Abstract
γ-Aminobutyric acid (GABA) is the inhibitory neurotransmitter in the brain, also playing a role in diseases like epilepsy. We now show that this inhibitory neurotransmitter can also reduce migratory activity in SW 480 colon carcinoma cells. GABA reduced the norepinephrine-induced migratory activity of these cells within a three-dimensional collagen matrix to spontaneous migration levels, as was analyzed by time-lapse videomicroscopy. This inhibitory effect of GABA was mediated by the serpentine receptor GABAB and was intracellularly transduced by a decrease of the cyclic AMP concentration. Cancer cell migration is thus regulated by neurobiological signals, opening new possibilities for pharmacological agonists in cancer therapy.
Introduction
Evidence is growing that the migration of tumor cells is not solely a consequence of genetic alterations but is regulated by a multitude of epigenetic factors. Chemokines, neurotransmitters, and other structurally nonrelated ligands of serpentine receptors are known as important initiators of migratory activity (1). We have previously reported on the initiation of colon carcinoma cell migration by the neurotransmitter norepinephrine (2). In a further development of our studies on the influence of neurotransmitters on the migration of cancer cell migration, we investigated the effect of the inhibitory neurotransmitter GABA3 on the norepinephrine-induced migration of SW 480 colon carcinoma cells. GABA is the major inhibitory neurotransmitter of the central nervous system, where it has been shown to play a role in pathological conditions like epilepsy (3). GABA can, however, also be found in neural and nonneural tissues outside the brain (4, 5), including the gastrointestinal tract (6). Principally, GABA acts on two classes of cellular receptors: (a) the ionotropic GABAA and GABAC receptors are oligomeric chloride channels (6, 7); and (b) the metabotropic GABAB receptor is a member of the serpentine or seven-helices receptor family and is therefore related to chemokine receptors and catecholaminergic receptors, which both have been shown to be involved in the regulation of leukocyte and tumor cell migration (1, 8). Herein, we describe the signal transduction of the antidromic effects of norepinephrine and GABA on the regulation of tumor cell migration.
Materials and Methods
Cell Culture.
The colon carcinoma cell line SW 480 (American Type Culture Collection, Manassas, VA) was maintained at standard conditions as described previously (2).
Cell Migration Assay.
The cell migration assay was performed as described in detail previously (9). In short, cell locomotion within three-dimensional collagen lattices was recorded by time-lapse videomicroscopy overnight. For the analysis of migratory activity, 30 cells of each sample were randomly selected, and two-dimensional projections of paths were calculated by computer-assisted cell tracking in 20-min step intervals.
Measurement of Cellular cAMP.
For the measurement of changes in the cellular cAMP, 6 × 104 cells were incubated for 20 min at 37°C with either medium alone or with 10 μm norepinephrine, 100 μm GABA, or the combination of both neurotransmitters. For positive control, the cells were incubated with 500 ng/ml cholera toxin or 500 ng/ml pertussis toxin (both Sigma, Deisenhofen, Germany) under the same conditions. After incubation, cells were lysed, and the cAMP level was measured using a cAMP enzyme-linked immunoassay system (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom) as described by the manufacturer.
Flow-Cytometrical Measurement of Cytosolic Calcium.
For the investigation of changes in cytosolic calcium by treatment with norepinephrine or GABA, the SW 480 colon carcinoma cells were loaded with fluo-3/a.m. as previously described with minor modifications (10), and the calcium-induced fluo-3/a.m. fluorescence was measured immediately after addition of norepinephrine or GABA alone or in combination.
Results and Discussion
The Inhibition of Norepinephrine-induced Migration of Colon Carcinoma Cells by GABA Requires the Engagement of GABAB Receptors.
Norepinephrine induces migration of SW 480 colon carcinoma cells (2). We now report that this induced tumor cell migration is completely inhibited by GABA (Fig. 1,A). Norepinephrine induces migration from 42 ± 13% spontaneously locomoting cells to 63 ± 3%, whereas GABA alone had no effect (46 ± 4% locomoting cells) but abolished norepinehrine-induced migration (37 ± 3%). We investigated the receptor for the regulatory function of GABA by specific agonists (Fig. 1, B and C). We used baclofen as a specific agonist for GABAB receptors and isoguvacine for GABAA receptors. Baclofen but not isoguvacine inhibited the norepinephrine-induced migration (Fig. 1, B and C). After treatment with baclofen, the norepinephrine-induced migration was reduced from 52 ± 5% down to 33 ± 4% locomoting cells, whereas treatment with isoguvacine had no effect (44 ± 3% locomoting cells with norepinephrine versus 45 ± 2% locomoting cells with norepinephrine and isoguvancine). Thus, the inhibitory function of GABA on the migration of these tumor cells is mediated by GABAB receptors.
The Inhibitory Effect of GABA Is Mediated by the Regulation of Cellular cAMP.
It is well known from heart muscle cells that binding of norepinphrine leads to activation of stimulating G proteins (11). These G proteins regulate the activity of the adenylyl cyclase, which generates the second messenger cAMP (12). In turn, cAMP binds to multiple effector molecules, which are involved in the regulation of the cytosolic calcium concentration. A key effector molecule is the protein kinase A, which phosphorylates phospholamban. Phosphorylation of phospholamban leads to its release from sarcoplasmatic/endoplasmatic reticulum calcium ATPase, which sequestrates cytosolic calcium into intracellular stores (13).
In addition to the G protein-mediated signaling, Luttrell et al. (14) described an activation of PTKs via β-arrestin after β2-adrenoceptor engagement. We have shown that a PTK-dependent activation of the PLCγ is a crucial regulator for the norepinephrine-induced migration of colon carcinoma cells (2). Two second messengers are generated by the PLCγ: (a) inositol-1,4,5-trisphosphate, which opens intracellular calcium channels; and (b) diacylglycerol, which activates the PKCα. In this context, it is important to stress that an activation of the PKC with the diacylglycerol analogue phorbol-12-myristate-13-acetate is a sufficient start signal for the induction of very high locomotor activity in colon carcinoma cells (15).
In our experiments on the regulatory signal transduction of the norepinephrine- and GABA-mediated effects on cell migration, we focused on the G protein-mediated regulation of cellular cAMP. Norepinephrine induced an increase of cellular cAMP by 45.2%, whereas GABA reduced the cellular cAMP concentration by 48.5% (Fig. 2 A). The addition of GABA to norepinephrine-treated cells significantly reduced the norepinephrine-induced increase of cAMP. Therefore, an increase of locomotor activity is coupled to an increase of cellular cAMP, whereas a reduction of cellular cAMP decreases migratory activity.
Cells treated with dibutyryl-cAMP in combination with GABA and norepinephrine developed a migratory activity of 55 ± 18% locomoting cells, which was similar to norepinephrine alone (55 ± 15% locomoting cells), whereas cells treated with norepinephrine and GABA revealed a migratory activity of 36 ± 12% locomoting cells (Fig. 2 B). Addition of dibutyryl-cAMP alone did not lead to an increase of migratory activity (27 ± 4% locomoting cells). This shows that a decrease of cAMP is a sufficient stop signal for the norepinephrine-induced migration, but an increase of cAMP is not a sufficient start signal. As pointed out above, we have shown previously that an activation of the PKCα alone is a sufficient signaling event for the onset of migration (15). Therefore, the molecular switch to turn migration on can be different from the molecular switch to turn migration off. In addition, no changes of the cells’ protein tyrosine phosphorylation pattern were observed in cells treated with norepinephrine and GABA compared with cells treated with norepinephrine alone, as analyzed by antiphosphotyrosine immunoblotting (data not shown). This supports the view that the inhibitory GABA effect is predominantly or solely mediated by the regulation of cellular cAMP and does not interfere into PTK-mediated PLCγ activity.
At least both pathways, the cAMP-regulated adenylyl cyclase activity and the PTK-induced PLCγ activity, lead to the activation or generation of molecules, which regulate the intracellular calcium concentration (10). Consequently, we analyzed the cytosolic calcium concentration of cells treated with norepinephrine and GABA.
Treatment of the SW 480 cells with GABA did not lead to changes of the cytosolic calcium concentration, whereas norepinephrine treatment led to a strong increase of the cytosolic calcium concentration (Fig. 3). The norepinephrine effect is caused by the previously described PLCγ activation catalyzing the production of inositol-1,4,5-trisphosphate (2), which leads to the fast release of calcium from the endoplasmatic reticulum (arrow A in Fig. 4). The second effect of norepinephrine, the activation of the ATP-dependent sequestration of calcium by increased levels of cAMP, is not sufficient to override the ATP-independent calcium release from the endoplasmatic reticulum resulting in the observed increase of cytosolic calcium (arrow B in Fig. 4). Addition of GABA to norepinephrine-treated cells did not change cytosolic calcium as compared with norepinephrine alone (Fig. 3). GABA reduces the cAMP concentration (Fig. 2,A) and causes thereby a reduced calcium sequestration. This interruption of the previously described calcium cycling (10) results in the reduction of locomotor activity (arrow C in Fig. 4).
In conclusion, the induced migration of SW 480 colon carcinoma cells is inhibited by the neurotransmitter GABA. Our results lend themselves to the investigation of the possible use of GABAB agonists for the chemoprevention of metastasis development. Accordingly, Kawabata et al. (16) provided evidence for a preventive role of GABA in the frequency of azoxymethane-induced colonic adenocarcinoma in rats. Using this novel knowledge of neurobiological influences and the concerned intracellular pathways of tumor cell migration, it might be possible to develop strategies to delay or prevent metastasis formation either by blocking the involved receptors or by interfering with the intracellular signal transduction.
Effects of GABA on the migration of SW 480 colon carcinoma cells. The migration of the cells was induced by 10 μm norepinephrine (Nor). GABA (A), the GABAB-specific agonist baclofen (B), or the GABAA-specific agonist isoguvacine (C) were added at a concentration of 100 μm. The graphs show mean values of three independent experiments (90 cells were analyzed/sample).
Effects of GABA on the migration of SW 480 colon carcinoma cells. The migration of the cells was induced by 10 μm norepinephrine (Nor). GABA (A), the GABAB-specific agonist baclofen (B), or the GABAA-specific agonist isoguvacine (C) were added at a concentration of 100 μm. The graphs show mean values of three independent experiments (90 cells were analyzed/sample).
Role of cAMP in the regulation of SW 480 colon carcinoma cell migration. In A, the concentration of intracellular cAMP after treatment with 10 μm norepinephrine or 100 μm GABA alone or in combination was analyzed by an enzyme-linked immunoassay. Statistical significance of the cAMP reduction by GABA in norepinephrine-treated cells was calculated using Student’s t test (P = 0.02; indicated by an asterisk). Cholera (CTX) and pertussis (PTX) toxins were used as positive controls. The diagram shows values ± SD of three independent experiments. In B, cells were treated with either 1 μm dibutyryl-cAMP (db-cAMP) alone or in combination with norepinephrine and GABA. The graph shows mean values of three independent experiments (90 cells were analyzed/sample).
Role of cAMP in the regulation of SW 480 colon carcinoma cell migration. In A, the concentration of intracellular cAMP after treatment with 10 μm norepinephrine or 100 μm GABA alone or in combination was analyzed by an enzyme-linked immunoassay. Statistical significance of the cAMP reduction by GABA in norepinephrine-treated cells was calculated using Student’s t test (P = 0.02; indicated by an asterisk). Cholera (CTX) and pertussis (PTX) toxins were used as positive controls. The diagram shows values ± SD of three independent experiments. In B, cells were treated with either 1 μm dibutyryl-cAMP (db-cAMP) alone or in combination with norepinephrine and GABA. The graph shows mean values of three independent experiments (90 cells were analyzed/sample).
Flow-cytometrical measurement of changes of the intracellular calcium concentration after treatment with norepinephrine and GABA. Cells were loaded with the calcium-dye fluo-3/a.m. and subjected to flow cytometry immediately after addition of the neurotransmitters. The diagram shows values ± SD of three independent experiments. The mean fluorescence intensity of the control was adjusted to a value close to 700. Treatment of the cells with 5 μg/ml ionomycin served as a positive control and led to an increase to a mean fluorescence intensity of 1971 ± 302 (data not shown). Statistically significant changes versus control are indicated by an asterisk (Student’s t test; P < 0.05).
Flow-cytometrical measurement of changes of the intracellular calcium concentration after treatment with norepinephrine and GABA. Cells were loaded with the calcium-dye fluo-3/a.m. and subjected to flow cytometry immediately after addition of the neurotransmitters. The diagram shows values ± SD of three independent experiments. The mean fluorescence intensity of the control was adjusted to a value close to 700. Treatment of the cells with 5 μg/ml ionomycin served as a positive control and led to an increase to a mean fluorescence intensity of 1971 ± 302 (data not shown). Statistically significant changes versus control are indicated by an asterisk (Student’s t test; P < 0.05).
Model of the signal transduction pathways regulating the norepinphrine-induced locomotion and the inhibition of migration by GABA. This model is an extension of the pathway published in Ref. 2 (). The norepinephrine-induced pathways, which are marked by the arrows A and B, lead to a release of calcium from the endoplasmatic reticulum and its sequestration, thereby promoting a calcium cycling within the cell. The GABA-induced pathway, which is marked by the arrow C, interrupts this cycling by inhibiting the calcium sequestration. AC, adenylyl cyclase; PLB, phospholamban.
Model of the signal transduction pathways regulating the norepinphrine-induced locomotion and the inhibition of migration by GABA. This model is an extension of the pathway published in Ref. 2 (). The norepinephrine-induced pathways, which are marked by the arrows A and B, lead to a release of calcium from the endoplasmatic reticulum and its sequestration, thereby promoting a calcium cycling within the cell. The GABA-induced pathway, which is marked by the arrow C, interrupts this cycling by inhibiting the calcium sequestration. AC, adenylyl cyclase; PLB, phospholamban.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by the Bruno und Helene Joester Foundation, Cologne, Germany, and the Fritz Bender Foundation, Munich, Germany.
The abbreviations used are: GABA, γ-aminobutyric acid; cAMP, cyclic AMP; PLC, phospholipase C; PKA/C, protein kinase A/C; PTK, protein tyrosine kinase.