With the aim of questioning the apparent loss of specificity of the microsomal glucose-6-phosphate phosphohydrolase after detergent-treatment, we performed competitive inhibition experiments among the four best substrates of the enzyme, i.e. the 6-phosphates of glucose (Glc6P), mannose-6 (Man6P), glucosamine (GlcN6P) and 2-deoxyglucose (dGlc6P). The Km and Vmax of glucose-6-phosphatase (Glc6Pase) and mannose-6-phosphatase (Man6Pase), assayed either by complex formation determination of P(i) produced or by radiometric determination of [U-14C]Glc or [U-14C]Man, were very close to 1 mM and 0.64 mumol.min-1.mg-1 microsomal protein, respectively. The Km of the enzyme for GlcN6P and for dGlc6P, determined by colorimetric assay of P(i), were equal to 1.53 +/- 0.07 mM and 2.35 +/- 0.15 mM, respectively, whilst the Vmax was not different from that of Glc6Pase and Man6Pase. Unexpectedly, the Ki of Man6P (1.61 +/- 0.22 mM), GlcN6P (2.24 +/- 0.17 mM) and dGlc6P (3.40 +/- 0.07 mM) for Glc6Pase, assayed by liberation of [U-14C]Glc, were significantly (50%) higher than their Km previously determined. The Ki of Glc6P (0.66 +/- 0.05 mM) for Man6Pase, assayed by liberation of [U-14C]Man, was significantly lower than its Km previously determined. In contrast, the Ki of GlcN6P (1.55 +/- 0.05 mM) for Man6Pase, assayed by the radiometric assay, was not different from its Km previously determined. It can be inferred from these data that Glc6P phosphohydrolase exhibits specific behaviour towards Glc6P after the detergent-treatment of the microsomal membrane.