A role for VICKZ proteins in the progression of colorectal carcinomas: regulating lamellipodia formation

Abstract

VICKZ proteins are a highly conserved family of RNA binding proteins, implicated in RNA regulatory processes such as intracellular RNA localization, RNA stability, and translational control. During embryogenesis, VICKZ proteins are required for neural crest migration and in adults, the proteins are overexpressed primarily in different cancers. We hypothesized that VICKZ proteins may play a role in cancer cell migration. In patients, VICKZ expression varies with tumour type, with over 60% of colon, lung, and ovarian tumours showing strong expression. In colorectal carcinomas (CRCs), expression is detected at early stages, and the frequency and intensity of staining increase with progression of the disease to lymph node metastases, of which 97% express the protein at high levels. Indeed, in stage II CRC, the level of VICKZ expression in the primary lesion correlates with the degree of lymph node metastasis. In culture, VICKZ proteins rapidly accumulate in processes at the leading edge of PMA-stimulated SW480 CRC cells, where they co-localize with beta-actin mRNA. Two distinct cocktails of shRNAs, each targeting all three VICKZ paralogues, cause a dramatic drop in lamellipodia and ruffle formation in stimulated cells. Thus, VICKZ proteins help to facilitate the dynamic cell surface morphology required for cell motility. We propose that these proteins play an important role in CRC metastasis by shuttling requisite RNAs to the lamellipodia of migrating cells.

Figure 1

VICKZ expression in different types of tumours. Using the pan-VICKZ antibody, hVICKZ expression in 262 TMA samples from seven types of cancer was quantified, with each sample given a score from 0 (no staining) to 3 (extremely strong staining; see the Materials and methods section and Supplementary File 1, available online at http://www.interscience.wiley.com/jpages/0022-3417/suppmat/path.2376.html). The shaded bars represent the percentage of samples expressing hVICKZ (score 1–3) of each cancer type analysed. The bars marked with a single asterisk (colon, lung, and ovarian carcinomas) have a significantly higher percentage, in comparison with the bars marked with two asterisks (prostate and breast carcinomas) (p < 0.001). The open bars represent the hVICKZ mean staining score (with standard error of mean bars) for each tumour type studied. (The total number of samples analysed for each tumour type was as follows: colon, 46; lung, 45; ovarian, 35; melanoma, 16; brain, 21; breast, 54; prostrate, 43)

Figure 2

VICKZ protein expression increases through CRC tumourogenesis and invasion. Using the pan-VICKZ antibody, immunohistochemistry was performed on a section of a resected colon harbouring a lesion containing different stages of CRC progression. The brown colour represents VICKZ protein expression, and the blue, haematoxylin counterstain shows nuclei. (A) The entire section from one patient at low magnification (30×; scale bar = 1 mm). (B–E and B′–E′) Higher magnifications (200× and 600×, respectively) of the corresponding areas labelled in A (scale bars = 100 µm and 50 µm, respectively). A gradient of VICKZ protein expression is observed, with basal levels of expression in the normal region of the polyp (B), low levels in a region of dysplasia (C), moderate levels as the neoplastic cells migrate through the stem of the polyp (D), and higher levels in the cancer cells at the invasive front, where the cells are observed entering the mesenteric fat layer of the colon (E). Note that VICKZ expression can be seen in tumour cell clusters as well as in single tumour cells invading the adjacent stroma. The highest levels of VICKZ expression are observed in lymph node metastases (same patient) (F, G, G′, G″)

Figure 3

VICKZ protein expression increases during CRC invasion and predicts lymph node metastases. (A) One hundred and thirty-nine archival samples from CRC patients, representing normal epithelium, adenomas, adenocarcinomas, and lymph node (LN) metastases (cohort #1, see Supplementary Figure 1, available online at http://www.interscience.wiley.com/jpages/0022-3417/suppmat/path.2376.html), were stained with the pan-VICKZ antibody, scored, and graphed with standard error bars. The lymph node metastases show much higher VICKZ expression than any of the other groups (p < 0.001). Adenomas and adenocarcinomas are also significantly stronger in VICKZ expression than normal colonic epithelium (p < 0.01 and p < 0.001, respectively). (B) Pan VICKZ antibody was used to stain another 25 primary CRCs (cohort #2, see Supplementary Figure 1, available online at http://www.interscience.wiley.com/jpages/0022-3417/suppmat/path.2376.html) in order to test whether the intensity of VICKZ expression correlates with lymph node metastasis. The graph shows the mean score and standard error of mean for pan-VICKZ expression in CRC cohort #2. A significant correlation between metastasis and hVICKZ level of expression is observed (p = 0.037; two-tailed t-test)

Figure 4

VICKZ proteins localize to the leading edge of induced SW480 cells and to protruding lamellipodia in live cells. (A) Starved SW480 cells were stimulated by PMA and fixed after 40 min. Indirect immunofluorescence was performed using the pan-VICKZ antibody as the primary antibody and an anti-rabbit, Cy-5-conjugated secondary antibody. Intracellular localization of VICKZ protein to the lamellipodial leading edge was observed in induced cells and is indicated by the white arrow. (B) SW480 cells were transfected with GFP-hVICKZ1 plasmid and seeded on laminin-coated coverslips. After 2 days, the cells were stimulated as described. Fluorescent time-lapse microscopy was used to study the cellular distribution of GFP-hVICKZ1 over time within living cells. Pictures were taken every 5 min. Upper row: phase contrast series of the cell shows the lamella as a dark structure at the edge of the cell. Black arrows point to the area where the lamella is most active. Lower row: corresponding fluorescent images show that GFP-hVICKZ1 clearly localizes to the protruding, well-developed lamella. The last three pictures show that as the lamella disappears, GFP-hVICKZ1 is delocalized from the cell edge, illustrating spatial as well as temporal localization. Because the fluorescent images were overexposed in order to detect the signal in the lamella, the nucleus in these pictures is masked by the overlying cytoplasmic signal

Figure 5

mRNAs localized to lamellipodia of induced cells. (A) SW480 cells were stimulated by PMA after overnight serum starvation and then fixed and hybridized with a mix of TRITC-labelled oligonucleotides recognizing polyA+ RNA as a control for non-localized RNA, and FITC-labelled oligonucleotides recognizing β-actin, Arp2, or Arp3 RNA. DAPI was used to stain nuclei. Black and white stacks of images were obtained at the different wavelengths, and deconvolution was performed on the Z-series stacks obtained for each cell. The three resulting images per cell were pseudo-coloured (poly A+ in red; test RNA in green; and DAPI in blue) and overlaid to facilitate comparison. (B) SW480 cells stimulated by PMA after overnight serum starvation were probed for co-localization of VICKZ proteins, using the pan-VICKZ antibody (in red), and β-actin mRNA (in green). The white arrows in the magnified region of the overlay (outlined by the white rectangle) indicate cell membrane protrusions with co-localized protein and mRNA (in yellow)

Figure 6

shRNA knockdown of VICKZ expression in SW480 cells. (A) SW480 cells transduced with either scrambled shRNA (Scrambled) or shRNAs against all three VICKZ paralogues (VICKZ shRNA) were grown for 6 days in the absence (−) or presence (+) of doxycycline. Protein extracts were prepared and subjected to western blot analysis, using the pan-VICKZ or tubulin antibody. Normalizing against the tubulin signal, a 66 ± 15% reduction in VICKZ expression was observed upon induction. (B) As detailed in Supplementary File 1 (available online at http://www.interscience.wiley.com/jpages/0022-3417/suppmat/path.2376.html), uninduced SW480 cells transduced with shRNAs against all three VICKZ paralogues and labelled with TRITC-conjugated MitoTracker (bottom row) were co-plated with unlabelled, doxycycline-induced cells, serum-starved, stimulated with PMA, fixed, and stained with FITC-phalloidin in order to visualize actin microfilaments (middle row). Cells with normal levels of VICKZ (MitoTracker-positive; A′, B′) show smooth, normal membranes when viewed with Nomarski optics (top row), and contain cortical microfilaments (middle row). Cells with low levels of VICKZ (MitoTracker-negative; C′, D′) have a rough cell surface with multiple spike-like protrusions (white arrows) that stain positive for actin and are absent from uninduced cells

Figure 7

VICKZ KD in SW480 cells results in loss of lamellipodia and ruffles. SW480 cells were transduced with lentiviral vectors driving the expression, upon doxycycline (dox) induction, of either a scrambled shRNA (Scrambled) or one of two distinct pools of shRNAs directed against all three VICKZ homologues (Total1 or Total2). (A) Cells, induced or non-induced for shRNA production, were serum-starved and then stimulated by PMA. Pictures were taken every 3 s over the course of 20 min. Representative kymographs (assembled from 400 pictures each) are shown for each of the three types of transduced cells, without or with doxycycline induction. The white arrowheads show an example of how the peaks of lamellipodia extension (at the cell membrane) were determined for each kymograph; the black arrowheads indicate how the ruffles were counted. (B) The numbers of lamellipodia (left) and ruffles (right) were compared from a large set of kymographs obtained as described in A. From every cell examined, ten kymographs were constructed and analysed, with an average of 50 kymographs analysed per shRNA pool. For lamellipodia formation, a p value less than 0.001 was obtained when comparing plus and minus doxycycline in the ‘Total’ samples. For ruffle formation, a p value less than 0.0001 was obtained for the same samples