- Email: [email protected]
Magnesium sulfate suppresses inflammatory responses by human umbilical vein endothelial cells (HuVECs) through the NFκB pathway
Journal of Reproductive Immunology 73 (2007) 101–107
Magnesium sulfate suppresses inflammatory responses by human umbilical vein endothelial cells (HuVECs) through the NFB pathway Burton Rochelson a , Oonagh Dowling b , Nadav Schwartz a , Christine N. Metz b,∗ a
Division of Maternal-Fetal Medicine, North Shore University Hospital, 300 Community Drive, Manhasset, NY 11030, USA b The Susan & Herman Merinoff Center for Patient Oriented Research, The Feinstein Institute for Medical Research North Shore-LIJ Health System, 350 Community Drive, Manhasset, NY 11030, USA Received 10 March 2006; received in revised form 26 June 2006; accepted 29 June 2006
Abstract Dysfunctional endothelial cell activation and cytokines are implicated in preterm labor, a condition commonly treated with the tocolytic agent, magnesium sulfate (MgSO4 ). Based on recent findings showing the inflammatory effects of magnesium deficiency, we examined the effect of MgSO4 on human umbilical vein endothelial cell (HuVEC) inflammatory responses in vitro. HuVECs isolated from term umbilical cords were incubated with MgSO4 prior to stimulation with lipopolysaccharide (LPS) and then assessed for endothelial cell activation. Endothelial cell supernatants were assayed for inflammatory mediator production (interleukin8; IL-8), and endothelial cell-associated intercellular adhesion molecule (ICAM-1) expression was determined. In the absence of LPS stimulation, MgSO4 had no effect on HuVEC responses. Treatment of HuVECs with MgSO4 prior to LPS stimulation inhibited inflammatory mediator production (p < 0.05) and cell adhesion molecule expression (p < 0.05) in a dose-dependent manner. Mechanistic studies showed that MgSO4 reduced NFB nuclear translocation and protected cytoplasmic IB␣ from degradation in LPS-treated HuVECs. In conclusion, MgSO4 inhibits endothelial cell activation, as measured by levels of IL-8 and ICAM-1 expression, via NFB. Our results support the hypothesis that MgSO4 treatment may function as an anti-inflammatory agent during preterm labor. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Preterm labor; Magnesium sulfate; Endothelial cell activation; Cytokines
1. Introduction Preterm labor is one of the major contributing factors for neonatal morbidity and mortality in the developed world. Magnesium sulfate (MgSO4 ) is commonly used for suppressing preterm labor (Caritis, 2005; Lewis, 2005). However, the exact mechanism(s) of its action
∗ Corresponding author. Tel.: +1 516 562 3403; fax: +1 516 562 1022. E-mail address: [email protected] (C.N. Metz).
remains unclear. Preterm labor is characterized by an enhanced maternal systemic inflammatory response (Park et al., 2005) with the enhanced production of proinflammatory mediators, including IL-6, IL-8 and TNF (Hagberg et al., 2005; Romero et al., 2005; Wennerholm et al., 1998) and, in many cases, with intrauterine infections (Hagberg et al., 2005; Park et al., 2005; Romero et al., 1988). Infusion of inflammatory cytokines induces preterm labor and delivery in experimental models (Bry and Hallman, 1993). Based on the observations that magnesium deficiency is associated with increased inflammatory responses
0165-0378/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jri.2006.06.004
102
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
(Bernardin et al., 2005; Maier et al., 2004a; Nakagawa et al., 2001) and that an enhanced maternal inflammatory response with endothelial cell dysfunction is implicated in preterm labor (Park et al., 2005), we examined the effect of MgSO4 on HuVEC inflammatory responses in vitro. We observed that MgSO4 inhibited endothelial cell activation in vitro and identified that MgSO4 reduced the nuclear translocation of nuclear factor kappa B (NFB). These observations have potentially broad implications because NFB, a transcription factor typically associated with inflammation and infection, has been recently implicated as an important regulator of human labor (Lindstrom and Bennett, 2005). 2. Materials and methods This study received exempt status by the Institutional Review Board (IRB) of North Shore University Hospital because it involved the use of anonymous, normal, pathogen-free, term umbilical cords for the isolation of human umbilical vein endothelial cells (HuVECs). 2.1. HuVEC isolation and culture conditions HuVECs were isolated using collagenase to digest the subendothelial basement membrane, as previously described (Jaffe et al., 1973). HuVEC preparations were >98% pure based on acetylated LDL uptake and CD31 expression by flow cytometry. HuVECs were grown on gelatin-coated flasks (0.15%) in M199 media containing 10% fetal calf serum, endothelial cell growth supplement (90 g/ml; Sigma, St. Louis, MO), heparin (100 g/ml; Sigma), penicillin, streptomycin and l-glutamine (2 mM). HuVECs were sub-cultured using Trypsin/EDTA Reagent Pack (Cambrex, Walkersville, MD). 2.2. HuVEC stimulation assays For MgSO4 stimulation assays, HuVECs (passages 3–6) were grown to confluency on gelatin-coated 96well plates; all assays were performed in M199 containing 10% fetal calf serum, penicillin, streptomycin and l-glutamine (media). HuVECs were treated with MgSO4 (1–20 mM prepared in media; American Pharmaceutical Partners, Schaumburg, IL), terbutaline (10−5 to 10−8 M; Bedford Laboratories, Bedford, OH) or dexamethasone (10–20 M prepared in media, Elkins-Sinn, Cherry Hill, NJ) for 30–60 min prior to either no stimulation or stimulation with lipopolysaccharide (LPS) (0.1 g/ml; LPS isolated from E. coli serotype 0111:B4
(lot number 63H4011, Sigma)). Control (vehicle) wells were treated with media containing phosphate-buffered saline (PBS) as a vehicle control for MgSO4 . In separate experiments, HuVECs were treated with MgSO4 at the time of LPS treatment or 0.5 h post-LPS treatment. Following overnight incubation, cell-free culture supernatants were collected and analyzed for IL8 production by ELISA (R&D Systems, Minneapolis, MN). Cells were assessed for intercellular adhesion molecule-1 (ICAM-1) expression by cell-based ELISA methods, as previously described (Saeed et al., 2005). To control for cytotoxicity/viability, parallel cultures were stained using the 3-[4,5-dimethylthiazol-2-yl]-2,5diphenyltetrazolium bromide (MTT) reduction assay, as previously described (Saeed et al., 2005). Data are presented as: mean IL-8 levels (pg/ml ± S.D.); mean cellassociated ICAM-1 expression (OD ± S.D.); and mean viability (OD ± S.D.). 2.3. NFκB and IκBα analyses Confluent HuVEC monolayers were treated with MgSO4 (2–10 mM, prepared in media as described above) for 0.5–1 h prior to the addition of LPS (0.1–1 g/ml). Control (vehicle) wells were treated with PBS diluted in media for 0.5–1 h prior to the addition of LPS (0.1–1.0 g/ml). Nuclear and cytoplasmic extracts were prepared 1 h post-LPS addition using the NE-PER kit (Pierce Chemical Co., Rockford, IL). Nuclear and cytoplasmic preparations (∼10 g/lane) were electrophoresed, transferred to PVDF membranes and probed with antibody to NFB (p65 (Rel A), Cell Signaling Technology, Beverly, MA) and IB␣ (Santa Cruz Biotechnology, Santa Cruz, CA), respectively, or with antibodies to control nuclear (Lamin A/C; Santa Cruz Biotechnology) and cytoplasmic (actin; Chemicon International, Temecula, CA) proteins. Following incubation with HRP-conjugated secondary antibody, specific proteins were revealed using ECL reagent (Amersham Pharmacia, Piscataway, NJ). Band densities were determined using the NIH Image Program and the ratios of the specific:control bands are shown. 2.4. Statistics Each set of experiments was repeated three times (n = 4 samples per condition). For each of the three outcome variables (IL-8, ICAM-1 and viability/cytotoxicity), a two-way ANOVA was performed to compare the means for the various interventions, controlling for the day each experiment was performed using
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
103
SAS statistical software. The standard ANOVA assumptions of normality and equality of variance were checked and were satisfied for each variable; an ANOVA was considered significant if p < 0.05. The Dunnett’s test (using the adjusted means) was then used to make pairwise comparisons (vehicle versus treated) using SAS® statistical software (p < 0.05 was considered significant). 3. Results 3.1. Magnesium sulfate (MgSO4 ) treatment inhibits endothelial cell activation Under basal conditions, endothelial cells serve a barrier function and exhibit a non-inflammatory phenotype; they are not inert and do express constitutive measurable levels of both ICAM-1 and IL-8. In our first series of experiments, we examined the effect of MgSO4 on constitutive IL-8 production and ICAM-1 expression by HuVECs in vitro. We found that MgSO4 (1–20 mM) had no significant effect on constitutive IL-8 production and ICAM-1 expression by HuVECs (Fig. 1A and B). Dexamethasone (Dex), a potent anti-inflammatory agent, had no effect on IL-8 production and only a slight inhibitory effect on ICAM-1 expression was observed (Fig. 1A and B). These concentrations of MgSO4 and Dex were not cytotoxic to the cells (Fig. 1C). Activation of the endothelium during inflammation and infection is characterized by increased cell surface adhesion molecule expression, and enhanced chemokine and cytokine production required for leukocyte recruitment. We examined the effect of MgSO4 on endothelial cell activation using LPS as a stimulating agent. Treatment of HuVEC cells with MgSO4 (2.5–10 mM) prior to LPS stimulation inhibited IL-8 production (up to 55% inhibition) in a dose-dependent manner (Fig. 2A). Similarly, MgSO4 (2.5–10 mM) treatment of HuVEC cells suppressed ICAM-1 induction by LPS by approximately 40% (Fig. 2B). MgSO4 was more potent than Dex alone and the combination of Dex + MgSO4 was not additive (Fig. 2A and B). Additional experiments were performed to examine the effect of the time of MgSO4 administration on LPS-induced endothelial cell activation. A significant reduction in ICAM-1 expression (25%) was observed with simultaneous administration of MgSO4 (10 mM) and LPS when compared to vehicle-treated cells (Fig. 2C). When MgSO4 (10 mM) was added 30 min post-LPS, MgSO4 reduced ICAM-1 expression by approximately 15%; however, this reduction was not statistically significant (Fig. 2C). Similar results were observed with IL-8 production (data not shown).
Fig. 1. MgSO4 has no effect on basal IL-8 and ICAM-1 production by HuVECs. MgSO4 (1–20 mM) addition to HuVECs has no effect on (A) basal IL-8 production (pg/ml ± S.D.), (B) constitutive cell-associated ICAM-1 expression (OD ± S.D.), or (C) cytotoxicity, as determined by MTT assay (OD ± S.D.). The effect MgSO4 is compared to dexamethasone (Dex) treatment at two concentrations (10 and 20 M). Data from four experiments are shown.
3.2. Magnesium chloride (MgCl2 ) treatment inhibits endothelial cell activation To determine whether the effect of MgSO4 on endothelial cell activation was mediated through Mg cations, we assessed the effect of MgCl2 on IL-8 production and ICAM-1 expression by HuVEC cells following LPS stimulation. Similar to the inhibitory effect of MgSO4 on HuVEC activation, MgCl2 blocked HuVEC activation (as determined by IL-8
104
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
Fig. 3. MgCl2 inhibits HuVEC activation in a dose-dependent manner. MgCl2 (1–10 mM) pre-treatment of HuVECs significantly blocks (A) IL-8 production (pg/ml ± S.D.) and (B) ICAM-1 expression induced by LPS (100 ng/ml) (in the presence or absence of dexamethasone (Dex, 10 M)) (OD ± S.D.). * p <0.05 comparing treated vs. no treatment. Data from four experiments are shown.
Fig. 2. MgSO4 inhibits LPS (100 ng/ml) induced HuVEC activation in a dose-dependent manner. MgSO4 (1–10 mM) pre-treatment of HuVECs significantly blocks (A) IL-8 production (pg/ml ± S.D.), (B) ICAM-1 expression induced by LPS (in the presence or absence of dexamethasone (Dex, 10 M)) (OD ± S.D.) and (C) vehicle or MgSO4 (10 mM) was added to HuVECs 30 min prior to LPS, the same time as LPS or 30 min post-LPS addition and ICAM-1 expression was assessed (OD ± S.D.). * p < 0.05. Data from four experiments are shown.
production and ICAM-1 expression) induced by LPS in a dose-dependent manner (Fig. 3A and B, respectively). 3.3. Magnesium sulfate inhibits NFκB nuclear localization and protects IκBα from degradation Because the NFB pathway regulates many genes involved in endothelial cell activation, inflammation and
the onset of labor, we examined the effect of MgSO4 on NFB and IB family members. Treatment of HuVECs with MgSO4 inhibited the nuclear translocation of NFB induced by LPS stimulation in a dose-dependent manner (Fig. 4A). Consistent with this observation, further studies showed that MgSO4 protected IB␣ from LPSinduced degradation in the cytoplasm (Fig. 4B). 4. Discussion Magnesium sulfate is the most frequently used tocolytic agent for inhibiting preterm labor (Caritis, 2005). Despite its widespread use, there is remarkably little in the literature on the mechanism of action of magnesium sulfate in the treatment of preterm labor. An increasing number of studies suggest that preterm labor has its origins in an aberrant inflammatory response characterized by cytokine production and endothelial cell activation (Park et al., 2005). Recent studies show that magnesium deficiency is associated with an increased inflammatory response (Maier et al., 2004a; Nakagawa et al., 2001; Nasulewicz et al., 2004). We
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
Fig. 4. MgSO4 inhibits NFB nuclear translocation and IB␣ degradation in HuVECs. MgSO4 pre-treatment of HuVECs inhibits (A) NFB nuclear translocation and (B) IB␣ degradation in the cytoplasm induced by LPS (100–1000 ng/ml). Data are shown as (A) the ratio of NFB density to Lamin A/C, a control nuclear protein and (B) the ratio of IB␣ density to -actin, a control cytoplasmic protein.
investigated the possibility that magnesium sulfate therapy exerts anti-inflammatory effects on human umbilical vein endothelial cells. Based on previous in vitro studies using MgSO4 (Maier et al., 2004b) and plasma concentrations following MgSO4 therapy, we chose MgSO4 concentrations ≤10 mM that were non-cytotoxic. LPS was chosen because components of the bacterial cell wall (including LPS) mimic the infectious state to promote preterm delivery in experimental models (Elovitz and Mrinalini, 2004). The results of our study showing the suppressive effects of MgSO4 on endothelial cell inflammatory responses in vitro support our hypothesis that MgSO4 exerts anti-inflammatory responses. Numerous cytokines, e.g. IL-8, IL-6 and TNF and markers of endothelial cell activation, e.g. ICAM-1 and VCAM-1, are elevated during preterm labor and birth (Fischer et al., 2001; Hagberg et al., 2005; Marvin et al., 2000; Park et al., 2005; Romero et al., 2005; Wennerholm et al., 1998). IL-8 is elevated in cervical secretions and amniotic fluid in cases of preterm labor (Gonzalez et al., 2005; Holst et al., 2005; Wennerholm et al., 1998). In addition, IL-8 and IL-1 induce cervical ripening in guinea pigs (Chwalisz et al., 1994; Chwalisz, 1994). In
105
many cases, pathogenic agents enter the uterus and elicit inflammatory responses within the chorion and amnion (chorioamnionitis) and/or the umbilical cord (funisitis) to induce preterm labor (Romero et al., 2005). A recent study by D’Alquen et al. (2005) demonstrated endothelial cell activation within the umbilical cord during chorioamnionitis and funisitis. In the presence of chorioamnionitis associated with funisitis, concentrations of soluble ICAM-1 (marker of endothelial cell activation) are elevated in fetal serum, venous endothelium and amniotic epithelium (D’Alquen et al., 2005). Similarly, chorioamnionitis is associated with elevated levels of IL-8, IL-6 and TNF in the umbilical cord blood and leukocyte infiltration within the placenta (mediated via activated endothelial cells) (Dollner et al., 2002). Despite the fact that magnesium sulfate is the most commonly used tocolytic agent in the US, there are reports of adverse maternal and fetal effects of long-term/high dose MgSO4 therapy (Caritis, 2005; Mittendorf et al., 2003, 2006). The mechanism of action of MgSO4 is thought to be secondary to a competitive inhibition of calcium and its effect on smooth muscle contraction (Lewis, 2005). The potential mechanism of action of magnesium as an anti-inflammatory agent is supported by studies demonstrating a relationship between magnesium deficiency and endothelial cell dysfunction (Maier et al., 2004a). Magnesium deficiency has been associated with increased endothelial cell activation (adhesion molecule expression, increased cytokine production) (Maier et al., 2004a), impaired endothelial cell proliferation and migration (Bernardin et al., 2005), and enhanced cytokine release (basal and LPS-induced) by alveolar macrophages in vivo (Nakagawa et al., 2001). In vivo models of Mg deficiency are also associated with inflammation (Nasulewicz et al., 2004). Our findings showing MgSO4 -mediated IL-8 inhibition by stimulated HuVECs are consistent with recent studies showing the inhibitory effects MgSO4 on IL-8 production by human amniotic and decidual cells (Makhlouf and Simhan, 2006). However, the mechanism of action of MgSO4 was not investigated in this previous study. Magnesium therapy has proven to be beneficial for numerous inflammatory conditions. For example, aerosolized magnesium sulfate is used for the clinical treatment of acute severe asthma (Blitz et al., 2005), magnesium supplementation in vivo preserves the integrity of the blood brain barrier during experimental sepsis (Esen et al., 2005), and magnesium infusions protect against experimental stroke (Lee et al., 2005). Although magnesium is known to be a potent neurosedative, the antiinflammatory action of Mg may also help to explain its
106
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
efficacy in seizure prophylaxis in pre-eclampsia, a condition which has also been associated with endothelial cell dysfunction and enhanced inflammatory cytokine production (Roberts and Lain, 2002; Roberts, 1998). The effect of timing of MgSO4 administration on LPS-induced endothelial cell activation was investigated. MgSO4 , when added 30 min prior to LPS, significantly reduced ICAM-1 expression (up to 40%) by HuVECs (Fig. 2B and C). When MgSO4 was given simultaneously with LPS, we observed a reduced, but still significant, effect on ICAM-1 expression (25%, Fig. 2C). However, when MgSO4 was added 30 min. post-LPS, a slight but insignificant reduction (15%) was observed (Fig. 2C). Similar results were observed with IL-8 (data not shown). Based on our findings that MgSO4 -inhibited IB␣ degradation and NFB translocation (which begins immediately post-LPS addition, peaking within 0.5–1 h post-LPS addition), the timing of MgSO4 administration is important and should precede the initiation of the NFB cascade by LPS for most effective results. Mg deficiency alone is associated with an inflammatory state in vivo (Nasulewicz et al., 2004), suggesting that inadequate Mg stores may exacerbate host inflammatory responses during infection. Furthermore, these data support previous proposals that a sufficient Mg balance may be critically important during gestation (Durlach, 2004), particularly prior to the exposure of invading microorganisms. Additional in vivo experiments are required to test these hypotheses. We also examined the effect of another tocolytic agent, terbutaline (10−5 to 10−8 M), on endothelial cell activation; we found no significant effects of terbutaline on IL-8 production or ICAM-1 expression by LPStreated HuVECs. Our observations are consistent with Makhlouf and Simhan who found no effect of terbutaline on IL-8 production by human amniotic and decidual cells (Makhlouf and Simhan, 2006). Interestingly, terbutaline has been shown to exert anti-inflammatory effects. Terbutaline suppresses inflammatory mediator production by stimulated polymorphonuclear leukocytes (Zheng et al., 1991) and reduces TNF production by LPS-stimulated mesangial cells (Nakamura et al., 2000). In our study, magnesium sulfate suppresses the proinflammatory actions of LPS on human umbilical vein endothelial cells in vitro (Fig. 2A–C). In fact, the immunosuppressive activity of MgSO4 equaled that of dexamethasone, although their use together did not appear to be synergistic. Magnesium ions (Mg2+ ) appear to be critical for the observed effect of MgSO4 as MgCl2 had similar anti-inflammatory effects on HuVEC activation (Fig. 3A and B). Interestingly, one report addresses
the use of MgCl2 in place of MgSO4 because the clinical and pharmacological effects of the two agents overlap, while MgCl2 exhibits reduced tissue toxicity (Durlach et al., 2005). Based on our observations with umbilical cord endothelial cells, the anti-inflammatory action of MgSO4 at the molecular level is through its protective effect on cytoplasmic IB␣ degradation (Fig. 4B) and the subsequent inhibition of LPS-mediated NFB nuclear translocation (Fig. 4A). This is the first report indicating that NFB activation is sensitive to manipulation by Mg. Based on the fact that Mg participates in over 300 biochemical reactions as a co-factor for hundreds of enzyme systems (Wolf et al., 2003), it is not surprising that it impacts one of the most influential inflammatory pathways. NFB is a nuclear transcription factor classically coupled with inflammation and infection that regulates the production of numerous inflammatory mediators. The key role of NFB activation in regulating human labor has been identified (Lindstrom and Bennett, 2005) and, therefore, the potential role of MgSO4 as an antiinflammatory agent and regulator of NFB activation in various maternal and fetal compartments during preterm labor warrants further investigation. Acknowledgments The authors thank Ms. Tina Nemeroff and Dr. Xiangying Xue for their assistance in assaying IL-8 production and ICAM-1 expression by HuVECs and Dr. Martin Lesser and Ms. Meredith Lukin (Biostatistics Unit) for their assistance with data analyses. References Bernardin, D., Nasulewic, A., Mazur, A., Maier, J.A., 2005. Magnesium and microvascular endothelial cells: a role in inflammation and angiogenesis. Front. Biosci. 10, 1177–1182. Blitz, M., Blitz, S., Hughes, R., Diner, B., Beasley, R., Knopp, J., Rowe, B.H., 2005. Aerosolized magnesium sulfate for acute asthma: a systematic review. Chest 128, 337–344. Bry, K., Hallman, M., 1993. Transforming growth factor-beta 2 prevents preterm delivery induced by interleukin-1 alpha and tumor necrosis factor-alpha in the rabbit. Am. J. Obstet. Gynecol. 168, 1318–1322. Caritis, S., 2005. Adverse effects of tocolytic therapy. Br. J. Obstet. Gynaecol. 112 (Suppl 1), 74–78. Chwalisz, K., 1994. The use of progesterone antagonists for cervical ripening and as an adjunct to labour and delivery. Hum. Reprod. 9 (Suppl 1), 131–161. Chwalisz, K., Benson, M., Scholz, P., Daum, J., Beier, H.M., HegeleHartung, C., 1994. Cervical ripening with the cytokines interleukin 8, interleukin 1 beta and tumour necrosis factor alpha in guineapigs. Hum. Reprod. 9, 2173–2181. D’Alquen, D., Kramer, B.W., Seidenspinner, S., Marx, A., Berg, D., Groneck, P., Speer, C.P., 2005. Activation of umbilical cord
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107 endothelial cells and fetal inflammatory response in preterm infants with chorioamnionitis and funisitis. Pediatr. Res. 57, 263–269. Dollner, H., Vatten, L., Halgunset, J., Rahimipoor, S., Austgulen, R., 2002. Histologic chorioamnionitis and umbilical serum levels of pro-inflammatory cytokines and cytokine inhibitors. Br. J. Obstet. Gynaecol. 109, 534–539. Durlach, J., 2004. New data on the importance of gestational Mg deficiency. J. Am. Coll. Nutr. 23, 694S–700S. Durlach, J., Guiet-Bara, A., Pages, N., Bac, P., Bara, M., 2005. Magnesium chloride or magnesium sulfate: a genuine question. Magnes. Res. 18, 187–192. Elovitz, M.A., Mrinalini, C., 2004. Animal models of preterm birth. Trends Endocrinol. Metab. 15, 479–487. Esen, F., Erdem, T., Aktan, D., Orhan, M., Kaya, M., Eraksoy, H., Cakar, N., Telci, L., 2005. Effect of magnesium sulfate administration on blood–brain barrier in a rat model of intraperitoneal sepsis: a randomized controlled experimental study. Crit. Care 9, R18–R23. Fischer, D.C., Winkler, M., Ruck, P., Poth, D., Kemp, B., Rath, W., 2001. Localization and quantification of adhesion molecule expression in the lower uterine segment during premature labor. J. Perinat. Med. 29, 497–505. Gonzalez, B.E., Ferrer, I., Valls, C., Borras, M., Lailla, J.M., 2005. The value of interleukin-8, interleukin-6 and interleukin-1beta in vaginal wash as predictors of preterm delivery. Gynecol. Obstet. Invest. 59, 175–178. Hagberg, H., Mallard, C., Jacobsson, B., 2005. Role of cytokines in preterm labour and brain injury. Br. J. Obstet. Gynaecol. 112 (Suppl. 1), 16–18. Holst, R.M., Mattsby-Baltzer, I., Wennerholm, U.B., Hagberg, H., Jacobsson, B., 2005. Interleukin-6 and interleukin-8 in cervical fluid in a population of Swedish women in preterm labor: relationship to microbial invasion of the amniotic fluid, intra-amniotic inflammation, and preterm delivery. Acta Obstet. Gynecol. Scand. 84, 551–557. Jaffe, E.A., Nachman, R.L., Becker, C.G., Minick, C.R., 1973. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J. Clin. Invest. 52, 2745–2756. Lee, E.J., Lee, M.Y., Chang, G.L., Chen, L.H., Hu, Y.L., Chen, T.Y., Wu, T.S., 2005. Delayed treatment with magnesium: reduction of brain infarction and improvement of electrophysiological recovery following transient focal cerebral ischemia in rats. J. Neurosurg. 102, 1085–1093. Lewis, D.F., 2005. Magnesium sulfate: the first-line tocolytic. Obstet. Gynecol. Clin. North Am. 32, 485–500. Lindstrom, T.M., Bennett, P.R., 2005. The role of nuclear factor kappa B in human labour. Reproduction 130, 569–581. Maier, J.A., Malpuech-Brugere, C., Zimowska, W., Rayssiguier, Y., Mazur, A., 2004a. Low magnesium promotes endothelial cell dysfunction: implications for atherosclerosis, inflammation and thrombosis. Biochim. Biophys. Acta 1689, 13–21.
107
Maier, J.A., Bernardini, D., Rayssiguier, Y., Mazur, A., 2004b. High concentrations of magnesium modulate vascular endothelial cell behaviour in vitro. Biochim. Biophys. Acta 1689, 6–12. Makhlouf, M.A., Simhan, H.N., 2006. Effect of tocolytics on interleukin-8 production by human amniotic and decidual cells. J. Reprod. Immunol. 69, 1–7. Marvin, K.W., Keelan, J.A., Coleman, M.A., McCowan, L.M., Zhou, R.L., Mitchell, M.D., 2000. Intercellular adhesion molecule-1 (ICAM-1) in cervicovaginal fluid of women presenting with preterm labor: predictive value for preterm delivery. Am. J. Reprod. Immunol. 43, 264–271. Mittendorf, R., Dammann, O., Lee, K.S., 2006. Brain lesions in newborns exposed to high-dose magnesium sulfate during preterm labor. J. Perinatol. 26, 57–63. Mittendorf, R., Pryde, P.G., Lee, K.S., 2003. Association between use of antenatal magnesium sulfate in preterm labor and adverse health outcomes in infants. Am. J. Obstet. Gynecol. 189, 613. Nakagawa, M., Oono, H., Nishio, A., 2001. Enhanced production of IL1beta and IL-6 following endotoxin challenge in rats with dietary magnesium deficiency. J. Vet. Med. Sci. 63, 467–469. Nakamura, A., Imaizumi, A., Kohsaka, T., Yanagawa, Y., Johns, E.J., 2000. Beta2-adrenoceptor agonist suppresses tumour necrosis factor production in rat mesangial cells. Cytokine 12, 491–494. Nasulewicz, A., Zimowska, W., Bayle, D., Dzimira, S., Madej, J., Rayssiguier, Y., Opolski, A., Mazur, A., 2004. Changes in gene expression in the lungs of Mg-deficient mice are related to an inflammatory process. Magnes. Res. 17, 259–263. Park, J.S., Park, C.W., Lockwood, C.J., Norwitz, E.R., 2005. Role of cytokines in preterm labor and birth. Minerva Ginecol. 57, 349–366. Roberts, J.M., 1998. Endothelial dysfunction in preeclampsia. Semin. Reprod. Endocrinol. 16, 5–15. Roberts, J.M., Lain, K.Y., 2002. Recent insights into the pathogenesis of pre-eclampsia. Placenta 23, 359–372. Romero, R., Erez, O., Espinoza, J., 2005. Intrauterine infection, preterm labor, and cytokines. J. Soc. Gynecol. Invest. 12, 463–465. Romero, R., Mazor, M., Wu, Y.K., Sirtori, M., Oyarzun, E., Mitchell, M.D., Hobbins, J.C., 1988. Infection in the pathogenesis of preterm labor. Semin. Perinatol. 12, 262–279. Saeed, R.W., Varma, S., Peng-Nemeroff, T., Sherry, B., Balakhaneh, D., Huston, J., Tracey, K.J., Al Abed, Y., Metz, C.N., 2005. Cholinergic stimulation blocks endothelial cell activation and leukocyte recruitment during inflammation. J. Exp. Med. 201, 1113–1123. Wennerholm, U.B., Holm, B., Mattsby-Baltzer, I., Nielsen, T., PlatzChristensen, J.J., Sundell, G., Hagberg, H., 1998. Interleukin1alpha, interleukin-6 and interleukin-8 in cervico/vaginal secretion for screening of preterm birth in twin gestation. Acta Obstet. Gynecol. Scand. 77, 508–514. Wolf, F.I., Torsello, A., Fasanella, S., Cittadini, A., 2003. Cell physiology of magnesium. Mol. Aspects Med. 24, 11–26. Zheng, H., Crowley, J.J., Chan, J.C., Raffin, T.A., 1991. Attenuation of LPS-induced neutrophil thromboxane b2 release and chemiluminescence. J. Cell. Physiol. 146, 264–269.
Magnesium sulfate suppresses inflammatory responses by human umbilical vein endothelial cells (HuVECs) through the NFB pathway Burton Rochelson a , Oonagh Dowling b , Nadav Schwartz a , Christine N. Metz b,∗ a
Division of Maternal-Fetal Medicine, North Shore University Hospital, 300 Community Drive, Manhasset, NY 11030, USA b The Susan & Herman Merinoff Center for Patient Oriented Research, The Feinstein Institute for Medical Research North Shore-LIJ Health System, 350 Community Drive, Manhasset, NY 11030, USA Received 10 March 2006; received in revised form 26 June 2006; accepted 29 June 2006
Abstract Dysfunctional endothelial cell activation and cytokines are implicated in preterm labor, a condition commonly treated with the tocolytic agent, magnesium sulfate (MgSO4 ). Based on recent findings showing the inflammatory effects of magnesium deficiency, we examined the effect of MgSO4 on human umbilical vein endothelial cell (HuVEC) inflammatory responses in vitro. HuVECs isolated from term umbilical cords were incubated with MgSO4 prior to stimulation with lipopolysaccharide (LPS) and then assessed for endothelial cell activation. Endothelial cell supernatants were assayed for inflammatory mediator production (interleukin8; IL-8), and endothelial cell-associated intercellular adhesion molecule (ICAM-1) expression was determined. In the absence of LPS stimulation, MgSO4 had no effect on HuVEC responses. Treatment of HuVECs with MgSO4 prior to LPS stimulation inhibited inflammatory mediator production (p < 0.05) and cell adhesion molecule expression (p < 0.05) in a dose-dependent manner. Mechanistic studies showed that MgSO4 reduced NFB nuclear translocation and protected cytoplasmic IB␣ from degradation in LPS-treated HuVECs. In conclusion, MgSO4 inhibits endothelial cell activation, as measured by levels of IL-8 and ICAM-1 expression, via NFB. Our results support the hypothesis that MgSO4 treatment may function as an anti-inflammatory agent during preterm labor. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Preterm labor; Magnesium sulfate; Endothelial cell activation; Cytokines
1. Introduction Preterm labor is one of the major contributing factors for neonatal morbidity and mortality in the developed world. Magnesium sulfate (MgSO4 ) is commonly used for suppressing preterm labor (Caritis, 2005; Lewis, 2005). However, the exact mechanism(s) of its action
∗ Corresponding author. Tel.: +1 516 562 3403; fax: +1 516 562 1022. E-mail address: [email protected] (C.N. Metz).
remains unclear. Preterm labor is characterized by an enhanced maternal systemic inflammatory response (Park et al., 2005) with the enhanced production of proinflammatory mediators, including IL-6, IL-8 and TNF (Hagberg et al., 2005; Romero et al., 2005; Wennerholm et al., 1998) and, in many cases, with intrauterine infections (Hagberg et al., 2005; Park et al., 2005; Romero et al., 1988). Infusion of inflammatory cytokines induces preterm labor and delivery in experimental models (Bry and Hallman, 1993). Based on the observations that magnesium deficiency is associated with increased inflammatory responses
0165-0378/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.jri.2006.06.004
102
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
(Bernardin et al., 2005; Maier et al., 2004a; Nakagawa et al., 2001) and that an enhanced maternal inflammatory response with endothelial cell dysfunction is implicated in preterm labor (Park et al., 2005), we examined the effect of MgSO4 on HuVEC inflammatory responses in vitro. We observed that MgSO4 inhibited endothelial cell activation in vitro and identified that MgSO4 reduced the nuclear translocation of nuclear factor kappa B (NFB). These observations have potentially broad implications because NFB, a transcription factor typically associated with inflammation and infection, has been recently implicated as an important regulator of human labor (Lindstrom and Bennett, 2005). 2. Materials and methods This study received exempt status by the Institutional Review Board (IRB) of North Shore University Hospital because it involved the use of anonymous, normal, pathogen-free, term umbilical cords for the isolation of human umbilical vein endothelial cells (HuVECs). 2.1. HuVEC isolation and culture conditions HuVECs were isolated using collagenase to digest the subendothelial basement membrane, as previously described (Jaffe et al., 1973). HuVEC preparations were >98% pure based on acetylated LDL uptake and CD31 expression by flow cytometry. HuVECs were grown on gelatin-coated flasks (0.15%) in M199 media containing 10% fetal calf serum, endothelial cell growth supplement (90 g/ml; Sigma, St. Louis, MO), heparin (100 g/ml; Sigma), penicillin, streptomycin and l-glutamine (2 mM). HuVECs were sub-cultured using Trypsin/EDTA Reagent Pack (Cambrex, Walkersville, MD). 2.2. HuVEC stimulation assays For MgSO4 stimulation assays, HuVECs (passages 3–6) were grown to confluency on gelatin-coated 96well plates; all assays were performed in M199 containing 10% fetal calf serum, penicillin, streptomycin and l-glutamine (media). HuVECs were treated with MgSO4 (1–20 mM prepared in media; American Pharmaceutical Partners, Schaumburg, IL), terbutaline (10−5 to 10−8 M; Bedford Laboratories, Bedford, OH) or dexamethasone (10–20 M prepared in media, Elkins-Sinn, Cherry Hill, NJ) for 30–60 min prior to either no stimulation or stimulation with lipopolysaccharide (LPS) (0.1 g/ml; LPS isolated from E. coli serotype 0111:B4
(lot number 63H4011, Sigma)). Control (vehicle) wells were treated with media containing phosphate-buffered saline (PBS) as a vehicle control for MgSO4 . In separate experiments, HuVECs were treated with MgSO4 at the time of LPS treatment or 0.5 h post-LPS treatment. Following overnight incubation, cell-free culture supernatants were collected and analyzed for IL8 production by ELISA (R&D Systems, Minneapolis, MN). Cells were assessed for intercellular adhesion molecule-1 (ICAM-1) expression by cell-based ELISA methods, as previously described (Saeed et al., 2005). To control for cytotoxicity/viability, parallel cultures were stained using the 3-[4,5-dimethylthiazol-2-yl]-2,5diphenyltetrazolium bromide (MTT) reduction assay, as previously described (Saeed et al., 2005). Data are presented as: mean IL-8 levels (pg/ml ± S.D.); mean cellassociated ICAM-1 expression (OD ± S.D.); and mean viability (OD ± S.D.). 2.3. NFκB and IκBα analyses Confluent HuVEC monolayers were treated with MgSO4 (2–10 mM, prepared in media as described above) for 0.5–1 h prior to the addition of LPS (0.1–1 g/ml). Control (vehicle) wells were treated with PBS diluted in media for 0.5–1 h prior to the addition of LPS (0.1–1.0 g/ml). Nuclear and cytoplasmic extracts were prepared 1 h post-LPS addition using the NE-PER kit (Pierce Chemical Co., Rockford, IL). Nuclear and cytoplasmic preparations (∼10 g/lane) were electrophoresed, transferred to PVDF membranes and probed with antibody to NFB (p65 (Rel A), Cell Signaling Technology, Beverly, MA) and IB␣ (Santa Cruz Biotechnology, Santa Cruz, CA), respectively, or with antibodies to control nuclear (Lamin A/C; Santa Cruz Biotechnology) and cytoplasmic (actin; Chemicon International, Temecula, CA) proteins. Following incubation with HRP-conjugated secondary antibody, specific proteins were revealed using ECL reagent (Amersham Pharmacia, Piscataway, NJ). Band densities were determined using the NIH Image Program and the ratios of the specific:control bands are shown. 2.4. Statistics Each set of experiments was repeated three times (n = 4 samples per condition). For each of the three outcome variables (IL-8, ICAM-1 and viability/cytotoxicity), a two-way ANOVA was performed to compare the means for the various interventions, controlling for the day each experiment was performed using
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
103
SAS statistical software. The standard ANOVA assumptions of normality and equality of variance were checked and were satisfied for each variable; an ANOVA was considered significant if p < 0.05. The Dunnett’s test (using the adjusted means) was then used to make pairwise comparisons (vehicle versus treated) using SAS® statistical software (p < 0.05 was considered significant). 3. Results 3.1. Magnesium sulfate (MgSO4 ) treatment inhibits endothelial cell activation Under basal conditions, endothelial cells serve a barrier function and exhibit a non-inflammatory phenotype; they are not inert and do express constitutive measurable levels of both ICAM-1 and IL-8. In our first series of experiments, we examined the effect of MgSO4 on constitutive IL-8 production and ICAM-1 expression by HuVECs in vitro. We found that MgSO4 (1–20 mM) had no significant effect on constitutive IL-8 production and ICAM-1 expression by HuVECs (Fig. 1A and B). Dexamethasone (Dex), a potent anti-inflammatory agent, had no effect on IL-8 production and only a slight inhibitory effect on ICAM-1 expression was observed (Fig. 1A and B). These concentrations of MgSO4 and Dex were not cytotoxic to the cells (Fig. 1C). Activation of the endothelium during inflammation and infection is characterized by increased cell surface adhesion molecule expression, and enhanced chemokine and cytokine production required for leukocyte recruitment. We examined the effect of MgSO4 on endothelial cell activation using LPS as a stimulating agent. Treatment of HuVEC cells with MgSO4 (2.5–10 mM) prior to LPS stimulation inhibited IL-8 production (up to 55% inhibition) in a dose-dependent manner (Fig. 2A). Similarly, MgSO4 (2.5–10 mM) treatment of HuVEC cells suppressed ICAM-1 induction by LPS by approximately 40% (Fig. 2B). MgSO4 was more potent than Dex alone and the combination of Dex + MgSO4 was not additive (Fig. 2A and B). Additional experiments were performed to examine the effect of the time of MgSO4 administration on LPS-induced endothelial cell activation. A significant reduction in ICAM-1 expression (25%) was observed with simultaneous administration of MgSO4 (10 mM) and LPS when compared to vehicle-treated cells (Fig. 2C). When MgSO4 (10 mM) was added 30 min post-LPS, MgSO4 reduced ICAM-1 expression by approximately 15%; however, this reduction was not statistically significant (Fig. 2C). Similar results were observed with IL-8 production (data not shown).
Fig. 1. MgSO4 has no effect on basal IL-8 and ICAM-1 production by HuVECs. MgSO4 (1–20 mM) addition to HuVECs has no effect on (A) basal IL-8 production (pg/ml ± S.D.), (B) constitutive cell-associated ICAM-1 expression (OD ± S.D.), or (C) cytotoxicity, as determined by MTT assay (OD ± S.D.). The effect MgSO4 is compared to dexamethasone (Dex) treatment at two concentrations (10 and 20 M). Data from four experiments are shown.
3.2. Magnesium chloride (MgCl2 ) treatment inhibits endothelial cell activation To determine whether the effect of MgSO4 on endothelial cell activation was mediated through Mg cations, we assessed the effect of MgCl2 on IL-8 production and ICAM-1 expression by HuVEC cells following LPS stimulation. Similar to the inhibitory effect of MgSO4 on HuVEC activation, MgCl2 blocked HuVEC activation (as determined by IL-8
104
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
Fig. 3. MgCl2 inhibits HuVEC activation in a dose-dependent manner. MgCl2 (1–10 mM) pre-treatment of HuVECs significantly blocks (A) IL-8 production (pg/ml ± S.D.) and (B) ICAM-1 expression induced by LPS (100 ng/ml) (in the presence or absence of dexamethasone (Dex, 10 M)) (OD ± S.D.). * p <0.05 comparing treated vs. no treatment. Data from four experiments are shown.
Fig. 2. MgSO4 inhibits LPS (100 ng/ml) induced HuVEC activation in a dose-dependent manner. MgSO4 (1–10 mM) pre-treatment of HuVECs significantly blocks (A) IL-8 production (pg/ml ± S.D.), (B) ICAM-1 expression induced by LPS (in the presence or absence of dexamethasone (Dex, 10 M)) (OD ± S.D.) and (C) vehicle or MgSO4 (10 mM) was added to HuVECs 30 min prior to LPS, the same time as LPS or 30 min post-LPS addition and ICAM-1 expression was assessed (OD ± S.D.). * p < 0.05. Data from four experiments are shown.
production and ICAM-1 expression) induced by LPS in a dose-dependent manner (Fig. 3A and B, respectively). 3.3. Magnesium sulfate inhibits NFκB nuclear localization and protects IκBα from degradation Because the NFB pathway regulates many genes involved in endothelial cell activation, inflammation and
the onset of labor, we examined the effect of MgSO4 on NFB and IB family members. Treatment of HuVECs with MgSO4 inhibited the nuclear translocation of NFB induced by LPS stimulation in a dose-dependent manner (Fig. 4A). Consistent with this observation, further studies showed that MgSO4 protected IB␣ from LPSinduced degradation in the cytoplasm (Fig. 4B). 4. Discussion Magnesium sulfate is the most frequently used tocolytic agent for inhibiting preterm labor (Caritis, 2005). Despite its widespread use, there is remarkably little in the literature on the mechanism of action of magnesium sulfate in the treatment of preterm labor. An increasing number of studies suggest that preterm labor has its origins in an aberrant inflammatory response characterized by cytokine production and endothelial cell activation (Park et al., 2005). Recent studies show that magnesium deficiency is associated with an increased inflammatory response (Maier et al., 2004a; Nakagawa et al., 2001; Nasulewicz et al., 2004). We
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
Fig. 4. MgSO4 inhibits NFB nuclear translocation and IB␣ degradation in HuVECs. MgSO4 pre-treatment of HuVECs inhibits (A) NFB nuclear translocation and (B) IB␣ degradation in the cytoplasm induced by LPS (100–1000 ng/ml). Data are shown as (A) the ratio of NFB density to Lamin A/C, a control nuclear protein and (B) the ratio of IB␣ density to -actin, a control cytoplasmic protein.
investigated the possibility that magnesium sulfate therapy exerts anti-inflammatory effects on human umbilical vein endothelial cells. Based on previous in vitro studies using MgSO4 (Maier et al., 2004b) and plasma concentrations following MgSO4 therapy, we chose MgSO4 concentrations ≤10 mM that were non-cytotoxic. LPS was chosen because components of the bacterial cell wall (including LPS) mimic the infectious state to promote preterm delivery in experimental models (Elovitz and Mrinalini, 2004). The results of our study showing the suppressive effects of MgSO4 on endothelial cell inflammatory responses in vitro support our hypothesis that MgSO4 exerts anti-inflammatory responses. Numerous cytokines, e.g. IL-8, IL-6 and TNF and markers of endothelial cell activation, e.g. ICAM-1 and VCAM-1, are elevated during preterm labor and birth (Fischer et al., 2001; Hagberg et al., 2005; Marvin et al., 2000; Park et al., 2005; Romero et al., 2005; Wennerholm et al., 1998). IL-8 is elevated in cervical secretions and amniotic fluid in cases of preterm labor (Gonzalez et al., 2005; Holst et al., 2005; Wennerholm et al., 1998). In addition, IL-8 and IL-1 induce cervical ripening in guinea pigs (Chwalisz et al., 1994; Chwalisz, 1994). In
105
many cases, pathogenic agents enter the uterus and elicit inflammatory responses within the chorion and amnion (chorioamnionitis) and/or the umbilical cord (funisitis) to induce preterm labor (Romero et al., 2005). A recent study by D’Alquen et al. (2005) demonstrated endothelial cell activation within the umbilical cord during chorioamnionitis and funisitis. In the presence of chorioamnionitis associated with funisitis, concentrations of soluble ICAM-1 (marker of endothelial cell activation) are elevated in fetal serum, venous endothelium and amniotic epithelium (D’Alquen et al., 2005). Similarly, chorioamnionitis is associated with elevated levels of IL-8, IL-6 and TNF in the umbilical cord blood and leukocyte infiltration within the placenta (mediated via activated endothelial cells) (Dollner et al., 2002). Despite the fact that magnesium sulfate is the most commonly used tocolytic agent in the US, there are reports of adverse maternal and fetal effects of long-term/high dose MgSO4 therapy (Caritis, 2005; Mittendorf et al., 2003, 2006). The mechanism of action of MgSO4 is thought to be secondary to a competitive inhibition of calcium and its effect on smooth muscle contraction (Lewis, 2005). The potential mechanism of action of magnesium as an anti-inflammatory agent is supported by studies demonstrating a relationship between magnesium deficiency and endothelial cell dysfunction (Maier et al., 2004a). Magnesium deficiency has been associated with increased endothelial cell activation (adhesion molecule expression, increased cytokine production) (Maier et al., 2004a), impaired endothelial cell proliferation and migration (Bernardin et al., 2005), and enhanced cytokine release (basal and LPS-induced) by alveolar macrophages in vivo (Nakagawa et al., 2001). In vivo models of Mg deficiency are also associated with inflammation (Nasulewicz et al., 2004). Our findings showing MgSO4 -mediated IL-8 inhibition by stimulated HuVECs are consistent with recent studies showing the inhibitory effects MgSO4 on IL-8 production by human amniotic and decidual cells (Makhlouf and Simhan, 2006). However, the mechanism of action of MgSO4 was not investigated in this previous study. Magnesium therapy has proven to be beneficial for numerous inflammatory conditions. For example, aerosolized magnesium sulfate is used for the clinical treatment of acute severe asthma (Blitz et al., 2005), magnesium supplementation in vivo preserves the integrity of the blood brain barrier during experimental sepsis (Esen et al., 2005), and magnesium infusions protect against experimental stroke (Lee et al., 2005). Although magnesium is known to be a potent neurosedative, the antiinflammatory action of Mg may also help to explain its
106
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107
efficacy in seizure prophylaxis in pre-eclampsia, a condition which has also been associated with endothelial cell dysfunction and enhanced inflammatory cytokine production (Roberts and Lain, 2002; Roberts, 1998). The effect of timing of MgSO4 administration on LPS-induced endothelial cell activation was investigated. MgSO4 , when added 30 min prior to LPS, significantly reduced ICAM-1 expression (up to 40%) by HuVECs (Fig. 2B and C). When MgSO4 was given simultaneously with LPS, we observed a reduced, but still significant, effect on ICAM-1 expression (25%, Fig. 2C). However, when MgSO4 was added 30 min. post-LPS, a slight but insignificant reduction (15%) was observed (Fig. 2C). Similar results were observed with IL-8 (data not shown). Based on our findings that MgSO4 -inhibited IB␣ degradation and NFB translocation (which begins immediately post-LPS addition, peaking within 0.5–1 h post-LPS addition), the timing of MgSO4 administration is important and should precede the initiation of the NFB cascade by LPS for most effective results. Mg deficiency alone is associated with an inflammatory state in vivo (Nasulewicz et al., 2004), suggesting that inadequate Mg stores may exacerbate host inflammatory responses during infection. Furthermore, these data support previous proposals that a sufficient Mg balance may be critically important during gestation (Durlach, 2004), particularly prior to the exposure of invading microorganisms. Additional in vivo experiments are required to test these hypotheses. We also examined the effect of another tocolytic agent, terbutaline (10−5 to 10−8 M), on endothelial cell activation; we found no significant effects of terbutaline on IL-8 production or ICAM-1 expression by LPStreated HuVECs. Our observations are consistent with Makhlouf and Simhan who found no effect of terbutaline on IL-8 production by human amniotic and decidual cells (Makhlouf and Simhan, 2006). Interestingly, terbutaline has been shown to exert anti-inflammatory effects. Terbutaline suppresses inflammatory mediator production by stimulated polymorphonuclear leukocytes (Zheng et al., 1991) and reduces TNF production by LPS-stimulated mesangial cells (Nakamura et al., 2000). In our study, magnesium sulfate suppresses the proinflammatory actions of LPS on human umbilical vein endothelial cells in vitro (Fig. 2A–C). In fact, the immunosuppressive activity of MgSO4 equaled that of dexamethasone, although their use together did not appear to be synergistic. Magnesium ions (Mg2+ ) appear to be critical for the observed effect of MgSO4 as MgCl2 had similar anti-inflammatory effects on HuVEC activation (Fig. 3A and B). Interestingly, one report addresses
the use of MgCl2 in place of MgSO4 because the clinical and pharmacological effects of the two agents overlap, while MgCl2 exhibits reduced tissue toxicity (Durlach et al., 2005). Based on our observations with umbilical cord endothelial cells, the anti-inflammatory action of MgSO4 at the molecular level is through its protective effect on cytoplasmic IB␣ degradation (Fig. 4B) and the subsequent inhibition of LPS-mediated NFB nuclear translocation (Fig. 4A). This is the first report indicating that NFB activation is sensitive to manipulation by Mg. Based on the fact that Mg participates in over 300 biochemical reactions as a co-factor for hundreds of enzyme systems (Wolf et al., 2003), it is not surprising that it impacts one of the most influential inflammatory pathways. NFB is a nuclear transcription factor classically coupled with inflammation and infection that regulates the production of numerous inflammatory mediators. The key role of NFB activation in regulating human labor has been identified (Lindstrom and Bennett, 2005) and, therefore, the potential role of MgSO4 as an antiinflammatory agent and regulator of NFB activation in various maternal and fetal compartments during preterm labor warrants further investigation. Acknowledgments The authors thank Ms. Tina Nemeroff and Dr. Xiangying Xue for their assistance in assaying IL-8 production and ICAM-1 expression by HuVECs and Dr. Martin Lesser and Ms. Meredith Lukin (Biostatistics Unit) for their assistance with data analyses. References Bernardin, D., Nasulewic, A., Mazur, A., Maier, J.A., 2005. Magnesium and microvascular endothelial cells: a role in inflammation and angiogenesis. Front. Biosci. 10, 1177–1182. Blitz, M., Blitz, S., Hughes, R., Diner, B., Beasley, R., Knopp, J., Rowe, B.H., 2005. Aerosolized magnesium sulfate for acute asthma: a systematic review. Chest 128, 337–344. Bry, K., Hallman, M., 1993. Transforming growth factor-beta 2 prevents preterm delivery induced by interleukin-1 alpha and tumor necrosis factor-alpha in the rabbit. Am. J. Obstet. Gynecol. 168, 1318–1322. Caritis, S., 2005. Adverse effects of tocolytic therapy. Br. J. Obstet. Gynaecol. 112 (Suppl 1), 74–78. Chwalisz, K., 1994. The use of progesterone antagonists for cervical ripening and as an adjunct to labour and delivery. Hum. Reprod. 9 (Suppl 1), 131–161. Chwalisz, K., Benson, M., Scholz, P., Daum, J., Beier, H.M., HegeleHartung, C., 1994. Cervical ripening with the cytokines interleukin 8, interleukin 1 beta and tumour necrosis factor alpha in guineapigs. Hum. Reprod. 9, 2173–2181. D’Alquen, D., Kramer, B.W., Seidenspinner, S., Marx, A., Berg, D., Groneck, P., Speer, C.P., 2005. Activation of umbilical cord
B. Rochelson et al. / Journal of Reproductive Immunology 73 (2007) 101–107 endothelial cells and fetal inflammatory response in preterm infants with chorioamnionitis and funisitis. Pediatr. Res. 57, 263–269. Dollner, H., Vatten, L., Halgunset, J., Rahimipoor, S., Austgulen, R., 2002. Histologic chorioamnionitis and umbilical serum levels of pro-inflammatory cytokines and cytokine inhibitors. Br. J. Obstet. Gynaecol. 109, 534–539. Durlach, J., 2004. New data on the importance of gestational Mg deficiency. J. Am. Coll. Nutr. 23, 694S–700S. Durlach, J., Guiet-Bara, A., Pages, N., Bac, P., Bara, M., 2005. Magnesium chloride or magnesium sulfate: a genuine question. Magnes. Res. 18, 187–192. Elovitz, M.A., Mrinalini, C., 2004. Animal models of preterm birth. Trends Endocrinol. Metab. 15, 479–487. Esen, F., Erdem, T., Aktan, D., Orhan, M., Kaya, M., Eraksoy, H., Cakar, N., Telci, L., 2005. Effect of magnesium sulfate administration on blood–brain barrier in a rat model of intraperitoneal sepsis: a randomized controlled experimental study. Crit. Care 9, R18–R23. Fischer, D.C., Winkler, M., Ruck, P., Poth, D., Kemp, B., Rath, W., 2001. Localization and quantification of adhesion molecule expression in the lower uterine segment during premature labor. J. Perinat. Med. 29, 497–505. Gonzalez, B.E., Ferrer, I., Valls, C., Borras, M., Lailla, J.M., 2005. The value of interleukin-8, interleukin-6 and interleukin-1beta in vaginal wash as predictors of preterm delivery. Gynecol. Obstet. Invest. 59, 175–178. Hagberg, H., Mallard, C., Jacobsson, B., 2005. Role of cytokines in preterm labour and brain injury. Br. J. Obstet. Gynaecol. 112 (Suppl. 1), 16–18. Holst, R.M., Mattsby-Baltzer, I., Wennerholm, U.B., Hagberg, H., Jacobsson, B., 2005. Interleukin-6 and interleukin-8 in cervical fluid in a population of Swedish women in preterm labor: relationship to microbial invasion of the amniotic fluid, intra-amniotic inflammation, and preterm delivery. Acta Obstet. Gynecol. Scand. 84, 551–557. Jaffe, E.A., Nachman, R.L., Becker, C.G., Minick, C.R., 1973. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J. Clin. Invest. 52, 2745–2756. Lee, E.J., Lee, M.Y., Chang, G.L., Chen, L.H., Hu, Y.L., Chen, T.Y., Wu, T.S., 2005. Delayed treatment with magnesium: reduction of brain infarction and improvement of electrophysiological recovery following transient focal cerebral ischemia in rats. J. Neurosurg. 102, 1085–1093. Lewis, D.F., 2005. Magnesium sulfate: the first-line tocolytic. Obstet. Gynecol. Clin. North Am. 32, 485–500. Lindstrom, T.M., Bennett, P.R., 2005. The role of nuclear factor kappa B in human labour. Reproduction 130, 569–581. Maier, J.A., Malpuech-Brugere, C., Zimowska, W., Rayssiguier, Y., Mazur, A., 2004a. Low magnesium promotes endothelial cell dysfunction: implications for atherosclerosis, inflammation and thrombosis. Biochim. Biophys. Acta 1689, 13–21.
107
Maier, J.A., Bernardini, D., Rayssiguier, Y., Mazur, A., 2004b. High concentrations of magnesium modulate vascular endothelial cell behaviour in vitro. Biochim. Biophys. Acta 1689, 6–12. Makhlouf, M.A., Simhan, H.N., 2006. Effect of tocolytics on interleukin-8 production by human amniotic and decidual cells. J. Reprod. Immunol. 69, 1–7. Marvin, K.W., Keelan, J.A., Coleman, M.A., McCowan, L.M., Zhou, R.L., Mitchell, M.D., 2000. Intercellular adhesion molecule-1 (ICAM-1) in cervicovaginal fluid of women presenting with preterm labor: predictive value for preterm delivery. Am. J. Reprod. Immunol. 43, 264–271. Mittendorf, R., Dammann, O., Lee, K.S., 2006. Brain lesions in newborns exposed to high-dose magnesium sulfate during preterm labor. J. Perinatol. 26, 57–63. Mittendorf, R., Pryde, P.G., Lee, K.S., 2003. Association between use of antenatal magnesium sulfate in preterm labor and adverse health outcomes in infants. Am. J. Obstet. Gynecol. 189, 613. Nakagawa, M., Oono, H., Nishio, A., 2001. Enhanced production of IL1beta and IL-6 following endotoxin challenge in rats with dietary magnesium deficiency. J. Vet. Med. Sci. 63, 467–469. Nakamura, A., Imaizumi, A., Kohsaka, T., Yanagawa, Y., Johns, E.J., 2000. Beta2-adrenoceptor agonist suppresses tumour necrosis factor production in rat mesangial cells. Cytokine 12, 491–494. Nasulewicz, A., Zimowska, W., Bayle, D., Dzimira, S., Madej, J., Rayssiguier, Y., Opolski, A., Mazur, A., 2004. Changes in gene expression in the lungs of Mg-deficient mice are related to an inflammatory process. Magnes. Res. 17, 259–263. Park, J.S., Park, C.W., Lockwood, C.J., Norwitz, E.R., 2005. Role of cytokines in preterm labor and birth. Minerva Ginecol. 57, 349–366. Roberts, J.M., 1998. Endothelial dysfunction in preeclampsia. Semin. Reprod. Endocrinol. 16, 5–15. Roberts, J.M., Lain, K.Y., 2002. Recent insights into the pathogenesis of pre-eclampsia. Placenta 23, 359–372. Romero, R., Erez, O., Espinoza, J., 2005. Intrauterine infection, preterm labor, and cytokines. J. Soc. Gynecol. Invest. 12, 463–465. Romero, R., Mazor, M., Wu, Y.K., Sirtori, M., Oyarzun, E., Mitchell, M.D., Hobbins, J.C., 1988. Infection in the pathogenesis of preterm labor. Semin. Perinatol. 12, 262–279. Saeed, R.W., Varma, S., Peng-Nemeroff, T., Sherry, B., Balakhaneh, D., Huston, J., Tracey, K.J., Al Abed, Y., Metz, C.N., 2005. Cholinergic stimulation blocks endothelial cell activation and leukocyte recruitment during inflammation. J. Exp. Med. 201, 1113–1123. Wennerholm, U.B., Holm, B., Mattsby-Baltzer, I., Nielsen, T., PlatzChristensen, J.J., Sundell, G., Hagberg, H., 1998. Interleukin1alpha, interleukin-6 and interleukin-8 in cervico/vaginal secretion for screening of preterm birth in twin gestation. Acta Obstet. Gynecol. Scand. 77, 508–514. Wolf, F.I., Torsello, A., Fasanella, S., Cittadini, A., 2003. Cell physiology of magnesium. Mol. Aspects Med. 24, 11–26. Zheng, H., Crowley, J.J., Chan, J.C., Raffin, T.A., 1991. Attenuation of LPS-induced neutrophil thromboxane b2 release and chemiluminescence. J. Cell. Physiol. 146, 264–269.