Carbonic anhydrase isoenzymes I and II in rabbit erythroid cells.

The morphological and cytochemical changes in erythroid-cell differentiation have been extensively studied: application of the velocity-sedimentation technique has provided preliminary data on the associated biochemical changes (Denton & Arnstein, 1973). After the last cell division there is a dramatic increase in the synthesis of haemoglobin which, at least in part, probably accounts for the apparent decline in the activity of several enzymes at this stage (Denton et al., 1974). However, some enzymes, including carbonic anhydrase (EC 4.2.1.1), show a continued increase in activity throughout the whole differentiation sequence. Previously we suggested (Denton et af., 1974) that this increase in carbonic anhydrase activity was most probably due to synthesis of the socalled ‘high-activity’ isoenzyme, 11, rather than to synthesis of the ‘low-activity’ isoenzyme, I. Here we report preliminary experiments in which we measured isoenzyme I and I1 protein in lysates of the main erythroid cell types and compared this with the carbonic anhydrase activity to try to determine whether synthesis de novo of either or both isoenzymes or activation of inactive precursors occurs during differentiation. Erythroid cells from rabbit bone marrow were fractionated into the four main cell types (Denton et al., 1974) and lysates from these were analysed for isoenzymes I and I1 by immunoassay (Peller 8c Spencer, 1976), carbonic anhydrase activity (Rickli et al., 1964) and protein (Lowry et al., 1951). The results of a typical separation are shown in Fig. 1 . Clearly, dramatic changes in carbonic anhydrase activity are not due to corresponding changes in enzyme protein amount: the amount of isoenzyme-I protein does apparently increase as differentiation proceeds, but is only sufficient to account for less than 10% of the activity increase. The most obvious explanation is that most of the isoenzyme 11, and possibly isoenzyme I, in earlier cells is present as enzymically inactive but immunoreactive precursors. As carbonic anhydrase is a ZnZ+ metalloprotein, the apoenzyme is a likely candidate for this precursor role, particularly since Denton (1973) showed that Zn2+ from the stroma is gradually transferred to the cytoplasm during erythroid differentiation. Knowing the specific activities of purified isoenzymes I and I1 (Peller, 1976) and the amounts of both proteins in lysates of the various fractions, it is possible to calculate the total and specific activities of carbonic anhydrase expected and compare these with the observed values (see Table 1) . Observed activities for fractions 1 and 2 are lower than expected, which again suggests the presence either of an inactive form of the enzyme or possibly of an inhibitor. In fractions 3 and 4, observed carbonic anhydrase activities are higher than expected, which might indicate the presence of a third active form of the enzyme, unrelated to both isoenzymes I and I1 and therefore not detected by the immunoassay : such a form has not been reported in rabbit reticulocytes. Alternatively, the data might suggest the existence of an activator in cells of fractions 3 and 4, but this would necessarily need to disappear during erythrocyte maturation, since the measured activity of carbonic anhydrase in mature erythrocytes agrees with that calculated from the known isoenzyme I and I1 content. The data can be explained in terms of an activation process in which some of the isoenzyme 11, and possibly isoenzyme I, is converted into a hyperactive, relatively labile form. The demonstration (Tashian et al., 1975) that carbonic anhydrase from many species is N-acetylated at the N-terminal residue allows the speculation that deacylation of the enzyme might increase its specific catalytic activity while rendering it more labile. If, as proposed for the lens crystallin proteins (Strous et al., 1973), acetylation occurs as