| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Neuroendocrine Unit (Z.W., M.B., S.C.F., C.J.S.) and I. Frauenklinik-Innenstadt (P.B., B.S.), Medizinische Klinik-Innenstadt, Klinikum der Ludwig Maximilians University, 80336 Munich, Germany; and Department of Medicine and Obstetrics/Gynecology, University of Virginia (S.E.K.), Charlottesville, Virginia 22908
Address all correspondence and requests for reprints to: Christian J. Strasburger, M.D., Medical Clinic, University Hospital Innenstadt, Ludwig Maximilians University, Ziemssenstrasse 1, 80336 Munich, Germany. E-mail: cjs{at}medinn.med.uni-muenchen.de.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These results, obtained by a new, highly sensitive hGH-V-specific immunoassay, highlight the importance of maternal hGH-V in the regulation of maternal and fetal IGF-I. In addition, the results indicate that maternal weight has a major impact on circulating concentrations of hGH-V.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To investigate the physiological functions of hGH-V, monoclonal antibodies (mAbs) specifically recognizing hGH-V without cross-reactivity to hGH-N and hPL are extremely important tools. Their availability would allow for a specific hGH-V antagonist in cell culture or animal models. Furthermore, only immunoassays based on mAbs are expected to measure hGH-V specifically without the interference of closely related proteins. Until today, only one hGH-V-specific assay has been reported (9, 10), and the data generated using this method have contributed significantly to our understanding of hGH-V physiology. However, the mAbs used for this assay have been generated against a nonglycosylated form of hGH-V expressed in Escherichia coli (11). The same nonglycosylated form of hGH-V is used as the calibrator for the assay. In contrast, native hGH-V in the maternal circulation consists of both nonglycosylated and glycosylated isoforms (12, 13). This difference could impair the quantification of total hGH-V in circulation. Furthermore, this commercial assay is limited by its radioactivity, overnight incubation, and a minimal sample volume of 200 µl. Moreover, the sensitivity of the method is 0.2 µg/liter, which limits its suitability in early pregnancy or in cell culture systems, when low concentrations of hGH-V are expected.
In this paper we describe the production of new, high affinity mAbs against hGH-V using recombinant hGH-V expressed in mouse fibroblasts, which exhibits a glycosylation pattern identical to that found in humans. The epitopes of these mAbs were identified, and a nonisotopic, highly sensitive, hGH-V-specific sandwich immunoassay was established. Using this assay, the time course of hGH-V secretion was determined in a prospective clinical study in normal pregnancies.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tween 20, Tween 40, BSA, polyethylene glycol, bovine 
-globulin, Hunter’s TiterMax adjuvant, and biotin-amidocaproate-N-hydroxysuccinimide ester were purchased from Sigma-Aldrich (St. Louis, MO). Fast protein liquid chromatography equipment, Sepharose-r-protein A, Hyperbond enhanced chemiluminescence membrane, Hyperfilm, horseradish peroxidase-conjugated antimouse IgG, and enhanced chemiluminescence kit were purchased from Amersham Pharmacia (Uppsala, Sweden). Sodium dodecyl sulfate, Tris, acrylamide, TEMED, Mini-protein II Dual Slab Cell, and Trans-Blot Semi-Dry Electrophoretic Transfer Cell were purchased from Bio-Rad Laboratories, Inc. (Richmond, CA). Rabbit antimouse IgG (Z0109) was obtained from DAKO Corp. (Glostrup, Denmark), and DMEM, newborn calf serum, horse serum, and Protein Free II medium were purchased from Life Technologies, Inc. (Eggenstein, Germany). Cellmax Artificial Capillary Cell Culture cartridge was obtained from Cellco, Inc. (Germantown, MD). MiniPERM devices were purchased from Heraeus (Hanau, Germany), and the IgG-subclass kit was obtained from Pierce Chemical Co. (Rockford, IL). Recombinant hGH (IRP 88/624) was obtained from the National Institute for Biological Standards (Hertfordshire, UK). hGH-CV, a recombinant hGH mutant, was provided by Dr. Pär Gellerfors (Pharmacia \|[amp ]\| Upjohn, Inc., Stockholm, Sweden). hGH-CV has nine amino acid residues on the C-terminal end different from hGH-N, but identical to hGH-V. At the N-terminal end, four amino acid residues are different from hGH-V, but identical to hGH-N. The full-length GH receptor extracellular domain (amino acids 1–246) was a gift from Prof. A. Gertler (Rehovot, Israel). The DELFIA 1232 time-resolved fluorometer was produced by Wallac, Inc. (Turku, Finland). Ninety-six-well, flat-bottom microtiter plates (Maxisorp plates 442404) were obtained from Nunc (Rosklide, Denmark). The commercially available hGH-V assay was purchased from Biocode S.A. (Liege, Belgium).
Production and purification of hGH-V
Recombinant hGH-V was expressed in mouse fibroblasts as described previously (14). To increase both the concentration and quantity of GH-V-conditioned medium, the fibroblast cells were prepared for transfer to the Hybridoma Facility at University of Virginia by serial reductions of National Institute for Biological Standards over several weeks. Cells were then grown in a Cellmax Artificial Capillary Cell Culture cartridge in hybridoma medium. This cartridge system supports up to 1 x 1011 cells and produces a high yield of protein that is secreted into the medium. The medium was initially purified on an immunoexchange column using the mAb 6F1 with comparable affinity to both hGH-N and hGH-V. Columns were prepared by linking anti-hGH antibody to cyanogen-bromide Sepharose beads. Twenty micrograms of antibody were added to 100 µl buffer solution containing 0.1 M NaHCO3 (pH 8.3), 0.5 M NaCl, and 40 µl washed beads. hGH-V was eluted from the column with 0.5 M glycine and was neutralized with Tris base. The identity of GH-V from this source has been confirmed by mass spectrometric analysis (performed by Dr. Jay Fox, Biomolecular Core Laboratory, University of Virginia).
To further examine the products of the mouse fibroblasts, 10 µl supernatant were heated in reducing buffer for 4 min at 95 C and loaded to a 15% SDS-PAGE gel. The protein was blotted to a nitrocellulose membrane using semidry electrophoretic transfer. The membrane was incubated with an anti-hGH mAb 6F1, which has a comparable affinity to both hGH-N and hGH-V. After incubation with horseradish peroxidase-conjugated antimouse IgG, protein bands were detected by enhanced chemiluminescence. For large scale production, hGH-V was purified from supernatants through an affinity column using mAb 6F1. The purity of the hGH-V preparation was tested by SDS-PAGE, followed by Coomassie Blue staining.
Production and purification of mAbs against hGH-V
Two-month-old female BALB/c mice were immunized with recombinant hGH-V dissolved in Hunter’s TiterMax adjuvant and injected intradermally (10 µg antigen/mouse). After 4–6 months of repeated immunization, the mice with the highest serum titers were killed, and spleen cells were fused with NSO cells in the presence of polyethylene glycol using the hybridoma technique. Cells were grown in medium containing 20% horse serum. Hybridoma cell supernatants were screened for hGH-V-binding activity after 10–12 d of culture using biotinylated hGH-V. Hybridoma cells corresponding to the supernatants giving the highest signals were cloned at least three times by limiting dilution. The IgG subclass of the mAbs was determined, and large scale production was carried out in Mini PERM devices in protein-free medium. The IgG concentration of the supernatant was 1–5 g/liter. The mAbs were affinity purified using an r-Protein A fast protein liquid chromatography column; pooled IgG-containing fractions were extensively dialyzed against PBS, divided into aliquots, and stored at -20 C until use.
Biotinylation of mAbs, hGH-N, and hGH-V
The purified mAbs were biotinylated as described previously (15) using a 75-fold molecular excess of the labeling reagent (biotin) in the reaction. hGH-N and hGH-V were biotinylated using the same method, but with a 25-fold molar excess of biotin-amidocaproate-N-hydroxysuccinimide ester. Aliquots of purified biotinylated mAbs, hGH-V and hGH-N, were stored at -20 C in 50 mM Tris-HCl buffer (pH 7.8), containing 0.1% BSA and 0.02% sodium azide.
Selection and characterization of the mAbs
To select mAbs against hGH-V and without cross-reaction with hGH-N, the mAbs were screened with both biotinylated hGH-N and hGH-V. Only mAbs with high affinity to hGH-V and no or very low affinity to hGH-N were selected for expansion and subsequent epitope mapping. The precise cross-reaction pattern of the mAbs was studied by competitive binding assays. The degree of cross-reaction was estimated from the amount of competitor required to evoke 50% inhibition of binding of biotinylated hGH-V by the respective mAb. The epitopes of these hGH-V-specific mAbs were identified by examining their cross-reaction with hGH-CV. Furthermore, the relationship between the respective epitopes and receptor interaction site 1 of hGH-V was studied. Anti-hGH-V mAbs were immobilized on microtiter plates and saturated with recombinant hGH-V. Subsequently, biotinylated recombinant hGH receptor ectodomain (hGH-binding protein, rhGHBP) was added. The amount of the formation of hGH-V/rhGHBP complexes was quantified after addition of europium-labeled streptavidin and measurement by a time-resolved fluorometer.
Time-resolved fluorescence immunoassays
For all assays, microtiter plates were coated with antibodies diluted in phosphate buffer (50 mmol/liter, pH 9.6) by overnight incubation at 4 C. Phosphate buffer was used because previous studies have shown that the presence of phosphate, rather than carbonate, ions leads to a higher absorption capacity of the microtiter plates for mAbs. Washing buffer was prepared freshly for all experiments (50 mmol/liter Na2HPO4, 0.05% Tween 20, and 0.0025% NaN3, pH 7.5). Detection antibodies were diluted in assay buffer [50 mmol/liter Tris-(hydroxymethyl)-aminomethane, 154 mmol/liter NaCl, 20 µmol/liter diethylenetriaminepenta-acetic acid, 0.01% Tween 40, and 0.05% NaN3, pH 7.75]. BSA (0.5%) and bovine -globulin (0.05%) were added to combat nonspecific binding.
End-point detection was the same for all assays. After incubation with a biotinylated tracer at room temperature, the microtiter plates were washed three times with 0.3 ml wash buffer. Europium-labeled streptavidin (10 ng) was added to each well and incubated (30 min). After a 6-fold washing step, the addition of 0.2 ml enhancement solution to each well, and a final incubation (15 min) on a horizontal plate shaker, the signal was read using the DELFIA time-resolved fluorometer.
hGH-V-specific immunometric assay
Using a pair of the hGH-V-specific mAbs, a highly sensitive, sandwich-type immunofluorometric assay (hGH-V specific IFMA) was established. Ninety-six-well microtiter plates were coated with 500 ng mAb 3F12 in 0.2 ml phosphate buffer (50 mM, pH 9.6) by overnight incubation. After washing, 0.025 ml standards or samples were pipetted into each well, followed by 50 ng biotinylated anti-hGH-V mAb 8A9 in 0.175 ml assay buffer. Plates were sealed and incubated for 2 h at room temperature. After this step, the plates were processed as described below for all time-resolved fluorescence immunoassays.
Assay calibrators
Recombinant hGH-V produced in mouse fibroblasts as described above is used as the calibrator for the hGH-V-specific assay. The concentration of the hGH-V preparation was estimated by a competitive assay using rhGHBP as the binder and biotinylated recombinant hGH-N (88/624) as the tracer, because it is known that rhGHBP has a comparable affinity to both hGH-N and hGH-V (14). The hGH-V calibrators were produced by serial dilution in sheep serum.
Immunofunctional hGH assay, hGH-N-specific assay, and commercial hGH-V-specific assay
For comparison, three additional hGH assays have been used: the immunofunctional assay for GH (hGH-IFA), an hGH-N-specific IFMA, and the commercially available hGH-V assay [Biocode hGH-V-specific immunoradiometric assay (IRMA)]. The hGH-IFA, which is calibrated against recombinant hGH-N (IRP 88/624), equally recognizes both hGH-V and hGH-N and was performed as described previously (16). The hGH-N-specific IFMA was established by choosing mAbs not cross-reacting with hGH-V (17) and is calibrated against recombinant hGH-N (IRP 88/624). Despite the different antibodies and calibrators, the assay was performed as described above for the hGH-V-specific IFMA. The commercially available hGH-V assay was used according to the manufacturer’s instructions. This assay is calibrated against recombinant hGH-V expressed in E. coli as a calibrator. In our laboratory the intra- and interassay variabilities for all three assays were less than 8%. For the clinical study all samples from one subject were analyzed within the same run of the respective assay.
IGF-I and leptin assays
IGF-I levels were analyzed using the Nichols Advantage IGF-I Chemiluminescence Assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). Intra- and interassay coefficients of variation were less than 6.5%. Because of the significant differences in age between the participating pregnant women (range, 20–42 yr), IGF-I results were expressed as a multiple of the upper limit of normal of the method-specific, age-adjusted reference ranges for nonpregnant women. Leptin concentrations were measured by a monoclonal in-house IFMA as described previously (18). Intraassay coefficients of variation were 6.3% and 4.2% at concentrations of 1 and 10 µg/liter, respectively. Interassay coefficients of variation were 8.3% and 5.8% at the same concentrations. In both assays, all samples from one subject were analyzed within the same run.
Clinical study
Serum samples were obtained from a prospective, longitudinal study including 84 healthy women monitored during normal pregnancies. After obtaining informed consent, blood samples were drawn during visits between 0800 and 1100 h at the out-patient clinic every 4–6 wk until delivery. Smoking habits, age, medical history, and parity were recorded. Prepregnancy body weight and height were obtained from medical records. According to the NIH Consensus Development Panel, 6 women had a body mass index (BMI) below the normal range (<19 kg/m2), 66 were within the normal range (19–25 kg/m2), 10 were overweight (25–30 kg/m2), and 4 were obese (>30 kg/m2). For analysis, overweight and obese women were treated as one group. Demographic and anthropometrical data for the participating mothers and the newborns are summarized in Table 1
. At each visit, body weight was measured using a standardized balance at the out-patient clinic. One day after delivery, a final sample was taken. In two cases, samples were obtained frequently immediately before, during, and after parturition. Serum samples were centrifuged and stored at -20 C until assayed. The study includes 518 samples from gestational week (GW) 7 onward, with an average of 6.2 samples per pregnancy. Between GW 20 and 24, a glucose challenge test or a glucose tolerance test was performed according to published recommendations (19) to identify patients suffering from gestational diabetes. Birth weight and height of the newborns were recorded and compared with percentiles obtained from a large cohort of newborns of comparable ethnic background (20). Only samples from normal pregnancies leading to a healthy normal singleton appropriate in weight for gestational age were included in the analyses presented here. In 21 cases, cord blood samples were obtained. The study was approved by the ethics committee of the Medical Faculty of Ludwig Maximilians University (Munich, Germany).
|
Linear regression analysis and nonparametric tests (Spearman correlation, Mann-Whitney U test) were carried out when appropriate using the StatView software program (version 5.0, SAS Institute, Inc., Cary, NC). All values are given as the mean ± 1 SD unless otherwise noted. To show the overall variability in hGH-V concentrations, Fig. 6 shows hGH-V results from all samples analyzed during the study (n = 518). For all other analyses, only one sample from each subject at a certain time point was used.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant hGH-V was expressed in mouse fibroblasts as described previously (14). No hGH-N was detectable in cell culture supernatant as revealed by the hGH-N-specific IFMA. The purity of the hGH-V preparation after purification through an affinity column using anti-hGH mAb 6F1 was more than 90%, as shown by SDS-PAGE gel and Coomassie Blue staining. In both gel staining and Western blot using mAb 6F1, which has a comparable affinity to both hGH-N and hGH-V, two different hGH-V isoforms were identified, with molecular masses of 22 and 25 kDa, respectively (Fig. 1, lane 5). The latter is consistent with the previously reported glycosylated form of hGH-V. In contrast, the recombinant hGH-N preparation (88/624) expressed in Escherichia coli, and the recombinant hGH-N preparation expressed in mouse fibroblasts both consist of a single hGH isoform (22 kDa; Fig. 1, lanes 1–4).
|
From 25 clones exhibiting high affinity to hGH-V, 2 clones had no detectable or minimal cross-reaction with hGH-N. These clones were selected, expanded, and characterized. Displacement experiments showed that mAb 3F12 has no detectable cross-reaction (<0.01%) with both hGH-N and hPL up to a concentration of 300 nM (Fig. 2A). mAb 8A9 exhibited only minimal cross-reaction with hGH-N at high concentrations (0.41%), but not with hPL (Fig. 2B). With respect to their epitopes, the mAbs can be divided into two spatially distinct groups. mAb 8A9 interferes with hGH-V binding to rhGHBP, indicating that the epitope is inside binding site 1 of the hGH-V molecule. In contrast, the formation of hGH-V/rhGHBP complexes is not affected by the presence of anti-hGH-V mAb 3F12, indicating that this mAb targets an epitope outside the hormone/receptor interaction site 1. Furthermore, mAb 8A9 does not bind the mutated molecule hGH-CV. As hGH-CV differs from hGH-V in only 4 amino acid residues at the N-terminal end, mAb 8A9 seems to have an epitope in this region (see Fig. 3A) corresponding to helix 1 of the hGH-V molecule (H18R, H21Y, Y28F) and the connection loop between helix 1 and helix 2 (P37L). In contrast, mAb 3F12 binds to both hGH-V and hGH-CV. Together with the lack of interference with hGH-V/rhGHBP complex formation, this implicates an epitope outside helix 1, helix 4, and the connection loop between helixes 1 and 2. Looking at the 3-dimensional structure of hGH-V, only the hGH-V-specific amino acid residues at helix 2 (F92L), helix 3 (D112R, L113H, G126W), and the connection loop between helixes 3 and 4 (K140N, T142S, N149K) remain as possible targets of mAb 3F12. Within the 3-dimensional structure, these amino acid residues are located close together and form a patch that most likely is the epitope of mAb 3F12 (see Fig. 3B). The characterization of the mAbs is summarized in Table 1
.
|
|
Using mAb 3F12 as the capture antibody and biotinylated mAb 8A9 as the detection antibody, a highly sensitive IFMA specific for hGH-V was developed. The required sample volume is only 50 µl serum for a duplicate determination, and the total duration of the assay is less than 3 h. The assay has a linear working range from 0.02–25 µg/liter. The lower detection limit, as calculated from the concentration corresponding to the mean signal of a 20-fold measurement of zero standard plus 2 SD, is 0.01 µg/liter, and the lower limit of quantification, as defined by the lowest value measurable with an intraassay variability of less than 15%, is 0.02 µg/liter. The intraassay coefficients of variation are 2.8% and 4.0% (n = 20) at concentrations of 5 and 0.5 µg/liter, respectively. Interassay coefficients of variation at the same concentrations are 5.5% and 9.3%, respectively (n = 8). The linearity of the assay is 110 ± 4%, as revealed by measuring serum samples at 2- to 32-fold serial dilution in zero standard (n = 6). Linear regression analysis showed coefficients of correlation of 0.99 or greater, indicating a good linearity. To determine accuracy, five different concentrations of recombinant hGH-V in assay buffer were spiked to serum samples (n = 6). The theoretical hGH-V concentration for each sample was used to calculate recovery, which was 103 ± 4% (range, 92–112%).
To investigate a possible interference from hGH-N or hPL in this assay, serum samples containing measurable amounts of hGH-V (n = 3) were spiked with recombinant hGH-N (1, 10, 20, 50, and 100 µg/liter) and hPL (1, 10, 100, 1,000, and 10,000 µg/liter). After an overnight preincubation, the samples were assayed by the hGH-V-specific IFMA. No change in hGH-V levels was detectable even at the highest concentrations of hGH-N and hPL added.
As hGHBP is known to interfere in many assays for hGH determination, and furthermore, binding proteins frequently vary in their concentration throughout pregnancy, we investigated the susceptibility of the hGH-V-specific IFMA to the addition of rhGHBP. Three serum samples containing hGH-V at concentrations of 0.5, 5, and 10 µg/liter were incubated overnight at 4 C after spiking with rhGHBP (1,000, 2,000, 5,000, 10,000, and 20,000 pM) to allow for hGH-V/rhGHBP complex formation. No change in the hGH-V concentration measured was observed up to rhGHBP concentrations of 20,000 pM, which is clearly above the hGHBP concentrations seen in human serum samples.
hGH-V concentrations as measured by the specific IFMA, compared using an indirect approach
Before hGH-V-specific mAbs became available, the quantification of hGH-V was achieved by an indirect approach. Samples were measured by an assay measuring both hGH-N and hGH-V and by an hGH-N-specific assay. The difference between the assays gives an estimate of the hGH-V concentration present in a sample. For this indirect approach, we used the IFA (16), which equally recognizes hGH-N and hGH-V, together with the hGH-N-specific IFMA described previously (17). Comparison of the results obtained by this indirect approach with those obtained by the hGH-V-specific assay revealed a close agreement (r2 = 0.99; P < 0.0001; Fig. 4A).
|
We compared our hGH-V-specific IFMA to the commercial hGH-V IRMA (Biocode-IRMA). This assay uses the hGH-V-specific mAb E8 as the capture antibody and another hGH-V-specific mAb as detection antibody. Recombinant nonglycosylated hGH-V expressed in E. coli is used as standard. Despite the differences in mAbs and calibrators, the Biocode-IRMA and the new hGH-V-specific IFMA show a close agreement (Biocode-IRMA, -0.48 + 1.03 x hGH-V-specific IFMA; r2 = 0.93; Fig. 4B), at least in the concentration range where both assays can adequately quantify hGH-V concentrations. However, the sensitivity of the Biocode IRMA (0.2 µg/liter) is 1 order of magnitude lower than that of the hGH-V-specific IFMA (0.02 µg/liter).
Clinical findings
Maternal hGH-V concentration in our series of normal pregnancies shows a constant increase (Fig. 5). The high sensitivity of the hGH-V-specific IFMA allowed us to detect low, but measurable concentrations of hGH-V in samples from early pregnancy (GW 7). Thereafter, levels steadily increased, with the main increase observed between the GW 20 and 30. One day after parturition, hGH-V levels were undetectable. Mean concentrations of hGH-V increased from 0.9 ± 0.5 µg/liter (GW 7–13; range, 0.02–1.7 µg/liter) to 2.8 ± 0.9 µg/liter (GW 18–22; range, 1.4–5.5 µg/liter), 7.3 ± 2.6 µg/liter (GW 28–32; range, 2.0–14.9 µg/liter), and 13.0 ± 9.6 (GW 37–41; range, 2.1–69.8 µg/liter). The increase in hGH-V occurring between GW 18–23 and GW 28–32 was significantly correlated to the further increase occurring between GW 28–32 and GW 37–41 (r2 = 0.455; P = 0.04). The peak hGH-V concentration was achieved at GW 36.4 ± 2.6 wk, corresponding to 3.4 ± 2.7 wk before parturition. In a subgroup of 43 women, 2 consecutive samples were obtained at GW 35 and immediately before delivery. Mean concentrations of hGH-V did not change significantly. In 17 cases, concentrations of hGH-V decreased; in 16 cases, hGH-V concentrations remained unchanged; and in 10 cases, an increase was observed.
|
). Measurement of leptin revealed that the increase in hGH-V between GW 18–22 and GW 28–32 was negatively correlated to the increase in leptin within the same period (r2 = 0.114; P = 0.027), and that hGH-V at GW 28–32 was negatively correlated to leptin concentrations (r2 = 0.053; P = 0.012). hGH-V concentrations correlated with maternal IGF-I from GW 18 onward (GW 18–22: r2 = 0.118; P = 0.039; GW 28–32: r2 = 0.179; P < 0.001; GW 37–41: r2 = 0.1; P = 0.013). Furthermore, the increase in hGH-V from GW 18–22 to GW 28–32 and the increase in hGH-V from GW 28–32 to GW 37–41 were both correlated to the increase in IGF-I observed in the same time periods (r2 = 0.165; P = 0.014; and r2 = 0.168; P = 0.037, respectively).
In addition, the cord blood IGF-I concentration was strongly correlated to peak hGH-V concentrations (r2 = 0.505; P = 0.025) and to the increase in hGH-V between GW 28–32 and GW 37–41 (r2 = 0.475; P = 0.013; Fig. 7A). Cord blood IGF-I concentrations correlated negatively with prepregnancy BMI of the mothers (r2 = 0.107; P < 0.0001; Fig. 7B). We found a weak, but significant, correlation between peak hGH-V concentrations and child height at birth (r2 = 0.078; P = 0.017). In our study neither peak hGH-V concentrations nor the increase in hGH-V between GW 28–32 and GW 37–41 was correlated to the weight of the newborn.
|
). In agreement with the known half-life time of pituitary hGH (15–18 min), hGH-V levels fell to values below the detection limit within 3 h.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Predicted from the amino acid sequence, hGH-V has a potential N-glycosylation site at positions 140–142 (13). Because the enzymatic equipment of bacterial cells differs from that of eukaryotic cells (21), one can expect that a preparation of recombinant hGH-V produced in E. coli lacks the Nglycosylation pattern characteristic for the hGH-V molecule produced by mammalian cells and found in the maternal circulation during pregnancy. Therefore, we used an hGH-V preparation produced in mouse fibroblasts for generation of the mAbs. As demonstrated by Western blotting, this preparation consists of two molecular isoforms, most likely corresponding to a glycosylated (25-kDa) and a nonglycosylated (22-kDa) hGH-V isoform as described previously (22). The same preparation was used to calibrate the new hGH-V specific assay, because it is highly desirable that a calibrator of an assay is identical to the analyte. The differences in the antibodies as well as in the standard preparation explain why the absolute concentrations of hGH-V reported here might differ slightly from the concentrations reported by others. Of course, an international standard preparation for hGH-V is highly desirable to allow comparable measurements of hGH-V concentrations. The IFMA developed using the new hGH-V-specific mAbs (3F12, 8A9) is highly sensitive (0.02 µg/liter), making it uniquely possible to measure hGH-V concentrations at early stages of pregnancy. Our findings suggest that hGH-V is present in the maternal circulation as early as GW 7. Although these low concentrations may not have a major impact on the modification of maternal metabolism, hGH-V may have an important autocrine or paracrine role in early placental development. In addition, the increased sensitivity of this assay will allow us to study hGH-V secretion under pathological conditions in early pregnancy, such as extrauterine gravidity and blastocystic moles.
The main increase in concentrations of hGH-V took place between GW 20 and 30, and peak concentrations of hGH-V were reached around GW 36. Thereafter, concentrations of hGH-V remained constant, fell, or increased slightly until parturition. This time course of hGH-V secretion corresponds to the known changes in glucose metabolism during pregnancy characterized by a decrease in insulin sensitivity in normal women and an increase in insulin dose requirements in diabetic women (23, 24). Although this coincidence does not prove a causal role, further studies should be performed to elucidate the role of hGH-V in maternal glucose metabolism. In addition, it remains to be clarified whether hGH-V might be a useful marker of placental viability in complicated pregnancies. In contrast to other studies (25, 26), we did not observe any differences in hGH-V concentrations between women bearing males or females. The study by Coutant et al. (25) reported data from samples drawn immediately before parturition, a time point not included in our study. Furthermore, their cohort, despite being of comparable size, included 30 smoking women, whereas our cohort only includes 4 smokers. These differences might contribute to the conflicting results. On the other hand, the study by Chellakooty et al. (26) included a much larger number of pregnancies (n = 455), and the observed mean difference in hGH-V between women bearing males and females was 4.9% after correction for confounding factors. This difference might be too small to be reproducible in our series of only 84 pregnancies.
In nonpregnant subjects, pituitary GH secretion is blunted in obese subjects (27). Furthermore, recent reports found that hGH-V is negatively influenced by body weight between GW 24 and 29 (28) as well as by prepregnancy BMI around GW 28 (26). In our study we found that circulating concentrations of hGH-V were negatively correlated to prepregnancy BMI around GW 20 and 30 and at the end of pregnancy (GW 37–41). In addition, we found significantly higher levels of hGH-V in underweight women at GW 18–22, 28–32, and 37–41. Conversely, hGH-V concentrations were lower in overweight/obese women in late pregnancy. These data support the hypothesis, that hGH-V, similar to hGH-N in nonpregnant subjects, is negatively regulated by the adipose tissue. The findings that the increase in hGH-V between GW 18–22 and GW 28–32 is negatively correlated to the increase in leptin within the same period and that circulating concentrations of hGH-V are negatively correlated to leptin concentrations at GW 28–32 is in agreement with previous reports (25) indicating that leptin might be one of the factors mediating adipose tissues effects on GH secretion.
hGH-V concentrations showed a close correlation to maternal IGF-I levels from GW 18 onward. Furthermore, the increase in hGH-V from mid- to late pregnancy was closely correlated to the increase in IGF-I during the same period. This adds further evidence to the previous reports (7, 29) indicating that hGH-V is an important regulator of maternal IGF-I concentrations during pregnancy. In agreement with the lower concentrations of hGH-V in overweight/obese women, cord blood IGF-I was negatively correlated to prepregnancy BMI. The correlation between maternal hGH-V and cord blood IGF-I as well as height at birth suggest a regulatory role of hGH-V not only for the maternal, but also the fetal, metabolism. However, whether this is a direct or an indirect effect remains to be investigated.
In our series of normal pregnancies, 17 of 84 women at the end of pregnancy showed peak IGF-I levels above the age- and method-adjusted reference range for nonpregnant women. We conclude that a high IGF-I level during pregnancy does not necessarily represent a pathological condition.
In conclusion, high affinity, specific mAbs have been produced against recombinant, mammalian hGH-V and characterized by epitope mapping. Using these mAbs in an IFMA, the lower detection limits for hGH-V determination could be significantly improved, whereas the sample volume required for analysis could be reduced. The characteristics of the assays allow a reliable, simple, rapid, and nonisotopic measurement of hGH-V. Together with the recently described hGH-N-specific assay (17), it provides a useful tool to study the physiological and pathological roles of both hGH-N and hGH-V during pregnancy. Our clinical data further support the hypothesis that hGH-V is an important regulator of maternal metabolism and, in consequence, of fetal development. The relationship between maternal BMI and circulating hGH-V is of importance when investigating the possible role of hGH-V in metabolic disorders associated with pregnancy, such as gestational diabetes. Although hPL is believed to be the major mediator of insulin resistance in pregnancy (30), the concept of a metabolic role of hGH-V is strongly supported by the recently published data on severe insulin resistance in hGH-V transgenic mice (31). However, further studies are necessary to discriminate between the effects of hGH-V on maternal metabolism and the effects of maternal adipose tissue on hGH-V.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: BMI, Body mass index; GW, gestational week; hGH-N, human pituitary GH; hGH-V, human placental GH; hPL, human placental lactogen; IFA, immunofunctional assay; IFMA, immunofluorometric assay; IRMA, immunoradiometric assay; mAb, monoclonal antibody; rhGHBP, recombinant human GH-binding protein.
Received May 21, 2002.
Accepted November 18, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. R. Brian, M. Bidlingmaier, M. P. Wajnrajch, S. A. Weinzimer, and S. E. Inzucchi Treatment of Acromegaly with Pegvisomant during Pregnancy: Maternal and Fetal Effects J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3374 - 3377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Fuglsang, P. Sandager, N. Moller, S. Fisker, H. Orskov, and P. Ovesen Kinetics and secretion of placental growth hormone around parturition. Eur. J. Endocrinol., March 1, 2006; 154(3): 449 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ericsson, B. Hamark, N. Jansson, B. R. Johansson, T. L. Powell, and T. Jansson Hormonal regulation of glucose and system A amino acid transport in first trimester placental villous fragments Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2005; 288(3): R656 - R662. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Fuglsang, F. Lauszus, A. Flyvbjerg, and P. Ovesen Erratum J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 5042 - 5042. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Fuglsang, F. Lauszus, A. Flyvbjerg, and P. Ovesen Human Placental Growth Hormone, Insulin-Like Growth Factor I and -II, and Insulin Requirements during Pregnancy in Type 1 Diabetes J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4355 - 4361. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
