[CANCER RESEARCH 38, 723-728, March 1978] Alterations in Serum Glycosyltransferases and 5 -Nucleotidase in Breast Cancer Patients1 Clement Ip and Thomas Dao Department of Breast Surgery and Breast Cancer Research Unit, Roswell Park Memorial Institute, Buffalo, New York 14263 ABSTRACT isoenzyme was absent in control subjects, but it was signif icantly higher especially in patients with widespread metas We have measured sialyltransferase, galactosyltrans- ferase, and fucosyltransferase as well as S'-nucleotidase tasis. In this report we have examined 3 glycosyltransferases in the serum of breast cancer patients. Serum sialyltrans (sialyltransferase, galactosyltransferase, and fucosyltrans ferase values in 65 normal healthy females ranged from ferase) as well as 5'-nucleotidase in the serum of breast 2.6 to 8.5 units, with a mean of 5.4. In 25 women with cancer patients with either potentially curable disease or operable primary breast cancer, serum sialyltransferase disseminated métastases.Theobjective of this investigation levels were found to be between 6.2 and 15.4 units. was to determine whether the levels of these enzymes in Marked elevation of this enzyme level (range, 8.8 to 36 the serum can be correlated with the extent of the disease units) was observed in 48 patients with metastatic breast in these patients and to identify the most sensitive marker(s) cancer. Galactosyltransferase and fucosyltransferase for subsequent follow-up study in cancer patients. measurements, however, showed considerable overlap between the controls and the cancer patients. On the other hand serum 5 -nucleotidase and sialyltransferase MATERIALS AND METHODS in breast cancer patients showed very similar patterns. Materials. UDP-[fJ-'4C]galactose (specific activity, 274 Thus, serum 5 -nucleotidase values in 44 normal females mCi/mmole), CMP-[4,5,6,7,8,9-14C]sialic acid (specific ac ranged from 11.4 to 23.2 units, whereas the levels found tivity, 173 mCi/mmole), and [G-3H]AMP (specific activity, in 30 patients with metastasis were between 25 and 71.8 13.5 Ci/mmole) were purchased from New England Nu units. The tissue origin of abnormal levels of serum clear, Boston, Mass. GDP-[L/-'4C]fucose (specific activity, glycosyltransferases and 5 -nucleotidase was discussed 117 mCi/mmole) was bought from Amersham/Searle Corp., in relation to their physiological significance as well as Arlington Heights, III. Fetuin was purchased from Grand their role as markers for diagnosing early malignant Island Biological Co., Grand Island, N. Y. AMP, hen ovo- breast neoplasm and for monitoring the extent of metas mucoid (trypsin inhibitor; Type II-O), and A/-acetylneura- tasis. minic acid were obtained from Sigma Chemical Co., St. Louis, Mo. INTRODUCTION SF-fetuin3 for the sialyltransferase assay was prepared by mild acid hydrolysis of fetuin (27). SGF-fetuin for the galac Recently, there has been a growing interest in studying tosyltransferase assay was prepared by the periodate oxi- the alterations of circulating glycosyltransferases in cancer dation-borohydride reduction method described by Spiro patients. Kessel ef al. have reported elevated plasma sialyl (28). transferase (10, 11) and fucosyltransferase (12) in patients Serum Collection. Blood was collected from patients in with various types of malignant diseases. Kessel et al. (10, Vacutainers without anticoagulant. RBC were removed by 13) have also observed that changes in plasma sialyltrans centrifugation, and serum samples were stored at -70° ferase level were correlated with the course of the disease until ready for assay. in patients studied serially during therapy. Bhattacharya ef Enzyme Assays. For the determination of sialyltransfer al. (6) have reported an increase in serum galactosyltrans- ase, the assay mixture contained 50 fj.\ serum, 500 /¿gSF- ferase levels in patients with ovarian cancer. Similarly, fetuin, 1 nmole CMP-[14C]sialic acid (4.4 x 10s dpm), 1 ITIM fluctuations of this enzyme in the serum appeared to corre MnCU, and 20 m.M Tris-maleate buffer (pH 7.0) in a final late with tumor volume as well as the clinical status of the volume of 150 /J. For the determination of galactosyltrans patient. On the other hand Podolsky and Wieser (22) ferase, the assay mixture contained 20 /^l serum, 250 ^g showed that on the average the total serum galactosyltrans- SGF-fetuin, 3 nmoles UDP-['4C]galactose (2.2 x 10" dpm), ferase activity in cancer patients was only slightly higher 10 mM MnCU, and 20 HIM Tris-maleate buffer (pH 6.8) in a than that in normal" control subjects. However, by discon total volume of 100 ^1. For the assay of fucosyltransferase, tinuous polyacrylamide gel electrophoresis, Weiser ef al. the incubation reaction consisted of 40 ¿¿Iserum,4 mg (31) detected an isoenzyme of galactosyltransferase in the ovomucoid, 2 nmoles GDP-['4C]fucose (5 x 10s dpm), 10 serum in 43 of 58 patients with various types of cancer. This mM MgCL, and 20 mivi Tris-maleate buffer (pH 6.8) in a total volume of 100 /¿I.For the determination of endogenous 1 This work was supported in part by Grant CA 14812-03, National Cancer activities in all 3 glycosyltransferase measurements, the Institute, NIH, USPHS, and by the Mary Plagier Cary Charitable Trust. ' To whom requests for reprints should be addressed, at Roswell Park Memorial Institute. 666 Elm Street, Buffalo, N. Y. 14263. 3The abbreviations used are: SF-fetuin, sialic acid-free fetuin; SGF, sialic Received August 8. 1977; accepted December 13, 1977. acid-free and galactose-free fetuin. MARCH 1978 723 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1978 American Association for Cancer Research. C. Ip and T. Dao respective exogenous glycoprotein acceptor was omitted in RESULTS the reaction mixture. Tubes for the glycosyltransferase assays were incubated Table 1 summarizes some of the biological characteristics at 37°ina Dubnoff shaker for various periods of time: 1 hr of the serum sialyltransferase. Since the addition of a non- for sialyltransferase, 30 min for galactosyltransferase, and ionic detergent (Triton X-100) to the incubation mixture did 10 min for fucosyltransferase. These different incubation not affect the sensitivity of the enzyme measurement, it was times were chosen so that linearity of the reaction was therefore omitted in routine assays. achieved under the given assay conditions. The reaction The reaction catalyzed by sialyltransferase is stimulated was terminated by the addition of 2 ml ice-cold 5% trichlo- by Mn* ions and is specific for the transfer of sialic acid to roacetic acid:2% phosphotungstic acid in 0.5 N HCI, and the penultimate galactose residue. Native fetuin (which the tubes were centrifuged at 2,600 x g for 1C min. The contains terminal sialic acid residues) and SGF-fetuin are resultant pellet was washed twice with the same acid hence not suitable acceptors for the sialyltransferase assay. solution and once with ethanohdiethyl ether (2:1, v/v). The Fetuin, a fetal calf serum glycoprotein, has been shown to precipitate was dissolved in 1 ml NCS tissue solubulizer contain more than 1 oligosaccharide side chain with termi (Amersham/Searle) and placed in 10 ml scintillation cock nal sialic acid moieties (29). The sialic acid residue is tail consisting of Spectrafluor (Amersham/Searle):ethylene attached either «(2,3)to galactose or a(2,6) to /V-acetylga- glycol monoethyl ether:toluene (1:10:14, by volume) and lactosamine. The present assay system therefore conceiva 8% naphthalene. Radioactivity was determined by liquid bly measures the activity of 1 or more sialyltransferase(s), scintillation counting. All 3 glycosyltransferase activities although it is assumed that the majority of the sialic acid is were reported as exogenous activities calculated by the transferred to the galactose moiety in SF-fetuin. difference between total (in the presence of exogenous The endogenous activity (assayed in the absence of glycoprotein acceptor) and endogenous (in the absence of exogenous acceptor) is due to the transfer of sialic acid to exogenous acceptor) activities. The enzyme activities are glycoprotein substrates present in the serum. The endoge expressed as pmoles sugar transferred per mg protein per nous activities in serum samples containing high sialyl hr; 1 unit of glycosyltransferase activity is defined as 1 transferase values (from cancer patients) were moderately pmole sugar transferred per mg protein per hr. elevated but never exceeded 15% of the total enzyme For the assay of 5'-nucleotidase, the reaction mixture activity. Data reported in this paper concern only the exog contained 100 /J serum, 0.5 ^mole [3H]AMP (1.1 x 10s enous activities (total minus endogenous). dpm), 20 mM MgCI,, 50 mM Tris-HCI buffer (pH 8.5) in a Chart 1 illustrates the serum sialyltransferase activities in total volume of 800 /J. For exclusion of nonspecific phos- normal subjects, in women with benign breast lesions, and phatase, 0.5 ¿¿mole2-glycerophosphatewas also added to in women with breast cancer. The control values obtained the medium as a "substrate diverter." Incubation was car in 65 healthy females between the ages of 18 and 74 ranged ried out at 37°ina Dubnoff shaker for 30 min. The reaction from 2.6 to 8.5 units with a mean of 5.4. The upper limit of the normal range was set arbitrarily at 8.2 units, i.e., 2 was stopped by the addition of 0.1 ml 0.25 M ZnSCs, S.D.'s from the mean. We observed no correlation between followed by 0.1 ml 0.25 M Ba(OH)2. Protein and unhydro- lyzed 5'-AMP were precipitated and removed by centrifuga- age and levels of serum sialyltransferase in this group of tion although adenosine remained in the supernatant. A normal subjects. 0.5-ml aliquot of the supernatant fraction containing Among 21 patients with benign lesions of the breast (all [3H]adenosine was then counted in 10 ml ACS (Amersham/ lesions were confirmed by biopsy), 12 (57%) had serum Searle).