CD9 and CD81 Interactions and Their Structural Modelling in ... - MDPI

International Journal of

Molecular Sciences Article

CD9 and CD81 Interactions and Their Structural Modelling in Sperm Prior to Fertilization Michaela Frolikova 1 , Pavla Manaskova-Postlerova 1,2 , Jiri Cerny 3 ID , Jana Jankovicova 4 Ondrej Simonik 1,2 , Alzbeta Pohlova 1,5 , Petra Secova 4 , Jana Antalikova 4 ID and Katerina Dvorakova-Hortova 1,6, * 1

2 3 4

5 6

*

ID

,

Group of Reproductive Biology, Institute of Biotechnology, Czech Academy of Sciences, v.v.i., BIOCEV, Prumyslova 595, 252 50 Vestec, Czech Republic; [email protected] (M.F.); [email protected] (P.M.-P.); [email protected] (O.S.); [email protected] (A.P.) Department of Veterinary Sciences, Faculty of Agrobiology, Food and Natural Resources, University of Life Sciences Prague, Kamycka 129, 165 00 Prague, Czech Republic Laboratory of Structural Bioinformatics of Proteins, Institute of Biotechnology Czech Academy of Sciences, v.v.i., BIOCEV, Prumyslova 595, 252 50 Vestec, Czech Republic; [email protected] Laboratory of Reproductive Physiology, Institute of Animal Biochemistry and Genetics Centre of Biosciences Slovak Academy of Sciences, Dubravska Cesta 9, 845 05 Bratislava, Slovakia; [email protected] (J.J.); [email protected] (P.S.); [email protected] (J.A.) Department of Biochemistry, Faculty of Science, Charles University, Hlavova 2030/8, 128 43 Prague, Czech Republic Department of Zoology, Faculty of Science, Charles University, Vinicna 7, 128 44 Prague, Czech Republic Correspondence: [email protected]; Tel.: +420-325873799

Received: 28 March 2018; Accepted: 13 April 2018; Published: 19 April 2018







Abstract: Proteins CD9 and CD81 are members of the tetraspanin superfamily and were detected in mammalian sperm, where they are suspected to form an active tetraspanin web and to participate in sperm–egg membrane fusion. The importance of these two proteins during the early stages of fertilization is supported by the complete sterility of CD9/CD81 double null female mice. In this study, the putative mechanism of CD9/CD81 involvement in tetraspanin web formation in sperm and its activity prior to fertilization was addressed. Confocal microscopy and colocalization assay was used to determine a mutual CD9/CD81 localization visualised in detail by super-resolution microscopy, and their interaction was address by co-immunoprecipitation. The species-specific traits in CD9 and CD81 distribution during sperm maturation were compared between mice and humans. A mutual position of CD9/CD81 is shown in human spermatozoa in the acrosomal cap, however in mice, CD9 and CD81 occupy a distinct area. During the acrosome reaction in human sperm, only CD9 is relocated, compared to the relocation of both proteins in mice. The structural modelling of CD9 and CD81 homologous and possibly heterologous network formation was used to propose their lateral Cis as well as Trans interactions within the sperm membrane and during sperm–egg membrane fusion. Keywords: CD9; CD81; tetraspanin network; sperm; membrane fusion; capacitation; acrosome reaction; fertilization; structural modelling; mouse; human

1. Introduction CD9 and CD81 are expressed in a large variety of cells [1] and belong to the tetraspanin superfamily (TM4SF), whose members are small (20–50 kDa) proteins [2,3]. Their ability to form homologous partnerships as well as interact heterologously with distinct, non-tetraspanin proteins (including adhesion molecules, receptor and co-receptor molecules, and antigens of major Int. J. Mol. Sci. 2018, 19, 1236; doi:10.3390/ijms19041236

www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2018, 19, 1236

2 of 17

histocompatibility complex or cytoplasmic kinases) represents the key feature of tetraspanins that enables them to create a complex active network on a cell membrane surface called a “tetraspanin web” [4–6]. Consequently, tetraspanins are generally viewed as “molecular facilitators” that interact and bring into close proximity specific proteins involved in processes of cell activation and transduction, in particular of cellular development, proliferation, activation, and motility of somatic cells [7]. In a tetraspanin web, very relevant partners of CD9 and CD81 are integrins, mainly β1 integrins. These molecules together create integrin-tetraspanin adhesion complexes and tetraspanins play important role in integrin signalling [8]. In somatic cells, the association of CD9 and CD81 with α3β1 [8,9] and α6β1 integrins was confirmed [10–12], and a similar association can be expected in sperm, as the presence and favourable localization of these integrin heterodimers in sperm head was shown by Frolikova et al. [13]. Although the majority of up-to-date known molecular interactions mediated by tetraspanins were described within a single cell membrane (Cis interaction) [4,14,15], a “partnership” of tetraspanins with proteins in distinct cell membranes (Trans interaction) was also reported [16]. Kovalenko et al. [17] published the existence of both homo- and hetero-dimers of CD9 and CD81. Equally homo-and heterophilic reciprocal interactions of tetraspanins have been described [14,18]. Except tetraspanin-protein interactions, tetraspanin-lipid interactions play a crucial role in tetraspanin web organization and function. Therefore, it is highly probable that the proteins expressed on sperm, respectively gametes, are also organized in a tetraspanin web that is active prior and during fertilization, which is supported by the expression of tetraspanin molecules CD9 and CD81 on mouse and bull sperm [19,20]. The crucial role of CD9 and CD81 in fertilization is emphasized by the sterility of CD9/CD81 double null female mice [21]. In contrary the individual CD9 [11,22] or CD81 [21] knock-out female mice displayed reduced fertility when mated with males. On the other hand, both knock-out male mice lacking CD9 or CD81 were fertile [21–23]. The published modulation of the possible molecular mechanism of CD81 activity [24] contributed to a better explanation of fertilization events in mammals using a computational approach. Although the proposed model was matched for the CD81 molecule, the cholesterol binding principle can be considered as a key moment for switching between the active and non-active position and therefore binding of a partner protein. Follow-up association of tetraspanins with membrane-curving proteins or lipids, or different tetraspanins with abilities to bind these molecules, could influence the curvature of the cell membrane and formation of various tubular structures (e.g., microvilli) associated with cell adhesion and intercellular communication [25]. The changes of microvilli distribution and their CD9 content were proposed to be responsible for the development of the oocyte membrane block to sperm penetration [26] and female infertility of CD81-deficient mice due to less outward curvature of the oolema was also predicted [25]. The membrane tubular structures formation could be regulated by associations of tetraspanins with actin cytoskeleton via ERM proteins [25]. In this study, the putative mechanism of involvement of CD9 and CD81 in tetraspanin web formation and its activity during sperm preparation for fertilization was addressed with a focus on the species-specific differences between mouse and human. The distribution of CD9 and CD81 molecules in mature sperm and the differences in behaviour of CD81 in the acrosomal exocytosis are described. The structural modelling of CD9 and CD81 homologous and heterologous network formation discussing both lateral Cis interaction within the sperm membrane and Trans interaction during gamete fusion was used to propose the tetraspanin network structure. 2. Results 2.1. Mutual Position of CD9 and CD81 on Mouse Sperm Head To understand the relationship of CD9 and CD81 proteins on sperm and their role in the fertilization process, we investigated their mutual position on mouse sperm head in freshly released epididymal, capacitated and acrosome reacted sperm (Figure 1). The localization of CD81 on plasma

Int. J. Mol. Sci. 2018, 19, 1236

3 of 17

Int. J. Mol. Sci. 2018, 19, x FOR PEER REVIEW

3 of 17

membrane covering apical acrosome and CD9 on the acrosomal membrane and relocalization of both tetraspanins into equatorial segmentsegment during acrosome reaction (AR) have been described for both tetraspanins into equatorial during acrosome reaction (AR)already have been already each of these individually However, they However, have neverthey beenhave depicted described for proteins each of these proteins[19,20]. individually [19,20]. neversimultaneously been depicted on sperm to address their mutual behaviour. Our results of dual immunofluorescent labelling confirm simultaneously on sperm to address their mutual behaviour. Our results of dual immunofluorescent that CD81 is expressed on the plasma membrane covering the apical acrosome (see green arrow labelling confirm that CD81 is expressed on the plasma membrane covering the apical acrosome (see in Figure 1, line I), and is and present the acrosomal membranes (see red arrows inarrows Figure in 1, green arrow in Figure 1, CD9 line I), CD9on is present on the acrosomal membranes (see red line I.) in mouse epididymal sperm. Our data showed the presence of CD9 in both outer and inner Figure 1, line I.) in mouse epididymal sperm. Our data showed the presence of CD9 in both outer acrosomal membranesmembranes using confocal (see red arrows in Figure 1, in lineFigure I) and1,structure illumination and inner acrosomal using confocal (see red arrows line I) and structure microscopy (SIM) (Figure(SIM) 2). SIM was further visualise the accurate position illumination microscopy (Figure 2). SIMused was to further usedintodetail visualise in detaildual the accurate of CD9 and CD81. In contrary to CD9, the CD81 protein is localized in the apical plasma membrane dual position of CD9 and CD81. In contrary to CD9, the CD81 protein is localized in the apical coveringmembrane the acrosomal regionthe in mouse epididymal (Figure 2A). To visualise the localization plasma covering acrosomal region sperm in mouse epididymal sperm (Figure 2A). To of CD9, CD46, which is known to be present in both acrosomal membranes, was used as a marker. visualise the localization of CD9, CD46, which is known to be present in both acrosomal membranes, The localization of CD9 correlates with CD46 using SIM (Figure 2B) and confirms their mutual position. was used as a marker. The localization of CD9 correlates with CD46 using SIM (Figure 2B) and We achieved resolution 120 for the axial greenresolution channel and the green red channel the confirms theiraxial mutual position. Wenm achieved 120 140 nm for the channelusing and 140 SIM method. for the red channel using the SIM method. mouse sperm sperm capacitation capacitation (Figure (Figure 1, 1, The localization of CD9 and CD81 does not change during mouse line II), but both proteins relocate to the equatorial segment and partially over the postacrosomal region during AR. Moreover, they start to appear close to each other in the equatorial segment after AR (see arrows in Figure Figure 1, 1, line line III). III). Three-dimensional Three-dimensional colocalization maps based on Pearson’s correlation coefficient the colocalization area (Figure 3). correlation coefficientwere werecreated createdfor formore moreaccurate accuratevisualisation visualisationofof the colocalization area (Figure NoNo colocalization of CD9 andand CD81 was detected in epididymal sperm (Figure 3A), butbut thethe areas of 3). colocalization of CD9 CD81 was detected in epididymal sperm (Figure 3A), areas mutual colocalization were present in sperm head after ARAR (Figure 3B).3B). of mutual colocalization were present in sperm head after (Figure

Figure 1. Dynamics their mutual mutual localization localization in mouse sperm sperm Figure 1. Dynamics of of CD9 CD9 and and CD81 CD81 and and changes changes in in their in mouse captured by confocal microscopy. (I) In freshly released epididymal mouse sperm, CD9 (red) is captured by confocal microscopy. (I) In freshly released epididymal mouse sperm, CD9 (red) is locked locked over the acrosome vesicle in both of its membrane compartment—inner and outer acrosome over the acrosome vesicle in both of its membrane compartment—inner and outer acrosome membrane membrane and CD81 (green) is present in plasma membrane the apical acrosome (red arrows)(red andarrows) CD81 (green) is present in plasma membrane over the apicalover acrosome (green arrow). (green arrow). (II) During no changesofin localization of both proteins see both red (II) During capacitation no capacitation changes in localization both proteins occur, see both redoccur, and green arrows. and green arrows. (III) After finishing of AR, plasma membrane covering acrosome apical area and (III) After finishing of AR, plasma membrane covering acrosome apical area and outer acrosomal outer acrosomal membrane lost. However, CD9detectable. and CD81CD9 are and still CD81 detectable. CD9 and membrane are lost. However,are both CD9 and CD81both are still are relocated in CD81 are relocated in inner and acrosomal membrane andofinequatorial plasma membrane of equatorial inner acrosomal membrane in plasma membrane segment (red and green segment arrows). (red and arrows). They DAPI start in mutual contact. DAPI (blue). Scale bar represents 1 μm. They startgreen in mutual contact. (blue). Scale bar represents 1 µm.

Int. J. Mol. Sci. 2018, 19, 1236

4 of 17

Int. J. Mol. Sci. 2018, 19, x FOR PEER REVIEW

4 of 17

Int. J. Mol. Sci. 2018, 19, x FOR PEER REVIEW

4 of 17

Figure 2. Precise localization of CD9 within the inner and outer acrosomal membranes depicted by

Figure 2. Precise localization of CD9 within the inner and outer acrosomal membranes depicted by structure illumination microscopy (SIM) in mouse epididymal sperm head. (A) SIM data confirm Figure 2. Precise localization of CD9 within the inner and outer acrosomal membranes depicted by structure illumination (SIM) in mouse sperm (A) SIM data localization of CD81microscopy (green) in plasma membrane of epididymal apical acrosome area head. and localization of CD9confirm in structure illumination microscopy (SIM) in mouse epididymal sperm area head.and (A) localization SIM data confirm localization of CD81 (green) in plasma membrane of apical acrosome of CD9 in acrosomal membrane and absence of their mutual contact in epididymal sperm; (B) CD46 (red) was localization of CD81 (green) in plasma membrane apical acrosome area and localization of CD9 in was acrosomal membrane and absence ofof their mutual contact in epididymal sperm; (B) CD46 (red) used as marker for confirmation presence ofof CD9 (red) in both outer and inner acrosomal acrosomal and absence of their mutual contact in epididymal sperm; (B) CD46 (red) was used membranes. as marker membrane for confirmation of presence of CD9 (red) in both outer and inner acrosomal membranes. DAPI (blue). Scale bar represents 5 μm. used as marker for confirmation of presence of CD9 (red) in both outer and inner acrosomal DAPI (blue). Scale bar represents 5 µm. membranes. DAPI (blue). Scale bar represents 5 μm.

Figure 3. Surface rendering of colocalization map of CD9 and CD81 in mouse epididymal and acrosome reacted sperm. Confocal images and Huygens software were used for better visualisation Figure 3. Surface rendering of colocalization map of CD9 and CD81 in mouse epididymal and colocalization area (yellow) of CD9 (red) and CD81 (green). No in colocalization area wasand detected Figure 3. Surface rendering of colocalization map of CD9 and(A) CD81 mouse epididymal acrosome acrosome reacted sperm. Confocal images and Huygens software were used for better visualisation on sperm. freshly released epididymal sperm contrary to the acrosome reacted spermvisualisation (B), where thecolocalization analysis reacted Confocal images and Huygens software were used for better colocalization area (yellow) of CD9 of (red) and CD81 (green). (A) No colocalization area was detected confirmed a mutual colocalization CD9 and CD81. Colocalization are based Pearson’s area (yellow) ofreleased CD9 (red) and CD81 (green). (A) No colocalization areamaps was detected onon freshly released on freshly epididymal sperm contrary to the acrosome reacted sperm (B), where the analysis correlation coefficient. DAPI (blue) Scale bar represents 2 μm. epididymal sperm contrary to the acrosome sperm (B), where maps the analysis confirmed a mutual confirmed a mutual colocalization of CD9reacted and CD81. Colocalization are based on Pearson’s colocalization of CD9 and CD81. Colocalization maps are based on Pearson’s correlation coefficient. correlation coefficient. DAPI (blue) Scaleinbar represents 2 μm. 2.2. Mutual Localization of CD9 and CD81 Human Sperm Head

DAPI (blue) Scale bar represents 2 µm. In human ejaculated sperm, the presence CD9 and CD81 was detected in the apical 2.2. Mutual Localization of CD9 and CD81 in Human of Sperm Head acrosomal area (Figure 4, line I), and CD81was also partiality present in the post-acrosomal area. In 2.2. Mutual of CD9 sperm, and CD81 Human Sperm In Localization human ejaculated theinpresence of CD9Head and CD81 was detected in the apical ejaculated human sperm, both proteins are localized within the acrosomal cap but there are acrosomal area (Figure 4, line I), and CD81was also partiality present in the post-acrosomal area. In differences in the labelling pattern of CD81 andand CD9. While CD81 creates In human ejaculated sperm, the presence of CD9 CD81 was detected in thevery apicalspecific acrosomal ejaculated human sperm, both proteins are localized within the acrosomal cap but there are non-homogenous dotted pattern, CD9 ispartiality evenly distributed (Figure 4). These findings suggest that area (Figure 4, line I), and CD81was also present in the post-acrosomal area. In ejaculated differences in the labelling pattern of CD81 and CD9. While CD81 creates very specific CD81 could be accumulated in clusters favouring their homophilic interactions. No significant human sperm, both proteins are localized withindistributed the acrosomal cap there are differences non-homogenous dotted pattern, CD9 is evenly (Figure 4).but These findings suggest thatin the changes in protein localization occurred during capacitation process (Figure 4, line II). When AR is labelling pattern of CD81 and CD9. While CD81 creates very specific non-homogenous pattern, CD81 could be accumulated in clusters favouring their homophilic interactions. No dotted significant completed, the plasma membrane covering the apical acrosomal area and the outer acrosomal CD9 is evenly distributed (Figure 4). These findings suggest that CD81 could be accumulated in clusters changes in protein localization occurred during capacitation process (Figure 4, line II). When AR is membrane are both lost. The fluorescent signal of CD9 remains in the equatorial segment (see red completed, homophilic the plasma interactions. membrane covering the apical acrosomal area and the outeroccurred acrosomal favouring No significant changes protein localization during arrowtheir in Figure 4, line III), but CD81 disappears from the apical in area of the acrosomal cap. However, membrane are both lost. The fluorescent signal ofisCD9 remains in the equatorial segment covering (see red the capacitation process (Figure 4, line II). When AR completed, the plasma membrane it stays detectable in the postacrosomal region of human sperm head (Figure 4, line III). As in the arrow in Figure 4, line III), but CD81 disappears from the apical arealost. of theThe acrosomal cap. However, apicalcase acrosomal andwe thecreated outer acrosomal membrane are both fluorescent signal of CD9 of mousearea sperm, 3D colocalization maps based on Pearson’s correlation coefficient it stays detectable in the postacrosomal region of human sperm head (Figure 4, line III). As infrom the the remains equatorial (see red arrow in Figure 4,area line and III), but CD81interpretation disappears for in a the more accurate segment visualisation of the colocalization a better of case of mouse sperm, we created 3D colocalization maps based on Pearson’s correlation coefficient resultscap. (Figure 5). The it ejaculated human sperm show high rate of colocalization apicalimmunofluorescent area of the acrosomal However, stays detectable in the postacrosomal region of human for a more accurate visualisation of the colocalization area and a better interpretation of (Figure While4,nearly colocalization was detected in acrosome reacted human (Figure 5B). maps sperm head5A). (Figure line no III). As in the case of mouse sperm, we created 3Dsperm colocalization immunofluorescent results (Figure 5). The ejaculated human sperm show high rate of colocalization based(Figure on Pearson’s correlation coefficient for more accurate visualisation of the colocalization 5A). While nearly no colocalization wasadetected in acrosome reacted human sperm (Figure 5B). area

and a better interpretation of immunofluorescent results (Figure 5). The ejaculated human sperm

Int. J. Mol. Sci. 2018, 19, 1236

5 of 17

showInt. high rate colocalization (Figure 5A). While nearly no colocalization was detected in 5acrosome J. Mol. Sci. of 2018, 19, x FOR PEER REVIEW of 17 Int. J. Mol.sperm Sci. 2018, (Figure 19, x FOR PEER 5 of 17 reacted human 5B).REVIEW

Figure 4. Dynamics of CD9 and CD81 in their mutual localization in human sperm captured by

Figure 4. Dynamics of CD9 and in their mutual localization human sperm captured confocal microscopy. (I) InCD81 ejaculated sperm, both CD9 (red) and in CD81 (green) are present in theby confocal microscopy. (I)Dynamics In ejaculated sperm, both (red) and CD81 (green) are present in the apical apical acrosome. (II) capacitation changes in localization of both occur. (III) After acrosome. Figure 4. of During CD9 and CD81CD9 innotheir mutual localization inproteins human sperm captured by completing AR, theno plasma membrane covering theof apical acrosomal area and the outer confocal microscopy. (I) In ejaculated sperm, both CD9 (red) and CD81 (green) areacrosomal present in the AR, (II) During capacitation changes in localization both proteins occur. (III) After completing are(II) both lost. The fluorescent signal of CD9 located overofthe equatorial segment (red After apicalmembrane acrosome. During capacitation changes in is localization both proteins occur.membrane (III) the plasma membrane covering the apicalnoacrosomal area and the outer acrosomal are arrow) and CD81 disappears from the acrosome. After the acrosome reaction, CD81 is mainly the plasma membrane covering the apical acrosomal area and the outer acrosomal both completing lost.detectable The AR, fluorescent signal of CD9 is located over the equatorial segment (red arrow) and in the postacrosomal region of human sperm head. DAPI (blue). Scale bar represents 1 μm. are both lost. The fluorescent signal of CD9 is located over the equatorial segment (red CD81membrane disappears from the acrosome. After the acrosome reaction, CD81 is mainly detectable in the arrow) and CD81 disappears from the acrosome. After the acrosome reaction, CD81 is mainly postacrosomal region of human sperm head. DAPI (blue). Scale bar represents 1 µm. detectable in the postacrosomal region of human sperm head. DAPI (blue). Scale bar represents 1 μm.

Figure 5. Surface rendering of colocalization map of CD9 and CD81 in human ejaculated and acrosome reacted sperm. Confocal images and Huygens software were used for better visualisation of colocalization area (yellow) for CD9 (red) and CD81 (green). (A) Ejaculated human sperm shows a high rate of colocalization of CD9 and CD81 in the acrosomal cap. (B) In case of acrosome reacted sperm, CD9 remains to be present in the inner acrosome membrane, however CD81 is hardly Figure 5. Surface rendering of colocalization map of CD9 and CD81 in human ejaculated and the spermofhead. DAPI (blue). Scale barof represents 2 (A)CD81 and 3 (B) Figure 5. detectable Surfaceacross rendering colocalization map CD9 and inμm. human ejaculated

and acrosome reacted sperm. Confocal images and Huygens software were used for better visualisation acrosome reacted sperm. Confocal images and Huygens software were used for better visualisation of area (yellow) (red) and CD81 (green). (A) Ejaculated human sperm shows a of 2.3.colocalization Immunoprecipitation of CD9 for andCD9 CD81 Complex from Spermatozoa colocalization area (yellow) for CD9 (red) and CD81 (green). (A) cap. Ejaculated human sperm shows a high high rate of colocalization of CD9 and CD81 in the acrosomal (B) In case of acrosome reacted The co-immunoprecipitation experiments on the mouse model did not reveal any mutual rate of colocalization of CD9 and CD81ininthe theinner acrosomal cap.membrane, (B) In casehowever of acrosome sperm, sperm, CD9 remains to be present acrosome CD81reacted is hardly interaction between CD9 and CD81 tetraspanin molecules in spermatozoa. Representative figure of detectabletoacross the sperm head. DAPI (blue). Scale bar represents 2 (A) and is 3 (B) μm. detectable across CD9 remains be present in the inner acrosome membrane, however CD81 hardly detection with CD81 antibody in CD9 epididymal sperm immunoprecipitate did not show a band in the sperm head. DAPI (blue). bar represents 2 (A) and 3 (B) µm. experiments with human expected molecular weightScale (Figure 6). The co-immunoprecipitation 2.3. Immunoprecipitation of CD9 and CD81 Complex from Spermatozoa spermatozoa proved CD9 and CD81 interactions in different stages of their post-testicular

2.3. Immunoprecipitation of CD9 and CD81 Complexon from The co-immunoprecipitation experiments theSpermatozoa mouse model did not reveal any mutual interaction between CD9 and CD81 tetraspanin molecules in spermatozoa. Representative figure of

The co-immunoprecipitation experiments on the mouse model did not reveal any mutual detection with CD81 antibody in CD9 epididymal sperm immunoprecipitate did not show a band in interaction between CD9 and CD81 tetraspanin molecules in spermatozoa. Representative figure expected molecular weight (Figure 6). The co-immunoprecipitation experiments with human of detection with CD81 in CD9 immunoprecipitate didpost-testicular not show a band spermatozoa provedantibody CD9 and CD81epididymal interactionssperm in different stages of their in expected molecular weight (Figure 6). The co-immunoprecipitation experiments with human spermatozoa proved CD9 and CD81 interactions in different stages of their post-testicular maturation

Int. J. Mol. Sci. 2018, 19, x FOR PEER REVIEW

6 of 17

maturation (Figure 6). Int. J. Mol. Sci. 2018, 19, 1236

In CD9 immunoprecipitate from ejaculated and capacitated human sperm, 6 of 17 Int. J. Mol. Sci.reacted 2018, 19, x with FOR PEER REVIEW 6 of 17 hand, CD81 antibody two bands in molecular weights of 22 and 24 kDa. On the other the CD81 molecule was detected in CD9 immunoprecipitate from acrosome-reacted sperm only in maturation 6). In CD9 immunoprecipitate from ejaculated and capacitated humanCD81 sperm, (Figure 6). In band CD9(Figure immunoprecipitate from ejaculated and capacitated sperm, antibody one protein of 24 kDa. Strong protein bands showed heavy and human light chains of CD9 antibody CD81 antibody reacted with two bands in molecular weights of On 22 and 24 kDa.hand, On thethe other hand, reacted with two bands in molecular weights of 22 and 24 kDa. the other CD81 molecule (Figurethe 6, CD81 grey arrows). molecule was detected in CD9 immunoprecipitate from acrosome-reacted sperm only in was detected in CD9 immunoprecipitate from acrosome-reacted sperm only in one protein band of one protein band of 24 kDa. Strong protein bands showed heavy and light chains of CD9 antibody 24 kDa.(Figure Strong6,protein bands showed heavy and light chains of CD9 antibody (Figure 6, grey arrows). grey arrows).

7 of mouse FigureFigure 6. Co-immunoprecipitation CD81molecules molecules from lysates 5 7 ×of10 6. Co-immunoprecipitationofofCD9 CD9 and and CD81 from lysates 5 × 10 mouse Figure 6. Co-immunoprecipitation of CD9 and CD81 molecules from lysates 5 × 107 of mouse epididymal sperm (Ep),and and human human sperm sperm (Ej—ejaculated, AR— AR— epididymal sperm (Ep), (Ej—ejaculated,Cap—capacitated, Cap—capacitated, epididymal sperm (Ep), and human sperm (Ej—ejaculated, Cap—capacitated, AR—acrosome-reacted). acrosome-reacted). Detection CD81tetraspanin tetraspanin with rabbit antibody anti-CD81 (H-121)(H-121) acrosome-reacted). Detection of of CD81 withpolyclonal polyclonal rabbit antibody anti-CD81 Detection of immunoprecipitates CD81 tetraspanin with polyclonal rabbit (H-110); antibody (H-121) in CD9 using polyclonal rabbit antibody grey anti-CD81 arrows—heavy and lightin CD9 in CD9 immunoprecipitates using polyclonal rabbit antibody (H-110); grey arrows—heavy and light immunoprecipitates usingblack polyclonal rabbit antibody grey arrows—heavy and light chains chains from antibody, arrows—antibody reaction,(H-110); * protein bands from rabbit antiserum. chains from antibody, black arrows—antibody reaction, * protein bands from rabbit antiserum. from antibody, black arrows—antibody reaction, * protein bands from rabbit antiserum. 2.4. Determination of Intra-Molecular Disulphide Bonds in CD9 and CD81 Tetraspanins

2.4. Determination of Intra-Molecular Disulphide Bonds in CD9 and CD81 Tetraspanins Westernofblot analysis under Disulphide a non-reducing condition was performed for depiction of potential 2.4. Determination Intra-Molecular Bonds in CD9 and CD81 Tetraspanins disulphide in tetraspanin molecules. Sperm RIPA was lysates from mouse spermatozoa in Western blotbonds analysis under a non-reducing condition performed for depiction of potential Western blot analysis under a non-reducing condition was performedcomplexes for depiction of potential different post-testicular maturation stages showed high-molecular-weight connected disulphide bonds in tetraspanin molecules. Sperm RIPA lysates from mouse spermatozoa in disulphide in tetraspanin RIPAanti-CD9 lysatesand from mouse spermatozoa with bonds intramolecular disulphide molecules. bridges. BothSperm antibodies, anti-CD81, recognized different post-testicular maturation stages showed high-molecular-weight complexes connectedin protein band over 160 kDa withstages decreased intensity from epididymal to acrosome-reacted mouse different post-testicular maturation showed high-molecular-weight complexes connected with with intramolecular disulphide bridges. Both antibodies, anti-CD9 and anti-CD81, recognized sperm samples. Furthermore, a protein band of 75 kDa was observed with a lower intensity in intramolecular disulphide bridges. Both antibodies, anti-CD9 and anti-CD81, recognized protein band proteincapacitated band over 160 kDa with decreased intensity from epididymal to acrosome-reacted and acrosome-reacted sperm lysates. In epididymal sperm extract, both tetraspanin mouse over 160 kDa with decreased intensity from epididymal to acrosome-reacted mouse sperm samples. sperm antibodies samples. recognized Furthermore, a protein of(Figure 75 kDa7A). was lower under intensity in diffuse band of band 45 kDa In observed all human with sperma lysates Furthermore, a protein bandthe of CD9 75 sperm kDa was observed withbands a lower intensity inboth capacitated and non-reducing conditions, antibody detected with molecular of 60 capacitated and acrosome-reacted lysates. In protein epididymal sperm extract,weights tetraspanin acrosome-reacted sperm lysates. In epididymal sperm extract, both tetraspanin antibodies recognized and 100 kDa, and a double band in the range of 20 and 25 kDa. The CD81 antibody recognized antibodies recognized diffuse band of 45 kDa (Figure 7A). In all human sperm lysates under evident of approximately kDa and two weakerlysates protein bands between and 100 kDa in allthe diffuse band ofband 45 kDa (Figure Inantibody all human sperm under non-reducing conditions, non-reducing conditions, the7A). CD948 detected protein bands with 75 molecular weights ofCD9 60 sperm lysates (Figure 7B). antibody bands with of 60 kDa, andantibody a doublerecognized band in the and 100 detected kDa, andprotein a double band in molecular the range weights of 20 and 25 and kDa.100 The CD81 range of 20 andof25 kDa. The CD81 antibody recognized evidentbands bandbetween of approximately kDa evident band approximately 48 kDa and two weaker protein 75 and 10048 kDa inand all two weaker protein bands between 75 and 100 kDa in all sperm lysates (Figure 7B). sperm lysates (Figure 7B).

Figure 7. Detection of CD9 and CD81 tetraspanins in mouse (A) and human (B) sperm extract under non-reducing conditions (5 × 106 sperm cells per lane); Ep—epididymal, Ej—ejaculated, Cap— capacitated, and AR—acrosome-reacted spermatozoa. White arrows indicate antibody reaction. As negative control, blots were incubated with rabbit immunoglobulins. Blots were completely negative.

Figure7.7. Detection CD81 tetraspanins in mouse (A) and (B) sperm extract extract under Figure DetectionofofCD9 CD9and and CD81 tetraspanins in mouse (A) human and human (B) sperm 6 sperm 6 non-reducing conditions (5 × 10 cells per lane); Ep—epididymal, Ej—ejaculated, Cap— under non-reducing conditions (5 × 10 sperm cells per lane); Ep—epididymal, Ej—ejaculated, capacitated, and AR—acrosome-reacted spermatozoa. White arrows indicate antibody reaction. As Cap—capacitated, and AR—acrosome-reacted spermatozoa. White arrows indicate antibody negative control, blots were incubated with rabbit immunoglobulins. Blots were completely negative. reaction. As negative control, blots were incubated with rabbit immunoglobulins. Blots were completely negative.

Int. J. Mol. Sci. 2018, 19, 1236

Int. J. Mol. Sci. 2018, 19, x FOR PEER REVIEW 2.5. Molecular Modelling

7 of 17

7 of 17

2.5. Molecular Modelling We have obtained an all-atom model of human CD81 by modelling the region not refined by crystallography (residues 38 to 54) using runs of the loopmodel MODELLER. We have obtained an all-atom model of 100 human CD81 by modelling thefunction region notofrefined by The modelling suggests a disordered highly100 flexible without strong for secondary crystallography (residues 38 to 54) using runs ofregion the loopmodel functionpreference of MODELLER. The structure. However, thearemaining of flexible the extracellular domain is stabilized disulphide modelling suggests disordered part highly region without strong preferenceby fortwo secondary structure. However, remainingin part of the extracellular domain ischange stabilized by two disulphide bridges, and the domainthe is involved a closed/open conformation [26]. bridges, and understanding the domain is involved in achange, closed/open changea [26]. For better of this we conformation have performed series of 50 ns–long MD For better understanding of this change, we have performed a series of 50 Effective ns–long MD simulations of the CD81 model in implicit solvation/lipid membrane model; Energy simulations of the CD81 model in implicit solvation/lipid membrane model; Effective Energy Function 1/Implicit Membrane Model 1 (EEF1/IMM1) and the explicit all atom TIP3P water/DOPC Function 1/Implicit Membrane Model 1 (EEF1/IMM1) and the explicit all atom TIP3P water/DOPC (dioleoyl-phosphatidylcholine) membrane simulations with or without cholesterol present in the (dioleoyl-phosphatidylcholine) membrane simulations with or without cholesterol present in the binding cavity formed by transmembrane helices (TM). The cholesterol bound CD81 structure indicates binding cavity formed by transmembrane helices (TM). The cholesterol bound CD81 structure (consistently the crystal structure) a conical shape of the protein of TM indicates with (consistently with the crystal structure) a conical shape ofwith the wider proteinseparation with wider helices towardsofthe domain inducing adomain convexinducing curvature of thecurvature membrane bilayer separation TMextracellular helices towards the extracellular a convex of the (Figure 8). Within a cholesterol the cavity is not occupied, TM helices form membrane bilayer (Figure 8).depleted Within a environment, cholesterol depleted environment, the cavityand is not occupied, and TM helices form in a more compact bundle. This potentially leads to a change in membrane in a more compact bundle. This potentially leads to a change in membrane curvature which is crucial curvature which crucial for sperm-egg at the convex part of the as the for sperm-egg fusionis at the convex part offusion the equatorial segment, as equatorial the placesegment, of first sperm-egg place of first sperm-egg plasma membrane interaction. Without cholesterol binding, theof compact plasma membrane interaction. Without cholesterol binding, the compact arrangement TM helices arrangement of TM helices would lead to more planar or even a concave membrane bilayer. We would lead to more planar or even a concave membrane bilayer. We have also observed the cholesterol have also observed the cholesterol unbinding induced conformational change with the extracellular unbinding induced conformational change with the extracellular domain opening as a lid. The newly domain opening as a lid. The newly accessible residues, including the disordered region, probably accessible includingbythe disordered region, probably role facilitate in recognition byand a so far play aresidues, role in recognition a so far unknown binding partnerplay and acould both Cis unknown binding partner and could facilitate both Cis and Trans interactions. Trans interactions.

Figure 8. Models of CD81 and CD9 with and without cholesterol. (A) CD81 monomer (yellow) with

Figure 8. Models of CD81 and CD9 with and without cholesterol. (A) CD81 monomer (yellow) with bound cholesterol (green) after 50 ns MD simulation, showing the conical shape of the TM helices. bound cholesterol 50structure ns MD simulation, conical shape of thewithout TM helices. Overlaid is the (green) starting after crystal (transparent showing gray); (B) the CD81 monomer (yellow) Overlaid is the starting crystal structure (transparent gray); (B) CD81 monomer (yellow) without cholesterol after 50 ns MD simulation. The TM helices adopt more compact cylinder like cholesterol after 50 ns MD simulation. The TM helices adopt more compact cylinder like conformation; conformation; (C) CD9 monomer (blue) binding cholesterol (green) after 50 ns MD simulation. The (C) CD9 monomer (blue) (green) 50 ns MD simulation. model suggests model suggests that binding contrarycholesterol to the CD81 case after the extracellular domain isThe more extended in that cholesterol rich environment; (D) CD9 domain monomer (blue)extended without cholesterol after 50environment; ns MD contrary to the CD81 case the extracellular is more in cholesterol rich simulation showing a tight interaction of theafter TM helices. (D) CD9 monomer (blue) without cholesterol 50 ns MD simulation showing a tight interaction of the TM helices. Further, we have performed a series of protein–protein flexible side-chain docking using ClusPro. Two principal CD81 binding modes were identified (Figure 9A,B). While the most probable Further, we have performed a series of protein–protein flexible side-chain ClusPro. dimer interface shows the TM Cys residues pointing in opposite direction and thedocking dimer isusing stabilized Two principal CD81 binding modes wereinterface identified (Figure 9A,B). While the mostbyprobable only non-covalently, the second dimer allows further covalent stabilization forming dimer a interface shows bridge the TMbetween Cys residues pointingCys in opposite direction and the dimer is stabilized disulphide neighbouring residues across monomers. The formation of a only disulphide bridge between CD81 monomers is also supported by our experimental results. a disulphide non-covalently, the second dimer interface allows further covalent stabilization by forming

bridge between neighbouring Cys residues across monomers. The formation of a disulphide bridge between CD81 monomers is also supported by our experimental results.

Int. J. Mol. Sci. 2018, 19, 1236 Int. J. Mol. Sci. 2018, 19, x FOR PEER REVIEW

8 of 17 8 of 17

Figure 9. A summary of oligomer forming interfaces of CD81 and CD9. (A,B) Top and side view Figure 9. A summary of oligomer forming interfaces of CD81 and CD9. (A,B) Top and side view showing two possible binding modes of CD81 monomer (yellow) as predicted by flexible side-chain showing two possible binding modes of CD81 monomer (yellow) as predicted by flexible side-chain docking. The yellow-orange CD81 dimer displays a strong non-covalent interaction, while the less docking. The yellow-orange CD81 dimer displays a strong non-covalent interaction, while the less stable yellow-lime CD81 dimer can be further stabilized by formation of an intermolecular disulphide stable yellow-lime CD81 dimer can be further stabilized by formation of an intermolecular disulphide bridge. The Cys residues are shown in pink. A rotation of the covalently bound lime CD81 moiety bridge. The Cys residues are shown in pink. A rotation of the covalently bound lime CD81 moiety could could another subunit leading to stepwise formation of CD81 oligomer stretch Figure recruitrecruit another CD81CD81 subunit leading to stepwise formation of CD81 oligomer stretch (see(see Figure 10); 10); (C) The CD9 dimer (shades of blue) displaying two potential binding modes with CD81 moiety (C) The CD9 dimer (shades of blue) displaying two potential binding modes with CD81 moiety (yellow) (yellow) as predicted by side-chain flexible side-chain as predicted by flexible docking.docking.

An all-atom model of the human CD9 protein was obtained using the I-TASSER suite of An all-atom model of the human CD9 protein was obtained using the I-TASSER suite of programs. programs. We have performed 50 ns long MD simulations of the CD9 model with or without We have performed 50 ns long MD simulations of the CD9 model with or without cholesterol present. cholesterol present. The modelling suggests that the effect of possible cholesterol binding to CD9 is The modelling suggests that the effect of possible cholesterol binding to CD9 is not as profound as not as profound as in CD81 case. The results show that the TM helices of CD9 are rather compact in CD81 case. The results show that the TM helices of CD9 are rather compact and arranged almost and arranged almost parallel to each other in both cases. The extracellular domain is only weakly parallel to each other in both cases. The extracellular domain is only weakly sensitive to cholesterol sensitive to cholesterol binding and remains in the open conformation. This may be interpreted that binding and remains in the open conformation. This may be interpreted that CD9 takes less active or CD9 takes less active or no role in shaping membrane curvature contrary to CD81. no role in shaping membrane curvature contrary to CD81. In order to identify a possible CD9 dimer, we performed the protein-protein flexible side-chain In order to identify a possible CD9 dimer, we performed the protein-protein flexible side-chain docking using ClusPro. The results suggested a dimer with the TM helices of CD9 monomers in a V docking using ClusPro. The results suggested a dimer with the TM helices of CD9 monomers in a V shape arrangement (Figure 9C). The docking followed by a 50 ns MD simulation of the CD9 dimer in shape arrangement (Figure 9C). The docking followed by a 50 ns MD simulation of the CD9 dimer in implicit solvent/membrane environment suggests that the extracellular domain is involved in CD9 implicit solvent/membrane suggests that the domain is involved in CD9 dimer formation employingenvironment the short helix (residues 138extracellular to 151) as dimerization interface. The dimer formation employing the short helix (residues 138 to 151) as dimerization interface. The outward outward facing surface can serve as an interface for other binding partners. facing surface can serve as an interface for other binding partners. No disulphide bridges involving Cys residues within the transmembrane region of CD9 dimer disulphide involvingThis Cys residues within with the transmembrane of CD9 wereNo suggested bybridges the docking. is consistent the Uniprot region annotation fordimer the were suggested by the docking. This is consistent with the Uniprot annotation for the transmembrane transmembrane Cys residues which are expected to be post translationally modified by Cys residues which are expected be post translationally modified by palmitoylation, probably palmitoylation, probably utilized to more in CD9 web stabilization during sperm maturation and utilized more in CD9 web stabilization during sperm maturation and acrosome reaction protein acrosome reaction protein dynamics than in cholesterol binding. Moreover, the palmitoylation is dynamics during than in membrane cholesterolfusion binding. Moreover, the palmitoylation is across beneficial membrane beneficial when could ease the CD9 transfer the during membranes. CD9, fusion when could ease the CD9 transfer across the membranes. CD9, in contrary to CD81, also exhibits in contrary to CD81, also exhibits more extended extracellular domain in dimer (Figure 9C) which more extended extracellular in dimer (Figure On 9C) the which would be recipient theshow trans would be recipient towardsdomain the trans interactions. other hand, the CD81towards does not interactions. On the other hand, the CD81 does not show considerable palmitoylation according considerable palmitoylation according the Uniprot, and as a result would allow binding the of Uniprot, and a result would allow of cholesterol or showed phospoholipides. Our experimental cholesterol orasphospoholipides. Our binding experimental data also no oligomers stabilized by data also showed no oligomers stabilized by However, intermolecular disulphide for CD9. However, intermolecular disulphide bridges for CD9. similarly to the bridges CD81, two intramolecular similarly to the CD81, two intramolecular disulphide bridges stabilizing the extracellular domain of disulphide bridges stabilizing the extracellular domain of CD9 are expected to form based on the CD9 are expected to form based on the sequence homology. sequence homology. 3. Discussion 3. Discussion Events prior and during sperm–egg fusion are connected with drastic membrane reorganization Events prior and during sperm–egg fusion are connected with drastic membrane of both gametes [27]. One of the “regulators” of these processes is cholesterol efflux. Besides the reorganization of both gametes [27]. One of the “regulators” of these processes is cholesterol efflux. membrane microdomains enriched in cholesterol defined as rafts, other membrane “structures” Besides the membrane microdomains enriched in cholesterol defined as rafts, other membrane called tetraspanin-enriched microdomains (TEM) are presented within the membrane [5,28], and “structures” called tetraspanin-enriched microdomains (TEM) are presented within the membrane [5,28], and tetraspanin–tetraspanin interactions are also regulated by lipids, including the palmitate moieties that are attached to tetraspanins, membrane cholesterol, and gangliosides [28–32]. Based on

Int. J. Mol. Sci. 2018, 19, 1236

9 of 17

tetraspanin–tetraspanin interactions are also regulated by lipids, including the palmitate moieties that are attached to tetraspanins, membrane cholesterol, and gangliosides [28–32]. Based on recently detected open conformation (cholesterol free) of CD81 tetraspanin, the regulation of subcellular localisation of their partner proteins in response to differences on cholesterol concentration could be proposed, via the ability to tightly bind partner proteins [24]. This feature of CD81 is likely to be also utilized by mammalian sperm during their maturation call capacitation when major membrane rearrangement including cholesterol efflux takes place [33]. In vitro capacitation disrupts lipid raft domains and causes a shift in the overall membrane fluidity in the plasma membrane [34], which may induce the interaction of raft resident proteins to initiate signalling pathways associated with the capacitation [34]. Based on the data of Hogue et al. [35] the interaction of proteins from rafts with tetraspanins within TEM could be expected. Within mammalian sperm lipid rafts, several molecules have been documented with affinity for the ZP and egg plasma membrane including sperm fusogenic protein IZUMO1 [34], and according to Tanphaichitr et al. [36], the sperm lipid rafts are platforms of ZP binding molecules on the sperm plasma membrane. Beside others, capacitation-associated changes in sperm lipid rafts play a role in positioning IZUMO1 into the equatorial fusogenic region during AR [37,38]. The cooperation with rafts also seems to be probable for CD9 and CD81 in mice. These tetraspanins share location and dynamic rearrangement during AR and they are part of the multiprotein network participating during membrane fusion. The questions have remained to be answer whether CD9 and CD81 crosslink into a uniform heterologous network, and if so, whether these two tetraspanins utilise covalent or non-covalent bonds. Moreover, are these predicted interaction species specific or could they also depend on the maturation stage of the sperm? Our results from immunofluorescent staining and consequently from co-precipitation suggest that both are probably true. We showed crucial differences in localization of CD9 and CD81 between human and mouse sperm and between individual maturation stage of sperm. Our results from co-precipitation experiments showed potential mutual interaction of CD9 and CD81 tetraspanins in human spermatozoa which coincide with results of immunofluorescent staining that confirm mutual colocalization of studied proteins in acrosomal cap area. Consequently, in epididymal mouse sperm, where we noticed completely different localization of CD9 and CD81, no complexes of these proteins were detected by co-precipitation. However, contrary to Ito et al. [19] that described CD9 as an inner acrosomal membrane associated protein, our data showed the presence of CD9 in both outer and inner acrosomal membranes. Moreover, in human ejaculated sperm CD9 immunoprecipitates, the CD81 antibody recognized two bands of 22 and 24 kDa. Antibody signal of lower protein band decreased during capacitation and was completely absent in acrosome-reacted sperm. This may be caused by leaving of antibody epitope from one of CD81 isoforms or number of spermatozoa undergoing spontaneous acrosome reaction, where isoform with lower molecular weight disappeared together with acrosomal cap. For disulphide bridges modelling, lysates of human spermatozoa were prepared in non-reducing conditions. CD81 molecule forms dimmers connect with S-S bonds. This presumptive CD81 dimmer has been already described by Stipp et al. [39]. On the other hand, CD9 protein was found predominantly as a monomer (double band between 20 and 25 kDa) without disulphide bridges. Additionally, some covalent complexes were detected in approximate molecular weight of 60 kDa. Moreover, covalent complexes in human sperm lysates differ in CD9 and CD81 detection. However, according to co-precipitation experiments, these two tetraspanin molecules may interact non-covalently. Nevertheless, CD9 and CD81 molecules might be covalently linked with other tetraspanins or protein molecules. In the structure of the CD9 molecule, there are 10 cysteins and six are palmitoylated. Four of them are located in extracellular domain without palmitoylation and covalent interaction with other proteins may be formed more likely in extracellular domain. In samples of mouse spermatozoa, both antibodies detected high-molecular-weight complexes (over 160 kDa), presumably octamers. Although detection by both antibodies showed bands in the same molecular weight, co-precipitation experiments revealed that these complexes are not covalently formed by CD9 and CD81. However,

Int. J. Mol. Sci. 2018, 19, 1236

10 of 17

as mentioned above also these complexes are likely formed by CD9 and CD81 molecules with other proteins, such as ERMs [40] and integrins, CD19, CD21, and CD45 [41]. Interestingly, there is evident decrease of signal intensity of antibodies reaction in capacitated and acrosome reacted sperm lysates, Int. J. due Mol. Sci. 19, x FOR PEER REVIEW 10 of 17 probably to 2018, spontaneous and induced acrosome reaction [42], respectively. To suggest a potential structure and function of a tetraspanin web involving CD81 and CD9 capacitated and acrosome reacted sperm lysates, probably due to spontaneous and induced moieties, we performed a series of docking and MD simulations with different CD81/CD9 acrosome reaction [42],also respectively. stoichiometry. Based on the results of alland CD9function and CD81 we can speculate complex To suggest a potential structure of a simulations tetraspanin web involving CD81that anda CD9 network can bewe formed combining the extended CD81 formed with by alternating non-covalent moieties, performed also a series of docking andoligomer MD simulations different CD81/CD9 and disulphide bridge stabilized dimers.ofAall CD9 non-covalent dimer can then interact withthat twoasuch stoichiometry. Based on the results CD9 and CD81 simulations we can speculate complex network can be formed via combining the extended CD81 with oligomer formed by alternating oligomers. Moreover, tetraspanins their ability to associate membrane-curving proteins non-covalent and disulphide bridge stabilized dimers. A CD9 non-covalent dimer can then interact or lipids, or different tetraspanins may influence the curvature of the membrane and formation of with two structures such oligomers. Moreover, intetraspanins their ability to The associate withCD81 various tubular [25] participating formation ofvia fusogenic domains. CD9 and membrane-curving proteins or lipids, or different tetraspanins may influence the curvature of the combined network can be significantly influenced by the presence or absence of cholesterol in the lipid membrane and formation of various tubular structures [25] participating in formation of fusogenic bilayer leading to changes in membrane curvature. The proposed arrangement of the CD81/CD9 domains. The CD9 and CD81 combined network can be significantly influenced by the presence or tetraspanin is summarized in Figure Under low cholesterol conditions dissociation absenceweb of cholesterol in the lipid bilayer 10. leading to changes in membrane curvature.the The proposed of CD81arrangement bound cholesterol might lead to moreweb compact TM region within changing of the CD81/CD9 tetraspanin is summarized in Figure 10.CD81 Understretches low cholesterol its conical to cylindrical conformation result in lower area of the “upper” leaf of the conditions the dissociation of CD81 which bound would cholesterol might lead to more compact TM region within stretches changing its conical cylindrical conformation which would resultain lower role bilayer and CD81 bending of the membrane. ThistoCD81 ability may be suspected to play crucial area of the “upper” leaf of the bilayer and bending of the membrane. This CD81 ability may be for facilitating sperm-egg plasma membrane interaction and fusion as convex part of the equatorial suspected to play a crucial role for facilitating sperm-egg plasma membrane interaction and fusion segment servers for primary recognition and perpendicular attachment of mammalian sperm to the as convex part of was the equatorial segment servers for of primary recognition and perpendicular oolema. The modelling dealing with human variants CD9 and CD81 proteins. The analysis of attachment of mammalian sperm to the oolema. The modelling was dealing with human variants of homology between mouse and human variants revealed that the transmembrane domain residues are CD9 and CD81 proteins. The analysis of homology between mouse and human variants revealed mostly sequentially conserved. However, the extracellular domains differ significantly as summarized that the transmembrane domain residues are mostly sequentially conserved. However, the in Figure 11. extracellular domains differ significantly as summarized in Figure 11.

Figure 10. The proposed arrangement of the CD81/CD9 tetraspanin network. A complex CD9 and

Figure 10. The proposed of the CD81/CD9 tetraspanin network. formed A complex and CD81 CD81 network can arrangement be formed combining the extended CD81 oligomer by CD9 alternating network can be formed combining the extended CD81 oligomer formed by alternating non-covalent non-covalent (yellow-orange) and disulphide bridge stabilized dimers (orange-orange). A dimer of (yellow-orange) and disulphide bridge stabilized dimers (orange-orange). A could dimerbe offormed. dimers A is CD9 depicted dimers is depicted for simplicity, however, longer CD81 oligomer stretches for simplicity, however, longer CD81 oligomer could formed. A CD9 non-covalent dimer non-covalent dimer (shades of blue) can then stretches interact with twobe such oligomers. (A,B) Side and front view alongcan thethen CD81 oligomer (a oligomers. membrane—black lines—would be inalong horizontal (shades of blue) interact with stretch two such (A,B) Side and front view the CD81 orientation) showing the possible CD81/CD9 network Top view (membrane in plane). Under low oligomer stretch (a membrane—black lines—would be (C) in horizontal orientation) showing the possible cholesterol conditions dissociation of CD81 bound cholesterol might leadconditions to more compact TM CD81/CD9 network (C) Topthe view (membrane in plane). Under low cholesterol the dissociation region within CD81 stretches changing its conical to cylindrical conformation which would result in of CD81 bound cholesterol might lead to more compact TM region within CD81 stretches changing its lower area of the “upper” leaf of the bilayer and bending of the membrane (indicated with arrows in conical to cylindrical conformation which would result in lower area of the “upper” leaf of the bilayer and (A,B)). bending of the membrane (indicated with arrows in (A,B)).

Int. J. Mol. Sci. 2018, 19, 1236

11 of 17

Int. J. Mol. Sci. 2018, 19, x FOR PEER REVIEW

11 of 17

Figure 11. The Figure 11. The position position of of non-identical non-identical amino amino acids acids between between human human and and mouse mouse CD81 CD81 and and CD9 CD9 projected onto the model of CD9/CD81 complex (Figure 9C). The differences in amino acid sequence projected onto the model of CD9/CD81 complex (Figure 9C). The differences in amino acid sequence are depicted as red spheres. spheres. The extracellular are depicted as red The CD81 CD81 moiety moiety (yellow) (yellow) contains contains 17 17 differences differences in in the the extracellular domain, blue) differs differs in in 25 25 amino amino acids acids located located also also at at the the CD9/CD81 CD9/CD81 domain, while while the the CD9 CD9 (shades (shades of of blue) interface. These differences could be responsible for species specific interaction between CD9CD81 and interface. These differences could be responsible for species specific interaction between CD9 and CD81 as well as recognition by other proteins. as well as recognition by other proteins.

The strong adhesion adhesion generating generating fusion fusion competent The responsibility responsibility of of CD9 CD9 tetraspanin tetraspanin for for strong competent sites sites between sperm and egg plasma membrane has been shown [15] as well as the importance of between sperm and egg plasma membrane has been shown [15] as well as the importance of cholesterol cholesterol as a key mediator of membrane curvature during fusion events [43,44]. A complex CD9 as a key mediator of membrane curvature during fusion events [43,44]. A complex CD9 and CD81 and CD81 network can be formed combining the CD81 oligomer utilizing Cis interaction both network can be formed combining the CD81 oligomer utilizing Cis interaction both non-covalent and non-covalent and which disulphide bridge which stabilization. lead to the complex stabilization.dimer A CD9 disulphide bridge lead to the complex A CD9 non-covalent cannon-covalent then interact dimer can then interact with two such CD81 oligomers. Buschiazzo et al. [44] assumed that with two such CD81 oligomers. Buschiazzo et al. [44] assumed that cholesterol efflux during the sperm cholesterol efflux during the sperm capacitation make sperm membrane more fluid and able to capacitation make sperm membrane more fluid and able to undergo greater positive curvature to undergo greater positive curvature to adapt a more ordered oocyte membrane at the moment of adapt a more ordered oocyte membrane at the moment of fusion. Specifically, under low cholesterol fusion. Specifically, under low cholesterol conditions, the dissociation of CD81 bound cholesterol conditions, the dissociation of CD81 bound cholesterol might lead to more compact transmembrane might lead to more compact transmembrane region within CD81 stretches changing its conical to region within CD81 stretches changing its conical to cylindrical conformation. This feature would cylindrical conformation. This feature would consequently result in bending of the bilayer consequently result in bending of the bilayer membrane. Taken together experimental results and membrane. Taken together experimental results and structural modelling, we proposed, that due to structural modelling, we proposed, that due to the ability of the CD81 molecule to bind and release the ability of the CD81 molecule to bind and release cholesterol accompanied by stable covalent cholesterol accompanied by stable covalent formation of the network, CD81 might be considered formation of the network, CD81 might be considered as a regulator of dynamic machinery within as a regulator of dynamic machinery within the tetraspanin web on the sperm plasma membrane. the tetraspanin web on the sperm plasma membrane. On the other hand, CD9 in weak interaction On the other hand, CD9 in weak interaction with CD81 might be more likely involved in tetraspanins with CD81 might be more likely involved in tetraspanins web stabilization and play a role in web stabilization and play a role in tetraspanins facilitated trans interaction at sperm–egg membrane tetraspanins facilitated trans interaction at sperm–egg membrane recognition by employing the recognition by employing the more extended extracellular domain. more extended extracellular domain. 4. Materials and Methods 4. Materials and Methods 4.1. Primary Antibodies 4.1. Primary Antibodies In our experiments, the following primary antibodies were used: polyclonal rabbit anti-CD9: In sc-9148, our experiments, the following used: polyclonal anti-CD9: H-110, raised against a peptide primary mappingantibodies within an were internal region of CD9 rabbit (101–210 AA) of H-110, raised against a peptide mapping anraised internal region of CD9 (101–210 of humansc-9148, origin; polyclonal rabbit anti-CD81: H-121,within sc-9158, against a peptide mappingAA) within human origin; polyclonal H-121, origin; sc-9158,polyclonal raised against peptide mapping within an internal region (90–210rabbit AA) ofanti-CD81: CD81 of human goat aanti-CD81: Q-14, sc-31234, an internal region (90–210 AA) ofwithin CD81 of origin; polyclonaldomain goat anti-CD81: sc-31234, raised against a peptide mapping anhuman N-terminal extracellular of CD81 ofQ-14, human origin. raised against a peptide mapping within an N-terminal extracellular domain of CD81 of human origin. All these antibodies were produced by Santa Cruz Biotechnology, Santa Cruz, CA, USA.

Int. J. Mol. Sci. 2018, 19, 1236

12 of 17

All these antibodies were produced by Santa Cruz Biotechnology, Santa Cruz, CA, USA. Anti-CD46 (HM-1118; Hycult Biotech, Uden, The Netherlands) was used as a marker of acrosomal membrane. 4.2. Animals C57BL/6J mice were used for the experiments. They were purchased from the Animal Resources Centre or produced by the animal breeding facilities of the Institute of Biotechnology. The mice were housed in the IMG animal facilities, Institute of Molecular Genetics of Czech Academy of Science, Prague, and food and water were supplied ad libitum. The mice were healthy 10–12 weeks old five animals with no sign of stress or discomfort. All animal procedures and experimental protocols were approved by the Animal Welfare Committee of the Czech Academy of Sciences (Animal Ethics Number 66866/2015-MZE-17214, 18 December 2015). 4.3. Human Samples Human ejaculates were obtained from IVF Center Gennet (Prague, Czech Republic) with the informed consent of healthy donors and in accordance with the Institutes’ Human Ethics Committee guidelines. Spermiograms were evaluated in andrological laboratory at IVF Center. Only normospermic samples according to WHO laboratory manual (2010), five different healthy donors, were used in our study. After liquefaction, 1 mL of ejaculate was subjected to discontinuous centrifugation gradient (55%/80%) SupraSperm System (Origio, Måløv, Danmark) and centrifuged at 300 g for 20 min at room temperature. 4.4. Capacitation and Acrosome Reaction 4.4.1. Mouse Sperm In this study, 35 mm Petri dishes obtained from Corning (New York, NY, USA) were used for capacitation in vitro. Spermatozoa, which were recovered from the distal region of cauda epididymidis, were placed in capacitating M2 medium (Sigma-Aldrich, Prague, Czech Republic) and left in an incubator for 10 min at 37 ◦ C under 5% CO2 to relax sperm. After that, the concentration of stock sperm in medium was adjusted to 5 × 106 sperm/mL in 100 µL of M2 medium under paraffin oil and left to capacitate for 90 min. Calcium Ionophore (CaI, A 23187, Sigma-Aldrich), at a final concentration of 5 µM was added to the capacitated sperm and acrosome reaction was induced for an additional 90 m. The viability and motility of the sperm population was checked throughout the whole experiment. 4.4.2. Human Sperm Purified human spermatozoa of 5 × 106 cells were capacitated in 0.5 mL of SpermPreparation Medium (Origio, Måløv, Danmark) for 2 h at 37 ◦ C under 5% CO2 . Acrosome reaction was induced by calcium ionophore A23187 (Sigma-Aldrich) in final concentration of 10 µM for 1 h at 37 ◦ C under 5% CO2 . Acrosome-reacted sperm were evaluated on the absence of PNA-lectin labelling (VectorLaboratories, Burlingame, CA, USA) under the fluorescent microscope. 4.5. Dual Immunofluorescent Staining of CD81 and CD9 Ejaculated or epididymal, capacitated, and acrosome-reacted sperm were used for preparation of cells smears. Sperm were washed twice in tube with PBS, smeared onto glass slides, and air-dried. Sperm smears were fixed with 3.7% formaldehyde in PBS (pH 7.34) at room temperature for 10 min, followed by washing in PBS. Sperm were blocked with 5% BSA in PBS for 45 min and incubated with primary antibodies: goat polyclonal IgG anti-CD81 (Q14) diluted 1:20 in PBS and rabbit polyclonal IgG anti-CD9 (H110) diluted 1:20 in PBS over night at 4 ◦ C, followed by 1 h of incubation with both secondary antibody together: Alexa Fluor 568 donkey anti-rabbit IgG (H + L) or Alexa Fluor 488 donkey anti-goat IgG (H + L) (Life Technologies, Prague, Czech Republic) diluted 1:500

Int. J. Mol. Sci. 2018, 19, 1236

13 of 17

in PBS at room temperature. After washing, the slides were mounted into a Vectashield mounting medium with DAPI (Vector Lab., Burlingame, CA, USA). Fluorescent images were collected with high-end confocal microscope Carl Zeiss LSM 880 NLO at Imaging Methods Core Facility at BIOCEV (Vestec, Czech Republic). Huygens Professional version 17.10 (Scientific Volume Imaging, Hilversum, The Netherlands, Available online: http://svi.nl) software was used for deconvolution of confocal images and for colocalization maps. An open-source software Fiji [45] was used for image processing. 4.6. Super-Resolution Microscopy Freshly released epididymal sperm were used for SIM super-resolution microscopy. Sperm were collected, as described previously, with the following differences. Sperm samples were always prepared onto high precision cover glasses (thickness No. 1.5 H, 170 ± 5 µm, Marienfeld, Germany). Moreover, after the application of the primary and secondary antibodies, sperm were incubated for 5 min with DAPI (0.85 µg/mL, Thermo Scientific, Waltham, MA, USA) and washed 3× in PBS. At the end, sperm were washed 1× in distilled water and air-dried. Dry samples were covered with 90% glycerol with 5% anti-fade N-propyl gallate (Sigma-Aldrich). Multi-colour SIM super-resolution images were obtained by Zeiss Elyra PS.1 inverted microscope at Laboratory of confocal and fluorescent microscopy of Faculty of Science (Charles University, Prague, Czech Republic). An open source software Fiji was used for another image processing. 4.7. Protein Extraction Sperm suspension (mouse epididymal, human ejaculated, mouse and human capacitated and acrosome-reacted) were collected, washed in PBS and pellets of 5 × 107 sperm cells were lysed in 100 µl of RIPA buffer (150 Mm NaCl; 1% NP-40; 0.5% sodium deoxycholate; 0.1% SDS—Sodium Dodecyl Sulphate; 50 mM Tris pH 7.5 and protease inhibitors) for membrane protein extraction. Two-times concentrated non-reducing sample buffer with 8 M urea were added in 1:1 (v/v) ratio and vortexed for 2 min at RT. 4.8. Co-Immunoprecipitation of CD9 and CD81 Molecules Sperm suspension (5 × 107 ) was lysed in 100 µL of 1% CHAPS at 4◦ C for 1 h. Lysates were centrifuged at 20.000 g for 15 min at 4 ◦ C. Then supernatant was incubated with polyclonal antibody anti-CD9 (H-110) in final amount 5 µg per sample for 2 h at 4 ◦ C in rotator. Twenty µL of washed Dynabeads™ Protein G (Invitrogen, Carlsbad, CA, USA) were added and incubated for 1 h at 4 ◦ C in rotator. Precipitates bound to Dynabeads were separated in magnetic rack (Invitrogen) and washed with RIPA for 5 min at 4 ◦ C in rotator. This procedure was repeated for three times. Co-immunoprecipitated complexes were eluted from Dynabeads by incubation in reducing sample buffer for 5 min at 100 ◦ C. 4.9. SDS-PAGE with Immunoblotting SDS-electrophoresis and immunoblotting technique was used for CD9 and CD81 detection and was carried out using protocols based on standard methods [46,47]. Samples containing protein equivalent to 5 × 106 sperm cells were run on a 4% stacking and 12% running SDS polyacrylamide gel using Precision Plus Protein™ Dual Color Standards (Bio-Rad, Hercules, CA, USA) as molecular weight markers. After transferring proteins onto a PVDF (Polyvinylidine Fluoride) membrane, nonspecific sites were blocked with SuperBlock (PBS) Blocking Buffer (Thermo Scientific). CD9 and CD81 molecules were identified by the primary polyclonal rabbit antibodies anti-CD9 (H-110) and anti-CD81 (H-121) diluted 1:500 in PBS, followed by a peroxidase goat anti-rabbit IgG secondary antibody (Bio-Rad) diluted 1:3000. Antibody reaction was visualised by chemiluminescent substrate Super Signal West Pico (Pierce, Rockford, IL, USA). As a negative control, blots were incubated with immunoglobulins from rabbit serum (Sigma-Aldrich) in the same concentration as primary antibody.

Int. J. Mol. Sci. 2018, 19, 1236

14 of 17

4.10. Molecular Modelling of CD9 and CD81 Tetraspanin Web The amino acid sequences of studied human proteins were obtained from the UniProt database [48] as deposited under following entry names: CD9 (P21926) and CD81 (P60033). The all atom model of the CD81 was built with MODELLER [49] version 9.14 using the CD81 crystal structure (5tcx) [24] as template. The residues missing in the template were refined using the loopmodel function. The I-TASSER standalone package version 5.0 [50] was used to obtain homology models of CD9. The prediction of protein-protein interactions employed the locally available version of the ClusPro 2.0 protein-protein docking server [51]. The parameters of implicit solvation/lipid membrane model (EEF1/IMM1) [52] as well as the all atom DOPC membrane simulations were assigned using the web-based graphical user interface CHARMM-GUI (charmm-gui.org) [53]. The molecular dynamics simulation of suggested complexes was performed using the CHARMM [54] version c41b1 molecular dynamics package. The molecular graphics was prepared using Pymol [55] version 1.8.4. 4.11. Data Analysis Huygens Professional version 17.10 (Scientific Volume Imaging, Hilversum, The Netherlands, Available online: http://svi.nl) was used for visualisation mutual position of individual proteins based on surface rendering of the colocalization analysis. A colocalization analyser computed a Pearson’s correlation coefficient and created a 3D colocalization map. The Pearson’s correlation coefficient expresses the rate of correlation of colocalizing channels in a dual-colour image and gives a value between minus 1 to plus 1. In this case, 1 means an absolutely positive correlation, 0 means no correlation and −1 means a perfect anti-correlation. The value between 0.5 and 1 is interpreted as colocalization. Costes method was used for a background estimation. Acknowledgments: This work was supported by the project “BIOCEV—Biotechnology and Biomedicine Centre of the Academy of Sciences and Charles University” (CZ.1.05/1.1.00/02.0109), from the European Regional Development Fund (Available online: www.biocev.eu), by the Grant Agency of the Czech Republic No. GA-18-11275S, by the Charles University in Prague No. SVV260440, by the Institutional support of the Institute of Biotechnology RVO: 86652036, by Scientific Grant Agency of the Ministry of Education, Science, Research and Sport of the Slovak Republic and the Slovak Academy of Sciences (VEGA-2/0037/16), by the Slovak Research and Development Agency (APVV-15-0196), bilateral projects SAS-CAS (15-05) and SAS-CAS (18-17). We acknowledge the Imaging Methods Core Facility at BIOCEV, institution supported by the Czech-BioImaging large RI projects (LM2015062 and CZ.02.1.01/0.0/0.0/16_013/0001775, funded by Ministry of Education, Youth and of Sport Czech Republic) and Operational Program Prague Competitiveness (CZ.2.16/3.1.00/21515) funded by European Regional Development Fund for their support with obtaining imaging data presented in this paper. Access to computing and storage facilities owend by parties and projects contributing to the National Grid Infrastructure MetaCentrum provided under the programme “Projects of Large Research, Development and Innovations Infrastructures” (Czech Education and Scientific NETwork—CESNET LM2015042), is greatly appreciated. Author Contributions: Katerina Dvorakova-Hortova designed the study, analysed data and drafted the manuscript. Michaela Frolikova designed and performed all immunofluorescent experiments and analyses. Pavla Manaskova-Postlerova designed immunoprecipitation and immunodetection experiments and analysed data. Jiri Cerny performed the molecular modelling. Jana Jankovicova and Jana Antalikova analysed data and drafted the manuscript. Ondrej Simonik performed immunoprecipitated experiments and Western blot immunodetection. Alzbeta Pohlova prepared sperm samples and performed in vitro capacitation. Petra Secova contributed to the methodological support of immunoprecipitation and immunodetection. All the authors contributed to the manuscript preparation and approved the final manuscript. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2. 3. 4.

Stipp, C.S.; Orlicky, D.; Hemler, M. FPRP, a major, highly stoichiometric, highly specific CD81-and CD9-associated protein. J. Biol. Chem. 2001, 276, 4853–4862. [CrossRef] [PubMed] Van Spriel, A.B.; Figdor, C.G. The role of tetraspanins in the pathogenesis of infectious diseases. Microbes Infect. 2010, 12, 106–112. [CrossRef] [PubMed] Boucheix, C.; Rubinstein, E. Tetraspanins. Cell. Mol. Life Sci. 1997, 58, 1189–1205. [CrossRef] [PubMed] Charrin, S.; le Naour, F.; Silvie, O.; Milhiet, P.E.; Boucheix, C.; Rubinstein, E. Lateral organization of membrane proteins: Tetraspanins spin their web. Biochem. J. 2009, 420, 133–154. [CrossRef] [PubMed]

Int. J. Mol. Sci. 2018, 19, 1236

5. 6. 7. 8. 9.

10. 11.

12.

13. 14. 15.

16. 17.

18.

19.

20.

21. 22. 23.

24.

25.

15 of 17

Hemler, M.E. Tetraspanin proteins mediate cellular penetration, invasion, and fusion events and define a novel type of membrane microdomain. Annu. Rev. Cell Dev. Biol. 2003, 19, 397–422. [CrossRef] [PubMed] Levy, S.; Shoham, T. The tetraspanin web modulates immune-signaling complexes. Nat. Rev. Immunol. 2005, 5, 136–148. [CrossRef] [PubMed] Charrin, S.; Jouannet, S.; Boucheix, C.; Rubinstein, E. Tetraspains at a glance. J. Cell Sci. 2014, 127, 1–8. [CrossRef] [PubMed] Berditchevski, F.; Odintsova, E. Characterization of Integrin–Tetraspanin Adhesion Complexes: Role of Tetraspanins in Integrin Signaling. J. Cell Biol. 1999, 146, 477–492. [CrossRef] [PubMed] Stipp, C.S.; Hemler, M.E. Transmembrane-4-superfamily proteins CD151 and CD81 associate with alpha 3 beta 1 integrin, and selectively contribute to alpha 3 beta 1-dependent neurite outgrowth. J. Cell Sci. 2000, 113 Pt 11, 1871–1882. [PubMed] Bassani, S.; Cingolani, L.A. Tetraspanins: Interactions and interplay with integrins. Int. J. Biochem. Cell Biol. 2012, 44, 703–708. [CrossRef] [PubMed] Miyado, K.; Yamada, G.; Yamada, S.; Hasuwa, H.; Nakamura, Y.; Ryu, F.; Suzuki, K.; Kosai, K.; Inoue, K.; Ogura, A.; et al. Requirement of CD9 on the egg plasma membrane for fertilization. Science 2000, 287, 321–324. [CrossRef] [PubMed] Ziyyat, A.; Rubinstein, E.; Monier-Gavelle, F.; Barraud, V.; Kulski, O.; Prenant, M.; Boucheix, C.; Bomsel, M.; Wolf, J.P. CD9 controls the formation of clusters that contain tetraspanins and the integrin alpha 6 beta 1, which are involved in human and mouse gamete fusion. J. Cell Sci. 2006, 119 Pt 3, 416–424. [CrossRef] [PubMed] Frolikova, M.; Sebkova, N.; Ded, L.; Dvorakova-Hortova, K. Characterization of CD46 and beta1 integrin dynamics during sperm acrosome reaction. Sci. Rep. 2016, 6, 33714. [CrossRef] [PubMed] Yanez-Mo, M.; Tejedor, R.; Rouselle, P.; Sanches-Madrid, F. Tetraspanins in intercellular adhesion of polarized epithelial cells: Spatial and functional relationship to integrins and cadherins. J. Cell Sci. 2001, 144, 577–587. Jégou, A.; Ziyyat, A.; Barraud-Lange, V.; Perez, E.; Wolf, J.P.; Pincet, F.; Gourier, C. CD9 tetraspanin generates fusion competent sites on the egg membrane for mammalian fertilization. Proc. Natl. Acad. Sci. USA 2011, 108, 10946–10951. [CrossRef] [PubMed] Ellerman, D.A.; Ha, C.; Primakoff, P.; Myles, D.G.; Dveksler, G.S. Direct binding of the ligand PSG17 to CD9 requires a CD9 site essential for sperm-egg fusion. Mol. Biol. Cell 2003, 14, 5098–5103. [CrossRef] [PubMed] Kovalenko, O.; Yang, X.; Kolesnikova, T.V.; Hemler, M.E. Evidence for specific tetraspanin homodimers: Inhibition of palmitoylation makes cysteine residues available for cross-linking. Biochem. J. 2004, 377, 407–417. [CrossRef] [PubMed] Barreiro, O.; Zamai, M.; Yáñez-Mó, M.; Tejera, E.; López-Romero, P.; Monk, P.N.; Gratton, E.; Caiolfa, V.R.; Sánchez-Madrid, F. Endothelial adhesion receptors are recruited to adherent leukocytes by inclusion in preformed tetraspanin nanoplatforms. J. Cell Biol. 2008, 183, 527–542. [CrossRef] [PubMed] Ito, C.; Yamatoya, K.; Yoshida, K.; Maekawa, M.; Miyado, K.; Toshimori, K. Tetraspanin family protein CD9 in the mouse sperm: Unique localization, appearance, behavior and fate during fertilization. Cell Tissue Res. 2010, 340, 583–594. [CrossRef] [PubMed] Jankovicova, J.; Frolikova, M.; Sebkova, N.; Simon, M.; Cupperova, P.; Lipcseyova, D.; Michalkova, K.; Horovska, L.; Sedlacek, R.; Stopka, P.; et al. Characterization of tetraspanin protein CD81 in mouse spermatozoa and bovine gametes. Reproduction 2000, 152, 785–793. [CrossRef] [PubMed] Rubinstein, E.; Ziyyat, A.; Prenant, M.; Wrobel, E.; Wolf, J.P.; Levy, S.; Le Naour, F.; Boucheix, C. Reduced fertility of female mice lacking CD81. Dev. Biol. 2006, 290, 351–358. [CrossRef] [PubMed] Le Naour, F.; Rubinstein, E.; Jasmin, C.; Prenant, M.; Boucheix, C. Severely reduced female fertility in CD9-deficient mice. Science 2000, 287, 319–321. [CrossRef] [PubMed] Anifandis, G.; Messini, C.; Dafopoulos, K.; Sotiriou, S.; Messinis, I. Molecular and cellular mechanisms of sperm-oocyte interactions opinions relative ti in vitro fertilization (IVF). Int. J. Mol. Sci. 2014, 15, 12972–12997. [CrossRef] [PubMed] Zimmerman, B.; Kelly, B.; McMilla, B.J.; Seegar, T.C.M.; Dror, R.O.; Kruse, A.C.; Blacklow, S.C. Crystal structure of a full-length human tetraspanin reveals a cholesterol-binding pocket. Cell 2016, 167, 1041–1051. [CrossRef] [PubMed] Zhang, X.A.; Huang, C. Tetraspanins and cell membrane tubular structures. Cell. Mol. Life Sci. 2012, 69, 2843–2852. [CrossRef] [PubMed]

Int. J. Mol. Sci. 2018, 19, 1236

26.

27. 28.

29. 30.

31.

32.

33. 34.

35.

36.

37. 38.

39. 40.

41.

42.

43. 44.

45.

16 of 17

˙ Zyłkiewicz, E.; Nowakowska, J.; Maleszewski, M. Decrease in CD9 content and reorganization of microvilli may contribute to the oolemma block to sperm penetration during fertilization of mouse oocyte. Zygote 2010, 18, 195–201. [CrossRef] [PubMed] Evans, P.J. Sperm-Egg interaction. Annu. Rev. Physiol. 2012, 74, 477–502. [CrossRef] [PubMed] Berditchevski, F.; Odintsova, E.; Sawada, S.; Gilbert, E. Expression of the palmitoylation-deficient CD151 weakens the association of α3β1 integrin with the tetraspanin-enriched microdomains and affects integrin-dependent signalling. J. Biol. Chem. 2002, 277, 36991–37000. [CrossRef] [PubMed] Charrin, S.; Manié, S.; Oualid, M.; Billard, M.; Boucheix, C.; Rubinstein, E. Differential stability of tetraspanin/tetraspanin inetractions: Role of palmitoylation. FEBS Lett. 2002, 516, 139–144. [CrossRef] Charrin, S.; Maniá, S.; Thiele, C.; Billard, M.; Gerlier, D.; Boucheix, C.; Rubinstein, E. A physical and functional link between cholesterol and tetraspanins. Eur. J. Immunol. 2003, 33, 2479–2489. [CrossRef] [PubMed] Odintsova, E.; Butters, T.D.; Monti, E.; Sprong, H.; van Meer, G.; Berditchevski, F. Gangliosides play an important role in the organization of CD82-enriched microdomains. Biochem. J. 2006, 400, 315–325. [CrossRef] [PubMed] Yang, X.; Claas, C.; Kraeft, S.K.; Chen, J.B.; Wang, Z.; Kreidberg, J.A.; Hemler, M.E. Palmitoylation of tetraspanin proteins: Modulation of CD151 latera interactions, Subcellular distribution, and integrin-dependent cell morphology. Mol. Biol. Cell 2002, 13, 767–781. [CrossRef] [PubMed] Yanagimachi, R. The Physiology of Reprodution; Knobil, J.D.N., Ed.; Raven Press: New York, NY, USA, 1994; pp. 189–317. Sleight, S.B.; Miranda, P.V.; Plaskett, N.W.; Maier, B.; Lysiak, J.; Scrable, H.; Herr, J.; Visconti, P.E. Isolation and proteomic analysis of mouse sperm detergent-resistant membrane fractions: Evidence for dissociation of lipid rafts during capacitation. Biol. Reprod. 2005, 73, 721–729. [CrossRef] [PubMed] Hogue, I.B.; Grover, J.R.; Soheilian, F.; Nagashima, K.; Ono, A. Gag induces the coalescence of clustered lipid rafts and tetraspanin-enriched domains at HIV-1 assembly sites on plasma membrane. J. Virol. 2011, 85, 9749–9766. [CrossRef] [PubMed] Tanphaichitr, N.; Carmona, E.; Khalil, M.B.; Xu, H.; Berger, T.; Gerton, G. New insights into sperm-zona pellucida interaction: Involvement of sperm lipid rafts. Front. Biosci. 2007, 12, 1748–1766. [CrossRef] [PubMed] Sosnik, J.; Buffone, M.G.; Visconti, P.E. Analysis of CAPAZA3 localization reveals temporally discrete events during the acrosome reaction. J. Cell. Phys. 2010, 224, 575–580. [CrossRef] [PubMed] Inoue, N.; Satouh, Y.; Ikawa, M.; Okabe, M.; Yanagimachi, R. Acrosome-reacted mouse spermatozoa recovered from the perivitelline space can fertilize other eggs. Proc. Natl. Acad. Sci. USA 2011, 108, 20008–20011. [CrossRef] [PubMed] Stipp, C.S.; Kolesnikova, T.V.; Hemler, M.E. EWI-2 is a major CD9 and CD81 partner and membber of a novel Ig protein subfamily. J. Biol. Chem. 2001, 276, 40545–40554. [CrossRef] [PubMed] Sala-Valdés, M.; Ursa, A.; Charrin, S.; Rubinstein, E.; Hemler, M.E.; Sánchez-Madrid, F.; Yáñez-Mó, M. EWI-2 and EWI-F link the tetraspanin web to the actin cytoskeleton through their direct association with ezrin-radixin-moesin proteins. J. Biol. Chem. 2006, 281, 19665–19675. [CrossRef] [PubMed] Horvath, G.; Serru, V.; Clay, D.; Billard, M.; Boucheix, C.; Rubinstein, E. CD19 is linked to the integrin-associated tetraspans CD9, CD81, and CD82. J. Biol. Chem. 1998, 273, 30537–30543. [CrossRef] [PubMed] Sebkova, N.; Cerna, M.; Ded, L.; Peknicova, J.; Dvorakova-Hortova, K. The slower the better: How sperm capacitation and acrosome reaction is modified in the presence of estrogens. Reproduction 2012, 143, 297–307. [CrossRef] [PubMed] Ivankin, A.; Kuzmenko, I.; Gidalevitz, D. Cholesterol mediates membrane curvature during fusion events. Phys. Rev. Lett. 2012, 108, 238103. [CrossRef] [PubMed] Buschiazzo, J.; Ialy-Radio, C.; Auer, J.; Wolf, J.P.; Serres, C.; Lefevre, B.; Ziyyat, A. Cholesterol depletion disorganizes oocyte membrane rafts altering mouse fertilization. PLoS ONE 2013, 8, 1–13. [CrossRef] [PubMed] Schindelin, J.; Arganda-Carreras, I.; Frise, E.; Kaynig, V.; Longair, M.; Pietzsch, T.; Preibisch, S.; Rueden, C.; Saalfeld, S.; Schmid, B.; et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 2012, 9, 676–682. [CrossRef] [PubMed]

Int. J. Mol. Sci. 2018, 19, 1236

46. 47.

48. 49. 50. 51. 52. 53. 54.

55.

17 of 17

Laemmli, U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970, 227, 680–685. [CrossRef] [PubMed] Towbin, H.; Staehelin, T.; Gordon, J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some application. Proc. Natl. Acad. Sci. USA 1979, 76, 4350–4354. [CrossRef] [PubMed] The UniProt Consortium. UniProt: The universal protein knowledgebase. Nucleic Acids Res. 2017, 45, D158–D169. [CrossRef] Webb, B.; Sali, A. Comparative protein structure modeling using MODELLER. Curr. Protoc. Bioinform. 2014, 47. [CrossRef] Yang, J.; Yan, R.; Roy, A.; Xu, D.; Poisson, J.; Zhang, Y. The I-TASSER Suite: Protein structure and function prediction. Nat. Methods 2015, 12, 7–8. [CrossRef] [PubMed] Kozakov, D.; Hall, D.R.; Xia, B.; Porter, K.A.; Padhorny, D.; Yueh, C.; Beglov, D.; Vajda, S. The ClusPro web server for protein-protein docking. Nat. Protoc. 2017, 12, 255–278. [CrossRef] [PubMed] Lazaridis, T. Effective energy function for proteins in lipid membranes. Proteins 2003, 52, 176–192. [CrossRef] [PubMed] Jo, S.; Kim, T.; Iyer, V.G.; Im, W. CHARMM-GUI: A web-based graphical user interface for CHARMM. J. Comput. Chem. 2008, 29, 1859–1865. [CrossRef] [PubMed] Brooks, B.R.; Brooks, C.L., 3rd; Mackerell, A.D., Jr.; Nilsson, L.; Petrella, R.J.; Roux, B.; Won, Y.; Archontis, G.; Bartels, C.; Boresch, S.; et al. CHARMM: The biomolecular simulation program. J. Comput. Chem. 2009, 30, 1545–1614. [CrossRef] [PubMed] Schrodinger, L.L.C. The PyMOL Molecular Graphics System, Version 1.8; 2015. Available online: https: //pymol.org/2/support.html?#citing (accessed on 4 December 2015). © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).