Inhibition of autotaxin production or activity blocks lysophosphatidylcholine-induced migration of human breast cancer and melanoma cells

Abstract

Increased expression of autotaxin in tumors including glioblastoma, breast, renal, ovarian, lung, and thyroid cancers is associated with increased tumor aggressiveness. Autotaxin promotes metastasis as well as cell growth, survival, and migration of cancer cells. These actions could depend on the noncatalytic effects of autotaxin on cell adhesion, or the catalytic activity of autotaxin, which converts lysophosphatidylcholine into lysophosphatidate in the extracellular fluid surrounding the tumor. Both lysophosphatidylcholine (LPC) and lysophosphatidate have been reported to stimulate migration through their respective G-protein coupled receptors. The present study determines the roles of autotaxin, LPC, and lysophosphatidate in controlling the migration of two cancer cell lines: MDA-MB-231 breast cancer cells, which produce little autotaxin and MDA-MB-435 melanoma cells that secrete significant levels of autotaxin. LPC alone was unable to stimulate the migration of either cell type unless autotaxin was present. Knocking down autotaxin secretion, or inhibiting its catalytic activity, blocked cell migration by preventing lysophosphatidate production and the subsequent activation of LPA(1/3) receptors. We conclude that inhibiting autotaxin production or activity could provide a beneficial adjuvant to chemotherapy for preventing tumor growth and metastasis in patients with high autotaxin expression in their tumors.

Figures

FIGURE 1

Effects of lysophosphatidate and lysophosphatidylcholine on the migration of MDA-MB-435 and MDA-MB-231 cells. Panel A, shows the migration of MDA-MB-435 (○), and MDA-MB-231 (■) cells after a 3 h incubation with different concentrations of LPA. Migration in response to 5% charcoal-stripped FBS was used as a positive control. Panel B, shows migration when the cells were incubated for 6 h with different concentrations of LPC. Results are expressed relative to the number of cells migrating with only basal media (RPMI 1640 with 0.1% BSA and 0.2% FBS-C), which was given a value of 1. Typically, basal migration consisted of approximately 75 cells per field for a 3 h incubation and 120 cells per field for a 6 h incubation. Results are means ± SEM from at least 3 independent experiments. Statistically significant differences between basal migration and agonist-stimulated migration are indicated by * (p

FIGURE 2

MDA-MB-435 cells secrete significant amounts of active ATX into the extracellular medium compared to MDA-MB-231 cells. Panel A shows the relative mRNA levels of ATX compared to GAPDH in MDA-MB-435 and MDA-MB-231 cells. Results are means ± SEM from at least 3 independent experiments. Panel B shows the results from the fluorescence assay for ATX activity of concentrated conditioned media collected from MDA-MB-435 (○), or MDA-MB-231 (■) cells. Results are means ± SEM from at least 3 independent experiments. Panel C shows the Western Blot analysis for ATX of concentrated media collected from 3 separate dishes of MDA-MB-435 or MDA-MB-231 cells. Recombinant ATX from Dr T Clair was used as a standard (StdATX).

FIGURE 3

Stimulation of migration by lysophosphatidylcholine depends on autotaxin activity. Panel A shows the migration of MDA-MB-231 cells over 3 h in the presence of basal medium or concentrated medium prepared from either MDA-MB-231 cells (231 cMedium), or from MDA-MB-435 cells (435 cMedium). These experiments were performed in the presence or absence of 10 μM LPC, or 1 μM of the ATX inhibitor, VPC8a202. Migration of MDA-MB-231 cells with only basal media (RPMI 1640 with 0.1% BSA and 0.2% FBS) present was given a value of 1. Migration in the presence of 5% FBS and 0.5 μM LPA plus VPC8a202 were used as controls. Panel B, shows results from similar experiments using MDA-MB-435 cells over a 3 h period. Panel C shows the effects of the ATX inhibitors, VPC8a202 and S32826, on the migration of MDA-MB-435 cells over 3 h in the absence of concentrated medium. Results are means ± SEM from at least 3 independent experiments. Statistically significant differences are indicated by: * (p

FIGURE 4

Kinetics of inhibition of ATX activity by VPC8a202. The initial rate of recombinant ATX activity was measured colorimetrically at different concentrations of LPC in presence and absence of 0.5 μM VPC8a202. Activity was expressed as the release of choline during 18 h and it is represented by the absorbance of the product at 555 nm (A555). Each point is the average of three measurements (S.D. values for the reaction rates were less than 5% of every measurement and are not depicted). Consumed substrate was less than 10% in every case thus ensuring an initial rate of reaction. Results were fitted to a straight line by linear regression. The apparent Km for LPC was about 588 μM.

FIGURE 5

Knock-down of ATX with siRNA significantly decreases the migratory potential of concentrated medium from MDA-MB-435 cells. Concentrated medium was collected from 3 dishes of MDA-MB-435 cells treated with siRNA for ATX (siATX) and 3 dishes treated with control siRNA (siCTRL). Panel A shows a fluorescence assay representing ATX activity in siCTRL concentrated medium (○) and in siATX concentrated medium (●). Error bars indicate the SEM of 3 experimental values. Panel B, shows the Western Blot analysis of a representative sample of concentrated media that were probed for ATX. Recombinant ATX was used as a standard (StdATX). Panel C compares the migration of MDA-MB-435 cells over a 3 h incubation in the presence or absence of LPC and concentrated medium collected from MDA-MB-435 cells that were treated with siCTRL (white bars) or siATX (dark bars). Results are means ± SEM from 3 independent experiments. Statistically significant differences are indicated by * (p

FIGURE 6

Knockdown of ATX with siRNA in MDA-MB-435 cells abolishes stimulation of migration by lysophosphatidylcholine. A 6 h incubation was used to measure the migration of MDA-MB-435 cells that were treated with siATX or siCTRL. Migration in basal media (RPMI 1640 with 0.1% BSA and 0.2% FBS-C) was compared to migration in the presence of 10 μM LPC, or 0.5 μM LPA. Results are means ± ranges from 2 independent experiments.

FIGURE 7

The LPA receptor antagonist VPC32183 inhibits migration of MDA-MB-435 cells. A 3 h incubation was used to measure the migration of MDA-MB-435 cells in the presence of their own concentrated medium with either LPC or LPA, and in the presence or absence of the LPA1/3 receptor antagonist, VPC32183. Migration in the presence of agonists/antagonists was compared to the number of cells migrating in basal media (RPMI 1640 with 0.1% BSA and 0.2% FBS-C), which is given a value of 1. Results are means ± SEM from at least 3 independent experiments. Statistically significant differences are indicated by: * (p < 0.01).