When advanced cancers and recurrent cancers are treated with anticancer drugs, the cancers become smaller and sometimes disappear. Many of them, however, recur several months to several years later, and the anticancer drugs become ineffective. Cancer cells include cancer cells that are susceptible to anticancer drugs and cancer cells that are resistant to anticancer drugs, and the former have recently come to be called "cancer stem cells." Cancer stem cells are very tough. They have a strong anticancer drug efflux function, and superior ability to cope with the active oxygen species of the stress response. Consequently, drug discovery and development to combat cancer stem cells has become a very urgent task. Since during the course of investigating the properties of cancer stem cells by visualizing cancer stem cells we have been making efforts to simultaneously detect destructive substances, we will present our results as of the present.
We have also been pursuing research on "cell reprogramming," which converts cancer cells into cells that have more moderate properties. We discovered a method of producing induced pluripotent stem cells (iPS cells) from normal cells by using three microRNAs whose safety has been confirmed in vivo. When we introduced those three microRNAs into cancer cells, they greatly decreased their tumorigenicity. Because of the possibility of serving as a treatment modality with few side effects, we will describe that method as well.

Approximately 1000 species of bacteria in the mammalian intestine interact with their host and grow, and together they constitute the "gut flora." Based on recent gut flora analyses performed with a next-generation sequencer close relationships are now known to exist between abnormal compositions of the bacterial species that make up the gut flora and a variety of human diseases, including inflammatory bowel disease, asthma, obesity, diabetes, colorectal cancer, and liver cirrhosis. Gut flora abnormalities are referred to as "dysbiosis" and are characterized by a "decrease or simplification of the diversity" of the constituent bacteria. In other words, the overall number of gut flora genes the intestine contains is reduced, and indicates a bacterial composition that is functionally inferior as a whole. Moreover, the fact that dysbiosis is a "cause" of certain diseases and that improving dysbiosis is a very effective treatment has been demonstrated by clinical studies of fecal transplantation, etc. However, why dysbiosis is linked to diseases has never been elucidated. One postulated reason is that immune system homeostasis breaks down as a result of dysbiosis. In other words, it appears that because the bacterial species that activate the immune system become predominant, and the bacteria that inhibit the immune system decrease, the net effect may be induction of abnormal activation of the immune system. Against this background, in the research systems in the past that have returned various findings related to the intestinal flora, research designed to identify the effect of individual bacterial species in the intestine has been attracting interest. More specifically, the research attempts to investigate the effects of a certain species of bacteria by administering only that species to germ-free mice thereby producing gnotobiotic mice that harbor only that specific species. By combining gnotobiotic techniques and analysis with next generation sequencers it has gradually become possible to determine the effects of each of the bacterial constituents that make up the gut flora on the differentiation and functions of host immune cells. In this lecture I will describe findings related to the effects of gut flora on the host, primarily on the host's immune system, which are gradually becoming clearer.

Multi-modality therapy that includes surgery is the standard treatment for cancer. The adverse effects of treatment are greatest in far advanced cancer, which requires the most powerful treatment, and there is no latitude for the addition of new drugs. Moreover, such powerful treatments become ineffective in advanced cancers that have spread and are near the terminal stage, and nothing can be done. The purpose of this study was to develop a novel method of treatment for such far advanced cancers. Because phenylbutyrate (PB) is an oral drug, we started our study with a non-digestive system cancer, breast cancer, and then expanded it to colorectal cancer and stomach cancer.
The presence of PB-resistant lines and sensitive lines in breast cancer was confirmed, and rigorous analyses using microarrays showed that epigenetic regulation of Zeb1 is related to PB sensitivity and confirmed strong Zeb1 expression in triple-negative breast cancers. In the future we would like to conduct a clinical study of the efficacy of PB, primarily in recurrent Her2/luminal type breast cancers, which appear to express little Zeb1.
Since Zeb1 expression is known to be important in regulating the epithelial mesenchymal transition (EMT) and to be involved in cancer resistance to treatment, we are also conducting a basic experiment in gastrointestinal cancer based on a similar hypothesis. Expression of Zeb1 itself was found not to be related to PB resistance in colorectal cancer. We therefore performed a comprehensive gene analysis using microarrays. The results revealed that although resistance genes are involved in EMT, they are not under epigenetic regulation, and the candidates for the resistance genes have been narrowed down to Ascl-2, LEF1, and TSPAN8. On the other hand, the resistance genes ALDH7A1 and CALD1 identified as EMT-related genes in stomach cancer have been mentioned as epigenetic regulation genes, but IGFBP2 alone has been identified as a sensitivity gene candidate that might regulate them as a regulatory gene. By contrast, the reduction of LGALS1 expression in a resistant line is regulated epigenetically, and there appeared to be a mechanism that involves hypomethylation in the EMT in stomach cancer.
In conclusion, the EMT phenotype appeared to be related to PB sensitivity, but it appeared that hypermethylation may be important in breast cancer, genomic abnormalities in colorectal cancer, and hypomethylation in stomach cancer. In the future we would like to further verify this hypothesis and aim to perform a clinical trial among PB-sensitive patients.