IN-VITRO BIOASSAY WITH ENHANCED SENSITIVITY FOR HUMAN GRANULOCYTE-COLONY-STIMULATING FACTOR

A method for the determination of human granulocyte colony-stimulating factor (hG-CSF) activity, based on stimulation of cellular proliferat ion, was developed using a subclone of the murine myeloid leukemia cel l line NFS-60, with an improved sensitivity for hG-CSF, as indicator. The optimal range for quantitative analysis of hG-CSF was about 4-60 p g ml(-1). The stimulatory effect was measured by a colorimetric microa ssay: the optical density of formazan, which is produced by viable cel ls from (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MT T), was obtained by reading plates in a multi-channel photometer. The assay was designed as a five-dose parallel line test, employing three or four doses for potency determinations, which fulfil pharmacopoeial requirements for assay validity. Inter-assay relative standard deviati on (RSD) varied between 5.2 and 12.0%. Most assay experiments revealed potencies within limits of error of 90-110% and the mean index of pre cision value was 0.057. The recently developed yeast cell-derived Inte rnational Standard (88/502) served as a reference for activity of rhG- CSF. Specificity of the assay was demonstrated by absence of response upon exposure to a panel of biomolecules, including recombinant human interleukin-3, and by the suppression of growth stimulation in the pre sence of neutralizing anti hG-CSF antibodies. Potency readings of ungl ycosylated rhG-CSF were dependent on pH of assay medium with higher re lative activities observed at pH 6.6 than at 7.4. Moreover, SOS-PAGE a nalysis of the carbohydrate-deficient preparation, following incubatio n at physiological pH, revealed several high molecular weight rhG-CSF bands and decreased monomeric form. The method described was found sui table for potency assessments of pharmaceutical formulations of hG-CSF .