This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lindsay, R. S. Articles by Johnstone, F. D. Search for Related Content PubMed PubMed Citation Articles by Lindsay, R. S. Articles by Johnstone, F. D.

Insulin and Insulin Propeptides at Birth in Offspring of Diabetic Mothers

MedStar Research Institute (R.S.L.), Washington, D.C. 20010; Diabetic Department (J.D.W.), Royal Infirmary of Edinburgh NHS Trust, Edinburgh EH3 9YW, United Kingdom, Department of Clinical Biochemistry (I.H., C.N.H.), Addenbrooke’s NHS Trust, Cambridge CB2 2QR, United Kingdom, and University Department of Obstetrics and Gynaecology (A.A.C., B.A.H., F.D.J.), Centre for Reproductive Biology, University of Edinburgh, Edinburgh EH3 9ET, United Kingdom

Address all correspondence and requests for reprints to: Robert Lindsay, MedStar Research Institute, 108 Irving Street NW, Washington, D.C. 20010. E-mail: robert.lindsay{at}medstar.net.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Maternal diabetes during pregnancy is associated with excess fetal growth and increased fetal insulin production. We hypothesized that insulin propeptides (proinsulin and 32–33 split proinsulin) might be more robust indicators of chronic fetal overproduction of insulin. We examined insulin-like molecules in cord blood (ILM) (insulin, proinsulin, and 32–33 split proinsulin) in relation to birth weight, maternal glycemia, and cord glucose in 140 offspring of mothers with type 1 diabetes (ODM) and 49 offspring of mothers who did not have diabetes (CONTROL) as well as degradation of ILM in response to sampling conditions at birth. Insulin propeptides were abundant in cord blood, comprising 50% of ILM in CONTROL and 36% in ODM (P < 0.0001) and more resistant to degradation than insulin (P < 0.05). Concentrations of all three ILM were highly intercorrelated with median values 2- to 5-fold higher in ODM than CONTROL [e.g. median (range): insulin ODM 110 (60–217) pmol/liter; CONTROL 22 (15–37) pmol/liter; P < 0.0001]. In ODM, 32–33 split proinsulin and proinsulin were more closely related to birth weight (Spearman r for ILM: r_{32–33 split}= 0.54; r_PROINSULIN: r = 0.54; r_INSULIN = 0.40: r_{32–33 split} and r_PROINSULIN > r_INSULINP < 0.05) and fetal leptin (r_{32–33 split}= 0.55; r_PROINSULIN; r = 0.54; r_INSULIN = 0.22: r_{32–33 split} and r_PROINSULIN > r_INSULINP < 0.05) than insulin). By contrast, insulin was more closely related to cord glucose (r_{32–33 split} = 0.15; r_PROINSULIN: r = 0.10; r_INSULIN = 0.42: r_INSULIN > r_{32–33 split} and r_PROINSULINP < 0.05). In CONTROL, 32–33 split proinsulin was also more closely related to fetal leptin r_{32–33 split}= 0.61; r_PROINSULIN: r = 0.29; r_INSULIN = 0.33: r_{32–33 split} > r_INSULINP < 0.05). In ODM, 32–33 split proinsulin and proinsulin have closer relationships to fetal growth and leptin concentrations at birth than insulin. Measurement of insulin propeptides may be advantageous in assessment of the influence of maternal hyperglycemia on the newborn.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
MATERNAL DIABETES DURING pregnancy results in profound changes in fetal growth, at least in part as a result of increased fetal insulin production. Fetal hyperinsulinemia has been detected in offspring of diabetic mothers both in utero [by sampling of amniotic fluid (1, 2, 3)] and immediately after birth [by sampling of cord blood (4)]. Furthermore, the presence of maternal diabetes during pregnancy influences long-term health of offspring with increases in risk of obesity (2, 5), type 2 diabetes (6), and impaired glucose tolerance (3, 7, 8) in later life. Importantly, this later risk appears to be related to the extent of metabolic abnormality in the fetus as measured by fetal insulin production; increased fetal amniotic fluid insulin concentrations are predictive of both later impaired glucose tolerance (3, 8) and obesity (2). Fetal insulin concentrations are therefore potentially important indicators of the metabolic effect of maternal diabetes on the fetus and later health of the child.

Insulin is produced by enzymatic processing of precursor peptides. In the major pathway, intact proinsulin is cleaved to 32–33 split proinsulin, further processed by removal of a dipeptide fragment to des-31,32 proinsulin, before eventual processing to insulin and C-peptide. In the minor pathway (important only in the presence of insulin processing defects), intact proinsulin may be cleaved to form 65–66 split proinsulin further processed to des-64,65 proinsulin before final production of insulin and C-peptide (9). The assay used in this study detects both intermediate products in the major pathway (32–33 split proinsulin and des-31,32 proinsulin), and for simplicity we refer to this immunoreactivity as 32–33 split proinsulin throughout (10, 11). In adults, insulin propeptides comprise around 12% of circulating insulin-like molecules and are almost entirely made up of intact proinsulin and 32–33 split proinsulin (11, 12). In the fetus such circulating propeptides comprise up to 40% of circulating insulin-like-molecules (13, 14) and again are almost entirely intact proinsulin and 32–33 split proinsulin (14).

Propeptides may be important for two reasons. First, where nonspecific insulin assays are used, cross-reactivity will result in inaccurate estimations of insulin concentration. Second, because propeptides are cosecreted with insulin but have different biological characteristics, having longer half-lives (15, 16, 17), and being poorer substrates for insulin degrading enzymes (18), there is the potential that they might reflect excess fetal insulin production more accurately than insulin itself. In studies of pregnancies not complicated by diabetes, relationships of 32–33 split proinsulin to birth weight are stronger than those of insulin to birth weight (14).

We wished to investigate insulin propeptides in cord blood samples from offspring of diabetic and control pregnancies. We examined the hypotheses that 1) insulin propeptides would comprise a substantial proportion of circulating insulin-like molecules in offspring of diabetic mothers, as is the case in offspring of nondiabetic mothers; 2) propetides might be different in their resistance to degradation; and 3) propeptides might offer a better index of chronic overproduction of insulin by the fetus than insulin as assessed by relationships to maternal glycemia, birth weight, and leptin concentration.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Recruitment and clinical protocol

Recruitment began in January 1999 and ended in May 2001 and took place in eight hospital-based antenatal centers in Scotland. All women with type 1 diabetes were invited to join the study. Two hundred ninety-five women with type 1 diabetes attended and planned to deliver at these centers, of whom 265 were considered initially eligible for the study (4 mothers of twins and 26 mothers who had early miscarriages or were planning terminations of pregnancy were not considered eligible). A total of 250 women consented to take part in the study (94%) in whom cord blood samples were obtained in 200 (80%). Of the 50 cases in which cord blood was not obtained, in four cases difficulty in obtaining cord specimens was recorded (in three cases with retention of the placenta), in seven cases blood was not collected because of intervening medical problems before or at delivery (one intrauterine death, three terminations of pregnancy related to fetal abnormalities, one maternal death, one abruptio placentae, one shoulder dystocia). In the remaining 39 cases, failure to obtain specimens did not relate to maternal or fetal problems.

Because sample hemolysis results in increased degradation of insulin (19, 20, 21) and pilot studies (see below) indicated significant (>10%) degradation of insulin in which samples were either not collected from cord for more than 20 min or there was delay between sample collection and freezing for more than 60 min, samples not fulfilling these criteria were excluded from the main data analysis. Thus, the 200 samples were further restricted to those in whom 1) there was no evidence of hemolysis of cord blood (17 excluded); 2) cord blood had been collected within 20 min [12 exclusions: (median interquartile range) collection time for remaining samples, 2 min (1, 2, 3, 4, 5, 6, 7)]; 3) cord blood had been centrifuged and plasma frozen within 60 min [17 exclusions: time from collection to freezing for remaining samples, 17 min (11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26)]; 4) antenatal glucocorticoids had not been administered in the 24 h before birth (15 excluded); and finally 4) children delivered at 33 wk or longer (five excluded). A final total of 140 samples from offspring of diabetic mothers were analyzed (some samples being in more than one exclusion category). No mothers had major preexisting medical problems. Review of charts revealed that 27 mothers (19.3%) had been hospitalized during pregnancy because of hypertensive problems (9 pregnancy-induced hypertension, 18 preeclampsia). During pregnancy, diabetic retinopathy (DR) was recorded in 34% of women with type 1 diabetes: 31 background DR (24%), 5 preproliferative DR (4%), 9 proliferative DR or previous treatment for proliferative DR (7%), 9 missing. Abnormal renal function was recorded in 16% of women during pregnancy: 107 normal (84%), 1 microalbuminuria (1%), 20 persistent proteinuria or preeclampsia (16%), 12 missing.

Control mothers were recruited from the same centers. A total of 145 women were invited to take part in the study. Control patients had no history of obstetric or metabolic disease, and routine screening for gestational diabetes was negative. Of the 145 women who gave initial consent, cord samples were attempted in 75 and obtained in 70 (one retained placenta specimen could not be drawn, four sample handling problems). Of these, 70 samples were further restricted to those in whom there was no evidence of hemolysis of cord blood (three exclusions); cord blood had been collected within 20 min [six exclusions: median (interquartile range) collection time for remaining samples, 3 min (2, 3, 4, 5, 6, 7)]; cord blood had been centrifuged and plasma frozen within 60 min [11 exclusions: time from collection to freezing for remaining samples 20 min (12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29)]; antenatal glucocorticoids had not been administered in the 24 h before birth (one excluded); and finally children delivered at 33 weeks or longer (no exclusions). A final total of 49 samples from offspring of nondiabetic mothers were analyzed (some samples being in more than one exclusion category).

All clinics offer antenatal care from a multidisciplinary team comprising obstetricians, diabetologists, specialist midwives, diabetes specialist nurses, and dietetic support. Local management protocols were followed: All mothers with type 1 diabetes were asked to record preprandial glucose three times daily, 1 h postprandial after breakfast, lunch, and the evening meal on 1 d/wk and at 0200–0300 h on 2 d/month.

Hemoglobin A1c (HbA1c) was monitored at individual centers as per local clinical protocols and, in addition, samples were sent in the first trimester, 20 and 30 wk, and, if possible, in the final week of gestation for measurement using a single assay (Variant, nondiabetic reference range 4.4–5.7%, Bio-Rad Laboratories, Inc., Hercules, CA) at a central reference laboratory (Western General Hospital, Edinburgh). Data on clinical outcome including cesarian section, intercurrent medical conditions, and hypertensive conditions of pregnancy were obtained by chart review. Gestational ages were calculated from estimated dates of delivery from chart review. This date was derived from dates of last menstrual period (LMP), where available, or by ultrasound if there was either conflict with dates as assessed by LMP (>6 d) or LMP was unavailable.

Weight was measured at birth and included for offspring born between 33 and 42 wk of gestation. Birth weight has been expressed as an SD score as previously described (22). Individual birth weights (BWT) were compared with mean and SDs derived from 26,000 nondiabetic deliveries in cells specific for gender, week of gestation, and parity (0 vs. 1). The Z score is calculated as (BWT-mean)/SD).

All mothers gave informed consent. Protocols were approved by local ethical committees.

Cord blood sampling, insulin assay, and assessment of stability of insulin and propeptides

After delivery, 20 ml cord blood were collected from the umbilical vein, after cord clamping, into lithium heparin at ambient temperature. Depending on local circumstances, samples were then either transferred to local laboratories for centrifugation or centrifuged in situ, before initial storage of plasma at -20 C and eventual central storage at -70 C [after a median of 11 (5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21) d].

Stability of insulin, proinsulin, and 32–33 split proinsulin to two experimental conditions reflecting mode of collection was examined. First, to assess the effect of delay of sampling from the cord, samples were obtained immediately or after 10, 20, and 30 min, respectively, in four subjects. Samples were processed identically thereafter, being centrifuged and plasma stored at -20 C immediately. Second, the effect of delay in centrifugation and freezing of plasma samples after collection was assessed. Baseline samples were taken from cord (without delay) and then stored for varying times in lithium heparin at ambient temperature before centrifugation and storage at -20 C.

Assays

Plasma, insulin, and cortisol were assayed with 1235 AutoDELFIA automatic immunoassay system using time-resolved fluoroimmunoassays (insulin kit B080–101, cortisol kit B060–101). All reagents, standards, and consumables are those recommended and supplied by the manufacturer (Perkin-Elmer Life Sciences, Wallac Oy, Finland). Intact proinsulin and 32–33 split proinsulin were assayed on the same system, using antibodies as previously described (10). The detection antibody used in the 32–33 split proinsulin assay (CPT-3F11) was donated by DAKO Corp Diagnostics Ltd. (Denmark House, Angel Drove, Ely, Cambridgeshire, UK). Detection antibodies were labeled with Europium using the DELFIA Europium labeling kit 1244–302 (Perkin-Elmer Life Sciences). Standards were supplied by the National Institute for Biological Standards and Controls (intact proinsulin, first international reference reagent 84/611) and donated by Lilly Research Laboratories (Indianapolis, IL) (chromatography purified 32–33 split proinsulin). In two subjects (both controls), insulin concentrations were below the limit of detection of the assay (2 pmol/liter). These data have been included with an assigned value of 1.9 pmol/liter. These assays display low cross-reactivity to other insulin species. The insulin assay shows cross-reactivity less than 0.2% with intact proinsulin, less than 0.5% with 32–33 split proinsulin at concentrations of 2736 pmol/liter and 2800 pmol/liter, respectively. The intact proinsulin assay showed less than 1% cross-reaction with insulin and 32–33 split proinsulin at concentrations of 2500 pmol/liter and 400 pmol/liter, respectively. The 32–33 split proinsulin assay showed 100% cross-reaction with intact proinsulin. To obtain a specific measure of 32–33 split proinsulin, it was necessary to take account of the intact proinsulin concentration of the specimen. Cross-reaction with insulin was less than 1% at 2500 pmol/liter.

Leptin was assayed using an AutoDELFIA method developed at Addenbrooke’s, based on two commercially available monoclonal antibodies (MAB 398 and BAM 398 obtained from R&D Systems Europe, Ltd., Abingdon, UK). Europium-labeled streptavidin allowed the use of time-resolved fluorometry as the detection system. Calibration was with recombinant human leptin (catalog no. 398-LP, R&D Systems).

Glucose was assayed in singleton on the Dimension XL clinical chemistry system (DadeBehring Inc., Newark, DE) using an adaptation of the hexokinase-glucose-6-phosphate dehydrogenase method (23).

Statistical analysis

Data were analyzed using standard software (SAS Institute Inc., Cary, NC). In several cases (insulin, proinsulin, 32–33 split proinsulin), measures were not normally distributed; therefore, log-transformed values were used to approximate normal distributions or compared using nonparametric tests. Simple correlations of insulin-like molecules in cord blood (ILM) to fetal and maternal variables were assessed by Spearman correlation. Differences between control women and women with type 1 diabetes were assessed by unpaired t test or, where further predictor variables were included, general linear models. General linear models were also used to examine the influence of mode of delivery (vaginal vs. emergency cesarean section vs. elective cesarean section) in mothers with type 1 diabetes and mothers who did not have diabetes (CONTROL). The hypothesis that correlation coefficients of the ILM vs. birth weight or leptin were different from those of insulin vs. birth weight or leptin was assessed by standard methods (24). Effects of delay in sampling or freezing specimens were analyzed using repeated-measures ANOVA.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Stability of insulin, proinsulin, and 32–33 split proinsulin

Insulin was markedly less stable to mode of collection than either proinsulin or 32–33 split proinsulin. Thus, delays in both sampling from cord (Fig. 1A) and freezing of plasma (Fig. 1B) resulted in decreases in insulin activity detected by assay. By contrast, both proinsulin and 32–33 split proinsulin appeared to be relatively robust with no significant declines in activity, even when samples were not sampled from cord for up to 30 min (Fig. 1A) and with maintenance of samples at ambient temperature for up to 24 h (Fig. 1B). In light of these results, all following analyses are confined to measures of insulin sampled within 20 min and centrifuged and frozen within 60 min.

View larger version (13K): [in this window] [in a new window]   Figure 1. Influence of delays in either sampling from cord (A) or freezing separated plasma specimen (B). Results represent percentage of baseline values of four specimens. Error bars represent SE at each time point. There is significant fall in insulin values both with delay in sampling from cord (Pfor effect of time = 0.03) and delay in freezing (Pfor effect of time <0.0001). By contrast, there was no significant time effect for samples of proinsulin or 32–33 split proinsulin under either condition.

Mothers with type 1 diabetes were of similar age, parity, and gravidity as controls (Table 1). Offspring of mothers with type 1 diabetes (ODM) were delivered 2 wk earlier than CONTROL but, despite this, were heavier at birth, with markedly increased weight in both males and females after adjustment for gestation at delivery (Table 1). In keeping with this, birth weight in ODM was, on average, 2.0 SDs above that expected in the population and significantly greater than CONTROL (Table 1). ODM were more likely to be delivered by cesarean section (Table 1).

In both CONTROL (P = 0.01) and ODM (P < 0.001), mode of delivery was associated with a significant difference in cortisol concentrations with highest concentrations with vaginal delivery, intermediate with emergency cesarean, and lowest with elective cesarean delivery (Fig. 2, adjusted for birth weight, gestational age at delivery, and sex). Concentrations of leptin, proinsulin, and 32–33 split proinsulin were not significantly associated with mode of delivery. By contrast, concentrations of insulin were significantly higher when the child had been delivered by elective cesarean section (Fig. 2).

View larger version (21K): [in this window] [in a new window]   Figure 2. Influence of mode of delivery (vaginal delivery vs. emergency cesarean section vs. elective cesarean section on hormonal measures at birth. All hormonal measures are adjusted for sex, gestational age at delivery, and birth weight. *, Significant effect of mode of delivery (ANOVA, P< 0.05). , Post hoctesting, P< 0.05 vs. elective cesarean section.

Values of all three ILM (insulin, proinsulin, and 32–33 split proinsulin) were highly intercorrelated; in general, these relationships were strongest between 32–33 split proinsulin and proinsulin and in ODM rather than CONTROL (Table 2). In ODM all three ILM were highly significantly related to BWT, leptin, and maternal HbA1c (Table 2), and in controls 32–33 split proinsulin was significantly related to leptin concentrations and insulin to leptin and BWT (Table 2). When the strength of the relationship of propeptides to maternal and fetal variables were compared, 32–33 split proinsulin was more strongly related than insulin to BWT (in ODM) and leptin (in both CONTROL and ODM). Proinsulin was more strongly related than insulin to BWT and leptin in ODM. By contrast, insulin was more strongly related to cord glucose concentrations in ODM. All three ILM were significantly correlated to maternal HbA1c in ODM (Table 2).

Influence of gender

In general, concentrations of ILM were higher in female than male offspring, and, perhaps surprisingly, this was the case both in CONTROL and ODM. The gender difference was significant for both 32–33 split proinsulin (CONTROL and ODM) and proinsulin (ODM alone) but not for insulin (Fig. 3). Given that female offspring were on average smaller, inclusion of birth weight (and gestational age at delivery) as an additional covariate in models examining the relationship of insulin and gender tended to accentuate the difference in ILM between males and females (data not shown). Plasma glucose was also higher in female ODM (Fig. 3, inset); however, even after inclusion of cord glucose (along with BWT, gestation at delivery, and mode of delivery) to the above model female sex predicted higher concentration of insulin (P = 0.04, CONTROL only), proinsulin (P = <0.01, CONTROL and ODM) and 32–33 split proinsulin (P < 0.001, CONTROL and ODM).

View larger version (25K): [in this window] [in a new window]   Figure 3. Influence of gender on insulin, proinsulin and 32–33 split proinsulin and on glucose (inset). *, P< 0.05 for effect of gender.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Concentrations of insulin propeptides in newborn ODM have not been extensively investigated previously and in some cases may have been limited by then-available methods of estimating propeptides. In studies measuring proinsulin by RIA, it was found to be present in amniotic fluid (1) and cord blood (4) and raised in the presence of abnormal maternal glucose metabolism [gestational diabetes mellitus (4) and type 1 diabetes (1)]. By contrast, measurement of "proinsulin-like components" by a gel filtration method detailed no increase in proinsulin-like components in ODM despite highly significant increases in insulin concentrations (27). These previous studies suggested that proinsulin might be more closely related to BWT than insulin in offspring of mothers with gestational diabetes mellitus (4) but did not investigate 32–33 split proinsulin and did not examine the relationship of maternal glycemic control to specific insulin propeptides.

Propeptides have been examined in detail in cord blood samples in offspring of nondiabetic pregnancies: insulin and 32–33 split proinsulin being significantly related to BWT (with 32–33 split insulin having a closer relationship than insulin), but proinsulin had little relationship to BWT (14). We examined a smaller number of control pregnancies, and in our study the relationship of insulin and 32–33 split proinsulin were of marginal significance (P = 0.03 and P = 0.07, respectively); nevertheless, the pattern is very similar with markedly stronger relationships of insulin and 32–33 split proinsulin than proinsulin in the controls.

We also observed a gender difference in insulin, proinsulin, and 32–33 split proinsulin, with females having higher values. This has been previously reported offspring of nondiabetic mothers (14, 28), but the physiological explanation of this finding remains obscure. It is of interest in this context that the gender difference was present in both offspring of diabetic and nondiabetic pregnancies; thus, the gender-specific modulation of insulin concentrations was not overcome, even in the context of marked stimulation of insulin secretion, as found in the offspring of diabetic mothers.

In adults, proinsulin and 32–33 split proinsulin comprise around 12% of circulating ILM (12). One further molecule may be present, 65–66 split proinsulin. We chose not to assess this because it has previously been found to be present in only very small amounts in cord blood (14). In adults proinsulin and 32,33 proinsulin concentrations have been found to be increased in type 2 diabetes (29) and an increased ratio of proinsulin+32–33 split proinsulin:insulin has been found to be predictive of later type 2 diabetes (30).

In general, concentrations of insulin, 32–33 split proinsulin, and proinsulin are highly related. However, our data also show a strong negative relationship of proinsulin (expressed as a percentage of total-ILM) and the absolute quantity of total-ILM. One interpretation of this is that as insulin secretion increases relatively less proinsulin and more insulin is secreted, but the proportion of 32–33 split proinsulin remains relatively constant. In keeping with this interpretation, ODM as a group have proportionately less proinsulin present. By contrast, the relation of 32–33 split proinsulin to total-ILM and in ODM vs. controls is remarkably constant. Furthermore, we found that 32–33 split proinsulin bears a closer relationship to BWT and measures of maternal glycemia than insulin, whereas in both the control and diabetic groups, insulin was more closely related to cord glucose measures. 32–33 split proinsulin is also more closely related to fetal leptin, which is in turn closely related to fetal adiposity (31). There are a number of potential reasons for this. First, there are major differences in metabolism of insulin and the propeptides. Half-lives in adult plasma are quite different (15, 16, 17), leading to the potential that proinsulin and 32–33 split proinsulin might more accurately reflect chronic hyperinsulinemia. By contrast, the short half-life of insulin probably underpins the closer relationship with cord glucose; thus, insulin reflects short-term metabolic perturbations more accurately. Second, as shown here, both proinsulin and 32–33 split proinsulin are more stable to methods of collection. We tried to minimize the influence of this in our analysis by excluding samples not centrifuged or frozen within standard time intervals. It remains an important practical point, however, in which samples are collected in clinical or research situations for insulin to assess metabolic impact of diabetes on the fetus; either proinsulin or 32–33 split proinsulin may prove more robust measures for this reason. Finally, although concentrations of insulin, proinsulin, and 32–33 split proinsulin are all highly related to one another and clearly all reflect underlying increases in ß-cell function, there may be more subtle differences in ß-cell function and insulin processing reflected by the different species. Proinsulin is processed to insulin by the action of ß-cell prohormone convertase enzymes PC1 (also named PC3- converting proinsulin to 32–33 split proinsulin) and PC2 (converting 32–33 split proinsulin to insulin) during fetal life (9). It is unknown whether maternal diabetes influences activity of these enzymes.

Our results are limited to consideration of the relative quantities of proinsulin, 32–33 split proinsulin, and insulin. Interpretation of this is difficult because differences between ODM and control mothers might reflect differences in the half-lives of the different insulin species, differences in short-term stimulation, and alteration in relative rates of secretion of proinsulin vs. 32–33 split proinsulin vs. insulin. With those caveats in mind, we observed a relative increase in the percentage of insulin and decrease in the percentage of proinsulin in ODM vs. the controls. These differences are not accounted for by differences in cord glucose, BWT, and total insulin secretion leading to the possibility that there are underlying differences in insulin processing between the diabetics and controls. If so, this difference would tend to reduce the quantities of insulin propeptides relative to insulin and thus be closer to the adult pattern. Previous studies in ODM have suggested that concentrations of 32–33 split proinsulin might relate to maternal nutrition with increased energy intake by mother being associated with falls in concentrations of 32–33 split proinsulin at birth (28). If this effect were mediated by energy transfer to the fetus, it might be hypothesized that ODM would have relatively low concentrations of 32–33 split proinsulin; this was not the pattern that we observed.

Maternal diabetes has a major influence on all three ILM. There are a number of other differences between our control group and the ODM in our study. Around 19% of our diabetic mothers experienced hypertensive problems in pregnancy. These mothers were retained in the analysis, although numbers are small; pregnancy-induced hypertension and preeclampsia did not appear to influence ILM at birth (data not shown), compared with other ODM. ODM were also more likely to be born by cesarean section. In controls, cesarean section did not appear to affect concentrations of ILM (data not shown). In ODM, elective cesarean section was associated with higher insulin concentrations. Although this could potentially reflect an effect of case selection for operative delivery, it is notable that concentrations of leptin and the other two ILM were not associated with mode of delivery (after adjustment for BWT). In this light, an alternative explanation arises: that insulin is subject to short-term perturbation by mode of delivery and the leptin and the other ILM are not.

ODM were more likely to have been administered glucocorticoids. We excluded offspring who had been administered glucocorticoids within a day before delivery. In the remaining ODM, there was no significant influence of administration of glucocorticoids after adjustment for other variables (gestation, BWT, gender; data not shown).

In conclusion, we found that although insulin and its propetides are highly intercorrelated in cord blood and all increased in the presence of maternal diabetes, there are also important differences between them. Both proinsulin and 32–33 split proinsulin are more robust to method of collection. The 32–33 split proinsulin appears to have particularly strong relationships to BWT, fetal leptin, and maternal glycemia, supporting the hypothesis that it might more accurately reflect chronic hyperinsulinemia in utero. Measurement of all three insulin species may be advantageous in assessing the impact of maternal diabetes in the newborn.

	Acknowledgments

 
This study was supported by a project grant from the Chief Scientist Office of the Scottish Office. The study would not have been possible without contributions of many people to data collection at the different centers: Aberdeen Maternity Hospital—Dr. Donald W. M. Pearson, Dr. Gordon D. Lang, Dr. Peter Danielan, Dr. David J. Lloyd, Dr. J. Anne Reid, Sister Lesley Mowat, M. S., Fiona MacGregor; Ayrshire Central Hospital—Dr. J. Niall MacPherson, Dr. H. Gordon Dobbie, Dr. D. Andrew W. Collier, Dr. Fiona Barnes, Sister Barbara Stewart, Sister Jean McCulloch, Sister Anne Reid, Sister Ruby Barr; Bellshill Maternity Hospital—Dr. David M. Matthews, D. R. Andrew, D. B. Harrower, Dr. Dina MacLellan, Sister Trish McCue, Sister Pat Sullivan, Sister Sandra Coats, Sister Linda Tallon, Sister Marion Gillen, Midwife Gail Halleron, Mr. Peter Clinton; Glasgow Royal Maternity Hospital, Glasgow—Dr. Angus MacCuish, Dr. C. Burnett Lunan, Dr. Kenneth Paterson, Dr. Layla Alroomi, Sister Tricia Kelly, Sister Liz Winn, Staff Anne Harris, Mr. Ian Thomas; Ninewells Hospital, Dundee—Dr. Graham P. Leese, Dr. Gary J. Mires, Dr. Peter W. Fowlie, Sister Mary Robertson, Midwife Anne E. Sturrock, Midwife Paula Thomson, Dr. Jimmy Burns; The Queen Mother’s Maternity Hospital, Glasgow—Dr. Michael Small, Dr. Alan D. Cameron, Dr. Thomas L. Turner, Sister Brenda Capaldi, Sister Anne Byrne, M.B.E., Midwife Cindy Horan, Midwife Sheila Lee, Dr. Chris Hillier, Mr. Richard Webber; Raigmore Hospital Inverness—Dr. Sandra M. MacRury, Dr. D. A. Russell Lees, Paed. Dr. Ian MacDonald, Dr. Liz McFarlane, Midwife Marie McDonald, Sister Jenny Lobban, Midwife Karen Marnoch, M.S., Angela Veitch; Simpson Memorial Maternity Pavilion, Edinburgh—Dr. James D. Walker, Dr. Hilary D. MacPherson, Dr. Alan Patrick, Dr. John A. McKnight, Dr. Ian A. Laing, Sister Shonagh Battison, Staff Joan Grant, Staff Liz McKay, Midwife Jackie Rankine, Midwife Jacqui Ravie, Midwife Tricia Taylor, Midwife Fiona McNeilage, Midwife Marie MacAulay, Midwife Mhairi Allen, Mr. John Scott, Mr. David Hargreaves.

Additional help and advice was obtained from Dr. Gillian Penney (Scottish Program for Clinical Effectiveness in Reproductive Health), Professor Gillian Raab (Napier University, Edinburgh), and Fiona Tulloch and Brian Misson (Cambridge).

	Footnotes

 
This work was supported by a project grant from the Chief Scientist Office of the Scottish Office.

Abbreviations: CONTROL, Mothers who did not have diabetes; DR, diabetic retinopathy; HbA1c, hemoglobin A1c; ILM, insulin-like molecules in cord blood; LMP, last menstrual period; ODM, offspring of mothers with type 1 diabetes.

Received July 16, 2002.

Accepted January 15, 2003.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Persson B, Heding LG, Lunell NO, Pschera H, Stangenberg M, Wager J 1982 Fetal beta cell function in diabetic pregnancy. Amniotic fluid concentrations of proinsulin, insulin, and C-peptide during the last trimester of pregnancy. Am J Obstet Gynecol 144:455–459[Medline]
2. Silverman BL, Landsberg L, Metzger BE 1993 Fetal hyperinsulinism in offspring of diabetic mothers. Association with the subsequent development of childhood obesity. Ann N Y Acad Sci 699:36–45[Medline]
3. Weiss PA, Scholz HS, Haas J, Tamussino KF, Seissler J, Borkenstein MH 2000 Long-term follow-up of infants of mothers with type 1 diabetes: evidence for hereditary and nonhereditary transmission of diabetes and precursors. Diabetes Care 23:905–911[Abstract]
4. Dornhorst A, Nicholls JS, Ali K, Andres C, Adamson DL, Kelly LF, Niththyananthan R, Beard RW, Gray IP 1994 Fetal proinsulin and birth weight. Diabet Med 11:177–181[Medline]
5. Pettitt DJ, Baird HR, Aleck KA, Bennett PH, Knowler WC 1983 Excessive obesity in offspring of Pima Indian women with diabetes during pregnancy. N Engl J Med 308:242–245[Abstract]
6. Pettitt DJ, Nelson RG, Saad MF, Bennett PH, Knowler WC 1993 Diabetes and obesity in the offspring of Pima Indian women with diabetes during pregnancy. Diabetes Care 16:310–314[Abstract]
7. Plagemann A, Harder T, Kohlhoff R, Rohde W, Dorner G 1997 Glucose tolerance and insulin secretion in children of mothers with pregestational IDDM or gestational diabetes. Diabetologia 40:1094–1100[CrossRef][Medline]
8. Silverman BL, Metzger BE, Cho NH, Loeb CA 1995 Impaired glucose tolerance in adolescent offspring of diabetic mothers. Relationship to fetal hyperinsulinism. Diabetes Care 18:611–617[Abstract]
9. Clark PM, Hales CN 1994 How to measure plasma insulin. Diabetes Metab Rev 10:79–90[Medline]
10. Sobey WJ, Beer SF, Carrington CA, Clark PM, Frank BH, Gray IP, Luzio SD, Owens DR, Schneider AE, Siddle K 1989 Sensitive and specific two-site immunoradiometric assays for human insulin, proinsulin, 65–66 split and 32–33 split proinsulin. Biochem J 260:535–541[Medline]
11. Ostrega D, Polonsky K, Nagi D, Yudkin J, Cox LJ, Clark PM, Hales CN 1995 Measurement of proinsulin and intermediates. Validation of immunoassay methods by high-performance liquid chromatography. Diabetes 44:437–440[Abstract]
12. Hales CN, Barker DJ, Clark PM, Cox LJ, Fall C, Osmond C, Winter PD 1991 Fetal and infant growth and impaired glucose tolerance at age 64. BMJ 303:1019–1022
13. Hawdon JM, Hubbard M, Hales CN, Clark PM 1995 Use of specific immunoradiometric assay to determine preterm neonatal insulin-glucose relations. Arch Dis Child Fetal Neonatal Ed 73:F166–F169
14. Godfrey KM, Hales CN, Osmond C, Barker DJ, Taylor KP 1996 Relation of cord plasma concentrations of proinsulin, 32–33 split proinsulin, insulin and C-peptide to placental weight and the baby’s size and proportions at birth. Early Hum Dev 46:129–140[CrossRef][Medline]
15. Glauber HS, Revers RR, Henry R, Schmeiser L, Wallace P, Kolterman OG, Cohen RM, Rubenstein AH, Galloway JA, Frank BH 1986 In vivo deactivation of proinsulin action on glucose disposal and hepatic glucose production in normal man. Diabetes 35:311–317[Abstract]
16. Zilker TR, Gray IP, Hales CN, Wahl K, Ermler R, Lebender A, Heinzel G, Bottermann P 1988 Pharmacokinetics of biosynthetic human proinsulin following intravenous and subcutaneous administration in metabolically healthy volunteers. Horm Metab Res Suppl 18:37–43[Medline]
17. Tillil H, Frank BH, Pekar AH, Broelsch C, Rubenstein AH, Polonsky KS 1990 Hypoglycemic potency and metabolic clearance rate of intravenously administered human proinsulin and metabolites. Endocrinology 127:2418–2422[Abstract/Free Full Text]
18. Duckworth WC, Bennett RG, Hamel FG 1998 Insulin degradation: progress and potential. Endocr Rev 19:608–624[Abstract/Free Full Text]
19. Sapin R, Ongagna JC, Gasser F, Grucker D 1998 Insulin measurements in haemolysed serum: influence of insulinase inhibitors. Clin Chim Acta 274:111–117[CrossRef][Medline]
20. Reimers TJ, McCann JP, Cowan RG, Concannon PW 1982 Effects of storage, hemolysis, and freezing and thawing on concentrations of thyroxine, cortisol, and insulin in blood samples. Proc Soc Exp Biol Med 170:509–516[CrossRef][Medline]
21. Brodal BP 1971 The influence of haemolysis on the radioimmunoassay of insulin. Scand J Clin Lab Invest 28:287–290[Medline]
22. Johnstone FD, Mao JH, Steel JM, Prescott RJ, Hume R 2000 Factors affecting fetal weight distribution in women with type I diabetes. Br J Obstet Gynaecol 107:1001–1006
23. Kunst A, Draeger B, Ziegenhorn J 1984 UV methods with hexokinase and glucose-6-phosphate dehydrogenase. In: Verlag, Chemie DFL, Bergmeyer HU eds. Methods of enzymatic analysis, vol 6. Basel: VCH; 163–172
24. Kleinbaum DG, Kupper LL, Muller KE, Nizam A 1998 Applied regression analysis and other multivariable methods, 3rd ed. Pacific Grove, CA: Duxbury
25. Pedersen J 1954 Weight and length at birth in infants of diabetic mothers. Acta Endocrinol 16:330–342
26. Freinkel N 1980 Banting Lecture 1980. Of pregnancy and progeny. Diabetes 29:1023–1035[Abstract]
27. Schwartz R, Gruppuso PA, Petzold K, Brambilla D, Hiilesmaa V, Teramo KA 1994 Hyperinsulinemia and macrosomia in the fetus of the diabetic mother. Diabetes Care 17:640–648[Abstract]
28. Godfrey KM, Robinson S, Hales CN, Barker DJ, Osmond C, Taylor KP 1996 Nutrition in pregnancy and the concentrations of proinsulin, 32–33 split proinsulin, insulin, and C-peptide in cord plasma. Diabet Med 13:868–873[CrossRef][Medline]
29. Williams DR, Byrne C, Clark PM, Cox L, Day NE, Rayman G, Wang T, Hales CN 1991 Raised proinsulin concentration as early indicator of beta cell dysfunction. BMJ 303:95–96
30. Wareham NJ, Byrne CD, Williams R, Day NE, Hales CN 1999 Fasting proinsulin concentrations predict the development of type 2 diabetes. Diabetes Care 22:262–270[Abstract/Free Full Text]
31. Schubring C, Siebler T, Kratzsch J, Englaro P, Blum WF, Triep K, Kiess W 1999 Leptin serum concentrations in healthy neonates within the first week of life: relation to insulin and growth hormone levels, skinfold thickness, body mass index and weight. Clin Endocrinol (Oxf) 51:199–204[CrossRef][Medline]

This article has been cited by other articles:

				S. M. Nelson, P. M. Coan, G. J. Burton, and R. S. Lindsay Placental Structure in Type 1 Diabetes: Relation to Fetal Insulin, Leptin, and IGF-I Diabetes, November 1, 2009; 58(11): 2634 - 2641. [Abstract] [Full Text] [PDF]

				S. M. Nelson, D. J. Freeman, N. Sattar, F. D. Johnstone, and R. S. Lindsay IGF-1 and Leptin Associate With Fetal HDL Cholesterol at Birth: Examination in Offspring of Mothers With Type 1 Diabetes Diabetes, November 1, 2007; 56(11): 2705 - 2709. [Abstract] [Full Text] [PDF]

				S. M. Nelson, N. Sattar, D. J. Freeman, J. D. Walker, and R. S. Lindsay Inflammation and Endothelial Activation Is Evident at Birth in Offspring of Mothers With Type 1 Diabetes Diabetes, November 1, 2007; 56(11): 2697 - 2704. [Abstract] [Full Text] [PDF]

				B. M. Shields, B. Knight, H. Hopper, A. Hill, R. J. Powell, A. T. Hattersley, and P. M. Clark Measurement of Cord Insulin and Insulin-Related Peptides Suggests That Girls Are More Insulin Resistant Than Boys at Birth Diabetes Care, October 1, 2007; 30(10): 2661 - 2666. [Abstract] [Full Text] [PDF]

				J. A. Westgate, R. S. Lindsay, J. Beattie, N. S. Pattison, G. Gamble, L. F.J. Mildenhall, B. H. Breier, and F. D. Johnstone Hyperinsulinemia in Cord Blood in Mothers With Type 2 Diabetes and Gestational Diabetes Mellitus in New Zealand. Diabetes Care, June 1, 2006; 29(6): 1345 - 1350. [Abstract] [Full Text] [PDF]

				R. S. Lindsay, A.-G. Ziegler, B. A. Hamilton, A. A. Calder, F. D. Johnstone, and J. D. Walker Type 1 Diabetes-Related Antibodies in the Fetal Circulation: Prevalence and Influence on Cord Insulin and Birth Weight in Offspring of Mothers with Type 1 Diabetes J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3436 - 3439. [Abstract] [Full Text] [PDF]

				R. S. Lindsay, J. D. Walker, P. J. Havel, B. A. Hamilton, A. A. Calder, and F. D. Johnstone Adiponectin Is Present in Cord Blood but Is Unrelated to Birth Weight Diabetes Care, August 1, 2003; 26(8): 2244 - 2249. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lindsay, R. S. Articles by Johnstone, F. D. Search for Related Content PubMed PubMed Citation Articles by Lindsay, R. S. Articles by Johnstone, F. D.