Regulation of noradrenergic function by inflammatory cytokines and depolarization

Authors


Address correspondence and reprint requests to Dr Beth A. Habecker, Department of Physiology & Pharmacology, L334, OHSU, 3181 SW Sam Jackson Park Road, Portland, OR 97239, USA. E-mail: habecker@ohsu.edu

Abstract

Although the sympathetic neurons innervating the heart are exposed to the inflammatory cytokines cardiotrophin-1 (CT-1), interleukin-6 (IL-6) and tumor necrosis factor alpha (TNFα) after myocardial infarction, the effects of these cytokines on noradrenergic function are not well understood. We used cultured sympathetic neurons to investigate the effects of these cytokines on catecholamine content, the tyrosine hydroxylase co-factor, tetrahydrobiopterin (BH4), and norepinephrine (NE) uptake. CT-1, but not IL-6 or TNFα, suppressed NE uptake and catecholamines in these neurons, whereas CT-1 and, to a lesser extent, IL-6 decreased BH4 content. CT-1 exerted these effects by decreasing tyrosine hydroxylase, GTP cyclohydrolase (GCH) and NE transporter mRNAs, while IL-6 lowered only GCH mRNA. The neurons innervating the heart are also activated by the central nervous system after myocardial infarction. We examined the combined effect of depolarization and cytokines on noradrenergic function. In CT-1-treated cells, depolarization caused a small increase in BH4 and NE uptake, and a large increase in catecholamines. These changes were accompanied by increased TH, GCH and NE transporter mRNAs. CT-1 and depolarization regulate expression of noradrenergic properties in an opposing manner, and the combined treatment results in elevated cellular catecholamines and decreased NE uptake relative to control cells.

Abbreviations used
BH4

(6R)-5,6,7,8-tetrahydrobiopterin

CT-1

cardiotrophin-1

CNTF

ciliary neurotrophic factor

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GCH

GTP cyclohydrolase

IL-6

interleukin-6

LIF

leukemia inhibitory factor

NE

norepinephrine

NET

norepinephrine transporter

PGP

protein gene product 9.5

sR

soluble IL-6 receptor

TH

tyrosine hydroxylase

TNFα

tumor necrosis factor alpha

VIP

vasoactive intestinal peptide

Noradrenergic neurotransmission can be altered by nerve injury, which leads to changes in the synthesis and release of norepinephrine (NE). The synthesis of NE is regulated primarily through the expression and activity of tyrosine hydroxylase (TH), the rate-limiting enzyme for catecholamine synthesis. TH activity is critically dependent on the availability of its co-factor (6R)-5,6,7,8-tetrahydrobiopterin (BH4) (Kaufman 1978; Zigmond et al. 1989). The action of released NE is terminated by re-uptake of NE into nerve terminals by the sodium-dependent, high-affinity norepinephrine transporter (NET) (Amara and Kuhar 1993). Following an injury to noradrenergic neurons, the expression of TH and the synthesis of NE are suppressed, while at the same time several neuropeptides, including vasoactive intestinal peptide (VIP), are induced (Sun et al. 1994; Sun and Zigmond 1996; Zigmond et al. 1996). Inflammatory cytokines, especially leukemia inhibitory factor (LIF), which is released by non-neuronal ganglionic cells, are responsible for many of the neurochemical changes that occur in sympathetic neurons after injury (Rao et al. 1993; Shadiack et al. 1993; Sun and Zigmond 1996). Similar neurochemical changes can be elicited in primary cultures of sympathetic neurons by treatment with LIF or the related cytokine, ciliary neurotrophic factor (CNTF), including induction of VIP coupled with reduction of BH4 and NE levels, and NE uptake (Yamamori et al. 1989; Lewis et al. 1994; Stegenga et al. 1996; Matsuoka et al. 1997; Habecker et al. 2000).

Although axotomy studies have only been performed on the sympathetic neurons of the superior cervical ganglion, it is likely that injury induces similar changes in other sympathetic neurons. For example, regional changes occur in the sympathetic innervation of the heart following myocardial infarction which are reminiscent of the axotomy response of superior cervical ganglion neurons. These changes include depletion of neuronal NE (Kozlovskis et al. 1986; Mathes et al. 1971) and elevation of VIP in the sympathetic neurons of the stellate ganglia which project to the heart (Roudenok et al. 2001). A number of inflammatory cytokines are induced in the heart following myocardial infarction, including tumor necrosis factor alpha (TNFα), interleukin-6 (IL-6) and cardiotrophin-1 (CT-1) (Frangogiannis et al. 1998; Gwechenberger et al. 1999; Aoyama et al. 2000). IL-6 and CT-1 are related to LIF and CNTF, sharing common receptor subunits and signaling pathways (Heinrich et al. 1998). Although the effects of these cytokines on noradrenergic function are not known, it is plausible that they might contribute to the regulation of sympathetic tone in neurons innervating the heart.

There are, however, important differences between axotomy and myocardial infarction with regard to their effects on noradrenergic function. Decreased cardiac output following myocardial infarction results in activation of the sympathetic nervous system and increased firing of the post-ganglionic noradrenergic neurons innervating the heart. In other systems where this has been examined, increased sympathetic nerve activity induces the expression of TH and increases its enzymatic activity, resulting in elevated NE production and release (reviewed by Zigmond 1988; Zigmond et al. 1989; Kumer and Vrana 1996). Similar changes occur in cultured sympathetic neurons following depolarization with 30–40 mm KCl (Walicke et al. 1977; Hefti et al. 1982; Raynaud et al. 1987a). Although the relationship between NE uptake and sympathetic nerve activity has not been examined, depolarization of adrenal chromaffin cells inhibits NE uptake (Role and Perlman 1983). This suggests that increasing the firing rate of sympathetic neurons innervating the heart could enhance noradrenergic transmission in the heart by elevating NE synthesis, increasing NE release and decreasing NE re-uptake.

We used cultured sympathetic neurons to investigate the effects of IL-6, CT-1, TNFα and depolarization on catecholamine content, BH4 levels and NE uptake. Inasmuch as IL-6 and CT-1 are related to LIF and CNTF, we anticipated that they would suppress noradrenergic function. Moreover, we expected that TNFα would potentiate the effects of IL-6, since it induces the IL-6 receptor in sympathetic neurons (Marz et al. 1996). To our surprise, only CT-1 suppressed noradrenergic function in sympathetic neurons, although all three cytokines induced the expression of VIP mRNA. CT-1 decreased noradrenergic properties, at least in part, through lowering TH, GTP cyclohydrolase (GCH) and NET mRNA. On the other hand, chronic depolarization produced a large increase in catecholamine production and smaller increases in BH4 content, NE uptake and mRNA encoding TH, NET and GCH. In CT-1 treated cells, the magnitude of the depolarization-induced increase in NE uptake was substantially smaller than the increase in catecholamine content. These results suggest that there is an activity-dependent rise in catecholamines that is only marginally influenced by inflammatory cytokines.

Materials and methods

Materials

Cell culture reagents were obtained from Gibco BRL/Invitrogen (Carlsbad, CA, USA). Biochemicals and hormones were purchased from Sigma Chemical Company (St. Louis, MO, USA) except as noted. Dispase was obtained from Boehringer Mannheim (Indianapolis, IN, USA), collagenase type II from Worthington Biochemicals (Freehold, NJ, USA) and nerve growth factor (NGF) from Austral Biologicals (San Ramon, CA, USA). Rat TNFα, rat CNTF, human IL-6 and rat IL-6 were purchased from R & D Systems (Minneapolis, MN, USA) or Peprotech (Rocky Hill, NJ, USA), human CT-1 and mouse IL-6 from Alomone labs (Jerusalem, Israel), and human soluble IL-6 receptor (sIL-6R or sR) from Sigma or R & D systems. BioCoat tissue culture plates were obtained from BD Biosciences (San Jose, CA, USA) and Cells-to-cDNA II from Ambion (Austin, TX, USA). The LightCycler-FastStart SYBR Green I PCR amplification kit was from Roche (Indianapolis, IN, USA).

Primary cell culture

Cultures of sympathetic neurons were prepared from the superior cervical ganglia (SCG) of newborn rats as described (Hawrot and Patterson 1979; Rao and Landis 1990). Cells were grown in L15-CO2-complete media, supplemented with NGF (50 ng/mL), penicillin G (100 U/mL), streptomycin sulfate (100 µg/mL) and either 5% fetal bovine serum or 5% rat serum. Neurons were pre-plated for at least 2 h and grown in 96-well BioCoat plates. Prior to treatment with cytokines or KCl, cells were maintained for 2 days in the presence of the anti-mitotic agent, fluorodeoxyuridine/uridine (10 µm), in order to reduce the number of non-neuronal cells. There were approximately 1000–2000 neurons per well. Cytokines and KCl were diluted in culture medium and filter-sterilized before addition to the culture dishes. KCl (1 m) was diluted in medium to a final concentration of 30 mm, while cytokines were added at the concentrations listed in figure legends. Unless otherwise noted, LIF and CNTF were used at 10 ng/mL, and all other cytokines were used at 50 ng/mL. Human IL-6 was added together with the human soluble IL-6 receptor (sR). Unless otherwise noted, neurons were treated for 5–7 days with cytokines with or without KCl prior to assay, with media changed every 2–3 days.

HPLC analysis of (6R)-5,6,7,8-tetrahydrobiopterin (BH4)

Tetrahydrobiopterin was quantified by an HPLC procedure using electrochemical detection described by Hylands and colleagues (Howells et al. 1986; Howells and Hyland 1987; Habecker et al. 2002). Sympathetic neuron cultures were homogenized at 4°C by trituration in the HPLC mobile phase, which consisted of 50 mm sodium acetate, 5 mm citric acid, pH 5.22, and 50 µm EDTA, supplemented with 1 mg/mL of both dithioerythritol (DTE) and diethylenetriaminepentaacetic acid (DTPA) (Howells and Hyland 1987), and spun down at 4°C to remove cell debris. The BH4 in the samples was stable under these conditions for at least 5 h at 4°C and for 1 month at − 70°C. Tetrahydrobiopterin was chromatographed by reversed-phase HPLC with electrochemical detection (HPLC-ED) on a C18 column (15 × 0.46 cm, 5 µm particle size, Rainin; Woburn, MA, USA) (Howells et al. 1986; Habecker et al. 2002). An ESA Coulochem II multielectrode detector (ESA; Chelmsford, MA, USA) was used to quantify the BH4 with the electrodes set as follows: electrode 1, + 0.18 V; electrode 2, – 0.07 V. Solutes in the column eluents were first oxidized by electrode 1, and then BH4 was reduced by electrode 2. The limits of detection for the BH4 reduction signal in electrode 2 were approximately 60 fmol.

HPLC analysis of catecholamines

Catecholamines were measured by HPLC-ED (Felice et al. 1978; Woodward et al. 1987; Habecker et al. 2002). Sympathetic neurons were triturated through a pipette tip in 75 µL of ice-cold 0.2 m perchloric acid containing 0.2 mm EDTA, 1.0 µm ascorbate and 250 nm dihydroxyl-benzylamine (dihydroxybenzylamine (DHBA), an internal standard), and then centrifuged. An aliquot of the supernatant fluid was chromatographed by HPLC on C18 reversed-phase HPLC (15 × 0.46 cm, 5 µm, Rainin) using a mobile phase containing 50 mm sodium phosphate and 50 mm sodium acetate, pH 3.0, 360 mg/L sodium octane sulfonate, 100 µL/L triethylamine and 6% (v/v) acetonitrile. Catecholamines were quantified using an ESA Coulochem II detector with the electrode potential set at +0.18 V. The detection limit for dopamine was estimated to be less than 50 fmol. While norepinephrine is the catecholamine produced by sympathetic neurons in vivo, dopamine is the predominant catecholamine produced by these neurons in vitro, and was quantified as a measure of noradrenergic function in the sympathetic neurons (Woodward et al. 1987).

NE uptake and binding

NE uptake was assayed essentially as described (Pacholczyk et al. 1991; Habecker et al. 2000). Cultured sympathetic neurons were rinsed with Krebs–Ringer–HEPES buffer (KRH; 120 mm NaCl, 4.7 mm KCl, 2.2 mm CaCl, 1.2 mm KH2PO4, 1.2 mm MgSO4, 10 mm HEPES, 5 mm Tris Base pH 7.4) and incubated in KRH containing 10 nm3H-NE, 1 mm ascorbic acid, 50 µm pargyline at 37°C, with or without 10 µm desipramine, for 5 or 10 min with similar results. Excess 3H-NE was removed by washing with ice-cold buffer lacking 3H-NE, and cells were solubilized in a solution of 0.1% sodium dodecyl sulfate (SDS)/0.1 m NaOH to quantify radioactivity by liquid scintillation counting. Non-specific uptake was defined as that remaining in the presence of 10 µm desipramine and was negligible in the neuron cultures. The data shown represent total uptake.

For cell-surface binding experiments, cells were incubated for 5 min in KRH containing 10 nm3H-NE at 4°C to prevent transport. Similar results were obtained in experiments carried out at 37°C and 4°C.

Non-quantitative reverse transcription-polymerase chain reaction (RT-PCR)

RNA was isolated from sympathetic neurons using Trizol. RNA was reverse transcribed using random primers and Moloney murine leukemia virus (MMLV) reverse transcriptase, and a 216 base pair (bp) VIP mRNA product amplified by PCR through 30 cycles (45 s each at 94, 58 and 72°C) with primers specific for the rat VIP gene (Nishizawa et al. 1985): upper primer (bp 50–69) AGTGTGCTGTTCTCACAGTCG; lower primer (bp 244–264) GCTGGTGAAAACTCCATCAGC.

Real-time PCR

RNA and RT reaction products were generated from individual wells of sympathetic neurons using the Cells-to-cDNA II Kit according to the manufacturer's protocol. For the PCR amplification, 2 µL of RT reaction products were combined with either 2.0 mm MgCl2 for GAPDH (glyceraldehyde-3-phosphate dehydrogenase) primers, or 3 mm MgCl2 for TH, NET, PGP (protein gene product 9.5/ubiquitin carboxy-terminal hydrolase L1) or GCH primers, 0.5 µm each primer and 2 µL DNA Master in a final reaction volume of 20 µL. PCR was performed with the LightCycler-FastStart SYBR Green I PCR amplification kit in a Roche LightCycler. A negative control for the RT was included for each set of cells to confirm the lack of genomic DNA contamination.

The rat GAPDH primers generate a 238 bp fragment (Comer et al. 1997; Comer et al. 1999): (+) CCTGCACCACCAACTGCTTAGC, (–) GCCAGTGAGCTTCCCGTTCAGC. The TH primers generate a 220 bp fragment (Comer et al. 1997): (+) GCTGTCACGTCCCCAAGGTT, (–) CAGCCCGAGACAAGGAGGAG. The NET primers generate a 229 bp fragment (Comer et al. 1997): (+) TCTCCATCCTTGGTTACATGG, (–) AGGACCTGGAAGTCATCAGC. The PGP primers generate a 127 bp fragment (Gene Bank accession no. 017237): (+) TAATGTGGACGGCCACCTCT, (–) AGAATTCACTGAGCGCGAGC. The GCH primers generate a 124 bp fragment (Gene Bank accession no. 024356): (+) GATGAGGACCATGACGAGAT, (–) CCTAACAAGCAAGTCCTTGG. Annealing temperatures ranged between 55 and 60°C for the different primers. The Roche LightCycler parameters were set according to the manufacturer's recommendation: denature at 94°C for 10 min, followed by 50 cycles of 94°C for 0 s, 55–60°C for 5 s and 72°C for 20 s, with a temperature transition rate of 20°C/s. One fluorescence reading was taken after each cycle at the end of the 72°C elongation phase. Fluorescence was plotted as a function of cycle number to determine when reactions were in the linear phase of amplification. To confirm that only specific PCR products were generated, a melt analysis was carried out to determine the specific melt temperature (Tm) for each amplification product. Standard curves were generated for each set of primers using control neuron mRNA. A slope was generated from the standard curve PCR amplifications and unknown samples were compared with the known standard values. Aliquots of the RT reaction products from individual culture wells were assayed for TH, NET and GCH, and for GAPDH or PGP as a normalization control.

Statistics

Statistical analyses were carried out with Prism 3.0 (Graphpad Software). The significance of differences between cytokine treatments and control cells were assessed by anova with the Dunnett post-hoc test, while differences between multiple conditions (cytokines with or without KCl) were examined by anova using the Newman–Keuls post-hoc test.

Results

To determine whether the cytokines, IL-6, CT-1 and TNFα, either alone or in combination, altered noradrenergic function, we treated cultured sympathetic neurons with these cytokines for 5–7 days, and then examined NE uptake and BH4 and catecholamine levels. Sister cultures were treated with LIF and CNTF to serve as positive controls.

LIF and CNTF decreased NE uptake in cultured sympathetic neurons by about 75% (p < 0.05), but TNFα, IL-6, IL-6 + TNFα and IL-6 + sR had no effect (Fig. 1a). Uptake of NE was also inhibited significantly in the presence of CT-1 (data not shown, see Fig. 3a). In addition to the effects on uptake, CNTF and CT-1 lowered the binding of 3H-NE to the NE transporter by 70–75% in experiments carried out at 4°C (Fig. 1b), whereas IL-6 + TNFα and IL-6 + sR did not. Inasmuch as the activation of growth factor receptors has been shown to cause the acute redistribution of NET away from the cell surface (Apparsundaram et al. 1998), the loss of binding following chronic exposure to cytokines suggested that the decrease in NE uptake was the result of a loss of NET on the cell surface. Therefore, in order to determine whether the loss of NE uptake was due to an acute decrease in cell surface NET, we treated cells with CT-1 for times ranging from 15 min to 2 h and measured NE uptake. Acute treatment with CT-1 did not alter NE uptake in cultured sympathetic neurons (Fig. 1c), suggesting that CT-1 and by analogy, LIF and CNTF, do not act by stimulating rapid redistribution of NET.

Figure 1.

Cytokine regulation of NE uptake. Chronic cytokine treatment. Sympathetic neurons were treated for 5–7 days with LIF (10 ng/mL), CNTF (10 ng/mL), IL-6 (50 ng/mL), TNFα (50 ng/mL), IL-6 + TNFα (50 ng/mL each), IL-6 (50 ng/mL) + soluble IL-6 receptor (sR; 100ng/mL) or CT-1 (50 ng/mL). (a) NE uptake was assayed at 37°C as described in the methods. The data shown are from a single experiment, and are representative of results obtained in five experiments (*p < 0.001; mean ± SEM, n = 5–7). (b) Binding to the NE transporter was assayed at 4°C. The data shown are from a single experiment and are representative of results obtained in four experiments (*p < 0.001; mean ± SEM, n = 6). (c) Acute cardiotrophin-1 treatment. Neurons were treated with 100 ng/mL CT-1 for the times indicated and NE uptake assayed at 37°C. The data shown are from a single experiment and are representative of results obtained in three experiments (mean ± SEM, n = 6).

Catecholamine levels were decreased in cultured sympathetic neurons grown in the presence of CNTF and CT-1 by 85% and 68%, respectively (p < 0.05), whereas TNFα, IL-6, IL-6 + TNFα and IL-6 + sR had no effect on catecholamine levels (Fig. 2a). This result was unanticipated, since the effects of IL-6 and CNTF are often indistinguishable. These results were not due to inactive forms of the cytokines, since IL-6 and TNFα did induce the expression of VIP mRNA in sister cultures, as confirmed by RT-PCR (Fig. 2b). In addition, CT-1 and CNTF decreased BH4 levels by approximately 90%, whereas IL-6 + sR only caused a 30% reduction and IL-6 alone or in combination with TNFα had no effect on neuronal BH4 content (Fig. 2c).

Figure 2.

Cytokine regulation of catecholamines, VIP mRNA and BH4. Sympathetic neurons were treated for 5–7 days with CNTF (10ng/mL), IL-6 (50 ng/mL), TNFα (50 ng/mL), IL-6 + TNFα (50 ng/mL each), IL-6 (50 ng/mL) + soluble IL-6 receptor (sR; 100 ng/mL) or CT-1 (50 ng/mL). (a) Dopamine content was assayed by HPLC with electrochemical detection. The data shown are from a single experiment and are representative of data obtained in four independent experiments. (*p < 0.001; mean ± SEM, n = 6). (b) VIP mRNA was identified by RT-PCR. All treatments induced VIP mRNA. Actin mRNA was amplified from each sample to confirm RNA integrity and cDNA production and amplification (data not shown). (c) BH4 was assayed by HPLC with electrochemical detection. Data are from a single experiment (*p < 0.01; mean ± SEM; n = 3) and are representative of three independent experiments.

To examine the combined effects of cytokines and depolarization on neurotransmitter function, sympathetic neurons were maintained in culture for 6–7 days in the presence of 30 mm KCl, with or without cytokines. In this series of experiments, the cytokines LIF, CNTF and CT-1 all lowered NE uptake by 50–60%, compared with untreated sister cultures (Fig. 3a). Depolarization of cultured sympathetic neurons with 30 mm KCl generally increased NE uptake, although the increase was not significant in the CNTF- or IL-6 + sR-treated cells (Fig. 3a). Depolarization had a more dramatic effect on catecholamine levels in cultured sympathetic neurons, increasing the levels by about threefold in control and IL-6 + sR-treated cultures, and by about seven and eightfold, respectively, in CNTF- and CT-1-treated cultures (Fig. 3b). It is of interest to note that depolarization of the cytokine-treated neurons raised catecholamines to levels that were greater than those found in non-depolarized, control cells. As a percentage of the levels in non-depolarized controls, depolarization increased catecholamine levels by 340 ± 37% in the absence of added cytokines, by 172 ± 10% in the presence of CNTF, by 330 ± 19% in the presence of IL-6 + sR, and by 329 ± 18% in the presence of CT-1 (mean ± SEM, n = 19–20). Finally, depolarization of the cultured neurons also increased the level of the TH co-factor, BH4, within each treatment condition (Fig. 3c), although the increase was not as robust as that seen for the catecholamines, and the BH4 levels in cytokine-treated cells never exceeded those in non-depolarized control cells.

Figure 3.

Regulation of noradrenergic function by cytokines and depolarization. Sympathetic neurons were treated for 7 days with CNTF (10 ng/mL), LIF (10 ng/mL), IL-6 (50 ng/mL) + soluble IL-6 receptor (sR; 100 ng/mL) or CT-1 (50 ng/mL) in the presence or absence of 30 mm KCl, and assayed for NE uptake (a), dopamine (b) or BH4 (c). (a) The data shown are percentage control NE uptake averaged from four independent experiments (mean ± SEM; n = 10–20; *p < 0.05 compared with non-depolarized control; **p < 0.001 compared with the non-depolarized control of the same treatment group). (b) The dopamine levels (pmol DA/well) are representative of results from four experiments (mean ± SEM; n = 5; *p < 0.05 compared with non-depolarized control; **p < 0.001 compared with the non-depolarized control of the same treatment group). (c) The BH4 levels are expressed as percentage of control, averaged from four independent experiments (mean ± SEM of 18–19 samples; *p < 0.001 compared with non-depolarized control cells; **p < 0.05 and ***p < 0.01 compared with non-depolarized controls of the same treatment group).

The cytokines LIF and CNTF regulate noradrenergic properties by decreasing expression of the genes encoding TH (Yamamori et al. 1989), NET (Matsuoka et al. 1997; Habecker et al. 2000) and GTP cyclohydrolase (GCH), the rate-limiting enzyme for the production of BH4 (Stegenga et al. 1996). To determine whether CT-1 also suppressed expression of these genes, sympathetic neurons were treated for 4–6 days with CNTF, CT-1 or IL6 + sR. mRNAs were quantified using real-time PCR and were normalized to either PGP or GAPDH as an internal standard. CNTF and CT-1 decreased TH mRNA by about 75% and 60%, respectively (Fig. 4a), NET mRNA by about 85% for both cytokines (Fig. 4b), and GCH mRNA by about 55% and 60%, respectively (Fig. 4c), compared with untreated control cells. In contrast, addition of IL-6 with the soluble IL-6 receptor did not significantly change the expression of TH or NET, but did lower GCH mRNA by about 50% (Fig. 4c).

Figure 4.

Cytokine regulation of TH, NET and GCH mRNA. Sympathetic neurons were treated for 4–5 days with CNTF (10 ng/mL), IL-6 (50ng/mL) + soluble IL-6 receptor (sR; 100 ng/mL) or CT-1 (50 ng/mL), and mRNA encoding TH, NET or GCH was quantified using real-time PCR. (a) TH mRNA was normalized to PGP mRNA and expressed as percentage of control. Data are representative of results obtained in three experiments (mean ± SD; n = 2–3; *p < 0.05 compared with control). (b) NET mRNA was normalized to PGP mRNA and expressed as percentage of control. Data are representative of results obtained in three experiments (mean ± SEM; n = 3; *p < 0.001 compared with control). (c) GCH mRNA was normalized to PGP mRNA and expressed as percentage of control. Data are representative of results obtained in two experiments (mean ± SEM; n = 3; *p < 0.001 compared with control).

Depolarization has been reported to increase (Hefti et al. 1982) or to not change (Fann and Patterson 1994) TH gene expression in sympathetic neurons, and its effect on NET and GCH expression is unknown. To explore the effect of depolarization on these parameters of sympathetic function, cells grown in the presence or absence of cytokines were treated with or without 30 mm KCl, and gene expression was assessed with real-time PCR. Addition of 30 mm KCl increased TH mRNA in sympathetic neurons treated with CT-1, but did not significantly alter TH in other conditions (Fig. 5a). Depolarization increased NET mRNA significantly in control and CT-1-treated cells (Fig. 5b) but had little effect on IL-6-treated cells, while KCl increased GCH mRNA in all treatment conditions (Fig. 5c).

Figure 5.

Regulation of noradrenergic genes by cytokines and depolarization. Sympathetic neurons were treated for 4–5 days with IL-6 (50 ng/mL) + soluble IL-6 receptor (sR; 100 ng/mL) or CT-1 (50 ng/mL) in the presence or absence of 30 mm KCl, and mRNA encoding TH, NET and GCH was quantified using real-time PCR. (a) TH mRNA was normalized to either PGP or GAPDH mRNA in the same cells. Data are expressed as percentage of control, averaged from three independent experiments (mean ± SEM of 7 samples; *p < 0.05, compared with non-depolarized control; **p < 0.01 compared with non-depolarized controls of the same treatment group). (b) NET mRNA was normalized to either PGP or GAPDH mRNA in the same cells. Data are expressed as percentage of control, averaged from three independent experiments (mean ± SEM of 9 samples; *p < 0.01, **p < 0.001 compared with non-depolarized control; ***p < 0.05 compared with non-depolarized controls of the same treatment group). (c) GCH mRNA was normalized to either PGP or GAPDH mRNA in the same cells. Data are expressed as percentage of control, averaged from two independent experiments (mean ± SEM of 6 samples (*p < 0.05, **p < 0.01 compared with non-depolarized control; **p < 0.01, ***p < 0.001 compared with non-depolarized controls of the same treatment group).

Discussion

Following myocardial infarction, the sympathetic neurons innervating the heart are activated by descending central pathways, while inflammatory cytokines such as CT-1, IL-6 and TNFα are increased in heart tissue. The effects of depolarization and inflammatory cytokines on noradrenergic function in the sympathetic neurons innervating the heart are not well understood. Therefore, we investigated the effects that these changes might have on the synthesis and re-uptake of catecholamines by sympathetic neurons. We found that CT-1, an inflammatory cytokine that is elevated following myocardial infarction, decreased sympathetic function and norepinephrine re-uptake. However, depolarizing the neurons with 30 mm KCl had a much more robust and opposite effect on catecholamine levels, such that the outcome from the combined presence of depolarization and CT-1 results in the potential for enhanced sympathetic action.

The CT-1-induced decrease in cellular catecholamine content could reflect changes in the synthesis, storage, release and/or re-uptake of dopamine and NE in sympathetic neurons. The cytokines LIF and CNTF, which act through the same signaling complex as CT-1, regulate several of these processes. They inhibit catecholamine synthesis by suppressing the expression of TH (Yamamori et al. 1989), GTP cyclohydrolase (GCH) (Stegenga et al. 1996), dopamine beta hydroxylase (Cervini et al. 1994; Dziennis and Habecker 2003), and the genes that encode them. In contrast, they do not alter the other enzyme required for NE production, aromatic-l-amino acid decarboxylase (Habecker et al. 2002), or the vesicular monoamine transporter (Habecker et al. 2000). CT-1 also suppressed TH and GCH mRNA, suggesting that the loss of catecholamine was due in part to decreased TH and GCH levels lowering catecholamine synthesis. LIF and CNTF suppress the uptake of NE and other catecholamines into sympathetic neurons by decreasing the expression of NE transporter mRNA and the levels of NET protein (Matsuoka et al. 1997; Habecker et al. 2000). Similarly, CT-1 decreased NE uptake and NET mRNA, consistent with CT-1 inhibition of NET expression. The steady-state level of dopamine in these cells reflects decreases both in synthesis and in dopamine re-uptake through the NE transporter. Increasing the TH co-factor BH4 prevents the cytokine suppression of dopamine (Habecker et al. 2002), suggesting that inhibition of catecholamine synthesis is the primary reason for the decrease in cellular dopamine content.

In contrast to the general suppression of noradrenergic properties caused by CT-1, IL-6 decreased only BH4 content and the mRNA encoding its rate-limiting enzyme, GCH. This was true even with additional soluble IL-6 receptor. Inasmuch as TNFα has been shown to induce IL-6 receptor expression in sympathetic neurons (Marz et al. 1996), we anticipated that it might facilitate the actions of IL-6 and mimic the responses seen with IL-6 in the presence of soluble IL-6 receptor. Rather than suppress noradrenergic function, TNFα had no effect on catecholamine levels, BH4 content or NE uptake. It did, however, induce the expression of VIP mRNA, indicating that the cytokine was active and was not toxic to the catecholamine-producing neurons. Moreover, CT-1 and IL-6 also induced VIP mRNA, indicating that decreased noradrenergic parameters were not the result of neuronal toxicity of these cytokines. These results suggest that while TNFα and IL-6 regulate neuropeptide expression in sympathetic neurons, they do not play a role in the regulation of catecholamine synthesis or re-uptake.

The differences between IL-6 and CT-1 were unexpected since these cytokines are both related to LIF and CNTF, and they share the common signaling receptor, gp130. The major difference between these cytokines, with respect to signal transduction, is that IL-6 uses a gp130 homodimer while LIF, CNTF and CT-1 all activate a gp130/LIF receptor heterodimer (Heinrich et al. 1998). Activation of either type of receptor complex generally results in similar downstream effects. For example, IL-6 can promote cell survival and peptide mRNA induction in sympathetic neurons, similar to cell survival and peptide induction elicited by LIF or CNTF (Kotzbauer et al. 1994; Marz et al. 1998). Similarly, the addition of LIF, CNTF or IL-6 with soluble IL-6 receptor stimulates dendrite retraction in cultured sympathetic neurons (Guo et al. 1999). Under each of these conditions, IL-6 in the presence of soluble IL-6 receptor induces responses that are similar to those observed with LIF and CNTF. In our studies, levels of IL-6 and soluble IL-6 receptor that were sufficient to induce VIP mRNA expression, decrease BH4 content and lower GCH mRNA, did not suppress catecholamine production, reduce NE uptake, or decrease TH or NET mRNA. CT-1, like CNTF, caused a significant reduction in the levels of catecholamines and BH4, and reduced NE uptake and expression of all three noradrenergic genes. These results suggest that the regulation of noradrenergic properties in sympathetic neurons depends on the specific type of receptor complex that is activated.

In contrast to cytokine suppression of noradrenergic properties, depolarization of the sympathetic neurons with 30 mm KCl increased all of the parameters of noradrenergic function that we examined. This concentration of KCl was chosen because it is sufficient to cause depolarization of sympathetic neurons in culture, elevating TH expression and catecholamine production (Walicke et al. 1977; Hefti et al. 1982; Raynaud et al. 1987a; Sun et al. 1992). Depolarization caused the induction of TH, NET and GCH mRNA in CT-1-treated cells, suggesting that the increase in noradrenergic parameters results, at least in part, from increased gene expression. Although GCH mRNA was increased by depolarization in all conditions, the effects of KCl on NET and TH mRNA differed depending on whether or not cytokines were present. Previous studies examining the regulation of TH gene expression by KCl also generated mixed results (Hefti et al. 1982), suggesting that the overall effect of nerve activity on TH mRNA is dependent on the cellular context, and is less robust than the induction of TH enzyme levels and activity (Fann and Patterson 1994).

Although the relationship between NE uptake and sympathetic nerve activity had not been examined previously, depolarization of adrenal chromaffin cells inhibits NE uptake (Role and Perlman 1983). On the other hand, manipulations that increase NE concentrations at sympathetic neuroeffector junctions increase the number of norepinephrine transporter binding sites (Swann et al. 1985). The small increases in NE uptake and NET mRNA that we observed in a subset of the depolarized cells are likely to be the result of multiple, converging signals influencing the expression and/or activity of the norepinephrine transporter. These various signals include, but may not be limited to, inhibitory effects of cytokines, increased nerve activity, and elevated intra and extracellular catecholamines. It remains to be determined whether uptake is regulated by nerve activity, perhaps through the activation of voltage gated ion channels, or is controlled by the changes in catecholamine synthesis and release. The increase in NE uptake cannot be attributed to changes in media osmolality, since treatment of sympathetic neurons with 30 mm NaCl or mannitol decreased norepinephrine transporter expression and NE uptake (B. Habecker and V. Brooks, unpublished observation). It is not yet clear how the depolarization caused by the elevation of extracellular KCl relates to the increased sympathetic neuronal activity following myocardial infarction in vivo, and whether the increased activity alters the parameters of noradrenergic function in similar ways.

Chronic depolarization of sympathetic neurons in culture had a much larger effect on catecholamine levels (300–800% increase) than on NE uptake (10–20% increase). Thus, insofar as chronic depolarization mimics nerve activity in these neurons, nerve activity would counteract the suppression of catecholamines by cytokines. It is likely that the up-regulation of catecholamines is due to several factors, including the stimulation of TH expression and activity (Zigmond and Ben-Ari 1977; Zigmond 1980; Hefti et al. 1982; Raynaud et al. 1987a,b), increased levels of the vesicular monoamine transporter (Desnos et al. 1995), and the elevation of BH4 levels reported here. Increased sympathetic nerve activity enhances TH activity and NE release through the opening of voltage gated Ca2+ channels (Vidal et al. 1989). The regulation of BH4 production by sympathetic nerve activity has not been examined, but the increase in GCH mRNA indicates it is due, at least in part, to stimulation of GTP cyclohydrolase expression. The mechanism of this induction is likely to involve more than just Ca2+ influx, since depolarization increases BH4 in hypothalamic neurons but not in mesencephalic neurons (Zhu et al. 1994). The elevation in catecholamines was far greater than the individual increases in TH and GCH mRNA, or BH4 content. This suggests that depolarization has a synergistic effect on catecholamine accumulation through the combination of greater co-factor availability, increased TH expression and phosphorylation, and increased vesicular storage.

We have investigated the combined effects of inflammatory cytokines and nerve activity on noradrenergic function in order to understand the changes that take place in sympathetic neurons innervating the heart following myocardial infarction. CT-1 and depolarization regulate expression of noradrenergic genes in an opposing manner, and the combined treatment results in elevated cellular catecholamines and decreased NE uptake relative to control cells. After myocardial infarction, these cells would also be exposed to IL-1β (Ono et al. 1998; Deten et al. 2002), which can stimulate release of NE from sympathetic neurons (Niijima et al. 1991; Terao et al. 1995). Increased catecholamine content coupled with stimulation of release and decreased re-uptake may contribute to the elevation in plasma NE observed following myocardial infarction (Eisenhofer 2001).

Acknowledgements

The authors thank Michelle Barney for technical assistance and helpful discussions. This work was supported by NIH grant HL68231 and AHA grant 0151349Z (BAH).

Ancillary