Correlation Between The Cell Membrane Phospholipids And The Metastasis Of Murine Ascites Hepatocarcinoma With Different Lymphatic Metastatic Potential

Background: Tumor metastasis is regulated by many different factors and is theresult of the complicated interactions between a variety of biological molecules andsignaling pathways in time and space. Therefore, it is of important significance to studykey molecular mechanisms underlying tumor lymphatic metastasis and to developeffective invention means against tumor. Phospholipids are a class of lipids that are amajor component of all cell membranes and participate in various biological functionsof tumor cells such as cell proliferation, migration, adhesion, signaling，apoptosis, andcell cycle. Therefore，the prognosis of patients with tumor will be enhanced if keyphospholipid molecules in tumor metastatic progression are identified and thus tumorbiomarkers are found out based on these key phospholipid molecules to assist in theearly detection of cancer and act as therapeutic targets for intervention. Phosphatidylcholine （PC）, phosphatidyl ethanolamine （PE）, and phosphatidylinositol（PI） are themain components of cell membrane. The biological function of cells is strongly affectedby asymmetric distribution and un-identical fatty acids of phospholipids on the bothsides of cell membrane. Recent studies revealed that both phospholipids and theirmetabolites are related to tumor metastasis. Recent research found that PS content ofmelanoma hematogenous pulmonary metastases are related to different progressionstages of tumor cell lines malignant degree. The various phospholipid compositionsplay a variety of roles in the process of tumor metastasis, therefore, we can developnovel anticancer drugs through the inhibition of cell proliferation, migratiov, adhesion,signal transduction, immune and other links. Targeting therapy of cancer based on thephospholipid molecular has been widely concerned. However, the relevance betweenthe membrane phospholipid composition and changes of cancer lymphatic metastasispotential has not been involved. Elemene, extracted from the traditional Chinese medicinal herb Curcumawenyujin, is a mixture of β-, γ-and δ-elemene with β-elemene as the main componentand the other two as the lesser amounts.Experimental pharmacological studiesconfirmed that the effect of elemene on multiple tumor cell in vivo has stronglyinhibition and killing effect, and elemene can inhibit the growth of multiple tumor cells,as well as broad-spectrum antitumor effect.Elemene replace traditional chemotherapyvia hepatic artery interventional therapy primary carcinoma of the liver, with excellenteffect, low toxicity, easy retention in tumor local advantages, for the interventiontherapy of tumor and increase the new vigor.Curcumin is extracted from herbs turmericroot, is a kind of plant polyphenols, has a variety of pharmacological activity, plays animportant role in inflammation, antitumour, atherosclerosis, lipid peroxidation, antiviral.Therefore, in-depth study of anticancer mechanism on elemene and curcumin is ofimportant practical significance.Our research group screened a murine ascites hepatocarcinoma cell strain withhigh metastatic potential in lymphatic system of mice HCa-P/L₆（rate of lymphaticmetastasis100%）spreading to lymphatic nodes of multiple sites in mice was screenedfrom a murine ascites hepatocarcinoma cell strain with low metastatic potential HCa-P（rate of lymphatic metastasis <20%）. HCa-P/L₆is a novel sensitive cell line model forinvestigating the mechanisms of the lymphatic spread of a tumor and the effect ofmedicine on cells. HCa-P/L₆and HCa-P have basically the same genetic background,and are a mutual reference of tumor cell models with different lymphatic metastaticpotential in studying the mechanisms of tumor metastasis and the screening ofantitumor drugs. At present, the group applied limiting dilution analysis to separatedHCa-P and got4cloning cell strains（a non-matestatic cell strain H5and three lowmatestatic cell strains G2, B8, and E10） as experimental materials to study themechanisms of the lymphatic metastasis of cancer liver.Objective:1. To compare the difference of cell membrane phospholipids （PC, PE and PI）,firstly between a murine hepatocarcinoma line HCa-P/L₆（with high lymphaticmetastasis rate at70%～80%） and HCa-P （with low lymphatic metastasis rate at0～20%）, secondly between three cell strains with low metastatic potential （G2, B8, andE10） and non-metastatic cell strain （H5）, and thirdly between three kinds of transplantedtumor cells with low lymphatic potential and one kind of non-metastatic transplantedtumor cells respectively from mouse foot pad inoculation of G2, B8, E10and H5by the use of HPLC, and thus to reveal the correlation between the membrane phospholipidsand the metastasis of tumors with different lymphatic metastatic potential to provideexperimental proof to elucidate the mechanism of tumor metastasis.2. To compare the variation of cell membrane phospholipid firstly between amurine hepatocarcinoma line HCa-P/L₆（with high lymphatic metastasis rate at70%～80%） and HCa-P （with low lymphatic metastasis rate at0～20%）, secondly betweenthree cell strains with low metastatic potential （G2, B8, and E10） and non-metastaticcell strain （H5）, and thirdly between three kinds of transplanted tumor cells with lowlymphatic potential and one kind of non-metastatic transplanted tumor cells respectivelyfrom mouse foot pad inoculation of G2, B8, E10and H5by the use of HPLC after beingtreated with elemene or curcumin, and to observe the effection of elemene or curcuminon cell membrane phospholipids and provide experimental proof to seek therapeutictargets based on phospholipid molecues against tumors. Two murine hepatocarcinomacell lines with high and low metastatic potential（HCa-P/L₆and HCa-P） were used toinvestigate and compare effects of elemene or curcumin on some biologicalcharacteristics of these two cell lines to explain the mechanism of the effectiveness ofphospolids on metastasis of cancer.Methods:1. Firstly a murine hepatocarcinoma line HCa-P/L₆（with high lymphaticmetastasis rate at70%～80%） and HCa-P （with low lymphatic metastasis rate at0～20%）, secondly three cell strains with low metastatic potential （G2, B8, and E10） andnon-metastatic cell strain （H5）, and thirdly three kinds of transplanted tumor cells withlow lymphatic potential and one kind of non-metastatic transplanted tumor cells,respectively from mouse foot pad inoculation of G2, B8, E10and H5, were chosed astest materials. Cell membrane was dissociated by differential and density gradientcentrifugation extracted by liquid-liquid extraction. PC, PE and PI in cell membranewere detected by and high-performance liquid chromatography （HPLC）.2. The cytotoxicity of elemene and curcumin on two murine asciteshepatocarcinoma cell lines with different lymphatic potential （high HCa-P/L₆and lowHCa-P） was analysized by the use of MTT. Then, the maximum uncytotoxicity doseand the half maximal inhibitory concentration （IC50） were calculated. Pretreated byelemene or curcumin at the maximum uncytotoxicity dose, the effectiveness of elemeneor curcumin on HCa-P/L₆and HCa-P was observed by making cell growth curve andcell population doubling time; the effectiveness of elemene on cell cycle of HCa-P/L₆ and HCa-P was demonstrated by flow cytometry （FCM）.Results:1. There was difference of HPLC chromatograms of PE and PI of cell membranebetween two murine ascites hepatocarcinoma cell lines with different lymphaticpotential （high HCa-P/L₆and low HCa-P）, but no significant difference of that of PC.The level of PE and PI in HCa-P/L₆was higher than that in HCa-P, this phenomenonmay suggest that higher PE and PI may be related to higher potential of cancermetastatic cells.2. Neither significant difference HPLC chromatograms of PC was found amongfour cloning cell strains from HCa-P, nor significant difference HPLC chromatogramsof PE and PI among three low lymphomatic metastatic cloning cell strains （G2, B8andE5）. But there was difference of HPLC chromatograms of PE and PI of cell membranebetween G2and non-metastatic cell strain （H5）, the content level of PE and PI in G2was higher than that in H5, this phenomenon may demonstrate that higher PE and PImay be related to metastatic potential of cancer cells.3. After four cloning cell strains separated from HCa-P were respectivelysubcutaneously inoculated to the right foot pad of every mouse615, the rate oftransplanted tumor formation is respectively G2-35%，B8-20%，E10-25%，and H5-10%,and the rate of lymphatic spread of cell strains in615mice is respectively G2-25%,B8-10%, E10-10%and H5-0.4. After four cloning cell strains separated from HCa-P were respectivelysubcutaneously inoculated to the right foot pad of every mouse615, the content level ofPC in low metastatic strain G2is lower than that in other three transplanted tumor cellstrains（t test, P<0.05）, but no significant diffence among H5, B8, and E10. The contentlevel of PE and PI in G2is higher than that in non metastatic strain H5（t test，P<0.05）,but no significant difference among G2, B8, and E10. No significant change of thecontent level of PC, PE and PI is found between transplanted tumor cells from G2, H5,B8, and E10and its homologous parental cloning cell strains.5. Elemene at the concentration of20～100μg/mL can obviously inhibit thecells of two strains growing in a dose-dependent manner（P<0.05） after HCa-P/L₆andHCa-P cells have been protreated with curcumin for48hours. Elemene had moreinhibition on HCa-P/L₆with IC5048μg/mL than on HCa-P with IC5090μg/mL. Themaximum uncytotoxicity dose is10μg/mL. Pretreated by elemene at10μg/mL for7days, the proliferation of HCa-P/L₆and HCa-P were inhibited in time-dependent manner （t test P<0.05） showing by cell growth curves and prolonging populationdoubling times （t test P<0.05）. Pretreated by10μg/mL elemene for48h, cell cycles inHCa-P/L₆and HCa-P were redistributed with HCa-P/L₆and HCa-P being blocked in Sphase and the blocking effectiveness in HCa-P/L₆was stronger than that in HCa-P; thecontent level of PE and PI was reduced in HCa-P/L₆and HCa-P （t test, P<0.05）, whileno significant change of the content level of PC was found between before and afterbeing treated with elemene （t test P﹥0.05）; the content level of PE and PI was reducedin four cloning cell strains（G2, B8, E10, and H5）（t test, P<0.05）, especially in G2withsignificant less phospholipid content level than in other three strains, while nosignificant change of the content level of PC was found; all the content level of PEand PI was reduced in four transplanted tumor cell strains （t test P<0.01） respectivelyfrom G2, B8, E10, and H5being subcutaneously inoculated to the right foot pad ofevery mouse615（t test P<0.01）, especially in G2transplanted tumor cell strain withsignificant less phospholipid content level than in other three strains, while nosignificant change of the content level of PC was found.6. Curcumin at the concentration of15μmol/l～250μmol/l can obviously inhibitthe cells of two strains growing in a dose-dependent manner（P<0.05） after HCa-P/L₆and HCa-P cells have been pretreated with curcumin for48hours. The IC50of HCa-P/L₆phase and the blocking effectiveness in HCa-P/L₆was stronger than that in HCa-P; thecontent level of PE and PI was reduced in HCa-P/L₆and HCa-P （t test, P<0.05）, whileno significant change of the content level of PC was found between before and afterbeing treated with elemene （t test P﹥0.05）; the content level of PE and PI was reducedin four cloning cell strains（G2, B8, E10, and H5）（t test, P<0.05）, especially in G2withsignificant less phospholipid content level than in other three strains, while nosignificant change of the content level of PC was found; all the content level of PEand PI was reduced in four transplanted tumor cell strains （t test P<0.01） respectivelyfrom G2, B8, E10, and H5being subcutaneously inoculated to the right foot pad ofevery mouse615（t test P<0.01）, especially in G2transplanted tumor cell strain withsignificant less phospholipid content level than in other three strains, while nosignificant change of the content level of PC was found.is51.48μmol/L, while the IC50of HCa-P is86.48μmol/L, the latter is higher than the former. Treated by the maximumuncytotoxicity dose of curcumin15μmol/L for7days, the growth of HCa-P/L₆andHCa-P was inhibited （t test P<0.05） in time-dependent manner （χ2test P <0.01） andthe population doubling times of HCa-P/L₆and HCa-P were prolonged （P <0.01）, and curcumin had more inhibiton on HCa-P/L₆than on HCa-P （t test P <0.05）. Treated bycurcumin at15μmol/L for48h, the cell cycle was redistributed with HCa-P/L₆beingarressted in S phase while HCa-P in S and G2/M phases; the content level of PE and PIare reduced in HCa-P/L₆and HCa-P, and the decreased content level in HCa-P/L₆wasmore obvious than that in HCa-P; the content level of PE and PI was reduced in fourcloning cell strains（G2, B8, E10, and H5）（t test, P<0.05）, especially in G2withsignificant less phospholipid content level than in other three strains, while nosignificant change of the content level of PC was found; all the content level of PEand PI was reduced in four transplanted tumor cell strains （t test P<0.01） respectivelyfrom G2, B8, E10, and H5being subcutaneously inoculated to the right foot pad ofevery mouse615（t test P<0.01）, especially in G2transplanted tumor cell strain withsignificant less phospholipid content level than in other three strains, while nosignificant change of the content level of PC was found.7. At maximum uncytotoxicity dose, the effectiveness of elemene on cellmembrane phospholipids is more significant than that on curcumin. After four cloningcell strains were respectively subcutaneously inoculated to the right foot pad of everymouse615, the sensitivity of drugs to the transplanted tumor cell strains was increasedrespectively than that to hteir homologous parental cloning cell strains, and theeffectiveness of drugs on membrane phospolipids of cancer cells was also increased.Conclusion:1. In two murine ascites hepatocarcinoma cell lines with different lymphaticpotential （high HCa-P/L₆and low HCa-P）, the level of PE and PI in HCa-P/L₆washigher than that in HCa-P. The content level of PE and PI in cloning cell strain G2withlow lymphatic potential was higher than that in non metastatic strain H5. After fourcloning cell strains separated from HCa-P were respectively subcutaneously inoculatedto the right foot pad of every mouse615, the content level of PE and PI in G2transplanted tumor cells is higher than that in non metastatic strain H5（t test，P<0.05）.This phenomenon may demonstrate that higher PE and PI may be related to metastaticpotential of cancer cells.2. Elemene could inhibit the proliferation of HCa-P/L₆and HCa-P indose-dependent manner and IC50of HCa-P/L₆was lower than that of HCa-P.10μg/mLElemene for48h could prolong population doubling times of HCa-P/L₆and HCa-P,redistributed cell cycle progressions of HCa-P/L₆and HCa-P with being blocked in Sphase and the blocking effectiveness in HCa-P/L₆stronger than that in HCa-P. HCa-P/L₆was validated more sensitive to elemene than HCa-P was.3. Curcumin could inhibit the proliferation of HCa-P/L₆and HCa-P indose-dependent manner and IC50of HCa-P/L₆was lower than that of HCa-P.15μmol/lcurcumin for48h could prolong population doubling times of HCa-P/L₆and HCa-P,redistributed cell cycle progressions of HCa-P/L₆with being blocked in S phase, butthat of HCa-P is blocked in S and G2/M phases and the blocking effectiveness inHCa-P/L₆stronger than that in HCa-P. HCa-P/L₆was validated more sensitive tocurcumin than HCa-P was. The cell cycle progressions redistribution induced bycurcumin might be potential mechanisms underlying the effectiveness of elemene ontumor growth and metastasis.4. At maximum uncytotoxicity dose, the decreasing of PE and PI reduced byelemene on cell membrane phospholipids is more significant than that by curcumin. Theinhibition of elemene or curcumin on the proliferation and metastasis of cancer cellsmaybe associated with that elemene or curcumin can rise to the reduction of PE and PIof membrane phospolipid content level in cancer cells.