wolfson 2016 jcem

Erythropoietin and soluble erythropoietin receptor: a role for maternal vascular adaptation to high altitude pregnancy Gabriel H. Wolfson, Enrique Vargas, Vaughn A. Browne, Lorna G. Moore, Colleen G. Julian

The Journal of Clinical Endocrinology & Metabolism The Endocrine Society Submitted: March 28, 2016 Accepted: November 01, 2016 First Online: November 03 , 2016

Early Release articles are PDF versions of manuscripts that have been peer reviewed and accepted

but not yet copyedited. The manuscripts are published online as soon as possible after acceptance

and before the copyedited, typeset articles are published. They are posted "as is" (i.e., as

submitted by the authors at the modification stage), and do not reflect editorial changes. No

corrections/changes to the PDF manuscripts are accepted. Accordingly, there likely will be

differences between the Early Release manuscripts and the final, typeset articles. The manuscripts

remain listed on the Early Release page until the final, typeset articles are posted. At that point,

the manuscripts are removed from the Early Release page.

DISCLAIMER: These manuscripts are provided "as is" without warranty of any kind, either express

or particular purpose, or non‐infringement. Changes will be made to these manuscripts before

publication. Review and/or use or reliance on these materials is at the discretion and risk of the

reader/user. In no event shall the Endocrine Society be liable for damages of any kind arising

references to, products or publications do not imply endorsement of that product or publication.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2016. at 16:47 For personal use only. No other uses without permission. . All rights reserved.

EARLY R

ELEASE

The Journal of Clinical Endocrinology & Metabolism; Copyright 2016 DOI: 10.1210/jc.2016-1767

Erythropoietin and soluble erythropoietin receptor: a role for maternal vascular adaptation to high altitude pregnancy

Gabriel H. Wolfson 1, Enrique Vargas 2, Vaughn A. Browne 3, Lorna G. Moore 4, Colleen G. Julian 1 1 Department of Medicine, University of Colorado School of Medicine, Aurora, Colorado, USA 2 Bolivian Institute of High Altitude Biology, La Paz, Bolivia 3 Department of Emergency Medicine, University of Colorado School of Medicine, Aurora, Colorado, USA 4 Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Aurora, Colorado, USA Received 28 March 2016. Accepted 01 November 2016.

Epo and sEpoR during pregnancy at high altitude

Corresponding Address: Colleen G. Julian, PhD, Department of Medicine, University of Colorado Denver, 12700 E 19th Avenue, Mailstop 8611, 3rd Floor Research Complex 2, Aurora, CO 80045, Email: [email protected], Phone: 303-724-644 | Fax: 303-724-1799

Supporting Grants: U.S. National Institutes of Health (NIH) Building Interdisciplinary Research Careers Women's Health grant (5 K12 HD057022-07; CGJ) and NIH-HL079647 (LGM). No funding source was involved in the design, conduct, interpretation of data; in the writing of the report; or in the decision to submit this article for publication.

Context: An imbalance of pro- and anti-angiogenic factors is thought to be central to the widespread vascular dysfunction characteristic to preeclampsia (PreE). Erythropoietin (Epo), a pleiotrophic cytokine known for its erythropoietic effects, has important angiogenic and vasoactive properties, however its contribution to impaired maternal vascular responses in PreE is unknown. Objective(s): Since chronic hypoxia raises the incidence of PreE we asked whether the chronic hypoxia of high altitude (HA) increased maternal Epo and soluble Epo receptor (sEpoR) levels, and whether such effects differed between PreE cases and normotensive controls at HA. Design, Setting and Participants: Longitudinal studies were conducted in pregnant, Andean HA (n=28; 3600 m) or sea level (SL, n=16; 300 m) residents. Cross-sectional studies included 34 gestational-age matched Andean PreE cases (n=17) and controls (n=17) in La Paz-El Alto, Bolivia (3600m – 4100m). Results: HA augmented the pregnancy-associated rise in Epo relative to SL (P=0.002), despite a similar reduction of Hb concentration across pregnancy at each altitude (7-9%, both P<0.001). HA PreE cases had equivalent circulating Epo levels compared to normotensive controls, but greater sEpoR values (P<0.05) and reduced hemoglobin (P=0.06, trend). Conclusion(s): Chronic hypoxia augments the pregnancy-associated rise in Epo but has no effect on maternal sEpoR, an effect that may be important for successful vascular adaptation to pregnancy at HA. In contrast, we speculate that elevated sEpoR observed in PreE cases vs. controls at HA impedes Epo-stimulated angiogenesis, vasodilation and the maintenance of endothelial function and may thereby be of pathophysiological relevance for increased incidence of PreE at HA.

PRECIS: Chronic hypoxia augments the pregnancy-associated rise in Epo but has no effect on sEpoR, an effect that may be important for vascular adaptation to pregnancy.

Introduction

1


mailto:[email protected]

EARLY R

ELEASE


Preeclampsia (PreE) is a multisystem vascular disease that complicates roughly 8.5 million pregnancies each year and accounts for 40% of fetal mortality and 18% of maternal deaths.1 PreE and fetal growth restriction often coexist and account for nearly 30% of premature deliveries with an estimated economic cost of $8.7 billion.2,3 Among the numerous factors known to increase the risk of PreE are pre-existing vascular disease, primiparity, advanced maternal age and hypoxia such as that experienced at high altitude (HA, ≥ 2500 m). HA residence raises the incidence of PreE threefold, highlighting the clinical relevance of persistent maternal hypoxia for hypertensive pregnancy disorders.4,5

Successful pregnancy is accompanied by pronounced, systemic arterial and venous vasodilation, a 30% rise in cardiac output 6 and extensive remodeling of the maternal spiral arteries that ultimately facilitate perfusion of the placental bed. In PreE a blunted maternal vasodilatory response to pregnancy and incomplete remodeling of the decidual and myometrial spiral arteries limits placental perfusion, thereby leading to hemodynamic and oxidative placental stress.7,8 As a result, the PreE placenta is thought to release antiangiogenic and proinflammatory factors into the maternal circulation and initiate the widespread maternal vascular endothelial dysfunction hallmark to PreE.9-12 Placental expression of the antiangiogenic factor soluble fms-like protein kinase 1 (sFlt1), for instance, is markedly greater in PreE, and paralleled by elevated sFlt1 in the maternal circulation.9-12 Previous studies have focused on the roles of sFlt1, placental growth factor and soluble endoglin in the angiogenic imbalance characteristic of PreE.9-12 We consider the role of another angiogenic factor, erythropoietin (Epo) and its receptor (EpoR), which are both regulated by hypoxia and thought to be essential for endothelial function and vascular homeostasis.

Epo is the master regulator of hypoxia-induced erythropoiesis, yet it has several other functions that promote successful placentation and vascular adaptation to pregnancy. Functional Epo membrane bound receptors (mEpoR) are present not only on erythroid progenitors but also vascular endothelial cells.13 Highlighting its pleiotrophic nature, Epo acts as a potent angiogenic factor, promotes neovascularization, induces endothelial nitric oxide synthase (eNOS) activity and enhances vascular repair in response to mechanical or ischemia-reperfusion injury.14-19 Further, Epo enhances trophoblast and decidual stromal cell proliferation and survival in early human pregnancy.20

Alternative EpoR mRNA splicing produces a soluble form of EpoR (sEpoR) that is present in human peripheral blood, and directly competes with mEpoR for unbound, circulating Epo;21 in this way, sEpoR impairs Epo signaling and Epo-mediated erythropoiesis.22 We hypothesized that HA would augment the pregnancy-induced rise in Epo and possibly reduce sEpoR levels. Further, we expected that such changes were absent or compromised in PreE or, in other words, that PreE women would have lower Epo and higher sEpoR levels than normotensive controls at HA. To the best of our knowledge, this is the first study to (1) measure sEpoR during pregnancy at any altitude, (2) compare changes in Epo and sEpoR levels during pregnancy at sea level (SL) vs. HA, and (3) contrast Epo or sEpoR values between normotensive and PreE pregnancy at HA.

Materials and Methods

Study Populations. A longitudinal and cross-sectional cohort were included for study. Cohort I consisted of 48 Andean women residing at HA (n=28; 3600 – 4300 m) or SL (n=16; 300 m) who remained normotensive throughout pregnancy. These women were participants in a larger investigation

2


EARLY R

ELEASE


designed to identify physiologic and genetic components of hypoxia-associated intrauterine growth restriction.23,24 Subjects were studied longitudinally to determine whether HA altered the magnitude and pattern of Epo and sEpoR across normotensive pregnancy. Subjects were referred by their prenatal care provider. For inclusion, women were required to be receiving prenatal care, having a singleton pregnancy, to be at no known risk for pregnancy complications (e.g., diabetes, preexisting hypertension, cardiovascular disease, history of gestational diabetes), and to self-identify as being of Aymara or Quechua descent. Only women with serial samples for study points of interest (20 and 36 weeks of pregnancy and 3-4 months post-partum) were included for analysis. We excluded smokers and those with active infections. All women completed a health and demographic questionnaire to determine her age, parity, duration of residence at current altitude, and medical and reproductive history. All subjects also underwent an obstetrical exam. During the physical exam, we measured bilateral upper extremity blood pressures (BP), height, and weight. Urine samples were collected to screen for infection and proteinuria. Non-fasting maternal venous blood samples were drawn between 09:00 and 11:00 am from an antecubital vein into EDTA collection tubes using standard techniques; the plasma was then separated and stored at -80o C until analysis. HA studies were conducted at the Bolivian Institute of High Altitude Biology (IBBA) and the Southern Clinic in La Paz-El Alto, Bolivia. SL studies were performed at the Siraní Clinic in Santa Cruz de la Sierra, Bolivia.

Cohort II, the cross-sectional cohort, consisted of 34 pregnant Andean women who presented for obstetrical care at five hospitals (Hospital de Los Andes, n=19; Bolivia-Holandés, n=5; Materno-Infantil, n=9; Hospital de la Mujer, n=1) in La Paz-El Alto, Bolivia. Subjects included normotensive controls (n=17) and PreE (n=17) cases matched by gestational age within two weeks at the time of sampling (mean gestational age for sampling were 34.2 2.2 vs. 35.6 3.0 weeks for normotensive controls and PreE cases, respectively). PreE cases were recruited within the first 24 hrs after diagnosis, blood samples obtained as described above after treatment with alpha methyldopa and magnesium sulfate, which are standard of care in Bolivia. For ethical reasons it wasn’t possible to delay treatment for research purposes therefore blood samples had to be obtained post-treatment. Inclusion and exclusion criteria were the same as those described for Cohort I. None of the normotensive controls developed PreE later in pregnancy.

All subjects gave written informed consent to study procedures that had been approved by the Colorado Multiple Institutional Review Board and the Colégio Médico, its Bolivian counterpart.

Definitions. PreE women were normotensive before pregnancy but developed elevated systolic (>140 mmHg) and diastolic (>90 mmHg) pressures after 20 weeks. PreE was distinguished from gestational hypertension by the presence of ≥1+ proteinuria at presentation and confirmed by ≥300 mg in a 24-h urine collection. In all cases, BP and proteinuria measurements obtained before the initiation of treatment were used to classify subjects as PreE. Gestational week was based on the date of last menstrual period and confirmed by fetal biometry at week 20 or clinical assessment at delivery.

Assessment of Epo and sEpoR levels. Plasma Epo and sEpoR were measured in duplicate using commercially available solid-phase sandwich ELISA kits (R&D Systems, Inc. MN, USA) according to manufacturer specifications. Detection ranges for the Epo and sEpoR ELISA kits are 2.5 – 200 mIU/ml and 62.50 - 4,000 pg/ml, respectively. Intra- and inter-assay variability for Epo ELISA kits are 2.8-5.2 % CV and 4.2-8.3 % CV, as indicated by R&D Systems. For sEpoR, the manufacturer does not report intra-

3


EARLY R

ELEASE


or inter-assay variability; we calculated an sEpoR intra-assay % CV of 6.9-12.9 and an inter-assay % CV of 1.5-13.8. Samples with undetectable Epo or sEpoR levels were assigned the value of the lower detection limit value (2.5 mIU/ml and 62.5 pg/ml, respectively).

Blood volume. For the measurement of blood volume (BV), subjects breathed through a rebreathing circuit initially containing 100% O2, from which CO2 was continuously removed and 100% O2 periodically added. A venous blood sample was withdrawn from an indwelling catheter after 5 min of quiet breathing. Carbon monoxide (CO, 100%, 60 ml ATP) was added to the rebreathing circuit, and additional blood samples obtained after 10 and 15 min of rebreathing. The percent

CO-Hb rose from ∼2% to 6–8%, as measured in triplicate by OSM3 (Radiometer, Copenhagen,

Denmark). Total BV was calculated using the equation [CO added/ΔCO content] × [1/Hb] × 100 where CO is the volume of CO added to the rebreathing circuit, ΔCO content is the difference in CO content between the baseline and the average of the 10- and 15-min values, and Hb is the measured Hb concentration using the cyanmethemoglobin technique. RCM was calculated as total BV multiplied by hematocrit (Hct) as determined by using the microcentrifuge technique after correcting for trapped plasma using the constant 0.98,25 and the remainder considered plasma volume (PV). Hb and Hct were measured in triplicate.

Statistical Analysis. In the text, tables, and figures, data are expressed as means ± SD or percentages with 95% confidence intervals after normality was affirmed using Kolmogorov and Smirnov tests. Pearson correlation was used to identify the potential influence of gestational age at the time of sampling and maternal characteristics (age, parity, weight) for our primary outcome variables (i.e., Epo and sEpoR); none were significant in either cohort and were thus not included as covariates. Repeated measures 2-way ANOVA with Sidak’s corrections for multiple comparisons were used to compare continuous variables between altitudes across pregnancy. Paired t-tests were used to compare continuous variables between gestational age matched normotensive controls and PreE cases. Proportions were compared using Fisher's exact test. Pearson correlation and linear regression were used to determine the relationship between Epo and indicators of pregnancy-induced hemodilution. All analyses were conducted using SPSS 24.0 (IBM Corp, Armonk, NY) and considered significant when P < 0.05. Sample sizes and statistical tests are noted for each table and figure, as appropriate.

Results

Maternal, newborn and delivery characteristics (Table 1).

Cohort I: Women studied longitudinally at SL and HA were similar in age (26.4 1.2 vs. 27.2 1.3 years) and lived at their current altitude of residence for the vast majority of their life (20.5 7.6 and 21.7 9.1 years, respectively). HA women were of greater parity (1.6 0.8 vs. 2.9 1.9 live births, p<0.05) and shorter stature than their SL counterparts (p<0.001, Table 1). Compared to LA, HA women gained less weight from 20 to 36 weeks of pregnancy (p<0.05), and weighed less at 36 weeks and postpartum (p <0.01, each).

4


EARLY R

ELEASE


Infants born to SL and HA women were equivalent with respect to gestational age at birth, ponderal index, Apgar scores (1 or 5 minutes), sex as well as the proportion born IUGR (18.8% and 8.7%, respectively) or premature (12.5% and 8.0%, respectively; all p=NS). Eighty percent (80%) of SL infants were delivered by cesarean section versus 18% of HA infants (p<0.05, Table 1). Compared to LA, birth weight (p=0.09), length (p=0.08) and head circumference (p<0.01) were or tended to be reduced in HA newborns.

Cohort II: In the cross-sectional cohort, normotensive and PreE women were of similar age 25.8 2.8 and 28.0 7.0 years), parity (0.7 0.7 and 0.9 1.2 live births) and height (150 4 and 151 9 cm), and had lived at high altitude for a similar duration (24.8 3.7 and 23.1 9.1, all p=NS). PreE weighed more than NORM at the time of gestational age-matched sampling (p<0.01, Table 1).

Compared to normotensive cases, PreE infants tended to be born at a younger gestational age (38.8 1.0 vs 37.5 2.4, p<0.10), and were of lower birth weight and birth length (p<0.05, both), but with similar head circumference and ponderal index (Table 1). An equivalent proportion of NORM and PreE infants were female, IUGR or premature (all p=NS, Table 1). Apgar scores were lower in infants born to PreE women compared to those born to normotensive controls at 5 minutes (p<0.01), despite similar values at 1 minute. More than half (59%) of PreE were delivered by cesarean section compared to only 12% of normotensives (p<0.05, Table 1).

Effect of altitude on maternal Epo and sEpoR across pregnancy. Epo increased with pregnancy at each altitude (Fig. 1A). Epo was equivalent between altitudes in the non-pregnant state, but values were greater among HA women during pregnancy (2.6 and 2.4-fold greater for 20 and 36 weeks, respectively, with a significant interaction between pregnancy and altitude, p=0.001). Specifically, Epo rose more sharply and to a greater extent at HA than SL (Fig. 1A). sEpoR values were consistent across pregnancy at SL and HA and did not differ between altitudes at any point (Fig. 1B).

As expected, Hb and Hct were greater at HA than SL at all times (Fig. 1C & D). Notably, however, the magnitude by which pregnancy lowered Hb and Hct was equivalent between altitudes (7 - 9 % drop by 36 weeks) despite vastly different Epo profiles. Given this observation we subsequently tested whether HA modified the relationship between Epo and Hb or Hct. At HA Epo was negatively associated with Hb and Hct at all times but at SL, Epo was independent of Hb and Hct at 36 weeks and post-partum, and tended to be positively associated with Hb and Hct at 20 weeks (Fig. 2A-C). As a result, the relationship between Epo and Hct differed between altitudes, with an inverse relationship being present at HA at all time points but either absent (nonpregnant and week 36, Fig 2A & C) or opposite in direction (week 20, Fig 2B) at SL. In contrast, sEpoR was unrelated to Hb or Hct at either altitude with the exception of the non-pregnant state at HA (Hb: r = -0.33, P=0.04, Hct: r = -0.38, P=0.02)

Relationship of BV and Epo during HA pregnancy. Compared to the non-pregnant state, total BV, expressed either as an absolute value or per kg, rose across normotensive pregnancy due to a 28% expansion of total PV by 36 weeks and a lesser, but statistically significant, 17% increase in total RCM across the same time frame (Fig. 3A-C). Epo was not associated with PV or RCM in the non-pregnant state but, as expected, was positively associated with PV (mL/kg) during pregnancy (20 weeks: r = 0.33, P=0.04; 36 weeks: r = 0.35, P=0.04) and inversely related to RCM (mL/kg) at 36 weeks (r = - 0.39, P=0.02).

Effect of PreE on maternal Epo and sEpoR.

5


EARLY R

ELEASE


Compared to normotensive controls PreE women had higher sEpoR values (322.7 401.8 vs. 72.2 40.1, respectively; P=0.02) (Fig. 4A). In contrast, Epo values were similar between PreE and controls, and Hb tended to be lower (P=NS and P=0.06, respectively) (Fig. 4B and C).

Discussion

Our findings demonstrate that environmental hypoxia increases the pregnancy-associated rise in Epo, despite a similar reduction in Hb across pregnancy at each altitude. Maternal sEpoR levels, on the other hand, were unaffected by HA during normotensive pregnancy or in the non-pregnant, postpartum state. Compared to normotensive women at HA PreE cases had similar Epo but higher sEpoR, which may reduce Epo bioavailability and thereby impair Epo-mediated responses necessary for vascular adaptation to pregnancy. In vitro studies indicating that recombinant sEpoR inhibits Epo mediated Stat-5 phosphorylation in BaF/EpoR cells support this possibility.22

Extensive hematological and ventilatory adaptations during pregnancy serve to maintain maternal CaO2 and, in turn, support maternal and fetal metabolism.6 Such adaptations include a pronounced expansion of PV and a smaller, yet significant, increase in RCM that is paralleled by rising maternal Epo. This pregnancy-associated rise in Epo reflects augmented production rather than reduced clearance.26 Consistent with existing literature, our data indicate that maternal Epo increases approximately twofold across normotensive, SL pregnancy.27 Notably, however, among normotensive, HA women Epo was twofold greater compared to SL at 20 and 36 weeks despite being similar in the non-pregnant, postpartum state. Our findings agree with those of a cross-sectional study indicating higher maternal Epo at HA relative to SL at the time of delivery28 and, to the best of our knowledge, are the first prospective comparison of Epo across pregnancy between HA and SL.

The source of the elevated basal maternal Epo levels during uncomplicated HA vs. SL pregnancy is unknown, but is likely placental in origin. First, Epo produced by fetal tissues, primarily hepatocytes, does not cross the placental barrier 29 and therefore cannot be responsible for rising maternal Epo levels during gestation. Second, because renal blood flow and oxygenation rise during uncomplicated pregnancy and is preserved during acute exposure to mild or moderate hypoxia,30-32 it is unlikely that hypoxia-induced renal Epo production accounts for the elevated Epo levels we observed at HA. Third, as previously reported maternal Hb is consistently greater during HA pregnancy relative to SL,33 but dropped similarly at each altitude across gestation; this suggests that plasma volume expansion was also similar between altitudes. Thus, excessive hemodilutional anemia does not appear to trigger the elevated Epo production at HA vs. SL during pregnancy. Finally, placental Epo production is highly sensitive to maternal oxygenation, as indicated by the effect of maternal hypoxia to raise the estimated umbilical Epo secretion rate of 27,900 mU/min in an in vivo ovine model.34

We speculate that the non-hematopoetic actions of Epo, including enhanced eNOS activation and endothelium-dependent vasodilation, stimulation of angiogenesis, and induction of placental growth factor,14,15,17-19,35 support the preservation of placental and maternal vascular adaptations that are essential for successful pregnancy under conditions of chronic hypoxia. Using this logic, we would expect elevated sEpoR, such as those we observed in PreE vs. normotensive controls at HA, to impair Epo-mediated vascular responses. On the other hand excessive placental Epo may be maladaptive by, for example, preventing the extensive vascular cell apoptosis and extracellular matrix decomposition that is essential for complete remodeling of the spiral arteries 20. In line with this possibility, some reports indicate that maternal Epo is

6


EARLY R

ELEASE


elevated in cases of abnormal placentation or PreE,36,37 a putatively compensatory response to impaired placental and renal perfusion. In support of a causal relationship, however, the therapeutic use of recombinant human Epo (rhEpo) for renal anemia during pregnancy evokes systemic arterial hypertension.38 Moreover, rhEpo treatment induced vasoconstriction of isolated human placental vessels studied in vitro, suggesting that Epo participates in the vasoregulation of the placental vascular bed in ways that are likely of pathophysiological importance for PreE.39 Therefore, we cannot exclude the possibility that the enhanced pregnancy-associated rise in Epo observed at HA contributes to the modest impairment of vascular adaptation characteristic to uncomplicated HA pregnancy or the increased incidence of PreE or IUGR.24 However, we found that Epo values were equivalent in PreE and normotensive pregnancy, an observation that is similar to other reports (e.g., 40) and does not suggest elevated Epo levels are linked to PreE.

Existing literature strongly indicates a positive association between maternal Hb, contracted plasma volume and pregnancy-induced hypertension at low altitude.41,42 Our finding that Hb was marginally lower in PreE compared to normotensive controls suggests that maternal Hb values within the range reported to be associated with minimal risk of stillbirth, prematurity or small-for-gestational-age birth at HA (11 to 13.6 g/dL) also do not influence the risk of PreE.33 One explanation for the modestly lower Hb levels we observed in PreE may be that the excessive inflammation in PreE dampens erythroid progenitor responsiveness to Epo and thereby impairs Epo-mediated erythropoiesis,43 as appears to be the case for anemia associated with other inflammatory conditions such as chronic kidney disease.44,45 Such effects may be the due to the capacity of endogenous cytokines (e.g., interferon gamma, interleukin 6 [IL6]) and T-cell activation to diminish erythroid progenitor responsiveness to Epo, or to regulate sEpoR production. Khankin, et al. demonstrated a link between inflammation and Epo resistance, reporting increased sEpoR levels in human erythroleukemia cells stimulated with IL6 and tumor necrosis factor alpha compared to untreated cells.22 Thus, the elevated sEpoR we observed in PreE may be stimulated by inflammatory cytokines. In support of this possibility, we have previously shown that HA women with PreE have higher IL6 and IL8 compared to normotensive controls.46 Moreover, elevated Epo levels in PreE have been shown to be unassociated with circulating erythroblasts, further suggesting that responsiveness to Epo is impaired in PreE.37 Finally, the elevated sEpoR levels we observed in PreE may impair Epo-mediated erythropoiesis and thereby reduce maternal Hb levels.22

Our findings suggest that Epo and sEpoR play important roles in maternal vascular responses to pregnancy at high altitude. We speculate that the enhanced pregnancy-induced rise in Epo at HA serves to maintain uteroplacental blood flow during hypoxic pregnancy. We further propose that enhanced sEpoR levels may be of pathophysiological relevance for the onset of PreE, and the elevated PreE incidence at high altitude by impeding Epo-stimulated angiogenesis, vasodilation and the maintenance of endothelial function. Further work, utilizing more sensitive methods to detect circulating sEpoR levels and experimental animal studies, including rhEpo or Epo antagonist treatment, are needed to understand the effect of sEpoR for Epo bioavailability during pregnancy and its impact on pregnancy outcome. Such studies may improve methods for the early detection of vascular disorders during pregnancy or reveal novel therapeutic avenues for PreE.

Acknowledgement(s). The authors acknowledge all of the women who participated in this project. Their appreciation is also extended to physicians and technical staff at the IBBA and Clinica Siraní for their assistance with the conduct of this project.

7


EARLY R

ELEASE


Disclosure Statement: The authors report no conflict of interest.

References 1. Anderson UD, Olsson MG, Kristensen KH, Akerstrom B, Hansson SR. Review: Biochemical markers to predict preeclampsia. Placenta. 2012;33 Suppl:S42-47. 2. Meis PJ, Goldenberg RL, Mercer BM, et al. The preterm prediction study: risk factors for indicated preterm births. Maternal-Fetal Medicine Units Network of the National Institute of Child Health and Human Development. Am J Obstet Gynecol. 1998;178(3):562-567. 3. In: Behrman RE, Butler AS, eds. Preterm Birth: Causes, Consequences, and Prevention. Washington (DC)2007. 4. Keyes LE, Armaza JF, Niermeyer S, et al. Intrauterine growth restriction, preeclampsia and intrauterine mortality at high altitude in Bolivia. Pediatr Res. 2003;54(1):20-25. 5. Palmer SK, Moore LG, Young DA, Cregger B, Berman JC, Zamudio S. Altered blood pressure course during normal pregnancy and increased preeclampsia at high altitude (3100 meters) in Colorado. Am J Obstet Gynecol. 1999;180(5):1161-1168. 6. Osol G, Moore LG. Maternal uterine vascular remodeling during pregnancy. Microcirculation. 2014;21(1):38-47. 7. Burton GJ, Yung HW, Cindrova-Davies T, Charnock-Jones DS. Placental endoplasmic reticulum stress and oxidative stress in the pathophysiology of unexplained intrauterine growth restriction and early onset preeclampsia. Placenta. 2009;30 Suppl A:S43-48. 8. Lyall F, Robson SC, Bulmer JN. Spiral artery remodeling and trophoblast invasion in preeclampsia and fetal growth restriction: relationship to clinical outcome. Hypertension. 2013;62(6):1046-1054. 9. Cerdeira AS, Karumanchi SA. Angiogenic factors in preeclampsia and related disorders. Cold Spring Harb Perspect Med. 2012;2(11). 10. Levine RJ, Maynard SE, Qian C, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med. 2004;350(7):672-683. 11. Maynard SE, Min JY, Merchan J, et al. Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest. 2003;111(5):649-658. 12. Powe CE, Levine RJ, Karumanchi SA. Preeclampsia, a disease of the maternal endothelium: the role of antiangiogenic factors and implications for later cardiovascular disease. Circulation. 2011;123(24):2856-2869. 13. Anagnostou A, Liu Z, Steiner M, et al. Erythropoietin receptor mRNA expression in human endothelial cells. Proc Natl Acad Sci U S A. 1994;91(9):3974-3978. 14. Beleslin-Cokic BB, Cokic VP, Yu X, Weksler BB, Schechter AN, Noguchi CT. Erythropoietin and hypoxia stimulate erythropoietin receptor and nitric oxide production by endothelial cells. Blood. 2004;104(7):2073-2080. 15. Kanagy NL, Perrine MF, Cheung DK, Walker BR. Erythropoietin administration in vivo increases vascular nitric oxide synthase expression. J Cardiovasc Pharmacol. 2003;42(4):527-533. 16. Quaschning T, Ruschitzka F, Stallmach T, et al. Erythropoietin-induced excessive erythrocytosis activates the tissue endothelin system in mice. FASEB J. 2003;17(2):259-261. 17. Ribatti D, Presta M, Vacca A, et al. Human erythropoietin induces a pro-angiogenic phenotype in cultured endothelial cells and stimulates neovascularization in vivo. Blood. 1999;93(8):2627-2636.

8


EARLY R

ELEASE


18. Nakano M, Satoh K, Fukumoto Y, et al. Important role of erythropoietin receptor to promote VEGF expression and angiogenesis in peripheral ischemia in mice. Circ Res. 2007;100(5):662-669. 19. Urao N, Okigaki M, Yamada H, et al. Erythropoietin-mobilized endothelial progenitors enhance reendothelialization via Akt-endothelial nitric oxide synthase activation and prevent neointimal hyperplasia. Circ Res. 2006;98(11):1405-1413. 20. Ji YQ, Zhang YQ, Li MQ, Du MR, Wei WW, Li da J. EPO improves the proliferation and inhibits apoptosis of trophoblast and decidual stromal cells through activating STAT-5 and inactivating p38 signal in human early pregnancy. Int J Clin Exp Pathol. 2011;4(8):765-774. 21. Yet MG, Jones SS. The extracytoplasmic domain of the erythropoietin receptor forms a monomeric complex with erythropoietin. Blood. 1993;82(6):1713-1719. 22. Khankin EV, Mutter WP, Tamez H, Yuan HT, Karumanchi SA, Thadhani R. Soluble erythropoietin receptor contributes to erythropoietin resistance in end-stage renal disease. PloS one. 2010;5(2):e9246. 23. Bigham A, Bauchet M, Pinto D, et al. Identifying signatures of natural selection in tibetan and andean populations using dense genome scan data. PLoS genetics. 2010;6(9). 24. Julian CG, Wilson MJ, Lopez M, et al. Augmented uterine artery blood flow and oxygen delivery protect Andeans from altitude-associated reductions in fetal growth. Am J Physiol Regul Integr Comp Physiol. 2009;296(5):R1564-1575. 25. Burge CM, Skinner SL. Determination of hemoglobin mass and blood volume with CO: evaluation and application of a method. J Appl Physiol (1985). 1995;79(2):623-631. 26. Gough SR, Mosher MD, Conrad KP. Metabolism of erythropoietin in conscious pregnant rats. Am J Physiol. 1995;268(5 Pt 2):R1117-1120. 27. Barton DP, Joy MT, Lappin TR, et al. Maternal erythropoietin in singleton pregnancies: a randomized trial on the effect of oral hematinic supplementation. Am J Obstet Gynecol. 1994;170(3):896-901. 28. Zamudio S, Baumann MU, Illsley NP. Effects of chronic hypoxia in vivo on the expression of human placental glucose transporters. Placenta. 2006;27(1):49-55. 29. Zanjani ED, Pixley JS, Slotnick N, MacKintosh FR, Ekhterae D, Clemons G. Erythropoietin does not cross the placenta into the fetus. Pathobiology. 1993;61(3-4):211-215. 30. Conrad KP, Davison JM. The renal circulation in normal pregnancy and preeclampsia: is there a place for relaxin? Am J Physiol Renal Physiol. 2014;306(10):F1121-1135. 31. Ashack R, Farber MO, Weinberger MH, Robertson GL, Fineberg NS, Manfredi F. Renal and hormonal responses to acute hypoxia in normal individuals. J Lab Clin Med. 1985;106(1):12-16. 32. Swenson ER. Renal function and fluid homeostasis. In: Hornbein TF SR, ed. High Altitude: An Exploration of Human Adaptation. New York: Marcel Dekker; 2001:525-568. 33. Gonzales GF, Steenland K, Tapia V. Maternal hemoglobin level and fetal outcome at low and high altitudes. Am J Physiol Regul Integr Comp Physiol. 2009;297(5):R1477-1485. 34. Davis LE, Widness JA, Brace RA. Renal and placental secretion of erythropoietin during anemia or hypoxia in the ovine fetus. Am J Obstet Gynecol. 2003;189(6):1764-1770. 35. Gonsalves CS, Li C, Mpollo MS, et al. Erythropoietin-mediated expression of placenta growth factor is regulated via activation of hypoxia-inducible factor-1alpha and post-transcriptionally by miR-214 in sickle cell disease. Biochem J. 2015;468(3):409-423. 36. Goldstein JD, Garry DJ, Maulik D. Obstetric conditions and erythropoietin levels. Am J Obstet Gynecol. 2000;182(5):1055-1057.

9


EARLY R

ELEASE


37. Troeger C, Holzgreve W, Ladewig A, Zhong XY, Hahn S. Examination of maternal plasma erythropoietin and activin A concentrations with regard to circulatory erythroblast levels in normal and preeclamptic pregnancies. Fetal Diagn Ther. 2006;21(1):156-160. 38. Kashiwagi M, Breymann C, Huch R, Huch A. Hypertension in a pregnancy with renal anemia after recombinant human erythropoietin (rhEPO) therapy. Arch Gynecol Obstet. 2002;267(1):54-56. 39. Resch BE, Gaspar R, Sonkodi S, Falkay G. Vasoactive effects of erythropoietin on human placental blood vessels in vitro. Am J Obstet Gynecol. 2003;188(4):993-996. 40. Hershkovitz R, Ohel I, Sheizaf B, et al. Erythropoietin concentration among patients with and without preeclampsia. Arch Gynecol Obstet. 2005;273(3):140-143. 41. Knottnerus JA, Delgado LR, Knipschild PG, Essed GG, Smits F. Haematologic parameters and pregnancy outcome. A prospective cohort study in the third trimester. J Clin Epidemiol. 1990;43(5):461-466. 42. Murphy JF, O'Riordan J, Newcombe RG, Coles EC, Pearson JF. Relation of haemoglobin levels in first and second trimesters to outcome of pregnancy. Lancet. 1986;1(8488):992-995. 43. Redman CW, Sacks GP, Sargent IL. Preeclampsia: an excessive maternal inflammatory response to pregnancy. Am J Obstet Gynecol. 1999;180(2 Pt 1):499-506. 44. Priyadarshi A, Shapiro JI. Erythropoietin resistance in the treatment of the anemia of chronic renal failure. Semin Dial. 2006;19(4):273-278. 45. Stenvinkel P, Barany P. Anaemia, rHuEPO resistance, and cardiovascular disease in end-stage renal failure; links to inflammation and oxidative stress. Nephrol Dial Transplant. 2002;17 Suppl 5:32-37. 46. Davila RD, Julian CG, Browne VA, et al. Role of cytokines in altitude-associated preeclampsia. Pregnancy Hypertens. 2012;2(1):65-70.

Figure 1. Maternal Epo, sEpoR and indices of hemodilution during pregnancy at HA and SL. Repeated measures 2-way ANOVA with Sidak’s corrections for multiple comparisons were used to contrast Epo, sEpoR, hemoglobin (Hb) and hematocrit (Hct) levels across pregnancy between Andean women residing at HA (n=28) versus SL (n=16). Complete Epo, sEpoR, Hb and Hct data were available for each subject at every study point. Pregnancy increased circulating Epo levels at HA but not at SL with the result that Epo was greater at 20 and 36 weeks compared to sea-level values (A). sEpoR did not change across pregnancy at either altitude and values were equivalent between altitudes at each point of study (B). Despite substantially higher Epo levels during pregnancy at HA, the magnitude of pregnancy-associated fall in hemoglobin (Hb) and hematocrit (Hct) were similar at each altitude (C & D). As expected Hb and Hct values were greater at HA than SL at all times. Superscripts a, b, and c indicate significant differences relative to the non-pregnant state, 20 weeks or 36 weeks, respectively. Significant differences between altitudes are noted by ** (P<0.01) or *** (P<0.001). Data are shown as the mean ± SD.

Figure 2. Relationship between Epo and Hct at HA vs. SL in the non-pregnant state and during pregnancy. Linear regression models were used to test whether the relationship (slope) between Epo and Hct differed between Andean women residing at HA vs. SL in the non-pregnant, postpartum period, or during pregnancy (20 or 36 weeks). Complete Epo and Hct data were available for HA (n=28) and SL (n=16) women at all study points. Regression lines and individual data points for HA (HA: solid line, solid squares) and (SL: dashed line, open circles) are shown. At SL Epo and Hct are unrelated in the non-pregnant state; in contrast, non-pregnant

10


EARLY R

ELEASE


Epo and Hct values are inversely associated at HA (A). At 20 weeks the relationship between Epo and Hct are significantly different between the two altitudes, as indicated by the HA-SL slope contrast (P=0.04; B). Specifically, 20 wk Epo and Hct values are inversely related at HA and positively associated at SL. At 36 weeks the relationship between Epo and Hct is similar to that of the non-pregnant state, with an inverse association apparent at HA and no association at SL (C).

Figure 3. Red cell mass, plasma and blood volume across HA pregnancy. Repeated measures ANOVA was used to test the effect of pregnancy on hematological parameters in HA Andean women (n=28, for all study points). Total red cell mass (RCM) increased ~17% by 36 weeks of pregnancy (P<0.0001), however when expressed as a function of maternal weight RCM was consistent across time (A). Plasma volume (PV, mL and mL/kg) expanded consistently across pregnancy, with values being ~28% and ~20% higher at 36 weeks vs. the non-pregnant state, respectively (B; both P<0.0001). As a result, blood volume (BV, mL and mL/kg) and increased progressively with pregnancy, rising ~24% and ~ 20% by 36 weeks, respectively (C; both P<0.001). Notably, maximal expansion of RCM, PV and BV expressed as mL/kg was achieved by 20 weeks. Superscripts indicate a significant difference relative to a non-pregnant, b 20 wk or c 36 wk values. Data are shown as the mean ± SD.

Figure 4. Epo and sEpoR levels in preeclamptic (PreE) vs. normotensive pregnancies at HA. Paired t-tests were used to contrast Epo, sEpoR and Hb values in PreE vs. normotensive pregnancy in Andean HA residents with Bonferroni multiple testing correction. PreE cases (open squares) and normotensive controls (solid circles) were gestational-age matched (n=17, each group). Maternal Epo levels were equivalent between normotensive controls and PreE cases (A, control: 26.4 17.2 vs. PreE: 32.1 23.7 mIU/mL). In contrast, compared to controls PreE women had greater sEpoR values (B, control: 72.2 40.1 vs. PreE: 322.7 401.8 pg/mL; P=0.02), a statistically significant difference after Bonferroni correction.

11


EARLY R

ELEASE


12

Table 1. Maternal, Newborn and Delivery Characteristics

Maternal Characteristics Longitudinal Normotensive, SL vs. HA Cross-sectional HA, Control vs. PreE Variable Sea level High altitude P-value Control PreE P-value Age, yrs 26.4 ± 1.2 (16) 27.2 ± 1.3 (28) NS 25.8 ± 2.8 (17) 28.0 ± 7.0 (17) NS Parity, # 1.6 ± 0.8 (16) 2.9 ± 1.9 (28) <0.05 0.7 ± 0.7 (17) 0.9 ± 1.2 (17) NS Current altitude, yrs 20.5 ± 7.6 (16) 21.7 ± 9.1 (28) ---- 25.2 ± 3.3 (17) 23.1 ± 9.2 (17) NS Height, cm 158 ± 2 (16) 150 ± 1 (28) <0.001 150 ± 4 (17) 151 ± 9 (17) NS Weight gain, 20-36 wks, kg 8.2 ± 3.9 (15) 6.0 ± 3.1 (28) <0.05 ---- ---- ---- Weight, 36 wks, kg 65.8 ± 7.4 (16) 76.9 ± 13.3 (28) <0.01 63.7 ± 7.2 (17) 72.7 ± 10.5 (17) <0.01 Weight, post-partum, kg 68.8 ± 12.7 (16) 59.3 ± 9.0 (28) <0.01 ---- ---- ---- BMI, post-partum, kg 27.4 ± 3.7 (16) 26.3 ± 3.7 (28) NS ---- ---- ---- Delivery and Newborn Information Variable Sea level High altitude P-value Control PreE P-value Gestational age, wks 39.5 ± 2.2, (15) 39.4 ± 1.4 (25) NS 38.8 ± 1.0 (17) 37.5 ± 2.4 (17) trend Birth weight, gm 3425 ± 573, (16) 3194 ± 364 (25) =0.10 3104 ± 384 (17) 2641 ± 641 (17) <0.05 Head circumference, cm 35.4 ± 1.0 (16) 34.2 ± 1.0 (21) <0.01 33.8 ± 1.1 (16) 33.7 ± 1.8 (16) NS Birth length, cm 49.9 ± 2.4 (16) 48.7 ± 1.9 (24) =0.08 49.7 ± 1.4 (15) 47.9 ± 2.1 (15) <0.05 Ponderal index, kg/m3 27.5 ± 3.0 (16) 27.6 ± 2.7 (24) NS 25.0 ± 2.6 (15) 24.2 ± 2.7 (15) NS Infant sex, % female 50 (26, 75) (16) 57 (39, 77) (26) NS 30 (6,56) (13) 30 (6,56) (13) NS SGA, % yes 18.8 (0,37.9) (16) 8.7 (0, 20.2) (23) NS 29 (7,51) (17) 35 (12,58) (17) NS Preterm, % yes 12.5 (0, 28.7) (16) 8 (0, 18.6) (25) NS 0 (0,0) (17) 0 (0,0) (17) NS Apgar 1 min 7.9 ± 0.3 (16) 7.9 ± 0.8 (21) NS 7.7 ± 0.5 (15) 7.3 ± 1.4 (15) NS Apgar 5 min 8.9 ± 0.7 (16) 9.0 ± 0.8 (21) NS 9.6 ± 0.5 (15) 9.1 ± 0.7 (15) <0.01 Cesarean delivery, % 80 (59.8, 100.2) (15) 17.9 (3.6, 32.0) (28) <0.05 12 (0,27) (17) 59 (36, 82) (17) <0.01

Abbreviations: body mass index (BMI), small for gestational age (SGA); continuous variables are shown as the mean ± SD, categorical variables are shown as proportions with 95% CI; subject numbers are shown in italics for each variable.


EARLY R

ELEASE

NP 20 368

10

12

14

16

18

Pregnancy week

Hem

oglo

bin,

g/d

L

SL

HA

NP 20 3625

30

35

40

45

50

Pregnancy week

Hem

atoc

rit, %

HA

SL

NP 20 360

10

20

30

Pregnancy week

Epo,

mIU

/mL

SL

HA

NP 20 36 0

100

200

300

400

Pregnancy week

sEpo

R, p

g/m

L

HASL

pregnancy <0.0001altitude = 0.002preg x altitude =0.001

pregnancy <0.0001altitude <0.0001preg x altitude NS

pregnancy <0.0001altitude <0.0001preg x altitude NS

pregnancy NSaltitude NSpreg x altitude NS

Figure 1

A. B.

C. D.

a

aa

aa

aa

aa

*****

a a

a

*** *********

***

***


EARLY R

ELEASE30 35 40 45 50 55

0

10

20

30

40

Hct, %

Epo,

mIU

/mL

Epo HA slope - 0.9, p=0.05Epo SL slope - 0.2, p=0.27

HA vs. SL slope, p=0.26

25 30 35 40 45 500

20

40

60

80

Hct, %

Epo,

mIU

/mL

Epo HA slope - 2.4, p=0.06

Epo SL slope + 0.6, p=0.06

HA vs. SL slope, p=0.04

25 30 35 40 45 50 550

20

40

60

80

100

Hct, %

Epo,

mIU

/mL

Epo HA slope - 2.0, p=0.01

Epo SL slope - 0.4, p=0.18HA vs. SL slope, p=0.17

A. Non-pregnant

B. 20 wks

C. 36 wks


EARLY R

ELEASE

A.

NP 20 wks 36 wks0

1000

2000

3000

0

20

40

60

Pregnancy week

RC

M, m

L

RC

M, m

L/kg

c c

a, b

RCM, mL/kg

RCM, mL

NP 20 wks 36 wks0

1000

2000

3000

4000

0

20

40

60

80

Pregnancy week

PV, m

L

PV, mL

PV, mL/kg

b, ca, c

a, b

a a

b, c

PV, mL/kg

NP 20 wks 36 wks0

2000

4000

6000

8000

0

50

100

150

Pregnancy week

BV,

mL

BV, mL

BV, mL/kg

b, c a, c

a, b

b, ca a

BV, m

L/kg

B.

C.

Figure 3. The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2016. at 16:47 For personal use only. No other uses without permission. . All rights reserved.

EARLY R

ELEASE

Control

PreE0

50

100

150

Epo,

mIU

/mL

Control

PreE-500

0

500

1000

1500

sEpo

R, p

g/m

L A.

B.

p = NS

p = 0.02


wolfson 2016 jcem

Documents