Polymorphism in the 5' flanking region of the human insulin gene in 134 unrelated Japanese subjects 73 with noninsulin-dependent diabetes mellitus (NIDDM) 16 with insulin-dependent diabetes mellitus (IDDM) and 45 controls was investigated using the Southern blot hybridizationmethod. DNA isolated from peripheral blood leukocytes was digested with the restriction enzymes Sac I, Pst I and/or Bgl II and hybridized with a 32P-labeled insuin-DNA-specific probe. The specific probe was a 5' flanking BamHI-BgI II fragment or a Pst I fragment, which contained the polymorphic locus. The results were classified into three groups according the size of the hybridized fragment. The Sac I (Pst I) fragment of the class 1 allele was 5.9 (1.1)±0.3kb in length, the Sac I (Pst I) fragment of the class 2 allele was 6.6 (1.8)±0.3kb in length, and the Sac I (Pst I) fragment of the class 3 allele was greater than 7.0 (2.2) kb in length. The allelic frequency of classes 2 and 3, in total 268 alleles, was 5.6%, while in 116 alleles of NIDDM it was 8.2%. The value was lower than in Caucasians and U. S. blacks, and the results suggested that the 5' flanking insertion was not a genetic marker of the majority of NIDDM in the Japanese. As NIDDM is a heterogeneos disease, further detailed analysis is needed. When the frequency according the family history of diabetes mellitus, obesity and age at onset was compared the allelic frequency of classes 2 and 3 in NIDDM with a family history without obesity tended to be higher than others with NIDDM (P=0.013), and the frequency in NIDDM whose onset was at the age of 39 or less tended to be lower than those whose onset was at 40 or move (P=0.062). On the other hand, the frequency in IDDM was 0%(versus NIDDM P=0.095). These results suggested that we could not completely, exclude the significance of the insertion in NIDDM and possibility of without the insertion related with IDDM.
The measurement of urinary N-acetyl-β-D-glucosaminidase (NAG) activity has been thought to be useful for the early detection of diabetic nephropathy. On the other hand, urinary NAG has been reported to be positively correlated with blood sugar or glycosylated hemoglobin. In order to assess the rapid change of urinary NAG activity in patients with glucose intolerance, its value was measured at vaious times during a 75-g oral glucose toleance test. The ratios of urinary NAG activity to urinary creatinine concentration, that is, NAG index, before and after glucose load, were compared. The mean value before glucose load in the 7 patients tested was 6.8±0.7 U/g·Cr (mean±SE), while the values at 2 and 3 hours after glucose load were 3.2±0.4 and 4.0±0.5 U/g·Cr, respectively. These two values were significantly lower than that before glucose load. The estimation of urinary NAG activity for the early detection of diabetic nephropathy, should be done under some constant condition, for example, the measurement of the daily total amount or at the time or fasting.