Ribonucleotide Content of Mononuclear Cells from Normal Subjects and Patients with Chronic Lymphocytic Leukemia: Increased Nicotinamide Adenine Dinucleotide Concentration in Chronic Lymphocytic Leukemia Lymphocytes1

The ribonucleotide content of lymphocytes obtained from nor mal subjects and patients with chronic lymphocytic leukemia (CLL) was determined by means of high-performance liquid chromatography. The levels of normal Band T-cells were com pared to each other as well as those of their CLL counterparts. Unfractionated CLL lymphocytes, predominantly B-cells, had significantly lower levels of adenosine-5'-triphosphate, cytidine5'-triphosphate, uridine-5'-triphosphate, cytidine-5'-diphosphate, and guanosine-5'-phosphate, while the concentration of nicotinamide-adenine dinucleotide was significantly higher than in normal unfractionated lymphocytes which consisted mainly of T-cells. For enriched populations: (a) CLL B-cells had much lower adenosine-5'-triphosphate (3439 versus 5689) (pmol/1 x 107 cells), cytidine-5'-triphosphate (107 versus 313), guanosine-5'triphosphate (462 versus 978), and uridine-5'-triphosphate (633 versus 1214) than normal B-cells; (b) CLL T-enriched subpopu lations had significantly lower ribonucleoside triphosphates, adenosine-5'-triphosphate (3217 versus 5468), cytidine-5'-triphosphate (119 versus 209), guanosine-5 '-triphosphate (422 versus 826), and uridine-5'-triphosphate (504 versus 969) than normal T-cells. The lower ribonucleoside triphosphate levels found in unfractionated CLL lymphocytes, therefore, are the result of differences between the CLL and normal B-cells as well as between CLL and normal T-cells. These findings establish a framework for studying the reasons underlying the decreased ribonucleoside triphosphate levels in unfractionated CLL lympho cytes. T-helper and T-suppressor lymphocytes showed similar ribonucleotide patterns. Nucleoside and base levels were signif icantly higher in normal monocytes than in normal lymphocytes. The only compound found to be increased in the CLL Blymphocytes when compared to their normal counterparts was nicotinamide-adenine dinucleotide. The level in CLL lymphocytes was 404 versus 209 pmol/107 cells for normal B-lymphocytes. No correlation was found between any ribonucleotide levels and the expression of 5'-nucleotidase activity. INTRODUCTION Several reports in the literature have indicated that defects may be present in CLL4 lymphocytes affecting cytoskeletal func1Supported by NIH Grants CA 11655 and CA 28376. 2Scholar of the Leukemia Society of America, Inc. 3To whom requests for reprints should be addressed, at Department of Medi cine, New York University School of Medicine. 550 First Avenue, New York, N. Y. 10016. * The abbreviations used are: CLL, chronic lymphocytic leukemia; EN.neuraminidase-treated sheep erythrocytes; ETAC, trypsinized complement-coated sheep erythrocytes; HPLC, high-performance liquid chromatography; PRPP, 5-phosphoribosyl 1-pyrophosphate. Received March 25, 1983; accepted August 3. 1983. tion. These include the finding of an anomalous capping response and decreased motility in CLL B-lymphocytes when compared to their normal counterparts (8, 15,19). Since these are energydependent processes requiring ATP, and low ATP levels have been reported in CLL lymphocytes (17, 23), we determined the nucleotide content of CLL and normal lymphocytes. Our studies confirm the earlier report of decreased ATP levels in these cells and also document alterations in other nucleoside triphosphates and NAD in cells from patients with CLL. To allow a more meaningful analysis of these values, the results were compared in normal and CLL lymphocyte populations enriched in Band Tcells. The latter were further subdivided into fractions enriched in helper and suppressor cells. MATERIALS AND METHODS Heparinized blood obtained from normal subjects and untreated pa tients with CLL was centrifuged on Ficoll-Hypaque gradients as de scribed previously (4) with the substitution of Roswell Park Memorial Institute Medium 1640 for phosphate-buffered saline. The normal donors ranged from 25 to 75 years in age, which overlapped with the age range of the patients. The mononuclear cells were depleted of monocytes by adherence to Falcon plastic culture dishes (31) and by centrifugation through Percoli gradients (27). Normal monocytes were harvested from the plastic plates as described previously (28) or were obtained from the interface of the Percoli gradient. The degree of cellular purity was assessed by Coulter Counter sizing (28). The lymphocyte preparations generally contained <2% monocytes and <10% platelets, while the monocyte preparations ranged from 85 to 90% in purity. Tand 8-Lymphocyte Subpopulations. The percentage of Tand Bcells was assayed by standard resetting techniques using EN and ETAC. Briefly, EN sheep cells were treated with neuraminidase as described previously (44). ETAC sheep cells were prepared by exposure to trypsin, washed 3 times with Roswell Park Memorial Institute Medium 1640, incubated with anti-sheep hemolysin, and then complement coated at 37° (44). The percentage of T-cells was determined by counting 200 cells after incubation of lymphocytes (200 /il; 4 x 106/ml) with 200 /¿I (0.5% EN)at 37°for 5 min followed by a 250 x g, 10-min centrifugation. After a 1-hr incubation, the lymphocytes were resuspended and stained with 1% gentian violet. The percentage of B-cells was assayed by counting 200 cells after incubation of 200 ^l of lymphocytes (4 x 106/ml) with 200 n\ of 0.5% ETAC at 37°for 0.5 hr with gentle rotation. Prior to counting, the lymphocytes were stained with 1% gentian violet. Prepa rations were enriched in Tor B-lymphocytes by either of several meth ods. B-cell enrichment was done either by adherence to anti-(Fab)2 antibody-coated dishes as described by Stark ef a/. (40) or by ETAC rosette formation and centrifugation through Ficoll-Hypaque and lysis as described by BreÃ3me ef a/. (5). T-cells were purified after forming EN rosettes (42) or by harvesting the supernatant from preparations where the B-cells were attached to antibody-coated plates (7). These proce dures yielded preparations with the following major "enriched cell" com position. Normal: normal B-cells, 68 ±5% (S.E.) and normal T-cells, 92 5608 CANCER RESEARCH VOL. 43 Research. on October 23, 2017. © 1983 American Association for Cancer cancerres.aacrjournals.org Downloaded from Nucleotide Levels in Lymphocytes ±3%; CLL B-cells, 93 ±2% and CLL T-cells, 78 ±5%. Populations enriched in suppressor and helper T-lymphocytes were obtained with OKT4 and OKT8 antibodies (Ortho Diagnostics, Nutley, N. J.) as described by Reinherz ef al. (33). The final preparations were >80% purity for each of the normal lymphocyte subsets, and >70% purity for the CLL T-enriched cell subsets. All of the purification protocols were carried out in identical fashion for CLL and normal lymphocytes. HPLC Analysis of 5'-Ribonucleotides. Nucleotides were extracted from cells with 60% methanol by the method of Donofrio ef al. (12). Extracts were prepared from freshly isolated cells (2 to 30 x 106 cells) using a volume of 200 //I, 60% (v/v) methanol/H20/1 x 107 cells. The protein from each cell preparation was dissolved in 0.1 N NaOH and assayed for total protein content by the method of Lowry ef al. (22). The extracts were taken to dryness using a Speedvac (Savant Instruments, Hicksville, N. Y.) and stored at -70°. The residue dissolved in H20 was applied to either a Whatman Partisil 10 SAX or a SOTA 5-^m SAX column. The HPLC was carried out using a Spectra Physics 8000A liquid Chromatograph equipped with a 2-channel data system. Online UV detection was done with both fixed (254 nm) and variable wavelength detectors. Separation conditions were a modification of the method of McKeag and Brown (25) and have been detailed elsewhere (18). An increased flow rate at the end of the gradient (3.0 ml/min at 40 min) was used to reduce the peak widths of the ribonucleoside triphosphates and also allowed for a more rapid analysis cycle time. To minimize the adverse effects of a rising base line on the electronic integration of the nucleotide triphosphates, a method modified from that described by Shmukler (37) for the purification of KH2PO4 was used in the preparation of Buffer B. This was carried out as follows: 1 M KH2PO4 was frozen at -20° and thawed at room temperature, the supernatant was carefully removed, and the conductivity was measured using a Radiometer COM3 conduc tivity meter. The concentration was determined by comparison with a standard conductance curve of 0.1 to 1 M KH2PO4, and the KH2PO4 was diluted to 0.25 M. KCI was then added to 0.5 M before adjusting the pH to 5.0 with 1.0 M KOH as required for the composition of Buffer B. For a typical analysis of ribonucleotide levels, samples were injected (50 to 100 n\) a minimum of 2 times, and the data were averaged. Ribonucleo tide concentrations were converted to pmol/1 x 107 cells and also expressed in terms of the total cellular protein to nmol/mg cellular protein. The recovery of nucleotides during extraction was determined using a mixture of standards containing most of the ribonucleotides present in lymphocytes. One-half of the mixture was immediately analyzed by HPLC, while the other half was subjected to the extraction procedure detailed above. The mean recovery of samples containing the 20 nucleo tides in Chart 1 was 88 ±6.6% (S.D.). Identification and Quantitäten of 5'-Ribonucleotides. 5'-Ribonucle otides were identified by absolute and relative retention times, cochromatography with standards, incorporation of [MC]adenine into purine ribonucleotides (24), absorbance ratios, and continuous flow UV scanning of selected peaks (21). Ribonucleotides were quantitated by calibration with standards (Sigma Chemical Co., St. Louis, Mo.). The degree of purity of the nucleotide standard was found to range from 50 to 90%. To accurately determine the response factor for each nucleotide standard, stock solutions of 1 mg/ml were diluted (1/40) with 0.04 M phosphate buffer, pH 7.0, and the absorbance was determined at the Aâ„¢ for the particular ribonucleotide standard. Aliquots from each standard were separately injected and eluted as described previously (18). In order to obtain the percentage of purity, a ratio of the area of the major peak component in the chromatogram to the total peak areas was determined. This value was used to calculate the true concentration of the standard. The response factor (defined in area units/pmol) was then calculated by dividing the area of the peak component by the true concentration. A minimum of 2 such determinations was carried out for each ribonucleotide response factor. The response factors for each standard were found to be constant from column to column and ranged in value from 54 to 285. A response factor of 100 was used for quantitation of the 3 to 4 peaks eluting early from the chromatogram which" contained unresolved nucleoside and base components. Data were stored in a 2-dimensional array on a Digital Minc-11 laboratory computer and processed using either the Student's f test for unpaired samples or the Mann-Whitney U test. Assay of 5'-Nucleotidase. 5'-Nucleotidase activity was measured by the release of 32P,from AM32P using adsorption to 20% Norit as described previously (30). RESULTS Nucleotide Levels in Normal and CLL Lymphocytes. Typical chromatograms of extracts obtained from normal and CLL lym phocytes are shown in Chart 1 (A and B). The Chromatographie conditions described by Brown ef al. (13, 25), with minor modi fications, were able to adequately separate most ribonucleotides present in extracts from human lymphocytes within 50 min. The absolute retention times were found to vary depending on the efficiency of the particular column in use, but the relative retention times and elution order were within ±5%. Fourteen nucleotides were clearly identified and quantitated. As expected, the ribo nucleotides present in highest concentration in extracts of normal and CLL lymphocytes were ATP and ADP. Also shown as a major UV absorbing component is ascorbic acid as has been described previously (18). The ribonucleotide levels of normal lymphocytes are shown in Table 1. The results obtained in earlier studies from other labo ratories (11, 29, 34, 35) are shown for comparison. Good agree ment was found with respect to the levels of ATP, DTP, and AMP, while slightly lower values for CTPand GTP were obtained. The reason for the latter discrepancy is probably the more accurate correction of the absorbance contributed by the buffer in the present study. Table 2 shows the results obtained with lymphocyte extracts from normal subjects and patients with CLL. CLL lymphocytes have significantly lower levels of ATP, CTP, UTP, and CDP. The concentration of NAD was significantly higher in CLL lympho cytes. While it was not possible to distinguish particular nucleosides and bases, the aggregate area of the peaks eluting in this region was significantly lower in CLL lymphocytes. The adenine and guanine energy charge did not differ between CLL and normal unfractionated lymphocytes, while the total adenine, cytidine and uridine nucleotide content was significantly lower in CLL lymphocytes. Nucleotide Levels in Enriched Subpopulations of Lympho cytes. Lymphocytes from patients with CLL have an altered ratio of B/T cells. For this reason, the nucleotide levels were determined in preparations enriched in Band T-lymphocytes. Table 3 shows the nucleotide levels in Band T-lymphocyteenriched preparations from normal and CLL donors. The Tenriched normal lymphocytes had higher levels of NAD and UDPA/-acetylglucosamine compared to normal B-lymphocytes, while all of the other ribonucleotide levels were the same. B-cells from patients with CLL were markedly higher than those of their Tcell counterparts in NAD, while their other ribonucleotides did not differ. When fractionation into populations of T-helper and Tsuppressor cells was carried out, similar ribonucleotide levels were found in these subsets (Table 4). Comparison of Nucleotides from B-enriched Normal and CLL Lymphocytes. Table 5 shows the nucleotide levels of Benriched preparations from normal donors and patients with CLL. Normal B-lymphocytes contained higher levels of ATP, GTP,