Simultaneous Estimation of Rates of Pyrimidine and Purine Nucleotide Synthesis de nouo in Cultured Human Cells*

The requirement of the pathways of pyrimidine and purine nucleotide synthesis de nouo for COz was exploited in a method for simultaneous estimation of the rates of operation of these pathways in cultured human lymphoblasts and fibroblasts. Rates of incorporation of [‘%]bicarbonate into pyrimidine and purine compounds were constant for up to 2 h and were proportional to cell number in the assay. Incorporation rates appeared to reflect rates of synthesis of pyrimidine and purine compounds in individual cell strains under conditions in which: 1) carbamyl phosphate concentrations were less than 0.1% of the hourly flux of labeled bicarbonate into the pyrimidine pathway; 2) specific activities of bicarbonate pools were apparently unchanged. Alterations in rates of [‘%]bicarbonate incorporation during incubation of normal and hypoxanthine-guanine phosphoribosyltransferase (EC 2.4.2.8)-deficient lymphoblasts with purine bases and purine and pyrimidine nucleosides were in accord with previous observations using alternative methods for the individual estimation of rates of purine or pyrimidine synthesis. In addition, comparable increases in [‘4C]bicarbonate incorporation into purines and pyrimidines were demonstrated in human lymphocytes during exposure to phytohemagglutinin, a stimulus previously shown to accelerate rates of pyrimidine and purine synthesis. These findings provided evidence for the validity of the present method in assessing rates of pyrimidine and purine synthesis. High correlations between log phase growth rates of individual lymphoblast lines and their rates of incorporation of label were observed. Although specific and consistent differences were observed in rates of [‘4C]bicarbonate incorporation into pyrimidine and purine compounds in normal, hypoxanthine-guanine phosphoribosyltransferase-deficient, and B-phosphoribosyl I-pyrophosphate synthetase (EC 2.7.6.1) superactive strains, lack of information concerning the specific activities of intracellular bicarbonate pools in different cell strains restricts absolute comparisons of rates of nucleotide synthesis between cell strains.

Among methods described for the study of specific biosynthetic pathways in intact cells, measurement of radiolabeled precursor incorporation into pathway intermediates or end products, or both, has been commonly employed. The rates of operation of the pathways of pyrimidine and purine nucleotide synthesis de nouo in mammalian cells in vitro have been studied individually utilizing ['4C]bicarbonate (1,2) and either ['4C]formate (3)(4)(5) or ["'Clglycine (6, 7), respectively. Shortterm isotopic studies of the rates of these pathways have assumed a close relationship between rates of label incorporation and rates of nucleotide synthesis (l-4,6,7).
The present study describes an isotopic method for the simultaneous estimation of rates of pyrimidine and purine synthesis in individual human lymphoblast lines and fibroblast strains in culture. In the course of applying this method, which depends upon measurement of the rates cf incorporation of the common precursor, ['4C]bicarbonate, into pyrimidine and purine compounds, close correlations between growth rates of individual lymphoblast cultures and their rates of synthesis of these compounds became apparent and are described. In addition, rates of incorporation of label into pyrimidines and purines in normal cells were compared with those in cells bearing specific genetic abnormalities in enzymes of purine metabolism, and these findings, as well as responses of certain of these cultures to chemical and pharmacological agents, are reported. . Purine bases were then eluted from the columns with two 2-ml aliquots of 6 N HCl, which were also pooled (purine compounds) and brought to a pH between 0.1 and 1.0 by addition of 5 N NaOH. into the pyrimidine and purine compounds of cultured human lymphoblasts and fibroblasts were constant for up to 120 min and were linearly related to the number of cells in the assay over ranges of at least 0.1 to 6.0 X 10" cells per ml for lymphoblasts and of at least 0.2 to 3.0 x 10" cells per ml for fibroblasts.
These relationships are exemplified in Fig. 1 for lymphoblast line  in which mean values of 14.2% and 10.5% of the total radioactivity incorporated into pyrimidine and purine compounds, respectively, appeared in the incubation medium. In experiments carried out to assess the possibility that label incorporation by fibroblasts in suspension might differ significantly from that of these cells in monolayer culture, rates of incorporation of ['4C]bicarbonate into the pyrimidine and purine compounds of fibroblasts assayed in suspension differed from those fibroblasts assayed in monolayer by no more than 8% in any study.
The methods employed here for isolation of pyrimidine and purine compounds permit determination of label in all pyrimidine compounds and precursors except carbamyl phosphate, and in all stable purine compounds distal to bicarbonate incorporation into the pathway of purine nucleotide synthesis de nouo,3 measurement of the specific activities of intracellular carbamyl phosphate and bicarbonate pools under the conditions studied would thus allow assessment of the validity of equating rates of label incorporation and nucleotide synthesis de nouo. While direct measurements of these specific activities of these compounds are not currently achievable, the following studies provide indirect evidence to support a close relationship between these rates in individual cell strains. phosphate was less than 0.1% of the hourly flux of labeled bicarbonate (based on the starting specific activity of ['"Clbicarbonate in the incubation medium) into pyrimidine compounds through this pool, suggesting an approximate turnover time of 4 s for the carbamyl phosphate pool. Thus, equilibration of the specific activity of this pool with that of the intracellular bicarbonate pool should be quite rapid, and the critical factor in relating rates of label incorporation and rates of synthesis in both pathways is the specific activity of the bicarbonate pool. The constant rates of incorporation of labeled bicarbonate into pyrimidine and purine compounds during at least 120 min of incubation without additions suggest that the specific activity of the bicarbonate p oo! is relatively constant for this period and that rates of incorporation of label reflect rates of synthesis in this circumstance.
The precise amount of pyrimidine or purine synthesized, however, cannot be stated without information regarding the specific activity of the intracellular bicarbonate pool. Thus, rates of incorporation under these conditions are stated as counts per min per 10" cells per h and provide relative estimates of rates of synthesis of the respective classes of compounds. Evidence to suggest that the addition of the compounds described below which alter rates of label incorporation into one or both pathways do so by altering rates of synthesis rather than changing specific activities of the bicarbonate pool was provided by studies in which the order of addition of label and effector was varied (Table II). Regardless of whether addition of effector preceded addition of label by up to 30 min, followed addition of label by this time period, or the additions were simultaneous, the effects of an individual compound on incorporation of label into the two pathways were nearly constant. This finding held for each of the lymphoblast lines studied. held constant, rates of incorporation of radioactivity remained constant. These findings are consistent with effects of these compounds on rates of synthesis rather than on rates of uptake of or equilibration of label into the bicarbonate pool. In contrast to the apparent validity of the incorporation method in measuring relative rates of synthesis in individual cell lines and strains, differences in rates of incorporation of label into pyrimidine and purine compounds among different Moreover, potential differences between cell strains in the labeling of precursor pools have not been considered in previous studies comparing rates of pyrimidine and purine nucleotide synthesis in cultured cells (3,4,6,7,9,10,13,(22)(23)(24)(25).
Considerable variation was observed in rates of [%]bicarbonate incorporation into the pyrimidine and purine compounds of individual lymphoblast lines studied on multiple occasions over a period of several months. In the course of investigating factors which might account for this variation, it was found that despite apparently unaltered culture conditions, periodic changes in growth rate occurred for each cell line.
However, when each determination of the rates of [%]bicarbonate incorporation into pyrimidine and purine compounds was evaluated in relation to the log phase growth rate of the culture (expressed as the reciprocal of doubling time) at the time of the labeling study, nearly linear relation-ships between growth rate and rates of bicarbonate incorporation into these classes of compounds were established for each cell line, as shown in Fig. 2 for line WI-LB. Using the linear regression coefficients derived from these data, comparative rates of incorporation of label into pyrimidine and purine compounds for the four lymphoblast lines studied were calculated relative to an arbitrary doubling time of 30 h and are presented in Table III. Despite variation in growth rates among the cell lines, rates of incorporation into pyrimidines expressed in this way were similar in normal, hypoxanthineguanine phosphoribosyltransferase-deficient, and adenosine kinase-deficient lymphoblasts.
In contrast, the rate of incorporation into purine compounds in the hypoxanthine-guanine phosphoribosyltransferase-deficient line, Ag'9C135SC1, was more than twice those observed in normal lymphoblasts or in lymphoblasts deficient in adenosine kinase. Rates of incorporation of ['%]bicarbonate into pyrimidine and purine compounds in fibroblast strains are shown in Table  IV. In order to minimize possible effects of differences in passage number or state of culture confluence on rates of bicarbonate incorporation, all fibroblast strains were studied at several points during the log phase of a single growth cycle on at least two occasions between the seventh and eleventh passage in culture. Since much less variation in log phase growth rate was encountered either within individual fibroblast strains or among different strains (Table IV) than was the case for lymphoblast lines, no assessment of a relationship between growth and bicarbonate incorporation rates in fibroblasts comparable to that described for lymphoblasts seemed necessary. On the other hand, the design of these experiments permitted evaluation of the previously described (23) variation in the purine synthetic rates of individual fibroblast strains during the growth cycle. Fluctuation in apparent purine (and pyrimidine) synthetic rates in individual subconfluent fibroblast cultures was confirmed in these bicarbonate incorporation studies, and these variations were found to occur in cultures growing at constant logarithmic rates. In view of these considerations, the comparative rates of incorporation of label into pyrimidine and purine compounds shown in Table IV are expressed as the mean value of determinations made at several points in the growth cycle with an appropriate indication of the variation among determinations. Four fibroblast strains derived from normal individuals were comparable in their rates of incorporation into both pyrimidines and purines. In a fibroblast strain severely deficient in hypoxanthine-guanine phosphoribosyltransferase, the mean rate of incorporation of label into purine compounds was more than 50% greater than the mean rates of this process in the normal strains, while the mean rate of incorporation into pyrimidine compounds in this mutant strain was not clearly increased. However, increased rates of ['4C]bicarbonate incorporation into both pyrimidine and purine compounds were observed in each of the four fibroblast strains derived from patients with excessive PP-ribose-P synthetase activity. Relative rates of ['4C]bicarbonate incorporation into the pyrimidine and purine compounds of lymphoblast lines WI-L2 and Ag'9C135SCl during incubation with selected pyrimidine nucleosides and purine bases and nucleosides are shown in Table V (3,5,22 (6) 0.56 f 0.04 (4) 0.28 k 0.05 (6) 0.44 f 0.09 (6) 0.50 -t 0.10 (6) 0.52 + 0.12 (9) 0.47 rt_ 0.08 (4) 0.93 f 0.06 (4) 0.96 + 0.04 (4) 0.47 3z 0.11 (9) 0.98 f 0.08 (5)   labeled formate incorporation on the other, may reflect differences in transport or equilibration of the respective compounds in the presence of PHA (15,36) rather than differences in the magnitudes of stimulation of purine synthesis. The nearly 8-fold increment in rate of label incorporation into lymphocyte pyrimidine compounds in response to PHA agrees closely with the finding of Ito and Uchino (2) whose estimates of pyrimidine synthetic rate were based on the incorporation of ['4C]bicarbonate into acid-soluble uridine nucleotides.

DISCUSSION
The pathways of pyrimidine and purine nucleotide synthesis de nouo share a requirement for 1 molecule of CO2 for incorporation into the respective base moieties. In addition, few other reactions in cultured mammalian cells fix CO2 into acid-stable products. For these reasons, we have chosen to study the incorporation of ['4C]bicarbonate into pyrimidine and purine compounds as a method for simultaneous determination of the rates of operation of these pathways in individual human lymphoblast lines and fibroblast strains. The present studies which provide evidence for the validity and reproducibility of this method, suggest its potential usefulness in investigations directed at defining the biochemical basis of coordination of these pathways in intact cells. Rates of incorporation of ['4C]bicarbonate into pyrimidine and purine compounds of lymphoblasts and fibroblasts were constant for at least 2 h and were proportional to the number of cells added over wide ranges of cell densities. In addition, in normal cells and in cells deficient in hypoxanthine-guanine phosphoribosyltransferase, rates of bicarbonate incorporation showed the expected patterns of response to addition of compounds previously demonstrated by alternative methods to inhibit purine or pyrimidine, or both, biosynthesis (3,5,20). That rates of incorporation of ['4C]bicarbonate reflect rates of synthesis of pyrimidine and purine compounds in individual cell strains under these conditions is suggested by the demonstration that: 1) carbamyl phosphate concentrations were inconsequential with respect to the flux of label into the pyrimidine pathway in the presence or absence of effecters of incorporation; 2) the specific activity of the bicarbonate pool was apparently unchanged since rates of incorporation were constant with time and, in the presence of effecters, were independent of order of addition of label and effector. Comparison of rates of incorporation and of patterns of distribution of [ 14C]bicarbonate into pyrimidine compounds under base-line conditions and in the presence of uridine, 6azauridine, or oxipurinol lent support to the present method as a valid one for the study of pyrimidine synthesis. The specific patterns of accumulation of pathway intermediates in the presence of 6-azauridine and oxipurinol agreed well with the previous studies showing that metaboiites of these compounds inhibit the activity of orotidylic acid decarboxylase (27-30). Moreover, the increased rates of incorporation of ['4C]bicarbonate into uridylate precursors accompanying blockade of the pathway by these agents bore resemblance to the overproduction of pyrimidine nucleotide precursors in the disorder hereditary oroticaciduria (37) in which orotidylic acid decarboxylase and (usually) orotate phosphoribosyltransferase (EC 2.4.2.10) activities are deficient.
Increased rates of bicarbonate incorporation into the purine and pyrimidine compounds of lymphocytes during exposure to PHA provided an additional line of evidence to support the validity of this method for the estimation of the rates of these pathways. The similar magnitudes of the increments in pyrimidine and purine synthetic rates measured by the current method seem at least in keeping with the requirement for a balanced production of pyrimidine and purine nucleotides for