Efficient Enrichment Of Hepatic Cancer Stem Cells From A Primary Rat HCC Model

In recent years, cancer stem cells (CSCs) has created great excitement in theresearch community, this notion is said that: tumors are maintained by their ownstem cells. This group of cells remain dormant or incessantly generate the tumorcells that lead to cancer. They are thought to be involved in tumor resistance totherapy, tumor relapse and progression. CSCs are important, studies show thatthey are responsible for drug resistance. Obtaining further understanding of itsspecific traits will offer insights regarding the early stages of tumorigenesis to aidin preventing this phenomenon and to enhance the selectivity of antitumortherapies. Recently, CSCs were isolated from variety of tumors, such as tumorsof the breast, prostate, brain and colon, etc. Even so, the existence of stem cells isstill a controversial topic.Hepatocellular carcinoma (HCC) is a highly malignant tumor associated withfrequent metastasis and poor prognosis. Isolated hepatic CSCs (HCSCs) representa suitable in vitro model for developing therapeutic strategies aimed ateradicating the tumorigenic subpopulation within HCC. It is for this reason that the isolation and identification of HCSCs are crucial. One common method forisolating CSCs has been to characterize their cell-surface phenotype and usemarkers to negatively or positively select for particular cells, such as CD133andEpCAM, using fluorescence-activated cell sorting (FACS) and magnetic-activated cell Sorting (MACS) to isolate. However, HCSCs markers have beenshown to exhibit limited plasticity. In order to remedy this Weakness, anothermethod for CSC separation is based on the differential efflux of fluorescent dyes,such as rhodamine123or Hoechst33342. Recently, the isolation of sidepopulation (SP) cells using Hoechst33342dye has become a useful method forobtaining CSCs from various tumors. However, there are still some limitationsassociated with this method, such as the poisonousness of dyes, detecting falsepositive stem cells and the requirement for special instruments. So we chooseanother method, density gradient centrifugation. In a previous study, wesuccessfully isolated fetal liver stem/progenitor cells (FLSPCs) from primarycultured fetal liver cells via density-gradient centrifugation centered on athree-step method. We aimed to determine whether HCSCs can be obtained byexploiting their physical properties. So, we improve the design to conduct theisolation of HCSCs. This new strategy for isolating HCSCs will make greatcontributions in two main areas: it will provide new insights related to theisolation of other kinds of CSCs; and it will allow us to investigate thecontribution of HCSCs to HCC.Aims:Based on physical properties of HCSCs, use percoll continuous gradientcentrifugation to enrich HCSCs. Then identify them whether they have astem-like property through morphology and function test. Finally test theirTumorigenic ability by tumor formation in NOD/SCID mices. Methods:Injecting diethylnitrosamine(DENA) intraperitoneally to Fischer344rats toestablish liver cancer model. Then culture the hepatic cancer cells in vitro and usedifferential trypsinization to purify the cells preliminarily. Finally, Percolldiscontinuous density gradient centrifugation, divide the cells into four fractionsaccording to the density beads.Through transmission electron microscopy and immunofluorescencestaining with Phalloidin, observe the morphological comparison among thedifferent fractions. Then do the AFP and CK-19immunofluorescence staining todetermine cell origin, Followed with flow cytometry and quantitative real-timedetection of CSCs marker levels of these cells. Then through the growth curve,Transwell experiments to test their ability of proliferation, migration and invasion.Finally take tumor formation in NOD/SCID mice to detect cells’ tumorigenicity.Results:We found the hepatic cancer cells of fraction3were most likely enrichedwith HCSCs. All cells from Percoll are hepatocellular source by the AFP andCK-19immunofluorescence staining. Transmission electron microscopy andphalloidin staining show that morphology of fraction3is similar with cancerstem cells, and there are a lot of cilia and pseudopodiamore around. Growth curveand transwell assay show that the fraction3has the strongest ability ofproliferation, migration and invasion. The results of flow cytometry revealed thatthe expressions of EpCAM (84.5±8.31)%and AFP (85.2±8.31)%of fraction3were significantly different from the others. The results of RT-PCR revealed thatthe expressions of EpCAM (111.34±5.59) and AFP (13.76±0.95) of fraction3were significantly different to the others. Meanwhile for CD133of fraction3(17.16±0.42), which were significantly different from fraction2and fraction4, but not significant different from fraction1.As few as1×104cells from fraction3could initiate tumors not only in subcutaneous tissues, but also in liver tissues ofNOD/SCID mice. And the size of tumors formed by fraction3was much largerthan any other groups. Pathological analysis indicated that the tumors from thesubcutaneous regions and tumors from livers were all hepatic carcinoma derived.Conclusion and significance:Fraction3cells that isolated from primary cultured rat hepatic cancer cellsby Percoll discontinuous density gradient centrifugation, show morphologicallysimilar to CSCs. Learn from the functional aspects, the ability of proliferation,invasion and tumorigenicity are far more strong than the other fractions. All theseevidences prove that this fraction of cells is HCSCs that we needed.Compared to the surface markers isolation, although the accuracy issomewhat less, but our method is simple, affordable and enough cells can beenriched, completely free from markers sorting restrictions due to non-specificmarkers. This is a good way to enrich the HCSCs, and this study will greatlycontribute to two important areas of biological interest: CSC isolation and HCCtherapy.