- Email: [email protected]
Isoenzyme pattern and partial characterization of hexosaminidases in the membrane and cytosol of human erythrocytes
Clinical Biochemistry 40 (2007) 467 – 477
Isoenzyme pattern and partial characterization of hexosaminidases in the membrane and cytosol of human erythrocytes Luca Massaccesi, Adriana Lombardo, Bruno Venerando, Guido Tettamanti, Giancarlo Goi ⁎ Department of Medical Chemistry, Biochemistry and Biotechnology, University of Milan, School of Medicine, Via Saldini, 50 –20133 Milan, Italy Received 19 June 2006; received in revised form 8 November 2006; accepted 4 December 2006 Available online 12 January 2007
Abstract Objectives: Hexosaminidase activity is present in lysosomes, plasma membrane and cytosol of many human cells. Plasma membrane and cytosolic hexosaminidase is not well characterized, particularly as regards their isoenzyme forms and their relationship with the lysosomal ones. Design and methods: Erythrocyte hexosaminidase isoforms were chromatographically separated, characterized and compared to those in the plasma of healthy individuals and in the erythrocytes of a Tay–Sachs patient. Results: Hexosaminidase isoenzymes were found in plasma membrane and cytosol and were composed of the same α- and β-subunits as the lysosomal and plasma hexosaminidase A and B isoenzymes, though with some structural and kinetic differences. In addition, the cytosol contained a hexosaminidase that is a specific N-acetyl-β-D-glucosaminidase, the one involved in the removal of N-acetylglucosamine residues O-linked to proteins, named O-GlcNAcase. Conclusions: This work provides an additional step in the characterization of hexosaminidases helping better understand their role in nonlysosomal compartments and their involvement in physiological or pathological situations. © 2007 Published by The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. Keywords: Hexosaminidases; Erythrocytes; Plasma membrane; Cytosol; Tay–Sachs disease; N-Acetyl-β-D-glucosaminidase
Introduction Hexosaminidase (Hex), one of the most widely studied glycohydrolases, is a ubiquitously distributed enzyme found in the lysosomes, plasma membrane and cytosol of many mammalian cells and tissues, as well as in blood plasma and other body fluids [1–6]. The lysosomal form of the enzyme, which is responsible for the degradation of complex glycoderivatives – both Nacetylglucosamine (GlcNAc) and N-acetylgalactosamine (GalNAc) derivatives – has been extensively studied for its involvement in different lysosomal inborn errors of metabolism such as Tay–Sachs and Sandhoff diseases [7,8], and in specific acquired pathologies [9–11]. The protein structure, isoenzymatic pattern, kinetic characteristics and gene coding of the lysosomal Hex have all been defined [12,13]. The enzyme
⁎ Corresponding author. Fax: +39 02 503 16017. E-mail address: [email protected] (G. Goi).
exists as two major isoforms, an acidic “A” form (Hex-A) and a basic “B” form (Hex-B), and several minor isoenzymes, including the “S” form (the only form present in Sandhoff disease) and the so-called “intermediate forms” I1 and I2. All these forms are derived from different combinations of two polypeptide chains, α- and β-subunits, from two different genes, HEXA and HEXB, with additional heterogeneity conferred by post-translational modifications [12]. In the traffic from the site of biosynthesis to the lysosomes, part of lysosomal Hex is released into the extracellular compartment [12]. In blood plasma Hex-A, Hex-B and the intermediate forms are present as differently sialylated isoforms, with a more acidic isoelectric point than the corresponding lysosomally located isoenzymes, and are sensitive to sialidase treatment [12]. These isoenzymes increase markedly during pregnancy where a new form appears, named “P” and related to Hex-B [12,14]. The plasma I2 isoform is related to Hex-B and the lysosomal one to Hex-A [15]. In contrast, there are fewer studies on the membrane-bound and cytosolic Hex. With regard to the former, the post-translational
0009-9120/$ - see front matter © 2007 Published by The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.clinbiochem.2006.12.004
468
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
modifications that enable it to be anchored to the membrane are still not known and its physiological role is unclear. Plasma membrane Hex could be involved in glycosyl epitope changes of glycoproteins and glycosphingolipids implicated in specific cell–cell recognition events during development or neoplastic transformation [16]. Red blood cell membranes from diabetic patients have alterations to their physicochemical, dynamic, conformational and functional properties and their glycohydrolase pattern [10]. Erythrocyte membrane-bound Hex and other erythrocyte glycohydrolases have been proposed as tools to study healthy and pathological senescence [17,18]. For the cytosolic Hex, two forms, “D” and “C”, have been characterized [19–22]. The “C” form from human brain [20], which is now called O-GlcNAcase on account of its specificity of action on GlcNAc residues Olinked to proteins, has a different polypeptide composition from the lysosomal Hex and is coded by a separate gene [20,23]. Interest in this enzyme was aroused by the observation that a class of cytosolic glycoproteins carries single residues of Olinked GlcNAc, and undergoes functional modifications after insertion or detachment of GlcNAc [24]. The possible involvement of GlcNAcase in certain pathological conditions such as Alzheimer's disease, diabetes mellitus and breast cancer was suggested [25–27]. Characterization of erythrocyte Hex will enable us to establish their role in physiological and pathological situations such as diabetes and aging. Our specific objectives were: a) to study the isoenzymatic pattern of plasma membrane and cytosolic Hex; b) to define the properties and kinetic characteristics of the different isoenzymes; c) to investigate possible structural and functional relationships between the different erythrocyte forms of Hex and the lysosomal forms. We also compared the behaviour of erythrocyte Hex in a case of Tay–Sachs disease, with mutation of the HEXA gene, and deficient Hex-A activity. Materials and methods Chemicals and other products The purest available commercial chemicals were used. Water was freshly redistilled in a glass apparatus. 4-Methylumbelliferone (4-MU), purchased from Fluka GmbH (Bucks, Switzerland) was recrystallized three times from ethanol; 4-MU-2-acetamido-2-deoxy-β-D-glucopyranoside (4MU-GlcNAc), 4-MU-2acetamido-2-deoxy-β-D-galactopiranoside (4MU-GalNAc) and 4-MU-6-sulfo-2-acetamido-2-deoxy-β- D -glucopyranoside (4MU-GlcNAc-6S) were purchased from Melford (Suffolk, UK); bovine serum albumin (BSA), Triton X-100, 5,5′dithiobis-(2-nitrobenzoic acid) (DTNB), acetylthiocholine iodide, saponin, β-NADH, haemoglobin, N-acetylgalactosamine and cytochrome c were from Sigma (St. Louis, MO, USA). Polybuffer exchanger PBE-94 and polybuffer 74–HCl were purchased from Amersham Pharmacia Biotech (Uppsala, Sweden), Clostridium perfringens sialidase from Boehringer GmbH (Mannheim, Germany) and CM cellulose from Whatman Ltd (Maidstone, UK).
Subjects Blood (5–6 mL) was taken from 20 adult volunteer blood donors, males and females, aged 20–50 years, from the blood bank of the National Cancer Institute of Milan, Italy, when they attended for scheduled blood chemistry tests. All the specimens were carefully checked to make sure there were no antibodies indicative of infectious diseases such as AIDS or hepatitis. We also studied one patient with infantile GM2–gangliosidosis, Tay–Sachs variant, due to HEXA mutations, admitted to the Department of Paediatrics, Center for Metabolic Diseases, University of Padua, Italy. He presented psychomotor retardation and an eye lesion consisting of a dark red spot surrounded by a very pale ring. The biochemical diagnosis was done on leukocytes and no Hex-A activity was detected. All the individuals involved or their parents/relatives were informed about the purpose of the investigation and gave consent. Specimen preparation Erythrocytes and plasma were prepared from heparinised venous blood [2]. In brief, after collection, the blood sample was immediately centrifuged for 15 min at 3000×g and plasma withdrawn and stored with glycol ethylene. The buffy coat aspirated from the surface of the pellet was discarded and the residual material was diluted (1:1, v/v) with phosphate buffer solution (PBS) at pH 7.4 and filtered with Leucostop 4LT-B filters (Baxter, Mirandola, Italy). All platelet and leukocyte contaminants were completely removed, as previously described [2] and their absence was verified by microscopic examination. The filtered material, containing only erythrocytes, was centrifuged for 5 min at 1200×g and the pellet was washed twice (1200×g per 5 min) with PBS at pH 7.4. Unsealed erythrocyte membranes (ghosts), right-side-out vesicles (ROV) and inside-out vesicles (IOV) were prepared according to the method of Steck et al. [28]; haemoglobindepleted cytosol (CYT) was obtained using the method of Perrella et al. [29]. The specimens were assayed immediately. Hex loosely bound to erythrocyte membranes was released by treatment with 1 M NaCl followed by centrifugation, according to Goi et al. [2]. Hexosaminidase assay Hex activity was determined fluorimetrically using 4MUGlcNAc and 4MU-GalNAc as substrates. In order to determine the activity only toward GlcNAc derivatives the assay was done with 4MU-GlcNAc as substrate in the presence of GalNAc (50 mmol/L) [19]. Under these conditions any residual activity after addition of GalNAc indicates the presence of a Hex specifically acting on GlcNAc-derivatives, that is an N-acetylβ-D-glucosaminidase activity. Hex activity involving the αsubunit was assessed using 4MU-GlcNAc-6S as substrate [30]. Hex assays in ghosts, CYT, ROV, IOV and the effluents from chromatofocusing or anion exchange columns were run according to Goi et al. [2]. The incubation mixture for determining
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
469
enzyme activities in ROV contained 150 mmol/L sodium chloride. The enzyme activity in ghosts, CYT, ROV and IOV was expressed as μU/mg of proteins. Plasma Hex was assayed as reported by Lombardo et al. [3].
Results
Separation of hexosaminidase isoenzymes
The chromatofocusing elution profiles of Hex isoenzymes in the erythrocyte membrane and cytosol are presented in Fig. 1. For comparison, the figure also reports the profile for the plasma of normal subjects (panel C) which shows the abbreviation of the well-known Hex isoforms and, between them, the indication of the P isoform, present in conspicuous amounts only in pregnancy; the S form was found in plasma from patients with Sandhoff disease (data not shown). In all cases the enzyme activity was measured using 4MU-GlcNAc, in the presence or absence of GalNAc, and 4MU-GlcNAc-6S, as substrates. In erythrocyte membranes, the Hex isoenzyme pattern was similar to that of the plasma enzyme (Figs. 1A and C). Based on our experience [14] and the literature [12], two isoenzyme groups can be identified, a first comprising four isoenzymes (group I), with pI from 5.9 to 5.2, and a second (group II) also of four isoenzymes with pI from 4.7 to 4.1. The isoenzymes in group I had no activity when 4MU-GlcNAc-6S was used as substrate, indicating the absence of the α-subunit. However, group II isoenzymes, with the sole exception of the form elutable at pI 4.7 (x-form), showed activity with the selective assay for the α-subunit. All isoenzymes in both groups showed activity on 4MUGalNAc although less than on 4MU-GlcNAc. An enzyme firmly anchored to the PBE resin was elutable only with 1.0 mol/L NaCl (y-form), like the Hex-S present in this fraction in Sandhoff patients. This enzyme, in spite of its elution in the same range as the Hex-S form, was present in small amounts and was not active toward 4MU-GlcNAc-6S or inhibited by GalNAc, indicating it had features of a specific GlcNAcase. It was completely inactive on 4MU-GalNAc. The assays of all isoforms, including the x- and y-forms, were done at their optimal pH values. In the erythrocyte cytosol (Fig. 1B) the Hex isoform pattern was similar to that of the membrane enzyme, the only difference being the great abundance of the y-form.
Automated chromatofocusing on PBE-94 The method of Goi et al. [14] was used. Briefly: 1.5 mL of ghost preparation (4 mg as proteins), or 2.5 mL of the supernatant after ghost treatment with 1 M NaCl (4 mg as proteins), or 0.2 mL of plasma, or 5 mL of CYT were dialyzed for 2 h in 25 mmol/L imidazole–HCl buffer, pH 7.4. Then the specimens were injected into a PBE-94 column (3.0 × 210 mm), previously equilibrated and washed with the same buffer for 5 min. At this point, 7.5 mmol/L per pH unit of polybuffer 74– HCl, pH 4.0, was pumped through the column so as to produce a linear pH elution gradient for 45 min, after which the column was washed for 10 min with 1.0 mol/L aqueous NaCl solution. All the solutions contained Triton X-100, 0.3% final concentration. The flow rate was 0.84 mL/min. The effluent was collected in 0.5-mL aliquots and immediately utilized for the enzyme assay. FPLC system with anion exchange column (using ÄKTApurifier from Amersham Pharmacia Biotech AB Uppsala, Sweden) Briefly: 0.3 mL of plasma or 5.0 mL of CYT were dialyzed for 2 h in 20 mmol/L Tris–HCl buffer pH 7.5. Then the specimens were injected into a mono Q HR 5/5 column, previously equilibrated and washed with the same buffer for 5 min. At this point 1.0 mol/L NaCl dissolved in the same buffer was pumped through the column so as to produce a linear gradient from 0 to 0.5 mol/L in 30 min. Afterwards the column was washed for 10 min with 1.0 mol/L NaCl solution in Tris buffer. The flow rate was 1.0 mL/min. The effluent was collected in 0.5-mL aliquots and immediately utilized for the enzyme assay. Sialidase treatment Erythrocyte ghosts, cytosol (4 mg protein in each case) and plasma (20 mg protein) were buffered at pH 7.4 with 25 mmol/L imidazole–HCl and treated with 300 mU of C. perfringens sialidase dissolved in the same buffer. The mixture was incubated at 37 °C for the time needed for complete release of sialic acid (10–30 min) and then chromatofocused on PBE 94. Sialic acid released was determined according to Caimi et al. [31]. Other methods The protein content in all specimens was determined by the method of Lowry et al. [32], using BSA as the reference standard. NADH–cytochrome c oxidoreductase (NADH–CytC Ox/Red) (EC 1.6.99.3) activity was determined by the method of McIntosh et al. [33], and acetylcholinesterase (AchEsterase) (EC 3.1.1.7) activity by that of Ellman et al. [34].
Chromatofocusing separation of hexosaminidase isoenzymes in erythrocyte ghosts and cytosol from normal subjects
Characteristics and properties of major hexosaminidase isoenzymes in erythrocyte membranes and cytosol The chromatofocusing fractions corresponding to each of the different isoenzymes were pooled and investigated for optimal pH and kinetic characteristics, using 4MU-GlcNAc as substrate, with or without GalNAc. In the membrane, group I and II isoenzymes had the same optimum pH of 4.4, with Km of respectively 0.40 and 1.03 mmol/L, and Vmax of 38 and 25 μU/mg protein. The y-form had optimum pH in the range of 5.0 to 6.0, an apparent Km of 0.80 mmol/L and Vmax of 5 μU/mg protein. Estimated total activities of Hex, groups I and II isoenzymes, and of the yform (GlcNAcase) were respectively 20.4 and 2.02 μU/109 erythrocytes. In the cytosol, isoenzyme groups I and II had the same optimum pH of 5.2, Km of 0.44 and 0.98, Vmax of 60 and 48 μU/
470
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
Fig. 1. Chromatofocusing elution profiles of erythrocyte membrane and cytosol hexosaminidase isoenzymes from normal human subjects. For comparison, the hexosaminidase isoenzyme pattern of human plasma is also presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MUGlcNAc-6S as substrates.
mg protein. The y-form had optimum pH in the range of 5.6 to 6.6, Km of 0.40 mmol/L and an apparent Vmax of 45 μU/mg protein. Estimated total activities of Hex groups I and II isoenzymes, and of the y-form (GlcNAcase) were respectively 150 and 52 μU/109 erythrocytes. Both isoenzyme groups, in either cytosol or membranes, were stable for 1 week at +4 °C and for 40 days at −40 °C and did not lose their activity with freezing and thawing. They were
completely inhibited by GalNAc at 50 mmol/L. However, the group I isoenzymes were resistant to 3 h at 52 °C, whereas group II, with the exception of the x-form, completely lost their activity in these conditions. The y-form, which was not inhibited by GalNAc, was completely denatured after 30 min at 52 °C. Its activity decayed rapidly after freezing at −20 °C and thawing. As shown in Fig. 2, the chromatofocusing patterns of erythrocyte membrane and cytosol Hex did not change at all
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
471
Fig. 2. Chromatofocusing elution profiles of erythrocyte membrane and cytosol hexosaminidase isoenzymes from normal human subjects before and after sialidase treatment. For comparison, the hexosaminidase isoenzyme pattern of human plasma, also treated with sialidase, is presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc.
after sialidase treatment, indicating either complete resistance to sialidase of the putative sialic acid residues or complete absence of sialylation. In contrast, all the plasma Hex isoforms, known
to contain sialic acid [12], shifted markedly toward less acidic pH, indicating the loss of negative charges consistent with the removal of sialic acid residues.
472
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
Anion exchange chromatographic separation of hexosaminidase isoenzymes in erythrocyte cytosol from controls To further characterize the y-form, abundant in the erythrocyte cytosol, cytosolic Hex isoforms were separated by the FPLC system with anion exchange column (Fig. 3). For comparison, the isoform profile from plasma is also given. Hex activity was assayed with 4MU-GlcNAc (with or without GalNAc), 4MUGalNAc, or 4MU-GlcNAc-6S. A first group of three isoenzymes was eluted between 50 and 200 mM NaCl. None had any activity on 4MU-GlcNAc-6S and all were totally inhibited by GalNAc, resembling the group I isoenzymes presented in Fig. 1, panel B. A second group (group II) of three isoenzymes was eluted between 280 and 400 mM NaCl. These were active on 4MUGlcNAc-6S and were completely inhibited by GalNAc, again resembling the group II isoenzymes shown in Fig. 1, panel B. Finally, one enzyme was eluted between 430 and 520 mM NaCl. It was not inhibited by GalNAc, and was not active on 4MUGlcNAc-6S and 4MU-GalNAc, similarly to the y-form in Fig. 1, panel B, thus having the features of GlcNAcase. The group I and II isoforms hydrolyzed 4MU-GalNAc although to a lesser extent (about ten times less) than on 4MU-
GlcNAc. The isoenzyme profile in plasma featured two groups of isoforms, similarly to the groups I and II from erythrocyte cytosol, although they eluted at somewhat different NaCl concentrations. These two groups appeared to correspond to the known Hex-B and Hex-A groups of isoenzymes [12]. No y-form was present in plasma, confirming the chromatofocusing separation findings. Chromatofocusing separation of hexosaminidase isoenzymes in erythrocyte ghosts and cytosol in a case of GM2–gangliosidosis (Tay–Sachs variant) The isoenzyme patterns of erythrocytes (ghost and cytosol) and plasma Hex from a patient with Tay–Sachs disease is presented in Fig. 4. In the plasma there were no isoenzymes of the Hex-A group, but I2–P was present, with those of the Hex-B group that were markedly elevated but completely inhibited by GalNAc. The isoenzyme profiles in the erythrocyte membranes and cytosol, with the exception of the GlcNAcase (y-form), that is present in the erythrocyte samples, essentially overlapped those of plasma. As in plasma, group II isoenzymes were absent while
Fig. 3. Anion exchange elution profiles of erythrocyte cytosol hexosaminidase isoenzymes from human normal subjects. For comparison, the hexosaminidase isoenzyme pattern of human plasma is also presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MU-GlcNAc-6S as substrates.
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
473
Fig. 4. Chromatofocusing elution profiles of erythrocyte membrane and cytosol hexosaminidase isoenzymes of a Tay–Sachs patient. For comparison, the hexosaminidase isoenzyme pattern in the plasma of the same patient is presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MU-GlcNAc-6S as substrates.
group I was markedly increased. The x-form was present as was GlcNAcase (y-form), which is insensitive to GalNAc and not active on 4MU-GalNAc, in either erythrocyte membranes or cytosol. No activity was seen on 4MU-GlcNAc-6S in either erythrocytes or plasma, as expected. The membrane-linked y-form was much more abundant in Tay–Sachs than normal erythrocytes.
Chromatofocusing separation of hexosaminidase isoenzymes of erythrocyte ghosts released by treatment with 1 M NaCl Confirming previous evidence [2], treatment of erythrocyte ghosts with 1 M NaCl caused a substantial release of Hex activity. The chromatofocusing isoenzyme profile (Fig. 5) shows an interesting difference compared to that of untreated
474
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
Fig. 5. Chromatofocusing separation of hexosaminidase isoenzymes of erythrocyte ghosts from normal human subjects released by treatment with 1 M NaCl. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MU-GlcNAc-6S as substrates.
ghosts. The group I isoforms and the x-form remained essentially the same, indicating that these isoform were partly firmly attached to the membrane, and partly (the released ones) only loosely linked, probably with a similar molecular architecture. Instead, all the subunits acting on 4MU-GlcNAc6S (α-subunits) were no longer part of the group II isoforms and were elutable from the column with 1 M NaCl, a response similar to that of the y-form. Hexosaminidase orientation in human erythrocyte membranes The orientation of the major Hex activity in erythrocyte membranes (optimum pH 4.4) was assessed using ROVand IOV vesicles prepared from erythrocyte membranes (Table 1). This activity was seen predominantly in ROV vesicles (83%), with only 15% in IOV vesicles. The pattern of Hex activity was essentially identical to that of AchEsterase, a marker of the membrane external surface, and quite different from NADH– Cyt-C Ox/Red, a marker of the internal surface, indicating that the orientation of Hex is mainly to the external surface of the plasma membrane. It was not possible to establish the orientation
of the membrane-bound GlcNAcase because of its extremely low activity in normal erythrocytes. Discussion There is evidence that glycohydrolases, or at least some of them, occur not only in the lysosomes but also in the plasma membrane and cytosol of mammalian cells [2,6,35–38]. The lysosomal, plasma membrane and cytosol sialidases have different primary structures and are coded by different genes [39]. The lysosomal hexosaminidase, which is the better known,
Table 1 Presence of hexosaminidase at pH 4.4 in oriented vesicles from human erythrocytes membrane Enzyme source
Hexosaminidase AchEsterase NADH–Cyt-C Ox/ (μU/mg of protein) (U/mg protein) Red (mU/mg protein)
Unsealed ghosts 31.1 ± 15.3 ROV 26.2 ± 12.5 IOV 4.7 ± 2.3
2.55 ± 0.43 2.04 ± 0.35 0.31 ± 0.11
The data are the mean values ± S.D. of six experiments.
23.2 ± 9.6 2.1 ± 0.6 22.5 ± 8.5
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
exists as two major isoforms, Hex-A and Hex-B, resulting from the combination of the two polypeptides (α- and β-subunits) coded by two different genes, HEXA and HEXB. The plasma membrane-linked enzyme's structural features are still little known but recently Mencarelli et al. [6] described and characterized a form of Hex-A from plasma membranes of human fibroblasts, active on GM2 ganglioside, and speculated about its lysosomal origin. The cytosolic enzyme so far described appears specifically to affect only GlcNAc derivatives (it is actually an N-acetylglucosaminidase) and is coded by a MGEA5 gene [20,23,40]; on account of its intrinsic histone acetyltransferase activity [41], it is nowadays known as “nuclear cytoplasmic O-GlcNAcase and acetyltransferase” (NCOAT). The present investigation studied Hex isoenzymes in the plasma membrane and cytosol of human erythrocytes, which have no lysosomes, employing isoform separation techniques already applied to lysosomal Hex-A and Hex-B [2,14], and commonly used enzyme assays [19], with the substrate 4MUGlcNAc with or without GalNAc (as competitive inhibitor), to assess the specificity toward GlcNAc-derivatives, or 4MUGalNAc, or 4MU-GlcNAc-6S as substrate to recognize the activity of the α-subunit in Hex-A isoenzyme. Further information was obtained by comparing the erythrocyte Hex isoenzymes with those of blood plasma from control subjects and with the erythrocytes and plasma from a Tay–Sachs patient that entirely lack Hex-A. A first finding was that Hex are found in human erythrocyte plasma membranes, confirming previous observations [2,6], and cytosol – this was a novel finding; we also report that the plasma membrane enzymes are mostly oriented toward the extracellular environment, as recently demonstrated for another membrane glycohydrolase, acidic sialidase [42]. In both the membrane and cytosol, the chromatofocusing profiles of Hex isoenzymes were similar, with two main isoenzyme groups, I and II, with isoelectric points similar to those of plasma Hex-B and Hex-A groups. The membrane and cytosolic group I isoenzymes had similar kinetics to the plasma Hex-B group, were active toward both GlcNAc and GalNAc derivatives, sensitive to inhibition by GalNAc and had no activity on 4MUGlcNAc-6S, indicating the absence of the α-subunit. As expected, in the erythrocytes from the Tay–Sachs patient (where only the β-subunits are present) conspicuous amounts of the group I isoenzymes were found in both the membrane and cytosol. The membrane and cytosolic group II isoenzymes showed substantial similarity to the plasma Hex-A group, sharing the same kinetics and being inhibited by GalNAc and active toward 4MU-GlcNAc-6S. This indicates the presence of the α-subunit. With the exception of the x-form, the plasma membrane and cytosolic group II isoenzymes were completely absent in the Tay–Sachs erythrocytes. Interestingly, the x-form, which is not active toward 4MU-GlcNAc-6S, thus excluding the presence of α-subunits, was the only group II isoenzyme in the Tay–Sachs patient, possibly corresponding to the I2 isoform [12]. A very intriguing point, surely worth further investigation, was that when erythrocyte ghosts were treated with 1 M NaCl the
475
released (i.e. the loosely bound) Hex-B isoforms had the same chromatofocusing profile as the unreleased isoform – the tightly bound one, whereas Hex-A isoforms released behaved quite differently, consistent with the presence of separated isoforms. These results raise a question: how does each isoenzyme reach the plasma membrane or the cytosol? For membranebound Hex, attached to the external leaflet, we can exclude aspecific adsorption of the plasma circulating isoenzymes. In fact, whereas the plasma isoenzymes are known to be affected by sialidase [12] – as we also observed – the membrane-bound ones are not affected at all, nor are the cytosolic isoforms, under conditions where the release of sialic acid (i.e. the activity of sialidase) was measured (data not shown). We can also exclude that they are linked to the mannose-6-phosphate receptor on the erythrocyte surface since even increasing concentrations of mannose-6-phosphate, as competitive ligand, did not succeed in releasing the enzyme from the erythrocyte membranes (data not shown). For cytosolic Hex, we can exclude an origin from contaminating blood cells because the erythrocyte suspension was completely free of any leukocytes or platelets before erythrocyte lysis. Moreover, the optimal pH of cytosolic Hex is less acidic than that of the lysosomal Hex. We can only suggest that the membrane or cytosolic localization of these isoenzymes is due to post-translational modifications that are somehow different from those that lead the enzymes to be released from the cells or targeted to the lysosomal compartment. This study also turned up another finding, concerning the Hex y-form, which appears to be a specific GlcNAcase. It was abundant in the erythrocyte cytosol of normal subjects and of the patient with GM2–gangliosidosis. On the basis of some properties, such as the pH curve with a characteristic plateau, the thermal instability, insensitivity to any inhibitory effect by GalNAc, and the lack of activity on 4MU-GalNAc, we suggest that the cytosolic erythrocyte GlcNAcase might be the OGlcNAcase purified and characterized in other tissues, cloned and identified as isoform C [19–22]. This enzyme's role in the dynamic process of O-GlcNAc–glycoprotein modification has been suggested [20,41], as well as in the rearrangement of the glycidic moieties of proteins required for connections between erythrocyte membrane and cytoskeleton. Band 4.1, a protein that serves as a bridge joining the cytoskeleton to the inner surface of the plasma membrane, also contains O-GlcNAc moieties [43]. An additional interesting finding was the presence of GlcNAcase activity in the erythrocyte ghosts from normal subjects, and the marked increase of this enzyme form in the ghosts of the Tay–Sachs case. This needs to be investigated in more detail, in order to clarify whether the “canonical” OGlcNAcase in the erythrocyte cytosol is structurally connected with the plasma membrane-bound GlcNAcase, and why the ghost-linked GlcNAcase was so abundant in Tay–Sachs erythrocytes. To conclude, this study showed that both the plasma membrane and cytosol of human erythrocytes have Hex isoenzymes quite likely originating from translational modifications of the products of HEXA and HEXB genes, coding for
476
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
lysosomal Hex. This finding makes an important step forward in the knowledge of the biology of the lysosomal–endosomal compartment in relation to cell plasma membrane generation and/or turnover and is probably related to the recent observation that some cells, mostly from the haematopoietic lineage, contain secretory lysosomes, call exosomes in erythrocytes [44,45]. Generation of exosomes is a critical mechanism for shedding membrane proteins in the maturing red cell [46] and it is likely that during erythrocyte maturation, with a mechanism still to be clarified, non-integral proteins of the late lysosomal–endosomal compartment are also incorporated in the membrane. Some exosomes can “explode” into reticulocytes [47,48], thus explaining their presence in cytosol too. Additional important evidence concerns the presence of the O-GlcNAcase in the erythrocyte cytosol as well as in the membrane, the latter form being particularly abundant in the Tay–Sachs patient. It will be interesting to study the relationships between the cytosolic and membrane-bound forms of this enzyme. The data also offer a starting point for assessing the possible clinical involvement of these erythrocyte glycohydrolases. Oxidative stress is a feature of the development of age-related chronic degenerative diseases, such as diabetes, atherosclerosis, hypertension, Parkinson's and Alzheimer's [49]. In these pathologies, both the generation rates of oxidants and the susceptibility of tissues to oxidative damage influence the integrity of cell membranes of all tissues [50] and the erythrocyte plasma membrane, that is easy to obtain with high purity and could serve as a useful model for following the functional condition of membranes in these age-related degenerative diseases. Acknowledgments This work was supported in part by the Italian Ministry for Education, University and Research (COFIN grant 2004 to Guido Tettamanti). We thank Dr. Alberto Burlina, Department of Pediatrics, University of Padua, Padua, Italy, for kindly supplying the blood specimen from a Tay–Sachs disease patient. References [1] Goi G, Fabi A, Lombardo A, et al. Stability of enzymes of lysosomal origin in human cerebrospinal fluid. Clin Chim Acta 1987;163:215–24. [2] Goi G, Bairati C, Massaccesi L, et al. Membrane anchoring and surface distribution of glycohydrolases of human erythrocyte membranes. FEBS Lett 2000;473:89–94. [3] Lombardo A, Caimi L, Marchesini S, Goi GC, Tettamanti G. Enzymes of lysosomal origin in human plasma and serum: assay conditions and parameters influencing the assay. Clin Chim Acta 1980;108:337–46. [4] Hutchinson T, Dwivedi K, Rastogi A, Prasad R, Pereira BM. N-Acetyl beta-D-glucosaminidase is not attached to human sperm membranes through the glycosylphosphatidyl inositol (GPI)-anchor. Asian J Androl 2002;4:27–33. [5] Izumi T, Suzuki K. Neutral hydrolases of rat brain. Preliminary characterization and developmental changes of neutral beta-N-acetylhexosamindases. Biochim Biophys Acta 1980;615:402–13. [6] Mencarelli S, Cavalieri C, Magini A, et al. Identification of plasma membrane associated mature beta-hexosaminidase A, active towards GM2 ganglioside, in human fibroblasts. FEBS Lett 2005;579:5501–6.
[7] Kustermann-Kuhn B, Harzer K. Prenatal diagnosis of Tay–Sachs disease. Reflectometry of hexosaminidase A, B, and C/S bands on zymograms. Hum Genet 1983;65:172–5. [8] Neufeld EF. Natural history and inherited disorders of a lysosomal enzyme, beta-hexosaminidase. J Biol Chem 1989;264:10927–30. [9] Burlina AB, Goi G, Fabi A, et al. Behaviour of some lysosomal enzymes in the plasma of insulin dependent diabetic patients during artificial pancreas treatment. Clin Biochem 1987;20:423–7. [10] Goi G, Bairati C, Segalini G, et al. Alterations in the activity of several glycohydrolases in red blood cell membrane from type 2 diabetes mellitus patients. Metabolism 1999;48:817–21. [11] Reglero A, Carretero MI, Cabezas JA. Increased serum alpha-L-fucosidase and beta-N-acetylglucosaminidase activities in diabetic, cirrhotic and gastric cancer patients. Clin Chim Acta 1980;103:155–8. [12] Mahuran D, Novak A, Lowden JA. The lysosomal hexosaminidase isozymes. Isozymes Curr Top Biol Med Res 1985;12:229–88. [13] Grave RA, Kaback MM, Proia RL, Sandhoff K, Suzuki K, Suzuki K. The GM2 gangliosidosis 8th ed, 2001. [14] Goi G, Fabi A, Lombardo A, et al. The lysosomal beta-D-N-acetylglucosaminidase isozymes in human plasma during pregnancy: separation and quantification by a simple automated procedure. Clin Chim Acta 1989; 179:327–40. [15] Mahuran DJ. Characterization of human placental beta-hexosaminidase I2. Proteolytic processing intermediates of hexosaminidase A. J Biol Chem 1990;265:6794–9. [16] Hakomori S. Glycosylation defining cancer malignancy: new wine in an old bottle. Proc Natl Acad Sci U S A 2002;99:10231–3. [17] Goi G, Cazzola R, Tringali C, et al. Erythrocyte membrane alterations during ageing affect beta-D-glucuronidase and neutral sialidase in elderly healthy subjects. Exp Gerontol 2005;40:219–25. [18] Massaccesi L, Corsi MM, Baquero-Herrera CJ, et al. Erythrocyte glycohydrolases in subjects with trisomy 21: could Down's syndrome be a model of accelerated ageing? Mech Ageing Dev 2006;127:324–31. [19] Dong DL, Hart GW. Purification and characterization of an O-GlcNAc selective N-acetyl-beta-D-glucosaminidase from rat spleen cytosol. J Biol Chem 1994;269:19321–30. [20] Gao Y, Wells L, Comer FI, Parker GJ, Hart GW. Dynamic O-glycosylation of nuclear and cytosolic proteins: cloning and characterization of a neutral, cytosolic beta-N-acetylglucosaminidase from human brain. J Biol Chem 2001;276:9838–45. [21] Izumi T, Suzuki K. Neutral beta-N-acetylhexosaminidases of rat brain. Purification and enzymatic and immunological characterization. J Biol Chem 1983;258:6991–9. [22] Ueno R, Yuan CS. Purification and properties of neutral beta-Nacetylglucosaminidase from carp blood. Biochim Biophys Acta 1991; 1074:79–84. [23] Heckel D, Comtesse N, Brass N, et al. Novel immunogenic antigen homologous to hyaluronidase in meningioma. Hum Mol Genet 1998; 7:1859–72. [24] Comer FI, Hart GW. O-Glycosylation of nuclear and cytosolic proteins. Dynamic interplay between O-GlcNAc and O-phosphate. J Biol Chem 2000;275:29179–82. [25] Cooksey RC, Hebert Jr LF, Zhu JH, et al. Mechanism of hexosamineinduced insulin resistance in transgenic mice overexpressing glutamine : fructose-6-phosphate amidotransferase: decreased glucose transporter GLUT4 translocation and reversal by treatment with thiazolidinedione. Endocrinology 1999;140:1151–7. [26] Slawson C, Pidala J, Potter R. Increased N-acetyl-beta-glucosaminidase activity in primary breast carcinomas corresponds to a decrease in Nacetylglucosamine containing proteins. Biochim Biophys Acta 2001;1537: 147–157. [27] Yao PJ, Coleman PD. Reduction of O-linked N-acetylglucosaminemodified assembly protein-3 in Alzheimer's disease. J Neurosci 1998; 18:2399–411. [28] Steck TL, Kant JA. Preparation of impermeable ghosts and inside-out vesicles from human erythrocyte membranes. Methods Enzymol 1974; 31:172–80. [29] Perrella M, Benazzi L, Ripamonti M, Rossi-Bernardi L. Bohr effect in
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
[30]
[31]
[32] [33]
[34]
[35]
[36] [37]
[38] [39]
[40]
hemoglobin deoxy/cyanomet intermediates. Biochemistry 1994;33: 10358–10366. Fuchs W, Navon R, Kaback MM, Kresse H. Tay–Sachs disease: one-step assay of beta-N-acetylhexosaminidase in serum with a sulphated chromogenic substrate. Clin Chim Acta 1983;133:253–61. Caimi L, Lombardo A, Preti A, Wiesmann U, Tettamanti G. Optimal conditions for the assay of fibroblast neuraminidase with different natural substrates. Biochim Biophys Acta 1979;571:137–46. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265–75. McIntosh CH, Plummer DT. The subcellular localization of acetylcholinesterase and its molecular forms in pig cerebral cortex. J Neurochem 1976;27:449–57. Ellman GL, Courtney KD, Andres Jr V, Feather-Stone RM. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 1961;7:88–95. Haeuw JF, Strecker G, Wieruszeski JM, Montreuil J, Michalski JC. Substrate specificity of rat liver cytosolic alpha-D-mannosidase. Novel degradative pathway for oligomannosidic type glycans. Eur J Biochem 1991;202:1257–68. Haeuw JF, Michalski JC, Strecker G, Spik G, Montreuil J. Cytosolic glycosidases: do they exist? Glycobiology 1991;1:487–92. Comtesse N, Maldener E, Meese E. Identification of a nuclear variant of MGEA5, a cytoplasmic hyaluronidase and a beta-N-acetylglucosaminidase. Biochem Biophys Res Commun 2001;283:634–40. Berrin JG, Czjzek M, Kroon PA, et al. Substrate (aglycone) specificity of human cytosolic beta-glucosidase. Biochem J 2003;373:41–8. Monti E, Preti A, Venerando B, Borsani G. Recent development in mammalian sialidase molecular biology. Neurochem Res 2002;27: 649–663. Lehman DM, Fu DJ, Freeman AB, et al. A single nucleotide polymorphism in MGEA5 encoding O-GlcNAc-selective N-acetyl-beta-D
[41]
[42]
[43]
[44]
[45] [46]
[47] [48]
[49]
[50]
477
glucosaminidase is associated with type 2 diabetes in Mexican Americans. Diabetes 2005;54:1214–21. Toleman C, Paterson AJ, Whisenhunt TR, Kudlow JE. Characterization of the histone acetyltransferase (HAT) domain of a bifunctional protein with activable O-GlcNAcase and HAT activities. J Biol Chem 2004;279: 53665–73. Chiarini A, Fiorilli A, Di Francesco L, Venerando B, Tettamanti G. Human erythrocyte sialidase is linked to the plasma membrane by a glycosylphosphatidylinositol anchor and partly located on the outer surface. Glycoconjugate J 1993;10:64–71. Holt GD, Haltiwanger RS, Torres CR, Hart GW. Erythrocytes contain cytoplasmic glycoproteins. O-Linked GlcNAc on Band 4.1. J Biol Chem 1987;262:14847–50. Pan BT, Teng K, Wu C, Adam M, Johnstone RM. Electron microscopic evidence for externalization of the transferrin receptor in vesicular form in sheep reticulocytes. J Cell Biol 1985;101:942–8. Blott EJ, Griffiths GM. Secretory lysosomes. Nat Rev, Mol Cell Biol 2002;3:122–31. Fader CM, Savina A, Sanchez D, Colombo MI. Exosome secretion and red cell maturation: exploring molecular components involved in the docking and fusion of multivesicular bodies in K562 cells. Blood Cells Mol Dis 2005;35:153–7. Koury MJ, Koury ST, Kopsombut P, Bondurant MC. In vitro maturation of nascent reticulocytes to erythrocytes. Blood 2005;105:2168–74. Artal-Sanz M, Samara C, Syntichaki P, Tavernarakis N. Lysosomal biogenesis and function is critical for necrotic cell death in Caenorhabditis elegans. J Cell Biol 2006;173:231–9. Ames BN, Shigenaga MK, Hagen TM. Oxidants, antioxidants, and the degenerative diseases of aging. Proc Natl Acad Sci U S A 1993; 90:7915–22. Droge W. Oxidative stress and aging. Adv Exp Med Biol 2003;543: 191–200.
Isoenzyme pattern and partial characterization of hexosaminidases in the membrane and cytosol of human erythrocytes Luca Massaccesi, Adriana Lombardo, Bruno Venerando, Guido Tettamanti, Giancarlo Goi ⁎ Department of Medical Chemistry, Biochemistry and Biotechnology, University of Milan, School of Medicine, Via Saldini, 50 –20133 Milan, Italy Received 19 June 2006; received in revised form 8 November 2006; accepted 4 December 2006 Available online 12 January 2007
Abstract Objectives: Hexosaminidase activity is present in lysosomes, plasma membrane and cytosol of many human cells. Plasma membrane and cytosolic hexosaminidase is not well characterized, particularly as regards their isoenzyme forms and their relationship with the lysosomal ones. Design and methods: Erythrocyte hexosaminidase isoforms were chromatographically separated, characterized and compared to those in the plasma of healthy individuals and in the erythrocytes of a Tay–Sachs patient. Results: Hexosaminidase isoenzymes were found in plasma membrane and cytosol and were composed of the same α- and β-subunits as the lysosomal and plasma hexosaminidase A and B isoenzymes, though with some structural and kinetic differences. In addition, the cytosol contained a hexosaminidase that is a specific N-acetyl-β-D-glucosaminidase, the one involved in the removal of N-acetylglucosamine residues O-linked to proteins, named O-GlcNAcase. Conclusions: This work provides an additional step in the characterization of hexosaminidases helping better understand their role in nonlysosomal compartments and their involvement in physiological or pathological situations. © 2007 Published by The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. Keywords: Hexosaminidases; Erythrocytes; Plasma membrane; Cytosol; Tay–Sachs disease; N-Acetyl-β-D-glucosaminidase
Introduction Hexosaminidase (Hex), one of the most widely studied glycohydrolases, is a ubiquitously distributed enzyme found in the lysosomes, plasma membrane and cytosol of many mammalian cells and tissues, as well as in blood plasma and other body fluids [1–6]. The lysosomal form of the enzyme, which is responsible for the degradation of complex glycoderivatives – both Nacetylglucosamine (GlcNAc) and N-acetylgalactosamine (GalNAc) derivatives – has been extensively studied for its involvement in different lysosomal inborn errors of metabolism such as Tay–Sachs and Sandhoff diseases [7,8], and in specific acquired pathologies [9–11]. The protein structure, isoenzymatic pattern, kinetic characteristics and gene coding of the lysosomal Hex have all been defined [12,13]. The enzyme
⁎ Corresponding author. Fax: +39 02 503 16017. E-mail address: [email protected] (G. Goi).
exists as two major isoforms, an acidic “A” form (Hex-A) and a basic “B” form (Hex-B), and several minor isoenzymes, including the “S” form (the only form present in Sandhoff disease) and the so-called “intermediate forms” I1 and I2. All these forms are derived from different combinations of two polypeptide chains, α- and β-subunits, from two different genes, HEXA and HEXB, with additional heterogeneity conferred by post-translational modifications [12]. In the traffic from the site of biosynthesis to the lysosomes, part of lysosomal Hex is released into the extracellular compartment [12]. In blood plasma Hex-A, Hex-B and the intermediate forms are present as differently sialylated isoforms, with a more acidic isoelectric point than the corresponding lysosomally located isoenzymes, and are sensitive to sialidase treatment [12]. These isoenzymes increase markedly during pregnancy where a new form appears, named “P” and related to Hex-B [12,14]. The plasma I2 isoform is related to Hex-B and the lysosomal one to Hex-A [15]. In contrast, there are fewer studies on the membrane-bound and cytosolic Hex. With regard to the former, the post-translational
0009-9120/$ - see front matter © 2007 Published by The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.clinbiochem.2006.12.004
468
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
modifications that enable it to be anchored to the membrane are still not known and its physiological role is unclear. Plasma membrane Hex could be involved in glycosyl epitope changes of glycoproteins and glycosphingolipids implicated in specific cell–cell recognition events during development or neoplastic transformation [16]. Red blood cell membranes from diabetic patients have alterations to their physicochemical, dynamic, conformational and functional properties and their glycohydrolase pattern [10]. Erythrocyte membrane-bound Hex and other erythrocyte glycohydrolases have been proposed as tools to study healthy and pathological senescence [17,18]. For the cytosolic Hex, two forms, “D” and “C”, have been characterized [19–22]. The “C” form from human brain [20], which is now called O-GlcNAcase on account of its specificity of action on GlcNAc residues Olinked to proteins, has a different polypeptide composition from the lysosomal Hex and is coded by a separate gene [20,23]. Interest in this enzyme was aroused by the observation that a class of cytosolic glycoproteins carries single residues of Olinked GlcNAc, and undergoes functional modifications after insertion or detachment of GlcNAc [24]. The possible involvement of GlcNAcase in certain pathological conditions such as Alzheimer's disease, diabetes mellitus and breast cancer was suggested [25–27]. Characterization of erythrocyte Hex will enable us to establish their role in physiological and pathological situations such as diabetes and aging. Our specific objectives were: a) to study the isoenzymatic pattern of plasma membrane and cytosolic Hex; b) to define the properties and kinetic characteristics of the different isoenzymes; c) to investigate possible structural and functional relationships between the different erythrocyte forms of Hex and the lysosomal forms. We also compared the behaviour of erythrocyte Hex in a case of Tay–Sachs disease, with mutation of the HEXA gene, and deficient Hex-A activity. Materials and methods Chemicals and other products The purest available commercial chemicals were used. Water was freshly redistilled in a glass apparatus. 4-Methylumbelliferone (4-MU), purchased from Fluka GmbH (Bucks, Switzerland) was recrystallized three times from ethanol; 4-MU-2-acetamido-2-deoxy-β-D-glucopyranoside (4MU-GlcNAc), 4-MU-2acetamido-2-deoxy-β-D-galactopiranoside (4MU-GalNAc) and 4-MU-6-sulfo-2-acetamido-2-deoxy-β- D -glucopyranoside (4MU-GlcNAc-6S) were purchased from Melford (Suffolk, UK); bovine serum albumin (BSA), Triton X-100, 5,5′dithiobis-(2-nitrobenzoic acid) (DTNB), acetylthiocholine iodide, saponin, β-NADH, haemoglobin, N-acetylgalactosamine and cytochrome c were from Sigma (St. Louis, MO, USA). Polybuffer exchanger PBE-94 and polybuffer 74–HCl were purchased from Amersham Pharmacia Biotech (Uppsala, Sweden), Clostridium perfringens sialidase from Boehringer GmbH (Mannheim, Germany) and CM cellulose from Whatman Ltd (Maidstone, UK).
Subjects Blood (5–6 mL) was taken from 20 adult volunteer blood donors, males and females, aged 20–50 years, from the blood bank of the National Cancer Institute of Milan, Italy, when they attended for scheduled blood chemistry tests. All the specimens were carefully checked to make sure there were no antibodies indicative of infectious diseases such as AIDS or hepatitis. We also studied one patient with infantile GM2–gangliosidosis, Tay–Sachs variant, due to HEXA mutations, admitted to the Department of Paediatrics, Center for Metabolic Diseases, University of Padua, Italy. He presented psychomotor retardation and an eye lesion consisting of a dark red spot surrounded by a very pale ring. The biochemical diagnosis was done on leukocytes and no Hex-A activity was detected. All the individuals involved or their parents/relatives were informed about the purpose of the investigation and gave consent. Specimen preparation Erythrocytes and plasma were prepared from heparinised venous blood [2]. In brief, after collection, the blood sample was immediately centrifuged for 15 min at 3000×g and plasma withdrawn and stored with glycol ethylene. The buffy coat aspirated from the surface of the pellet was discarded and the residual material was diluted (1:1, v/v) with phosphate buffer solution (PBS) at pH 7.4 and filtered with Leucostop 4LT-B filters (Baxter, Mirandola, Italy). All platelet and leukocyte contaminants were completely removed, as previously described [2] and their absence was verified by microscopic examination. The filtered material, containing only erythrocytes, was centrifuged for 5 min at 1200×g and the pellet was washed twice (1200×g per 5 min) with PBS at pH 7.4. Unsealed erythrocyte membranes (ghosts), right-side-out vesicles (ROV) and inside-out vesicles (IOV) were prepared according to the method of Steck et al. [28]; haemoglobindepleted cytosol (CYT) was obtained using the method of Perrella et al. [29]. The specimens were assayed immediately. Hex loosely bound to erythrocyte membranes was released by treatment with 1 M NaCl followed by centrifugation, according to Goi et al. [2]. Hexosaminidase assay Hex activity was determined fluorimetrically using 4MUGlcNAc and 4MU-GalNAc as substrates. In order to determine the activity only toward GlcNAc derivatives the assay was done with 4MU-GlcNAc as substrate in the presence of GalNAc (50 mmol/L) [19]. Under these conditions any residual activity after addition of GalNAc indicates the presence of a Hex specifically acting on GlcNAc-derivatives, that is an N-acetylβ-D-glucosaminidase activity. Hex activity involving the αsubunit was assessed using 4MU-GlcNAc-6S as substrate [30]. Hex assays in ghosts, CYT, ROV, IOV and the effluents from chromatofocusing or anion exchange columns were run according to Goi et al. [2]. The incubation mixture for determining
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
469
enzyme activities in ROV contained 150 mmol/L sodium chloride. The enzyme activity in ghosts, CYT, ROV and IOV was expressed as μU/mg of proteins. Plasma Hex was assayed as reported by Lombardo et al. [3].
Results
Separation of hexosaminidase isoenzymes
The chromatofocusing elution profiles of Hex isoenzymes in the erythrocyte membrane and cytosol are presented in Fig. 1. For comparison, the figure also reports the profile for the plasma of normal subjects (panel C) which shows the abbreviation of the well-known Hex isoforms and, between them, the indication of the P isoform, present in conspicuous amounts only in pregnancy; the S form was found in plasma from patients with Sandhoff disease (data not shown). In all cases the enzyme activity was measured using 4MU-GlcNAc, in the presence or absence of GalNAc, and 4MU-GlcNAc-6S, as substrates. In erythrocyte membranes, the Hex isoenzyme pattern was similar to that of the plasma enzyme (Figs. 1A and C). Based on our experience [14] and the literature [12], two isoenzyme groups can be identified, a first comprising four isoenzymes (group I), with pI from 5.9 to 5.2, and a second (group II) also of four isoenzymes with pI from 4.7 to 4.1. The isoenzymes in group I had no activity when 4MU-GlcNAc-6S was used as substrate, indicating the absence of the α-subunit. However, group II isoenzymes, with the sole exception of the form elutable at pI 4.7 (x-form), showed activity with the selective assay for the α-subunit. All isoenzymes in both groups showed activity on 4MUGalNAc although less than on 4MU-GlcNAc. An enzyme firmly anchored to the PBE resin was elutable only with 1.0 mol/L NaCl (y-form), like the Hex-S present in this fraction in Sandhoff patients. This enzyme, in spite of its elution in the same range as the Hex-S form, was present in small amounts and was not active toward 4MU-GlcNAc-6S or inhibited by GalNAc, indicating it had features of a specific GlcNAcase. It was completely inactive on 4MU-GalNAc. The assays of all isoforms, including the x- and y-forms, were done at their optimal pH values. In the erythrocyte cytosol (Fig. 1B) the Hex isoform pattern was similar to that of the membrane enzyme, the only difference being the great abundance of the y-form.
Automated chromatofocusing on PBE-94 The method of Goi et al. [14] was used. Briefly: 1.5 mL of ghost preparation (4 mg as proteins), or 2.5 mL of the supernatant after ghost treatment with 1 M NaCl (4 mg as proteins), or 0.2 mL of plasma, or 5 mL of CYT were dialyzed for 2 h in 25 mmol/L imidazole–HCl buffer, pH 7.4. Then the specimens were injected into a PBE-94 column (3.0 × 210 mm), previously equilibrated and washed with the same buffer for 5 min. At this point, 7.5 mmol/L per pH unit of polybuffer 74– HCl, pH 4.0, was pumped through the column so as to produce a linear pH elution gradient for 45 min, after which the column was washed for 10 min with 1.0 mol/L aqueous NaCl solution. All the solutions contained Triton X-100, 0.3% final concentration. The flow rate was 0.84 mL/min. The effluent was collected in 0.5-mL aliquots and immediately utilized for the enzyme assay. FPLC system with anion exchange column (using ÄKTApurifier from Amersham Pharmacia Biotech AB Uppsala, Sweden) Briefly: 0.3 mL of plasma or 5.0 mL of CYT were dialyzed for 2 h in 20 mmol/L Tris–HCl buffer pH 7.5. Then the specimens were injected into a mono Q HR 5/5 column, previously equilibrated and washed with the same buffer for 5 min. At this point 1.0 mol/L NaCl dissolved in the same buffer was pumped through the column so as to produce a linear gradient from 0 to 0.5 mol/L in 30 min. Afterwards the column was washed for 10 min with 1.0 mol/L NaCl solution in Tris buffer. The flow rate was 1.0 mL/min. The effluent was collected in 0.5-mL aliquots and immediately utilized for the enzyme assay. Sialidase treatment Erythrocyte ghosts, cytosol (4 mg protein in each case) and plasma (20 mg protein) were buffered at pH 7.4 with 25 mmol/L imidazole–HCl and treated with 300 mU of C. perfringens sialidase dissolved in the same buffer. The mixture was incubated at 37 °C for the time needed for complete release of sialic acid (10–30 min) and then chromatofocused on PBE 94. Sialic acid released was determined according to Caimi et al. [31]. Other methods The protein content in all specimens was determined by the method of Lowry et al. [32], using BSA as the reference standard. NADH–cytochrome c oxidoreductase (NADH–CytC Ox/Red) (EC 1.6.99.3) activity was determined by the method of McIntosh et al. [33], and acetylcholinesterase (AchEsterase) (EC 3.1.1.7) activity by that of Ellman et al. [34].
Chromatofocusing separation of hexosaminidase isoenzymes in erythrocyte ghosts and cytosol from normal subjects
Characteristics and properties of major hexosaminidase isoenzymes in erythrocyte membranes and cytosol The chromatofocusing fractions corresponding to each of the different isoenzymes were pooled and investigated for optimal pH and kinetic characteristics, using 4MU-GlcNAc as substrate, with or without GalNAc. In the membrane, group I and II isoenzymes had the same optimum pH of 4.4, with Km of respectively 0.40 and 1.03 mmol/L, and Vmax of 38 and 25 μU/mg protein. The y-form had optimum pH in the range of 5.0 to 6.0, an apparent Km of 0.80 mmol/L and Vmax of 5 μU/mg protein. Estimated total activities of Hex, groups I and II isoenzymes, and of the yform (GlcNAcase) were respectively 20.4 and 2.02 μU/109 erythrocytes. In the cytosol, isoenzyme groups I and II had the same optimum pH of 5.2, Km of 0.44 and 0.98, Vmax of 60 and 48 μU/
470
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
Fig. 1. Chromatofocusing elution profiles of erythrocyte membrane and cytosol hexosaminidase isoenzymes from normal human subjects. For comparison, the hexosaminidase isoenzyme pattern of human plasma is also presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MUGlcNAc-6S as substrates.
mg protein. The y-form had optimum pH in the range of 5.6 to 6.6, Km of 0.40 mmol/L and an apparent Vmax of 45 μU/mg protein. Estimated total activities of Hex groups I and II isoenzymes, and of the y-form (GlcNAcase) were respectively 150 and 52 μU/109 erythrocytes. Both isoenzyme groups, in either cytosol or membranes, were stable for 1 week at +4 °C and for 40 days at −40 °C and did not lose their activity with freezing and thawing. They were
completely inhibited by GalNAc at 50 mmol/L. However, the group I isoenzymes were resistant to 3 h at 52 °C, whereas group II, with the exception of the x-form, completely lost their activity in these conditions. The y-form, which was not inhibited by GalNAc, was completely denatured after 30 min at 52 °C. Its activity decayed rapidly after freezing at −20 °C and thawing. As shown in Fig. 2, the chromatofocusing patterns of erythrocyte membrane and cytosol Hex did not change at all
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
471
Fig. 2. Chromatofocusing elution profiles of erythrocyte membrane and cytosol hexosaminidase isoenzymes from normal human subjects before and after sialidase treatment. For comparison, the hexosaminidase isoenzyme pattern of human plasma, also treated with sialidase, is presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc.
after sialidase treatment, indicating either complete resistance to sialidase of the putative sialic acid residues or complete absence of sialylation. In contrast, all the plasma Hex isoforms, known
to contain sialic acid [12], shifted markedly toward less acidic pH, indicating the loss of negative charges consistent with the removal of sialic acid residues.
472
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
Anion exchange chromatographic separation of hexosaminidase isoenzymes in erythrocyte cytosol from controls To further characterize the y-form, abundant in the erythrocyte cytosol, cytosolic Hex isoforms were separated by the FPLC system with anion exchange column (Fig. 3). For comparison, the isoform profile from plasma is also given. Hex activity was assayed with 4MU-GlcNAc (with or without GalNAc), 4MUGalNAc, or 4MU-GlcNAc-6S. A first group of three isoenzymes was eluted between 50 and 200 mM NaCl. None had any activity on 4MU-GlcNAc-6S and all were totally inhibited by GalNAc, resembling the group I isoenzymes presented in Fig. 1, panel B. A second group (group II) of three isoenzymes was eluted between 280 and 400 mM NaCl. These were active on 4MUGlcNAc-6S and were completely inhibited by GalNAc, again resembling the group II isoenzymes shown in Fig. 1, panel B. Finally, one enzyme was eluted between 430 and 520 mM NaCl. It was not inhibited by GalNAc, and was not active on 4MUGlcNAc-6S and 4MU-GalNAc, similarly to the y-form in Fig. 1, panel B, thus having the features of GlcNAcase. The group I and II isoforms hydrolyzed 4MU-GalNAc although to a lesser extent (about ten times less) than on 4MU-
GlcNAc. The isoenzyme profile in plasma featured two groups of isoforms, similarly to the groups I and II from erythrocyte cytosol, although they eluted at somewhat different NaCl concentrations. These two groups appeared to correspond to the known Hex-B and Hex-A groups of isoenzymes [12]. No y-form was present in plasma, confirming the chromatofocusing separation findings. Chromatofocusing separation of hexosaminidase isoenzymes in erythrocyte ghosts and cytosol in a case of GM2–gangliosidosis (Tay–Sachs variant) The isoenzyme patterns of erythrocytes (ghost and cytosol) and plasma Hex from a patient with Tay–Sachs disease is presented in Fig. 4. In the plasma there were no isoenzymes of the Hex-A group, but I2–P was present, with those of the Hex-B group that were markedly elevated but completely inhibited by GalNAc. The isoenzyme profiles in the erythrocyte membranes and cytosol, with the exception of the GlcNAcase (y-form), that is present in the erythrocyte samples, essentially overlapped those of plasma. As in plasma, group II isoenzymes were absent while
Fig. 3. Anion exchange elution profiles of erythrocyte cytosol hexosaminidase isoenzymes from human normal subjects. For comparison, the hexosaminidase isoenzyme pattern of human plasma is also presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MU-GlcNAc-6S as substrates.
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
473
Fig. 4. Chromatofocusing elution profiles of erythrocyte membrane and cytosol hexosaminidase isoenzymes of a Tay–Sachs patient. For comparison, the hexosaminidase isoenzyme pattern in the plasma of the same patient is presented. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MU-GlcNAc-6S as substrates.
group I was markedly increased. The x-form was present as was GlcNAcase (y-form), which is insensitive to GalNAc and not active on 4MU-GalNAc, in either erythrocyte membranes or cytosol. No activity was seen on 4MU-GlcNAc-6S in either erythrocytes or plasma, as expected. The membrane-linked y-form was much more abundant in Tay–Sachs than normal erythrocytes.
Chromatofocusing separation of hexosaminidase isoenzymes of erythrocyte ghosts released by treatment with 1 M NaCl Confirming previous evidence [2], treatment of erythrocyte ghosts with 1 M NaCl caused a substantial release of Hex activity. The chromatofocusing isoenzyme profile (Fig. 5) shows an interesting difference compared to that of untreated
474
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
Fig. 5. Chromatofocusing separation of hexosaminidase isoenzymes of erythrocyte ghosts from normal human subjects released by treatment with 1 M NaCl. The assays were run at pH 4.4, utilizing 4MU-GlcNAc (with or without GalNAc), and 4MU-GlcNAc-6S as substrates.
ghosts. The group I isoforms and the x-form remained essentially the same, indicating that these isoform were partly firmly attached to the membrane, and partly (the released ones) only loosely linked, probably with a similar molecular architecture. Instead, all the subunits acting on 4MU-GlcNAc6S (α-subunits) were no longer part of the group II isoforms and were elutable from the column with 1 M NaCl, a response similar to that of the y-form. Hexosaminidase orientation in human erythrocyte membranes The orientation of the major Hex activity in erythrocyte membranes (optimum pH 4.4) was assessed using ROVand IOV vesicles prepared from erythrocyte membranes (Table 1). This activity was seen predominantly in ROV vesicles (83%), with only 15% in IOV vesicles. The pattern of Hex activity was essentially identical to that of AchEsterase, a marker of the membrane external surface, and quite different from NADH– Cyt-C Ox/Red, a marker of the internal surface, indicating that the orientation of Hex is mainly to the external surface of the plasma membrane. It was not possible to establish the orientation
of the membrane-bound GlcNAcase because of its extremely low activity in normal erythrocytes. Discussion There is evidence that glycohydrolases, or at least some of them, occur not only in the lysosomes but also in the plasma membrane and cytosol of mammalian cells [2,6,35–38]. The lysosomal, plasma membrane and cytosol sialidases have different primary structures and are coded by different genes [39]. The lysosomal hexosaminidase, which is the better known,
Table 1 Presence of hexosaminidase at pH 4.4 in oriented vesicles from human erythrocytes membrane Enzyme source
Hexosaminidase AchEsterase NADH–Cyt-C Ox/ (μU/mg of protein) (U/mg protein) Red (mU/mg protein)
Unsealed ghosts 31.1 ± 15.3 ROV 26.2 ± 12.5 IOV 4.7 ± 2.3
2.55 ± 0.43 2.04 ± 0.35 0.31 ± 0.11
The data are the mean values ± S.D. of six experiments.
23.2 ± 9.6 2.1 ± 0.6 22.5 ± 8.5
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
exists as two major isoforms, Hex-A and Hex-B, resulting from the combination of the two polypeptides (α- and β-subunits) coded by two different genes, HEXA and HEXB. The plasma membrane-linked enzyme's structural features are still little known but recently Mencarelli et al. [6] described and characterized a form of Hex-A from plasma membranes of human fibroblasts, active on GM2 ganglioside, and speculated about its lysosomal origin. The cytosolic enzyme so far described appears specifically to affect only GlcNAc derivatives (it is actually an N-acetylglucosaminidase) and is coded by a MGEA5 gene [20,23,40]; on account of its intrinsic histone acetyltransferase activity [41], it is nowadays known as “nuclear cytoplasmic O-GlcNAcase and acetyltransferase” (NCOAT). The present investigation studied Hex isoenzymes in the plasma membrane and cytosol of human erythrocytes, which have no lysosomes, employing isoform separation techniques already applied to lysosomal Hex-A and Hex-B [2,14], and commonly used enzyme assays [19], with the substrate 4MUGlcNAc with or without GalNAc (as competitive inhibitor), to assess the specificity toward GlcNAc-derivatives, or 4MUGalNAc, or 4MU-GlcNAc-6S as substrate to recognize the activity of the α-subunit in Hex-A isoenzyme. Further information was obtained by comparing the erythrocyte Hex isoenzymes with those of blood plasma from control subjects and with the erythrocytes and plasma from a Tay–Sachs patient that entirely lack Hex-A. A first finding was that Hex are found in human erythrocyte plasma membranes, confirming previous observations [2,6], and cytosol – this was a novel finding; we also report that the plasma membrane enzymes are mostly oriented toward the extracellular environment, as recently demonstrated for another membrane glycohydrolase, acidic sialidase [42]. In both the membrane and cytosol, the chromatofocusing profiles of Hex isoenzymes were similar, with two main isoenzyme groups, I and II, with isoelectric points similar to those of plasma Hex-B and Hex-A groups. The membrane and cytosolic group I isoenzymes had similar kinetics to the plasma Hex-B group, were active toward both GlcNAc and GalNAc derivatives, sensitive to inhibition by GalNAc and had no activity on 4MUGlcNAc-6S, indicating the absence of the α-subunit. As expected, in the erythrocytes from the Tay–Sachs patient (where only the β-subunits are present) conspicuous amounts of the group I isoenzymes were found in both the membrane and cytosol. The membrane and cytosolic group II isoenzymes showed substantial similarity to the plasma Hex-A group, sharing the same kinetics and being inhibited by GalNAc and active toward 4MU-GlcNAc-6S. This indicates the presence of the α-subunit. With the exception of the x-form, the plasma membrane and cytosolic group II isoenzymes were completely absent in the Tay–Sachs erythrocytes. Interestingly, the x-form, which is not active toward 4MU-GlcNAc-6S, thus excluding the presence of α-subunits, was the only group II isoenzyme in the Tay–Sachs patient, possibly corresponding to the I2 isoform [12]. A very intriguing point, surely worth further investigation, was that when erythrocyte ghosts were treated with 1 M NaCl the
475
released (i.e. the loosely bound) Hex-B isoforms had the same chromatofocusing profile as the unreleased isoform – the tightly bound one, whereas Hex-A isoforms released behaved quite differently, consistent with the presence of separated isoforms. These results raise a question: how does each isoenzyme reach the plasma membrane or the cytosol? For membranebound Hex, attached to the external leaflet, we can exclude aspecific adsorption of the plasma circulating isoenzymes. In fact, whereas the plasma isoenzymes are known to be affected by sialidase [12] – as we also observed – the membrane-bound ones are not affected at all, nor are the cytosolic isoforms, under conditions where the release of sialic acid (i.e. the activity of sialidase) was measured (data not shown). We can also exclude that they are linked to the mannose-6-phosphate receptor on the erythrocyte surface since even increasing concentrations of mannose-6-phosphate, as competitive ligand, did not succeed in releasing the enzyme from the erythrocyte membranes (data not shown). For cytosolic Hex, we can exclude an origin from contaminating blood cells because the erythrocyte suspension was completely free of any leukocytes or platelets before erythrocyte lysis. Moreover, the optimal pH of cytosolic Hex is less acidic than that of the lysosomal Hex. We can only suggest that the membrane or cytosolic localization of these isoenzymes is due to post-translational modifications that are somehow different from those that lead the enzymes to be released from the cells or targeted to the lysosomal compartment. This study also turned up another finding, concerning the Hex y-form, which appears to be a specific GlcNAcase. It was abundant in the erythrocyte cytosol of normal subjects and of the patient with GM2–gangliosidosis. On the basis of some properties, such as the pH curve with a characteristic plateau, the thermal instability, insensitivity to any inhibitory effect by GalNAc, and the lack of activity on 4MU-GalNAc, we suggest that the cytosolic erythrocyte GlcNAcase might be the OGlcNAcase purified and characterized in other tissues, cloned and identified as isoform C [19–22]. This enzyme's role in the dynamic process of O-GlcNAc–glycoprotein modification has been suggested [20,41], as well as in the rearrangement of the glycidic moieties of proteins required for connections between erythrocyte membrane and cytoskeleton. Band 4.1, a protein that serves as a bridge joining the cytoskeleton to the inner surface of the plasma membrane, also contains O-GlcNAc moieties [43]. An additional interesting finding was the presence of GlcNAcase activity in the erythrocyte ghosts from normal subjects, and the marked increase of this enzyme form in the ghosts of the Tay–Sachs case. This needs to be investigated in more detail, in order to clarify whether the “canonical” OGlcNAcase in the erythrocyte cytosol is structurally connected with the plasma membrane-bound GlcNAcase, and why the ghost-linked GlcNAcase was so abundant in Tay–Sachs erythrocytes. To conclude, this study showed that both the plasma membrane and cytosol of human erythrocytes have Hex isoenzymes quite likely originating from translational modifications of the products of HEXA and HEXB genes, coding for
476
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
lysosomal Hex. This finding makes an important step forward in the knowledge of the biology of the lysosomal–endosomal compartment in relation to cell plasma membrane generation and/or turnover and is probably related to the recent observation that some cells, mostly from the haematopoietic lineage, contain secretory lysosomes, call exosomes in erythrocytes [44,45]. Generation of exosomes is a critical mechanism for shedding membrane proteins in the maturing red cell [46] and it is likely that during erythrocyte maturation, with a mechanism still to be clarified, non-integral proteins of the late lysosomal–endosomal compartment are also incorporated in the membrane. Some exosomes can “explode” into reticulocytes [47,48], thus explaining their presence in cytosol too. Additional important evidence concerns the presence of the O-GlcNAcase in the erythrocyte cytosol as well as in the membrane, the latter form being particularly abundant in the Tay–Sachs patient. It will be interesting to study the relationships between the cytosolic and membrane-bound forms of this enzyme. The data also offer a starting point for assessing the possible clinical involvement of these erythrocyte glycohydrolases. Oxidative stress is a feature of the development of age-related chronic degenerative diseases, such as diabetes, atherosclerosis, hypertension, Parkinson's and Alzheimer's [49]. In these pathologies, both the generation rates of oxidants and the susceptibility of tissues to oxidative damage influence the integrity of cell membranes of all tissues [50] and the erythrocyte plasma membrane, that is easy to obtain with high purity and could serve as a useful model for following the functional condition of membranes in these age-related degenerative diseases. Acknowledgments This work was supported in part by the Italian Ministry for Education, University and Research (COFIN grant 2004 to Guido Tettamanti). We thank Dr. Alberto Burlina, Department of Pediatrics, University of Padua, Padua, Italy, for kindly supplying the blood specimen from a Tay–Sachs disease patient. References [1] Goi G, Fabi A, Lombardo A, et al. Stability of enzymes of lysosomal origin in human cerebrospinal fluid. Clin Chim Acta 1987;163:215–24. [2] Goi G, Bairati C, Massaccesi L, et al. Membrane anchoring and surface distribution of glycohydrolases of human erythrocyte membranes. FEBS Lett 2000;473:89–94. [3] Lombardo A, Caimi L, Marchesini S, Goi GC, Tettamanti G. Enzymes of lysosomal origin in human plasma and serum: assay conditions and parameters influencing the assay. Clin Chim Acta 1980;108:337–46. [4] Hutchinson T, Dwivedi K, Rastogi A, Prasad R, Pereira BM. N-Acetyl beta-D-glucosaminidase is not attached to human sperm membranes through the glycosylphosphatidyl inositol (GPI)-anchor. Asian J Androl 2002;4:27–33. [5] Izumi T, Suzuki K. Neutral hydrolases of rat brain. Preliminary characterization and developmental changes of neutral beta-N-acetylhexosamindases. Biochim Biophys Acta 1980;615:402–13. [6] Mencarelli S, Cavalieri C, Magini A, et al. Identification of plasma membrane associated mature beta-hexosaminidase A, active towards GM2 ganglioside, in human fibroblasts. FEBS Lett 2005;579:5501–6.
[7] Kustermann-Kuhn B, Harzer K. Prenatal diagnosis of Tay–Sachs disease. Reflectometry of hexosaminidase A, B, and C/S bands on zymograms. Hum Genet 1983;65:172–5. [8] Neufeld EF. Natural history and inherited disorders of a lysosomal enzyme, beta-hexosaminidase. J Biol Chem 1989;264:10927–30. [9] Burlina AB, Goi G, Fabi A, et al. Behaviour of some lysosomal enzymes in the plasma of insulin dependent diabetic patients during artificial pancreas treatment. Clin Biochem 1987;20:423–7. [10] Goi G, Bairati C, Segalini G, et al. Alterations in the activity of several glycohydrolases in red blood cell membrane from type 2 diabetes mellitus patients. Metabolism 1999;48:817–21. [11] Reglero A, Carretero MI, Cabezas JA. Increased serum alpha-L-fucosidase and beta-N-acetylglucosaminidase activities in diabetic, cirrhotic and gastric cancer patients. Clin Chim Acta 1980;103:155–8. [12] Mahuran D, Novak A, Lowden JA. The lysosomal hexosaminidase isozymes. Isozymes Curr Top Biol Med Res 1985;12:229–88. [13] Grave RA, Kaback MM, Proia RL, Sandhoff K, Suzuki K, Suzuki K. The GM2 gangliosidosis 8th ed, 2001. [14] Goi G, Fabi A, Lombardo A, et al. The lysosomal beta-D-N-acetylglucosaminidase isozymes in human plasma during pregnancy: separation and quantification by a simple automated procedure. Clin Chim Acta 1989; 179:327–40. [15] Mahuran DJ. Characterization of human placental beta-hexosaminidase I2. Proteolytic processing intermediates of hexosaminidase A. J Biol Chem 1990;265:6794–9. [16] Hakomori S. Glycosylation defining cancer malignancy: new wine in an old bottle. Proc Natl Acad Sci U S A 2002;99:10231–3. [17] Goi G, Cazzola R, Tringali C, et al. Erythrocyte membrane alterations during ageing affect beta-D-glucuronidase and neutral sialidase in elderly healthy subjects. Exp Gerontol 2005;40:219–25. [18] Massaccesi L, Corsi MM, Baquero-Herrera CJ, et al. Erythrocyte glycohydrolases in subjects with trisomy 21: could Down's syndrome be a model of accelerated ageing? Mech Ageing Dev 2006;127:324–31. [19] Dong DL, Hart GW. Purification and characterization of an O-GlcNAc selective N-acetyl-beta-D-glucosaminidase from rat spleen cytosol. J Biol Chem 1994;269:19321–30. [20] Gao Y, Wells L, Comer FI, Parker GJ, Hart GW. Dynamic O-glycosylation of nuclear and cytosolic proteins: cloning and characterization of a neutral, cytosolic beta-N-acetylglucosaminidase from human brain. J Biol Chem 2001;276:9838–45. [21] Izumi T, Suzuki K. Neutral beta-N-acetylhexosaminidases of rat brain. Purification and enzymatic and immunological characterization. J Biol Chem 1983;258:6991–9. [22] Ueno R, Yuan CS. Purification and properties of neutral beta-Nacetylglucosaminidase from carp blood. Biochim Biophys Acta 1991; 1074:79–84. [23] Heckel D, Comtesse N, Brass N, et al. Novel immunogenic antigen homologous to hyaluronidase in meningioma. Hum Mol Genet 1998; 7:1859–72. [24] Comer FI, Hart GW. O-Glycosylation of nuclear and cytosolic proteins. Dynamic interplay between O-GlcNAc and O-phosphate. J Biol Chem 2000;275:29179–82. [25] Cooksey RC, Hebert Jr LF, Zhu JH, et al. Mechanism of hexosamineinduced insulin resistance in transgenic mice overexpressing glutamine : fructose-6-phosphate amidotransferase: decreased glucose transporter GLUT4 translocation and reversal by treatment with thiazolidinedione. Endocrinology 1999;140:1151–7. [26] Slawson C, Pidala J, Potter R. Increased N-acetyl-beta-glucosaminidase activity in primary breast carcinomas corresponds to a decrease in Nacetylglucosamine containing proteins. Biochim Biophys Acta 2001;1537: 147–157. [27] Yao PJ, Coleman PD. Reduction of O-linked N-acetylglucosaminemodified assembly protein-3 in Alzheimer's disease. J Neurosci 1998; 18:2399–411. [28] Steck TL, Kant JA. Preparation of impermeable ghosts and inside-out vesicles from human erythrocyte membranes. Methods Enzymol 1974; 31:172–80. [29] Perrella M, Benazzi L, Ripamonti M, Rossi-Bernardi L. Bohr effect in
L. Massaccesi et al. / Clinical Biochemistry 40 (2007) 467–477
[30]
[31]
[32] [33]
[34]
[35]
[36] [37]
[38] [39]
[40]
hemoglobin deoxy/cyanomet intermediates. Biochemistry 1994;33: 10358–10366. Fuchs W, Navon R, Kaback MM, Kresse H. Tay–Sachs disease: one-step assay of beta-N-acetylhexosaminidase in serum with a sulphated chromogenic substrate. Clin Chim Acta 1983;133:253–61. Caimi L, Lombardo A, Preti A, Wiesmann U, Tettamanti G. Optimal conditions for the assay of fibroblast neuraminidase with different natural substrates. Biochim Biophys Acta 1979;571:137–46. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265–75. McIntosh CH, Plummer DT. The subcellular localization of acetylcholinesterase and its molecular forms in pig cerebral cortex. J Neurochem 1976;27:449–57. Ellman GL, Courtney KD, Andres Jr V, Feather-Stone RM. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 1961;7:88–95. Haeuw JF, Strecker G, Wieruszeski JM, Montreuil J, Michalski JC. Substrate specificity of rat liver cytosolic alpha-D-mannosidase. Novel degradative pathway for oligomannosidic type glycans. Eur J Biochem 1991;202:1257–68. Haeuw JF, Michalski JC, Strecker G, Spik G, Montreuil J. Cytosolic glycosidases: do they exist? Glycobiology 1991;1:487–92. Comtesse N, Maldener E, Meese E. Identification of a nuclear variant of MGEA5, a cytoplasmic hyaluronidase and a beta-N-acetylglucosaminidase. Biochem Biophys Res Commun 2001;283:634–40. Berrin JG, Czjzek M, Kroon PA, et al. Substrate (aglycone) specificity of human cytosolic beta-glucosidase. Biochem J 2003;373:41–8. Monti E, Preti A, Venerando B, Borsani G. Recent development in mammalian sialidase molecular biology. Neurochem Res 2002;27: 649–663. Lehman DM, Fu DJ, Freeman AB, et al. A single nucleotide polymorphism in MGEA5 encoding O-GlcNAc-selective N-acetyl-beta-D
[41]
[42]
[43]
[44]
[45] [46]
[47] [48]
[49]
[50]
477
glucosaminidase is associated with type 2 diabetes in Mexican Americans. Diabetes 2005;54:1214–21. Toleman C, Paterson AJ, Whisenhunt TR, Kudlow JE. Characterization of the histone acetyltransferase (HAT) domain of a bifunctional protein with activable O-GlcNAcase and HAT activities. J Biol Chem 2004;279: 53665–73. Chiarini A, Fiorilli A, Di Francesco L, Venerando B, Tettamanti G. Human erythrocyte sialidase is linked to the plasma membrane by a glycosylphosphatidylinositol anchor and partly located on the outer surface. Glycoconjugate J 1993;10:64–71. Holt GD, Haltiwanger RS, Torres CR, Hart GW. Erythrocytes contain cytoplasmic glycoproteins. O-Linked GlcNAc on Band 4.1. J Biol Chem 1987;262:14847–50. Pan BT, Teng K, Wu C, Adam M, Johnstone RM. Electron microscopic evidence for externalization of the transferrin receptor in vesicular form in sheep reticulocytes. J Cell Biol 1985;101:942–8. Blott EJ, Griffiths GM. Secretory lysosomes. Nat Rev, Mol Cell Biol 2002;3:122–31. Fader CM, Savina A, Sanchez D, Colombo MI. Exosome secretion and red cell maturation: exploring molecular components involved in the docking and fusion of multivesicular bodies in K562 cells. Blood Cells Mol Dis 2005;35:153–7. Koury MJ, Koury ST, Kopsombut P, Bondurant MC. In vitro maturation of nascent reticulocytes to erythrocytes. Blood 2005;105:2168–74. Artal-Sanz M, Samara C, Syntichaki P, Tavernarakis N. Lysosomal biogenesis and function is critical for necrotic cell death in Caenorhabditis elegans. J Cell Biol 2006;173:231–9. Ames BN, Shigenaga MK, Hagen TM. Oxidants, antioxidants, and the degenerative diseases of aging. Proc Natl Acad Sci U S A 1993; 90:7915–22. Droge W. Oxidative stress and aging. Adv Exp Med Biol 2003;543: 191–200.