Figure S1. Adnk4 Injection Influenced Plasma and Bone Marrow NK4 Levels

Supplemental Data

Figure S1. AdNK4 injection influenced plasma and bone marrow NK4 levels

We evaluated the pattern of NK4 in mice injected i.v. with AdNK4 (109 pfu/mouse). (a) The plasma was collected at various days during the week following the AdNK4 injection (day 0). The concentration of NK4 in the plasma was determined using ELISA kit for human HGF (B-Bridge International Inc., CA).1 Because this ELISA system does not cross-react with mouse HGF, human NK4 can be specifically detectable in mouse plasma. Recombinant human NK4 was used for determination of the standard curve. AveragesSEM of the values for three animals for each point. *p<0.05, **p<0.005 and ***p<0.001 vs. the value of untreated mice. (b) Mice administered AdNK4, were used to harvest bone marrow cells at various times (n=3 for each point). Protein extracts were prepared and analyzed by Western blot. Vinculin was used for normalization. The image is representative of experiments performed in triplicate. Anti-HGF antibody utilized for NK4 detection was prepared by Nakamura et al.2 After AdNK4 injection, NK4 levels persisted elevated in the plasma for 7 days, and in the bone marrow NK4 peaked at day 4.

Figure S2. Bioluminescence parameters at early times after xenografting, and in AdLacZ injected mice bearing metastases

(a) Representative images of metastasis-bearing xenograft mice (ME), treated or untreated with adenoviral or chemical inhibitor. For each animal, the bioluminescence imaging is reported at 1 and 24 h after 1833/TGL cell injection into the left cardiac ventricle. The intensity of the signal, measured as photon flux, is shown as a color scale. 1 h after 1833/TGL cell injection, a diffuse whole-body bioluminescence signal was detected in all the animals. At 24 h after injection, much of the diffuse signal disappeared; however foci of arrested tumor cells could be seen at skeletal level. The Table reports AveragesSEM of bioluminescence values at 24 h all over the skeleton for all the animals examined in each group: ME n=12; ME+AdNK4 n=12; ME+DAS n=8; ME+AdNK4+DAS n=8. The treatments did not modify bioluminescence measured in ME. (b) Representative images of ME, injected with the control vector AdLacZ (ME + AdLacZ). ME n = 12; ME + AdLacZ n = 6. The Table reports AveragesSEM of bioluminescence values all over the skeleton, indicating no difference between ME and ME + AdLacZ. (c) During the experimental period, the body weights of ME xenograft mice were unaffected by the treatments at the corresponding times.

Figure S3.Beclin-1 in bone metastatic tissue of 1833-xenograft mice, treated or not with AdNK4

Immunohistochemistry analysis of Beclin-1 was performed. Beclin-1, a main actor of autophagy, was found overexpressed not only in neoplastic cells (ahead arrows) of bone metastasis, but also in supportive fibroblasts (arrows). In the latter, Beclin-1 localized principally at nuclear level. The presence of Beclin-1 in nuclei is reported.3We hypothesize that in fibroblasts within metastasis ECM,hypoxic conditions induced an anti-oxidant defense in neighboring cancer cells, protecting them against apoptosis. Representative immunostaining images for Beclin-1 are shown. The experiments have been repeated on five serial sections for each specimen from 3 xenograft mice with similar results. Magnifications of delimitated fields are shown. bo, bone; bl, blood lacunae.

Figure S4. Effect of NK4 on Bim and Beclin-1 expression under HGF stimulation

The interplay between autophagy and cell survival/cell death pathways is quite complex. We studied the effect of HGF, anti-apoptotic stimulus,4 on Bim and beclin-1 expression. Recent studies have demonstrated a role of Bim in anoikis, and the different functions played by the isoforms.5We treated 1833-bone metastatic clone with HGF, in the presence or the absence of 10 µg/ml NK4, a competitive inhibitor of HGF binding to Met. 1833 clone is known to produce HGF.6 The treatment with recombinant human HGF mimics paracrine HGF. The blockade of autocrine and paracrine HGF, by preventing the activities of signal kinases downstream, enhanced BimEL expression and phosphorylation as well as the level of BimS, the principal player of anoikis.7Concomitantly, Beclin-1 protein level increased. Western blots were performed in triplicate, and representative images are shown. Vinculin was used for normalization. The numbers at the bottom indicate fold variations of the 3 Bim isoforms, each vs. the relative value after HGF treatment, taken as 1. For Beclin-1, the fold-variations were calculated versus the value observed after HGF treatment, taken as 1.

Altogether our data were consistent with anoikis triggering in hypoxic bone metastasis after AdNK4 administration to the xenograft mice, in which concomitantly hypoxia-dependent HIF-1 expression diminished at bone marrow level. We suppose that bone marrow protective functions towards metastatic cells diminished, affecting survival notwithstanding the presence of Beclin-1.

Figure S5. Viability of 1833 cells, grown under stress and nonadherent conditions, and relationship with autophagy

(a)Starved cellswere transfected or not with Akt dominant negative (ΔAkt), and exposed or not to HGF for 36 h. Then, the survival of the cells in poly-HEMA was evaluated by MTT assay at various times. The relative cell viability of HGF pre-treated cells was higher (2-fold), and prolonged with respect to HGF-untreated cells. Akt was implicated only in HGF-treated cells, consistent with the role played on Bim degradation. The experiments were repeated 3 times with similar results. AveragesSEM. **p<0.005 vs. the value of HGF-untreated cells; °°p<0.005 and °°°p<0.001 vs. the value of HGF-treated cells in the presence of ΔAkt. (b) Time-courses of phospho-Akt (pAkt) and Akt are shown. The pAkt/Akt ratios were calculated using the densitometric values after normalization with vinculin. pAkt/Akt ratios are shown as fold variations calculated vs. HGF-untreated cells, taken as 1. The experiments were repeated three times with similar results. Akt activity (pAkt/Akt ratio) peaked at 30 min, increasing again at 12-16h, with a significant impact on viability due to HGF. (c) Stressful and potentially toxic environmental stimuli, such as nutrient and growth factor depletion or pro-oxidant radicals, induce autophagy. The autophagy process is finely tuned by sensors of nutrient and energy available.8 The oncosuppressor Beclin-1 is an autophagy marker, and we evaluated its expression in 1833 cells under a stressful condition consisting in starvation. In 1833 cells undergoing 4 h starvation, or exposed to 72-h fetal bovine serum (FBS), Beclin-1 (red) was faintly detectable and diffusely distributed in the cytoplasm, regarded as negative. After 48 or 72 h starvation, Beclin-1 staining was confined to perinuclear region as large and brilliant puncta suggestive of macro-aggregate reactivity, regarded as positive.9 The experiments were performed in triplicate, and were repeated 3 times with similar results. Nuclei were stained with DAPI (bleu).

Autophagy is a lysosomal-dependent pathway for macromolecule and organelle degradation, that has a pivotal role in cell homeostasis. This specialized degradative route, which starts with the formation of autophagosomes and ends in the lysosomes, provides building blocks and energy for cellular survival.10Consistently, in starved 1833 cells, the BimL isoform is predominant, index of autophagyc vesicle formation (Figure 8d of the present paper). Our data confirm in bone metastasis the interplay autophagy/anoikis, depending on environmental conditions.

SupplementalExperimental Procedure

Intracardiac injections

Cells were harvested from subconfluent cell culture plates, washed and resuspended at 5x106/ml in PBS. 0.1 ml of the suspended cells was injected into the left cardiac ventricle of 4-week-old female nu/nu mice (Harlan, Italy), using 26G needles.11 Mice were anesthetized with Avertin (0.2 ml/10 g, body weight) before injection. A successful injection was characterized by the pumping of arterial blood into the syringe. By Optical Imaging, we verified at early times after xenografting the presence of bioluminescent circulating cells in the mice ME, and in ME-treated with AdNK4, DAS or AdNK4 plus DAS.

Supplemental References

1. Wen J, Matsumoto K, Taniura N, Tomioka D, Nakamura T. Inhibition of colon cancer growth and metastasis by NK4 gene repetitive delivery in mice. Biochem Biophys Res Commun 2007; 358: 117-123.
2. Sakai K, Nakamura T, Matsumoto K, Nakamura T. Angioinhibitory action of NK4 involves impaired extracellular assembly of fibronectin mediated by perlecan-NK4 association. J Biol Chem 2009; 284: 22491-22499.
3. Geng Q-R, Xu D-Z, He L-J, Lu J-B, Zhou Z-W, Zhan Y-Q et al. Beclin-1 expression is a significant predictor of survival in patients with lymphonode-positive gastric cancer. PLoS ONE2012; 7: e45968-e45974.
4. Desiderio MA. Hepatocyte growth factor in invasive growth of carcinomas. CMLS-Cell Mol Life Sci2007; 64: 1341-1354.
5. Ruppert SM, Li W, Zhang G, Carlson AL, Limaye A, Durum SK et al. The major isoforms of Bim contribute to distinct biological activities that govern the processes of autophagy and apoptosis in interleukin-7 dependent lymphocytes. Biochim Biophys Acta2012; 1823: 1877-1893.
6. Maroni P, Matteucci E, LuzzatiA, Perrucchini G, Bendinelli P, Desiderio MA. Nuclear co-localization and functional interaction of COX-2 and HIF-1 characterize bone metastasis of human breast carcinoma. Breast Cancer Res Treat2011; 129: 433-450.
7. Hara H, Takeda T, Yamamoto N, Furuya K, Hirose K, Kamiya T et al. Zinc-induced modulation of SRSF6 activity alters Bim splicing to promote generation of the most potent isoform BimS. FEBS J2013; doi:10.1111/febs.12318.
8. Liao XH, Majithia A, Huang X, Kimmel AR. Growth control via TOR kinase signaling, an intracellular sensor of amino acid and energy availability, with crosstalk potential to proline metabolism. Amino Acids2008; 35: 761-770.
9. Nicotra G, Mercalli F, Peracchio C, Castino R, Follo C, Valente G et al.Autophagy-active beclin-1 correlates with favourable clinical outcome in non-Hodgkin lymphomas. Modern Pathol2010; 23: 937-950.
10. Nazarko VY, Zhong Q. ULK1 targets Beclin-1 in autophagy. Nature Cell Biol2013; 15: 727-728.
11. Kang Y, Siegel PM, Shu W, Drobnjak M, Kakoken SM, Cordòn-Cardo C et al. A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 2003; 3: 537-549.