Molecularly cloned mammalian glucosamine-6-phosphate deaminase localizes to transporting epithelium and lacks oscillin activity

^a Department of Neuroscience, The Johns Hopkins University, School of Medicine, Baltimore, Maryland 21205, USA
^b Department of Pharmacology and Molecular Sciences, The Johns Hopkins University, School of Medicine, Baltimore, Maryland 21205, USA
^c Department of Psychiatry and Behavioral Sciences, The Johns Hopkins University, School of Medicine, Baltimore, Maryland 21205, USA
^d Department of Biological Sciences, Kent State University, Kent, Ohio 41242, USA

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Glucosamine-6-phosphate deaminase (GNPDA) catalyzes the conversion of glucosamine-6-phosphate to fructose-6-phosphate, a reaction that under physiological conditions proceeds to the formation of fructose-6-phosphate. Though first identified in mammalian tissues in 1956, the enzyme has not previously been molecularly characterized in mammalian tissues, although a bacterial GNPDA has been cloned. Recently, a protein displaying similarity to bacterial GNPDA was purified and cloned from sperm extract. It was proposed that this protein was the factor, found in sperm extracts, that causes calcium oscillations in cells; thus, the protein was named 'oscillin.' We demonstrate that oscillin is the mammalian form of glucosamine 6-phosphate deaminase by showing that cloned oscillin has a robust GNPDA activity and can account for all such activity in mammalian tissues extracts. In situ hybridization and immunohistochemistry localize GNPDA selectively to tissues with high energy requirements such as the apical zone of transporting epithelia in the proximal convoluted tubules of the kidney and the small intestine; to neurons (but not glia) and especially to nerve terminals in the brain; and to motile sperm. Recombinant GNPDA and GNPDA purified to homogeneity from hamster sperm fail to elevate intracellular calcium when injected into mouse eggs over a wide range of concentrations under conditions in which sperm extracts elicit pronounced calcium oscillations. Thus, the calcium-releasing or oscillin activity of sperm extracts is due to a substance other than GNPDA. Since GNPDA is the sole enzyme linking hexosamine systems with glycolytic pathways, we propose that it provides a source of energy in the form of phosphosugar derived from the catabolism of hexosamines found in glycoproteins, glycolipids, and sialic acid-containing macromolecules. Evidence that GNPDA can regulate hexosamine stores comes from our observation that transfection of GNPDA into HEK-293 cells reduces cellular levels of sialic acid. —Wolosker, H., Kline, D., Bian, Y., Blackshaw, S., Cameron, A. M., Fralich, T. J., Schnaar, R. L., Snyder, S. H. Molecularly cloned mammalian glucosamine-6-phosphate deaminase localizes to transporting epithelium and lacks oscillin activity. FASEB J. 12, 91–99 (1998)

Key Words: metabolism • hexosamines • cloning • sialic acid • GNPDA

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
HEXOSAMINES ARE BUILDING BLOCKS of structural macromolecules such as glycoproteins, glycolipids, and sulfated polysaccharides. Hexosamines can also interface with the glycolytic pathway through the conversion of glucosamine-6-phosphate to fructose-6-phosphate, potentially influencing energy metabolism. A glucosamine 6-phosphate deaminase (GNPDA)² activity that catalyzes the formation of fructose 6-phosphate and ammonia from glucosamine 6-phosphate in mammalian cells was described by Comb and Roseman in 1958 (1), but the protein responsible for this activity was never completely purified and molecularly cloned; however, a bacterial GNPDA has been cloned (2). In an independent line of investigation, it has been reported that a factor in sperm extracts stimulates calcium release in mammalian eggs (3–7) and may be responsible for the initiating activation of the egg at fertilization, a known calcium-dependent event. Recently, Parrington et al. (7) isolated and cloned a protein from hamster sperm that appeared to induce calcium release and displayed about a 50% amino acid sequence identity to bacterial GNPDA. They suggested that this protein was the sperm factor responsible for inducing calcium release in eggs, and named it `oscillin.'

We have cloned and expressed hamster sperm oscillin protein and show that it possesses GNPDA activity. We demonstrate that GNPDA mRNA and protein are present in numerous tissues. GNPDA is especially abundant in transporting epithelium of the kidney and small intestine. We also show that recombinant GNPDA and GNPDA (oscillin) purified to apparent homogeneity from sperm extracts do not possess calcium-releasing activity in egg cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cloning
GNPDA was cloned by reverse transcription-polymerase chain reaction (RT-PCR) from hamster testis total RNA, using primers based on the published sequence of oscillin (7) containing sequences for Sal I/Not I restriction sites as follows: upper primer, 5' ACG CTG CGA CGA TGA TTA TCC TGG AAC ACT ACT CTC AGG 3'; lower primer, 5' ATA AGA ATG CGG CCG CGT CGC TGT ATG GCT TCT TAG CAG CC 3'. Human brain GNPDA was obtained by PCR from a Stratagene infant brain library using primers based on the sequence for the KIAA0060 gene (GenBank accession number D31766), which resembles GNPDA, containing Bgl II/Not I restriction sites as follows: upper primer, 5' GAA GAT CTT GAT GAA GCT CAT CAT CCT GGA GCA CTA TTC TCA GGC G 3'; lower primer, 5' ATA AGA ATG CGG CCG CCT AAT CGC TGT ATG GTT TCT TCG AAG ATT GGC 3'.

Northern blots
Total RNA from hamster or rats was extracted using Triazol reagent (Gibco-BRL, Gaithersburg, Md.). Agarose denaturing gels were run and RNA was transferred to nylon membranes (Schleicher & Schuell, Keene, N.H.) as described (8).

Expression of fusion protein in Escherichia coli
Hamster sperm GNPDA was subcloned into pGEX 4T-2 vector (Pharmacia Biotech, Pitscataway, N.J.) and expressed in BL21 bacteria (Novagen, Madison, Wis.). Expression and purification of glutathione S-transferase (GST) fusion protein was performed as described (9), using glutathione agarose beads (Sigma, St. Louis). Elution of GST fusion protein from the beads was achieved by a 10 min incubation at room temperature in the presence of 20 mM Tris-HCl (pH 8.0) and 15 mM reduced glutathione. The material was dialyzed for 12 h against either phosphate-buffered saline (PBS) or 20 mM Hepes-Tris, pH 7.4, plus 75 mM KCl, with frequent changes of buffer. In some experiments, the GST part of the fusion protein was removed by cleavage with biotinylated thrombin (Novagen). Thrombin was removed by the addition of streptavidin-agarose according to the manufacturer's instructions. The GST and uncleaved GST fusion protein were further removed by the addition of glutathione agarose. All preparations were checked for purity in Coomassie blue-stained sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

Cell culture and expression of GNPDA in 293 cells
HEK 293 cells were cultured in Dulbecco's modified Eagle's medium plus penicillin-streptomycin, 10% fetal bovine serum, and 10 mM glutamine at 37°C in 5% CO₂. GNPDA was subcloned into pRK5 vector containing an epitope tag for hemagluttinin (HA tag). The cells were transfected using the calcium-phosphate method (8) in the presence of 1 to 10 µg DNA/100 mm plate for 3–6 h. The medium was removed, cells were washed three times with PBS, and replaced with new medium. After 24 h, cells were scraped off the plates and processed for Western blots or sialic acid determination.

Antibody production
Polyclonal antibodies against the GST fusion protein were raised in rabbits. The serum was adsorbed to a GST column, obtained by coupling GST to CnBr sepharose as described (10). The antibodies against the GST part of the fusion protein remained bound to the column and the serum was free of anti-GST antibodies. For immunohistochemistry, the antibody was purified by passing the serum through an antigen affinity column as described (10).

Western blots
Tissue from male Sprague-Dawley rats (8–9 wk old) was removed and immediately homogenized with a Polytron in a medium containing 20 mM Tris-HCl pH 7.4, 100 mM KCl, and protease inhibitor cocktail (Boehringer Mannheim, Mannheim, Germany). The homogenate was centrifuged for 10 min at 15,000 g and the supernatants were run in 12% SDS-PAGE and transferred to polyvinylidene difluoride membranes. After blocking the membranes with 5% milk in 0.1% Tween 20, the blot was incubated with a 1:30,000 dilution of anti-serum for 50 min at room temperature. Detection was performed using enhanced chemiluminescence assay (Amershan, Arlington Heights, Ill.).

In situ hybridization
In situ hybridization was performed using full-length hamster GNPDA antisense RNA, essentially as described (11). Sense RNA was used as a control probe.

Immunohistochemistry
Male Sprague-Dawley rats (8–9 wk old) were perfused with 4% paraformaldeyde in 0.1 M phosphate buffer, pH 7.4. The organs were removed, postfixed for 2 h at 4°C, and cryoprotected in phosphate buffer plus 25% sucrose. Saggital brain sections (20 µm) were cut with a freezing microtome and other tissues were processed for paraffin embedding. Staining of either free-floating or sections affixed to slides was performed using an ABC kit (Vector Laboratories, Burlingame, Calif.) and developed with DAB (Gibco-BRL). Controls were treated with either preimmune serum or antibody preabsorbed with the antigen.

Enzymatic activity
GNPDA activity was determined by measuring fructose 6-phosphate formed from glucosamine 6-phosphate as described (12). Glucosamine-6-phosphate solution was kept frozen at acid pH. The solution was neutralized just before use. N-acetylglucosamine 6-phosphate was synthesized according to Leloir and Cardini (13).

Sialic acid
Sialic acid was determined by measuring the amount of N-acetylneuraminic acid (NeuAc). Mock and GNPDA-transfected HEK 293 cells were lysed by sonication in a medium containing 20 mM Tris-HCl, pH 7.4, 100 mM KCl, and protease inhibitor cocktail. Total membrane fraction was collected by centrifugation at 100,000 g for 1 h. The membranes were washed by centrifugation three times with cold water. After washing, the pellet was suspended in 0.1 M HCl and 0.25 M NaCl, and sialic acid was released by a 3 h incubation at 80°C. Samples were analyzed using a Dionex HPLC system with pulsed amperometric detection (14) using a standard of NeuAc. The areas of the peaks were calculated, and sialic acid was expressed as nmol NeuAc/mg protein.

Sperm extracts and purification of sperm GNPDA
Hamster spermatozoa were obtained from cauda epidydimis. Cells were lysed by sonication in medium containing 20 mM Tris-HCl, pH 7.4, 75 mM KCl, and protease inhibitor cocktail. The suspension was centrifuged at 100,000 g and the supernatant was loaded onto an Econo-Pac column (Bio-Rad, Richmond, Calif.). The column was washed with 30 ml lysis buffer and the proteins were eluted with 20 mM Tris-HCl, pH 7.4, plus 1 M KCl. The eluate was concentrated in Centriplus 3000 (Amicon, Lexington, Mass.) and diluted in 20 mM Hepes-Tris (pH 7.4) to decrease the salt concentration to 75 mM. The suspension was concentrated a second time to 42 mg/ml. The sperm extract was frozen at -80°C without loss of activity. Purification of sperm GNPDA protein was attained by loading sperm extract onto a N-aminohexanoyl glucosamine 6-phosphate agarose column (15). The column was washed three times with PBS, and GNPDA was eluted by the addition of medium containing 20 mM Tris-HCl, pH 7.4, 75 mM KCl, and 10 mM N-acetylglucosamine 6-phosphate. The eluate was concentrated and diluted twice with buffer without N-acetylglucosamine 6-phosphate.

Microinjection and measurement of intracellular calcium in mouse eggs
Mature mouse eggs were obtained from female mice after superovulation induced by hormone injection (16, 17), and cumulus cells were removed with 0.3 mg/ml hyaluronidase. A direct quantitative pressure system was used for microinjection as previuosly described (17). Eggs were first injected with 1.8 pl of 2 mM fura 2-dextran to give a final concentration of 17 µM in the egg. Fura 2 fluorescence was monitored photometrically; an increase in intracellular calcium is indicated by the increase in the fura 2 emission ratio for the 350/385 nm excitation wavelengths after background subtraction (17, 18). A second injection introduced recombinant GNPDA, the purified GNPDA, extracts of sperm, or extracts of HEK 293 transfected with GNPDA. Typically, the injection volumes were 1–10% of the egg volume. Intracellular concentrations were calculated from the concentration and volume of the solution injected, assuming uniform distribution and an egg volume of 205 pl. After recording the basal Ca²⁺ level, the recording was interrupted and the injection pipet was inserted into the egg. In most cases, the injection was observed and then the recording resumed. During the injections, eggs were held in IVF media containing 0.4% BSA (17) at 37°C in a humidified environment with 5% CO₂.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Glucosamine-6-phosphate deaminase is a highly conserved protein
In a search of the GenBank database, we discovered an unidentified human sequence closely resembling that of E. coli GNPDA. Using PCR, we have cloned and fully sequenced human GNPDA and also molecularly cloned hamster sperm oscillin ( Fig. 1). Amino acid sequences of the three proteins are highly similar. The major difference between the E. coli and mammalian forms is that the E. coli protein is 23 amino acids shorter than the hamster and human proteins, which are both 289 amino acids. For the 266 amino acids shared by all three enzymes, the hamster and human amino acid sequences show 98.5% identity and the nucleotide sequences display 90% identity. In the shared sequence, both human and hamster proteins display 77% identity in amino acid sequence and 60% identity of nucleotide sequence to the bacterial protein. In some cases the human sequence more closely resembles the bacterial sequence than does the hamster, although the reverse is true in other cases. This conservation from human to bacteria is one of the highest in biology, comparable to cytochrome c (19), cyclophilins (20), and PIN, a protein inhibitor of nitric oxide (21).

View larger version (44K): [in this window] [in a new window]   Figure 1. Alignment of E. coli GNPDA amino acid sequence with oscillin and human homologue. Boxed areas correspond to homologous regions.

GNPDA localizes to transporting epithelium in kidney and small intestine and to neurons in the brain
Parrington et al. (7) reported that oscillin occurs exclusively in sperm and testes, as their Western blot analysis failed to detect protein in the liver and brain. Our Northern blot analysis, using as a probe cDNA from hamster GNPDA, reveals a ubiquitous distribution ( Fig. 2A). Highest densities occur in the kidney and small intestine, with somewhat lower levels in the spleen, testes, ovary, brain, lung, and heart. Extremely low levels are observed in skeletal muscle and liver. We conducted Western blot analysis using a polyclonal antibody raised against a fusion protein of recombinant hamster GNPDA ( Fig. 2B). GNPDA protein displays a similar distribution to mRNA with high levels in the kidney and small intestine, comparable to those in testes and spleen, but lower levels in the brain, lung, and heart. Only negligible levels are detected in liver and skeletal muscle. HEK293 cells display endogenous GNPDA, with a major augmentation associated with transfection. In all tissues, GNPDA is associated with a discrete band of about 35 kDa. In the heart, an additional band of about 50 kDa is detected, which is abolished by preabsorption with antigen.

View larger version (46K): [in this window] [in a new window]   Figure 2. Tissue distribution of GNPDA. A) Northern blot of GNPDA in rat tissues. Each lane contains 20 µg total RNA probed with hamster GNPDA cDNA. The blot was stripped once and reprobed to glyceraldehyde 3-phosphate dehydrogenase to verify the integrity of the RNA message. B) Western blot analysis containing 60 µg protein/lane, except for the last lane, which contains 200 ng of HEK 293 cell extract transfected with hamster GNDPA containing an epitope tag of hemagluttinin (HA tag). The epitope-tagged GNPDA runs at a higher molecular weight due to the additional nine amino acids of the tag that are fused with the amino terminus of the protein.

In situ hybridization in whole 4-day-old mice reveals highest densities of GNPDA mRNA in the kidney and small intestine, with substantial levels also evident in testes, submandibular salivary gland, and brain ( Fig. 3). In all of these organs, the signal is greatly reduced in control preparations using sense probes. A prominent signal in the adrenal is essentially the same in sense and antisense-treated preparations and so is presumably nonspecific. Immunohistochemical examination at high resolution reveals discrete localizations within the small intestine and kidney ( Fig. 4). In the kidney, staining is most prominent in the apical portion of epithelium in the proximal convoluted tubules. By contrast, the basal portion of these epithelial cells displays negligible staining; nor do we detect staining in associated connective tissue, distal tubules, or the kidney medulla. The great majority of cells in glomeruli also stains, but at much lower intensity.

View larger version (82K): [in this window] [in a new window]   Figure 3. In situ hybridization of GNPDA. A 4-day-old mouse was probed with GNPDA antisense (A) or sense (B) RNA probe. Adr, adrenal; c, cerebellum; ctx, cerebral cortex; h, hippocampus; sal. gl, submandibular salivary gland.

View larger version (95K): [in this window] [in a new window]   Figure 4. Localization of GNPDA to transporting epithelium. Paraffin sections were incubated with 0.5 µg of affinity purified antibody as described in Materials and Methods. In the kidney (A), high levels of GNPDA immunoreactivity were localized to the proximal convoluted tubules (PCT) at the apex of the proximal convoluted cells (asterisk), in contrast to the low levels of staining at the basal portion of the tubular cells (arrow). Moderate levels of staining were observed in the glomerulus (Gl). In small intestine (C), high levels of GNPDA were localized to the intestinal epithelia (Ep). Preabsorption of antibodies with an excess antigen (B, D) abolished immunostaining.

In the small intestine, staining is also selectively localized to epithelial cells, being most prominent in the apical portion ( Fig. 4C). Much lower levels of staining occur in the lamina propria and underlying muscle.

In the brain, immunohistochemical staining is prominent in neuropil, reflecting selective neuronal localizations. No staining is evident in white matter areas such as the corpus callosum, white matter of areas of the cerebellum, and the brain stem ( Fig. 5). Staining is most prominent in terminal fields. Large neuronal cells, such as the pyramidal cells of the hippocampus and the Purkinje cells of the cerebellum, fail to stain. In the cerebellum, GNPDA protein staining is most prominent in the molecular layer, whereas in situ hybridization reveals mRNA most prominent in the granule cell layer ( Fig. 6A). This implies that the protein staining in the molecular layer represents the parallel fibers of the granule cells. In the hippocampus, in situ hybridization shows prominent staining in the granule cells of the dentate gyrus and the pyramidal cells in CA-1, CA-2, CA-3, and CA-4 ( Fig. 6B). Since no GNPDA protein staining is evident in the hippocampal pyramidal cells ( Fig. 5), it is likely that most of the enzyme protein is distributed in the apical dendrites. This conclusion is supported by high-power immunohistochemistry, which reveals substantial staining in the apical dendrites of pyramidal cells ( Fig. 6C). In situ hybridization also reveals high mRNA levels selectively concentrated in cells of the anterior olfactory nucleus and in glomerular and mitral cells of the olfactory bulb (Figs. 6D, E).

View larger version (51K): [in this window] [in a new window]   Figure 5. Immunohistochemical localization of GNPDA in rat brain sections. Saggital sections were incubated with 0.7 µg affinity-purified antibody against GNPDA (A), as described in Materials and Methods. Preabsorption of antibodies with excess antigen (B) abolished immunostaining. Aon, anterior olfactory nuclei; c, corpus callosum; Ctx, cerebral cortex; H, hypothalamus; Hp, hippocampus; Ic, inferior colliculi, Lv, lateral ventricle; Mo, medulla oblongata, Ob, olfactory bulb; Sc, superior colliculi, St, striatum. Dark areas represent positive staining.

View larger version (78K): [in this window] [in a new window]   Figure 6. Localization of oscillin to distinct neuronal structures. In situ hybridization (A, B, D, E) and immunohistochemical staining (C) of saggital sections. In the cerebellum (A), the GNPDA message was localized to the granule cell layer (Gr). In the hippocampus (B, C), the GNPDA message was detected in CA1-CA4 of Ammon's horn (CA1-CA4) and in granule cells of the dentate gyrus (DG). Immunoreactivity was observed in the apical dendrites (arrows) of pyramidal cells (Py) in the stratum radiatum (Rad) of hippocampus (C). GNPDA message was high in glomerular (GL), mitral cell (MCL), and granule cell layers (GCL) of the olfactory bulb (D) as well as in the anterior olfactory nuclei (E). Or, stratum oriens; Mol, molecular cell layer. x10–20 (A, B, D, E); x65 (C).

Demonstration of GNPDA catalytic activity in recombinant oscillin/GNPDA
Parrington et al. (7) noted an overall 53% amino acid sequence identity between hamster oscillin and bacterial GNPDA, but did not examine catalytic activity of the recombinant protein. We monitored this activity in recombinant hamster oscillin using glucosamine-6-phosphate as substrate and measuring the evolution of fructose-6-phosphate by a colorimetric reaction ( Fig. 7). Catalytic activity is linear with time for the GST fusion GNPDA, whereas the GST protein itself lacks activity. N-Acetylglucosamine-6-phosphate has been described as an allosteric effector of GNPDA (1, 15). We observe augmentation of catalytic activity with added N-acetylglucosamine-6-phosphate. Thus, the catalytic properties of mammalian GNPDA appear to be the same as those of the purified and cloned bacterial enzyme.

View larger version (16K): [in this window] [in a new window]   Figure 7. Recombinant protein displays GNPDA activity. GNPDA activity was assayed at 37°C in a medium containing 20 mM Tris-HCl, pH 7.4, 10 mM glucosamine 6-phosphate, and 5 µg/ml of GNPDA-GST (, ) or GST alone (, ), either in the absence (, ) or presence (, ) of 1 mM N-acetylglucosamine 6-phosphate. The reaction was stopped by the addition of concentrated HCl. The experiment was replicated three times using three different enzyme preparations with similar results.

Transfected GNPDA alters sialic acid levels
The hexosamine pool of energy might be quite large. Thus, Eriksson et al. (22) noted that although ATP is the dominant nucleotide (936 pmol/µg DNA), UDP-N-acetylglucosamine pool levels are more than half those of ATP (596 pmol/µg DNA). Hexosamine pools turn over rapidly, especially during lysosomal degradation of polysaccharides (23). To examine the extent to which GNPDA can influence this pool, we transfected GNPDA into HEK293 cells and monitored sialic acid content of the membrane fraction of the cells. The sialic acid content of control mock transfected HEK 293 cells was 18.6 ± 1.4 nmol NeuAc/mg protein (mean±SE, n=6). HEK 293 cells transfected with GNPDA-HA contained 14.6 (1.5 nmol NeuAc/mg protein (mean±SE, n=6), which was significantly lower than the control (P=0.041) and represents a 22% decrease on sialic acid levels. Since the transfection efficiency is only about 30–40%, GNPDA has the potential of substantially influencing the hexosamine pool of cells.

Mammalian GNPDA lacks oscillin activity
Parrington et al. (7) purified oscillin from hamster sperm extracts, monitoring calcium release in sequentially purified preparations. However, they did not examine the influence of recombinant oscillin upon calcium dynamics in eggs. The purity of isolated hamster oscillin was examined by Parrington et al. (7) with SDS gels that displayed a major band at 35 kDa with several additional bands of lower molecular weight. The staining procedure used a Coomasie brilliant blue R-250 dye, which is less sensitive than silver stains. Thus, it is conceivable that the calcium-releasing or oscillin activity obtained with purified preparations was attributable to proteins other than the molecularly cloned 35 kDa oscillin. We have injected recombinant GNPDA (oscillin) into mouse eggs and detect no increase in intracellular calcium, though sperm extracts produce robust calcium oscillations ( Fig. 8). We examined a wide range of concentrations of recombinant GNPDA, including amounts comparable to those contained in the injected sperm extract and levels much higher (7–34 µg/ml; n=6). We also injected the GNPDA-GST fusion protein at a high concentration and find no change in intracellular calcium (0.04–1.3 mg/ml; n=3). An extract of transfected HEK 293 cells containing abundant GNPDA was injected at concentrations equivalent to that found in sperm extract and at much higher concentrations (0.3–457 µg/ml, n=7). No change in intracellular calcium is detected at any concentration. We wondered whether recombinant GNPDA differs in some way from the protein found in sperm, perhaps in posttranslational modifications. Accordingly, we purified GNPDA to homogeneity from hamster sperm. We used a Cibachron blue chromatographic step similar to that used by Parrington et al. (7). For further purification, we used a glucosamine-6-phosphate agarose column ( Fig. 9A). The overall purification factor is comparable to that obtained for the bacterial GNPDA when a similar procedure was used (15). SDS-PAGE gel analysis reveals a major 35 kDa band by silver stain, which is recognized by our antibody against GNPDA ( Fig. 9B). Injection of purified GNPDA protein in amounts comparable to 1 or 20 times the amount of protein present in sperm fails to produce any change in calcium in 19 of 21 cases (0.4–8 µg/ml; Fig. 8D). Though two eggs displayed calcium oscillations when injected with the purified GNPDA, the frequency of these oscillations was much lower than that found in fertilized eggs or eggs injected with sperm extract, and the response was likely an artifact given that it occurred in two rare cases.

View larger version (10K): [in this window] [in a new window]   Figure 8. GNPDA lacks Ca2+ oscillatory activity. Different GNPDA preparations were microinjected into mouse eggs (arrows), and intracellular calcium was recorded as described in Materials and Methods. A) GNPDA-GST at a final concentration in the egg at 44 µg/ml; B) recombinant GNPDA (after removal of GST from the fusion protein by thrombin) at 25 µg/ml; C) 293 cell extract transfected with GNPDA-HA at 457 µg/ml; D) sperm GNPDA purified protein at 5.4 µg/ml; E) Cibacron blue sperm extract eluate at 180 µg/ml final concentration in the egg. The data are representative of the experiments described in Results.

View larger version (14K): [in this window] [in a new window]   Figure 9. Purification of hamster sperm GNPDA. A) Purification steps of GNPDA. a, total homogenate; b, supernatant 100,000 g centrifugation; c, Cibacron blue column eluate; d, glucosamine 6-phosphate agarose column eluate. B) Western blot of GNPDA fractions. Each lane contains 60 (Fractions a and b), 30 (Fraction c), or 0.12 (Fraction d) µg protein. C) Silver-stained SDS-Page of GNPDA purified from sperm (Fraction d).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Although GNPDA was first identified as an enzymatic activity in mammalian kidney about 40 years ago, the enzyme has not previously been characterized in mammalian tissues. Our identification of molecularly cloned GNPDA, demonstration of its enzyme activity, and its localization in numerous tissues indicate that this enzyme may play an important role in cellular metabolism. GNPDA is well situated to transfer hexosamines from structural macromolecules into the glycolytic pathway ( Fig. 10). GNPDA is the only known enzyme that connects these two systems. Though GNPDA, in principle, can operate in both directions, extremely high ammonia concentrations would be required for the enzyme to function in the direction of hexosamine synthesis. Thus, under physiological conditions, GNPDA should synthesize fructose-6-phosphate, implying a biological role in delivering hexosamines into the glycolytic cascade.

View larger version (17K): [in this window] [in a new window]   Figure 10. Proposed metabolic role for GNPDA. Upon lysosomal degradation of structural macromolecules, N-acetylglucosamine is released into the cytosol (23). N-acetylglucosamine is phosphorylated to N-acetylglucosamine 6-phosphate by a kinase (Reaction 1) present in mammalian tissues (25, 26). Subsequent action of deacetylase (13, 26, 27) generates glucosamine 6-phosphate (Reaction 2). GNPDA acts on glucosamine 6-phosphate and generates fructose 6-phosphate and ammonia (Reaction 3), which will fuel the glycolytic pathway.

The possibility that GNPDA functions as an energy source is supported by the localizations we have observed. In the kidney and small intestine, GNPDA localizes to transporting epithelia whose cellular metabolic rate is among the greatest in the body. The dynamic motility of sperm consumes large amounts of energy, and sperm possess very high quantities of GNPDA. In the brain, GNPDA is localized to neurons rather than glia, and neurons are much more energy-consuming than glia. Within neurons, GNPDA appears to be concentrated in terminals, which are the most energy-consumptive portion of neurons because of the considerable energy demands associated with exocytosis and endocytosis of synaptic vesicles.

We failed to detect calcium-releasing actions of recombinant or GNPDA purified from sperm. Loss of activity during purification probably cannot explain these results, because GNPDA catalytic activity is stable to multiple freeze-thaw cycles (data not shown). Conceivably, oscillin activity of sperm derives from a complex of GNPDA and some other protein that is dissociated during the purification procedure. Our affinity chromatography purification step uses elution with the ligand N-acetylglucosamine-6-phosphate and washing under a mild conditions. Thus, only a very loosely bound protein would dissociate under these conditions. Accordingly, we suspect that calcium oscillations are elicited primarily by a protein other than GNPDA, perhaps one of the other proteins that was purified together with GNPDA by Parrington et al. (7). Recently, another sperm factor with no homology to GNPDA has been reported to induce activation of the mouse egg (24), and this or another factor may be responsible for the calcium-releasing ability of sperm extracts.

	ACKNOWLEDGMENTS

 
Supported by U.S. Public Health Service grant MH-18501 and Research Scientist Award DA00074 to S.H.S.; National Institutes of Health grant HD 31683 to D.K.; National Science Foundation Grant IBN-9631745 to R.L.S.; and predoctoral fellowship MH 10325–03 to A.M.C. H.W. is a Pew fellow. S.B. was a Howard Hughes Institute predoctoral fellow.

	FOOTNOTES

 
¹ Correspondence: Departments of Neuroscience, Pharmacology and Molecular Sciences, and Psychiatry and Behavioral Sciences, The Johns Hopkins University, School of Medicine, 725 N. Wolfe Street, Baltimore, MD 21205, USA.

² Abbreviations: GNPDA, glucosamine-6-phosphate deaminase; RT-PCR, reverse transscription-polymerase chain reaction; GST, glutathione S-transferase; NeuAc, N-acetylneuraminic acid; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Received for publication September 30, 1997. Accepted for publication October 16, 1997.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Comb, D. G., and Roseman, S. (1958) Glucosamine metabolism. IV. Glucosamine 6-phosphate deaminase. J. Biol. Chem. 232, 807–827[Free Full Text]
2. Rogers, M. J., Ohgi, T., Plumbridge, J., and Soll, D. (1988) Nucleotide sequences of the Escherichia coli and Nag B genes: the structural genes for the N-acetylglucosamine transport protein of the bacterial phosphoenolpyruvate:sugar phosphotransferase system for glucosamine 6-phosphate deaminase. Gene 62, 197–207[Medline]
3. Swann, K. (1990) A cytosolic sperm factor stimulates repetitive calcium increases and mimics fertilization in hamster eggs. Development 110, 1295–1302[Abstract/Free Full Text]
4. Wu, I. I., He, C. L., and Fissore R. A. (1997) Injection of a porcine sperm factor triggers calcium oscillations in mouse oocytes and bovine eggs. Mol. Reprod. Dev. 46, 176–189[Medline]
5. Stricker, S. A. (1997) Intracellular injections of a soluble sperm factor trigger calcium oscillations and meiotic maturation in unfertilized oocytes of a marine worm. Dev. Biol. 186, 185–201[Medline]
6. Swann, K. (1996) Soluble sperm factors and Ca²⁺ release in eggs at fertilization. Rev. Reprod. 1, 33–39[Abstract]
7. Parrington, J., Swann, K., Shevchenko, V. I., Sesay, A. K., and Lai, F. A. (1996) Calcium oscillations in mammalian eggs triggered by a soluble sperm protein. Nature (London) 379, 364–368[Medline]
8. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning, 2nd Ed, Cold Spring Harbor Laboratory Press, New York
9. Smith, B. B., and Johnson, K. S. (1988) Single-step purification of polypeptides expressed in Escherichia coli as fusions with glutathione S-transferase. Gene 67, 31–40[Medline]
10. Harlow, E., and Lane, D. (1988) Antibodies. Cold Spring Harbor Laboratory Press, New York
11. Schaeren-Wiemers, N., and Gerfin-Moser, A. (1993) A single protocol to detect transcripts of various types and expression levels in neural tissue and culture cells: in situ hybridization using digoxygenin-labelled probes. Histochemistry 100, 431–440[Medline]
12. Davis, J. S., and Gander, J. E. (1967) A reevaluation of the procedure of Roe for determination of fructose. Anal. Biochem. 19, 72–79[Medline]
13. Leloir L. F., and Cardini, C. E. (1956) Enzymes acting on glucosamine phosphates. Biochim. Biophys. Acta 20, 30–42
14. Manzi, A. E., Diaz, E., and Varhi, A. (1990) High-pressure liquid chromatography of sialic acids on a pellicular resin anion-exchange column with pulsed amperometric detection. A comparison with six other systems. Anal. Biochem. 188, 20–22[Medline]
15. Calcagno, M., Campos, P. J., Mulliert, G., and Suastegui, J. (1984) Purification, molecular and kinetic properties of glucosamine 6-phosphate isomerase (deaminase) from Escherichia coli. Biochim. Biophys. Acta 787, 165–173[Medline]
16. Kline, D., and Kline J. T. (1992) Repetitive calcium transients and the role of calcium in exocytosis and cell cycle activation in the mouse egg. Dev. Biol. 149, 80–92[Medline]
17. Mehlmann, L. M., and Kline, D. (1994) Regulation of intracellular calcium in the mouse egg: Calcium release in response to sperm or inositol trisphosphate is enhanced after meiotic maturation. Biol. Reprod. 51, 1088–1098[Abstract]
18. Kline, D., and Kline J. T. (1992) Thapsigargin activates a calcium influx pathway in the unfertilized mouse egg and suppresses repetitive calcium transients in the fertilized egg. J. Biol. Chem. 267, 17624–17630[Abstract/Free Full Text]
19. Suzuki, H., Hosokawa, Y., Nishikimi, M., and Ozawa, T. (1989) Structural organization of the human mitochondrial cytochrome c₁ gene. J. Biol. Chem. 264, 1368–1374[Abstract/Free Full Text]
20. Kay, J. F. (1996) Structure-function relationships in the FK506-binding protein (FKBP) family of peptidylprolyl cis-trans isomerases. Biochem. J. 314, 361–385
21. Jaffrey, S. R., and Snyder, S. H. (1996) PIN: an associated protein inhibitor of neuronal nitric oxide synthase. Science 274, 774–777[Abstract/Free Full Text]
22. Eriksson, G., Sarnstrand, B., and Mamstrom, A. (1984) Equilibration of [3H] glucosamine and [35S] sulfate with intracellular pools of UDP-N-acetylhexosamine and 3'- phosphoadenosine-5'-phosphosulfate (PAPS) in culture fibroblast. Arch. Biochem. Biophys. 235, 692–698[Medline]
23. Rome, L. H., and Hill, D. F. (1986) Lysosomal degradation of glycoproteins and glycosaminoglycans. Efflux and recycling of sulphate and N-acetylhexosamines. Biochem. J. 268, 707–713
24. Sette, C., Bevilacqua, A., Bianchini, A., Mangia, F., Geremia, R., and Rossi, P. (1997) Parthenogenetic activation of mouse eggs by microinjection of a truncated c-kit tyrosine kinase present in spermatozoa. Development 124, 2267–2274[Abstract]
25. Corfield, A. P., Clamp, J. R., and Wagner, S. A. (1985) The metabolism of sialic acids in isolated rat colonic mucosal cells. Biochem. J. 226, 163–174
26. Weidanz, J. A., Campbell, P., Moore, D., DeLucas, L. J., Roden, L., Thompson, J. N., and Vezza, A. C. (1996) N-acetylglucosamine kinase and N-acetylglucosamine 6-phosphate deacetylasein normal human erythrocytes and plasmodium falciparum. Br. J. Haematol. 95, 645–653
27. Weidanz, J. A., Campbell, P., DeLucas, L. J., Jin, J., Moore, D., Roden, L., Yu, H., Heilman, E., and Vezza, A. C. (1995) Glucosamine 6-phosphate deaminase in normal human erythrocytes. Br. J. Haematol. 91, 72–79[Medline]