Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2008 Apr 17;149(8):3942–3951. doi: 10.1210/en.2008-0281

Presence of Arylsulfatase A and Sulfogalactosylglycerolipid in Mouse Ovaries: Localization to the Corpus Luteum

Araya Anupriwan 1, Matthias Schenk 1, Kessiri Kongmanas 1, Rapeepun Vanichviriyakit 1, Daniela Costa Santos 1, Arman Yaghoubian 1, Fang Liu 1, Alexander Wu 1, Trish Berger 1, Kym F Faull 1, Porncharn Saitongdee 1, Prapee Sretarugsa 1,a, Nongnuj Tanphaichitr 1,a
PMCID: PMC2488217  PMID: 18420734

Abstract

Arylsulfatase A (AS-A) is a lysosomal enzyme, which catalyzes the desulfation of certain sulfogalactolipids, including sulfogalactosylglycerolipid (SGG), a molecule implicated in cell adhesion. In this report, immunocytochemistry revealed the selective presence of AS-A in the corpus luteum of mouse ovaries. Immunoblotting indicated that mouse corpus luteum AS-A had a molecular mass of 66 kDa, similar to AS-A of other tissues. Corpus luteum AS-A was active, capable of desulfating the artificial substrate, p-nitrocatechol sulfate, at the optimum pH of five. To understand further the role of AS-A in female reproduction, levels of AS-A were determined during corpus luteum development in pseudopregnant mice and during luteolysis after cessation of pseudopregnancy. Immunocytochemistry, immunoblotting and desulfation activity showed that AS-A expression was evident at the onset of pseudopregnancy in the newly formed corpora lutea, and its level increased steadily during gland development. The increase in the expression and activity of AS-A continued throughout luteolysis after the decrease in serum progesterone levels. We also observed the selective presence of SGG on the luteal cell surface in developed corpora lutea, as shown by immunofluorescence of mouse ovary sections as well as high-performance thin-layer chromatography of lipids isolated from mouse and pig corpora lutea. The identity of the “SGG” band on the thin layer silica plate was further validated by electrospray ionization mass spectrometry. Significantly, SGG disappeared in regressing corpora lutea. Therefore, lysosomal AS-A may be involved in cell-surface remodeling during luteolysis by desulfating SGG after its endocytosis and targeting to the lysosome.

THE CORPUS LUTEUM is a temporary endocrine gland formed in the ovary after ovulation. It consists of large and small luteal cells, derived from granulosa and theca cells, respectively. In contrast to the mature follicle, which consists of a cumulus oocyte complex surrounded by a substantial volume of follicular fluid, the corpus luteum is a solid cellular structure. Cell-cell communication and adhesion contribute to corpus luteum function, primarily production of progesterone essential for maintaining pregnancy (1,2). In mice, which have a short luteal phase in the normal estrous cycle, the corpus luteum does not become fully developed. Full development occurs only when the female is bred or becomes pseudopregnant due to cervical stimulation. The corpora lutea are maintained throughout gestation in many species, including the mouse and pig. At the end of pregnancy or pseudopregnancy [8–10 d in mice (3)], the corpora lutea undergo regression (luteolysis) induced by prostaglandin F2α (PGF2α) (4,5). Functional regression of the luteal cells is associated with cessation of their progesterone secretory activity. Subsequently, structural luteolysis takes place with apoptosis of luteal cells (6,7,8), and degeneration and processing of the extracellular matrices (9,10,11). At this time, autophagic vacuoles and lysosome-like bodies also increase in number in conjunction with involution of cellular structures (12). Cytochemical studies indicate that acid hydrolases, including acid phosphatase(s) and arylsulfatase(s), are present in these organelles in the regressing luteal cells, and the increase in their activities is correlated with the progressive state of luteal regression (12,13,14,15).

Arylsulfatase A (AS-A) (Enzyme Commission of the International Union of Biochemistry 3.1.6.8) is an evolutionarily conserved enzyme in the arylsulfatase family (16). AS-A is localized to the lysosome and has a pH optimum of five. Sulfogalactosylceramide (SGC) and sulfogalactosylglycerolipid (SGG), implicated in cell adhesion and signaling (17,18) as well as in lipid raft formation (19,20), are natural substrates of AS-A. The enzyme, mediated by its cofactor, saposin B, desulfates SGG and SGC to generate galactosylglycerolopid (GG) and galactosylceramide (GC), respectively (21,22,23). In the brain this desulfation is likely the first step of the SGC turnover pathway, and the significance of AS-A in this process is well recognized. Individuals genetically deficient in AS-A suffer a neurological disorder, metachromatic leukodystrophy, due to SGC accumulation in the brain (16,24). We have shown that SGG present on the sperm surface plays a significant role in sperm-egg interaction. This is attributed to the direct affinity of SGG for the egg extracellular matrix (ECM), the zona pellucida (ZP) (19,25). In addition, SGG plays a role in the formation of sperm lipid rafts, which are platforms on the sperm surface for ZP binding (26). However, SGG may be desulfated by AS-A during the sperm acrosome reaction, induced by a ZP glycoprotein (27). Galactosylglycerolipid (GG), the desulfated product of SGG, does not have affinity for the ZP (25), suggesting that acrosome-reacted sperm remain bound to the egg ECM via other ZP binding molecules on the sperm surface. The presence of AS-A in female reproductive tissues has also been described (28,29). Therefore, it is likely that the natural substrates of AS-A, SGG/SGC, may also exist in these tissues, and the cell adhesion phenomena in which these sulfoglycolipids partake may be temporally regulated by the desulfation activity of AS-A.

In previous cytochemical studies (12,14), arylsulfatases were localized to the corpus luteum, using an artificial substrate, p-nitrocatechol sulfate (NCS), at pH 5. Although it was tempting to speculate that the activity detected in these studies belonged to AS-A, based on its pH optimum of five, further studies to validate this postulation were needed because this previous qualitative cytochemical localization of arylsulfatases was done at only one pH and with only the artificial substrate, NCS. In this report we describe the immunolocalization of AS-A in mouse ovaries using an antibody specific to AS-A. The selective presence of enzymatically active AS-A in the corpus luteum led us to question whether its natural sulfogalactolipid substrates, SGG/SGC, existed in this gland and whether these sulfoglycolipids, having a cell adhesion property, may play roles in corpus luteum formation and functions. Furthermore, whether the levels of SGG/SGC were decreased as part of cell-surface remodeling during luteolysis was addressed.

Materials and Methods

Animals

Experiments involving mice were performed at Mahidol University, Thailand, and at the Ottawa Health Research Institute, Canada. In Thailand, ICR mice, obtained from the Thai National Laboratory Animal Center, Mahidol University, were used. In Canada, CF-1 females and vasectomized CD-1 males were purchased from Charles River Canada (St-Constant, Quebec, Canada). In all cases, animals were kept in a controlled temperature room with a 14-h dark, 10-h light cycle. Their use was approved by the Animal Care Committee of each institute.

Superovulated females.

Female mice (5 wk old) were induced to superovulate by sequential ip injections with 10 IU pregnant mare serum gonadotropin (PMSG) (Sigma-Aldrich, St. Louis, MO), followed by 10 IU human chorionic gonadotropin (hCG) (Sigma-Aldrich) 44–48 h later. The animals were killed 16 h after hCG, and ovaries were excised and fixed in Bouin’s solution for immunocytochemistry of AS-A. Corpora lutea were dissected from ovaries under a stereomicroscope and kept at −80 C for subsequent immunoblotting and lipid isolation (see below). In experiments in which immunofluorescence of SGG was performed, ovaries containing corpora lutea at different stages were collected from 5-wk-old superovulated animals that were killed 2.5 and 7 d after hCG.

Pseudopregnant mice.

Superovulated 23-d-old mice (PMSG and hCG primed as described previously for 5 wk old females) were housed with vasectomized males (∼10 wk old, produced as described below for experiments at Mahidol University, or purchased from Charles River Canada for experiments at Ottawa Health Research Institute), immediately after the hCG injection to the females. The pairing day was assigned as experimental d 0. Vaginal plugs were checked the day after mating (experimental d 1) for evidence of copulation, and only females with plugs were considered to be pseudopregnant. Blood samples were collected by cardiac puncture from these mated females on experimental d 0, 2, 4, 8, 9, 10, 13, or 15 after anesthetization with isoflurane. Serum was collected from coagulated blood by centrifugation (100 g, 15 min, 4 C) and stored frozen at −20 C for subsequent progesterone assays, which were done at the Bangkok Pathology Laboratory Co., Ltd. (Bangkok, Thailand), using the Bayer’s ACS:180 automated chemiluminescence system (E for L International Ltd., Bangkok, Thailand). These pseudopregnant mice were killed, and their dissected ovaries were subjected to immunocytochemistry, immunoblotting, and assessment of AS-A activity.

Vasectomies were performed under anesthesia by ip injection with Nembutal (Ceva Sante Animale, Libourne Cedex, France) at 4 mg/100 g body weight. The abdominal wall was incised just above the preputial gland. The vas deferens was teased away from the surrounding fat and ligated with catgut (no. 3–0) thread at two positions approximately 1-cm apart. After a cut was made between these two ligation sites, the abdominal wall was sutured, and the mice were allowed to recover for 2 wk. Successful vasectomy was confirmed by the absence of sperm in ejaculates, thus indicating inability of the animals to impregnate females.

Pigs were also used for collection of corpora lutea. The animals were maintained for research use in accordance with the Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching, and with approval from the Animal Use and Care Advisory Committee, University of California Davis. Corpora lutea were collected from pigs on d 5 and 15 of the estrous cycle and d 26 of pregnancy. Some of the females were injected with PGF2α (Lutalyse; Upjohn, Kalamazoo, MI) on d 14 of the estrous cycle to initiate luteal regression, and regressing corpora lutea were collected on the following day. Day zero was defined as the first day of behavioral estrus, detected by receptivity to a mature boar. Pigs were killed by electrocution, followed by exsanguination in a federally inspected (U.S. Department of Agriculture) slaughter plant. Pregnancy was verified by the presence of embryos in the uterus. Ovaries were removed and corpora lutea dissected from the surface of the ovary with a scalpel. Removed corpora lutea were immediately frozen in liquid nitrogen until use for lipid extraction.

Antibodies

Rabbit polyclonal anti-AS-A antiserum directed against purified AS-A from pig sperm plasma membranes was generated, and its IgG fraction was purified as described previously (30). Immunoblotting revealed that this antibody recognized purified human liver AS-A (31) (provided by Dr. Arvan Fluharty, University of California at Los Angeles, Los Angeles, CA) and a 66-kDa band of pig and mouse sperm proteins (21,32). IgG from preimmune serum (PRS) of the rabbits used for anti-AS-A production was prepared and used as a negative control of anti-AS-A IgG.

Rabbit polyclonal anti-SGG/SGC IgG was prepared and affinity purified according to our previously described method (25) except that SGG multilamellar liposomes were used for the affinity purification instead of the SGG BioSil matrix. This affinity purified antibody reacts with both SGG and SGC (25). Mouse monoclonal anti-SGC IgM (O4) antibody was also obtained from Neuromics (Edina, MN); this antibody shows positive immunofluorescent staining with the mouse sperm head, which contains SGG as the only sulfoglycolipid (25), and the staining pattern is the same as that obtained with our affinity purified rabbit polyclonal anti-SGG/SGC antibody (our unpublished data). Therefore, both of these antibodies can recognize SGG and SGC.

Immunocytochemistry of AS-A in the murine ovary

Ovaries dissected from killed animals, including superovulated 5-wk-old females, 23- to 36-d-old females in the “pseudopregnancy” experiment, and 20-d-old prepubertal females, were fixed in Bouin’s solution, and processed for paraffin embedding and sectioning. Standard procedures were used to deparaffinize and rehydrate the sections (4 μm thick), as well as to quench endogenous peroxidase and remove residual picric acid. After neutralization of residual formaldehyde in the tissues with 300 mm glycine, antigen was retrieved by microwaving the tissues immersed in 0.01 m citrate buffer (pH 6.0) for 3 min at the highest power and an additional 7 min at low power. To block nonspecific binding, tissues were incubated for 30 min at room temperature with the blocking solution [10% normal goat serum in Tris-buffered saline (TBS): 20 mm Tris-HCl, 150 mm NaCl (pH 7.4)]. Tissues were incubated (90 min, room temperature; or overnight, 4 C) with 10 μg/ml anti-AS-A IgG in the blocking solution. After successive washing with 0.1% Tween 20 in TBS, antigen-antibody interaction was detected using the Vectastain ABC Elite kit (Vector Laboratories, Inc., Burlingame, CA). This involved treatment of the section with biotinylated secondary antibody (goat antirabbit IgG) (30 min, room temperature), followed by incubation with avidin-biotin-horseradish peroxidase complexes and detection by reaction with a peroxidase substrate, diaminobenzidine. Concentrations of chemicals and conditions for these treatments were as described by the manufacturer. The brown color product of the peroxidase reaction signified AS-A localization. Sections were also stained with a hematoxylin solution to counterstain nuclei. In alternate sections, anti-AS-A IgG was preadsorbed with 5 μg/ml AS-A purified from sperm as previously described (30). Tissues treated with 10 μg/ml rabbit PRS IgG served as additional negative controls.

Immunofluorescence of SGG in the murine ovary

Immunofluorescence of SGG was performed on frozen ovary sections. The ovaries collected from pseudopregnant d-8 mice and from superovulated females 2.5 and 7 d after hCG were embedded in Tissue-Tek O.C.T. compound (Sakura FineTek Japan, Tokyo, Japan) and directly frozen in liquid nitrogen. Cryosections, approximately 15-μm thick, were prepared and mounted onto poly-l-lysine-coated slides. After air-drying, the sections were fixed in 4% paraformaldehyde in PBS for 5 min, rinsed three times in PBS, and preincubated in blocking buffer (2% BSA and 2% normal goat serum in PBS containing 0.1% Tween 20) for 1 h. The sections were then incubated with mouse anti-SGC/SGG IgM (15 μg/ml) or rabbit anti-SGG/SGC IgG (10 μg/ml) in blocking buffer overnight at room temperature. Sections were washed three times in PBS and incubated (1 h, room temperature) with Alexa-488 conjugated goat antimouse-IgM (Invitrogen-Molecular Probes-Burlington, Ontario, Canada) at 1:200 dilution in blocking buffer, or with Alexa-488 conjugated goat antirabbit-IgG (Invitrogen-Molecular Probes) at 1:200 dilution, for the detection of the anti-SGC/SGG IgM and anti-SGG/SGC IgG reactivity, respectively. After washing the sections in PBS, nuclei were stained with TO-PRO-3 (Invitrogen-Molecular Probes) at 1:1000 dilution in PBS for 5 min. Finally, sections were rinsed in PBS and mounted with Vectashield (Vector Laboratories) underneath coverslips to minimize bleaching of the fluorescence. Negative controls were sections that were exposed to normal mouse IgM or rabbit IgG instead of the primary antibodies. The sections were examined by an Olympus FV1000 confocal laser-scanning microscope (Olympus, Tokyo, Japan). Confocal fluorescence images were taken at intervals of 0.5 μm. Confocal stacks (three to five images) were reconstructed using the Olympus Fluoview software. The same section was then stained with 1% methylene blue for histological examination of the corpus luteum stages (developed or regressing) under a bright-field microscope.

Immunoblotting of ovaries and isolated corpora lutea for AS-A

Corpora lutea collected from superovulated 5-wk-old female mice were placed in a 1.5-ml microcentrifuge tube and washed once with cold TBS containing 1 mm phenylmethanesulfonyl fluoride, 10 μg/ml aprotinin, 10 mm EDTA, 0.1% β-mercaptoethanol, and 0.1% Tween 20, and gently vortexed in the same buffer to remove loosely attached stromal cells, which remained in the supernatant, while the corpora lutea settled to the bottom of the tube under gravity. This washing procedure was repeated twice. The washed corpora lutea were homogenized in the same buffer and centrifuged at a low speed to remove cellular particulates. Total proteins were determined in the supernatant using Bio-Rad Protein Assay Solution (Bio-Rad Laboratories, Inc., Hercules, CA), and proteins in the supernatant were separated by SDS-PAGE (12% polyacrylamide gel) (33), followed by immunoblotting (34) for AS-A. Ovaries collected from 23- to 36-d-old females in the “pseudopregnancy” experiment were similarly prepared for immunoblotting. They were homogenized in radioimmunoprecipitation assay buffer [150 mm NaCl, 1% Triton X-100 in 50 mm Tris-HCl (pH 5)]. After quantification of total proteins, samples were subjected to gel electrophoresis, immunoblotting, and measurement of enzymatic activity.

Nitrocellulose membranes containing proteins from the corpus luteum or ovary were blocked with 5% powdered skim milk in TBS before incubation with the primary antibody. AS-A antiserum was used at 1:3000 in TBS-5% powdered skim milk. TBS-0.01% Tween 20 was used for washing the membrane after each incubation step. Horseradish peroxidase-conjugated goat antirabbit IgG (heavy and light chain) IgG (Bio-Rad Laboratories), diluted 1:5000 in TBS-5% powdered skim milk was used as the secondary antibody. Antibody-antigen reactivity was detected by chemiluminescence using an ECL kit from Pierce (Rockford, IL).

Determination of AS-A enzymatic activity

The artificial substrate, NCS, was used for assaying AS-A activity in ovarian homogenates, as previously described (21). Purified pig sperm AS-A (21) was used as a positive control. One unit of AS-A activity was defined as 1 μmole nitrocatechol released in 1 h.

Characterization of SGG in corpus luteum lipids

Lipids were extracted from isolated corpora lutea of 5-wk-old superovulated mice, cyclical d-5 (E-D5) and -15 (E-D15) pigs, and pregnant d-26 (P-D26) pigs using the modified Bligh-Dyer’s method (35,36), and subjected to high-performance thin-layer chromatography (HPTLC), using a 10- × 10-cm plate (HPK silica 60-Å, particle size: 5 μm, thickness 200 μm) from Whatman (Kent, UK), and chloroform/methanol/H2O (65:25:4, vol/vol/vol) as the plate developing solvent, as previously described (37). The developed plate was stained for glycolipids by orcinol charring (36) and post-stained for total lipids with Coomassie blue (CB) G-250 (37). To detect sulfolipids, samples run in parallel on another HPTLC plate were stained with an azure A solution (38). Co-chromatographed lipid standards included SGC (bovine brain, a mixture of nonhydroxylated and hydroxylated forms; Sigma-Aldrich), GC (bovine brain, nonhydroxylated form; Sigma-Aldrich), cholesterol, and SGG and GG, both prepared from pig testis in our laboratory, as previously described (39,40). The identity and purity of the prepared SGG and GG were confirmed by negative ion electrospray (ESI) and positive ion atmospheric pressure chemical ionization mass spectrometry (MS), respectively. Although the data were not shown for GG, the typical ESI spectrum of this purified SGG, obtained from a parallel run with the putative “SGG” band isolated from pig corpus luteum lipids, could be viewed (see Fig. 6).

Figure 6.

Figure 6

Open in a new tab

ESI-MS confirms the presence of SGG as one component of the sulfoglycolipid band in corpora lutea of pregnant pigs. Negative ion ESI mass spectra (A and B) and negative ion tandem mass spectra (C and D, fragment ion scans, parent ion-labeled P) of HPTLC-purified lipid extract from pig corpora lutea (A and C) and SGG purified from pig testis (B and D).

ESI-MS was performed on the putative “SGG band” scraped from HPTLC plates on which pig corpus luteum lipids were separated. Total lipids from corpora lutea (90 mg wet tissue weight) from P-D26 pigs were applied to three HPTLC plates as a broad band (width = 5 cm) on each plate. Lipid standards (SGG, SGC, GG, and GC) were applied as a narrow band on the edge of each plate. The plates were developed as described previously. Each plate was then cut lengthwise to produce a strip that included the lipid standards and a small portion of the corpus luteum lipids. This cut strip was subjected to orcinol staining/charring to locate the glycolipid bands, including SGG. This stained strip was then realigned with the unstained part of the plate, and the area on the unstained part that matched the standard SGG band was scraped. The pooled scraped silica was subjected to lipid extraction in 2 ml chloroform/methanol (1:1, vol/vol). This silica mixture was incubated (4 C, overnight) and centrifuged (340 g, 3 min) to pellet the silica powder, leaving the extracted lipids in the supernatant. Lipids were re-extracted from the pellet following the same steps as in the first extraction round, and the two supernatants were pooled and subjected to the Bligh-Dyer extraction (36). Lipids extracted into the chloroform phase were dried under a stream of N2 and resuspended in 250 μl chloroform. Of these extracted lipids, 10% were subjected to HPTLC with orcinol staining to reconfirm the composition (Fig. 5C). The remainder of the lipid sample was dried and reserved for MS analyses.

Figure 5.

Figure 5

Open in a new tab

HPTLC showing the presence of SGG in mouse and pig corpora lutea. A, Mouse corpora lutea. Lipids were extracted from 34 corpora lutea, dissected from ovaries of 5-wk-old superovulated mice. B, Pig corpora lutea. Lipids were extracted from 10 mg corpora lutea of P-D26 pigs and from 30 mg corpora lutea of E-D5 pigs. Lipid samples used in both A and B were divided into two halves, each of which was applied as a lane on the silica plate for HPTLC. Lane 1 was subjected to orcinol staining/charring (1a), followed by CB staining (1b), whereas the other lane (lane 2) was stained with azure A. Glycolipids stained purple (marked with *), and phospholipids appeared brown (•) with orcinol staining/charring (lane 1a). Post-staining of lane 1a with CB revealed all lipids blue (lane 1b). Lane 2, Sulfolipids stained blue with azure A. Note the presence of the putative SGG band (pointed by an arrow) with the same Rf as standard SGG in corpus luteum lipids from superovulated mice and both P-D26 and E-D5 pigs, as revealed from the three types of staining. C, HPTLC pattern of the scraped lipid band that was used for ESI-MS in Fig. 6. The area around the pig corpus luteum lipid band with the same Rf as SGG standard (marked by a bracket) was scraped off HPTLC plates of the P-D26 sample. One tenth of the lipids extracted from the scraped silica powder was re-subjected to HPTLC. Note that SGG constituted the main band of the extracted lipids, although two additional lipid bands, one above and one below the putative “SGG” band, were coextracted. Data shown are representative of duplicate experiments. Chol, Cholesterol.

ESI-MS was performed by redissolving the dried lipid extracts in methanol/chloroform (4:1, vol/vol, 100 μl). The sample was injected (20 μl/injection) into a stream of the same solvent flowing (20 μl/min) into an ESI (Ionspray) source connected to a triple quadrupole mass spectrometer (API III+; PerkinElmer Sciex, Thornhill, Ontario, Canada). Negative and positive ion mass spectra were recorded by scanning from m/z 100–1500 (orifice 70 V, 0.3 Da step size, 5 sec/scan). Fragment ion spectra of preselected parents and parent ion spectra of preselected fragments were recorded by scanning Q3 and Q1, respectively, under tandem mass spectrometric conditions from m/z 50–1500 (orifice varied between 70 and 200 V, collision gas thickness instrumental setting between 200 and 330, 5–6 sec/scan). Instrument-supplied software was used to average all the spectra from each sample injected.

Results

Presence of AS-A in the mouse corpus luteum

Immunocytochemistry revealed the presence of AS-A in the ovaries of 5-wk-old superovulated mice. AS-A was localized selectively to the corpus luteum, and not in any follicular cells (Fig. 1A, b and c). In the ovary of prepubertal 20-d-old mice, in which the corpora lutea had not yet formed, the staining of AS-A was observed in the theca and stromal cells, although the staining intensity in these cells was much lower than that in luteal cells (Fig. 1A, a). The specificity of anti-AS-A IgG was revealed by the lack of immunostaining when PRS IgG was used in place of the primary antibody (Fig. 1A, d). Furthermore, anti-AS-A IgG preadsorbed with purified AS-A gave minimal staining (data not shown).

Figure 1.

Figure 1

Open in a new tab

A, Immunocytochemistry of AS-A in the mouse ovary. Panel a, An ovary section from a 20-d-old female. Panels b–d, Ovary sections from a superovulated postpubertal female. Panel d, Negative control, PRS IgG was used in place of the primary antibody. Panel c is a high-magnification image of an area of the ovary section shown in panel b. Note that AS-A was localized specifically to the corpus luteum (CL) of the superovulated female, and to the theca and stromal cells (hollow and solid arrow, respectively) of the 20-d-old mouse. AS-A staining was absent in follicular cells in the follicles (F). Although the egg showed some staining, it was likely to be nonspecific. Black bar, 500 μm. White bar, 50 μm. Data shown are representative of four replicate experiments. B, Immunoblotting of mouse corpora luteum proteins for AS-A. Proteins extracted from three corpora lutea collected from superovulated postpubertal mice were used for the “mouse CL” lane. Purified pig sperm AS-A (50 ng) was coelectrophoresed (left lane) as the positive control. Data shown are representative of three replicate experiments.

The existence of AS-A in corpora lutea of superovulated postpubertal mice was confirmed by immunoblotting of isolated corpora lutea, which revealed the main band of AS-A at 66 kDa and the minor band at 58 kDa (Fig. 1B). Notably, the main AS-A band in corpora lutea (Fig. 1B, right lane) had the same molecular mass as AS-A found in pig sperm (Fig. 1B, left lane) and other mammalian tissues (16,21,31). The 58-kDa AS-A band of the corpora lutea may be the degraded product of the 66-kDa form because the additional 58-kDa AS-A band was also observed after storing purified sperm AS-A for a prolonged time period (data not shown). In contrast, immunoblotting of cumulus cells isolated from cumulus oocyte complexes, as previously described (30), revealed the absence of AS-A (data not shown).

Levels of AS-A in developing and regressing corpora lutea

Because AS-A existed selectively in the corpora lutea, we further investigated whether its expression level and NCS desulfation activity changed temporally with the development and regression of the tissue. In ovulating rodents, functional corpora lutea do not develop fully without cervical stimulation, and corpora lutea from the preceding estrous cycle do not totally regress (1,2). Therefore, pseudopregnancy was generated in prepubertal mice to observe full development and subsequent regression of the corpora lutea in a timely manner, using serum progesterone levels as a marker for the development and regression phases (3,41). Figure 2A illustrates that levels of serum progesterone in these young females increased rapidly after mating with vasectomized males and reached the maximum on experimental d 4, indicating that pseudopregnancy had occurred. The progesterone level remained high until experimental d 8, then decreased markedly on d 9, indicating that the pseudopregnancy lasted only 8 d. The progesterone levels further declined as seen on experimental d 10, 13, and 15, indicating that functional luteolysis began on d 9. Changes of the progesterone levels in these pseudopregnant mice were similar to those previously described (41). Interestingly, the specific activity (Fig. 2B) and expression, as shown by immunoblotting (Fig. 2C), of AS-A were observed throughout the entire period of pseudopregnancy and luteolysis, and, in fact, were steadily increasing from experimental d 0–13. On experimental d 0, corpora lutea would not have yet been formed, and the activity and expression of AS-A would rather reflect the presence of the enzyme in theca and stromal cells (Fig. 1A, a). Immunocytochemistry results (Fig. 3) corroborated the enzymatic activity and immunoblotting results of AS-A levels. The intensity of the AS-A signal in the corpora lutea of ovarian sections was the highest on experimental d 13 (Fig. 3, e and f). Corpora lutea of the pseudopregnant d-8 animals had more intense labeling than those of the pseudopregnant d-4 animals (Fig. 3, c and d vs. a and b). Notably, there was a minor population of corpora lutea that labeled with a lower intensity for AS-A (Fig. 3e, arrow) in the pseudopregnant d-13 females. The reason for this staining intensity variation is unclear, although some areas with the low intensity showed luteal cells at a more regressing state than those luteal cells with higher anti-AS-A reactivity; they showed vacuoles and spaces between adjacent cells (data not shown). In all stages of the corpus luteum, intracellular staining of AS-A was apparent. It is likely that AS-A existed in the lysosomes of the luteal cells, given the fact that AS-A is localized to these organelles in other somatic cells (16). However, labeling of AS-A on the luteal cell surface may also be present (Fig. 3, b, d, and f).

Figure 2.

Figure 2

Open in a new tab

Serum progesterone levels and specific activity and expression of AS-A during pseudopregnancy and luteolysis. Day zero was the day when the 23-d-old females primed with PMSG and hCG (see Materials and Methods) were paired with vasectomized males. A, Serum progesterone level. B, Specific activity of AS-A (expressed as unit per mg of total proteins in the ovarian homogenate or unit per mg of wet weight of the ovary piece used for homogenization) in ovarian homogenates from various days of pseudopregnancy and luteolysis. C, Immunoblotting of AS-A in the ovary of the prepubertal mice used in the vasectomized male mating experiment. Samples shown were from mice of experimental d 0, 4, and 8 (pseudopregnant), and d 13 (containing corpora lutea undergoing luteolysis). Also shown in comparison was purified native AS-A from pig sperm. In each experiment at least three females were used for each time point. Data shown in A are the average of the means from duplicate experiments; those shown in B and C are from three replicate experiments. Values of AS-A specific activities are means ± sd values of the means from these three experiments.

Figure 3.

Figure 3

Open in a new tab

Immunocytochemistry of ovary sections showing various levels of AS-A in the corpus luteum during pseudopregnancy and luteolysis. Panels A and C (magnified in panels B and D), Ovary sections during corpus luteum development of pseudopregnant mice (d 4 and 8 in the “pseudopregnancy” experiment, respectively). Panels E and F, An ovary section during luteolysis (experimental d 13, see progesterone levels in Fig. 2A for indication of functional luteolysis). Lower panels (B, D, and F) show magnified areas of the top panels (A, C, and E, respectively). Note that the intensity of the AS-A staining in the corpus luteum was in the following order: d 13 > d 8 > d 4. On experimental d 13, a minor population of the corpora lutea stained for AS-A with less intensity (arrow in panel E). Data shown are representative of duplicate experiments.

Presence of SGG in the corpus luteum

SGG and SGC are natural substrates of AS-A. Therefore, we evaluated corpora lutea for the presence of these substrates. Immunofluorescence using mouse anti-SGC/SGG IgM indicated the selective presence of SGG/SGC (green fluorescence signals) in developed corpora lutea of pseudopregnant d-8 mice and post-hCG d-2.5 superovulated mice, respectively (Fig. 4, A and B). High-magnification microscopy showed that luteal cells in corpora lutea of pseudopregnant d-8 mice were larger than those of post-hCG d 2.5 mice, indicating that corpora lutea of pseudopregnant d-8 mice were more developed than those of the ovulated mice. SGG/SGC was apparently on the surface of these large luteal cells (Fig. 4A, inset). Nonetheless, some stromal cells and theca cells also contained SGG/SGC (Fig. 4, A and B). When mouse IgM was used in place of anti-SGC/SGG IgM, no green fluorescence staining was observed (Fig. 4C), indicating the specific staining of SGC/SGG by anti-SGC/SGG IgM. The same results were obtained when affinity purified rabbit anti-SGG/SGC IgG (see Materials and Methods) was used (data not shown). However, in regressing corpora lutea, which were present in the ovary of post-hCG d-7 superovulated mice, SGG/SGC staining was minimal, although positive staining of SGG/SGC remained in some stromal cells and theca cells in this ovary (Fig. 4D). The regression of these corpora lutea was evidenced by the morphology of the luteal cells; increased cytoplasmic area and/or intercellular space between luteal cells were observed, and some luteal cells contained smaller nuclei with an unrounded structure (Fig. 4, E and F).

Figure 4.

Figure 4

Open in a new tab

Immunofluorescence of SGG in mouse ovary cryosections using anti-SGC/SGG. Green fluorescence represented the staining of SGG, whereas red fluorescence was from TO-PRO-3 staining of the nucleus. Ovaries used for the study included those with fully developed corpora lutea, obtained from a pseudopregnant d-8 mouse (panel A), those with developed corpora lutea, obtained from a post-hCG d-2.5 superovulated postpubertal mouse (panel B), and those containing a large number of regressing corpora lutea, obtained from a post-hCG d-7 superovulated postpubertal mouse (panels D–F). Note selective SGG staining in fully developed and developed corpora lutea (panels A and B). A higher magnification image of the section in panel A (inset) strongly suggested the localization of SGG to the cell surface. In contrast, SGG staining disappeared in regressing corpora lutea (panel D with corresponding methylene blue image shown in panel E). Panel F shows a higher magnification image of the boxed area in panel E; note the presence of nuclei with smaller sizes and unrounded morphology in the regressing luteal cells (arrows), as well as extended cytoplasmic areas and/or intercellular spaces. This morphology was in contrast to that of luteal cells in a developed corpus luteum, as shown in the inset of panel B. In panels A and D, SGG was also localized to the surface of stromal and theca cells. Panel C is a negative control; a section from a pseudopregnant d-8 mouse was incubated with a normal mouse IgM (15 μg/ml) instead of anti-SGC/SGG IgM. Bars in the main image of panels A–E, 100 μm, whereas those in panel F and in the inset of panels A and B, 20 μm. Data shown are representative of four replicate experiments. AF, Antral follicle; CL, developed corpora lutea; PF, preantral follicle; rCL, regressing corpora lutea.

Because the results from the immunofluorescence studies could not differentiate between SGC and SGG in the developed corpus luteum, we further characterized the identity of the sulfoglycolipid using biochemical approaches. Figure 5A shows the HPTLC pattern of lipids extracted from isolated corpora lutea of superovulated postpubertal mice. Approximately eight lipid bands were glycolipids as revealed by purple staining with orcinol (marked by an asterisk in lane 1a). Phosphatidylcholine (PC) and phosphatidylethanolamine (PE) were two phospholipid bands that were charred brown after exposure to orcinol solution (marked by • in lane 1a) (36,38). One of the glycolipid bands had the same Rf (0.329) as SGG (marked by an arrow in Fig. 5A, lane 1a). Although the band was faintly stained purple with orcinol, it was post-stained more intensely with CB (Fig. 5A, lane 1b), which stains all lipids with greater sensitivity (37). This lipid band was also stained positively with azure A, indicating that it was a sulfolipid (36,38) (Fig. 5A, lane 2). Collectively, these results strongly suggested that the lipid band from the mouse corpora lutea contained SGG. A few additional lipid bands with higher mobilities than SGG were also stained positive with azure A; however, the SGC band (Rf = 0.309 and 0.278 for the nonhydroxylated and hydroxylated species, respectively, in our HPTLC system) was not present in lipids isolated from mouse corpora lutea (Fig. 5A, lane 2). Figure 5B shows that a lipid band with the same Rf as SGG and positive staining with orcinol and azure A was also present in the lipid extract of corpora lutea from P-D26 pigs, and, to a lesser extent, in the extract of corpora lutea from E-D5 pigs. Note that lipids loaded onto the HPTLC plate from this latter sample were extracted from 3-fold more tissue wet weight, compared with the lipids from corpora lutea of P-D26 pigs. Notably, our NCS desulfation assay revealed that the corpora lutea of P-D26 pigs also contained much higher AS-A activity than that of E-D5 pigs (0.941 ± 0.163 vs. 0.145 ± 0.072 U/mg protein). Immunoblotting also confirmed the presence of AS-A in pig corpora lutea, with the same molecular mass (66 kDa) as that in the mouse (data not shown). Furthermore, our unpublished results indicated the absence of detectable amounts of both SGG and AS-A in isolated granulosa cells, retrieved from PMSG-primed postpubertal mice. All of these results suggested the selective existence of AS-A and SGG in the corpus luteum in the ovary.

Because larger tissue quantities are available from pig corpora lutea, lipids from pig corpora lutea were used as a source to prepare the putative “SGG” band for ESI-MS analyses. This putative pig corpus luteum SGG band was scraped from the HPTLC plate and prepared for ESI-MS. Although attempts were made to scrape only the band with the same Rf as SGG, lipids that ran adjacently to this band were also coextracted (Fig. 5C). The negative ion ESI spectrum of this partially purified SGG material revealed a complex of ions in the m/z 700–1000 region (Fig. 6A). This complex contained a significant signal at m/z 795.5, which within the accuracy of the instrument used was indistinguishable from that calculated (795.53 Da for C41H79O12S) and experimentally observed (m/z 795.5, Fig. 6B) for the (M-H)− ion from authentic SGG. Parent ion (m/z 97.1, HSO4−) tandem mass spectra of the partially purified lipid extract revealed a weak signal at m/z 796.0, providing confirmation of the presence of SGG in the extract (data not shown). However, more convincing were the fragment ion tandem mass spectra of the lipid extract m/z 795.5 parent, which revealed fragment ions at m/z 539.3 and 96.8, attributable to loss of the ester side chain (795.5-C16H32O2, calculated 539.29 Da) and sulfate (HSO4−, calculated 97.07 Da), respectively (Fig. 6C). These ions are the only significant fragment ions in the tandem mass spectrum of authentic SGG (Fig. 6D).

Because SGG appeared to be significantly decreased in regressing mouse corpora lutea (Fig. 4D), we further asked whether the same phenomenon would be observed in the pig ovary. Corpora lutea of E-D15 pigs and P-D26 pigs are well developed. However, injection of PGF2α into the pigs on cyclical d 14 induces regression of the corpora lutea. Figure 7 shows the comparison of the lipid profiles of corpora lutea collected from E-D15, P-D26, and PGF2α-treated (R) pigs. As expected, the SGG (band 5*) and other glycolipid bands (all denoted by an asterisk) existing in the P-D26 corpora lutea were also present in lipids isolated from developed corpora lutea of E-D15 pigs, although the distribution of band 1* and band 3* glycolipids was higher in the E-D15 sample. In addition, band 4* appeared selectively in the E-D15 sample. Besides SGG, the identity of other glycolipids is unknown and under our investigation at the present time. Nonetheless, the SGG band in both E-D15 (data not shown) and P-D26 (Fig. 5B, rightlane 2) samples stained blue with azure A, indicating that it was sulfated. Significantly, in the regressing pig corpora lutea (R), the level of SGG was minimal (Fig. 7, middle lane), an observation that substantiated the results observed in the mouse system (Fig. 4). Levels of glycolipid bands 1–4 were also much reduced; however, the distribution of glycolipid bands 6–8 was increased in the R sample. Similar to the results observed in the mouse system, regressing (R) pig corpora lutea still contained an appreciable level of AS-A specific activity (0.584 ± 0.180 U/mg protein, assayed as described in Materials and Methods). This AS-A-specific activity was comparable to that in developed (E-D15) pig corpora lutea (0.602 ± 0.184 U/mg protein).

Figure 7.

Figure 7

Open in a new tab

HPTLC with orcinol staining-charring, demonstrating the significant decrease in SGG level in regressing pig corpora lutea. Lipids loaded were from developed corpora lutea (10 mg dissected from a E-D15 pig, and 5 mg dissected from a P-D26 pig), and from regressing corpora lutea [30 mg dissected from an R cyclical pig]. The asterisks (*) denote purple-stained glycolipid bands. Besides SGG (arrow), there appeared seven other glycolipids in the E-D15 and P-D26 samples. Note marked decreases in the amounts of SGG and glycolipid bands 1–4 in the R sample. Data shown are representative of duplicate experiments. Chol, Cholesterol.

Discussion

In this report we demonstrated the presence of AS-A and SGG in the ovary. Both AS-A and SGG were localized selectively in developing corpora lutea (Figs. 1, 3, and 4). AS-A was still present in regressing corpora lutea, in which the level of SGG was minimal. Previously, SGG has been known for its highly selective existence in mammalian male germ cells and, thus, also called seminolipid (16). Therefore, the presence of SGG in the corpus luteum as shown herein brings about the new concept that this sulfoglycolipid is also involved in female reproduction. Various approaches were used to reveal SGG presence in developed corpora lutea. Immunofluorescence using antibody that reacts with SGG and SGC was the first approach (Fig. 4). To discern further which of the two sulfoglycolipids was present in the developed corpus luteum, HPTLC with specific staining for glycolipid (orcinol charring) and the sulfate group (azure A reactivity) was used as the second approach; a lipid band with a similar Rf to that of the SGG standard, which was stained purple after orcinol charring and blue with azure A, was present in developed corpora lutea (Figs. 5 and 7). Finally, ESI-MS results validated this conclusion. Although the negative ion ESI spectra of partially purified putative pig corpus luteum SGG were complex, the fragment ion spectra revealed the selective presence of the two unique ions characteristic of authentic SGG (Fig. 6). The additional unassigned relatively intense fragment ions in these spectra reflected significant heterogeneity within the parent m/z 795.5 signal, and highlighted the selectivity available with tandem mass spectrometric analyses.

SGG and SGC have an affinity for a number of ECM proteins, including fibronectin, laminin, thrombospondin, and ZP glycoproteins (17), and they are involved in cell adhesion as in gamete interactions (21,32,42,43,44). In addition, the carbohydrate moieties of SGC molecules on apposing surface membranes of adjacent cells, mediated by a divalent cation, may interact with each other, thus facilitating cell-cell adhesion (45). Therefore, SGG present on the surface of luteal cells may be important for cellular organization/adhesion during development of the corpus luteum by interacting with ECM proteins or with itself on the adjacent cell surface. In fact, fibronectin, laminin, and thrombospondin are present in the ECM of the luteal tissue (10,11,46). SGG on the luteal cell surface of developing corpora lutea may also play roles in the formation of lipid rafts, which are likely platforms of cell adhesion and signaling molecules (47). However, during luteolysis, cell-cell/ECM adhesion and communication are disrupted, and cell-surface remodeling is the first step of structural luteolysis (9,10,48). We hypothesize that during this remodeling process, SGG is endocytosed and transported to the lysosome, where it is desulfated by AS-A to generate GG. GG is less effective in the adhesion process (17,25), and, therefore, the desulfation of SGG by AS-A may contribute to the loss of cell-ECM/cell adhesion and communication, which occurs during luteolysis. GG may also be further hydrolyzed to yield glycerol backbone lipid(s), which might have less propensity to retain the lipid raft microdomains (cell adhesion/signaling platforms). Because the Km value for SGG desulfation by AS-A is relatively high (91 μm) (21), an appreciable quantity of AS-A would be needed for effective SGG desulfation. This may explain why the active AS-A enzyme was initially expressed at the onset of mouse corpus luteum formation and its level continued to increase throughout the corpus luteum development into the luteolysis period (Fig. 2). Nonetheless, during luteal cell development, lysosomal AS-A and cell surface SGG would be compartmentalized from each other. Therefore, the desulfation of SGG would not occur until the signaling event that leads to cell-surface remodeling and subsequent transport of SGG to the lysosome takes place at the beginning of luteolysis. The role of luteal cell-surface AS-A is unclear. Although it is colocalized with cell-surface SGG, this environment is well outside the optimum pH of five for desulfation activity. AS-A on the sperm surface has a ZP adhesion property (21,32), and because it is colocalized with SGG in the lipid raft membranes on the sperm surface, AS-A and SGG likely act synergistically in sperm-ZP binding (19,20,26). It remains to be seen whether AS-A, coexisting with SGG on the luteal cell surface, has an analogous role in ECM interaction.

Although the disappearance of SGG in regressing corpora lutea was well validated (Figs. 4 and 7), the initial product of SGG degradation, GG, was not evident in these corpus luteum lipid samples on the HPTLC plate (Rf of GG in our HPTLC system was 0.659) (Fig. 7). Active lipid hydrolases, such as phospholipase A2 and phospholipase D, have been present in the corpus luteum during natural or PGF2α-induced luteolysis (49,50,51). Therefore, it was possible that β-galactosidase with specificity to remove β-galactose from GG may be present in regressing corpora lutea as well and responsible for the formation of alkylacylglycerol (diradylglycerol), which was not well discerned in our HPTLC system. Our future studies will entail the profiling of new lipid molecules in corpora lutea that are undergoing luteolysis with the expectation of finding SGG/GG metabolites in this luteal tissue.

Other glycolipids (labeled as *bands 1–4 and 6–8 in Fig. 7) in addition to SGG were present in developed corpora lutea from E-D15 and P-D26 pigs. Presently, the identities of these glycolipids are unknown, although the glycolipid bands 6–8 (Fig. 7) might be gangliosides, as previously described (52), due to their typical low mobility in this HPTLC system. These gangliosides appeared both in the cytoplasm and on the luteal cell surface (52), and this might explain why the glycolipid bands 6–8 still remained in the regressing luteal cells (Fig. 7, middle lane), where the cell-surface remodeling had occurred. In contrast, glycolipid bands 1–4, like SGG, diminished to background levels in regressing corpora lutea (Fig. 7). This suggested that these glycolipids might have a similar locale and role to that of SGG; namely, they might be localized to the luteal cell surface, and they might participate in cell-ECM/cell adhesion and communication processes during corpus luteum development. We are currently discerning the identities of these glycolipids by ESI-MS.

Acknowledgments

We thank Dr. Per-georg Nyholm, Biognos AB, Göteborg, Sweden, for stimulating discussions of the work, as well as Mr. Hongbin Xu for coordinating the pseudopregnancy experiments, which were performed in Ottawa, and Mr. Sitthichai Iamsaard for assistance in figure preparation.

Footnotes

This work was supported by Thailand Research Fund Grant RGJ-PHD/0119/2542 and BGJ/36/2544 (to P.Sa. and A.A.), a Canadian Institutes for Health Research Grant 62945 (to N.T.), a University of Ottawa, International Creative Research grant (to N.T.), W.M. Keck Foundation (to K.F.F.), Multistate Research Funded Grant W1171 (to T.B.), Royal Golden Jubilee scholarships (to A.A. and R.V.), and the Development and Promotion of Science and Technology Talented Project Thailand scholarship (to K.K.).

Disclosure Statement: The authors have nothing to declare.

First Published Online April 17, 2008

Abbreviations: AS-A, Arylsulfatase A; CB, Coomassie blue; ECM, extracellular matrix; E-D5, cyclical d-5; E-D15, cyclical d-15; ESI, electrospray; GC, galactosylceramide; GG, galactosylglycerolipid; hCG, human chorionic gonadotropin; HPTLC, high-performance thin-layer chromatography; MS, mass spectrometry; NCS, p-nitrocatechol sulfate; PC, phosphatidylcholine; P-D26, pregnant d-26; PE, phosphatidylethanolamine; PGF2α, prostaglandin F2α; PMSG, pregnant mare serum gonadotropin; PRS, preimmune serum; R, prostaglandin F2α-treated; SGC, sulfogalactosylceramide; SGG, sulfogalactosylglycerolipid; TBS, Tris-buffered saline; ZP, zona pellucida.

References

  1. Stouffer RL 2006 Structure, function, and regulation of the corpus luteum. In: Neill JD, ed. Physiology of reproduction. New York: Elsevier; 475–526 [Google Scholar]
  2. Stocco C, Telleria C, Gibori G 2007 The molecular control of corpus luteum formation, function, and regression. Endocr Rev 28:117–149 [DOI] [PubMed] [Google Scholar]
  3. MacCracken KA, Scalisi JC 1999 Development and evaluation of a ketogenic diet program. J Am Diet Assoc 99:1554–1558 [DOI] [PubMed] [Google Scholar]
  4. Bachelot A, Binart N 2005 Corpus luteum development: lessons from genetic models in mice. Curr Top Dev Biol 68:49–84 [DOI] [PubMed] [Google Scholar]
  5. Wang Z, Tamura K, Yoshie M, Tamura H, Imakawa K, Kogo H 2003 Prostaglandin F2α-induced functional regression of the corpus luteum and apoptosis in rodents. J Pharmacol Sci 92:19–27 [DOI] [PubMed] [Google Scholar]
  6. Niswender GD, Juengel JL, Silva PJ, Rollyson MK, McIntush EW 2000 Mechanisms controlling the function and life span of the corpus luteum. Physiol Rev 80:1–29 [DOI] [PubMed] [Google Scholar]
  7. Hussein MR 2005 Apoptosis in the ovary: molecular mechanisms. Hum Reprod Update 11:162–177 [DOI] [PubMed] [Google Scholar]
  8. Guo K, Wolf V, Dharmarajan AM, Feng Z, Bielke W, Saurer S, Friis R 1998 Apoptosis-associated gene expression in the corpus luteum of the rat. Biol Reprod 58:739–746 [DOI] [PubMed] [Google Scholar]
  9. Curry Jr TE, Osteen KG 2003 The matrix metalloproteinase system: changes, regulation, and impact throughout the ovarian and uterine reproductive cycle. Endocr Rev 24:428–465 [DOI] [PubMed] [Google Scholar]
  10. Yasuda K, Hagiwara E, Takeuchi A, Mukai C, Matsui C, Sakai A, Tamotsu S 2005 Changes in the distribution of tenascin and fibronectin in the mouse ovary during folliculogenesis, atresia, corpus luteum formation and luteolysis. Zoolog Sci 22:237–245 [DOI] [PubMed] [Google Scholar]
  11. Irving-Rodgers HF, Friden BE, Morris SE, Mason HD, Brannstrom M, Sekiguchi K, Sanzen N, Sorokin LM, Sado Y, Ninomiya Y, Rodgers RJ 2006 Extracellular matrix of the human cyclic corpus luteum. Mol Hum Reprod 12:525–534 [DOI] [PubMed] [Google Scholar]
  12. Paavola LG 1978 The corpus luteum of the guinea pig. II. Cytochemical studies on the Golgi complex, GERL, and lysosomes in luteal cells during maximal progesterone secretion. J Cell Biol 79:45–58 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Paavola LG, Boyd CO 1981 Cytoplasmic granules in luteal cells of pregnant and non-pregnant guinea pigs. A cytochemical study. Anat Rec 201:127–140 [DOI] [PubMed] [Google Scholar]
  14. Al Mehdi MI 1978 Electron cytochemical demonstration of four hydrolytic enzymes in the rat corpora lutea at second half of gestation. Histochemistry 56:317–327 [DOI] [PubMed] [Google Scholar]
  15. Gregoraszczuk EL, Sadowska J 1997 Lysosomal acid phosphatase activity and progesterone secretion by porcine corpora lutea at various periods of the luteal phase. Folia Histochem Cytobiol 35:35–39 [PubMed] [Google Scholar]
  16. Hanson SR, Best MD, Wong CH 2004 Sulfatases: structure, mechanism, biological activity, inhibition, and synthetic utility. Angew Chem Int Ed Engl 43:5736–5763 [DOI] [PubMed] [Google Scholar]
  17. Tanphaichitr N, Bou Khalil M, Weerachatyanukul W, Kates M, Xu H, Carmona E, Attar M, Carrier D 2003 Physiological and biophysical properties of male germ cell sulfogalactosylglycerolipid. In: De Vriese S, ed. Lipid metabolism and male fertility. Champaign, IL: AOCS Press; 125–148 [Google Scholar]
  18. Murray RK, Narasimhan R 1990 Glycoglycerolipids of animal tissues. In: Kates M, ed. Glycolipids, phosphoglycolipids, and sulfoglycolipids. New York: Plenum Press; 321–361 [Google Scholar]
  19. Tanphaichitr N, Carmona E, Bou Khalil M, Xu H, Berger T, Gerton GL 2007 New insights into sperm-zona pellucida interaction: involvement of sperm lipid rafts. Front Biosci 12:1748–1766 [DOI] [PubMed] [Google Scholar]
  20. Tanphaichitr N, Faull KF, Yaghoubian A, Xu H 2007 Roles of sperm lipid rafts and sulfogalactosyglycerolipid (SGG) in sperm functions: consensus and controversy1. Trends Glycosci Glycotechnol 19:67–83 [Google Scholar]
  21. Carmona E, Weerachatyanukul W, Soboloff T, Fluhary AL, White D, Promdee L, Ekker M, Berger T, Buhr M, Tanphaichitr N 2002 Arylsulfatase A is present on the pig sperm surface and is involved in sperm-zona pellucida binding. Dev Biol 247:182–196 [DOI] [PubMed] [Google Scholar]
  22. Waheed A, Van Etten RL 1980 Chemical characterization and substrate specificity of rabbit liver aryl sulfatase A. Biochim Biophys Acta 614:92–101 [DOI] [PubMed] [Google Scholar]
  23. Fluharty AL, Stevens RL, Miller RT, Kihara H 1974 Sulfoglycerogalactolipid from rat testis: a substrate for pure human arylsulfatase A. Biochem Biophys Res Commun 61:348–354 [DOI] [PubMed] [Google Scholar]
  24. Gieselmann V 2003 Metachromatic leukodystrophy: recent research developments. J Child Neurol 18:591–594 [DOI] [PubMed] [Google Scholar]
  25. White D, Weerachatyanukul W, Gadella B, Kamolvarin N, Attar M, Tanphaichitr N 2000 Role of sperm sulfogalactosylglycerolipid in mouse sperm-zona pellucida binding. Biol Reprod 63:147–155 [DOI] [PubMed] [Google Scholar]
  26. Bou Khalil M, Chakrabandhu K, Xu H, Weerachatyanukul W, Buhr M, Berger T, Carmona E, Vuong N, Kumarathasan P, Wong PT, Carrier D, Tanphaichitr N 2006 Sperm capacitation induces an increase in lipid rafts having zona pellucida binding ability and containing sulfogalactosylglycerolipid. Dev Biol 290:220–235 [DOI] [PubMed] [Google Scholar]
  27. Tanphaichitr N, Smith J, Kates M 1990 Levels of sulfogalactosylglycerolipid in capacitated motile and immotile mouse sperm. Biochem Cell Biol 68:528–535 [DOI] [PubMed] [Google Scholar]
  28. Kamei K, Kubushiro K, Fujii T, Tsukazaki K, Nozawa S, Iwamori M 1997 Menstrual cycle-associated regulation of anabolic and catabolic enzymes causes luteal phase-characteristic expression of sulfatide in human endometrium. Am J Obstet Gynecol 176:142–149 [DOI] [PubMed] [Google Scholar]
  29. Vitaioli L, Gobbetti A, Baldoni E 1996 Arylsulphatase A activity and sulphatide concentration in the female rabbit oviduct are under physiological hormonal influence. Histochem J 28:149–156 [DOI] [PubMed] [Google Scholar]
  30. Wu A, Anupriwan A, Iamsaard S, Chakrabandhu K, Santos DC, Rupar T, Tsang BK, Carmona E, Tanphaichitr N 2007 Sperm surface arylsulfatase A can disperse the cumulus matrix of cumulus oocyte complexes. J Cell Physiol 213:201–211 [DOI] [PubMed] [Google Scholar]
  31. Sarafian TA, Fluharty AL, Kihara H, Helfand G, Edmond J 1982 Large-scale purification of pyrogen-free human arylsulfatase A. J Appl Biochem 4:126–132 [Google Scholar]
  32. Tantibhedhyangkul J, Weerachatyanukul W, Carmona E, Xu H, Anupriwan A, Michaud D, Tanphaichitr N 2002 Role of sperm surface arylsulfatase A in mouse sperm-zona pellucida binding. Biol Reprod 67:212–219 [DOI] [PubMed] [Google Scholar]
  33. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 [DOI] [PubMed] [Google Scholar]
  34. Towbin H, Gordon J 1984 Immunoblotting and dot immunobinding-current status and outlook. J Immunol Methods 72:313–340 [DOI] [PubMed] [Google Scholar]
  35. Bligh EG, Dyer WJ 1959 A rapid method of total lipid extraction and purification. Can J Biochem Physiol 31:911–917 [DOI] [PubMed] [Google Scholar]
  36. Kates M 1986 Technique of lipidology: isolation, analysis and identification of lipids. In: Burdon RH, ed. Laboratory techniques in biochemistry and molecular biology. New York: Elsevier; 100–278 [Google Scholar]
  37. Furimsky A, Vuong N, Xu H, Kumarathasan P, Xu M, Weerachatyanukul W, Bou KM, Kates M, Tanphaichitr N 2005 Percoll gradient-centrifuged capacitated mouse sperm have increased fertilizing ability and higher contents of sulfogalactosylglycerolipid and docosahexaenoic acid-containing phosphatidylcholine compared to washed capacitated mouse sperm. Biol Reprod 72:574–583 [DOI] [PubMed] [Google Scholar]
  38. Weerachatyanukul W, Rattanachaiyanont M, Carmona E, Furimsky A, Mai A, Shoushtarian A, Sirichotiyakul S, Ballakier H, Leader A, Tanphaichitr N 2001 Sulfogalactosylglycerolipid is involved in human gamete interaction. Mol Reprod Dev 60:569–578 [DOI] [PubMed] [Google Scholar]
  39. Tupper S, Wong PTT, Kates M, Tanphaichitr N 1994 Interaction of divalent cations with germ cell specific sulfogalactosylglycerolipid and the effects on lipid chain dynamics. Biochemistry 33:13250–13258 [DOI] [PubMed] [Google Scholar]
  40. Gadella BM, Gadella TWJ, Colenbrander B, Van Golde LM, Lopes-Cardozo M 1994 Visualization and quantification of glycolipid polarity dynamics in the plasma membrane of the mammalian spermatozoon. J Cell Sci 107:2151–2163 [DOI] [PubMed] [Google Scholar]
  41. Liu K, Wahlberg P, Hagglund AC, Ny T 2003 Expression pattern and functional studies of matrix degrading proteases and their inhibitors in the mouse corpus luteum. Mol Cell Endocrinol 205:131–140 [DOI] [PubMed] [Google Scholar]
  42. Vartanian T, Szuchet S, Dawson G 1992 Oligodendrocyte-substratum adhesion activates the synthesis of specific lipid species involved in cell signaling. J Neurosci Res 32:69–78 [DOI] [PubMed] [Google Scholar]
  43. Maehashi E, Sato C, Ohta K, Harada Y, Matsuda T, Hirohashi N, Lennarz WJ, Kitajima K 2003 Identification of the sea urchin 350-kDa sperm-binding protein as a new sialic acid-binding lectin that belongs to the heat shock protein 110 family: implication of its binding to gangliosides in sperm lipid rafts in fertilization. J Biol Chem 278:42050–42057 [DOI] [PubMed] [Google Scholar]
  44. Lingwood C, Mylvaganam M, Minhas F, Binnington B, Branch DR, Pomes R 2005 The sulfogalactose moiety of sulfoglycosphingolipids serves as a mimic of tyrosine phosphate in many recognition processes. Prediction and demonstration of Src homology 2 domain/sulfogalactose binding. J Biol Chem 280:12542–12547 [DOI] [PubMed] [Google Scholar]
  45. Boggs JM, Gao W, Hirahara Y 2008 Myelin glycosphingolipids, galactosylceramide and sulfatide, participate in carbohydrate-carbohydrate interactions between apposed membranes and may form glycosynapses between oligodendrocyte and/or myelin membranes. Biochim Biophys Acta 1780:445–455 [DOI] [PubMed] [Google Scholar]
  46. Petrik JJ, Gentry PA, Feige JJ, LaMarre J 2002 Expression and localization of thrombospondin-1 and -2 and their cell-surface receptor, CD36, during rat follicular development and formation of the corpus luteum. Biol Reprod 67:1522–1531 [DOI] [PubMed] [Google Scholar]
  47. Rajendran L, Simons K 2005 Lipid rafts and membrane dynamics. J Cell Sci 118:1099–1102 [DOI] [PubMed] [Google Scholar]
  48. Carlson JC, Buhr MM, Riley JC 1989 Plasma membrane changes during corpus luteum regression. Can J Physiol Pharmacol 67:957–961 [DOI] [PubMed] [Google Scholar]
  49. Sawada M, Carlson JC 1991 Rapid plasma membrane changes in superoxide radical formation, fluidity, and phospholipase A2 activity in the corpus luteum of the rat during induction of luteolysis. Endocrinology 128:2992–2998 [DOI] [PubMed] [Google Scholar]
  50. Kurusu S, Kaizo K, Ibashi M, Kawaminami M, Hashimoto I 1999 Luteal phospholipase A2 activity increases during functional and structural luteolysis in pregnant rats. FEBS Lett 454:225–228 [DOI] [PubMed] [Google Scholar]
  51. Yamamoto H, Endo T, Kiya T, Goto T, Sagae S, Ito E, Watanabe H, Kudo R 1995 Activation of phospholipase D by prostaglandin F2α in rat luteal cells and effects of inhibitors of arachidonic acid metabolism. Prostaglandins 50:201–211 [DOI] [PubMed] [Google Scholar]
  52. Choo YK, Chiba K, Tai T, Ogiso M, Hoshi M 1995 Differential distribution of gangliosides in adult rat ovary during the oestrous cycle. Glycobiology 5:299–309 [DOI] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES