Cancer Letters.

(1991)

255

255 - 262

Elsewer Scientific Publishers Ireland Ltd.

Monitoring of urinary excretion of modified nucleosides in cancer patients using a set of six monoclonal antibodies C. Reynauda,

C. Brunoa,

P. Boullangerb,

J. Grange’,

S. Barb&id

and A. Niveleaua

‘Laboratoire de Genie Enzymatique Umr 106, *Laboratoire de Chimie UA 463, CNRS-Uniuersite Lyon I Batiment 308, 43 Boulevard Du I1 Nouembre 1918, F-69622 Villeurbanne, ‘INSERM Unite 271. 151 Cows Albert Thomas. F 69423 Lyon, Cedex 03 (France) and dBio-Rad SPD Clinical Division, 66 Via Morandi, Segrate, Milan0 20090 (Italy) (Received 8 August 1991) (Revision received 1 October 1991) (Accepted 10 October 1991)

Keywords: modified nucleosides; cretion; monoclonal antibodies

Summary Monoclonal antibodies were produced and characterized in order to allow the monitoring of the urinary excretion of six modified nucleosides. The specificity of each antibody was determined and competitiue solid-phase enzyme-linked immunoassays were designed, the sensitivity of which lay in the pmol range. Detection and quantitation of 5methylcytidine (5-MeCyd), 4-acetylcytidine (4-AcCyd), 1 -methylinosine (1 -Melno), l-methyladenosine (1 -MeAdo), 7-methylguanosine (7MeGuo) and pseudouridine ($-Urd) can be performed in small volumes (70 ~1) of untreated urine. Results can be obtained from as many as 20 different samples, for one molecule, within 3 h. With this technique, values observed for three commonly measured nucleosides in urine from healthy subjects (II/Urd, l-MeAdo and I-Melno) are in good agreement with those reported by other authors after analysis by high performance liquid chromatography. Results obtained in urine from cancer patients show significantly increased leuels of the six haptens quantitated by this immunoassay. Correspondence

to:

sion, 66 Via Morandi,

0304.3835/92/$05,00

A. Niveleau, Segrate.

0

1992

Printed and Published in Ireland

Bio-Rad

SPD

Milan0 20090,

Clinical Divi-

Italy.

urinary ex-

Introduction Modified bases and nucleosides are tRNA breakdown products which cannot be rephosphorylated and reused for de nova synthesis. Therefore these molecules are excreted by cells and can be found in serum and urine [22]. Tumor-bearing animals and cancer patients excrete these catabolites in higher amounts than healthy subjects [4,5,11,25, 30 - 32,34,36,38 - 411 as shown through the use of chromatographic methods [2,12,13,35]. The excretion of modified nucleosides can also be measured by immunological methods based on the pioneering work of Erlanger and Beiser [9]. The possibility of raising antibodies against nucleosides thus provides an alternative to the biochemist. Since it has been determined by Muller et al. [23], for various malignancies, that more than one nucleoside has to be monitored in the evaluation of a patient if one wants to reach a high level of confidence, the production of a set of antibodies to rapidly detect several such haptens with a high degree of specificity was a desirable goal. The other prerequisites were: a high sensitivity, the possibility of performing the assay on crude

El sevier Scientific Publishers Ireland Ltd

256

samples and of using enzyme-labelled rather than radio-labelled tracers. Previous success achieved in the past in the production of polyclonal sera against modified nucleosides [1,7,8] prompted us to produce a set of six monoclonal antibodies directed against l-MeAdo, 7-MeGuo, 1-MeIno, ~-ACCyd, 5-MeCyd and rl/-Urd, respectively. The properties of these antibodies enabled the detection and quantitation of the six corresponding haptens in a very small volume (70 ~1) of untreated urine in a competitive solid-phase enzyme-linked immunoassay. Preliminary experiments on urine from healthy subjects and cancer patients demonstrated the feasibility of the method. These results are reported here. Materials and Methods Preparation of conjugates Conjugates were prepared according to the method of Erlanger and Beiser [9] for 5-MeCyd, 7-MeGuo, l-MeIno, l-MeAdo and 4-AcCyd, with the following modifications: reduction was performed with sodium cyanoborohydride, at O°C for 2 h only while the pH was continuously monitored and maintained below 9.5. Spectral analysis showed neither degradation of the purine ring nor deacetylation of 4-AcCyd under these conditions. Pseudouridine was oxidized to pseudouridylic acid and linked to bovine serum albumin as in [3]. Conjugates were dissolved in isotonic saline at a final concentration of 10 mg/ml. For each conjugate the amount of nucleoside linked per mole of protein was estimated according to the method of Inouye et al. [ 181. Molar ratios between hapten and carrier were 8.0, 5.6, 1.9, 7.4, 17.6 and 22 for 5-MeCyd, 4-AcCyd, l-MeIno, 1-MeAdo, 7-MeGuo and $-Urd, respectively. Immunization BALB/c mice (IFFA-Credo, France) were immunized according to the following protocol: on day 1, 100 pg of conjugate was

emulsified in an equal volume of complete Freund’s adjuvant and injected intraperitoneally. At day 12 the same amount of conjugate was injected in incomplete Freund’s adjuvant through the same route. The immune response was evaluated by the dot-blot method of Hawkes et al. [ 141 and further injections at 12-day intervals were repeated if necessary. A final boost was performed 3 days before fusion by injecting 100 pg of conjugate intraperitoneally without Freund’s adjuvant. Production of hybridomas and screening Splenic lymphocytes were fused with Sp2/0-Ag 14 plasmacytoma cells [29], in a ratio of 4:1, using 50% polyethylene glycol 4000 [lo], generally following the method described by K6hler and Milstein [21]. The fused cells were resuspended in the selective (hypoxanthine-, azaserine-supplemented) medium and seeded in 24-well tissue culture plates. After 2 - 3 weeks hybridoma supernatants were tested for the presence of antibodies directed against the conjugate by performing dot-blots on nitrocellulose. Positive cultures were cloned by limiting dilution in 96-well plates. After 4 weeks, positive clones were transferred to 24-well trays. These cultures were recloned, the final clones were expanded and the antibodies produced were characterized with a monoclonal antibodyidentifying kit (Zymed Laboratories Inc., CA, U.S.A.). Clones selected in this study were those showing the highest specificity, determined as descibed below. Determination of the specificity of antinucleoside antibodies Polystyrene microplates were coated with modified nucleoside-carrier conjugates and the specificity of each antibody for the corresponding hapten was determined [27]. Briefly, supernatant (100 ~1) from hybridoma cultures was added to each well in the absence or in the presence of increasing amounts of various structurally related haptens. Plates were incubated for 1 h at room temperature and washed. The binding of monoclonal antibodies

257

was detected after incubation for 1 h with peroxydase-labeled anti-mouse Igs. The readings were done on a microplate reader (Bio-Rad 3550 MR) at 655 nm. The same method was used to calibrate the assay and determine the respective sensitivities for the six haptens. Results 0

of the rnonoclonaf

antibodies Class and subclass of immunoglobulins secreted by each relevant clone were hIgG1 for 5-MeCyd and 4-AcCyd; fdgG1 for 7-MeGuo; l-Memo and 1-MeAdo and KIgGza for $-Ed. The specificities of antibodies towards each hapten were determined by comparing the concentration of various structural analogs necessary to inhibit the binding of monoclonal antibodies to the corresponding conjugate by 50%. These results are summarized in Table I. No cross reactivity was observed between each antibody and the five other conjugates. Characterization

Antibodies

Hapten

to

antibodies

to modified

0.060 0.0016” 0.50 0.098 0.0077b 0.200

Values indicate the amount tal conditions described in “The amount of 7-MeGua bThe amount of 3-MeAde

of competing Materials and necessary to necessary to

1

OF COMPETITOR

10

100

(in pg/mll

nucleoside. The standard curves obtained for the six nucleosides are shown in Fig. 1. These results are summarized in Table II. The reproducibility of the method was assessed by performing triplicate assays with the same sample (urine or control-specimen) at various sites on the same microplate. The inter-well coefficient of variation was lower than 5%. Day to day variations for the same sample never exceeded 8%.

nucleosides.

Non-modified Nucleoside

5-MeCyd 7-MeGuo +Urd l-Memo I-MeAdo 4-AcCyd

.l

Fig. 1. Calibration curves of immunoassays. Inhibition curves were obtained for each conjugate by incubation of hybridoma’s supernatants with the correponding hapten as decribed in Materials and Methods.

and standardization of immunoassays The inhibition observed after adding increasing amounts of free hapten to the supernatants of hybridoma culture media allowed a useful range to be determined for each assay and the detection limit to be calculated for each

of monoclonal

I

.Ol

CONCENTRATION

Sensitivity

Table 1. Specificity

I ,001

11 (Cyd) 8.3 (Guo) 92 (Urd) 24 (Ino) 17 (Ado) 1150 (Cyd)

analog Base 53 (Cvt) 4.6 (Gus) 225 WYP) 110 (Ade) 775 (Cyt)

molecules (in nmol) necessary to inhibit the reaction by 50% in the experimenMethods. inhibit this reaction, was 0.04 nmol. inhibit this reaction was 28 nmol.

258

Table II. Nucleosides

Sensitivity

of the assays Detection

Useful range

limit

kg/ml)

5-MeCyd 7-MeGuo $-Urd l-Memo 1-MeAdo 4-AeCyd

0.6-0.15 0.016 - 0.004 10-2.5 0.6-0.15 0.5-0.12 5- 1.25

in pg/ml

in pmol/assay

0.078 0.002 1.25 0.078 0.006 0.0625

21 0.47 360 19 1.5 0.15

Assays were performed as described in Materials and Methods in polystyrene microplates coated with conjugates. Aliquots of 70 ~1 of standard solutions containing increasing amounts of nucleosides were mixed with monoclonal antibodies and distributed in each well. The final volume of the assay was 300 ~1. Values shown here were obtained as in Fig. 2.

Quantitation of modified nucleosides in urine Urine samples were obtained from nine healthy adult volunteers. Three samples were collected daily at regular intervals, during 4 consecutive days. Assays were performed on 703.J aliquots of urine. Three dilutions of each aliquot were used to calculate the concentration of each hapten. Urinary creatinine was measured using a calorimetric reaction with a diagnostic kit obtained from Sigma (France).

Table 111. Urinary excretion Nucleosides

of modified

Controls n = 61

5-MeCyd

23.5

zt 4.7

nucleosides

0.09

f

0.05

21.4

zt 6.92

ORL cancers n = 14

27.5

35.5 f 33.8 P < 0.01 0.18 f 1.4 P