Osteopontin knockdown suppresses tumorigenicity of human metastatic breast carcinoma, MDA-MB-435

Abstract

Elevated expression of osteopontin (OPN), a secreted phosphoglycoprotein, is frequently associated with many transformed cell lines. Various studies suggest that OPN may contribute to tumor progression as well as metastasis in multiple tumor types. High levels of OPN have been reported in patients with metastatic cancers, including breast. We found that the expression of OPN corroborates with the aggressive phenotype of the breast cancer cells i.e. the expression of OPN is acquired as the breast cancer cells become more aggressive. To assess the role(s) of OPN in breast carcinoma, expression of endogenous OPN was knocked down in metastatic MDA-MB-435 human breast carcinoma cells using RNA interference. We targeted multiple regions of the OPN transcript for RNA interference, along with 'scrambled' and 'non-targeting siRNA pool' controls to distinguish between target-specific and potential off-target effects including interferon-response gene (PeIF2-alpha) induction. The OPN knockdown by shRNA suppressed tumor take in immunocompromised mice. The 'silenced' cells also showed significantly lower invasion and migration in modified Boyden chamber assays and reduced ability to grow in soft agar. Thus, in addition to the widely reported roles of OPN in late stages of tumor progression, these results provide functional evidence that OPN contributes to breast tumor growth as well.

Fig. 1

Osteopontin expression increases with an increasing aggressive phenotype of breast cancer cells. Shown is a western blot of cell lysates of the indicated cell lines probed with the anti-human OPN MP (III) B10 monoclonal antibody. β-actin loading serves as an indicator of equal loading

Fig. 2

Silencing of OPN from MDA-MB-435 cells by RNA interference (a) Transfection of the NM636 heteroduplex results in near-total suppression of OPN expression. Shown is a western blot using anti-OPN antibody with cell-free conditioned medium from transiently transfected MDA-MB-435 cells. V: vector-only, pSUPER transfected cells, sc: pSUPER-scrambled transfected cells, shRNA-104, 134, 173, 636 and 762: pSUPER-NM104, NM134, NM173, NM636 and NM762 transfected cells. (b) Densitometric analysis of western blot in 2a. pSUPER-NM636 transfected cells are almost completely suppressed for OPN expression. Results are depicted relative to vector-transfectants. (c) The SMARTpool reagents (Dharmacon) bring about suppression of OPN expression. N: non-targeting siRNA pool (Dharmacon), P: SMARTpool reagent, 1, 2, 3 and 4: The individual siRNAs that comprise the SMARTpool reagent and 636: pSUPER-NM636 transfected cells. Details provided in Table 1. (d) Stable clones generated using pSUPER-NM636 are knocked down for OPN expression. Shown is a western blot using anti-OPN antibody with cell-free conditioned medium from stably transfected MDA-MB-435 cells. Vector (1–3): pSUPER transfectant clones 1 through 3, scrambled: pSUPER-scrambled sequence transfected cells, clones 3 and 5, OPNi: various clones of pSUPER-NM636 transfected cells. (e) The level of the IFN-induced gene, P-eIF2α is comparable across all transfected clones as analyzed by western blot in the stable transfectants generated using pSUPER-NM636 construct

Fig. 3

Silencing OPN attenuates the transformed phenotype of MDA-MB-435. (a) OPNi clones 3 and 6 show significantly (P < 0.005) slower in vitro growth than all the vector-only and scrambled-transfected cells. OPNi clones 12 and 16 grew at a rate comparable to that of the controls. Cells were seeded at a density of 1 × 10⁴ cells/well in a 24-well plate. The growth (viability) and morphology of cells was monitored daily for 10 days. 435: MDA-MB-435 cells, v-1 and 2: vector-only 9pSUPER) clones, sc-3 and 5: scrambled duplex transfected clones, OPNi 3, 6, 12 and 16: clones stably silenced for OPN by NM636 heteroduplex. (b) The OPNi clones 3 and 6 are inhibited in their ability (P < 0.05) to migrate through gelatin in a modified Boyden chamber assay as compared to the parent, vector-only and scrambled sequence clones. The gelatin-coated trans-well chamber was filled with 5 × 104 tumor cells and invasion was monitored. Each clone was tested in triplicate to migrate in response to 10% FBS. (c) Knocking down OPN expression significantly (P < 0.05) retards the ability of MDA-MB-435 cells to invade through Matrigel in response to 10 μg/ml fibronectin relative to the parent, vector-only and scrambled sequence clones. The trans-well membranes were stained with Diff-Quick solutions. (d) The OPNi transfectants are inhibited in their ability to form colonies in soft agar. Cells (1000) were seeded in 6-well tissue culture plates in Bactoagar. Visible colonies (>50 cells) were counted with the aid of a dissecting microscope. OPNi-3 and -6 show significant inhibition (P < 0.05) compared to controls as well as OPNi-12. (e) RNAi-mediated suppression of OPN hinders homotypic cell–cell adhesion. OPNi-3 is notably (P < 0.05) affected compared to controls as well as OPNi-6 and -12. Cells (3 × 10⁴) labeled using Fluoresbrite^® carboxylate nanospheres were dropped on to a confluent monolayer of the same type of cells and incubated at 37°C in 5% CO₂ humidified atmosphere for 20 min. Each test group was assayed in triplicate. v-1 and 2: vector-only clones, sc-3 and 5: scrambled duplex transfected clones, OPNi 3, 6 and 12: clones stably silenced for OPN by NM636

Fig. 4

Transcriptional silencing of OPN abrogates tumorigenicity and metastasis. (a) Knocking down OPN reduces tumorigenesis concomitant with reduced pulmonary metastases. Shown is the percent incidence of primary tumors and lung metastases in the controls and the OPNi clones. 435: MDA-MB-435 cells, v1, v-2, v3: vector-only clones, sc-3 and -5: scrambled sequence clones, OPNi 3, -6, -12 and -16: clones stably silenced for OPN by NM636 heteroduplex. While only 25 and 30% of mice injected with OPNi clones 3 and 6 developed tumors as well as pulmonary metastases, the mice injected with MDA-MB-435, other controls and OPNi clones 12 and 16 showed 80–90% incidence of lung metastases. The results represent two independent experiments (20 mice per group were analyzed). (b) OPNi clones 3 and 6 (expressing least OPN) show significantly smaller and slower growing tumors. One million cells were injected in the mammary fat pad of athymic nude mice. Tumor development was monitored over 2 months. OPNi clones 3 and 6 show a statistically significant (P < 0.05) retardation in the growth rate of the tumors from day 30 post-injection. The results represent two independent experiments

Fig. 5

The expression of OPN in the xenografts corroborates with the OPN expression in the cells injected at the orthotopic site. (a) Immunohistochemical staining of xenografts for OPN. Formalin-fixed, paraffin-embedded xenografts were immunostained with monoclonal anti-OPN-53 after heat-induced antigen retrieval. Photomicrographs were taken in the area of most intense and diffuse staining for OPN. (A) Xenograft transfected with empty pSUPER vector (pSUPER-3, negative control). (B) Xenograft developed by injecting cells of OPNi-6 (b) Immunoscoring of xenografts for OPNi. The formalin-fixed paraffin embedded sections were scored based on the intensity of staining and the percentage of positive cells at each intensity. 435: MDA-MB-435, v-1 through v-3: vector-only clones, sc-3: scrambled transfected clone 3, OPNi 3, 6, 12 and 16: clones silenced for OPN by NM636