Clinical cancer therapy, i.e. chemotherapy and radiation, target the actively proliferating tumor cells, which most of the times inevitably induce therapy-resistant cancer cells, derived from proliferative precursor types. Transdifferentiating agents, could avoid chemo- and radio- resistance by shifting the balance from proliferation to differentiation in these tumor precursor types.
In a previous study, albumin-associated lipids were found to induce the transdifferentiation of HCCLs toward adipocyte-like cells (Ruiz-Vela et al.). Our findings indicated that the connection between albumin-associated lipids and pluripotency (Garcia-Gonzalo & Izpisua[Belmonte 2008]) is far more complex than previously anticipated. In addition to maintaining self-renewal and pluripotency in hESCs (Garcia-Gonzalo & Izpisua[Belmonte 2008]), a role for albumin ([Kallee 1996]) and its associated lipids (Davis &[Dubos 1947]; Thomas et al.; Lafond et al.) has been discovered in adipogenesis in other cell types (Schopfer et al.). Our results also indicate that certain albumin complex associated poly- and monounsaturated fatty acids induce terminal differentiation and arrest cancer progression (Ruiz-Vela et al.). Concomitanlty to our studies, Khan et al. have also described the specific growth inhibition of esters of oleic acid and ricinoleic acid against the human skin malignant melanoma cell line (SK-MEL-1) (Khan et al.). Despite the important findings that we describe here, many key questions remain unanswered in the identification of novel genes and proteins that mediate the terminal transdifferentiation of human cancer cells. To get a better understanding of the mechanisms that lead to transdifferentiation we concentrated our efforts on the role of those genes that are differentially regulated during the process.
EM revealed a marked loss of pigmentation in the melanoma MALME-3M cells treated with albumin-associated lipids, in accordance with the downregulation of MLANA gene expression. Melan-A is known to form a complex with Pmel17 which affects its expression, stability, trafficking, and the processing which is required for melanosome maturation. Its expression is indispensable for Pmel17 function and the formation of cell pigmentation (Hoashi et al.).
Quantification of gene expression by RNA-seq led to the characterization of the MALME-3M cells treated with albumin-associated lipids. Several adipocytic markers such as PLIN2, LPL and PPARA were significantly upregulated. PPARA could be upregulated merely as a consequence of fatty acid accumulation as it has been shown to be involved in the regulation of obesity in rodents by increasing hepatic fatty acid oxidation (Kersten et al.). LPL is a well describedadipocyte marker as it regulates the hydrolysis of triglycerides in the adipose tissue (Mead et al.). Interestingly, PPARG1, an adipogenesis marker shown to be upregulated upon HCCL to adipocyte transdifferentiation (Ruiz-Vela et al.), showed a differential mapping pattern between the albumin-associated lipid-treated and non-treated cells. The PPARG gene has eight exons which are translated and spliced into different isoforms (http://www.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000132170;r=3:12328867–12475855), with the PPARG1 isoform being the primary transcript expressed in adipocytes. In treated cells, PPARG showed whole transcript expression, while in mock-treated cells no expression of exons 4 and 5 was evident (Additional file 3: Supporting information B). These two exons encode the Zinc finger binding site domain of the PPARG1 transcription factor (Finn et al.). The lack or downregulation of these functional domains in the non-treated cells might be indicative of the differential processing of this gene in the MALME-3M cell lines and of differential expression of the protein (Ruiz-Vela et al.).
The characterization of the adipocyte-like cells showed that the expression of PLIN2 was increased in MALME-3M, and in MCF-7 cells treated with albumin-associated lipids and with petroselinic acid. PLIN2 is a principal adipocytic marker, which coats lipid droplets in adipocytes (Brasaemle et al.; Heid et al.) and its expression has been linked to PPARG1. We previously reported that PPARG1 expression was increased in MALME-3M cells treated with albumin-associated lipids (Ruiz-Vela et al.). It has been recently reported that pretreatment of murine 3T3-L1 preadipocytes with Rosiglitazone, a potent PPARG1 agonist, decreased lipolysis and increased PLIN2 expression (Kim et al.).
Quantification of gene expression by RNA-seq also gave us some clues about the possible mechanistic pathways involved in transdifferentiation. Among the genes that were identified as differentially expressed between the treated and the untreated conditions, many of them were related to endocytic functions. Albumin-associated lipid induced transdifferentiation was accompanied by the upregulation of CLTC and other important adaptor-related complexes such as AP1B1, AP1G1, AP1S3, AP2A1, Synergin and AP3M1. Adaptor-related proteins are key components of clathrin coated vesicles that can bind directly to both the clathrin lattice and to the lipid and protein components of membranes (Pearse et al.). AP1 and AP3 are found at the coated vesicles located at the Golgi complex and it has been suggested that both associate with GLUT4 transporting vesicles and mediate distinct intracellular sorting events at the level of the TGN and endosomes in rat adipocytes (Gillingham et al.). The observed upregulation of proteins involved in GLUT4 trafficking could be related to the acquisition of the adipocytic phenotype.
Encouraged by the western blot results that showed a clear overexpression of CLTC protein in the transdifferentiated MCF-7 and MALME-3M cells, we decided to determine the role of CME in the adipogenic transdifferentiation process in MALME-3M and in MCF-7 cells. Transient CLTC silencing accomplished by siRNA, caused a significant reduction in LD accumulation induced by petroselinic acid, and further microscopic analysis of Oil Red O and hematoxylin stained cells suggested that not only LD accumulation but also neutral lipid composition in LDs was affected by CLTC silencing.
So far no study has reported the induction of CLTC expression or CME stimulation by monounsaturated fatty acids, although polyunsaturated fatty acids have been found to play a role in the formation of synaptic vesicles and in promoting vesicle budding and membrane trafficking (Darios &[Davletov 2006]; Chernomordik et al.; Chernomordik et al.). It was recently reported that α-Synuclein expression, coupled with exposure to physiological levels of polyunsaturated fatty acids, enhanced CLTC mediated endocytosis in neuronal and non-neuronal cultured cells (Ben Gedalya et al.).
Transferrin receptor (TfR), whose gene expression was upregulated by a fold change of 3.33 in the albumin-associated lipid-treated MALME-3M cells (Table1A), is internalized from the cell plasma membrane and recycled back to the cell plasma membrane specifically via CME (Hanover et al.). El-Jack et al. reported that murine 3T3-L1 preadipocytes, in a differentiation media previously described (Stephens et al.), showed a gradual increase in whole cell TfR levels but a decrease in cell surface TfR levels (El-Jack et al.). The results obtained in this study indicated that the differentiation process might account for the observed alterations in internalization and/or TfR recycling. These results may be useful in understanding why CME is of critical importance in HCCL to adipocyte transdifferentiation.
The clarification of the roles played by the differentially expressed genes and proteins in the process of adipogenic transdifferentiation in HCCL cultures should provide the basic foundation to develop novel molecules for cancer therapy.