Initial characterization of human thymocyte sialidase activity: Evidence that this enzymatic system is not altered during the course of T-cell maturation

Inr. J. Biochem. Vol.26, No. 6, pp. 169-716.1994

Pergamon 0020-71 lX(93)EOO51-P

Copyright0 1994Elsevier Science Ltd Printedin Great Britain.All rightsreserved 0020-71 IX/94$7.00+ 0.00

INITIAL CHARACTERIZATION OF HUMAN THYMOCYTE SIALIDASE ACTIVITY: EVIDENCE THAT THIS ENZYMATIC SYSTEM IS NOT ALTERED DURING THE COURSE OF T-CELL MATURATION A. GREFFARD,J. C. PAIRON,H. TERZIDIS-TRABELSI, J.-M. HESLAN,J. BIGNON,C. R. LAMBRE and Y. PILATTE* Inserm U 139, Hapital Henri Mondor, 94010 Creteil, France [Fax 33 (1) 49-81-35-331 (Received 29 October 1993)

The sialidase activity of human thymocyte was examined by a fluorogenic assay. 2. These studies revealed that human thymocyte sialidase activity is essentially acid-active and membrane-bound since 59.6% and 33% of the total activity was recovered in the lysosome-enriched and microsomal fractions, respectively. 3. A weak activity was also detected in the cytosolic fraction. 4. However, the acidic optimum pH of this soluble sialidase was at variance with the general concept of mammalian soluble sialidases which are known to be optimally active at more neutral pH. 5. This acidic soluble sialidase seems to be a general characteristic of the human T-cell lineage since examination of mature circulating T-cells revealed that they contain a soluble sialidase activity similar to that observed in thymocytes. 6. Analysis of mature and immature thymocyte subpopulation obtained by differential PNA agglutination indicated that this enzymatic system was not altered during the course of thymic maturation. 7. These results suggest that unlike in T-cell activation where changes in the level of sialidase activity were shown to influence the extent of cell surface sialylation and thereby the cell physiology, this enzymatic system seems not to be involved in the fluctuation of cell surface sialic acid content observed during thymic maturation. Abstract-l.

Sialidase

Sialic acid

PNA

Human thymocyte

INTRODUCTION Sialic acids are important constituents of glycoproteins and gangliosides. Due to their “terminal” position on numerous oligosaccharidic chains and their electronegative charge, these amino sugars influence the function of many glycoconjugates and thereby, play an important role in the control of many biological systems (Schauer, 1982; Pilatte et al., 1993). For instance, several lines of evidence suggest that cell surface sialic acid residues (Boog et al., 1989; Cowing et al., 1983; Frohman er al., 1985; Hirayama et al., 1988; Kearse et al., 1988; Powell et al., 1987) constitute a key element *To whom correspondence should be addressed.

Thymic maturation

in the control of immunocompetent cell interactions. As a result, the mechanisms regulating the sialic acid amount of cell glycolipids and glycoproteins must be strictly controlled. In the first place, this control is under the dependence of sialyl-transferases (Schauer, 1982), but several lines of evidence suggest that cell sialidases (EC 3.2.1.18) which catalyze the hydrolytic removal of sialic acids from sialocompounds may also be involved. Indeed, it has been reported that expression of hyposialylated MHC Class I glycoproteins on the cell surface of Con A activated murine T lymphocytes (Landolfi and Cook, 1986), was closely related to the increased level of sialidase activity observed in activated cells (Landolfi et al., 1985). This observation is of importance since 169

770

A.

GREFFARD

lymphocyte functions are critically dependent on cell surface recognition events in which sialic acids may be involved. Along this line, it has been shown recently that culture of murine T-cells in mixed lymphocyte reaction induces a significant increase in sialidase activity associated with sialic acid loss from the cell surface which renders these cells responsive to allogenie B-cells (Taira and Nariuchi, 1988). In contrast, T-cells from mice genetically deficient in sialidase responded poorly when primed in the same conditions although they did respond when treated with exogenous sialidase (Taira and Nariuchi, 1988). Likewise, a sialidase expressed on T-cell surface was recently shown to be involved in the conversion of the human vitamin D,-binding protein (Gc) into a macrophage activating factor (Yamamoto and Kumashiro, 1993). Therefore, considering the numerous biological functions of sialic acids it can be hypothesized that in participating in the control of cellular sialylation, sialidase activity modulate T-cells physiology and regulate intercellular signalling required for immune system to function properly. It is now well established that mammalian cells contain various sialidase activities which differ in their biochemical properties, relative amount and subcellular distribution (Corfield et al., 1981; Schauer, 1982). Despite its potential functional importance, little is known on mammalian T-cell sialidase content and there is no available information on the human T lineage. In this paper we have characterized human thymocyte sialidase activities using 2’-(4methylumbelliferyl ) - tl - D - N- acetylneuraminic acid as substrate. We also attempted to determine how thymocyte sialidase enzymatic system was affected along the maturational process leading to circulating T-cells. Our results show that thymocyte sialidase activity is essentially membrane-bound and acid-active. Furthermore, we could not demonstrate any change neither in the level of activity nor in the subcellular distribution along maturation suggesting that sialidases are not involved in the changes in cell surface sialylation occurring during thymocytes maturation. MATERIALS

AND METHODS

Reagents

2’-(4-methylumbelliferyl)-a-D-N-acetylneuraminic acid (Mu-NeuAc), 4-methylumbelliferone and PNA were obtained from Sigma

et

al.

Chemical Co. (U.S.A.). Other chemical reagents were of analytical grade. Magnetic beads coated with sheep anti-mouse IgG (Dynabeads M-450) were purchased from Dynal (Norway), mouse anti-human B-cell (CD19) and mouse antihuman C3bi receptor (CDll) from Dakopatts (Denmark), FITC goat anti-mouse IgG from Ortho Diagnostic System (U.S.A.) and anti-Pan T (Leu 4) from Coulter (U.S.A.). Preparation

of cells

* Thymocytes. Normal thymus were obtained from children undergoing cardiac surgery. The tissue was minced in phosphate buffer saline (PBS) and pressed through a fine stainless steel mesh (125 pm). Cells were collected, extensively washed with PBS and centrifuged at 200g. The cell pellet was stored at -80°C. *Fractionation qf thymocytes by agglutination with peanut agglutinin PNA. PNA agglutinable

thymocytes were isolated according to Reisner’s method (1984). Briefly, a suspension of freshly purified thymocytes was adjusted to 4.10’ cells/ml and incubated with an equal volume of PNA (2 mg/ml PBS) for 20 min at room temperature. Then, the cell suspension was layered on top of a solution of heat-inactivated FCS in PBS (50% v/v). After 30 min, agglutinated thymocytes sediment whereas the unagglutinated cells remain on top. The two fractions were removed separately and transferred into 0.2 M D-Galactose solution (PBS) (l/5 v/v). to dissociate agglutinated cells. After I5 min, at room temperature, cells were washed, centrifuged and the pellet was stored at -80’C. *T lymphocytes. T lymphocytes were obtained from healthy volunteers. Cells were separated on ficoll-hypaque (Pharmacia, Sweden), washed with Hanks buffer saline and filtrated through glass wool. The cell suspension was then incubated with a monoclonal antibody mixture (mouse anti-human B-cell and antihuman C3bi receptor) for 1 hr at 4°C. Cells were centrifuged and washed. Then, the cells were incubated for 30min at 4°C with magnetic beads priorly coated with sheep anti-mouse IgG. Magnetic beads were collected by applying a magnet on the outer wall of the test tube for l-2 min. The B-cell depleted suspension was collected and the whole procedure was repeated once. The final T lymphocytes suspension was washed 3 times with PBS, centrifuged and the last pellet was stored at -80°C. The purity of T lymphocytes, was controlled by incubation of the cell suspension with

Human

anti-Pan IgG.

T (Leu 4) and

Subcellular

then

thymocyte

with FITC

sialidase

concentration of 0.1 M and 250 PM respectively. The reaction was stopped with 0.1 M NaOH. The fluorescence intensity of free methylumbelliferone was read at 450 nm after excitation at 365 nm on a Kontron SFM 25 fluorometer using pure methylumbelliferone as standard. All samples were assayed in duplicate. One enzymatic unit was defined as the amount of enzyme which hydrolyzes 1 nmole of substrate per hour at 37°C.

anti

fractionation

Cells were suspended at a concentration of 4.10’ cells/ml with 3 mM imidazole buffer (pH = 7.2) containing 250mM sucrose and 1 mM EDTA. The suspension was homogenized using a glass teflon homogenizer with 10 strokes. The crude homogenate was centrifuged at 1500 rpm for 15 min. The supernatant was collected and centrifuged at 17,400 g for 30 min. The pellet (“lysosome-enriched fraction”) was re-homogenized in the same buffer. The supernatant was centrifuged at 105,OOOg for 1 hr. The resulting pellet (“microsomal fraction”) was resuspended and the supernatant (“cytosolic fraction”) concentrated as described in the results section. This process was carried out at 4°C. The sialidase activity was assayed either in the same day or after storage at - 80°C. Sialidase

771

activity

Protein assay Protein concentration was measured according to Bradford’s method (1976) with bovine serum albumin as standard. RESULTS

Thymocytes As expected from the many reports published on mammalian sialidases, human thymocyte sialidase activity is predominantly membranebound and optimally active at acidic pH. Global examination of thymocyte sialidase content, using crude homogenate as a source of enzyme, revealed an enzymatic activity optimum at pH 4.0-4.2 (Fig. 1) with a sp. act. of 4.05 U/mg of protein. Subcellular fractionation experiments showed that thymocyte sialidase activity was essentially associated to cell membrane compartments,

assay

Sialidase activity was tested using a fluorogenie substrate (Potier et al., 1979): 2’-(4methylumbelliferyl ) - c(- D - N- acetylneuraminic acid (Mu-NeuAc). Briefly, each sample was incubated at 37°C for l-3 hr, in the presence of acetic acid/sodium acetate buffer (at various pH values) and substrate at the final

Lysosomal fraction

Homogenate 100 80 x

.Z .S

60 40

:

s

20

4 .z

S

z E 3

‘\

I.,.,.,.,.,.,.,.,.,.,...,.,.,.

1

~.,.,.,.,.

--::..

Cytosolic

Microsomal fraction

100

E

‘Z

d e

8o 60 :;;‘\:\ 40 20 S ,,,,,,,,,

:‘\I, 3.6

4.2

4.6

5.0

5.4

5.8

3.6

4.2

4.6

5.0

5.4

5.8

pH Values Fig. I. pH dependence

of the sialidase

activities associated with the various from human thymocytes.

subcellular

fractions

obtained

772

A. GREFFARD et af.

since 59.6 and 33% of the total cell sialidase activity was recovered, at pH 4.2, in the lysosome-enriched fraction and in the microsomal fraction, respectively. As shown in Fig.1, the pH activity curves of these two subcellular fractions exhibited acidic optimum and were similar to that obtained for the crude homogenate. The sialidase specific activities were 4.41 and 14.96 U/mg of protein for the lysosomal and microsomal fractions, respectively. We also looked for the possible presence of a sialidase activity in thymocyte cytosolic fraction. Soluble sialidases have been described in cytosolic fractions obtained from various mammalian organs, but due to their low level of activity they were rarely found in homogeneous isolated cell populations (Pilatte et al., 1987; Pilatte et al., submitted) probably because, in this case, their presence could not be detected unless the 105,OOOgsupernatant had been concentrated by using a centricon- 10 microconcentrator (Amicon, Paris, France) (Pilatte et al., 1987). This concentration method being timeconsuming, it can possibly induce some level of inactivation since mammalian sialidases are known to be highly unstable (Corfield et al., 1981). We therefore investigated an alternative concentration procedure: ammonium sulfate precipitation. Half of the cytosolic preparation was concentrated 15-fold by using the usual ultrafiltration method while the other half was brought to 70% saturation by adding the appropriate amount of liquid saturated ammonium sulfate solution. After 1 hr under stirring at 4”C, the precipitated proteins were pelleted by centrifugation and resolubilized in a volume of sucrose buffer sufficient to achieve a 15-fold concentration. Both preparations were then concurrently assayed for sialidase activity. This study revealed that the precipitation procedure had several advantages over ultrafiltration. It was faster and, more importantly, it yielded a 30% higher activity without significantly affecting the catalytic properties of the sialidase activity occurring in the cytosolic fracton. Interestingly, human thymocytes soluble sialidase activity exhibited a rather acidic optimum pH 4.0-4.2 (Fig. 1) while mammalian cytosolic sialidases are usually optimum at a more neutral pH (5.5-6.0). Its specific activity was very weak: 0.67 U/mg of protein and assuming that the whole soluble activity was recovered in the cytosolic fraction, it accounted for 7.3% of the total thymocyte sialidase activity at pH 4.2. The activity recovered in the three subcellular

fractions represented 46.4% of the activity initially found in the crude homogenate. The sialidase activities occurring in the various thymocyte su~ellular compartment were then further characterized. The apparent K, for Mu-NeuAc at optimum pH calculated by the Lineweaver-Burk reciprocal plot was 0.61’10-4 M for the lysosome-enriched fraction, 0.59.10-4 M for the microsomal fraction and 1.95.10-4 M for the cytosolic fraction. Divalent cations are known to differentially affect the enzymatic activity of certain mammalian sialidases (Corfield et al., 1981). Therefore, the sialidase activities associated with the different subcellular fractions were assayed in the presence or absence of various concentrations of several cations as their chloride form. In order to verify that the observed effects were not due to chloride which is known to alter certain mammalian sialidases (Corfield et al., 1981), we comparatively tested their acetate counterparts. A typical experiment is depicted in Fig. 2. We observed very little difference between the two sets of cations (chloride and acetate forms). In addition the sialidase activities of the different fractions were similarly affected by each cation. The more striking effect could be ascribed to Cuz’ which significantly inhibited the sialidase activity associated with the three subcellular fractions, while Ca”, Mg2+, and Zn*+ had only little effect. ~~yrn~~~}tes,~ra~tion~ti~nby PNA ~~~~uti~ut~~~

Human thymus contain at least two different lymphocyte subpopulations: a major which is rather immunosubpopulation incompetent, and a minor one displaying immunological functions similar to those of the circulating T lymphocytes. Since maturation and/or differentiation have been shown to be accompanied by profound alteration in the sialidase content of various cell types, we have investigated whether similar modifications occurred during the course of thymocytes maturation. Mature thymocytes were therefore separated from immature ones by differential agglutination with PNA (Reisner and Sharon, 1984). The PNA-unagglutinated cell fraction (PNA-) and PNA-agglutinated cell fraction (PNA’) corresponding to the mature and immature thymocytes subpopulations respectively, were then assayed for sialidase activity. As can be seen in Fig. 3, when the sialidase activity of the two thymocyte subpopulations was assayed as a

Human

thymocyte

sialidase

Microsomal

Lysosome-enriched fraction

713

activity

Cytosolic fraction

fraction

I

Acetate

I

Acetate

80 60 40 20

Chloride

Chloride

E

100 80 60 40

2

4

6

8

2

10

4

6

8

Concentration

10

2

4

6

8

10

(mM)

Fig. 2. Effect of divalent cations as their acetate and chloride forms, and EDTA on the level of sialidase activity present in thymocyte subcellular fractions (tested at their optimum pH). Results are expressed as a percentage of the initial activity.

whole in crude homogenates, the pH-activity curves were almost indistinguishable and no significant difference in the level of activity was found between the PNA- and PNA+ subpopulations (6.32 and 5.89 U/mg protein for the PNA+ and PNA- subpopulations, respectively). In addition, fractionation studies revealed that the subcellular distribution of this enzymatic activity in both thymocyte subpopulations was similar to that observed in unseparated cells and the sialidase activities associated with the various subcellular fractions

OPNA+ l PNA-

exhibited shown).

the same kinetic parameters

(data not

T lymphocytes Our observation of a sialidase activity optimally active at acidic pH in the cytosolic fraction obtained from human thymocytes was unexpected since mammalian cytosolic sialidases are usually optimum at more neutral pH (Corfield et al., 1981). Therefore, we investigated the cytosolic fraction of purified T lymphocytes in order to determine whether this acidic soluble form was a general characteristic of the T-lineage or whether the neutral activity did appear during the late phase of T-cell differentiation. The sialidase activity occurring in T-cell cytosolic fraction was optimum at pH 4.2. Furthermore, as for the thymocyte cytosolic activity, a significant shoulder around pH 5.0-5.4 was observed. However, this shoulder was more pronounced in T-lymphocytes. DISCUSSION

4.0 4.2

4.4

4.6

5.0

5.4

pH Values Fig. 3. Sialidase activities in crude homogenates of mature and immature thymocytes which were separated by differential PNA agglutination. Specific activities were determined at various pH ranging from 3.6 to 5.4.

Recently, Taira and Nariuchi (1988) presented data showing that the increased level of sialidase activity observed in activated T-cell was necessary for these cells to acquire responsiveness to allogeneic B-cells, probably through desialylation of critical molecules of the T-cell

714

A. Table 1. Properties human thymocytes.

GREFFARD

et ul.

of sialidase activities associated with subcellular fractions Mu NeuAc: 2’-(4-methylumbelliferyl)-a-p-N-acetylneuraminic Lysosomal

Optimum pH Apparent KM for Mu NeuAc U/mg protein % Total sialidase activity pH = 4.2

surface. This study (Taira and Nariuchi, 1988) substantiated earlier reports suggesting that Tcell sialidase activity participated in the control of the sialic acid content of the cell surface (Landolfi et al., 1985, 1986) and thus may play a functional regulatory role (Stancou et al., 1984). It highlighted the need for a more detailed characterization of this enzymatic activity in the T lineage. The present study constitutes the first description of the human thymocyte sialidase system. Using classical subcellular fractionation procedure, we were able to demonstrate that the bulk of sialidase activity towards the fluorogenic substrate Mu-NeuAc was membrane-bound since 59.6 and 33% of the total activity was recovered in the lysosomal and microsomal fraction, respectively. This finding is in agreement with the current knowledge on intracellular distribution of mammalian sialidase (Corfield et al., 198 1; Schauer, 1982). The “lysosomal” and “microsomal” activities were similar with regard to their acidic optimal pH and their apparent K,,, for Mu-NeuAc, as well as in the way they were affected by various divalent cations. However, such similitudes observed with a synthetic substrate do not imply that these activities are identical since lysosomal and plasma membrane sialidase activities have been repeatedly found to exhibit different specificities towards natural substrates (Corfield et al., 198 1; Schauer, 1982). Additionally, low level of soluble activity (7.3% of the total activity at pH 4.2) has been found in the cytosolic fraction, provided that this fraction had been concentrated IO-fold at least. In this respect, two concentration methods have been compared and it appeared that ammonium sulfate precipitation was easier, quicker and provided better yield than ultrafiltration. This soluble sialidase exhibited a rather acidic optimum around pH 4.0-4.2 but a significant shoulder was observed at pH 5.0-5.4 suggesting that this fraction may also contain a minor more neutral activity. Examination of circulating T-cell revealed that these cells

4.4 0.61~10-4M 4.41 59.6

Microsomal 4.4 0.59.10-4 M 14.96 33

from acid

Cytosolic 4.0 1.95.10-4 0.67 7.3

M

contain a cytosolic sialidase activity quantitatively and qualitatively similar to that observed in thymocytes indicating that no significant changes occurred in the cytosolic sialidase along with thymocyte maturation. Although these acidic optimum pH are in good agreement with that reported by Yeh et al. (1971) for the sialidase activity occurring in human leukocyte cytosolic fraction, they are at variance with the generally accepted concept of mammalian soluble sialidases that are usually found optimum in a more neutral pH range: 5.5-6.0 (Miyagi and Tsuiki, 1985; Tulsiani and Carubelli, 1970). However, data reported on cytosolic sialidases are somewhat controversial. For instance, some workers were able to purify a soluble neutral sialidase from the cytosolic fraction of rat liver (Miyagi and Tsuiki, 1984, 1985) while others did not detect such a soluble sialidase in the rat liver although they used the same strain (Horvat and Touster, 1968). The reasons for this discrepancy are obscure but they are likely to be related to the homogenization procedure (Schauer, 1983). Noticingly, using the methodology described herein we previously demonstrated the presence of a soluble near-neutral activity in other cell types such as rabbit (Pilatte et al., 1987) and human alveolar macrophage (Pilatte et al., submitted). During thymic residence T-cells undergo a complex maturation process leading to the acquisition of T-cell-specific functions such as help, killing and suppression. Considering that significant changes in sialidase activity have been reported along with differentiation (Nojiri ef al., 1982) or maturation (Lambre et al., 1990) in other unrelated cell types, it was of interest to determine whether thymocyte sialidase activity was altered during the course of thymic maturation. Inasmuch as the extent of surface sialylation has been found to change during thymocyte maturation (Toporowicz and Reisner, 1986) and previous studies showing a reverse correlation between the sialylation level of certain cell surface molecules and the amount of intracellular sialidase activity have suggested

Human

thymocyte

a role for these enzymes in the control of cell-surface sialic acid content (Landolfi et al., 1985, 1986; Taira and Nariuchi, 1988). The present study revealed that the increased cell surface sialylation observed during thymocyte maturation was not accompanied by change in sialidase activity since we found no significant difference between mature and immature thymocytes, neither in the level of total cell activity nor in the subcellular distribution of this enzymatic activity. These findings indicate that the implication of sialidases in the mechanisms leading to the increased expression of siahc acid on thymocyte surface during maturation is unlikely. Rather these changes are probably due exclusively to the induction of sialytransferases as previously shown in murine thymocytes (Toporowicz and Reisner, 1986) while the role of endogenous sialidases in the control of cell surface sialylation seems to be restricted to celf. activation in the T-lineage. This hypothesis may have important physiological consequences since if changes in sialyltransferase activity are susceptible to affect most cell-surface glycoproteins (Reichner ef al., 1988), several lines of evidence suggest that the putative sialidase-mediated processing proposed in T-cell activation is likely to be more selective (for a review, see Pilatte et al., 1993). At the steady state, the sialidases acting on glycoproteins are located within the Iysosomes. Therefore, the cell surface glycoproteins susceptible to sialidase-mediated processing are those which colocalized transiently with lysosomal sialidases during intracellular trafficking (Fishman and Cook, 1986). As a result, a limited number of piasma membrane glycoproteins are likely to be affected by changes in cellular sialidase activity such as those observed in T-cells after lectin (Landolfi et al., 1985) or allogeneic activation (Taira and Nariuchi, 1988). Interestingly, among the candidate molecules of which sialylation may be partfy controlled by cellular sialidases are two functionally important glycoproteins which both recycle through reprocessing pathways (Reichner et al., 1988): the transferin receptor (Benoist ef al., 1988; Madoulet et al., 1986) and MHC Ciass 1 molecules (Landolfi ef af., 1985, 1986). In addition, the fact that in the T-cell lineage, the experimental evidence suggesting that sialidases might participate in the regulation of the sialic acid content of certain cell-surface glycoproteins was only obtained in activated

sialidase

775

activity

cells might be of paramount importance in the course of T-cell activation. Indeed, it has been shown that the sialylation state of specific molecules rather than the total cell-surface sialic acid was critical in modulating certain recognition events that take place on the surface of immune cells (Powell et al., 1987; Taira and Nariuchi, 1988). REFERENCE Benoist H., Madouiet C., Trentesaux C., Carpentier Y., Joly P., Jardillier J. C. and Desplaces A. (1988) Association of Adriamycin-induced resistance to NK-mediated lysis with sialic acid level and immunological reactivity of transferrin receptors and glycophorin A. Int. J. Cancer 42, 299-304. Boog C. J. P., Neetjes J. J., Boes J., Ploegh H. L. and MeliefC. J. M. (1989) Specific immune responses restored by alteration in carbohydrate chains of surface molecules on antigen-presenting cells. Eur. J. Immunol. 19, 537-542. Bradford M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Ana&. Biochem. 72, 248254. Corfield A. P., Michalski J. C. and Schauer R. (1981) The substrate specificity of sialidases from microorganisms and mammals. In Perspectives in Inherited Me/abolic Diseases (Edited by G. Tettamanti, P. Durand and Didonato) 4, pp. 3-70. Ermes, Milan. Cowing C. and Chapdelaine J. (1983) T cells discriminate between Ia antigens expressed on allogeneic accessory cells and B cells: a potential function for carbohydrate side chains on la molecules. Proc. natn. Acad. Sri. U.S.A. 80, 6000-6004. Fishman J. B. and Cook J. S. (1986) The sequential transfer of internalized, cell surface sialoglyc~onjugates through the lysosomes and golgi complex in HeLa cells. J. biol. Chem. 261, 11896611905. Frohman M. and Cowing C. (1985) Presentation of antigen by B cells: functional dependence on radiation dose, interleukins, cellular activation and differential glycosylation. J. Immunol. 134, 2269-2275. Hirayama Y., Inaba K., Inaba M., Kato T., Kitaura M., Hosokawa T., lkehara S. and Muramatsu S. (1988) Neuraminidase-treated macrophages stimulate allogenic CD8 + T cells in the presence of exogenous interleukin 2. J. Exp. Med. 168, 1443-1456. Horvat A. and Touster 0. (1968) On the lysosomal occurence and the properties of the neuraminidase of rat liver and of Ehrlich ascites tumor cells. J. bio/. Chem. 243, 4380-4390. Kearse K. P., Cassatt D. R., Kaplan A. M. and Cohen D. A. (1988) The requirement for surface Ig signaling as a prerequisite for T cell: B cell interactions. A possible role for desialylation. 1. Immunol. 140, 1770&1778. Lamb& C. R., Greffard A., Gattegno L. and Saffar L. (1990) Modi~cations of sialidase activity during the monocyte-macrophage differentiation in vitro. Immunol. Lett. 23, 179-182. Landolfi N. F., Leone J., Womack J. E. and Cook R. G. (1985) Activation of T lymphocytes results in an increase in H-2 encoded neuraminidase. Immunogenetics 22, 159-l 67.

776

A. GREFFARD et al.

Landolfi N. F. and Cook R. G. (1986) Activated T lymphocytes express class I molecules which are hyposialylated compared to other lymphocyte populations. Molec. Irnrn~~o~. 23, 297-309. Madoulet C., Trentesaux C., Bcnoist H., Rebel G., Dreyfus H. and Jardillier J. C. (1986) Evidence for sialidase activity in K 562 cells: inhibition by adriamycin treatment. Cancer Biochem. Biophys. 9, 15-23. Miyagi T. and Tsuiki S. (1984) Rat-liver lysosomal sialidase. Solubilization, substrate specificity and comparison with the cytosolic sialidase. Eur. J. Bjoehem. 141, 75-81. Miyagi T. and Tsuiki S. (1985) Purification and characterization of cytosolic sialidase from rat liver. J. biol. Chem. 260, 6710-6716. Nojiri H., Takaku F., Tetsuka T. and Saito M. (1982) Stimulation of sialidase activity during cell differentiation of human promyelocytic leukemia cell line HL-60. Biocltem. biophys. Rex Commun. 104, 1239-1246. Pilatte Y., Bignon J. and Lambre C. R. (1987) Lysosomal and cytosolic sialidases in rabbit alveolar macrophages: demonstration of increased lysosomal activity after in uiuo activation with bacillus Calmette-Gutrin. Biochim. Bioph.ys. Acta. 923, 150-t 55. Pilatte Y., Bignon J. and Lambre C. R. (1993) Sialic acids as important molecules in the regulation of the immune system: pathophysiological implications of sialidases in immunity. Glycobiology 3, 201-217. Potier M., Mamelli L., Dallaire L. and Melaqon S. B. (1979) Fluorometric assay of neuraminidase with a sodium (4-methylum~lliferyl-~-D-~ acetyl neuraminate) substrate. Ana&. Biochem. 94, 287-296. Powell L. D., Whiteheart S. W. and Hart G. W. (1987) Cell surface sialic acid influences tumor cell recognition in the mixed lymphocyte reaction. J. Immunol. 139, 262-270.

Reichner J. S., Whiteheart S. W. and Hart G. W. (1988) Intracellular trafficking of cell surface sialoglycoconjugates. J. biol. Chem. 263, 16316-16326. Reisner Y. and Sharon N. (1984) Fractionation of subpopulations of mouse and human lymphocytes by peanut agglutinin or soybean agglutinin. In Methods in enzymology (Edited by G. Di Sabato, J. J. Langone and H. Van Vunakis) 108, pp. 1688179. Academic Press, Inc. N.Y. Schauer R. (1982) Chemistry, metabolism and biological functions of sialic acids. Adv. Carbohyd. Chem. Biochem. 40, 131-134. Schauer R. (1983) Glycosidases with special reference to the pathophysiological role of sialidases. In Structural Carbohydrates in the Liver (Edited by Popper H., Kottgen E., Reutter W. and Gudat F.), pp. 83-97. Falk Symposium 34, M.T.P. Press Ltd. Stancou R., Dumitresco S. M. and Robineaux R. (1984) Possible role of sialic acid and endogenous neuraminidase in T cell proliferation. Cell Biology ~~ter~otional Reports 8, 637-647. Taira S. and Nariuchi H. (1988) Possible role of neuraminidase in activated T-cells in the recognition of allogeneic Ia. J. Immunol. 141, 440-446. Toporowicz A. and Reisner Y. (1986) Changes in sialyltransferase activity during murine T cell differentiation. Cell. Irnrn~~o~. 100, 10-19. Tulsiani D. R. P. and Carubelli R. (1970) Studies on the soluble and lysosomal neuraminidases of rat liver. J. biol. Chem. 245, 1821-1827. Yamamoto N. and Kumashiro R. (1993) Conversion of Vitamin D, binding protein (group-specific component) to a macrophage activating factor by the stepwise action of /I-galactosidase of B cells and sialidase of T cells. J. Immunol. 151, 2794-2802. Yeh A. K., Tulsiani D. R. P. and Carubelli R. (1971) Neuraminidase activity in human leukocytes. J. Lab. Clin. Med. 78, 771-778.