Figure 1.
Visceral and parietal mesothelia have significantly different gene expression profiles.
Microarray analysis was performed on visceral and parietal mesothelia isolated from adult mice. Hierarchical clustering of 1,728 probes separated the two cell types based on gene expression patterns. Vertical dendrograms represent the individual samples, of which there are two replicates for each tissue type (A). Values shown are log base 2 and bright red, bright blue, and gray indicate the highest, lowest, and median normalized signal values respectively. Overrepresented functional categories were generated for both tissue types (B). B-H p values represent the Benjamini and Hochberg corrected p value, calculated by the online database and functional analysis program, DAVID, using Fisher's Exact test.
Figure 2.
Autotaxin is expressed at significantly higher levels in visceral mesothelia compared to parietal.
Immunofluorescence was used to detect autotaxin (ATX) expression in visceral and parietal tissues from Wt1cre/+; R26ReYFP/+ reporter mice. Cells that express Wt1, a mesothelial marker, will express enhanced yellow fluorescent protein (eYFP) that can be detected with a green fluorescent protein (GFP) antibody (A, D, G). Robust autotaxin expression is evident in visceral mesothelia (A–F). In contrast, the parietal mesothelium that lines the body wall expresses the mesothelial marker (G), but is devoid of autotaxin expression (H). Nuclei are marked with DAPI. Quantitative real-time RT-PCR was used to determine autotaxin transcript levels in visceral (omental) mesothelium, parietal mesothelium, heart, and liver (J). The autotaxin expression in visceral cells was set to 1 and fold changes were calculated for each subsequent tissue type. The asterisk represents a statistically significant difference when compared to the expression level in visceral cells (p<0.05, n = 4). Error bars were calculated using standard error of the mean.
Figure 3.
Autotaxin activity is significantly up-regulated in visceral mesothelia and mesotheliomas.
Conditioned media was collected from cultured visceral mesothelia, parietal mesothelia, peritoneal mesothelioma (MSTO), and pleural mesothelioma (ROB) cells. Some cultures were treated with the autotaxin inhibitor S32826. Media was assayed for LPA accumulation as a measure of autotaxin activity over a 2 hour period. Visceral mesothelia had significantly higher levels of autotaxin activity compared to parietal cells (A). Furthermore, MSTO and ROB cells had even higher levels of autotaxin activity compared to normal mesothelial cells (B). Normal and pathological cells were responsive to S32826 and displayed decreased levels of autotaxin activity (A–B).
Figure 4.
Visceral mesothelial cells are more adherent then their parietal counterparts.
Visceral and parietal mesothelia was isolated, dissociated, and plated on fibronectin-coated glass for 2 or 12 hours. After 2 hours, 57% more visceral cells attached to the fibronectin than parietal cells (A). Of the adherent cells, virtually all of the visceral cells had filamentous F-actin expression (B), in contrast to only 50% of the parietal cells. By 12 hours, visceral cells displayed a spread confirmation (D) with prominent vinculin expression at sites of focal adhesion (F). In contrast, parietal cells maintained a round confirmation (E) and had no focal adhesions (G) at this time point. The asterisk represents a statistically significant difference between the percent of adherent visceral and parietal cells (p<0.05, n = 6). Error bars were calculated using standard error of the mean.
Figure 5.
Mesothelia have varying abilities to migrate.
Visceral mesothelial (A), parietal mesothelial (B), pleural mesothelioma (C), and peritoneal mesothelioma (D) cells were plated on fibronectin and subjected to time-lapse imaging in order to visualize their ability to migrate. Panels A–D are compilations of static images taken at 20 minute intervals. False coloration indicates a cell's location at each interval: red 0 min; green 20 min; yellow 40 min; blue 60 min; purple 80 min.
Figure 6.
Autotaxin signaling regulates mesothelial and mesothelioma cell invasion.
Visceral mesothelial, parietal mesothelial, pleural mesothelioma (MSTO), and peritoneal mesothelioma (ROB) cells were cultured on transwell filters with and without addition of autotaxin (ATX), the autotaxin inhibitor S32826, Lpar1 inhibitor 1 (2440), or Lpar1 inhibitor 2 (8437). Of all the cells studied, the mesothelioma cells were the most invasive (C, D). Visceral cells (A) invaded more readily than parietal (B), but addition of ATX to parietal cells (F) promoted invasion to visceral-like levels. Treatment with S32826 reduced cell invasion in all groups (I-L). Furthermore, both of the Lpar1 inhibitors were efficient at decreasing cell invasion in mesothelial (M, Q) and mesothelioma cells (O, P, S, T). The percent of cell invasion is quantified in panel U. The percent of untreated visceral cells that migrated was set to 100. Asterisks represent statistically significant differences compared to untreated values in each group (p<0.05, n = 4). Error bars were calculated using standard error of the mean.
Figure 7.
Autotaxin signaling regulates mesothelial and mesothelioma cell migration.
Visceral mesothelial, parietal mesothelial, pleural mesothelioma (MSTO), and peritoneal mesothelioma (ROB) cells were plated on fibronectin, cultured with or without the addition of autotaxin, the autotaxin inhibitor S32826, Lpar1 inhibitor 1 (2440), or Lpar1 inhibitor 2 (8437), and subjected to time-lapse imaging in order to visualize their ability to migrate. Panels A-H are compilations of static images taken at 20 minute intervals. False coloration indicates a cell's location at each interval: red 0 min; green 20 min; yellow 40 min; blue 60 min; purple 80 min.