Search

It is now well established that human lymphoblastoid cell lines showing immaturity characters display ecto-5'-nucleotidase activities lower than normal levels. A recent paper (Sun, A.S., Holland, J.F. and Ohnuma, T. (1983) Biochim. Biophys. Acta 762, 577-584) mentioned that this phenomenon resulted from the presence of a 5'-nucleotidase inhibitor in these cell lines. We demonstrate here that the use of 5'-[3H]AMP as a substrate, and inadequate analysis of the products formed, led them to a misinterpretation. [3H]Adenosine derived from 5'-[3H]AMP hydrolysis was further transformed into [3H]inosine by the adenosine deaminase activity of the leukemic cell lines tested; [3H]inosine was precipitated with the excess substrate and was not taken into account in the ecto-5'-nucleotidase determination, which led the authors to confuse this adenosine deaminase activity with a 5'-nucleotidase inhibitor. We did not observe 5'-nucleotidase inhibition by leukemic cell cytosol when convenient assay methods were used and showed that the presence of such an inhibitor remains to be established.

5'-Nucleotidase is purified from lymphocyte plasma membranes by two affinity chromatographies. The first one, on Lens culinaris lectin-Sepharose 4B yields a fraction of twelve lectin-binding glycoproteins (lectin-receptor fraction). The second one on 5'-AMP-Sepharose 4B leads to pure enzyme. This enzyme is a glycoprotein with a molecular weight of 130 000; it gives a single band in polyacrylamide/dodecylsulfate electrophoresis and displays a very high specific activity (2500-3000 mumol Pih-1mg-1). Some properties of purified 5'-nucleotidase are similar to those of membrane-bound enzyme: substrate specificity, temperature dependence, effects of ions and SH-blocking reagents. Others are completely different for the two systems and these differences result from an interaction between the enzyme molecule and other Lens culinaris lectin binding proteins.

Murine T lymphocytes were separated by "panning" into four subpopulations, according to their Lyt2 and L3T4 phenotypes: Lyt2+L3T4+, Lyt2-L3T4-, Lyt2+L3T4-, and Lyt2-L3T4+. The activity of ecto-5'-nucleotidase in each subpopulation was measured. 5'-Nucleotidase activity was undetectable in Lyt2+L3T4+ cortical cells but was expressed in medullary Lyt2-L3T4+ and Lyt2+L3T4- T lymphocytes. The small cortical subpopulation of thymocyte precursors with the Lyt2-L3T4- phenotype expressed levels of 5'-nucleotidase comparable to the levels of medullary, mature lymphocytes. These results suggest that the use of ecto-5'-nucleotidase as a marker of the degree of T-cell maturation is questionable.

The 5'-nucleotidase properties of isolated lymphocyte plasma membranes from young pig mesenteric nodes are described; nucleosides-5'-monophosphates are the substrates of this specific enzyme. Concanavalin A inhibits this enzyme; on the same membranes this mitogen does not affect alkaline phosphatase and activates the membrane bound (Ca2+) ATPase. The 5'-nucleotidase inhibition is due to a specific interaction of Con A with carbohydrate groups of the membrane; its high positive cooperativity suggests that the lectin promotes reorganization of the membrane bound 5'-nucleotidase. Solubilization of the 5'-nucleotidase does not prevent the effect of Con A and the solubilized enzyme is firmly bound by Con A-Sepharose 4B; these results suggest that Con A inhibits the enzyme by a direct interaction and that 5'-nucleotidase can be considered as an eventual receptor for the lectin.