Oral Presentations / Growth Hormone & IGF Research 22 (2012) S1–S31
Friday 19 October 2012 PL05 Prolactin and tumorigenesis V. Gofﬁn1 . 1 Inserm U845, Research Center Growth and Signaling, Laboratory "PRL/GH Pathophysiology: translational approaches", University Paris Descartes, Faculty of Medicine, Paris, France Prolactin was discovered more than 80 years ago. It is currently viewed by clinical endocrinologists as a hormone of pituitary origin, whose biological actions relate exclusively to lactation and reproductive functions, for which any genetic disorder is yet to be identiﬁed, and whose unique associated pathology is hyperprolactinemia, which is efﬁciently cured using dopamine agonists. There is no debate about these established facts. However, experimental and clinical studies performed during the past decade have considerably broadened our perception of prolactin biology: i. there is now clear epidemiological evidence supporting that circulating prolactin is a risk factor for breast cancer, although the actual mechanisms of action and outcome of prolactin signalling on breast tumorigenesis and disease progression remain a matter of debate (and of intense investigation), ii. in addition to the endocrine (circulating) hormone, locallyproduced prolactin has been documented in various human tissues such as the mammary gland and the prostate, where it is assumed to act via autocrine/paracrine loop; this local mechanism of action may be ampliﬁed in tumor contexts due to prolactin and/or prolactin receptor over-expression, iii. we recently reported the ﬁrst genetic, gain-of-function variant of the prolactin receptor, that may act as an inherited genetic modiﬁer of benign and/or malignant tumorogenesis, iv. we have engineered human prolactin variants acting as pure competitive antagonists of the prolactin receptor, and we have demonstrated their ability to down-regulate the actions triggered by local prolactin as well as by the constitutively active receptor variant in various experimental models; these compounds thus represent a novel class of molecules of therapeutic interest as potential alternative to dopamine agonists. These novel aspects of prolactin biology will be presented and discussed, with major emphasis on experimental models aimed at deciphering the mechanisms of action of autocrine/paracrine prolactin on prostate tumorigenesis. PL06 IGF-II and feto-placental growth M. Constˆancia1 , I. Zvetkova1 , Y. Sekita1 , C. Hammerle1 , W. Cooper1 , I. Sandovici1 . 1 Metabolic Research Laboratories, Department of Obstetrics and Gynaecology, University of Cambridge, Cambridge, United Kingdom Igf2 and Igf2r are major genetic determinants of feto-placental growth in mammals. Their regulation by genomic imprinting (i.e. silencing of one allele depending on parental legacy) ensures that only the paternally derived Igf2 gene is expressed in most fetal tissues, whereas the active Igf2r gene is the maternal one . Igf2r targets Igf2 for lysosomal degradation and therefore suppresses fetal growth. The functional antagonism between maternal Igf2r and paternal Igf2 supports the notion that imprinting evolved as a consequence of a ‘conﬂict’ between the sexes over maternal investment to offspring . We are interested in understanding how Igf2 controls allocation of maternal resources and fetal growth, and how it coordinates whole body growth with the growth of the diverse organs. To study the function of Igf2 in maternal-fetal nutrient allocation we knocked out a placental-speciﬁc Igf2 transcript (P0) in mice .
The P0 mutation leads to an early placental growth restriction phenotype, which is associated with an impaired ability to supply nutrients to the fetus [3, 4]. As a consequence, fetuses are growth restricted in late gestation. However, in mid gestation, we found evidence for compensatory responses of the small P0 mutant placenta, which include increases in transplacental ﬂux of glucose, amino acids and calcium [3, 5]. These studies provide ﬁrst direct evidence for placental adaptive mechanisms that are essential to maintain normal fetal growth trajectories in response to fetal signals of nutrient demand. More recently, we showed that P0 is responsive to a wide range of environmental stimuli, including dietary composition [6–8]. Moreover, the placental adaptations that enhance the activity of System A amino acid-transporters during maternal under nutrition do not occur in the P0 mutants (8). This data suggests that Igf2 P0 acts as a sensor of environmental signals and plays an important role in placental adaptive responses to sub-optimal conditions . To provide a better understanding of the role of Igf2 at the maternal-fetal interface we have recently generated new mouse genetic models based on the Cre-loxP recombinase system. We found that embryonic Igf2 is required to stimulate placental growth and placental vasculogenesis. Other tissue-speciﬁc genetic manipulations (e.g. liver, pancreas), which impact on organ/body size growth, will be described during the talk. The Igf2 locus hosts a recently identiﬁed intronic microRNA (miR483). We provide evidence that miR-483 is an imprinted microRNA that is co-expressed with Igf2 and is under the epigenetic control of the imprinting control region located upstream of the H19 gene. To fully understand the extent by which miR-483 may contribute to growth control, mice with a speciﬁc genetic deletion of this microRNA, which does not interfere with levels of Igf2, were generated. Results of these experiments will also be presented. Our collective work shows that Igf2 is a key genetic regulator of both the supply and demand systems for maternal nutrients in mammals. Modiﬁcations to the balance between the supply and demand have major implications for fetal growth and the intrauterine programming of adult disease. The implications of our ﬁndings using mouse genetic modelling of Igf2 will be discussed in the context of human growth disorders and long-term susceptibility to common diseases. Reference(s)  Constˆancia et al. 2004 Nature 432:53.  Constˆancia et al. 2002 Nature 417:945.  Constˆancia et al. PNAS 2005 102:19219.  Sibley et al. 2004 PNAS 101:8204.  Dilworth et al. 2010 PNAS 107:3894.  Coan et al. 2010 J Physiol 588:527.  Coan et al. J Physiol 2011 589:3659.  Sferruzzi-Perri et al. 2011 Endocrinology 152:3202.  Sandovici I et al. 2012 Reprod Biomed Online 25:68.
OR05: GH and prolactin: Consequences and outcome OR05-1 Cardiovascular risk status in childhood cancer survivors with treated vs. untreated growth hormone deﬁciency A. Petryk1 , K.S. Baker2 , B. Frohnert1 , A. Moran1 , A.R. Sinaiko1 , L.M. Steffen3 , J.L. Perkins4 , L. Zhang3 , J.S. Hodges5 , J. Steinberger1 . 1 University of Minnesota, Pediatrics, Minneapolis, United States; 2 Fred Hutchinson Cancer Research Center, Clinical Research Division, Seattle, United States; 3 University of Minnesota, Division of Epidemiology and Community Health, Minneapolis, United States; 4 Children’s Hospitals and Clinics of Minnesota, Minneapolis, United States; 5 University of Minnesota, Division of Biostatistics, Minneapolis, United States Background: Childhood cancer survivors (CCS) are at increased risk for early mortality from cardiovascular (CV) disease, yet
Oral Presentations / Growth Hormone & IGF Research 22 (2012) S1–S31
the underlying mechanisms remain poorly understood. Growth hormone deﬁciency (GHD) is prevalent among CCS and has been associated with increased CV risk in adults, but the associations between GHD, GH treatment and CV risk factors have not been previously studied in CCS during childhood. Objective: To examine the associations between GHD, GH treatment, and CV risk factors in CCS <18 years old. Design/methods: 301 CCS (164 males, average 14.5, SD 2.6 years) >5 years after cancer diagnosis, and 208 sibling controls (112 males, average 13.6, SD 2.4 years) participated in this crosssectional study, which included anthropometry, body composition (by DXA and abdominal CT), and metabolic studies. GHD was deﬁned as peak stimulated GH level <10 mg/L after clonidine and arginine. Insulin sensitivity (Mlbm ) was measured by euglycemic hyperinsulinemic clamp. Statistical analyses used linear and logistic regression accounting for sibling clustering, adjusted for sex, age, Tanner stage, and body mass index (BMI). P < 0.0083 was considered signiﬁcant to account for comparisons between 6 pairs of groups. Results: 63 CCS had untreated GHD and 24 were on GH treatment at the study visit. Non-GHD CCS did not differ from controls in CV risk factors. Compared to controls, untreated GHD CCS had signiﬁcantly greater adiposity (BMI 25.2±0.7 vs. 20.9±0.3 kg/m2 , percent fat mass 36.3±1.2 vs. 25.2±0.8, waist circumference 79.1±1.6 vs. 68.6±0.8 cm, waist-to-height ratio 0.5±0.01 vs. 0.43±0.005, abdominal visceral fat 34.1±2.3 vs. 20.5±0.9 cm3 ), unfavorable lipid proﬁle (higher total cholesterol 171.7±4.3 vs. 148.1±2.1 mg/dL, higher LDL-C 105.7±3.7 vs. 85.1±1.9 mg/dL, higher triglycerides 109.1±7.1 vs. 76.8±4.0 mg/dL, lower HDL-C 43.0±1.4 vs. 48.1±1.0 mg/dL), and lower insulin sensitivity (M/lbm 11.6±0.6 vs. 14.6±0.4 mg/kg/min, fasting insulin 14.8±1.9 vs. 8.9±0.6 mU/L). After adjusting for BMI, untreated GHD CCS and controls still differed signiﬁcantly for waist-to-height ratio, visceral fat, total cholesterol, LDL-C, and Mlbm . Treated GHD CCS differed from controls only by having higher total cholesterol (174.7±8.4 mg/dL) and lower insulin sensitivity (Mlbm 11.1±1.0 mg/kg/min, fasting insulin 14.3±1.8 mU/L); these differences remained signiﬁcant after adjusting for BMI. Conclusions: This is the ﬁrst study examining the relation between GHD, GH treatment, and CV risk factors in CCS prior to adulthood. Compared to controls, untreated GHD CCS have worse CV risk factors during childhood. Treated GHD CCS have a more favorable body composition than untreated GHD CCS, but continue to show higher total cholesterol and lower insulin sensitivity than controls. These data show that GHD in CCS is associated with heightened risk for CV disease, although mechanisms other than GHD per se may account for reduced insulin sensitivity in this population. OR05-2 Pregnancies and live births in a large cohort of patients on growth hormone replacement therapy 2 ˚ G. Vila1 , A.-C. Akerblad , B.M. Biller3 , M. Koltowska-Haggstr ¨ om ¨ 2, 2 1 1 1 F. Lundgren , M. Riedl , A. Luger . Medical University of Vienna, Division of Endocrinology and Metabolism, Vienna, Austria; 2 KIMS Pﬁzer Endocrine Care, Sollentuna, Sweden; 3 Massachusetts General Hospital and Harvard Medical School, Neuroendocrine Unit, Boston, United States
Introduction: Growth hormone replacement therapy (GHRT) has not been formally approved for use during conception and pregnancy in women who are on GHRT for growth hormone deﬁciency (GHD). To date, there are few data and no guidelines regarding whether to continue or withdraw GHRT during conception and pregnancy. Pregnancy outcomes in these patients have only been evaluated in single center studies with small sample sizes, and with conﬂicting results. Moreover, there are no data on the effect of paternal GH exposure on the evolution and outcome of a partner’s pregnancy.
Methods/design: We searched for pregnancy reports in the Pﬁzer International Metabolic Database (KIMS), a pharmacoepidemiological study of adults with GHD. This database includes information from 4,651 women aged 15 to 50 years (followed for 19,348 patientyears; median 3.10 years) and 7,438 men aged above 16 years (followed for 36,074 patient-years; median 3.89 years). Results: A total of 151 pregnancies were reported from 131 patients, of which 116 were in women (age 22 to 41 years) and 15 were in partners of male patients (men’s age 26 to 68 years). Ovulation induction or assisted reproduction treatments were used in 65% of women; (10% of women and one man had discontinued GHRT before conception). GHRT use in pregnant women was as follows: continuation in 24% of the cases, withdrawal in 56% and dose reduction in 2%. Within the cohort that stopped GH during pregnancy, 46% gave birth to healthy babies, 12% reported fetal loss, and outcome is not reported in 42% of the cases. Within the cohort that continued GH during pregnancy, 41% gave birth to healthy babies, 34% reported fetal loss, and outcome is not reported in 24% of the cases. The category “fetal loss” includes both miscarriages and elective abortions. At the time of abstract submission, birth of 82 healthy children has been conﬁrmed, 20 of them from twin pregnancies and 3 from a triplet pregnancy. Of note, 14 healthy children were born to women who continued GH during conception and pregnancy and 6 with paternal exposure to GH. Conclusions: We provide the ﬁrst demographic data on pregnancy rates in a large cohort of patients receiving GHRT. As might be anticipated given the underlying hypothalamic-pituitary disorders, two-thirds of women were treated to achieve fertility. It appears that in the clinical practice setting, nearly all patients taking GH replacement continue treatment during the time when they seek fertility, and one-fourth continue it during pregnancy. The collection of missing data on pregnancy outcome is ongoing. OR05-3 Examining the impact of growth hormone on the collagen content of adipose tissue L.A. Householder1 , K.M. Troike1 , E.O. List1 , E.R. Lubbers1 , J.J. Kopchick2,3 , D.E. Berryman1,4 . 1 Ohio University, Edison Biotechnology Institute, Athens, United States; 2 Ohio University, Athens, United States; 3 Ohio University, Department of Biological Sciences, Heritage College of Osteopathic Medicine, Athens, United States; 4 Ohio University, School of Applied Health Sciences and Wellness, College of Health Sciences and Professions, Athens, United States Obesity is characterized by insulin resistance, inﬂammation and pathologically accelerated white adipose tissue (WAT) remodeling. Although it has not yet been extensively studied, the extracellular matrix (ECM) of WAT may be linked with these key features of obesity and is an important area of obesity research. The ECM is the structural framework of WAT and is made primarily of collagen ﬁbers. An excess of these collagen ﬁbers, called ﬁbrosis, has been observed in obese adipose tissue. Adipose tissue ﬁbrosis is thought to contribute to the metabolic abnormalities and inﬂammation present in obese individuals. Growth hormone (GH) has been shown to increase collagen in other tissues and has been reported to impact adipose tissue in a depot speciﬁc manner. To date, no study has assessed the role of GH in adipose tissue ECM deposition. The use of mouse models with differing GH action may help characterize the role of ECM in adipose function/dysfunction. Bovine growth hormone transgenic mice (bGH) are giant and lean with high serum levels of bGH, IGF-1, and insulin and have relatively short lifespans. This contradictory phenotype – unhealthy leanness – allows us to examine WAT ﬁbrosis independent of normal adiposity and question whether the amount of ECM in the WAT is more important than the overall amount of WAT. Therefore, the purpose of this study was to evaluate measures