EHPnet.

a-L-fucosidase (a-L-fucoside fucohydrolase EC 3.2.1.5 1) is an enzyme that can be separated by isoelectric focusing into several isoelectric forms (Alhadeff & O'Brien, 1977). In contrast to most mammalian species which contain predominantly acidic and neutral forms of cx-L-fucosidase, mouse tissues also contain unusual basic forms of the enzyme with approximate pI values of 8.3 and 9.0 (Laury-Kleintop et al., 1985a). These basic forms are most prominent in the foetal tissues and placenta where they may account for up to 75% of the total activity. In contrast, adult tissues contain only 20-25% of basic forms. Embryonal carcinoma, tumours whose stem cells correspond to pluripotent embryonic cells from early stages of development (Solter & Damjanov, 1979), also contain, in accordance with their embryonic nature, predominantly the basic isoelectric forms of CX-L-fucosidase (LauryKleintop et al., 1985b). Our studies of mouse a-L-fucosidase have so far concentrated on the enzyme from solid organs or homogeneous cell populations grown in culture (Laury-Kleintop et al., 1985a,b). The present study was undertaken to determine whether the acidic and basic forms of a-L-fucosidase could also be found in body fluids. To that end we have analyzed ascites fluid, produced by injecting a trophectodermal carcinoma into the peritoneal cavity (Damjanov et al., 1985). This tumour was chosen for two reasons: because it represents a malignant equivalent of trophectoderm and could thus (like placenta) be a good source of the embryonic (basic) forms of a-L-fucosidase; and because it grows readily in ascites form and causes rapid accumulation of ascites fluid. By studying the forms of OC-Lfucosidase in the ascites fluid we hoped to determined whether the enzyme is derived from the adult host or released from the placenta-like tumour cells. We hypothesized that the pre-

a-L-fucosidase (a-L-fucoside fucohydrolase EC 3.2.1.5 1) is an enzyme that can be separated by isoelectric focusing into several isoelectric forms (Alhadeff & O'Brien, 1977). In contrast to most mammalian species which contain predominantly acidic and neutral forms of cx-L-fucosidase, mouse tissues also contain unusual basic forms of the enzyme with approximate pI values of 8.3 and 9.0 (Laury- Kleintop et al., 1985a). These basic forms are most prominent in the foetal tissues and placenta where they may account for up to 75% of the total activity. In contrast, adult tissues contain only 20-25% of basic forms. Embryonal carcinoma, tumours whose stem cells correspond to pluripotent embryonic cells from early stages of development (Solter & Damjanov, 1979), also contain, in accordance with their embryonic nature, predominantly the basic isoelectric forms of CX-L-fucosidase (Laury-Kleintop et al., 1985b).
Our studies of mouse a-L-fucosidase have so far concentrated on the enzyme from solid organs or homogeneous cell populations grown in culture (Laury-Kleintop et al., 1985a,b). The present study was undertaken to determine whether the acidic and basic forms of a-L-fucosidase could also be found in body fluids. To that end we have analyzed ascites fluid, produced by injecting a trophectodermal carcinoma into the peritoneal cavity (Damjanov et al., 1985). This tumour was chosen for two reasons: because it represents a malignant equivalent of trophectoderm and could thus (like placenta) be a good source of the embryonic (basic) forms of a-L-fucosidase; and because it grows readily in ascites form and causes rapid accumulation of ascites fluid. By studying the forms of OC-Lfucosidase in the ascites fluid we hoped to determined whether the enzyme is derived from the adult host or released from the placenta-like tumour cells. We hypothesized that the pre-Correspondence: I. Damjanov. dominance of embryonic (basic) forms would indicate that the enzyme in the fluid is mostly of tumour cell origin. On the other hand, if the tumour-induced accumulation of fluid is just a transudate of proteins produced by the host, then one would expect it to contain predominantly acidic isoforms of a-L-fucosidase.
Ascites was produced by injecting 1-2 x 106 trophectodermal carcinoma E6246D cells into adult outbred female Swiss-Webster mice. This tumour cell line, originally derived from serial transplants of a spontaneous ovarian teratocarcinoma of a C3H/Fe mouse (Fekete & Ferigno, 1952), was fully characterized and shown to be developmentally and immunochemically equivalent to trophectodermal cells in the preimplantation stage mouse embryos (Damjanov et al., 1985). The tumour was allowed to grow for 7-10 days during which time it produced 25-30ml of clear ascites fluid. The animals were sacrificed by cervical dislocation and the fluid containing tumour cells was harvested into a glass container. Special care was taken to avoid contamination of the ascites fluid with blood, and all the animals with bloody ascites were discarded and not included into the study. Ascites was pooled from 3 animals and prepared for iso-electric focusing as follows. The cells were separated from the fluid by centrifugation at 12,000g for 20min. The supernatant was collected for further analysis and the cells were washed three times in 0.9% (w/v) NaCl to remove all traces of the ascites fluid. The washed cells were resuspended 1: 3 (w/v) 10 mM, pH 5.0, citric acid-sodium citrate buffer containing 0.02% (w/v) NaN3. The cells were lysed by five cycles of freeze-thawing in liquid nitrogen and the resulting homogenates were centrifuged at 12,000g for 20min. The supernatant fluids and resuspended pellets (after washing once with 10 mM, pH 5.0, citric acid-sodium citrate buffer) were assayed for a-L-fucosidase activity under conditions of linearity essentially as previously described (Alhadeff & O'Brien, 1977)  Colnbrook, Bucks, UK). Fluorescence was read on a Turner Model 111 fluorometer. A unit of activity is defined as the amount of enzyme which hydrolyzes 1 nmol of substrate min-' at 37°C. All reported ax-L-fucosidase activities have been corrected by subtracting appropriate tissue and substrate blanks.
Isoelectric focusing was performed at 2-4°using a 40 ml column essentially as previously described (Alhadeff et al., 1975), employing two percent ampholytes (pH range 5-8; LKB-Produkter, Stockholm, Sweden) and a 0.67% (w/v) sucrose gradient. Electrofocusing was conducted on 12 to 14 units of aC-L-fucosidase activity at starting amperages of 1.5 to 2.0mA (and 600V) for 15-18h after which 0.4 ml fractions were collected. The pH value of each fraction was determined at 2-4°C using a Beckman 3500 Digital pH meter, and 5Opl aliquots of each fraction were assayed for 30 min at 37°C for (X-L-fucosidase activity. The results were plotted and the relative amounts of a-L-fucosidase activity associated with the basic isoelectric forms (above an approximate pl value of 7) and the more acidic isoelectric forms (below an approximate pl value of 7.0) were determined by cutting out the respective portions of the profiles and weighing them on a Mettler analytical balance (Laury-Kleintop et al., 1985a).
The ascites fluid contained 32.6 units of a-Lfucosidase activity ml-1 and the trophectodermal carcinoma cells contained 11.4 U g 1 cells. This is in the range of Of-L-fucosidase activity (3.0-61.4Ug-1 cells) of other tumour cells studied previously (Laury-Kleintop et al., 1985b). Figure 1 depicts typical isoelectric focusing profiles of x-L-fucosidase activity from ascites flud (a) and from the supernatant fluid of the lysate of trophectodermal carcinoma cells (b). The bulk (88%) of recovered ascites fluid a-L-fucosidase activity is associated with acidic isoelectric forms (with approximate pl values ranging from 4.5 to 6.3) whereas only 12% of the activity is associated with the three more basic forms (with approximate pl values of 7.5, 8.2 and 9.0). In contrast, 24% of the recovered enzyme activity from the tumour cell lysates is associated with acidic isoelectric forms (pI values of 4.5 to 6.5) and 76% is associated with the three distinct basic peaks of activity (at pl values of 7.5, 8.2 and 9.2) which are seen in only small amounts in the ascites fluid. The three major basic peaks found in the cell lysates are comparable to those observed previously in mouse placenta (Laury-Kleintop et al., 1985a). Supernatant fluids from the host liver cells run in parallel contained more than 75% of recovered a-Lfucosidase activity associated with isoelectric forms with pl values below 7 in accordance with previously published data (Laury-Kleintop et al., 1985a).
In the present study we have shown that isoelectric focusing may be used for analysis of the forms of a-L-fucosidase in ascites fluid and that this approach may provide information about the derivation of the enzyme in the fluid. However, it was disappointing to find that the fluid bathing a-Lfucosidase-rich cells contains so little tumourderived enzyme. This indicates that isoelectric focusing of ascites fluid is not a useful approach for identifying tumour-derived Oe-L-fucosidase and that the basic forms characteristic of embryonic tumours probably cannot be used as markers for diagnostic purposes. On the other hand, the finding that the bulk of the enzyme activity is associated with acidic forms (even in the presence of tumour cells rich in basic isoelectric forms) suggests that the enzyme is derived primarily if not exclusively from the host. Analysis of OC-L-focusidase forms of dual origin thus provides additional evidence for the hypothesis that the ascites fluid elicited by the tumours is mostly of host origin and its protein content reflects like the mechanically caused peritoneal transudate of the composition of plasma (Pare et al., 1983).