N

NFa

Bb

c

d

ARA

KSPIT

i5Pisoets

(T

0h

Journal of Reproductive Immunology 101–102 (2014) 135–139

Contents lists available at ScienceDirect

Journal of Reproductive Immunology

journa l homepage: www.e lsev ier .com/ locate / j repr imm

ovel ideas about salt, blood pressure, and pregnancy

atalia Rakovaa, Dominik N. Mullera, Anne Cathrine Staffb,c,riedrich C. Lufta, Ralf Dechenda,d,∗

Experimental and Clinical Research Center, a joint cooperation between the Charité Medical Faculty and the Max-Delbrueck Center for Molecular Medicine,erlin, GermanyDepartment of Obstetrics and Gynaecology, Oslo University Hospital, Ulleval, NorwayFaculty of Medicine, University of Oslo, NorwayHELIOS-Klinik, Berlin-Buch, Germany

a r t i c l e i n f o

rticle history:eceived 1 March 2013ccepted 2 April 2013

eywords:altreeclampsiammunityh17 cells

a b s t r a c t

The molecular mechanisms leading to preeclampsia are poorly understood. It has beenrelated to certain immune mechanisms, as well as the pathological regulation of therenin–angiotensin system together with perturbed salt and plasma volume regulation.Finally, a non-specific, vascular, inflammatory response is generated, which leads to theclinical syndrome. Here, we present novel findings in salt (NaCl) metabolism implying thatsalt is not only important in blood pressure control and volume homeostasis, but also inimmune regulation. Sodium and chloride can be stored without accumulation of waterin the interstitium at hypertonic concentrations through interactions with proteoglycans.Macrophages in the interstitium act as osmosensors for salt, producing increased amountsof vascular endothelial factor C, which increases the density of the lymph-capillary network

and the production of nitric oxide in vessels. An increased interstitial salt concentration acti-vates the innate immune system, especially Th17 cells, and may be an important trigger forautoimmune diseases. The novel findings with the idea of sodium storage and local mech-anisms of volume and immune regulation are appealing for preeclampsia and may unifythe “immune” and “vascular” hypotheses of preeclampsia.

© 2013 Elsevier Ireland Ltd. All rights reserved.

Preeclampsia is a major pregnancy complication, affect-ng 3–10% of pregnancies, and is responsible for at least0,000 maternal deaths per year (Roberts et al., 2003).reeclampsia originates in the placenta and is character-zed by hypertension and proteinuria, developing in theecond half of the pregnancy (Dekker and Sibai, 2001). Thenly present “cure” for preeclampsia is placental deliv-

ry and hence the baby. Many researchers agree thathere will be no single cause of the syndrome “preeclamp-ia”; however, there is a consensus of expert opinion that

∗ Corresponding author at: Experimental and Clinical Research CenterECRC), Lindenberger Weg 80, 13125 Berlin, Germany.el.: +49 30 450540301; fax: +49 30 450540944.

E-mail address: ralf.dechend@charite.de (R. Dechend).

165-0378/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.jri.2013.04.001

preeclampsia results from an imbalance between factorsproduced by the placenta and maternal adaptation to them(Staff et al., 2013).

Identification of risk factors for preeclampsia, suchas primiparity and partner specificity, which are relatedto paternal antigen exposure, generated a concept thatimmunological dysfunction at the fetal–maternal interfacein the first trimester contributes to preeclampsia (Dekkeret al., 2011). This concept hypothesizes that completefailure of maternal adaptation to fetal (paternal allo-antigens) would cause pregnancy loss. However, a partialfailure would induce poor placentation and dysfunctional

uteroplacental perfusion, leading to preeclampsia laterin pregnancy (Robertson et al., 2009). Different immunemechanisms such as maternal tolerance to paternal anti-gens by seminal plasma in the preconceptual phase,

ctive Imm

136 N. Rakova et al. / Journal of Reprodu

regulatory T cells, and decidual natural killer (NK) cellrecognition of fetal HLA-C on trophoblasts could be con-sidered likely possibilities (Redman and Sargent, 2010).

Recently, novel findings in salt (NaCl) metabolism havesuggested that salt is not only implicated in blood pres-sure control and volume homeostasis, but also in immuneregulation. Historically, salt has been a common additiveto human food. Animals and man cannot live without it.Thus, salt has a tradition as a valued gift and even as acurrency. Awareness of salt’s vital effects, such as sodiumbeing an essential element in regulating the body’s water,electrolyte balance, pH regulation, nervous transmission,and cardiac contraction did not prevent scientists frombecoming suspicious about possible adverse effects of ahigh salt intake. Thousands of years ago, ancient Chinesedoctors assumed a connection between salt intake and highblood pressure, as reflected by a red face and a hard pulse.In our time, numerous epidemiological trials have beenstaged with the goal of eliciting the truth about any connec-tion between salt and hypertension/cardiovascular disease.However, the studies have not been entirely convincing.Some investigations have indeed revealed a significantlyincreased risk of stroke and total cardiovascular diseaserelated to high salt intake (Strazzullo et al., 2009), whileothers found no justifying medical evidence to decrease saltconcentration in the diet (Cohen et al., 2006). The currentcontroversy may be attributed to the fact that it is nearlyimpossible to monitor salt intake in free-living individuals.Asking folks how much salt they eat is fraught with con-tradictions. Although 95% of salt ingested comes out in theurine, gathering this nectar is also not without problems.

The traditional concept is that regulation of sodiumexcretion by the kidney is the critical pathway for con-trolling fluid status, determining the level of the body’svolumes, and eventually blood pressure control. Titzeand colleagues presented evidence that these issues areeven more complicated (Machnik et al., 2009). Sodiumand chloride can be stored without accumulation ofwater in the subdermal interstitium at hypertonic con-centrations through interactions with proteoglycans. Asa matter of fact, storage mechanisms may be evenmore extensive. More peculiar still, they may show thatmacrophages regulate intradermal storage of sodium.They investigated underlying mechanisms of salt-inducedhypertension in rats and identified these immune cellsas principal operators responsible for the body’s responseto a salt load. Mononuclear phagocyte system (MPS) cells(macrophages), act as osmosensors. These little osmome-ters move into the skin, a major site of salt deposition andpresumably regulation.

One major conceptual difficulty has been proving thepresence of a local hypertonic environment within thecutaneous interstitium. The researchers used “ashing”of tissues (wet weighing, absolute desiccation, reweigh-ing, atomic absorption spectrometry) to determine theseelectrolyte concentrations. Measuring these variables inliving tissues (in vivo) is difficult; however, the group has

overcome these problems (data unpublished). Convincingevidence also stems from cause-and-effect relationships.Onsite macrophages express tonicity enhancer bindingprotein (TonEBP) in response to the detected local sodium

unology 101–102 (2014) 135–139

concentration. TonEBP is an osmosensitive transcriptionfactor with many roles. One of these roles happens to be theactivation of vascular endothelial growth factor C (VEGF-C). VEGF-C increases the density of the lymph-capillarynetwork in the skin, thereby enhancing production ofnitric oxide in the skin vessels. This state-of-affairs couldaffect the management of electrolyte and blood pressurehomeostasis. This novel mechanistic discovery does notnecessarily overrule the importance of renal regulatoryfunction. Nevertheless, the finding indicates that a com-plementary extrarenal mechanism of electrolyte, volumeregulation, and blood pressure balance control exists (Titzeand Machnik, 2010).

Supportive evidence that salt activates the innateimmune system has been recently presented by a studyinvestigating the triggering mechanisms of autoimmunediseases (Kleinewietfeld et al., 2013). These authors haveshown that increased sodium chloride induces Th17 cellsvia the activation of p38/MAPK, NFAT5, and SGK1. Further-more, mice fed with a high-salt diet developed more severeexperimental autoimmune encephalomyelitis than micereceiving a low-salt diet. The data are based on the notionthat the sodium concentrations in the interstitium and lym-phoid tissue are considerably higher than in the plasma.Wiig and Swartz (2012) indicate that concentrations of160 mM and even as high as 250 mM may exist, albeit in avery finite, gel-like state. One of the most astonishing find-ings from these papers is that the immune system functionsas a homeostatic regulator. Together with the interstitium,the proteoglycans, and the MPS cells, we are provided witha highly organized, local regulatory network that regu-lates sodium chloride storage. These data also suggest thatinterstitial salt concentration is a part of the “the milieuinterieur” that reflects an environmental risk factor regu-lating blood pressure via immune function (Titze and Ritz,2009; Titze and Machnik, 2010).

Titze et al. have relentlessly pursued human subjects-related research to underscore their animal findings. Anabsolutely environmentally controlled study in human vol-unteers undergoing a simulated “space flight to Mars”was helpful in that regard. In this study, salt intake wasfixed at three different levels, while all urine was col-lected. The study showed that single 24-h collections couldnot predict salt intake. However, the infradian rhythmsthis study identified underscore the idea of sodium stor-age and local mechanisms of storage and release (Titzeand Ritz, 2009). Titze also directed investigations at actu-ally “seeing” sodium storage. This “tour-de-force” wasaccomplished using 23Na magnetic resonance imaging inskin and skeletal muscle (Kopp et al., 2013). Still, doubtsremain whether or not these findings can be projectedonto common salt-consumption behavior in modern soci-eties. In the space-flight simulation, salt intake was variedbetween 12 g/day (currently assumed intake) and 6 g/day(approaching the guidelines) (Rakova et al., 2013). Thefindings of this study contradict the classical physiologicalconcept of extracellular volume expansion as a prerequisite

and a triggering mechanism for blood pressure elevation.The researchers showed that blood pressure effects wereindependent of extracellular water or total-body sodiumstorage. The salt effect on blood pressure was delayed

N. Rakova et al. / Journal of Reproductive Immunology 101–102 (2014) 135–139 137

salt in th

atsisbbatfi

fiirIlwHeoonptwawi2e

Fig. 1. The novel function of

nd required several weeks to achieve a steady state. Theextbook concept of volume expansion as a result of highalt intake leading to blood pressure rise, cardiac outputncrease, and eventually pressure natriuresis may not seemufficient to explain all the physiological mechanisms oflood pressure control. Rakova et al. (2013) showed thatoth aldosterone and cortisol regulate Na+ metabolism;ldosterone was associated with Na+ retention, while cor-isol was linked to increased Na+ excretion. The novelndings are illustrated in Fig. 1.

The important question is, how do these novel findingst into the pathogenesis of preeclampsia? In fact, several

nteresting observations have been reported that warranteinterpretation in the context of the “new salt concept.”ntervention studies are rare in preeclampsia; neverthe-ess, one study with low and high salt in 2077 pregnant

omen was performed as early as 1958 (Robinson, 1958).owever, the intervention for salt consisted of advice toither increase salt intake (during cooking, at meal times,r eating salty food) or to reduce the salt intake. The authorsbserved a lower incidence of preeclampsia, edema, peri-atal death, antepartum hemorrhage, and bleeding duringregnancy in those told to ingest more salt. Extra salt inhe diet seems to be essential for the health of a pregnantoman, her fetus, placental development, and appropri-

te function. In contrast to guidelines in non-pregnant

omen, salt restriction is not advised in pregnant women,

ndependent of whether or not they have hypertension. In000 the Cochrane database investigators, investigated theffects of advice on salt consumption during pregnancy

e interstitium is illustrated.

on the prevention and treatment of preeclampsia and itscomplications. They concluded that advice to lower saltintake should not be recommended (Duley and Henderson-Smart, 2000). Experimental in vivo data suggesting thatsalt reduction is not protective during pregnancy wereobtained by Giardina et al. (2002). A low-salt diet signif-icantly increased arterial pressure and vascular reactivityin pregnant and hypertensive pregnant rats subjected toreduced uterine perfusion pressure (RUPP). The authorsspeculate that the observed phenotype could be causedby an increase in Ca2+ entry from the extracellular spacepromoted by a low-salt diet.

In 1973, Weir et al. (1973) showed that renin,angiotensin II. and aldosterone were significantly lowerin hypertensive pregnancies, compared with normal preg-nancies. By that time it was well known that thecomponents of the renin–angiotensin–aldosterone system(RAAS) are increased during pregnancy (Shojaati et al.,2004). This upregulation is secondary to marked alterationsin the volume regulation, the composition, and the dynam-ics of body fluids during pregnancy. The marked expansionin plasma volume begins in the first trimester, acceler-ates in midgestation, and reaches a maximum after aboutthe 36th week of gestation. In preeclampsia, plasma vol-ume is not expanded to the normal pregnancy level. Thereduction in plasma volume in preeclampsia precedes the

clinical manifestations of the disease by several weeks.Despite gross retention of the intravascular plasma vol-ume with visible edema in patients with preeclampsia,plasma renin activity and aldosterone are paradoxically

ctive Imm

138 N. Rakova et al. / Journal of Reprodu

reduced, compared with normotensive pregnant women(Lindheimer and August, 2009). Relatively higher levelsof aldosterone for the given level of renin are observed,indicating reduced aldosterone availability in establishedpreeclampsia (Shojaati et al., 2004). In normal pregnancy,the ratio of aldosterone to renin increases substantiallythroughout gestation. Mohaupt and colleagues investi-gated this phenomenon (Escher and Mohaupt, 2007).They hypothesize that circulating aldosterone is the keyregulator for the increase in maternal plasma volumeduring pregnancy that appears necessary for optimalfetal development. Reduced aldosterone production abro-gates the pregnancy-associated expansion of circulatingfluid volume in preeclampsia and contributes to poorplacental perfusion and fetal substrate delivery. The reg-ulation of aldosterone in pregnancy and preeclampsia ispoorly understood, but it appears that the observed aldos-terone levels cannot solely be explained by increases inrenin and angiotensin II. Recently, we were able to showthat activating autoantibodies against the angiotensin IIreceptor, type 1 (AT1) increases the pregnancy-relatedreduced angiotensin II sensitivity in preeclampsia (Wenzelet al., 2011). Mohaupt et al. have recently expandedon their earlier observations and shown that vascularendothelial growth factor (VEGF) is a strong inducer ofadrenal aldosterone production (Escher and Mohaupt,2007).

Another important connection could be the fact that theplacenta is an organ with high levels of proteoglycans (PGs).PGs are macromolecules containing a core protein and atleast one negatively charged polysaccharide glycosamino-glycan side chain including heparan sulfate, dermatansulfate, or chondroitin sulfate (Said, 2011; Chui et al.,2012). The molecules are highly expressed in the placentaand represent the major components of the extracellularmatrix. PGs have multiple biological functions includingregulation of hemostasis and inflammation and via theirglycosaminoglycan side chains they interact with multi-ple growth factors (Said, 2011). PGs regulate a plethoraof cellular functions such as proliferation, migration, andinvasion. We have now learned that PGs can accumulatesodium without accompanying water retention. Chui et al.(2012) recently detected a significant decrease in the mRNAand protein expression of several proteoglycans, such assyndecan-2, glypicans 1 and 3, decorin, and perlecan inpreeclampsia compared with gestation-matched controls.

However, perhaps the most novel connection betweensodium homeostasis and preeclampsia could resideamong local salt concentrations, macrophages, and theconversion of naïve T cells to Th17 cells. Both Th17 anduteroplacental macrophages (so-called Hofbauer cells)have been investigated in the context of preeclampsia overthe years. Santner-Nanan et al. (2009) detected increasedTh17 cells and decreased regulatory T cells (Tregs) cells inperipheral blood of women with preeclampsia, comparedwith uneventful pregnancies. The data underscore theimportance of the balance between Tregs and Th17 cells

that is critical to maintain tolerance to the fetus. Thesedata undermine the concept that that defective controlof effector T cells by Tregs could lead to fetal rejection,causing maternal hypertension and the associated clinical

unology 101–102 (2014) 135–139

manifestations (Saito, 2010). A predominant Th17-typeimmunity in preeclampsia could be upstream of chronicinflammation and poor angiogenesis (Santner-Nanan et al.,2009; Saito, 2010).

Although described for several decades, the exact roleof macrophages in the uteroplacental unit and their func-tion in preeclampsia remain unknown. Recently, a geneexpression analysis revealed that most macrophages atthe maternal–fetal interface are M2 macrophages, whichexert an anti-inflammatory function and contribute to themaintenance of the semi-allogeneic fetus (Gustafsson et al.,2008). Schonkeren et al. (2011) recently demonstratedby means of immunohistochemistry that macrophagesare increased in the decidua of preeclamptic patients.Moreover, the amount of M2 macrophages was lowerin preeclampsia, providing evidence that a loss of animmunosuppressive phenotype is present locally at thematernal–fetal interface.

The hemodynamic and immunological features ofnormal pregnancy and the pathologies observed inpreeclampsia have generally been investigated indepen-dently of each other and no unifying concept has been putforward. The believers of the vascular hypothesis studiedthe marked expansion in plasma volume and upregula-tion of aldosterone in normal pregnancy and the failureof the volume expansion, interstitial edema, lack of aldos-terone availability, and increased angiotensin II sensitivityin preeclampsia. In contrast, reproductive immunologistshave gathered important evidence for a dysregulation ofregulatory T-cells and Th17 cells and their interaction withmacrophages and uterine NK cells. The new salt-regulatoryconcept is attractive and can be applied in normal preg-nancy as well as to the changes observed in preeclampsia.Understanding the role of local salt regulation in the inter-stitium of the uteroplacental unit could offer an importantopportunity to gain new insights into disease mechanisms.As a matter of fact, the recent introduction of 23Na mag-netic resonance imaging tomography offers an opportunityto do just that, even in humans, without risk to mother oroffspring (Kopp et al., 2013).

References

Chui, A., et al., 2012. The expression of placental proteoglycans in pre-eclampsia. Gynecol. Obstet. Invest. 73, 277–284.

Cohen, H.W., et al., 2006. Sodium intake and mortality in the NHANES IIfollow-up study. Am. J. Med. 119 (275), e7–e14.

Dekker, G., Sibai, B., 2001. Primary, secondary, and tertiary prevention ofpre-eclampsia. Lancet 357, 209–215.

Dekker, G., et al., 2011. The etiology of preeclampsia: the role of the father.J. Reprod. Immunol. 89, 126–132.

Duley, L., Henderson-Smart, D., 2000. Reduced salt intake compared tonormal dietary salt, or high intake, in pregnancy. Cochrane DatabaseSyst. Rev., CD001687.

Escher, G., Mohaupt, M., 2007. Role of aldosterone availability inpreeclampsia. Mol. Aspects Med. 28, 245–254.

Giardina, J.B., et al., 2002. Low-salt diet enhances vascular reactivity andca(2+) entry in pregnant rats with normal and reduced uterine per-fusion pressure. Hypertension 39, 368–374.

Gustafsson, C., et al., 2008. Gene expression profiling of human decidual

macrophages: evidence for immunosuppressive phenotype. PLoS ONE3, e2078.

Kleinewietfeld, M., et al., 2013. Sodium chloride drives autoimmunedisease by the induction of pathogenic TH17 cells. Nat. Med. 496,518–522.

ctive Imm

K

L

M

R

R

R

R

RS

S

S

increased angiotensin ii sensitivity in pregnant rats. Hypertension 58,

N. Rakova et al. / Journal of Reprodu

opp, C., et al., 2013. 23na magnetic resonance imaging-determined tissuesodium in healthy subjects and hypertensive patients. Hypertension61, 635–640.

indheimer, M.D., August, P., 2009. Aldosterone, maternal volume sta-tus and healthy pregnancies: a cycle of differing views. Nephrol. DialTransplant. 24, 1712–1714.

achnik, A., et al., 2009. Macrophages regulate salt-dependent volumeand blood pressure by a vascular endothelial growth factor-c-dependent buffering mechanism. Nat. Med. 15, 545–552.

akova, N., et al., 2013. Long-term space flight simulation reveals infradianrhythmicity in human Na(+) balance. Cell Metab. 17, 125–131.

edman, C.W., Sargent, I.L., 2010. Immunology of pre-eclampsia. Am. J.Reprod. Immunol. 63, 534–543.

oberts, J.M., et al., 2003. Summary of the NHLBI working groupon research on hypertension during pregnancy. Hypertension 41,437–445.

obertson, S.A., et al., 2009. Activating t regulatory cells for tolerancein early pregnancy – the contribution of seminal fluid. J. Reprod.Immunol. 83, 109–116.

obinson, M., 1958. Salt in pregnancy. Lancet 1, 178–181.aid, J.M., 2011. The role of proteoglycans in contributing to placental

thrombosis and fetal growth restriction. J. Pregnancy 2011, 928381.

aito, S., 2010. Th17 cells and regulatory t cells: new light on pathophy-

siology of preeclampsia. Immunol. Cell Biol. 88, 615–617.antner-Nanan, B., et al., 2009. Systemic increase in the ratio between

foxp3+ and il-17-producing cd4+ t cells in healthy pregnancy but notin preeclampsia. J. Immunol. 183, 7023–7030.

unology 101–102 (2014) 135–139 139

Schonkeren, D., et al., 2011. Differential distribution and phenotype ofdecidual macrophages in preeclamptic versus control pregnancies.Am. J. Pathol. 178, 709–717.

Shojaati, K., et al., 2004. Evidence for compromised aldosteronesynthase enzyme activity in preeclampsia. Kidney Int. 66,2322–2328.

Staff, A.C., et al., 2013. Review: preeclampsia, acute atherosis of the spi-ral arteries and future cardiovascular disease: two new hypotheses.Placenta 34, S73–S78.

Strazzullo, P., et al., 2009. Salt intake, stroke, and cardiovascular disease:meta-analysis of prospective studies. BMJ 339, b4567.

Titze, J., Machnik, A., 2010. Sodium sensing in the interstitium andrelationship to hypertension. Curr. Opin. Nephrol. Hypertens. 19,385–392.

Titze, J., Ritz, E., 2009. Salt and its effect on blood pressure and tar-get organ damage: new pieces in an old puzzle. J. Nephrol. 22,177–189.

Weir, R.J., et al., 1973. Plasma renin, renin substrate, angiotensin ii,and aldosterone in hypertensive disease of pregnancy. Lancet 1,291–294.

Wenzel, K., et al., 2011. Angiotensin ii type 1 receptor antibodies and

77–84.Wiig, H., Swartz, M.A., 2012. Interstitial fluid and lymph formation and

transport: physiological regulation and roles in inflammation and can-cer. Physiol. Rev. 92, 1005–1060.