A carregar...
Publicação
High systemic cholesterol and IFN-γ response in breast cancer cells
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.6 MB | Adobe PDF |
Autores
Resumo(s)
O papel do sistema imune no cancro foi alvo de debate durante mais de um século. Na década de 90, estudos em ratinhos imunocompetentes demonstraram que algumas moléculas do sistema imune têm um papel fundamental na regulação do tumor primário, sendo o IFN-γ uma das mais importantes. Uma das caraterísticas mais atrativas do IFN-γ é a sua capacidade de suprimir o crescimento tumoral através de diversos mecanismos. Algumas das capacidades anti tumorais do IFN-γ sob as células tumorais incluem: a diminuição da proliferação, o aumento da apoptose e necrose; e ainda o aumento de expressão de HLA (o qual permite o reconhecimento das células tumorais por parte das células imunes). No entanto, o IFN-γ é uma citocina pleiotrópica, o que significa que para além de funções anti tumorais, este apresenta também efeitos pro tumorais no microambiente tumoral. O exemplo mais conhecido disto é o facto do IFN-γ aumentar a expressão de PD-L1 nas células tumorais, o qual se liga ao seu recetor PD-1 e isto cria um “imune checkpoint” que permite que as células tumorais escapem às células do sistema imune. O efeito do IFN-γ no microambiente tumoral depende de alguns fatores, como a sua concentração, a composição do microambiente tumoral e o tipo de cancro. Nesta tese, o foco principal é o cancro da mama.
O cancro da mama é o subtipo de cancro que apresenta as taxas de incidência e de mortalidade mais elevadas nas mulheres, a nível mundial. O risco de cancro da mama está associado a vários fatores, sendo um deles a obesidade. Uma anormalidade metabólica sistémica associada à obesidade é a dislipidemia, a qual se traduz num aumento dos níveis de triglicéridos e colesterol-LDL e uma diminuição nos níveis de colesterol-HDL. Um estudo que foi conduzido previamente no nosso laboratório sugere que os níveis plasmáticos de colesterol-LDL foram positivamente correlacionados com o volume do tumor em pacientes com cancro de mama. Adicionalmente, observou-se que pacientes com níveis mais elevados de colesterol-LDL ao diagnóstico apresentam tumores maiores, maior grau de diferenciação e maior taxa proliferativa. O papel do IFN-γ no cancro da mama tem sido estudada ao longo dos anos. Focando nas suas propriedades anti tumorais, há estudos que indicam que o IFN-γ diminui a proliferação das células de cancro da mama, ao bloquear estas células na fase G1 do ciclo celular. Outros estudos revelaram que o IFN-γ consegue modular alguns péptidos do sistema imune no cancro da mama triplo negativo devido ao facto de aumentar a expressão de HLA-I e HLA-II. Focando agora nas consequências pro tumorais do IFN-γ, foi demonstrado em células do cancro da mama que esta citocina aumenta a expressão de PD-L1. No nosso laboratório, temos dados ainda não publicados que suportam uma consequência ainda não descrita para a sinalização do IFN-γ em células de cancro da mama: alterações metabólicas que levam ao aumento da internalização de colesterol-LDL. Isto levou-nos a traçar uma nova linha de pesquisa sobre como a sinalização do IFN-γ é regulada em diferentes contextos metabólicos. Como tal, o principal objetivo desta tese foi compreender se os efeitos do IFN-γ nas células do cancro da mama são afetados por níveis elevados de colesterol-LDL no microambiente tumoral, os quais se correlacionam com a progressão da doença no cancro da mama. Detalhadamente, decidimos analisar o papel do colesterol-LDL na via de sinalização do IFN-γ e nos efeitos que ocorrem a jusante do mesmo, em células de cancro da mama. Neste
estudo, utilizámos a linha celular MDA-MB-231 como modelo (a qual representa um subtipo de cancro da mama triplo negativo humano). Em alguns casos específicos, utilizámos outras linhas celulares para complementar os nossos resultados (a linha de cancro da mama luminal A humano MCF-7, a linha de cancro da mama triplo negativo humano HS578T e a linha de cancro da mama triplo-negativo de ratinho 4T1). Relativamente à nossa primeira experiência, os nossos resultados sugerem uma tendência do colesterol-LDL para influenciar a via de sinalização do IFN-γ (JAK/STAT1), visto que parece aumentar expressão do recetor de IFN-γ (IFNGR), a expressão de STAT1 e a fosforilação de STAT1, nas células MDA-MB-231. Focando na proliferação celular, não conseguimos inferir de que forma a resposta das células de cancro da mama ao IFN-γ é afetada por altos níveis de colesterol-LDL no microambiente tumoral. Isto sucedeu porque não conseguimos observar um efeito do IFN-γ na proliferação e no ciclo celular das células MDA-MB-231, apesar de termos testado várias abordagens (como diferentes time-points ou diferentes linhas celulares). A opção mais provável é que tenha havido heterogeneidade inter-laboratorial das nossas linhas celulares. Posteriormente, focámo-nos no impacto do colesterol-LDL na expressão de proteínas induzidas pelo IFN-γ, como o HLA e o PD-L1. Os nossos dados demonstram que o colesterol-LDL não tem impacto na expressão de HLA, mas consegue aumentar a expressão de PD-L1 na membrana das células MDA-MB-231 a jusante de IFN-γ. Surpreeendentemente, observámos ainda que o colesterol-LDL sozinho induz a localização de PD-L1 na membrana das células. Explorámos, também, o papel do colesterol-LDL na expressão de outras proteínas independentes do IFN-γ, como os ligandos do recetor NKG2D (MICs e ULBPs), onde observámos que o colesterol-LDL leva a um aumento de expressão de MICA/B e de ULBP2/5/6. Devido à importância da expressão de PD-L1 em células tumorais no contexto da imunoterapia, decidimos analisar melhor de que forma o colesterol-LDL afeta a localização do PD-L1 na membrana celular. Primeiro, de modo a entender se o efeito do colesterol-LDL na expressão de PD-L1 corresponde a um mecanismo geral em células de cancro da mama, testamos o seu efeito noutras outras linhas celulares (MCF-7, HS578T e 4T1). Os resultados obtidos foram complexos, uma vez que o colesterol-LDL influenciou a expressão de PD-L1 nas membranas tanto individualmente quanto em combinação com o IFN-γ na linhagem celular 4T1, mas apenas em combinação com o IFN-γ nas células HS578T. Adicionalmente, a exposição de células MCF-7 ao colesterol-LDL não afetou a expressão de PD-L1 nas membranas, quer isoladamente, quer em combinação com IFN-γ. No decorrer da tese de mestrado, um estudo foi de Wang et al. Foi publicado, o qual demonstrou que o colesterol se pode ligar diretamente ao PD-L1 e estabilizar sua estrutura na membrana celular. Ora, isto poderia explicar porque razão a exposição ao colesterol-LDL aumenta a expressão de PD-L1 nas membranas, assumindo que as células estavam a internalizar o colesterol-LDL do meio e enriquecendo as suas membranas com colesterol. Além disso, discrepâncias nessa capacidade de internalizar o colesterol poderiam explicar a variabilidade da resposta ao colesterol-LDL que observámos entre as linhas celulares que utilizámos. De modo a estudar isto, comparámos a capacidade de internalização de colesterol de duas linhas de cancro da mama triplo negativo humano que responderam de forma diferente ao efeito do colesterol-LDL na expressão de PD-L1: MDA-MB-231 e HS578T. Com isto, pudemos observar que, enquanto a exposição das células MDA-MB-231 ao colesterol-LDL levou ao aumento do teor de colesterol das mesmas; o mesmo não aconteceu para a linha celular HS578T. Por fim, avaliámos quais as consequências funcionais do aumento da expressão de PD-L1 pelo colesterol-LDL num ambiente imunológico. Para estudar isso, executámos um “killing assay” com as células MDA-MB-231 e com PBMCs e analisámos a percentagem de células
tumorais apoptóticas. Os nossos resultados demonstram que o colesterol-LDL tem um efeito protetor sobre as células MDA-MB-231, visto que diminui a percentagem de células tumorais apoptóticas. Apesar de serem dados interessantes, estes requerem repetição; e, no futuro, pretendemos ainda entender se a morte mediada pelas PBMCs é dependente de PD-L1 ou não e se o colesterol-LDL afeta isto. Concluindo, os nossos resultados sugerem que níveis elevados de colesterol-LDL no microambiente tumoral afetam a resposta do cancro da mama ao IFN-γ e a localização membranar de PD-L1. Isto pode ter implicações na eficácia da imunovigilância em ambientes enriquecidos em colesterol-LDL e destaca a importância de estudar a resposta imune contra células tumorais em diferentes contextos metabólicos. Os nossos dados sugerem também que o colesterol-LDL pode ter um impacto na resposta dos pacientes com cancro de mama aos inibidores de PD-1 ou PD-L1, uma vez que este aumenta a expressão de PD-L1. Embora seja necessária mais investigação nesta área, acreditamos que este conhecimento poderá vir a contribuir para um melhor diagnóstico e tratamento do cancro de mama num futuro próximo.
IFN-γ is one of the most studied cytokines in the scope of cancer immunosurveillance, since it plays a key role in anti-tumour immunity, being capable of suppressing tumour growth through numerous mechanisms. Despite its anti-tumorigenic functions, there is also evidence suggesting that IFN-γ has pro-tumorigenic effects. Thus, IFN-γ is now known as a pleiotropic cytokine and the consequences of its presence in the tumour microenvironment (TME) depend on different factors, including IFN-γ concentration, the composition of the TME and tumour-specific contest. The role of IFN-γ in breast cancer cells has been studied over the years. Focusing on its anti-tumoral properties, it has been suggested that IFN-γ decreases the proliferation of breast cancer cells by blocking them in the G1 phase of the cell cycle. Other findings indicate that IFN-γ modulates the immunopeptidome in triple-negative breast cancer (TNBC) by increasing HLA-I and HLA-II expression. Regarding pro-tumoral consequences, IFN-γ increases the expression of PD-L1 in several cancer cells, including breast cancer cells; which can promote tumour escape from immunosurveillance. Our laboratory has unpublished data supporting a not yet described consequence for IFN-γ signalling in breast cancer cells: metabolic alterations leading to increased LDL-cholesterol uptake. This led us to design a new research line on how IFN-γ signalling is regulated in different metabolic contexts. As such, the major aim of this thesis was to understand if the effects of IFN-γ in breast cancer cells are affected by high environmental LDL-cholesterol, which correlated with disease progression in breast cancer. In detail, we analysed the role of LDL-cholesterol on IFN-γ signalling and downstream effects in breast cancer cells. As a model, we used mainly the human triple-negative breast cancer cell line MDA-MB-231 and, in specific cases, other cell lines to complement our findings (the human luminal A cell line MCF-7, the human triple-negative cell line HS578T and the mouse triple-negative cell line 4T1). In our study, we observed a tendency of LDL-cholesterol to influence IFN-γ signalling by increasing IFNGR expression, STAT1 expression and STAT1 phosphorylation in MDA-MB-231 cells. Concerning cellular proliferation, we were unable to understand how the response of breast cancer cells to the IFN-γ is affected by high environmental LDL-cholesterol. This happened because we were not able to observe the effect of IFN-γ on breast cancer cell proliferation in the first place. Furthermore, we focused on the impact of LDL-cholesterol in IFN-γ-induced proteins, such as HLA and PD-L1. Our data indicate that LDL-cholesterol does not affect HLA-A2 expression, but it increases PD-L1 expression at the cell surface of MDA-MB-231 cells downstream of IFN-γ. Surprisingly, we also observed that LDL-cholesterol alone induces PD-L1 localization at the membranes. We also tested the impact of LDL-cholesterol in other proteins that are not dependent on IFN-γ, like MICA/B and ULBP2/5/6 (NKG2D ligands), where we observed that LDL also increases the surface expression of these proteins. Given the importance of the expression of PD-L1 on cancer cells in the context of immunotherapy, we decided to further analyse how LDL-cholesterol affects PD-L1 membrane localization. First, to understand if the effect of LDL-cholesterol on PD-L1 expression corresponds to a general mechanism in breast cancer cells, we tested it in other cell lines (MCF-7, HS578T and 4T1). The obtained results were complex, as we were able to observe that LDL-cholesterol influenced PD-L1 expression at membranes both individually or in combination with IFN-γ in the 4T1 cell line, but only in combination with IFN-γ in HS578T cells. Finally, the exposure of MCF-7 cells to LDL-cholesterol did not affect PD-L1 expression at membranes either alone or in combination with IFN-γ. While the master thesis was ongoing, a study was published by Wang et al., which described that cholesterol could bind directly to PD-L1 and stabilize its structure on the cell membrane. This could explain why exposure to LDL-cholesterol increases PD-L1 expression at membranes, assuming that cells were taking up LDL-cholesterol from the media and enriching its membranes with cholesterol. Also, discrepancies in this capacity to internalize cholesterol could account for the variability of the response to LDL-cholesterol we observed between the cell lines used. To gain further insight into this, we compared the two human triple-negative cell lines that responded differently to LDL-cholesterol -MDA-MB-231 and HS578T- for their ability to internalize cholesterol. We were able to see that, while the exposure of MDA-MB-231 cells to LDL-cholesterol led to the increased cholesterol content of cells; the same did not happen for the HS578T cell line. Finally, we assessed what are the functional consequences of PD-L1 increased expression by LDL-cholesterol within an immune environment. To study this, we performed a killing assay with MDA-MB-231 cells and peripheral blood mononuclear cells (PBMCs) and analysed the percentage of apoptotic tumour cells. Our data indicate that LDL-cholesterol has a protective effect over MDA-MB-231 cells, as it decreases the percentage of apoptotic tumour cells. These are interesting findings that require repetition; and, in the future, we aim to understand if this killing is dependent on PD-L1 or not and if LDL-cholesterol affects this. Altogether, our results suggest that high environmental LDL-cholesterol affects breast cancer response to IFN-γ and PD-L1 membrane localization. This could have implications for the efficacy of immunosurveillance in high LDL-cholesterol environments and highlights the importance of studying the immune response against cancer cells in different metabolic contexts. Our findings also suggest that LDL-cholesterol could have an impact on breast cancer patients’ response to PD-1 or PD-L1 checkpoint inhibitors since it increases PD-L1 expression. Although more research is required, we believe that this knowledge may contribute to better diagnosis and treatment of breast cancer in the future.
IFN-γ is one of the most studied cytokines in the scope of cancer immunosurveillance, since it plays a key role in anti-tumour immunity, being capable of suppressing tumour growth through numerous mechanisms. Despite its anti-tumorigenic functions, there is also evidence suggesting that IFN-γ has pro-tumorigenic effects. Thus, IFN-γ is now known as a pleiotropic cytokine and the consequences of its presence in the tumour microenvironment (TME) depend on different factors, including IFN-γ concentration, the composition of the TME and tumour-specific contest. The role of IFN-γ in breast cancer cells has been studied over the years. Focusing on its anti-tumoral properties, it has been suggested that IFN-γ decreases the proliferation of breast cancer cells by blocking them in the G1 phase of the cell cycle. Other findings indicate that IFN-γ modulates the immunopeptidome in triple-negative breast cancer (TNBC) by increasing HLA-I and HLA-II expression. Regarding pro-tumoral consequences, IFN-γ increases the expression of PD-L1 in several cancer cells, including breast cancer cells; which can promote tumour escape from immunosurveillance. Our laboratory has unpublished data supporting a not yet described consequence for IFN-γ signalling in breast cancer cells: metabolic alterations leading to increased LDL-cholesterol uptake. This led us to design a new research line on how IFN-γ signalling is regulated in different metabolic contexts. As such, the major aim of this thesis was to understand if the effects of IFN-γ in breast cancer cells are affected by high environmental LDL-cholesterol, which correlated with disease progression in breast cancer. In detail, we analysed the role of LDL-cholesterol on IFN-γ signalling and downstream effects in breast cancer cells. As a model, we used mainly the human triple-negative breast cancer cell line MDA-MB-231 and, in specific cases, other cell lines to complement our findings (the human luminal A cell line MCF-7, the human triple-negative cell line HS578T and the mouse triple-negative cell line 4T1). In our study, we observed a tendency of LDL-cholesterol to influence IFN-γ signalling by increasing IFNGR expression, STAT1 expression and STAT1 phosphorylation in MDA-MB-231 cells. Concerning cellular proliferation, we were unable to understand how the response of breast cancer cells to the IFN-γ is affected by high environmental LDL-cholesterol. This happened because we were not able to observe the effect of IFN-γ on breast cancer cell proliferation in the first place. Furthermore, we focused on the impact of LDL-cholesterol in IFN-γ-induced proteins, such as HLA and PD-L1. Our data indicate that LDL-cholesterol does not affect HLA-A2 expression, but it increases PD-L1 expression at the cell surface of MDA-MB-231 cells downstream of IFN-γ. Surprisingly, we also observed that LDL-cholesterol alone induces PD-L1 localization at the membranes. We also tested the impact of LDL-cholesterol in other proteins that are not dependent on IFN-γ, like MICA/B and ULBP2/5/6 (NKG2D ligands), where we observed that LDL also increases the surface expression of these proteins. Given the importance of the expression of PD-L1 on cancer cells in the context of immunotherapy, we decided to further analyse how LDL-cholesterol affects PD-L1 membrane localization. First, to understand if the effect of LDL-cholesterol on PD-L1 expression corresponds to a general mechanism in breast cancer cells, we tested it in other cell lines (MCF-7, HS578T and 4T1). The obtained results were complex, as we were able to observe that LDL-cholesterol influenced PD-L1 expression at membranes both individually or in combination with IFN-γ in the 4T1 cell line, but only in combination with IFN-γ in HS578T cells. Finally, the exposure of MCF-7 cells to LDL-cholesterol did not affect PD-L1 expression at membranes either alone or in combination with IFN-γ. While the master thesis was ongoing, a study was published by Wang et al., which described that cholesterol could bind directly to PD-L1 and stabilize its structure on the cell membrane. This could explain why exposure to LDL-cholesterol increases PD-L1 expression at membranes, assuming that cells were taking up LDL-cholesterol from the media and enriching its membranes with cholesterol. Also, discrepancies in this capacity to internalize cholesterol could account for the variability of the response to LDL-cholesterol we observed between the cell lines used. To gain further insight into this, we compared the two human triple-negative cell lines that responded differently to LDL-cholesterol -MDA-MB-231 and HS578T- for their ability to internalize cholesterol. We were able to see that, while the exposure of MDA-MB-231 cells to LDL-cholesterol led to the increased cholesterol content of cells; the same did not happen for the HS578T cell line. Finally, we assessed what are the functional consequences of PD-L1 increased expression by LDL-cholesterol within an immune environment. To study this, we performed a killing assay with MDA-MB-231 cells and peripheral blood mononuclear cells (PBMCs) and analysed the percentage of apoptotic tumour cells. Our data indicate that LDL-cholesterol has a protective effect over MDA-MB-231 cells, as it decreases the percentage of apoptotic tumour cells. These are interesting findings that require repetition; and, in the future, we aim to understand if this killing is dependent on PD-L1 or not and if LDL-cholesterol affects this. Altogether, our results suggest that high environmental LDL-cholesterol affects breast cancer response to IFN-γ and PD-L1 membrane localization. This could have implications for the efficacy of immunosurveillance in high LDL-cholesterol environments and highlights the importance of studying the immune response against cancer cells in different metabolic contexts. Our findings also suggest that LDL-cholesterol could have an impact on breast cancer patients’ response to PD-1 or PD-L1 checkpoint inhibitors since it increases PD-L1 expression. Although more research is required, we believe that this knowledge may contribute to better diagnosis and treatment of breast cancer in the future.
Descrição
Tese de mestrado, Oncobiologia, Universidade de Lisboa, Faculdade de Medicina, 2023
Palavras-chave
IFN-γ Colesterol-LDL Cancro da mama PD-L1 Microambiente tumoral Microambiente T Teses de mestrado - 2023