Annexin A3 involvement in lipid storage of clear cell Renal Cell Carcinoma (ccRCC) cells

Cristina Bianchi1, Silvia Bombelli1, Chiara Meregalli1, Barbara Torsello1, Giorgio Bovo2, Paolo Viganò3, Guido Strada3, Roberto A Perego1
  • 1 Università Milano-Bicocca, Dipartimento di Scienze della Salute (Monza)
  • 2 Ospedale San Gerardo, U.O. di Patologia (Monza)
  • 3 Ospedale Bassini, Cinisello Balsamo - A.O. ICP- U.O. di Urologia (Milano)


Renal cell carcinomas (RCC) account for about 85% of renal cancers and are characterized by different subtypes with different incidences. Clear cell RCC (ccRCC) is the most frequent subtype and is characterized by cells with clear cytoplasm due to lipid and glycogen storage. 90% of sporadic ccRCCs is characterized by biallelic inactivation of von-Hippel Lindau (VHL) gene that prevents degradation of hypoxia-inducible factor 1a and 2a (HIF1a/HIF2a) proteins with constitutive activation of their function [1]. HIF1 and HIF2, through the regulation of different and specific hypoxia-inducible genes, have an important role in the development of various metabolic alterations [2] responsible also of the “clear” cytoplasm that characterizes the cells of this tumor. Gene expression profiling and pathway analysis of early stage ccRCCs identified a molecular signature, involving PPARa/g modulation and lipid storage proteins, consistent with an adipogenic differentiation in ccRCC that can explain its “clear cell” morphology due to lipid accumulation in cytoplasm [3]. Interestingly, inhibition of ccRCC cell growth has been obtained by in vitro targeting PPARa pathway [4], evidencing that the adipocyte-like signature of ccRCC cells may be used to develop new therapeutic approaches. Annexin A3 (AnxA3) protein, described as biomarker in lung adenocarcinoma, prostate and ovarian cancer [5] and involved in the enhancement of the transactivating activity of HIF-1 [6], has been recently described also involved in inhibition of adipocyte differentiation [7]. Interestingly, we previously observed a HIF-related down-regulation of AnxA3 and a different pattern of its 36 and 33 kDa isoforms in human RCC respect to normal cortex cells [8]. By cytological, molecular, and functional approaches here we aimed to investigate AnxA3 involvement in adipocyte-like phenotype of ccRCC cells for potential translational approaches.

Materials and Methods

Primary cell cultures, established from ccRCC and normal cortex tissue samples were characterized by FACS analysis. Caki1 and A498 ccRCC cell lines and HK2 primary tubular cell lines were also used. Adipogenic medium was obtained by addition of 10ug/ml insulin, 0.25 uM dexamethasone and 0.5 mM IBMX to standard culture medium. Lipid storage in cultures and corresponding tissues was evaluated by Oil Red “O” staining or by FACS and immunofluorescence analysis with the fluorescent marker Bodipy. AnxA3 and PLIN2 expression was evaluated by western blot and immunofluorescence analysis, and gene silencing was performed by siRNA. Cell viability was evaluated by MTT assay.


ccRCC cultures maintain at the first passage the lipid storages observed in corresponding tissues. In ccRCC primary cultures AnxA3 expression was significantly down-regulated and PLIN2, a lipid droplet protein marker, significantly up-regulated respect to normal cortex. Moreover, in ccRCC primary cultures and cell lines, a lower expression of the 36 kDa isoform of AnxA3 protein correlated with a more abundant lipid storage. HIF1a-positive ccRCC Caki1 cell line, characterized by a lower expression of 36kDa AnxA3, and HIF1a-negative A498 and HK2 cells, characterized by a higher expression of 36 kDa AnxA3, were treated for 8 days with a specific medium known to induce adipocyte differentiation [7]. Caki1 cells, but not A498 and HK2 cells, showed an increase of lipid storage. Of note, in Caki1 cells treated with the adipogenic medium we observed a further decrement of 36 kDa AnxA3 protein level. Preliminary data showed that AnxA3 silencing induces an increase of lipid storage also in A498 cells, with a decrease of cell viability.


Our data show that primary cell cultures maintain the metabolic phenotype of ccRCC tissues and thus are a reliable model to study the metabolic alterations and the molecular mechanisms responsible of the “clear” cytoplasm of ccRCC cells. Using ccRCC primary cell cultures and cell lines we evidenced an involvement of AnxA3, and in particular of its 36 kDa isoform, in the modulation of lipid storage that characterizes the “adipocyte-like” phenotype of these tumor cells. In particular, our data seem to evidence for AnxA3 a role of negative regulator of lipid accumulation in ccRCC. The described calcium-dependent phospholipid-binding capacity of AnxA3 might have a role in modulating lipid accumulation in ccRCC lipid droplets that are lined by a phosholipid monolayer of cells.


These data may help to shed light on the complex molecular mechanisms involved in metabolic reprogramming of ccRCC. The comprehension of these molecular mechanisms may also help to identify new putative therapeutic targets for this cancer that nowadays is still poorly treatable. Moreover, the correlation here described between 36 kDa AnxA3 level and lipid storage in ccRCC cells may open the possibility to use AnxA3 as diagnostic/prognostic target in ccRCC, since it has been described in literature [9; 10] that AnxA3 may be secreted from different types of cancer cells (like prostate and ovarian cancer cells) and assayed in patients’ biological fluids.

Supported by A.G.S. Onlus


1. Mol Cancer 2009; 3: 8-31.
2. Biochim Biophys Acta 2011; 1813: 1263-68.
3. PLOS one 2010; 5: e10696.
4. PLOS one 2013; 8, e71115
5. Clin Trans Oncol 2013; 15:106-110
6. Biochem Biophys Res Commun 2005, 337:1283-87
7. J Biochem 2012; 152 (4): 355-63.
8. Am J Pathol 2010; 176: 1660-70.
9. J Cell Mol Med 2012; 16: 337-48
10. J Urology 2009, 181, 343-53