Hepcidin concentrations and iron homeostasis in preeclampsia

Clin Chem Lab Med 2010;48(10):1423–1426 � 2010 by Walter de Gruyter • Berlin • New York. DOI 10.1515/CCLM.2010.290

2010/185

Article in press - uncorrected proof

Hepcidin concentrations and iron homeostasis

in preeclampsia

Gergely Toldi1,2,*, Balazs Stenczer2, Attila Molvarec2,Zoltan Takats1, Gabriella Beko3, Janos Rigo Jr2 andBarna Vasarhelyi4

1 First Department of Pediatrics, Semmelweis University,Budapest, Hungary2 First Department of Obstetrics and Gynecology,Semmelweis University, Budapest, Hungary3 Department of Central Laboratory, SemmelweisUniversity, Budapest, Hungary4 Research Group of Pediatrics and Nephrology, HungarianAcademy of Sciences, Budapest, Hungary

Abstract

Background: Plasma iron is increased in preeclampsia (PE)when compared to healthy pregnant women. This is in con-trast to inflammation characteristic for PE. The link betweeniron homeostasis and inflammation is hepcidin. Our goal wasto describe hepcidin concentrations and its association withiron homeostasis in PE.Methods: We obtained peripheral blood samples from 30preeclamptic wgestational age: 36.5 (24–40) weeksx and 37healthy pregnant women wgestational age: 36 (28–39) weeksxto determine plasma hepcidin and interleukin-6 (IL-6) con-centrations, complete blood cell counts and parameters ofiron homeostasis wplasma iron, transferrin and ferritin levelsand total iron binding capacity (TIBC)x. Hepcidin was meas-ured using mass spectrophotometry. The Mann-Whitney testwas used for statistical analysis.Results: Plasma hepcidin, IL-6, iron and ferritin concentra-tions were increased (p-0.05 for all), whereas plasma trans-ferrin, TIBC and mean corpuscular hemoglobin concentra-tions were lower (p-0.05 for all) in PE compared to healthypregnant women. No differences were seen in the other para-meters investigated.Conclusions: Plasma iron concentrations are increaseddespite high hepcidin concentrations in PE. This might indi-cate a resistance to the iron-decreasing action of hepcidin.Clin Chem Lab Med 2010;48:1423–6.

Keywords: hepcidin; interleukin-6; iron homeostasis;preeclampsia.

*Corresponding author: Gergely Toldi, First Department ofPediatrics, Semmelweis University, Bokay u. 53–54, 1083Budapest, HungaryPhone: q36-20-4367181, Fax: q36-1-3138212,E-mail: [email protected] March 24, 2010; accepted May 3, 2010;previously published online July 14, 2010

Introduction

Preeclampsia (PE) affects approximately 5% of first timepregnancies, and is a leading cause of both maternal and fetalmorbidity and mortality. This systemic disorder is character-ized by the development of a maternal syndrome thatincludes hypertension, proteinuria, edema and vascularabnormalities. In most cases, these symptoms develop in thethird trimester of pregnancy and usually disappear within1 week after delivery. Vasoconstriction, endothelial cellinjury and systemic inflammation play a major role in thepathogenesis of PE. Th1 shift is a general characteristic ofthe disease, and pro-inflammatory cytokines are present inincreased concentrations in plasma (1).

Due to the complexity of PE, several other factors havebeen investigated that might play a role in the pathogenesis.Among these are alterations in iron homeostasis. It has beendemonstrated that plasma iron concentrations, ferritin, andsaturation of transferrin are increased, whereas total ironbinding capacity (TIBC), unsaturated iron binding capacityand apotransferrin are decreased in PE when compared tohealthy pregnant women (2, 3). Since iron may induce thegeneration of reactive oxygen species (ROS) through Fen-ton’s reaction, higher than normal iron concentrations havebeen suggested to contribute to the pathogenesis, and exac-erbate lipid peroxidation and endothelial cell injury in PE (3).

However, increased plasma iron concentrations are con-trary to the ongoing inflammation in PE. Several findingsand general clinical experience support the notion that chron-ic inflammation decreases iron availability, which might evenresult in inflammation-induced anemia (4). Based on theseobservations, one would expect a decrease in plasma ironconcentrations in PE, instead of increased concentrations thathave been reported.

The link between iron homeostasis and inflammation ishepcidin, a recently described acute phase peptide. It acts bydown regulating intestinal iron absorption and iron releasefrom enterocytes and macrophages through internalizationand degradation of ferroportin (5). Hepcidin expression isregulated by several factors. Its primary triggers includeinflammatory signals, such as interleukin-6 (IL-6) and highconcentrations of iron.

To date, no studies are available on alterations of hepcidinconcentrations in PE. Given its key role in iron homeostasis,such investigations are required to clarify the controversialreports on plasma iron concentrations in this inflammatorycomplication of pregnancy. The aim of our work was todescribe hepcidin concentrations and its association with ironhomeostasis in PE when compared to healthy pregnantwomen.

Brought to you by | University of Missouri-ColumbiaAuthenticated | 10.248.254.158

Download Date | 8/27/14 4:56 AM

1424 Toldi et al.: Hepcidin in preeclampsia


Table 1 Clinical characteristics of healthy pregnant and PE women.

Characteristics Healthy pregnant Preeclampticwomen womenns37 ns30

Age, years 30 (22–38) 30 (17–45)Gestational age at blood collection, weeks 36 (28–39) 36.5 (24–40)Systolic blood pressure, mm Hg 110 (100–130) 160a (140–201)Diastolic blood pressure, mm Hg 72 (50–80) 101a (95–120)Onset of PE, weeks of gestation – 34 (24–40)

Data are expressed as median, range. aps0.01.

Materials and methods

We enrolled 30 PE and 37 healthy singleton pregnant women in ourstudy. Clinical characteristics of participants are summarized inTable 1. Gestational and maternal age was comparable. PE wasdiagnosed according to internationally accepted standard criteria (6).All participants received regular oral iron supplementation in theform of iron (II) sulphate (30 mg/day) as part of routine perinatalcare. Informed consent was obtained from all subjects, and our studywas reviewed and approved by an independent Ethical Committeeof the institution. The study adhered to the tenets of the most recentrevision of the Declaration of Helsinki.

Plasma was isolated from 6 mL lithium-heparin anticoagulatedfasting blood samples (collected before receiving daily medications,including the oral iron supplementation) and stored at –808C untilanalysis. In addition, EDTA-K2 anticoagulated blood for a completeblood cell count was collected.

Hepcidin concentrations were measured using mass spectropho-tometry according to the modification of Murphy et al. (7). Humanplasma and acetonitrile containing 50 ng/mL internal standard weremixed 1:1. Tris(2-carboxyethyl)phosphine was used for the reduc-tion of samples, then iodoacetamide was added to block SH-groups.The mixture was centrifuged at 7000 g for 10 min at 58C; 150 mLsupernatant was transferred to an Oasis HLB 30 mm mElution SPEplate (Waters Corporation, Milford, MA, USA) containing 500 mLwater. Wells were rinsed with 500 mL water. For elution, trifluo-roacetic acid/water/acetonitrile (0.1/20/80, vol/vol/vol), 180 mL wasused. The eluant was collected onto a 96-well polypropylene plateand evaporated to dryness at 208C. For sample reconstitution,100 mL aqueous acetic acid (0.5%) was used and then the sampleswere injected onto a HPLC-column wLiChroCART 55-2 PurospherSTAR RP-18 endcapped (3 mm), Merck Chemicals, Darmstadt,Germanyx. Linear gradient elution was performed using eluent A:0.5% acetic acid in water and eluent B: 0.5% acetic acid in aceto-nitrile/methanol 3:1 (0 min: 95% A, 2 min: 95% A, 12 min: 5% A,18 min: 5% A) at a 200-mL/min flow rate. Ten mL sample wasinjected. The HPLC system was coupled to a mass spectrometer(Orbitrap Discovery XL, ThermoScientific, Waltham, MA, USA)operated in the positive ion electrospray mode. Double charged ionsof modified hepcidin were detected at a nominal resolution settingof 30,000 FWHM.

IL-6 was measured using the Roche Elecsys IL-6 kit (RocheDiagnostics GmbH, Mannheim, Germany), with a measurementrange of 1.5–5000 pg/mL.

Complete blood cell counts including erythrocytes (RBC), leu-kocytes (WBC), platelets, mean corpuscular volume (MCV), meancorpuscular hemoglobin (MCH) and mean corpuscular hemoglobinconcentrations (MCHC) were measured using a Sysmex K4500hematological automated analyzer (GMI, Ramsey, MN, USA) usingDiagon reagents (Diagon Ltd., Budapest, Hungary). Plasma iron,

transferrin and ferritin concentrations and TIBC were measuredusing an Olympus 2700 laboratory automated analyzer with avail-able Olympus kits (Olympus Europa GmbH, Hamburg, Germany).C-reactive protein (CRP) concentrations were measured using aRoche Hitachi 912 instrument with the Roche Tina-quant CRPimmuno-turbidimetric assay (Roche Diagnostics GmbH, Mannheim,Germany).

Data are shown as the median and range. Comparisons betweensample populations were performed with the Mann-Whitney test.The Kolmogorow-Smirnoff test for normality showed a non-normaldistribution of data. p-Values less than 0.05 were considered sig-nificant. Statistics were calculated using the R software (R Devel-opment Core Team, R Foundation for Statistical Computing,Vienna, Austria).

Results

Our results are summarized in Table 2. Briefly, plasma hep-cidin concentrations were increased in women with PE com-pared to healthy pregnant individuals (ps0.003). In addition,median IL-6 concentrations were also higher in PE comparedwith controls (ps0.0001). Plasma IL-6 concentrations werebelow the limit of detection in 28 of the healthy and eightof the women with PE. For statistical analysis, IL-6 concen-trations below the limit of detection were set to zero. Plasmairon and ferritin concentrations were also higher (ps0.02and ps0.003, respectively), while plasma transferrin con-centrations and TIBC values were lower in women with PEcompared to healthy pregnant women (ps0.02 for both).MCHC was decreased in PE compared to healthy pregnantwomen (ps0.04). No difference was detected in RBC,WBC, platelet, MCV, MCH, blood hemoglobin, hematocrit,and plasma CRP values.

Discussion and conclusions

In summary, we found that hepcidin concentrations wereincreased in PE when compared to healthy pregnant women.However, plasma iron concentrations did not decrease, butinstead showed an increase in PE. This finding might indi-cate resistance to the iron decreasing action of hepcidin inPE. Of note, some of the other parameters investigated werealso altered in PE when compared to healthy pregnant wom-en, including MCHC. A lower MCHC in PE could be dueto coagulation associated with PE (1).



Toldi et al.: Hepcidin in preeclampsia 1425


Table 2 Hepcidin, interleukin-6 (IL-6) and parameters of iron homeostasis in healthy pregnant women and those with preeclampsia.

Parameters Healthy pregnant Preeclampticwomen womenns37 ns30

Plasma hepcidin, ng/mL 3.74 (0.73–8.14) 5.68a (0.72–9.25)IL-6, pg/mL BLD (BLD–5.14) 2.75a (BLD–130.4)hs-CRP, mg/L 4.85 (0.9–11.6) 5.75 (0.2–18.2)RBC, T/L 4.05 (3.50–4.81) 4.05 (3.27–4.69)WBC, G/L 9.2 (6.2–14.4) 9.9 (7.2–14.7)Platelet, G/L 216 (153–329) 201 (70–373)MCV, fL 85.3 (63.6–98.1) 88.5 (78.1–94.0)MCH, pg 29.8 (21.2–33.9) 30.0 (25.4–33.2)MCHC, g/dL 34.4 (31.4–35.5) 33.6a (31.2–36.0)Hemoglobin, g/L 12.1 (9.5–14.1) 11.8 (9.9–14.9)Hematocrit, % 35.5 (30.1–41) 35.1 (23.9–43.2)Plasma iron, mmol/L 15.0 (6.8–29.5) 19.1a (7.1–51.6)Plasma transferrin, mmol/L 4.4 (3.6–6.2) 4.1a (2.8–5.7)TIBC, mmol/L 87.1 (71.3–122.8) 81.2a (55.4–112.9)Plasma ferritin, mg/L 15 (5–69) 34a (5–78)

Data are expressed as median, range. ap-0.05. BLD, below level of detection; CRP, C-reactive protein; RBC, erythrocyte count; WBC,leukocyte count; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentra-tion; TIBC, total iron binding capacity.

Several factors contribute to the regulation of hepcidinexpression. Its major trigger is the pro-inflammatory cyto-kine, IL-6 (8). In accordance with previous reports (9, 10),we demonstrated higher plasma concentrations of IL-6 in PE.This finding supports the notion that increased IL-6 may beresponsible for higher hepcidin concentrations.

In addition to IL-6, other factors might play a role in hep-cidin regulation in PE. Ishizaka et al. demonstrated thatadministration of angiotensin II induces hepcidin mRNAexpression (11). Recent observations demonstrated agonisticautoantibodies against the type 1 angiotensin II wAT(1)xreceptor in PE (12). Theoretically, this phenomenon may alsocontribute to the increased hepcidin concentrations seen inour patients. Of note, we were unable to confirm this hypoth-esis based on blood pressure values of patients, as the major-ity were receiving anti-hypertensive treatment (methyldopaor nifedipine).

The major effect of hepcidin is the internalization of fer-roportin that exports iron from cells. Decreased availabilityof ferroportin leads to a decrease in plasma iron concentra-tions. A major contradiction in our investigations was thefinding that increased hepcidin concentrations are coupledwith increased plasma iron in PE. Increased plasma iron con-centrations have been well described in PE (2, 3). This maycontribute to the inflammatory status due to increased pro-duction of ROS. A limitation of our study is that we did notmeasure ferroportin expression, whose concentrations mighthave been significantly altered in PE. Thus, our study shouldbe regarded as observational, without mechanistic implications.

A number of theories might explain the insensitivity tohepcidin action in PE. It has been suggested that PE presentswith hemochromatosis-like features. The prevalence of cer-tain mutations of the HFE gene was tested in PE patients,without definite results (3, 13). Hemochromatosis is gener-ally characterized by lower hepcidin concentrations. How-

ever, there is a subtype of the disease with higher hepcidinplasma concentrations, coupled with resistance to hepcidindue to functional or structural abnormalities of ferroportin.Indeed, several mutations in ferroportin have been reportedwith increased plasma iron concentrations and iron stores,including the Q248H mutation that is present in up to 20%of certain ethnic groups (14, 15). The limited number ofparticipants did not allow us to investigate possible mutationsin ferroportin expression in our study patients.

Effective mechanisms ensure a sufficient supply of iron tomeet the needs of the developing fetus during pregnancy.Transferrin receptors in the placenta are responsible for thetransfer of iron from the maternal to the fetal circulation.Transferrin receptor expression in the placenta is diminishedin PE, and this might explain the clinical experience that ironstores are decreased in newborns of mothers with PE (16,17). It is tempting to speculate that a factor contributing toincreased maternal iron concentrations in PE may be theinefficient transfer of iron to the fetus. Increased maternaliron concentrations may further increase hepcidin in PE(Figure 1).

Taking into consideration the above-described factors, onemight speculate to what extent hepcidin is responsible forthe regulation of iron homeostasis in women with PE com-pared to healthy pregnant women. To date, only one studydescribes the relationship between hepcidin and iron statusin pregnancy. This investigation demonstrated that urinaryhepcidin concentrations are positively associated with ferritin,and inversely associated with soluble transferrin receptors iniron-deficient pregnant women from a rural community.These results support that hepcidin signalling may differ inpregnancy, depending on the presence of iron deficiency(18). However, these data cannot be compared to our obser-vations, as none of our subjects were iron deficient, and weused plasma instead of urine to measure hepcidin concentra-



1426 Toldi et al.: Hepcidin in preeclampsia


Figure 1 Factors contributing to the regulation of hepcidinexpression.Superscript numbers refer to references in the text. Tf, transferrin;AT, angiotensin; R, receptor.

tions. Of note, plasma hepcidin concentrations are subject tofaster intra individual alterations compared to the relativelystable urinary hepcidin concentrations. While this may partlyexplain why we were unable to demonstrate an associationbetween hepcidin and plasma iron concentrations, it is stillof interest that even in healthy pregnancies, no associationbetween hepcidin and other parameters of iron status werenoted. The lack of correlation between hepcidin and ironhomeostasis may indicate that factors specific for pregnancymay interact with the iron decreasing action of hepcidin.

Acknowledgements

The authors wish to thank Edina Bıro (Central Laboratory Depart-ment, Semmelweis University, Budapest) for her professional tech-nical assistance during measurements. G.T. is a recipient of the MFBHabilitas Scholarship.

Conflict of interest statement

Authors’ conflict of interest disclosure: The authors stated thatthere are no conflicts of interest regarding the publication of thisarticle. Research funding played no role in the study design; in thecollection, analysis, and interpretation of data; in the writing of thereport; or in the decision to submit the report for publication.Research funding: Funding of this study was supported by grantsOTKA 76316 and TAMOP-4.2.2.-08/1/KMR-2008-0004.Employment or leadership: None declared.Honorarium: None declared.

References

1. Challis JR, Lockwood CJ, Myatt L, Norman JE, Strauss JF 3rd,Petraglia F. Inflammation and pregnancy. Reprod Sci 2009;16:206–15.

2. Basher K, Deb K. Alteration in iron status in preeclampsia.Mymensingh Med J 2006;15:22–4.

3. Rayman MP, Barlis J, Evans RW, Redman CW, King LJ.Abnormal iron parameters in the pregnancy syndrome pree-clampsia. Am J Obstet Gynecol 2002;187:412–8.

4. Balla J, Jeney V, Varga Z, Komodi E, Nagy E, Balla G. Ironhomeostasis in chronic inflammation. Acta Physiol Hung 2007;94:95–106.

5. Nemeth E, Ganz T. The role of hepcidin in iron metabolism.Acta Haematol 2009;122:78–86.

6. Brown MA, Lindheimer MD, de Swiet M, Van Assche A,Moutquin JM. The classification and diagnosis of the hyper-tensive disorders of pregnancy: statement from the InternationalSociety for the Study of Hypertension in Pregnancy (ISSHP).Hypertens Pregnancy 2001;20:IX–XIV.

7. Murphy AT, Witcher DR, Luan P, Wroblewski VJ. Quantitationof hepcidin from human and mouse serum using liquid chro-matography tandem mass spectrometry. Blood 2007;110:1048–54.

8. Wrighting DM, Andrews NC. Interleukin-6 induces hepcidinexpression through STAT3. Blood 2006;108:3204–9.

9. Greer IA, Lyall F, Perera T, Boswell F, Macara LM. Increasedconcentrations of cytokines interleukin-6 and interleukin-1receptor antagonist in plasma of women with preeclampsia: amechanism for endothelial dysfunction? Obstet Gynecol 1994;84:937–40.

10. Casart YC, Tarrazzi K, Camejo MI. Serum levels of interleukin-6, interleukin-1beta and human chorionic gonadotropin in pre-eclamptic and normal pregnancy. Gynecol Endocrinol 2007;23:300–3.

11. Ishizaka N, Saito K, Furuta K, Matsuzaki G, Koike K, NoiriE, et al. Angiotensin II-induced regulation of the expressionand localization of iron metabolism-related genes in the rat kid-ney. Hypertens Res 2007;30:195–202.

12. Xia Y, Kellems RE. Is preeclampsia an autoimmune disease?Clin Immunol 2009;133:1–12.

13. Senden IP, de Groot CJ, Steegers EA, Bertina RM, SwinkelsDW. Preeclampsia and the C282Y mutation in the hemochro-matosis (HFE) gene. Clin Chem 2004;50:973–4.

14. Kasvosve I, Gomo ZA, Nathoo KJ, Matibe P, Mudenge B,Loyevsky M, et al. Effect of ferroportin Q248H polymorphismon iron status in African children. Am J Clin Nutr 2005;82:1102–6.

15. Schimanski LM, Drakesmith H, Merryweather-Clarke AT,Viprakasit V, Edwards JP, Sweetland E, et al. In vitro functionalanalysis of human ferroportin (FPN) and hemochromatosis-associated FPN mutations. Blood 2005;105:4096–102.

16. Khatun R, Wu Y, Kanenishi K, Ueno M, Tanaka S, Hata T, etal. Immunohistochemical study of transferrin receptor expres-sion in the placenta of pre-eclamptic pregnancy. Placenta 2003;24:870–6.

17. Chockalingam UM, Murphy E, Ophoven JC, Weisdorf SA,Georgieff MK. Cord transferrin and ferritin values in newborninfants at risk for prenatal uteroplacental insufficiency andchronic hypoxia. J Pediatr 1987;111:283–6.

18. Schulze KJ, Christian P, Ruczinski I, Ray AL, Nath A, Wu LS,et al. Hepcidin and iron status among pregnant women in Bang-ladesh. Asia Pac J Clin Nutr 2008;17:451–6.



Documents

Hepcidin concentrations and iron homeostasis in preeclampsia