The Current Understanding of Cancer Stem Cells and How They Can be Used To Treat Cancer

Cancer is the 2^nd most prevalent cause of death, accounting for 10 million annual deaths and half of all deaths for individuals older than 70 [out world in data]. The current standard of care is composed firstly of tumor resection and subsequent chemotherapy and radiation. These treatments have significantly reduced mortality of many cancers (such as prostate cancer, whose 5 year survival rate has increased from 68.7% in 1970 to 98.65 in 2013) and increased the overall 5-year survival rate of all cancers from 50.3% to 67% in the same period. However, despite this improvement, an aging population in many countries has led to a higher incidence rate of cancer, so both the absolute number and share of deaths attributed to cancer is increasing. To combat this, research investigating new approaches to treating cancer, such as immunotherapy, angiogenesis disruption, and differentiation therapy, is being conducted. One major avenue of research has been the study of tumor initiation and Cancer Stem Cells (CSC). Here, the rational for studying CSCs in relation to cancer will be described and several studies will be used to exemplify our current understanding of tumor initiation via CSCs.

              The rationale for CSC study in relation to cancer can firstly be better understood in the larger context of our approach to cancer treatment. Unlike external diseases, such as contagions like bacteria and viruses, one of the main barriers to cancer treatment is the similarity between healthy and tumorigenic cells. For one, this leads to an inability of the immune system to detect the cancer as it carries all of the markers of ‘self’ – that is, the body recognizes it as a normal cell that is ‘supposed’ to be there. Clinically, it makes cancers difficult to treat as often what will kill a cancer will, by the same mechanism, kill non-cancerous cells. Ergo, the approach to cancer treatment hinges on differentially targeting cancerous cells from non-cancerous cells (or minimally, targeting cancer predominantly). For example radiotherapy does this by ionizing all cells in a region, but depends on the rapid proliferative property of tumors. This property means DNA repair machinery in tumors are less effective, and thus those cells will be more damaged than more quiescent cells (which, are non-cancerous). Note the limitation of this is that all cells are affected and some non-cancerous cells rapidly proliferate too (such as epithelial stem cells of the gastrointestinal tract). These cells are thus affected by radiotherapy and causes the GI discomfort, nausea, and vomiting associated with cancer treatment. Similarly, immunodeficiency is caused by current cancer treatments. This demonstrates that our approach to treating cancer centers on finding molecular or behavioral differences between non-cancerous and cancerous cells in addition to demonstrating the need for more specific targeting of cancer than the standard of care today.

              To this end, the discovery of cancer stem cells and the adoption of the cancer stem cell hierarchy model of cancer has provided a possible treatment target. The CSC hierarchy model was one of two models for cancer, the other being clonal evolution. The latter of these proposed that the division of a cancer cell created progeny with equal differentiability and properties, unless one of the progeny were to attain subsequent mutations. It further suggested that evolutionary pressures would eventually create a tumor, where one of the progeny had accumulated sufficient mutations to carry all of the classical characteristics of a cancer. Conversely, CSC hierarchy proposed that within a tumor was a small subset of so-called ‘cancer stem cells’ which could divide into differentiated progeny with a more narrow differential capacity (in this way, CSCs are the progenitors of other cells of the tumor, analogous to a hematopoietic stem cell being the progenitor of various differentiated progeny). The CSC hierarchy model was adopted, evidence for which will be presented. One of the implications of this model is the recognition that cancer stem cells are often much more quiescent than their progeny – ergo the bulk of a tumor may be made up of CSC daughter cells. Historical treatments which target the rapid proliferation of cancer cells are therefore less effective against the more quiescent cancer cells. Though the tumor size may decrease as a result of, for example, radiotherapy, these CSCs may survive the treatment as their repair machinery is effective in their quiescent state. Therefore, once the treatment is terminated, the cancer may, once again, proliferate. For this reason, it has become of interest to find specific molecular markers for CSCs to better treat cancer.

              One major method used for identification of such markers is a technique called ‘Lineage Tracing’. Lineage tracing is a technique that can be used in transgenic mice to elucidate which cells are downstream of a specific parent. This is done by using the tamoxifen-inducible Cre-LoxP system to create selective fluorescence in the parents in progeny. Cre is an enzyme which can remove a region flanked by LoxP sites. By making a transgenic mice where Cre is coexpressed with a marker of interest by sharing a promotor and by making a stop codon flanked by LoxP sites, preceded by a constitutively ON promotors and succeeded by a fluorescence gene, we create a system whereby when the marker is expressed, Cre is expressed, cutting out the stop codon and allowing for fluorescent expression. Subsequently, all progeny of this cell will have this cut out stop codon and so will also express the fluorescent reporter. This thus gives us spatial specificity. An additional layer of specificity is created on a temporal axis by tamoxifen Cre-inducibility. By fusing Cre with an estrogen receptor to make CreER, the Cre enzyme is only able to re-enter the nucleus to perform its action when tamoxifen (an estrogen analogue) is bound to it. Therefore, temporal specificity for when the stop codon is cut out is attained by pulsing the cell with tamoxifen to allow the cells expressing the marker of interest to all CreER to re-enter the nucleus, cut out the stop codon, and cause fluorescent expression. Note, the Cre-LoxP system can also be used to simply knock out genes (without fluorescence) by flanking the gene of interest in LoxP sites.

              This technique was used to great effect in Li Et Al. to show that Lgr5+ stem cells were the cellular origin of invasive intestinal-type gastric cancer (IGC) in mice. They posed the question: what cells are the progenitor to IGCs? Their hypothesis was that gastric Lgr5 (a surface receptor) positive cells (lgr5 was a known marker for epithelial stem cells) were the progenitor to IGC tumors (and not normal Lgr5 progenitors, such as parietal cells, pit cells, and corpus Lgr5+ chief cells). To this extent, they knocked out tumor suppressor genes Smad4 and PTEN (to induce IGCs), using the previously described Cre-LoxP system, in specific cell types (either lgr5+ cells or their progeny (using Capn8-Cre and atp4b-Cre). What was observed was Lgr5+ cells with knocked out (KO) Smad4 and PTEN caused both adenomas and invasive IGCs. By contrast, PTEN and Smad4 KO in parietal and pit cells did not lead to tumor initiation. In otherwords, they showed that lgr5+ stem cells act as CSCs in IGCs, and therefore that targeting LGR5+ cells (ie: targeting the CSCs of an IGC) might be an effective way to target cancer treatment.

              Similarly, Blaas Et Al. used lineage tracing to elucidate the differential palette of Lgr6+ cells. Established was the two populations of lgr6+ cells: luminal stem cells and basal stem cells. They used lineage tracing to affirm that these two populations were both unipotential (produced either luminal or basal cells) unlike their bipotential progenitors (mammary gland stem cells). They also showed that lgr6+ cells were potential CSCs for luminal mammary tumors by knocking out tumor suppressors Brca1 and p53 in lgr6+ cells. This knockout (done via Brca1^loxP/loxP:p53^loxP/loxP:Lgr6-CreER with a tamoxifen pulse) caused the formation of luminal cancer. Higher lgr6+ numbers in luminal mammary tumors was proportional with lower survival rates and depleting the tumor of lgr6+ cells led to less tumor proliferation and aggressiveness. All of this suggests that Lgr6+ cells are a CSC for luminal mammary tumors. Moreover, it was shown that, similarly to Lgr5+ cells in Li Et Al., lgr6+ stem cells would be a good target for cancer treatment.

              Together, these two studies serve to demonstrate how the Cre-LoxP system and lineage tracing can be used to identify useful targets for cancer therapy. The method was expanded on for an organismal-wide lineage tracing experiment in Multi-Organ mapping of cancer risk by Zhu et Al. They created a transgenic mouse that was Prom1+. They then harvested cells from the organs of these mice and transplanted and lineage traced them in control mice; they found that some of these Prom1+ cells were more proliferative than others, and that the ones that were more proliferative were more likely to be CSCs. In otherwords, this group used a ‘shotgun’ methodology to screen for stem cells that were likely to become tumorigenic. In this way, they could identify cells which could be good targets for cancer treatments.

              It is clear then, that CSCs are an area of active research and of great potential for identifying targets for cancer therapy. It is one avenue of finding more targeted solutions than radiotherapy and additionally is an avenue for personalized medicine as we gain the ability to characterize cancer biopsies and determine specifically which cells of the tumor are the CSCs that need to be targeted. With more research and a better understanding of tumor initiation we will surely be able to create more potent cancer therapies with fewer adverse systemic effects.

Sources

1.     Zhu L, Finkelstein D, Gao C, Shi L, Wang Y, López-Terrada D, Wang K, Utley S, Pounds S, Neale G, Ellison D, Onar-Thomas A, Gilbertson RJ. Multi-organ Mapping of Cancer Risk. Cell. 2016 Aug 25;166(5):1132-1146.

2.     Blaas L, Pucci F, Messal HA, Andersson AB, Josue Ruiz E, Gerling M, Douagi I, Spencer-Dene B, Musch A, Mitter R, Bhaw L, Stone R, Bornhorst D, Sesay AK, Jonkers J, Stamp G, Malanchi I, Toftgård R, Behrens A. Lgr6 labels a rare population of mammary gland progenitor cells that are able to originate luminal mammary tumours. Nat Cell Biol. 2016 Dec;18(12):1346-1356.

3.     Li XB, Yang G, Zhu L, Tang YL, Zhang C, Ju Z, Yang X, Teng Y. Gastric Lgr5(+) stem cells are the cellular origin of invasive intestinal-type gastric cancer in mice. Cell Res. 2016 Jul;26(7):838-49.

4.     Roser, M., & Ritchie, H. (2015, July 3). Cancer. Retrieved from https://ourworldindata.org/cancer#cancer-is-one-of-the-leading-causes-of-death