| Background and ObjectivesAcute leukemia is the most common type of cancer in children.Despite advances in the treatment of this disease,approximately 20% of patients die from drug resistance or relapse[1].In the past decade,molecular research has found that there are more and more genetic markers associated with acute myeloid leukemia(AML)and acute lymphocytic leukemia(ALL).These markers can enhance risk stratification,and changing treatment strategies can improve patient prognosis.ALL is often accompanied by gene mutations and chromosomal translocations related to lymphocyte development and cell cycle regulation[2].There is increasing evidence that epigenetic changes,including differences in DNA methylation,can affect susceptibility and prognosis of leukemia[3].Histone methylation disorders may affect important cellular processes related to tumorigenesis,such as proliferation and differentiation,leading to tumors[4].TET2 is a member of the 10-11 translocation gene family and catalyzes the conversion of 5-methylcytosine to 5-hydroxymethylcytosine in DNA.SET and MYND domain protein 2(SMYD2)is a methyltransferase that can regulate the activity of several proteins including histones[5].Abnormal expression of both genes has been reported to be associated with hematological tumors[6,7].Our previous experiments have confirmed that the decrease in TET2 gene m RNA level and the increase in SMYD2 gene m RNA level are poor prognostic factors for children with ALL.Compared with ALL,children with AML have a relatively poor prognosis and a lower cure rate.With the optimization of chemotherapy schemes and the development of hematopoietic stem cell transplantation and other technologies,the cure rate of children’s AML has been significantly increased,but some children with AML remain uncured due to incomplete eradication of leukemia stem cells(LSCs).LSCs are responsible for relapse and drug resistance in children with acute myeloid leukemia(AML).For relapsed or refractory AML,conventional chemotherapy has been ineffective,and new treatments are urgently needed.In recent years,chimeric antigen receptor(CAR)therapy targeting CD19 have achieved remarkable results in the application of B-lineage tumors,which also brings new hope for the cure of children with relapsed or refractory AML.However,few targets have been approved for clinical treatment of AML.At present,only CD123-CAR has been approved by the US FDA for clinical treatment of relapsed and refractory AML,while CARs for other targets are still undergoing treatment trials.AML-CAR targets under study include CD33,CD123,FLT3,Lewis-Y,and CLL-1,but the results are mostly poor,and new targets are urgently needed.3A4 is a monoclonal antibody previously developed in our laboratory.It is an Ig Gκ antibody that recognizes human CD45 subclass antigen.Earlier studies found that the antibody only recognizes myeloid leukemia and its stem cells,B-type tumor cells,normal B cells,na?ve T cells,and some monocytes,and it has no effect on normal memory T cells,neutrophils,red blood cells,platelets and most primitive hematopoietic stem cells.3A4 is a good target for AML.The success rate of CAR is not only related to target modulation,but also related to the degree of humanization of single-source antibodies that recognize targets.Most CARs are of mouse origin.Mouse-derived CAR is immunogenic to the human body and can produce anti-sc Fv antibodies.CAR is neutralized and the function of CAR is greatly impaired.Therefore,in order to solve this problem,we propose to use humanized CAR,which can significantly reduce the production of anti-sc Fv antibodies,thereby ensuring the killing activity of CAR.The3A4 antibody has been humanized and named Hu3A4,with a humanization level of 93%(results from previous research by our research group).Hu3A4 can specifically bind myeloid and B-lineage LSCs,and has a strong targeted killing effect on myeloid KG1 a cells.The construction of Hu3A4-CAR with Hu3A4 monoclonal antibody,targeted killing of leukemia stem cells and treatment of refractory AML has important scientific significance and clinical application prospects.This study consists of the following four chapters:(1)the construction of lentivirus expression vector for the Hu3A4-CAR targeting AML-LSC;(2)the study of Hu3A4-CAR lentiviral packaging,concentration,titer determination and the study on the successful expression of Hu3A4-CAR on T cells and expansion in vitro;(3)the study of Hu3A4-CAR-T targeted killing effect on leukemia cells in vitro and comparison with Mu3A4-CAR-T targeted killing effect and its killing mechanism;(4)the study of anti-tumor effect on AML mouse models by the Hu3A4-CAR-T cells and comparison with Mu3A4-CAR-T targeted anti-tumor effect in vivo.Methods1 Construction of lentivirus expression vector for the Hu3A4-CAR targeting AML-LSC1.1 Construction of second generation Hu3A4-CAR eukaryotic expression vector pc DNA3.1 / Hu3A4-4-1BB-3ζ and lentiviral expression vector p Lenti /Hu3A4-4-1BB-3ζ: After searching and comparing the literature,we determined the gene structure of the second-generation Hu3A4-CAR.The gene sequences of the extracellular region,transmembrane region and intracellular region of Hu3A4-CAR have been identified;The eukaryotic expression vector of pc DNA3.1/Hu3A4-4-1BB-3ζand lentiviral expression vector of p Lenti / Hu3A4-4-1BB-3ζ were constructed using molecular methods.1.2 Activity identification of the eukaryotic and lentiviral expression vectors for the Hu3A4-CAR: We transfected CHO cells with the two types of expression vectors to verify if Hu3A4-CAR gene was expressed by flow cytometry(FCM)analysis.The difference in CAR expression efficiency between the two types of vectors was compared,and a better expression vector was selected for the following study.2 The successful expression of Hu3A4-CAR on T cells2.1 Lentivirus packaging and titer determination: Human embryonic kidney 293 T cells were used as packaging cells.After a large number of extractions of packaging plasmids Rev,Gag-Pol,Vsvg and expression plasmid p Lenti / Hu3A4-4-1BB-3ζ,we packaged 293 T cells with packaging plasmids and expression plasmid through liposomes and collected supernatant carrying Hu3A4-CAR lentivirus.The lentivirus titer was determined.2.2 Study on the successful expression of Hu3A4-CAR and Mu3A4-CAR on T cells and expansion in vitro: We used Ficoll density gradient centrifugation to isolate T cells from human peripheral blood.T cells were co-cultured with anti-human CD3 / CD28 magnetic beads and IL-2,and then be transduced with a lentivirus containing CAR genes.Western blot,RT-PCR,and FCM analysis were used to identify whether Hu3A4-CAR and Mu3A4-CAR gene were expressed on T cells.We calculated the amplification times of Hu3A4-CAR-T cells and Mu3A4-CAR-T cells in vitro.3 Study of Hu3A4-CAR-T targeted killing effect on leukemia cells in vitro and comparison with Mu3A4-CAR-T targeted killing effect: Microscope and FCM analysis were used to verify whether Hu3A4-CAR-T bind to 3A4 + cells and compared the changes of Hu3A4-CAR-T cells bound to target cells under different ratios of effective cells to target cells.Calcein-AM release assays were performed to evaluate the killing efficacy of Hu3A4-CAR-T on KG1 a cells of leukemic cell line as well as leukemia cells of newly diagnosed AML patients,and to compare with the difference from Mu3A4-CAR-T targeted killing function.The level of cytokine release was detected by Cytometric Bead Array(CBA)on FCM;The m RNA expression level of granzyme B(GZM-B)and perforin(PFN)genes was analyzed by Real-time PCR method.We used Ficoll density gradient centrifugation to separate bone marrow mononuclear cells(BMMNCs).BMMNCs,Hu3A4-CAR-T cells and Mu3A4-CAR-T cells were extensively washed,counted and resuspended in IMDM medium.5×10~3BMMNCs were respectively combined with Hu3A4-CAR-T cells and Mu3A4-CAR-T cells at an E:T(effector to target)ratio of 10:1 for 20 h,and then the entire cell suspension was transferred to Metho Cult SF H4436,a methyl-cellulose based semi-solid medium supplemented with recombinant cytokines,and plated in duplicate.Fourteen days later,CFU colonies were classified and counted by our morphological observation under a light microscope.4 Study of anti-tumor effect on AML mouse model by the Hu3A4-CAR-T cells and comparison with Mu3A4-CAR-T targeted anti-tumor effect in vivo: NOD / SCID mice were set into 5 groups of 6 mice each,which were:(1)the model group inoculated with KG1 a cells;(2)the blank control group with PBS;(3)the control group treated with T cells;(4)the group treated with Mu3A4-CAR-T;(5)experimental group treated with Hu3A4-CAR-T.Pretreatment with cyclophosphamide at a concentration of10mg/ml,each mouse was injected intraperitoneally with 200 ul for 2 days.Except the PBS group,the other 4 groups of mice were each inoculated with 2x10~6 KG1 a cells to establish an animal model of AML.On the first and second day after injection of KG1 a cells,T,Hu3A4-CAR-T or Mu3A4-CAR-T cells were then injected through tail veins separately.We monitored the weight changes of all mice every day,observed the mental state,and checked for symptoms such as decreased activity,slow movement,and arch back elevation;Peripheral blood was collected from the periorbital venous plexus of all mice every 1-3 weeks,and CAR-T cells engraftment and persistence,as well as cytokines were determined by FCM.The dates of the onset and death of the mice were recorded,and survival curves were plotted to compare the differences using the log-rank test.Results of pathology and immunohistochemistry for mice organs from every groups were verified and compared.Results1 Construction of lentivirus expression vector for the Hu3A4-CAR targeting AML-LSC1.1 Construction of second generation Hu3A4-CAR eukaryotic expression vector pc DNA3.1 / Hu3A4-4-1BB-3ζ and lentiviral expression vector p Lenti /Hu3A4-4-1BB-3ζ: The gene structure of Hu3A4-CAR was designed as a second-generation CAR.The construction was composed of the Hu3A4-sc Fv,as the targeted binding moiety,linked to a CD8 hinge region and transmembrane region,followed by a CD3ζ signal region in tandem with the CD137 intracellular signaling motif.We have constructed eukaryotic expression vector of pc DNA3.1 /Hu3A4-4-1BB-3ζ and the lentiviral expression vector of p Lenti / Hu3A4-4-1BB-3ζsuccessfully.Sequencing results showed that the vectors were successfully constructed and could be used for the next experimental research.1.2 Activity identification of the eukaryotic expression vector and lentiviral expression vector for the Hu3A4-CAR: The result of FCM showed the constructed vectors could express the Hu3A4-CAR fusion protein in the CHO cells.The expression efficiency of lentiviral vector was higher than that of eukaryotic expression vector(36%vs.24.3%,P = 0.046).2 The successful expression of Hu3A4-CAR and Mu3A4-CAR on T cells2.1 Lentivirus packaging and titer determination: 293 T cells successfully packaged the lentiviruses containing Mu3A4-CAR gene and Hu3A4-CAR gene.The titer of non-concentrated lentiviral supernatant for Mu3A4-CAR and Hu3A4-CAR was more than 5 x 105 by q RT-PCR assay.2.2 Study on the successful expression of Hu3A4-CAR and Mu3A4-CAR on T cells and expansion in vitro: The expression of Hu3A4-CAR-T and Mu3A4-CAR-T were identified by RT-PCR,Western blot and FCM analysis.All three methods proved that Hu3A4-CAR and Mu3A4-CAR genes could be successfully introduced and expressed on the surface of T cells.The infection efficiency of lentiviral transfection system in human activated primary T cells was 42%-62%.Under the condition of CD3 / CD28 magnetic beads activating T cells,RPMI 1640 medium plus IL-2,Hu3A4-CAR-T and Mu3A4-CAR-T cells could expand 110-200 times in vitro.There was no statistical difference in amplification times between the two groups(average amplification times were 130 and 135 times,respectively,P> 0.05).3 Study of Hu3A4-CAR-T targeted killing effect on leukemia cells in vitro and comparison with Mu3A4-CAR-T targeted killing effect: Hu3A4-CAR-T cells could recognize and bind to 3A4+ target cells including KG1 a and Raji cells.As the ratio of effector cells to target cells increased,the binding ratio of Hu3A4-CAR-T to target cells increased.Calcein-AM is a fluorescent dye with relatively stable fluorescence in living cells,but not obvious fluorescence in dead cells.Target cells labeled with Calcein-AM were incubated with Hu3A4-CAR-T,Mu3A4-CAR-T,and normal T cells for 24 hours.FCM analysis showed target cells co-incubated with T cells could be detected of green fluorescence in the FITC channel after 24 hours.For the target cells co-incubated with T cells,the proportion of target cells positively bound by Calcein-AM was(95.3 + 0.3)%.While for the target cells co-incubated with Hu3A4-CAR-T and Mu3A4-CAR-T,the proportion of target cells bound by Calcein-AM was(17.2 + 0.5)% and(12.8 + 1.4)%,respectively(P = 0.079).And most of them were targeted by CAR-T and no green fluorescence could be detected,indicating that the target cells bound by CAR-T were killed.Hu3A4-CAR-T and Mu3A4-CAR-T cells did not kill 3A4-Nalm-6 cells.In addition,as the ratios of effective cells to target cells increased,cytolytic activity improved accordingly.Hu3A4-CAR-T and Mu3A4-CAR-T cells could cause increased IFN-γ secretion during targeted killing.When the target effect ratio was 10: 1,IFN-γlevels in Hu3A4-CAR-T and Mu3A4-CAR-T groups were 69.4 pg / ml and 51.8 pg / ml,respectively.There was no significant difference between the two groups(P = 0.739),but they both were significantly different from the T-cell killing control group(P <0.05).The Real-time PCR method was used to detect increased expression of granzyme B m RNA,and the Ct values were 0.0929 and 0.0925,respectively.There was no significant difference(P = 0.096),but they both were significantly different from the T-cell killing control group(P <0.05).Colony Forming Unit(CFU)Assays were performed after 5 × 10~3 BMMNCs being co-incubated with Hu3A4-CAR-T and Mu3A4-CAR-T for 20 hours respectively.After 14 days,the results showed that(26.3 +0.88)/5x10~3 colonies were formed in the non-CAR killed control group,while there were(27 + 1.15)/5x10~3 and(23 + 1.15)/5x10~3 colonies formed in BMMNCs incubated with Mu3A4-CAR-T and Hu3A4-CAR-T,respectively.Compared with the control group,the co-incubation group also obtained similar numbers of CFU colonies(P> 0.05).The two types of 3A4-CAR-T cells(humanized,mouse)had no significant effect on bone marrow hematopoietic stem / progenitor cells.4 Study of anti-tumor effect on AML mouse model by the Hu3A4-CAR-T cells and comparison with Mu3A4-CAR-T targeted anti-tumor effect in vivo: There were 5groups including the model group,PBS blank control group,T cell treatment control group,Hu3A4-CAR-T cell treatment group and Mu3A4-CAR-T cell treatment group,with 6 mice in each group.Except for the PBS group,each of the remaining 4 groups of mice was successfully inoculated with 2 × 106 KG1 a cells through the tail vein.On Day1 and Day2 after KG1 a cells being inoculated,the mice of the T cell treatment group were successfully infused with T cells through the tail vein,while the mice of the Hu3A4-CAR-T cell treatment group and Mu3A4-CAR-T cell treatment group were infused with Hu3A4-CAR-T cells and Mu3A4-CAR-T cells,respectively.In the Hu3A4-CAR-T treatment group,one mouse died half an hour after injection of Hu3A4-CAR-T cells on Day 2,and the remaining mice continued to be in the experiment.FCM was used to detect the retention status of two types of 3A4-CAR-T cells in mice.The results showed that Mu3A4-CAR-T cells could be detected in the peripheral blood of all mice in the Mu3A4-CAR-T cell treatment group on Day11(0.005%-0.052%),and Hu3A4-CAR-T cells could also be detected in 5 mice of Hu3A4-CAR-T treatment group(0.002%-0.127%).Unfortunately,on Day 25,the levels of Mu3A4-CAR-T cells in the peripheral blood of all mice in the Mu3A4-CAR-T cell treatment group were <0.001%,while Hu3A4-CAR-T cells could be detected in only 2 mice of Hu3A4-CAR-T cell treatment group(0.003% and 0.013%,respectively).And Hu3A4-CAR-T cells were no more detected in both mice on Day 41.By the end of the observation period(140 days),all the NOD / SCID mice in the blank control group survived,but all the mice in the T cell treatment control group and in the model group developed AML and died.The survival time of the two groups of mice was similar with a median time of 59(52-140)days for T cells treated group and 54(44-94)days for model group,respectively.Three mice treated by Mu3A4-CAR-T did not show any symptom of AML until the end of the observation period,while the other three mice in the same group developed AML and died with a median survival time of 98(68-140)days.Three mice treated by Hu3A4-CAR-T survived to the end of the observation period but the other three mice died with a median survival time of 89(2-140)days.Survival analysis on Day140 showed the difference in survival time of the five groups was statistically significant(P <0.05).The pathological results showed that the meningeal tissues of the diseased mice were infiltrated with tumor cells,and immunohistochemical staining results also confirmed that the tumor cells infiltrated the meninges,which may be the cause of hind limb paralysis in mice.No tumor cell infiltration was found in heart,lung,liver,spleen,kidney,intestine,pancreas and other tissues.Conclusions1.Eukaryotic expression vector and lentivirus expression vector for humanized3A4-CAR were successfully constructed by molecular cloning.Both types of expression vectors could be transfected to CHO cells successfully.The expression efficiency of lentiviral vector was higher than that of eukaryotic expression vector.2.Lentiviral particles carrying Hu3A4-CAR and Mu3A4-CAR genes were obtained.Titer of the two types of lentivirus supernatant was successfully determined.The Hu3A4-CAR and Mu3A4-CAR genes constructed in the lentiviral vectors could be transfected into T cells and identified by RT-PCR,western blot and FCM analysis.3.CAR-T cells could be cultured successfully in vitro.Hu3A4-CAR-T and Mu3A4-CAR-T cells could be expanded for 110-200 times,activated with anti-human CD3/28 Dynabeads and co-cultured with IL-2.4.Hu3A4-CAR-T could bind target cells which express 3A4 antigens.As the ratio of effector cells to target cells increased,the binding ratio of Hu3A4-CAR-T to target cells increased.Hu3A4-CAR-T as well as Mu3A4-CAR-T could specifically kill 3A4+ leukemia cell lines and leukemia cells of AML patients,but not 3A4-leukemia cells.IFN-γ level increased during the process of Hu3A4-CAR-T killing target cells.Hu3A4-CAR-T cells killed leukemia cells by the granulocytosis pathway.The two types of 3A4-CAR-T cells(humanized,mouse)had no significant effect on bone marrow hematopoietic stem / progenitor cells.5.Hu3A4-CAR-T as well as Mu3A4-CAR-T could be engrafted into the NOD/SCID mice and persist successfully in vivo.Hu3A4-CAR-T as well as Mu3A4-CAR-T cells could inhibit the survival of KG1 a cells and avoid the occurrence of AML in NOD/SCID mice. |
PDF Full Text Request