lg per lane) were separated on SDS-PAGE and stained with CBB (panel C). The results representative of three experiments are shown. 123 A 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa ZP3 dimeric ZP1 monomeric ZP1 ZP3 dimeric ZP1 monomeric ZP1 dimeric ZP1 monomeric ZP1 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 21.5 kDa 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 21.5 kDa 200 kDa 117 kDa 97 kDa 66 kDa 45 kDa 31 kDa 21.5 kDa B C D 45 123 45 12 3 45 12 345 Fig. 6. (A, B) Western blot analysis of the lysate of vitelline membrane of a laid egg. The SDS-solubilized vitelline membrane (1 lg per lane) isolated from the laid egg of a Japanese quail (lane 1), blue-breasted quail (lane 2), chicken (lane 3), turkey (lane 4) and guinea fowl (lane 5) were subjected to wes- tern blot analysis with anti-ZP3 immunoglob- ulin (A) or anti-ZP1 immunoglobulin (B). Representative blots of three experiments are shown. (C, D) Ten micrograms of pro- tein from the vitelline membrane of laid eggs from a Japanese quail (lane 1), blue- breasted quail (lane 2), chicken (lane 3), tur- key (lane 4) and guinea fowl (lane 5) were subjected to SDS-PAGE and detected with CBB staining (C), or by ligand blotting and probing with radiolabeled ZP3 (D). Results representative of three experiments are shown. M. Kinoshita et al. Incorporation of ZP1 into perivitelline membrane FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS 3585 recovered from SDS-PAGE gels. Although the final fate of the injected PL ZP1 was not elucidated, this unexpected result indicates the possibility that the properties of ZP1 necessary for incorporation into the PL might change after the serum ZP1 arrives at the PL. The authors of a recent study demonstrated that C-terminal proteolytic processing of the egg envelope protein of rainbow trout (Oncorhynchus mykiss), VEa,VEb and VEc, which are synthesized by the liver and transported in the bloodstream to the ovary, occurs after the arrival of precursor pro- teins at the egg [22]. These authors did not determine the source of the enzyme responsible for the cleavage of VE protein in the ovary; however, they speculated that the enzyme is associated with the oocyte plasma membrane close to the innermost layer of the egg envelope into which nascent VE proteins are incorpo- rated. In an analogous situation, we also showed that N-terminal proteolytic processing of quail ZP1 might take place after the arrival of ZP1 at the ovary, and the resulting products – the 46-kDa protein as well as the cleaved N-terminal fragments of ZP1 – are incor- porated into the PL [23]. Although the question remains unanswered of whether such structural changes have relevance regarding the capability of serum ZP1 to be incorporated into the PL, efforts are currently underway to reveal the differences between the serum ZP1 and the PL ZP1. In addition to these observations, we found that the accumulation of DIG-labeled ZP1 tended to be high in mature follicles (e.g. the largest and the second-largest follicles) com- pared with that in the third-largest follicles, and the signal intensity was belowthe detection level when we analysed the PL lysates isolated from the small yellow follicles (data not shown). This pattern is similar to that of the accumulation of ZP1 in the PL during fol- licular development, in that the ZP1 was first detected as a 97-kDa protein by means of western blot analy- sis when the granulosa layer was isolated from the fourth-largest follicle, and the intensity of the band was dramatically increased after the follicle matured to being the third-largest one [24]. Taken together, these observations strongly suggest that the data obtained in the current study reflect the events that occur in vivo for the formation of the insoluble fibers of the PL. Our results also provide evidence, for the first time, that serum ZP1 can interact with both ZP3 and ZP1 in the PL of Japanese quail. Our previous study demon- strated that the accumulation of ZP1 was not synchro- nized with that of ZP3 in the PL during follicular development [24]. Morever, the accumulation of ZP3 in the PL precedes that of ZP1 during follicular devel- opment in the quail ovary. Based on these observa- tions, we suggest that the accretion of ZP3 protein on the surface of the oocyte by an unknown mechanism might trigger the binding of serum ZP1 to the PL, and that the accumulation of ZP1 initiates the formation of a ZP1–ZP1 complex in addition to the ZP3–ZP1 heterocomplex in vivo. Although direct evidence is not available at present, the presence of two forms of ZP1 in the SDS polyacrylamide gels under non-reducing conditions, that is, monomeric (i.e. ZP1, which might interact with ZP3 in the PL) and dimeric (i.e. ZP1, which binds with ZP1 itself) ZP1s, could support this possibility. In accordance with this expectation, the monomeric ZP1 first appeared as a dominant band in the lysate of the third-largest follicles when the exoge- nous ZP1 in the lysates was detected using anti-DIG immunoglobulin (Fig. 3B). In the present study, all the vitelline membrane isolated from the eggs of various birds contained ZP3 and ZP1 homologue (Fig. 6A,B). However, when the lysates were analyzed using ligand blotting and probed with radiolabeled quail ZP3, a distinct difference in the intensity of the ZP1 band was observed between the Japanese quail eggs and the chicken and guinea fowl eggs, whereas comparable intensity was obtained in the case of turkey ZP1 (Fig. 6D). These results indicate that the affinity of quail ZP3 for chicken and guinea fowl ZP1 might be weak. Although the direct relationship is not known, the amino acid sequence similarity of ZP1 between quail and turkey is higher than that of quail and chicken (quail versus turkey, 91.4%; and quail ver- sus chicken, 87.6%). By contrast, when we injected intravenously the chicken or guinea fowl ZP1 purified from the serum of a laying female, a signal of ZP1, simi- lar in intensity to that of Japanese quail, was seen in the lysate of the PL of the largest follicles of the birds (Fig. 7B). These results indicate that the serum ZP1 purified from different species of birds could incorpo- rate into the PL in a manner similar to that of Japanese quail. Although we did not perform a quantitative anal- ysis, this process appears to be different from that of the ZP3–ZP1 interaction. Although we did not compare the differences in binding affinity of ZP1s for ZP3 between quail and other birds, in light of these facts, we consider that the interaction between ZP3 and ZP1 is relatively species-specific, whereas the specificity for ZP1–ZP1 binding is comparatively low. In a previous study we demonstrated that N-linked glycans on ZP1 play an important role in triggering the acrosome reac- tion in Japanese quail, whereas ZP3 failed to induce the acrosome reaction at any concentration tested [25]. On the other hand, Stewart et al. [26] observed the inter- action of chicken spermatozoa with the PL from Incorporation of ZP1 into perivitelline membrane M. Kinoshita et al. 3586 FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS different avian species in vitro by counting the number of holes on the PL produced by the hydrolase from the spermatozoa. They found that the number of holes on the PL of Galliformes, including turkey, quail and guinea fowl, produced by chicken spermatozoa, was equal to or greater than 100% of that observed on chicken PL. The induction of the acrosome reaction of the spermatozoa, probably caused by the effect of ZP1 in birds, also did not display precise species specificity in avian species. We would not rule out a role of the ZP3–ZP1 complex during fertilization as well as in the formation of the PL in avian species; this heterocomplex in the PL could exert a specific function in these events. Recent transgenic experiments with null mice suggested that the binding of sperm depends on the supramolecu- lar structure on the zona pellucida, not on an individual glycoprotein [27], although the machinery underlying the sperm–egg binding in mice is under debate [28]. The question of whether the ZP3–ZP1 complex, as well as the supramolecular structure of the PL, is responsible for these events in birds remains to be resolved. Materials and methods Preparation of birds and of tissue Female Japanese quail, Coturnix japonica, 15–30 weeks of age (Tokai-Yuki, Toyohashi, Japan), were maintained indi- vidually under a photoperiod of 14 h light : 10 h dark (with the light on at 05:00 h) and were provided with water and a commercial diet (Tokai-Hokuriku Nosan, Chita, Japan) ad libitum. Female chickens (Gallus gallus), turkeys (Melea- gris gallopavo) and guinea fowl (Numida meleagris) were housed in a room under a lighting schedule of 14 h light and 10 h darkness and were provided with tap water and a commercial diet (Tokai-Kigyo, Toyohashi, Japan) ad libi- tum. The birds were killed by bleeding from the carotid artery, and the serum and pre-ovulatory follicles were collected. The laid eggs of blue-breasted quail (Cotur- nix chinensis ) were generous gifts from T. Ono (Shinshu University). The granulosa layer of Japanese quail was iso- lated as a sheet of granulosa cells sandwiched between the PL and the basal laminae, as previously described [29]. The PL was isolated using a procedure described by Sasanami et al. [30]. All experimental procedures for the use and the care of birds in the present study were approved by the Animal Care Committee of Shizuoka University (approval number 19-13). Purification of ZP1 The isolated PL was dissolved overnight at room tempera- ture in 1% SDS, which was buffered at pH 6.8 with 70 mm Tris–HCl. After centrifugation at 10 000 g for 10 min, the supernatants were used as a PL lysate, and the protein con- centrations of the samples were measured using a BCA pro- tein assay kit (Pierce, Rockford, IL, USA). The PL lysate was separated on 1D SDS-PAGE, performed as described by Laemmli [31], under non-reducing conditions with 12% polyacrylamide as the separating gel. The samples (750 lg of protein per gel) were applied to a 5% stacking gel, with- out a comb, for lane casting. After electrophoresis, the gel was stained with Copper Stain (Bio-Rad Laboratories, Hercules, CA, USA), and the 97-kDa (monomeric ZP1) band was excised. The proteins were eluted by incubating the gel slices in 0.1% SDS buffered at pH 8.0 with 100 mm Tris–HCl, overnight at 4 °C with constant shaking. The eluent was then extensively dialyzed against water, lyophi- lized and dissolved in NaCl⁄ P i . The protein concentrations of the samples were measured as described above, using a BCA protein assay kit (Pierce). To prepare the affinity gel for the separation of serum ZP1, the IgG fractionated from anti-ZP1 serum [32], using a HiTrap Protein A FF affinity column (Amersham Pharma- cia Biotech, Piscataway, NJ, USA), was covalently coupled to 3-o-succinyl-s-aminocaproic acid-N-hydroxy-succinimide ester (NHS)-activated sepharose (Amersham Pharmacia Biotech) according to the manufacturer’s instructions. The serum of laying birds was incubated with the affinity gel for 16 h at 4 °C. After extensive washing with NaCl ⁄ P i , the gel was eluted with elution buffer (1 m CH 3 COOH, 0.1 m m glycine, pH 2.5) and the eluent containing serum ZP1 was applied to a 5% stacking gel, without a comb, for lane casting. The 97-kDa serum ZP1 was purified using the same procedure as that used to purify PL ZP1. The purity of the ZP1 proteins was confirmed using silver staining after separation of the protein by SDS-PAGE. Labeling and administration of ZP1 The purified serum ZP1 was labeled with DIG using a DIG protein-labeling kit (Roche, Penzberg, Germany) according to the manufacturer’s instructions. Briefly, the purified serum ZP1 (100 lg), dissolved in 1 mL of NaCl ⁄ P i , was mixed with 100 lL of labeling solution containing 70 lg of digoxigenin (DIG)-NHS, and the mix- ture was incubated at room temperature with constant agitation. Non-reacted DIG-NHS in the mixture was sepa- rated by gel filtration on a prepared Sephadex G-25 col- umn. The labeled proteins were pooled and stored at )80 °C until required for use. The purified PL ZP1 was also labeled with DIG using the same procedure for the labeling of the serum ZP1. The protein concentration of the DIG-labeled ZP1 was measured, as described above, using a BCA protein assay kit (Pierce). The birds were injected intravenously with 10 lg of the DIG-labeled serum or with PL ZP1 dissolved in NaCl ⁄ P i . The volume of NaCl ⁄ P i injected was kept at 0.1 mL per 100 g of body weight. Six hours after injection the birds M. Kinoshita et al. Incorporation of ZP1 into perivitelline membrane FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS 3587 were decapitated and the pre-ovulatory follicles were dis- sected. The PL was isolated as described above. Gel electrophoresis and western blot analysis SDS-PAGE was carried out under non-reducing conditions as described previously [31], using 12 and 5% polyacryl- amide for resolving and stacking gels, respectively. For wes- tern blotting, proteins separated on SDS-PAGE were transferred to a PVDF membrane (Immobilon-P; Millipore Bedford, MA, USA) [33]. The membrane was reacted with anti-ZP1 serum (1 : 10 000 dilution) or anti-ZP3 serum (1 : 10 000 dilution) [34] and visualized by means of a chemiluminescent technique (Amersham Pharmacia Bio- tech) using horseradish peroxidase-conjugated anti-rabbit IgG (1 : 16 000 dilution; Cappel, Durham, NC, USA) as a secondary antibody. To detect the exogenously injected ZP1, the membrane was incubated with horseradish peroxi- dase-conjugated anti-DIG immunoglobulin (1 : 1000 dilu- tion; Roche), and the bands were visualized as described above. Ligand blot analysis To identify the binding partner for serum ZP1 in the PL, the PL lysate (6 lg of protein) was separated on SDS- PAGE and transferred onto PVDF membrane, as described above. Non-specific binding was inhibited by incubating the membrane with 5% non-fat skim milk in saline buffered at pH 7.4 with 10 mm Tris–HCl containing 0.1% Tween 20 (blocking buffer), and the membrane was incubated with 0.006 lg of DIG-labeled serum ZP1, DIG-labeled bovine serum albumin in blocking buffer, or blocking buffer alone, overnight at room temperature. After washing several times, the DIG-labeled proteins were detected using horse- radish peroxidase-conjugated anti-DIG immunoglobulin (1 : 1000 dilution; Roche), and the bands were visualized as described above, using a chemiluminescent technique (Amersham Pharmacia Biotech). To detect the binding of quail ZP3 to the ZP1 in the laid eggs of various species, we performed a ligand blot analysis using radiolabeled ZP3, as described previously [18,19]. Immunofluorescence microscopy For localization of exogenously injected ZP1 in the follicle, the largest pre-ovulatory follicles were dissected, fixed in Bouin’s fixative and embedded in Paraplast (Oxford Lab- ware, St Louis, MO, USA). Immunohistochemical tech- niques using fluorescein isothiocyanate (FITC)-conjugated anti-DIG immunoglobulin (1 : 300 dilution; Roche) were as described previously [24]. The immunolabeled sections were examined under a fluorescence microscope (BX50; Olym- pus, Tokyo, Japan), equipped with a fluorescence mirror unit (U-MNIB2; Olympus). Acknowledgements The authors are grateful to Miss Maki Joho for techni- cal assistance. This work was supported, in part, by a Grant-in-Aid for Scientific Research (18780210 to T. S.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. References 1 Bellairs R, Harkness M & Harkness RD (1963) The vitel- line membrane of the hen’s egg: a chemical and electron microscopical study. J Ultrastruct Res 8, 339–359. 2 Bain JM & Hall JM (1969) Observations on the devel- opment and structure of the vitelline membrane of the hen’s egg: an electron microscope study. Aust J Biol Sci 22, 653–665. 3 Wyburn GM, Aitken RNC & Johnston HS (1965) The ultrastructure of the zona radiata of the ovarian follicle of the domestic fowl. J Anat 99, 469–484. 4 Spargo SC & Hope RM (2003) Evolution and nomen- clature of the zona pellucida gene family. Biol Reprod 68, 358–362. 5 Conner SJ, Lefievre LL, Hughes DC & Barratt CLR (2005) Cracking the egg: increased complexity in the zona pellucida. Hum Reprod 20, 1148–1152. 6 Smith J, Paton IR, Hughes DC & Burt DW (2005) Iso- lation and mapping the chicken zona pellucida genes: an insight into the evolution of orthologous genes in different species. Mol Reprod Dev 70, 133–145. 7 Hughes DC (2007) ZP genes in avian species illustrate the dynamic evolution of the vertebrate egg envelope. Cytogenet Genome Res 117, 86–91. 8 Monne M, Han L & Jovine L (2006) Tracking down the ZP domain: from the mammalian zona pellucida to the molluscan vitelline envelope. Semin Reprod Med 24, 204–216. 9 Lefievre L, Conner SJ, Salpekar A, Olufowobi O, Ash- ton P, Pavlovic B, Lenton W, Afnan M, IBrewis IA, Monk M et al. (2004) Four zona pellucida glycopro- teins are expressed in the human. Hum Reprod 19, 1580–1586. 10 Ganguly A, Sharma RK & Gupta SK (2008) Bonnet monkey (Macaca radiata) ovaries, like human oocytes, express four zona pellucida glycoproteins. Mol Reprod Dev 75, 156–166. 11 Hoodbhoy T, Joshi S, Boja ES, Williams SA, Stanley P & Dean J (2005) Human sperm do not bind to rat zonae pellucidae despite the presence of four homologous glycoproteins J Biol Chem 280, 12721– 12731. Incorporation of ZP1 into perivitelline membrane M. Kinoshita et al. 3588 FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS 12 Bork P & Sander C (1992) A large domain common to sperm receptors (ZP2 and ZP3) and TGF-b type III receptor. FEBS Lett 300, 237–240. 13 Mori M & Masuda N (1993) Proteins of the vitelline membrane of quail (Coturnix coturnix japonica) eggs. Poult Sci 72, 1566–1572. 14 Sasanami T, Pan J & Mori M (1999) FSH enhanced production of ZPC homologue of inner perivitelline membrane by quail granulosa cells. J Poult Sci 36, 343– 353. 15 Pan J, Sasanami T & Mori M (2003) Stimulation of ZPC production by follicle-stimulating hormone in the granulosa cells of Japanese quail (Coturnix japonica). J Poult Sci 40, 202–211. 16 Pan J, Sasanami T, Kono Y, Matsuda T & Mori M (2001) Effects of testosterone on production of perivitel- line membrane glycoprotein ZPC by granulosa cells of Japanese quail (Coturnix japonica). Biol Reprod 64, 310–316. 17 Sasanami T, Pan J & Mori M (2003) Expression of perivitelline membrane glycoprotein ZP1 in the liver of Japanese quail (Coturnix japonica) after in vivo treat- ment with diethylstilbestrol. J Steroid Biochem Mol Biol 84, 109–116. 18 Ohtsuki M, Hanafy AM, Mori M & Sasanami T (2004) Involvement of interaction of ZP1 and ZPC in the for- mation of quail perivitelline membrane. Cell Tissue Res 318, 565–570. 19 Sasanami T, Ohtsuki M, Ishiguro T, Matsushima K, Hiyama G, Kansaku N, Doi Y & Mori M (2006) Zona pellucida domain of ZPB1 controls specific binding of ZPB1 and ZPC in Japanese quail (Coturnix japonica). Cell Tissue Organ 183, 41–52. 20 Okumura H, Aoki N, Sato C, Nadano D & Matsuda T (2007) Heterocomplex formation and cell-surface accu- mulation of Hen’s serum zona pellucida B1 (ZPB1) with ZPC expressed by a mammalian cell line (Cos-7): a pos- sible initiating step of egg-envelope matrix construction. Biol Reprod 76, 9–18. 21 Okumura H, Okajima T, Nadano D & Matsuda T (2007) Association of chicken zona pellucida glycopro- tein (ZP) B1 with ZPC induces formation of ZPB1-ZPC fibrous aggregates containing disulfide-bridged ZPB1 dimer. Biochem Biophys Res Commun 364, 682–688. 22 Darie CC, Biniossek ML, Gawinowicz MA, Milgrom Y, Thumfart JO, Jovine L, Litscher ES & Wassarman PM (2005) Mass spectrometric evidence that proteolytic processing of rainbow trout egg vitelline envelope pro- teins takes place on the egg. J Biol Chem 280, 37585– 37598. 23 Sasanami T, Ohtsuki M, Doi Y, Ryota T, Fujishima A & Mori M (2008) Analysis of 46-kDa protein in the perivitelline membrane of Japanese quail (Coturnix japonica). Anim Sci J 79, 104–110. 24 Sasanami T, Ohtsuki M, Hanafy AM & Mori M (2004) Accumulation of ZP1 and ZPC in quail perivitelline membrane during follicular development. J Poult Sci 41, 289–293. 25 Sasanami T, Murata T, Ohtsuki M, Matsushima K, Hiyama G, Kansaku N & Mori M (2007) Induction of sperm acrosome reaction by perivitelline glycoprotein ZP1 in Japanese quail (Coturnix japonica). Reproduction 133, 41–49. 26 Stewart SG, Bausek N, Wohlrab F, Schneider WJ, Hor- rocks A & Wishart G (2004) Species specificity in avian sperm: perivitelline interaction. Comp Biochem Physiol A 137, 657–663. 27 Hoodbhoy T & Dean J (2004) Insights into the mole- cular basis of sperm-egg recognition in mammals. Reproduction 127, 417–422. 28 Wassarman PM (2005) Contribution of mouse egg zona pellucida glycoproteins to gamete recognition during fertilization. J Cell Physiol 204, 388–391. 29 Gilbert AB, Evans AJ, Perry MM & Davidson MH (1977) A method for separating the granulosa cells, the basal lamina and the theca of the preovulatory ovarian follicle of the domestic fowl (Gallus domesticus). J Reprod Fertil 50, 179–181. 30 Sasanami T, Pan J, Doi Y, Hisada M, Kohsaka T & Toriyama M (2002) Secretion of egg envelope protein ZPC after C-terminal proteolytic processing in quail granulosa cells. Eur J Biochem 269, 2223–2231. 31 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. 32 Ohtsuki M, Hiyama G, Kansaku N, Ogawa H, Mori M & Sasanami T (2008) Cloning of perivitelline membrane protein; ZP1 in turkey (Meleagris gallopavo). J Poult Sci 45, 67–74. 33 Matsudaira P (1987) Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J Biol Chem 262, 10035–10038. 34 Kuroki M & Mori M (1995) Origin of 33 kDa protein of vitelline membrane of quail egg: immunological studies. Develop Growth Differ 37, 545–550. M. Kinoshita et al. Incorporation of ZP1 into perivitelline membrane FEBS Journal 275 (2008) 3580–3589 ª 2008 The Authors Journal compilation ª 2008 FEBS 3589 . Incorporation of ZP1 into perivitelline membrane after in vivo treatment with exogenous ZP1 in Japanese quail (Coturnix japonica) Mihoko Kinoshita 1 ,. (2003) Expression of perivitelline membrane glycoprotein ZP1 in the liver of Japanese quail (Coturnix japonica) after in vivo treat- ment with diethylstilbestrol.