Figure 1.
Truncated APC products from SW480, DLD1, HT29, GP2D, CaCo2, LoVo, SW948 and VACO4A cells.
Functional domains of APC are indicated and include the Armadillo repeat (Arm) domain, the 15 (15RA–15RD) and 20 (20R1–20R3) amino acid repeats that are β-catenin binding sites, the β-catenin inhibitory domain (CID) which is necessary to target β-catenin for degradation and the first axin/conductin binding site (SAMP). The mutation cluster region (MCR) extends from the end of the 15RA to the middle of the 20R3 and corresponds to the region where most mutations associated with colorectal cancer have been found. The indicated amino acid positions refer to the length of the various truncated APCs expressed in the different colorectal cancer cell lines. SW480, DLD1, GP2D, CaCo2 and VACO4A cells have a truncating mutation at one allele and underwent loss of heterozygocity at the second allele, whereas SW948, HT29 and LoVo cells carry each two different truncating mutations.
Figure 2.
Stable knockdown of APC reduces colorectal cancer cell proliferation in vitro and increases wnt signalling.
HT29, GP2D, CaCo2, LoVo and VACO4A cells were transduced with lentiviruses for the expression of either shVEC, shN-APC or shVACO4A and the transduced cells were sorted by FACS based on GFP co-expressed by the lentivirus. Data are representative of at least two independent experiments. (A) Efficiency of APC knockdown and β-catenin level as determined by western blotting in Triton-X100 (Tx) or hypotonic (H) cell lysates. α-tubulin and β-actin were used as controls for equal sample loading. Molecular weights of APC isoforms (kD) are indicated. (B) Effect of APC knockdown on cellular proliferation, analyzed using the colorimetric MTT assay over the indicated time course. Data are the mean of triplicates+/−standard deviation. (C) Semi-quantitative RT-PCR of APC, axin2, Lgr5 and GAPDH (21 and 26 cycles).
Figure 3.
Transient knockdown of APC reduces colorectal cancer cell proliferation in vitro and increases wnt signalling.
GP2D and CaCo2 cells were transiently transfected with either siGFP or siAPC and further processed two days after transfection. Data are representative of at least two independent experiments. (A) Efficiency of APC knockdown and β-catenin level. (B) Effect of APC knockdown on cellular proliferation, as described in legend to figure 2B. (C) β-catenin-dependent reporter assay. Cells were transiently transfected with either the TOPglow or control FOPglow reporters, together with a β-galactosidase expression vector to correct for variations in the transfection efficiency. Luciferase activity was measured 48 h post-transfection and normalized to β-galactosidase values. Data are represented as the mean value of triplicates+/−standard deviation. To visualize FOP activity, values have been multiplied by 4 and 100 for GP2D and CaCo2 cells, respectively.
Figure 4.
Knockdown of truncated APC reduces proliferation of SW948 cells in vitro and increases wnt signalling.
SW948 cells were transduced with lentiviruses for the expression of either shVEC or shN-APC and the transduced cells were sorted by FACS based on GFP co-expressed by the lentivirus. Data are representative of at least two independent experiments. (A) Efficiency of APC knockdown and β-catenin level. (B) Effect of APC knockdown on cellular proliferation, as described in legend to figure 2B. (C) β-catenin-dependent reporter assay, as described in legend to figure 3C. To visualize FOP activity, values have been multiplied by 10.
Figure 5.
Knockdown of truncated APC inhibits tumour formation by HT29 cells in nude mice.
Lentivirus transduced HT29 cells expressing either shVEC, shN-APC or shVACO4A were injected subcutaneously into both flanks of nude mice (n, number of injections). Tumours were measured once a week over a period of 6 weeks post injection. (A) Tumour volumes upon termination shown as a dot plot. Bar, mean tumour size. (B) Increase of the mean tumour volume over time, +/−standard deviation.
Table 1.
Consequences of APC down-regulation.