Possible therapeutic effect of magnesium sulfate in pre-eclampsia by the down-regulation of placental tumor necrosis factor-alpha secretion

ARTICLE

Auteur(s) : Alaa Amash^1,2, Adi Y Weintraub³, Eyal Sheiner^2,3, Atef Zeadna³, Mahmoud Huleihel^1,2, Gershon Holcberg^2,3

¹The Shraga Segal Department of Microbiology and Immunology
²Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer Sheva
³Department of Obstetrics and Gynecology, Soroka University Medical Center, Beer Sheva, Israel

accepté le 7 Octobre 2009

Pre-eclampsia is a syndrome characterized by the onset of hypertension and proteinuria after 20 weeks of gestation. Additional signs and symptoms that can occur include visual disturbances, headache, epigastric pain, thrombocytopenia, and abnormal liver function [1]. These clinical manifestations result from mild to severe microangiopathy of target organs, including the brain, liver, kidney, and placenta [2]. Pre-eclampsia affects 3-5% of pregnancies worldwide, and is considered to be a leading cause of maternal and fetal morbidity and mortality. However, the etiology of this disease remains undefined [3].

The current theory concerning the development of pre-eclampsia proposes that pre-eclampsia is a two-stage disease. A combination of immunological, environmental and genetic factors results in impaired trophoblast invasion and defective placentation. This may lead to a reduction in uteroplacental perfusion, resulting in placental ischemia/hypoxia. Ischemic conditions in the placenta, during the late stages of gestation, initiate induced-release of several angiogenic factors including pro-inflammatory cytokines into the maternal circulation. Consequently, systemic endothelial dysfunction and the clinical manifestations of pre-eclampsia are seen [4, 5].

Pro-inflammatory cytokines are thought to link placental ischemia with cardiovascular and renal dysfunction. The placenta is an integral component of this inflammatory response as it actively produces a variety of cytokines and immunomodulatory hormones [6, 7]. Blood pressure regulatory systems, such as the renin-angiotensin system (RAS) and the sympathetic nervous system, interact with pro-inflammatory cytokines, which affect angiogenic and endothelium-derived factors regulating endothelial function [6, 8].

Tumor necrosis factor-alpha (TNF-α) is a well known member of the TNF superfamily that is involved in numerous cellular processes. TNF-α activities, mediated through two distinct receptors TNFR1 and TNFR2, include regulation of cytokines expression, immune receptors, proteases, growth factors and cell cycle genes, which, in turn, regulate inflammation, survival, apoptosis, cell migration, proliferation and differentiation [9]. A number of groups have reported that circulating levels TNF-α are increased in women with pre-eclampsia [10, 11], suggesting its possible involvement in the pathogenesis of this disorder. However, there is a controversy about the expression of placental TNF-α levels in pre-eclampsia [12-14].

Magnesium sulfate (MgSO₄) is the drug of choice for the treatment of severe pre-eclampsia [15], prevention of eclampsia [16] and prevention of recurrent eclamptic seizures [17]. Previously, we have shown that MgSO₄ might selectively attenuate the vasoconstrictive effect of angiotensin II (Ang-II) and endothelin-1 (ET-1) on placental vasculature [18]. Although the mechanism of action of MgSO₄ as an anticonvulsant agent in pre-eclampsia/eclampsia is still not clearly understood, some possible mechanisms including vasodilatation of cerebral vasculature, inhibition of platelet aggregation, protection of endothelial cells from damage by free radicals, prevention of calcium ion entry into ischemic cells, decreasing the release of acetylcholine at motor end plates within the neuromuscular junction, and as a competitive antagonist to the glutamate N-methyl-D-aspartate receptor (which is epileptogenic), have been proposed [19].

Limited data have suggested that MgSO₄ may have neuroprotective effects on preterm neonates [20-22]. Although possible mechanisms by which MgSO₄ might be neuroprotective, such as blocking of glutamate receptors [23], have been suggested, the specific mechanism of this potential effect remains unclear. Inflammatory cytokines, such as TNF-α, interleukin (IL)-6 and IL-1β, have been linked to increased risk of neuronal morbidities including cerebral palsy (CP) [24-26]. Recently, our group reported that MgSO₄ may differently affect the capacity of the fetal and the maternal compartments of normotensive human placenta to secrete the inflammatory cytokines TNF-α and IL-6, in presence of Ang-II [27]. Therefore, one of the potential mechanisms for the neuroprotective effects MgSO₄ might be by affecting the expression levels of these cytokines.

The current study was performed to compare the capacity of ex vivo-perfused, pre-eclamptic and normotensive placentas to produce TNF-α, and to examine the possible effect of MgSO₄ on the capacity of these placentas to produce TNF-α.

Methods and materials

Study population

Placental perfusion

Within 15-20 minutes of delivery, a fetal artery and corresponding vein from an intact cotyledon (lateral cotyledons containing part of the membranes) were cannulated. Following successful establishment of the fetal circulation, the placenta was mounted in a perfusion chamber, and the maternal circulation was simulated by placing four catheters into the intervillous space of the lobe, corresponding to the isolated perfused cotyledon. Maternal perfusate that returned from the intervillous space was continuously drained by a maternal venous catheter, placed at the lowest level on the maternal decidual surface to avoid significant pooling of perfusate.

Perfusion medium consisted of two liters (L) of M-199 cell culture medium [M-199 media (Sigma Chemicals Co., St. Louis, USA)], enriched with bovine serum albumin (1 gr/L), glucose (1 gr/L) (Sigma), heparin (10 IU/mL) (Beit Kama, Israel) and Gentamycin (40 μg/mL) (Teva, Petah Tekva, Israel). The pH of the medium was adjusted to 7.4 with bicarbonate (Sigma).

The two reservoirs, containing the perfusion medium for the maternal and the fetal circuits, were placed into heated water baths at 37˚C, and were equilibrated with a pre-humidified gas mixture of 95% O₂ and 5% CO₂ in the maternal reservoir and 95% N₂ and 5% CO₂ in the fetal reservoir. A perfusion pressure of 20-40 mmHg, giving a flow rate of 6-8 mL/min in the fetal circulation and 10-12 mL/min in the maternal circulation, was established. The venous return could be recycled into the respective reservoir, giving a closed circuit perfusion.

Perfusate samples from the fetal and the maternal circulations were collected in each experiment every 30 minutes until the end of the perfusion, and stored at - 70˚C until examined for TNF-α levels by enzyme linked immuno-assay (ELISA). The perfused cotyledon in each placenta was weighed at the end of the perfusion. ELISA results were normalized for gram of perfused cotyledon in order to minimize the error that may result from variations in perfused cotyledon weights from different placentas.

In order to minimize the possible effect of labor on placental release of cytokines, each normotensive or pre-eclamptic placental cotyledon, either after vaginal or caesarean delivery, was perfused for 30 minutes with lactated Ringer’s [Hartman solution (Teva Medical, Ashdod, Israel)] followed by 30 minutes of perfusion with medium enriched with albumin (1 gr/L), glucose (1 g/L) and heparin (10 IU/mL). Subsequently, the perfusion medium in the fetal and maternal compartments was exchanged with fresh, enriched medium.

Validation of placental integrity for each experiment was established throughout the experimental period by ensuring that the rate of perfusate input in both the maternal and fetal circuits equaled the rate of output, and that histological examination of the cotyledon at the end of each experiment revealed no significant morphological changes.

Examination of collected samples by ELISA

The first antibodies were incubated overnight in 96-well ELISA plates at 4˚C, followed by washing and addition of blocking buffer, consisting of 10% fetal calf serum (Beit HaEmek, Israel) in phosphate-buffered saline (PBS) (Beit HaEmek, Israel) for 2 h at 37˚C. Thereafter, blocking buffer was removed, and samples or recombinant TNF-α were added for 1 h incubation at 37˚C. After washing, the second antibodies were added and plates were incubated for an additional 1 h at 37˚C. After incubation, plates were washed, and Streptavidin HRP was added for 30 minutes (min) at 37˚C. After washing, tetramethyl benzidine (TMB) (DakoCytomation Inc., CA, USA) was added for 15 min and the reaction was stopped by adding 2N H₂SO₄. Optical absorbance was read using an ELISA reader (Model 550) (Biorad, CA, USA) at 450 nm.

Statistical analysis

Results

Pre-eclamptic placentas secrete higher levels of TNF-α than normotensive placentas

As shown in figure 1A and B and summarized in table 2, TNF-α levels in the fetal and maternal circulations of normotensive and pre-eclamptic placentas perfused with control medium increased with time, but were significantly higher in the maternal circulations as compared to the fetal circulations, at the end of perfusion. In normotensive placentas, TNF-α levels in the maternal circulation were 6.73 ± 1.11 pg/mL/g of cotyledon, as compared to the fetal circulation (0.43 ± 0.29 pg/mL/g of cotyledon; p < 0.01). Furthermore, TNF-α levels in the maternal circulation of pre-eclamptic placentas were 14.28 ± 2.69 pg/mL/g of cotyledon, as compared to the fetal circulation (1.74 ± 1.14 pg/mL/g of cotyledon; p < 0.001) (table 2).
Table 1 Demographic and clinical characteristics, and neonatal outcomes of study subjects

Effect of MgSO₄ on TNF-α secretion into the fetal and maternal circulations of perfused human normotensive and pre-eclamptic placentas

Perfusion with MgSO₄, resulted in decreased TNF-α levels in both the fetal and maternal circulations of pre-eclamptic placentas as compared to perfusion with the control medium (fetal: 0.89 ± 0.09 pg/mL/g of cotyledon versus 1.6 ± 0.59 pg/mL/g of cotyledon, after 330 min of perfusion; p < 0.05, and maternal: 4.74 ± 2.78 pg/mL/g of cotyledon versus 14.28 ± 2.69 pg/mL/g of cotyledon at the end of perfusion; p < 0.01, respectively) (figure 3A and B).

Moreover, TNF-α levels detected at the end of perfusion in the fetal and maternal circulations of pre-eclamptic placentas exposed to MgSO₄ (0.89 ± 0.09 pg/mL/g of cotyledon and 4.74 ± 2.78 pg/mL/g of cotyledon, respectively), were statistically equal to TNF-α levels in the fetal and maternal circulations of normotensive placentas perfused with control medium (0.43 ± 0.29 pg/mL/g of cotyledon and 6.73 ± 1.11 pg/mL/g of cotyledon, respectively) (figure 4A and B).

In the presence of MgSO₄, no significant differences were detected in TNF-α secretion into the fetal as compared to the maternal circulation of normotensive placentas (table 2). However, pre-eclamptic placentas perfused with MgSO₄ still secreted significantly higher levels of TNF-α into the maternal circulation (4.74 ± 2.78 pg/mL/g of cotyledon) as compared to the fetal circulation (0.54 ± 0.24 pg/mL/g of cotyledon; p < 0.05).

Discussion

Using an ex vivo, placental perfusion system, this is the first report showing over-secretion of TNF-α by pre-eclamptic as compared to normotensive placentas, and that MgSO₄ significantly decrease these higher levels of TNF-α. Our current study suggests that pre-eclamptic placentas secrete higher levels of the inflammatory cytokine TNF-α into both the fetal and the maternal circulation, as compared to normotensive placentas. Exposure to MgSO₄ significantly reduces the induced TNF-α secretion into both circulations.

Although, elevated levels of TNF-α and its soluble receptors in serum and placentas of pre-eclamptic women have been recently reported, others have not found changes in placental TNF-α expression [9], suggesting that the increase in serum TNF-α levels seen in pre-eclampsia is not caused by the placenta. Our results show that high levels of TNF-α were detected in the maternal circulation of pre-eclamptic placenta over six hours of perfusion. These findings suggest that the placenta may be responsible for the increase in serum TNF-α levels, thus contributing to the pathophysiology of pre-eclampsia. The discordance between our results and other published data may be related to different techniques used (placental perfusion versus immediate sampling of placental tissue after placental delivery), or due to different assay systems used to measure TNF-α levels.

Higher plasma levels of TNF-α were found to be associated with the severity of the pre-eclampsia, suggesting TNF-α as possible marker of severity [29]. TNF-α is also known as one of the vital circulating factors that mediate endothelial cell activation/dysfunction. It does so through regulation of a wide range of physiological processes in these cells, including prostaglandin production, expression of adhesion molecules and alteration of the balance between oxidant and anti-oxidant molecules [30]. Taking into consideration our current results, the placenta may play a key role in determining the severity of pre-eclampsia in its later stages.

Our results also suggest that the pre-eclamptic placenta secretes increased levels of TNF-α into the fetal circulation. Since TNF-α, as well as other inflammatory cytokines, have been linked to increased risk of CP [24-26], our data may add support to the leading role of pre-eclampsia in neonatal morbidities. Increased TNF-α secretion into the fetal circulation may result in increased levels of this cytokine in the fetal serum, increasing the risk of neonatal brain damage.

The decreased levels of TNF-α secreted into the maternal circulation of pre-eclamptic placentas achieved with MgSO₄, may indicate a possible therapeutic effect in pre-eclampsia. MgSO₄ it may act as an anti-inflammatory agent on the feto-placental interface by reducing the secretion of the inflammatory cytokine TNF-α. In this manner, MgSO₄ may reduce the inflammatory process in the blood vessels of the pre-eclamptic women. On the other hand, the similar reducing effect of MgSO₄ on the secretion of TNF-α into the fetal circulation suggests a possible neonatal neuroprotective effect in pre-eclampsia.

Our results are in agreement with recently reported data regarding the effect of MgSO₄ on inflammatory cytokine production. Makhlouf et al. showed that MgSO₄ suppresses endotoxin-stimulated IL-8 production by amniotic and decidual cells in vitro [31]. Moreover, it has been suggested that magnesium deficiency is associated with increased inflammatory responses [32-34].

Previously, we have shown that MgSO₄ and Ang-II may differently affect the capacity of the fetal and the maternal compartments of normotensive human placenta to secrete TNF-α [27]. In the current study, we have shown similar levels of TNF secreted from the maternal side of normotensive perfused placentas in the presence or absence of MgSO₄. However, in the previous study, we compared the effect of MgSO4 to Ang-II or AngII + MgSO₄ on TNF secretion. In addition, a different protocol was used in the current study (MgSO₄ was added after three hours of perfusion in the previous study; in the current study, MgSO₄ was added from the beginning of the perfusion), because of the sensitivity of the pre-eclamptic placentas used. This could lead to differing interpretations between the two studies.

Nuclear factor kappa B (NFκB) is a nuclear transcription factor that regulates numerous inflammatory mediators. Recently, Rochelson et al. reported that MgSO₄ inhibits endothelial cell activation by reducing the nuclear translocation of NFκB and by protecting its cytoplasmic inhibitor (IκBα) from degradation [35]. These data may provide the missing link to explain the mechanism by which MgSO₄ reduces the inflammatory state in the blood vessels of pre-eclamptic patients.

The fact that TNF-α secretion by the normotensive placenta is not affected MgSO₄ suggests that the effect of MgSO₄ is specific to the pre-eclamptic placenta. This might indicate a different mechanism of regulation of TNF-α secretion in pre-eclampsia as compared to normotensive pregnancies.

In conclusion, MgSO₄ may normalize increased secretion of placental TNF-α in pre-eclampsia. This suggests a possible therapeutic effect of MgSO₄ that acts by reducing TNF-α levels in maternal and fetal circulations, thereby reducing maternal endothelial dysfunction and improving neonatal outcome in pre-eclampsia. 

Acknowledgments

Disclosure. None of the authors has any conflict of interest to disclose.

References

2 Lain KY, Roberts JM. Contemporary concepts of the pathogenesis and management of preeclampsia. JAMA 2002; 287: 3183-6.

3 Sibai BM, Caritis S, Hauth J. What we have learned about preeclampsia. Semin Perinatol 2003; 27: 239-46.

4 Kharfi A, Giguère Y, Sapin V, Massé J, Dastugue B, Forest JC. Trophoblastic remodeling in normal and preeclamptic pregnancies: implication of cytokines. Clin Biochem 2003; 36: 323-31.

5 Page NM. The endocrinology of pre-eclampsia. Clin Endocrinol (Oxf) 2002; 57: 413-23.

6 LaMarca BD, Ryan MJ, Gilbert JS, Murphy SR, Granger JP. Inflammatory cytokines in the pathophysiology of hypertension during preeclampsia. Curr Hypertens Rep 2007; 9: 480-5.

7 Rusterholz C, Hahn S, Holzgreve W. Role of placentally produced inflammatory and regulatory cytokines in pregnancy and the etiology of preeclampsia. Semin Immunopathol 2007; 29: 151-62.

8 Brewster JA, Orsi NM, Gopichandran N, McShane P, Ekbote UV, Walker JJ. Gestational effects on host inflammatory response in normal and pre-eclamptic pregnancies. Eur J Obstet Gynecol Reprod Biol 2008; 140: 21-6.

9 Haider S, Knöfler M. Human Tumour Necrosis Factor: Physiological and Pathological Roles in Placenta and Endometrium. Placenta 2008; doi:10.1016/j.placenta.2008.10.012.

10 Conrad KP, Benyo DF. Placental cytokines and the pathogenesis of preeclampsia. Am J Reprod Immunol 1997; 37: 240-9.

11 Conrad KP, Miles TM, Benyo DF. Circulating levels of immunoreactive cytokines in women with preeclampsia. Am J Reprod Immunol 1998; 40: 102-11.

12 Wang Y, Walsh SW. TNF alpha concentrations and mRNA expression are increased in preeclamptic placentas. J Reprod Immunol 1996; 32: 157-69.

13 Benyo DF, Smarason A, Redman CW, Sims C, Conrad KP. Expression of inflammatory cytokines in placentas from women with preeclampsia. J Clin Endocrinol Metab 2001; 86: 2505-12.

14 Hayashi M, Ueda Y, Yamaguchi T, et al. Tumor necrosis factor-alpha in the placenta is not elevated in pre-eclamptic patients despite its elevation in peripheral blood. Am J Reprod Immunol 2005; 53: 113-9.

15 Witlin AG, Sibai BM. Magnesium sulfate therapy in preeclampsia and eclampsia. Obstet Gynecol 1998; 92: 883-9.

16 Lucas MJ, Leveno KJ, Cunningham FG. A comparison of magnesium sulfate with phenytoin for the prevention of eclampsia. N Engl J Med 1995; 333: 201-5.

17 Duley L, Henderson-Smart D. Magnesium sulphate versus phenytoin for eclampsia. Cochrane Database Syst Rev 2003; 4: CD000128.

18 Holcberg G, Sapir O, Hallak M, et al. Selective vasodilator effect of magnesium sulfate in human placenta. Am J Reprod Immunol 2004; 51: 192-7.

19 Roberts JM. Magnesium for preeclampsia and eclampsia. N Engl J Med 1995; 333: 250-1.

20 Doyle LW, Crowther CA, Middleton P, Marret S. Magnesium sulphate for women at risk of preterm birth for neuroprotection of the fetus. Cochrane Database Syst Rev 2007; 3: CD004661.

21 Nelson KB, Grether JK. Can magnesium sulfate reduce the risk of cerebral palsy in very low birth weight infants? Pediatrics 1995; 95: 263-7.

22 Hauth JC, Goldenberg RL, Nelson KG, Rouse DJ. Reduction of cerebral palsy with maternal magnesium sulfate treatment in newborns weighing 500-1,000 G. Am J Obstet Gynecol 1995; 172: 419.

23 Espinoza MI, Parer JT. Mechanisms of asphyxial brain damage, and possible pharmacologic interventions, in the fetus. Am J Obstet Gynecol 1991; 164: 1582-9.

24 Dammann O, Leviton A. Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res 1997; 42: 1-8.

25 Yoon BH, Jun JK, Romero R, et al. Amniotic fluid inflammatory cytokines (interleukin-6, interleukin-1beta tumor necrosis factor-alpha), neonatal brain white matter lesions, and cerebral palsy. Am J Obstet Gynecol 1997; 177: 19-26.

26 Nelson KB, Dambrosia JM, Grether JK, Phillips TM. Neonatal cytokines and coagulation factors in children with cerebral palsy. Ann Neurol 1998; 44: 665-75.

27 Holcberg G, Amash A, Sapir O, et al. Different effects of magnesium sulfate and angiotensin II on the capacity of the fetal and maternal compartments of normal human placenta to secrete TNF-alpha and IL-6. J Reprod Immunol 2006; 69: 115-25.

28 Holcberg G, Amash A, Sapir O, et al. Perfusion with lipopolysaccharide differently affects the secretion of interleukin-1 beta and interleukin-1 receptor antagonist by term and preterm human placentae. Placenta 2008; 29: 593-601.

29 Peracoli JC, Rudge MV, Peracoli MT. Tumor necrosis factor-alpha in gestation and puerperium of women with gestational hypertension and pre-eclampsia. Am J Reprod Immunol 2007; 57: 177-85.

30 Meekins JW, McLaughlin PJ, West DC, McFadyen IR, Johnson PM. Endothelial cell activation by tumour necrosis factor-alpha (TNF-alpha) and the development of pre-eclampsia. Clin Exp Immunol 1994; 98: 110-4.

31 Makhlouf MA, Simhan HN. Effect of tocolytics on interleukin-8 production by human amniotic and decidual cells. J Reprod Immunol 2006; 69: 1-7.

32 Bernardin D, Nasulewic A, Mazur A, Maier JA. Magnesium and microvascular endothelial cells: a role in inflammation and angiogenesis. Front Biosci 2005; 10: 1177-82.

33 Maier JA, Malpuech-Brugere C, Zimowska W, Rayssiguier Y, Mazur A. Low magnesium promotes endothelial cell dysfunction: implications for atherosclerosis, inflammation and thrombosis. Biochim Biophys Acta 2004; 1689: 13-21.

34 Nakagawa M, Oono H, Nishio A. Enhanced production of IL-1beta and IL-6 following endotoxin challenge in rats with dietary magnesium deficiency. J Vet Med Sci 2001; 63: 467-9.

35 Rochelson B, Dowling O, Schwartz N, Metz CN. Magnesium sulfate suppresses inflammatory responses by human umbilical vein endothelial cells (HuVECs) through the NFkappaB pathway. J Reprod Immunol 2007; 73: 101-7.