Phosphorylated insulin-like growth factor binding protein 1 is increased in pregnant diabetic subjects

During pregnancy, IGFs and their binding proteins (IGFBPs) are important fo r the growth of fetal and maternal tissues. IGFBP-1 normally circulates as a single, highly phosphorylated species (hpIGFBP-1). However, in pregnancy there are lesser phosphorylated isoforms (lpIGFBP-1) with decreased affinit y for IGF-I, allowing for increased IGF bioavailability. Because regulation of IGFBP-1 is abnormal in type 1 diabetes, we examined the impact of this on IGFBP-1 and its phosphorylation status in diabetic pregnancy. We assesse d IGFBP-1 in relation to birth weight, maternal weight gain, duration of di abetes, glycemic control, and the presence or absence of retinopathy in 44 diabetic and 11 nondiabetic subjects. We found that in type I diabetic pati ents there was a significant negative relationship between hpIGFBP-1 and bi rth weight (r = -0.42, P < 0.01) and between the ratio of hpIGFBP-1 to lpIG FBP-1 and birth weight (r = -0.38, P = 0.02) by week 18 of gestation. Multi ple regression analysis confirmed that hpIGFBP-1 was the best single predic tor of birth weight (R-2 = 0.3, P = 0.001) in diabetic subjects using model s including other parameters known to influence fetal size. In contrast to hpIGFBP-1 levels, lpIGFBP-1 levels were not associated with birth weight, b ut were significantly related to initial maternal BMI and maternal weight t hroughout gestation in diabetic subjects (r = -0.57, P < 0.001). hpIGFBP-1 levels were positively related to duration of diabetes (r = 0.38, P < 0.01) . Diabetic subjects had significantly higher hpIGFBP-1 and lpIGFBP-1 levels than nondiabetic subjects (hpIGFBP-1: 215 +/- 21 vs. 108 +/- 13 mu g/l, P = 0.01; lpIGFBP-1: 139 +/- 12 vs. 66 +/- 5 mu g/l, P < 0.001), but the rati o of hpIGFBP-1 to lpIGFBP-1 was similar in both groups (2.1 +/- 0.3 [diabet ic] vs. 1.7 +/- 0.2 [nondiabetic], NS). In summary, maternal IGFBP-1 levels were higher in diabetic than in normal pregnancies. Diabetic subjects with prolonged duration of diabetes and retinopathy had higher total IGFBP-1 le vels than those with shorter disease duration. Thus hpIGFBP-1 in diabetic p regnancy is positively related to the duration of diabetes and inversely re lated to fetal growth, with lpIGFBP-1 being related to maternal weight and BMT. The ratio of hpIGFBP-1 to lpIGFBP-1 may be a more robust indicator of fetal outcome, since it was consistent between diabetic and nondiabetic sub jects. Measurement of the different phosphorylated isoforms of IGFBP-1 may increase the usefulness of IGFBP-1 as a predictor of fetal growth in both n ormal and diabetic pregnancy.