The Open Access Journal of Science and TechnologyVol. 3 (2015), Article ID 101113, 9 pagesdoi:10.11131/2015/101113

AgiAlPublishing House

http://www.agialpress.com/

Research Article

In Vitro Generation of Cytotoxic CD4 LymphocyteResponse against Autologous AcuteMyeloid Leukemia

Kok Chong Yap1, Liam Pock Ho1,2, Pui San Yit1, and Yeow Tee Goh1

1Department of Hematology, SGH, Singapore2Department of Pathology, SGH, Singapore

Corresponding Author: Bernard Yap; email: bernard_y@hotmail.com

Received 21 August 2014; Accepted 28 September 2014

Academic Editors: Luiz Euribel Prestes Carneiro, Ramazan Güneşaçar, and Giuseppe Murdaca

Copyright © 2015 Yap Bernard et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract. Relapsed and refractory disease shortens the survival of acute myeloid leukemia (AML) patients. Therefore, additionaltherapy apart from conventional chemotherapy and stem cell transplantation is urgently needed, especially in eradicating residualleukemic stem cells. The identification of leukemia associated antigens, and the observation that administration of allogeneic Tcells may mediate graft versus leukemia effect paved the way to develop various immunotherapy strategies. Effective provision ofboth tumor antigen and co-stimulation are essential for strategies aimed at enrolling cytotoxic T- lymphocytes to eradicate leukemiccells. In this study, we attempted to up-regulate co- stimulators on leukemic cells (termed “dendritisation” to mimic dendritic cellsin providing strong co-stimulation) and couple with inherent leukemic antigens to generate specific antileukemic T-lymphocytesagainst autologous blast cells. Our study shows evidence of distinct cytotoxicity of CD4 positive T lymphocytes against autologousblast cells. Dendritised AML blasts were able to function with antigen presenting capability, while still phenotypically existing asblast cells. This was also achievable for a wide range of subtypes of AML, making it a potential immunotherapy against AML.

Keywords: AML, immunotherapy, cytotoxic, CD 4 T-lymphocytes

1. Introduction

Acute myeloid leukemia (AML) is a myeloid neoplasmcharacterised by the expansion of malignant non-lymphoidhematopoietic progenitor cells resulting in failure of normalhematopoiesis [1]. It is one of the most common acuteleukemia affecting adults, and incidence rates are higherwith increasing age [2, 3]. It is apparent that after standardtherapy, the survival rates of patients are reduced for relapsed

or refractory cases. It has been documented that 70% ofsuccessfully treated patients by chemotherapy relapse soonafter receiving treatment [4]. This high relapse rate occursdespite recent advances in therapeutic approach such asintensive chemotherapy, targeted mechanism therapy aswell as stem cell transplantation. Therefore, in order toprolong remission and survival rates of AML patients,there is a continuous need for the development of noveltherapy.

2 The Open Access Journal of Science and Technology

Relapsed and refractory AML has often been associatedwith drug resistant residual leukemic stem cells which arenot eradicated completely by intensive chemotherapy. Evenamong patients who manage to achieve complete remissionafter therapy, many may experience relapse [5–8]. Relapseof disease in patients who have attained complete remissionmay be attributed to the presence of residual leukemic stemcells after treatment. Hence, we believe that the eradiation ofleukemic stem cells is the most effective way to improve thesurvival rates of AML patients.

There is a higher rate of relapse after syngeneic bonemarrow transplantion than allogeneic bone marrow trans-plantation [9]. The reasons attributing to the better efficacy ofallogeneic transplantation in AML treatment may be partlydue to the allogeneic immune mechanism, also known asthe graft-versus-leukemia effect [10]. In a separate study,it was also found that T cell depletion of the graft wasassociated with a higher rate of relapse when comparedwith donor lymphocyte infusion (DLI). DLI was able tore- induce a stable hematological remission in one-fourthof AML patients with relapse [11]. Despite much promisesin harnessing DLI against relapsed AML, the increase inmalignant cells in aggressive relapsing AML may at timesoutstrip the clinical effect of the DLI, thereby resulting ina failure to sustain an anti-leukemic response [12]. Anothershortcoming of DLI is that patients often develop graft-versus- host disease (GVHD) as a result of the infusedallogeneic T cells. Therefore, it would be ideal if the infusedcytotoxic T cells have more potent and specific cytotoxiceffect against the leukemic cells (Graft versus Leukemiaeffect) but negligible or no GVHD effect.

The identification of leukemia associated antigens (LAAs)and the observation of anti- leukemic effect of cytotoxic allo-geneic T cells in AML patients have spurred the developmentof various immunotherapy strategies against leukemia. Forexample, different groups have generated T cells specificfor LAAs, by isolating LAAs from leukemic cells of AMLpatients and subsequently loading them into antigen present-ing cells such as the dendritic cells [13–15]. However, thismethod has its disadvantages especially in the isolation ofspecific LAAs that varies among individuals. Therefore, thistechnique requires the identification of specific LAAs thatare highly presented in individual patients to generate T cellswith anti-leukemic activity.We believe that the success of thecytotoxic T cells hinges on the correct and patient-specificLAAs being presented by the APC to T cells, and hence thesuccess rate of this technique may be modest.

AML is characterised by the accumulation of clonalmyeloid progenitor cells [16]. The graft-versus-leukemiaeffect after allogeneic stem cell transplantation exhibits thepotential of immunotherapy against residual leukemic cells[17]. To add on, there have been previous studies thatdetected specific T lymphocytes that are reactive againsttumor–associated antigens such as Wilm’s tumor gene prod-uct WT1 and proteinase 3 [18, 19]. Hence, it would be

of interest and clinical benefit should there be a suitableadjuvant immunotherapy for treating AML by harnessingcytotoxic T lymphocytes from patients to eradicate autolo-gous leukemic blast cells [16, 20].

One of the major reasons for non-efficient recognition ofAML blasts by T lymphocytes in vivo is that 70% of the blastsdo not express costimulatory molecules which are necessaryfor efficient T cell activation [21]. Dendritic cells are themost potent antigen presenting cells present in the humanbody capable of expressing costimulatory molecules. Hence,these dendritic cells are vital for the initiation of primaryspecific immune responses and are important means for theinduction of anticancer immunity [22]. Therefore, in thisstudy, we exploited the full capacity of dendritic cells topresent leukemic antigens to T cells, thereby converting thelatter to cytotoxic T cells. Addressing the issue of isolation ofcorrect and specific LAAs, our technique utilizes a protocolthat enhances dendritic cells’ antigen presenting capabilitiesto leukemic blast cells, but retaining the phenotype of theleukemic blast cells - a processwe termed as “dendritisation”.By “dendritising” the leukemic blast cells, these malignantcells are able to co-express inherent leukemic antigens andupregulate co-stimulators to generate specific anti-leukemiceffect. Unlike another technique that is published which fullyconverts leukemic blast cells into mature dendritic cells [21],our approach ensures that the leukemic blast cells retainits LAAs that would be presented by its acquired antigenpresenting capabilities of the dendritic cells. In addition,since the leukemic blast cells are derived from the patient,the generation of the autologous T cells would thereforereduce the possibility of GVHD. Another advantage fordendritisation is that no effort is required to determine anyknownLAAs since the dendritised leukemic blast cells wouldpresent all known and unknown LAAs. In this study, weattempted to dendritise leukemic blast cells and subsequentlygenerate autologous cytotoxic T cells in vitro from differentAML subtypes.

2. Material andMethods

2.1. Patient Samples. Bone marrow and peripheral bloodmononuclear cells were collected from AML patients afterwritten informed consent was obtained. Patient consentedby signing on the informed consent. This informed consentform was approved by Singhealth Centralised IRB with theapprocal number: 2008/060/B, for the purpose of laboratoryresearch/study that was performed using the collected bonemarrow and peripheral blood mononuclear cells. Bone mar-row and peripheral blood mononuclear cells were isolatedby density- gradient centrifugation (Ficoll-Paque Plus, GEHealthcare, Sweden) and subsequently washed with Hank’sBalanced Salt Solution (HBSS). Mononuclear cells (MNC)were suspended in complete medium (CM): Sterile water(B.Braun, Bethlehem; USA), powder RPMI 1640 (SigmaAldrich, St. Louis; USA), 10% fetal bovine serum (Research

AgiAlPublishing House | http://www.agialpress.com/

The Open Access Journal of Science and Technology 3

Instruments/ Hyclone, South America), 1% L-Glutamine(PAA, Australia), 1% Pen- Strap-Neomycin solution (PAA,Australia), Herpes (Sigma Aldrich, St. Louis; USA), sodiumbicarbonate (B.Braun, Melsungen; Germany), and cryopre-served for storage.

2.2. Generation of “dendritised” Blast Cells - “dendriti-sation” Phase. Cryopreserved MNCs were thawed in 37∘Cwater bath and washed in CM twice. Washed MNCs pelletswere subsequently re-suspended in CM. The suspensionof MNCs in CM were supplemented with a concortion ofhuman recombinant cytokines [40ng of IL-4 (Peprotech,NJ; USA) per µl of CM, 20ng of GM-CSF (Peprotech,NJ; USA) per µl of CM and 1ng TNF-𝛼 (Peprotech, NJ;USA) per µl of CM and incubated at 37 ∘C and 5% CO2.The blast cells (present in the suspension of MNCs) weredendritised for seven days with one third of the CM andcytokines refreshed on Day 4 during the dendritisation phase.On Day 7 of the dendritisation phase, a complete washof the blast cells was performed via centrifugation andsubsequently re-suspended in fresh CM in preparation forlymphocyte expansion (lymphoexpansion) phase. See Table1 for illustration of the dendritisation protocol.

2.3. Expansion of lymphocytes - “Lymphoexpansion” Phase.Inherent autologous T-lymphocytes were cultured andexpanded in the presence of dendritised blast cells in CM.100 units of INF-𝛾 (Peprotech, NJ; USA) was added per µlof CM of culture on Day 1 (lymphoexpansion phase)/Day7 (dendritisation phase). Also, 1µl of IL-2 (300 units/ml)was added to every 1ml of CM of culture on Day 2(lymphoexpansion phase). No muromonab- CD3 (OKT3)was added into the medium. A separate vial of cryopreservedMNCs were thawed in 37 ∘Cwater bath, washed in CM twiceand subsequently resuspended in CM with IL-2 and INF-𝛾supplements; this served as the control. One third of the CMand cytokines (IL-2 and INF-𝛾) were refreshed every fourdays for both non-control and control flasks. See Table 1 forillustration of the “lymphoexpansion” protocol.

2.4. Immunophenotyping – “Dendritisation” Phase. Cul-tured cells were harvested at several time points (Days 4–10),washed and stained with mouse anti-humanmonoclonal anti-bodies purchased fromBeckmanCoulter, Ancell Corporationand LifeSpan Biosciences [Beckman Coulter, MarseilleCedex; France: PC5-conjugated anti- CD45, PE-conjugatedanti-CD80 (IgG1, clone MAB104), PE- conjugated anti-CD86 (IgG2b𝜅, clone HA5.2B7), FITC-conjugated anti-CD34 (IgG1, clone 581), FITC-conjugated anti-CD33 (IgG1,clone D3HL60.251) and FITC-conjugated anti-CD13 (IgG1,clone SJ1D1). Ancell Corporation, Bayport MN; USA: PE-conjugated anti- MHC class II (HLA-DP, DQ and DR,IgG1, clone TDR31.1). Lifespan Biosciences, USA: FITC-conjugated anti-CD117 (IgG1)].

2.5. Immunophenotyping- “Lymphoexpansion” Phase. Cul-tured cells were harvested after seven days, washed andstained with mouse anti- human monoclonal antibod-ies purchased from Beckman Coulter, Marseille Cedex;France: PC5-conjugated anti-CD45 (IgG1, clone J33),PE-conjugated anti-CD4 (IgG1, clone 13B8.2), FITC-conjugated anti-CD8 (IgG1, clone B9.11) and 7AAD anti-body (BD Bioscience, USA; Material number 559925).

2.6. Cytotoxicity Assay. Cultured cells were first stained withBeckman Coulter FITC-conjugated-CD8 antibody, blastmarkers and myeloid markers (FITC-conjugated anti-CD33,FITC-conjugated anti- CD34, FITC-conjugated anti-CD117,FITC-conjugated anti- CD13). CD4 T lymphocytes wereobtained via negative selection using anti-FITC conjugatedbeads (Miltenyi Biotec, Bergisch Gladback; Germany;130-048-701). CD4 T lymphocytes from both dendritised andcontrol flasks (Effector Cells) were washed and coculturedwith freshly thawed autologous blast cells (Target Cells)for 4 h at various effector : target ratio (Effectors: Targetratios of 0:1, 2.5:1, 1:1, 10:1,40:1). After 4 h, cells werewashed and stained with mouse anti-human PC5- conjugatedanti-CD45(Beckman Coulter, Marseille Cedex; France),respective blast markers, 7AAD (BD Bioscience, USA;material number 559925) and annexin V(BD Bioscience,USA; material number 56421) for flow analysis.

3. Results

3.1. Patient characteristics. Patients were aged between 18and 73 years, and had subtypes (M1, M2, M4) of AMLaccording to the French-American-British (FAB) classifica-tion. Patients’ demography and main clinical parameters aresummarized in Table 2.

3.2. “Dendritisation” Phase - Generation of “Dendritised”Leukemic Blast Cells. There was distinct up-regulation ofco-stimulators (CD80, CD86) and MHC II in the dendritisedblast cells from Day 1 onwards. The strength of the expres-sion of co-stimulators and MHC II increased to a maximumDelta Mean Fluorescent Index (dMFI) around Day 7 afterthe start of “dendritisation” (See Figure 1). In addition, cellsurface projections akin to dendritic cells were observed inthe dendritised blast cells of all the seven patients in thisstudy.

3.3. “Lymphoexpansion” Phase- Generation of autologouscytotoxic T cells. After “dendritisation” of leukemic blastcells, the dendritised leukemic blast cells were co- culturedwith the inherent T lymphocytes that were present, in thepresence of IL-2 and INF-𝛾 (i.e., lymphoexpansion phase).There was a distinctive increase in the proliferation of Tlymphocytes (both CD4 and CD8 positive T lymphocytes).However, as the number of days of lymphoexpansion

AgiAlPublishing House | http://www.agialpress.com/

4 The Open Access Journal of Science and Technology

Table 1: Dendritisation Protocol.

Cytokine added Day 0 Day 4 Day 7 Day 11 Day 14 Day18 Day 21 Day 25 Day 28*IL 4 + + − − − − − − −GM-CSF + + − − − − − − −TNF-𝜶 + + − − − − − − −IL- 2 − − + + + + + + +𝜸-INF − − + − − − − − −Complete Medium + Wash Total Wash Wash Wash Wash Wash Wash Wash*Lymphoexpansion phase may vary for different AML patients

Table 2: Patients’ Characteristics.

Patient no. Age (Yrs) Sex AML subtype Relapse or Newly Diagnosed % Blast1 27 F M1 Newly Diagnosed 912 18 M M4 Newly Diagnosed 263 73 F M2 Newly Diagnosed 354 43 M M1 Newly Diagnosed 805 32 F M2 Newly Diagnosed 616 33 M M1 Newly Diagnosed 817 44 F M1 Newly Diagnosed 81

increased, there was a predominance of CD4 positive Tlymphocytes than CD8 positive T lymphocytes (See Figure2(a)). Expanded CD4 lymphocytes, both that were co-cultured with dendritised blast cells and controls, exhibitedeffector memory phenotype (TEM) (CD45RA−, CD45RO+,CD62L−, CCR7−, CD127−, CD25+, and CD27±, CD28±)(See Figure 2(b)). However, despite T lymphocytes from bothcontrol and dendritised blast cells having the same TEMphenotype, there was a distinct difference in the expressionof intra-cellular cytotoxic granules. It was observed thatthere was a significantly higher IFN-𝛾 expression in the Tlymphocytes that was co- cultured with dendritised blast cellsthan in the T lymphocytes from the controls. There was alsoa slight increase in granzyme A and granulysin in the Tlymphocytes that was co- cultured with dendritised blast cellsthan in the T lymphocytes from the controls (See Figure 2(c)).

3.4. Functional Test: Cytotoxicity Assay. Figure 3 showsthe comparison of the anti-leukemic cytotoxicity effectbetween CD4 positive T lymphocytes that were primed withdendritised blast cells and those that were primed with onlyIL-2 in the control flask for Patient 2. The percentage of liveleukemic blast cells (represented by 7AAD and Annexin Vdouble negative cells) showed a distinctive decrease from83% to 55% (Effectors:Target ratio from 0:1 to 40:1) whenco- cultured with CD4 T lymphocytes that were primed withdendritised blast cells (see Figure 3(a)). However, there wasno change in the percentage live leukemic blast cells that wereco-cultured with the CD4 lymphocytes from the control flask(see Figure 3(a)). Figure 3(b) shows the cumulative averagepercentage of dead blast cells that were co-cultured with CD4T lymphocytes and primed with dendritised blast cells and

controls. Cytotoxic- capability of CD4 T lymphocytes thatwere primed with dendritised blast cells were higher thatCD4 T lymphocytes from the controls. CD4 T lymphocytesthat were primed with dendritised blast cells exhibitedincreasing cytotoxicity effect against autologous blast cellswith increasing amount of CD4 T lymphocytes ( Percentageof blast cells killed increased from 16% (0:1) to 21.7%(1:1) to 25.2% (2.5:1) to 40% (10:1) to 45%(40:1), wherebyratio in brackets represents ratio of effector cells : blastCells respectively. On the contrary, CD4 T lymphocytes didnot exhibit cytotoxicity effect towards autologous blast cells(blast cells killed: 16.8% (0:1) to 21.3% (1:1) to 15.8% (2.5:1)to 21.7% (10:1) to 17.8 % (40:1). At effector:target ratio of10:1 and 40:1, the CD4 T lymphocytes that were primed fromdendritised blast cells exhibited close to a significant 30%cytotoxicity. The spontaneous percentage blast cell death at0:1 effector:target ratio are approximately similar at 16% inboth control and dendritised groups. Results from Figure 3(b)are derived from four patient samples.

4. Discussion

In this study, we report the successful generation of den-dritised leukemic blast cells in the presence of GM-CSF,IL-4 and TNF-𝛼 based on our “dendritisation” approach.The patients that were reported in this study were newlydiagnosed with AML of different subtypes (M1, M2 andM4)and blast percentages (ranging from 3%–91%). Generationof dendritised blast cells was successful in all patient sam-ples with dendritised blast cells expressing co-stimulatorymolecules (CD80 and CD86) and MHC II while retainingits blast cell phenotype (CD117+ and CD34+). Our protocol

AgiAlPublishing House | http://www.agialpress.com/

The Open Access Journal of Science and Technology 5

MHC IICD86CD8060

45

30

15

0100 101 102 103 104

Co

un

ts

FL 2Log

60

45

30

15

0100 101 102 103 104

60

45

30

15

0100 101 102 103 104

Co

un

ts

Co

un

ts

FL 2Log FL 2Log

60

50

40

30

20

10

0

0 4 7 10

0

500

1000

1500

2000

MF

I of

MH

C II

Immunophenotype for dendritisation phase

MF

I of

CD

80

an

d C

D8

6

CD80 CD80 MHC II

Day

Day 0Day 0

Day 0

Day 7Day 7

Day 7

(a)

(b)

Figure 1: Flow-cytometry analysis of co-stimulatory molecules (CD80 and CD86). MHC II and “dendritised” leukemic blast cells.(a) Distinctive increase in the co-stimulatory molecules (CD80 and CD86) and MHC II when compared between non-dendritised (Day 0)leukemic blast cells and dendritised blast cells at Day 7. Diagram shown is representative for Patient 1 and consistent with all 7 patients.Leukemic blast cells from 7 patients were co- cultured in a concortion of IL-4, GM-CSF and TNF-α over a period of 10 days. The dMFIof CD80, CD86 and MHC II peaked at approximately Day 7. (b) Flow cytometry analysis shows the progressive increase in the number ofleukemic blast cells which are positive for stem cell marker CD34, expressing CD80, CD86 and MHC II as the dendritisation days increasesfrom Day 0 to Day 7.

is better than other previous studies which were only 70%successful in differentiating leukemic blast cells to maturedendritic cells [23–25]. Indeed, there have been existingstudies which documented the generation of “leukemicdendritic cells” [26] and “immunocompetent dendritic cells”[21] that were derived from blast cells. However, a majordifference between our dendritised blast cells and the den-dritic cells that were generated in other studies [21, 26]was that our dendritised blast cells remained as blast cells(albeit possessing antigen presenting capabilites) whereasthe dendritic cells from other studies fully differentiated todendritic cells. We believe that our dendritised blast cellsstill possessed inherent leukemic antigens. However, it isuncertain if there is a reduction in the leukemic antigens

present in the dendritic cells that were generated from otherstudies.

This study shows that apart from the expression of highlevel of co-stimulatory molecules, our dendritised blast cellsalso possess up-regulated MHC II which was not reportedin dendritic cells generated from other studies. Contraryto the fact that MHC II presents exogeneous antigens,there have been also multiple pathways that have shownto contribute to the MHC class II mediated presentationof cytoplasmic and nuclear viral, tumor and self antigen[27]. These pathways include chaperone mediated autophagy(CMA), macroautophagy, a TAP-dependent pathway andintracellular antigen transfer which eventually present theantigens to CD4 positive T lymphocytes [27]. In our study, it

AgiAlPublishing House | http://www.agialpress.com/

6 The Open Access Journal of Science and Technology

10

010

110

210

310

4

100 101 102 103 104100 101 102 103 104

100

10

23

1.1% 0.6% 1.3% 0.6% 1.3% 0.7% 1.2% 0.6% 0.0%

70.3% 5.7% 92.9% 50.8%88.1% 28.0%21.2% 9.8%9.0%89.2% 93.8% 4.0% 77.1% 47.5%

66.8% 5.0% 87.4% 43.3%88.4% 25.7%34.3% 10.5%10.2%83.1% 84.6% 7.1% 58.3% 49.7%

1.4% 0.0% 1.7%

7.4% 0.0% 7.0%

0.9% 0.1%

4.0% 2.7% 3.8% 6.8% 6.5% 1.6% 5.1% 2.4% 0.2%4.5% 0.6%

SS

01

02

3S

S0

101 102 103 104

CD127(PE)CD127(PE)

CD4-FC5

CD

45R

A(F

ITC

)10

010

110

210

310

4

100 101 102 103 104

CD127(PE)

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CCR7(PE)

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

62L(PE)

62L(PE)

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD45RO(PE)

CD45RO(PE)

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD25(PE)

CD25(PE)

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD28(PE)

CD28(PE)

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD27(PE)

CD27(PE)

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD

45R

A(F

ITC

)

10

010

110

210

310

4100 101 102 103 104

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD

45R

A(F

ITC

)

10

010

110

210

310

4

100 101 102 103 104

CD127(PE)

CD

45R

A(F

ITC

)

SSCD45

CD45RACD127 CCR7 CD62L CD45RO CD25 CD28 CD27

Control CD4 T- Lymphocytes

Profile of CD4 T- Lymphocytes Co- Cultured with “Dendritised” Blast Cells

100

75

50

25

0100

101 102 103 104

FL 1 Log

Counts

Granulysin

100

75

50

25

0100

101 102 103 104

FL 1 Log

Counts

Granzyme B

100

75

50

25

0100

101 102 103 104

FL 1 Log

Counts

Granulysin A

100

75

50

25

0100

101 102 103 104

FL 1 Log

Counts

Perforin

100

75

50

25

0100

101 102 103 104

FL 1 Log

Counts

IFN-yn = 4

Granzyme A

(b)

(c)

10

0

100 101 102 103 104

10

110

210

310

4C

D8

-FIT

C

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

110

210

310

4C

D8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

110

210

310

4C

D8

-FIT

C

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

11

02

10

31

04

CD

8-F

ITC

CD4-PE

10

0

100 101 102 103 104

10

110

210

310

4C

D8

-FIT

C

CD4-PE

T- Lymphocytes Co-Cultured with “Dendritised” Blast Cells

Control (Profile of T- Lymphocytes)

CD8CD4 Day 7 Day 14 Day 21 Day 28 Day 35 Day 45

20.8%

36.2% 0.0% 19.1% 1.2% 24.6% 0.5% 38.1% 1.2% 18.7% 1.1% 1.6%

54.8%

34.2%

0.7% 51.3% 0.1% 51.5% 0.7% 0.7% 0.8%

79.2%9.5%46.0%17.8%38.8%24.3%26.1%21.6%23.9%24.7%52.5%25.9%

36.1% 35.1% 1.1% 10.5%

55.9% 9.4%33.6% 24.3%24.6% 27.1%43.8% 50.4%46.8%17.0% 35.8%

(a)

Figure 2: Flow cytometry analysis of expanded T-lymphocytes. (a) Flow cytometry diagram illustrating significant increase in bothCD8 and CD4 T lymphocytes from both control and dendritised culture flask. Beyond Day 28 of lymphoexpansion, there was a greaterincrease in CD4 T lymphocytes in comparison to CD8 T lymphocytes for both control and dendritised culture flask. (b) Both CD4 positiveT lymphocytes from control and dendritised flasks exhibit effector memory (TEM) phenotype (CD45RA−, CD45RO+, CD62L−, CCR7−,CD127−, CD25+, and CD27±, CD28±). (c) There was a significantly higher expression of INF-γ in CD4 positive T lymphocytes that wasco-cultured with dendritised leukemic blast cells in comparison to the control CD4 positive T lymphocytes. There was also a slightly higheramount of granzyme A and granulysin in T lymphocytes that was co-cultured with “dendritised” leukemic blast cells than the control CD4positive T lymphocytes. Unfilled Histogram = Control CD 4 T- Lymphocytes Filled Histogram = CD4 T- Lymphocytes Co-Cultured with“dendritised” Blast Cells.

AgiAlPublishing House | http://www.agialpress.com/

The Open Access Journal of Science and Technology 7

10

23

ss0

100 101 102 103 104

CD45-PC5

10

23

ss0

100 101 102 103 104100 101 102 103 104

CD117-FITC

R1

10

23

ss0

100 101 102 103 104

CD45-PC5

10

23

ss0

100 101 102 103 104

CD117-FITC

R1

10

01

01

10

21

03

10

4

100 101 102 103 104

10

01

01

10

21

03

10

4

7A

AD

7A

AD

AnnexinV-PE AnnexinV-PE100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE

100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE

100 101 102 103 104

10

01

01

10

21

03

10

47

AA

D

AnnexinV-PE

BM284-CYTOASSAY-0:1 BM284-CYTOASSAY-1:1 BM284-CYTOASSAY-2.5:1 BM284-CYTOASSAY-10:1 BM284-CYTOASSAY-40:1

BM284-CYTOASSAY-0:1 BM284-CYTOASSAY-1:1 BM284-CYTOASSAY-2.5:1 BM284-CYTOASSAY-10:1 BM284-CYTOASSAY-40:1

0.7% 0.7% 0.2% 1.2% 0.4% 3.4% 0.6% 4.4% 3.7% 21.1%

83.0% 15.6% 82.5% 16.1% 78.1% 18.1% 74.0% 21.1% 55.0% 20.2%

1.0% 2.3% 0.8% 4.7% 0.6% 3.2% 0.6% 4.2% 0.7% 3.7%

80.2% 16.5% 75.8% 18.77% 7.7% 18.5% 76.1% 19.2% 82.2% 13.4%

CD117

SS

CD45

SS

AnnexV

7AAD

0.7% 0.7% 0.2% 1.2%

16.1% 18.1%

3.4%0.4% 0.6% 4.4% 3.7% 20.1%

1.0% 2.3% 0.8% 4.7% 3.2%0.6% 0.6% 4.2% 0.7% 3.7%

16.5% 18.7% 18.5% 19.2% 13.4%

20.2%21.1%15.6%

E:T= 0:1 E:T= 1:1 E:T= 2.5:1 E:T= 10:1 E:T= 40:1

50

40

30

20

10

0

Cytotoxicity analysis of CD4 T vs autologous blasts

Pe

rce

nt

of

de

ad

Le

uke

mic

bla

st c

ell

es/

%

0:1 1:1 2.5:1 10:1

E:T ratio

Lymphoexpanded CD4 Ctrl CD4

Cytotoxicity of CD4 T- Lymphocytes Co- Cultured with “Dendritised” Blast Cells

(a)

(b)

Cytotoxicity of Control CD4 T-Lymphocytes

Figure 3: Cytotoxicity Assay profile of CD4 T lymphocytes from dendritised blast cells and control. (a) Flow cytometry analysis ofleukemic blast cells that are CD117 positive. (i) Flow cytometry diagram shows significant decrease in the population of 7AAD and AnnexinV double negative leukemic blast cells with increasing ratio of effector cells: target cells. (ii) Flow cytometry diagram shows relativelyconstant population of 7AAD and Annexin V double negative leukemic blast cells throughout all ratios of effector cells: target cells. Figure3(a) is illustrative of patient 2. (b) Cumulative average percentage of dead leukemic blast cells for effector:target (0:1, 1:1, 2.5:1, 10:1).There was a superiority of cytotoxicity for CD4 T lymphocytes from dendritised blast cells in comparison to control CD4 T lymphocytes.At effector:target ratio of 10:1, the CD4 T lymphocytes that were primed from “dendritised” blast cells exhibited close to a significant 15%cytotoxicity. Values are obtained from the cumulative average of patient 1–3 in this study, with two set of values for patient 1.

was observed that the MHC II and co-stimulatory molecules(CD 80 and CD86) attained a maximum level around Day7. Hence, we believe that the antigen presenting capabilityof our dendritised blast cells peak at Day 7 of dendritisationphase. This would mean that our dendritised blast cellswould be able to efficiently utilize the up-regulated MHCII to present inherent leukemic antigen to surrounding CD4positive T lymphocytes, in the presence of up-regulatedCD80 and CD86.

After the generation of dendritised blast cells, the nextphase of our study was to expand autologous T lymphocytesthat were presented with the inherent leukemic antigensby the generated dendritised blast cells. We observed anincrease in proliferated CD8 positive and CD4 positive Tlymphocytes. However, it was apparent that towards theend of the “lymphoexpansion” phase, there was a more

significant increase in the CD4 positive T lymphocytesthan CD8 positive T lymphocytes. This could possibly bedue to the treatment of inflammatory signal such as INF-𝛾 that could induce MHC II expression on the surface of avariety of APC [27]. This observation was expected becausethere was an up- regulation of MHC II molecules in thegenerated dendritised blast cells that were responsible forthe presentation of inherent leukemic antigens to the CD4positive T lymphocytes. Similar observation was made forthe proliferated T lymphocytes in the control arm.

Both CD4 positive T lymphocytes from control anddendritised flasks exhibited effector memory (TEM) pheno-type (CD45RA−, CD45RO+, CD62L−, CCR7−, CD127−,CD25+, and CD27±, CD28±). Human TEM are memorycells that have lost the constitutive expression of CCR7and CD62L, and display characteristic sets of chemokine

AgiAlPublishing House | http://www.agialpress.com/

8 The Open Access Journal of Science and Technology

receptors and adhesion molecules that are required forhoming to inflamed tissues [28]. Therefore, the CD4 positiveT lymphocytes from both control and dendritised flasks werecharacterized to have rapid effector function. Nonetheless,access of cytotoxic capability of CD4 positive T lymphocytesremained paramount despite knowing the TEM phenotype ofthe CD4 positive T lymphocytes. In our study, we have shownconsistent stimulation of autologous killing of leukemic blastcells through priming with dendritised blast cells via flowcytometry cytotoxicity assay (see Figure 3). At effector:targetratio of 1:1, there was no autologous cytotoxicity for CD4positive T lymphocytes from both dendritised and controlflasks. However, as the effector:target ratio increased to2.5:1, 10:1 and 40:1, there was a significant increase in thecytotoxicity effect for the CD4 positive T lymphocytes incomparison to the control CD4 positive T lymphocytes. Theextent of cytotoxicity towards autologous blast cells reachedto about 30% (E:T ratio of 40:1) more for CD4 positive Tlymphocytes that was primed from dendritised blast cells ascompared to control CD4 positive T lymphocytes.

In conclusion, our study shows evidence of distinctcytotoxicity of CD4 positive T lymphocytes against autol-ogous blast cells. “Dendritised” AML blasts were thereforeable to function with antigen presenting capability, whilestill phenotypically existing as blast cells. This was alsoachievable for a wide range of FAB subtypes of AMLmakingit a potential immunotherapy against AML. We believe thatour study findings support the feasibility of using culturedand “dendritised” AML blasts cells to present leukemicantigens to autologous T cells hence conferring cytotoxicityagainst autologous blast cells and their possible use in clinicaltrials in the future.

Funding

This work is supported by: National Medical ResearchCouncil [SRG-CG#04/2010] and Cell Pro & Immune Reconfrom Singapore Cancer Syndicate [SGH-SCS-BMTC].

Acknowledgements

The authors appreciate the support of SingHealth/Duke-NUSAcademic Medicine Research Institute and Taara Madhavan(Associate in Clinical Sciences, Duke-NUS Graduate Medi-cal School) in editing this manuscript.

Contribution

Mr. Bernard Yap and Dr. Ho Liam Pock have contributedequally to this manuscript.

Declaration of Conflicts

All authors have nothing to declare.

References

[1] I. Houtenbos, T. M.Westers, G. J. Ossenkoppele, and A. A. vande Loosdrecht, Employing the immunological synapse inAML:Development of leukemic dendritic cells for active specificimmunization, Immunobiology, 210, no. 2-4, 249–257, (2005).

[2] M. Smith, M. Barnett, R. Bassan, G. Gatta, C. Tondini, andW. Kern, Adult acute myeloid leukaemia, Critical Reviews inOncology/Hematology, 50, no. 3, 197–222, (2004).

[3] B. Deschler and M. Lübbert, Acute myeloid leukemia: Epi-demiology and etiology, Cancer, 107, no. 9, 2099–2107,(2006).

[4] A. Van Driessche, L. Gao, H. J. Stauss, P. Ponsaerts, D.R. Van Bockstaele, Z. N. Berneman, and V. F. I. Van Ten-deloo, Antigen-specific cellular immunotherapy of leukemia,Leukemia, 19, 1863–1871, (2005).

[5] H. R. Kim, J. H. Shin, J. N. Lee, and E. Y. Lee, Clinicalsignificance of quantitation of WT1 gene expression forminimal residual disease monitoring of acute myelogenousleukemia, The Korean journal of laboratory medicine, 27, no.5, 305–312, (2007).

[6] M. Yanada, G. Garcia-Manero, G. Borthakur, F. Ravandi,H. Kantarjian, and E. Estey, Relapse and death during firstremission in acute myeloid leukemia, Haematologica, 93, no.4, 633–634, (2008).

[7] M. M. Ho, D. E. Hogge, and V. Ling, MDR1 and BCRP1expression in leukemic progenitors correlates with chemother-apy response in acute myeloid leukemia, Experimental Hema-tology, 36, no. 4, 433–442, (2008).

[8] K. Hussein, B. Jahagirdar, P. Gupta, L. Burns, K. Larsen, andD.Weisdorf, Day 14 bone marrow biopsy in predicting completeremission and survival in acute myeloid leukemia, AmericanJournal of Hematology, 83, no. 6, 446–450, (2008).

[9] A. Fefer, K. M. Sullivan, P. Weiden, C. D. Buckner, G.Schoch, R. Storb, and E. D. Thomas, Graft versus leukemiaeffect in man: the relapse rate of acute leukemia is lowerafter allogeneic than after syngeneic marrow transplantation.,Progress in clinical and biological research, 244, 401–408,(1987).

[10] K. Bagheri, K. Alimoghadam, A. A. Pourfathollah, Z. M.Hassan, J. Hajati, and S. M. Moazzeni, The efficient generationof immunocompetent dendritic cells from leukemic blasts inacute myeloid leukemia: A local experience, Pathology andOncology Research, 15, no. 2, 257–267, (2009).

[11] X. Huang, Y. Wang, D. Liu, L. Xu, H. Chen, Y. Chen, W. Han,H. Shi, and K. Liu, Modified donor lymphocyte infusion (DLI)for the prophylaxis of leukemia relapse after hematopoieticstem cell transplantation in patients with advanced leukemia-feasibility and safety study, Journal of Clinical Immunology,28, no. 4, 390–397, (2008).

[12] J. H. F. Falkenburg, W.M. Smit, and R.Willemze, Cytotoxic T-lymphocyte (CTL) responses against acute or chronic myeloidleukemia, Immunological Reviews, 157, 223–230, (1997).

[13] M. Yazaki, T. Takahashi, M. Andho, Y. Akatsuka, T. Ito, Y.Miyake, Y. Ito, S. Nakamura, and Y. Wada, A novel minorhistocompatibility antigen recognized by HLA-A31 restrictedcytotoxic T lymphocytes generated from HLA-identical bonemarrow donor lymphocytes,BoneMarrow Transplantation, 24,no. 2, 129–137, (1999).

[14] A. Eljaafari, A. Farre, K. Duperrier, J. Even, H. Vie, M.Michallet, G. Souillet, A. Catherine Freidel, L. Gebuhrer, andD. Rigal, Generation of helper and cytotoxic CD4+T cell clones

AgiAlPublishing House | http://www.agialpress.com/

The Open Access Journal of Science and Technology 9

specific for the minor histocompatibility antigen H-Y, after invitro priming of human T cells by HLA-identical monocyte-derived dendritic cells, Transplantation, 71, no. 10, 1449–1455,(2001).

[15] P. J. Amrolia, S. D. Reid, L. Gao, B. Schultheis, G. Dotti,M. K. Brenner, J. V. Melo, J. M. Goldman, and H. J. Stauss,Allorestricted cytotoxic T cells specific for human CD45 showpotent antileukemic activity, Blood, 101, no. 3, 1007–1014,(2003).

[16] A. Draube, M. Beyer, and J. Wolf, Activation of autolo-gous leukemia-specific T cells in acute myeloid leukemia:Monocyte-derived dendritic cells cocultured with leukemicblasts compared with leukemia-derived dendritic cells, Euro-pean Journal of Haematology, 81, no. 4, 281–288, (2008).

[17] M. M. Horowitz, R. P. Gale, P. M. Sondel, J. M. Goldman,J. Kersey, H. Kolb, A. A. Rimm, O. Ringdén, C. Rozman, B.Speck, R. L. Truitt, F. E. Zwaan, and M. M. Bortin, Graft-versus-leukemia reactions after bone marrow transplantation,Blood, 75, no. 3, 555–562, (1990).

[18] J. Molldrem, S. Dermime, K. Parker, Y. Z. Jiang, D.Mavroudis,N. Hensel, P. Fukushima, and A. J. Barrett, Targeted T-celltherapy for human leukemia: Cytotoxic T lymphocytes specificfor a peptide derived from proteinase 3 preferentially lysehuman myeloid leukemia cells, Blood, 88, no. 7, 2450–2457,(1996).

[19] C. Scheibenbogen, A. Letsch, E. Thiel, A. Schmittel, V.Mailaender, S. Baerwolf, D. Nagorsen, and U. Keilholz, CD8T-cell responses to Wilms tumor gene product WT1 andproteinase 3 in patients with acute myeloid leukemia, Blood,100, no. 6, 2132–2137, (2002).

[20] A. Van Driessche, L. Gao, H. J. Stauss, P. Ponsaerts, D.R. Van Bockstaele, Z. N. Berneman, and V. F. I. Van Ten-deloo, Antigen-specific cellular immunotherapy of leukemia,Leukemia, 19, no. 11, 1863–1871, (2005).

[21] K. Bagheri, K. Alimoghadam, A. A. Pourfathollah, Z. M.Hassan, J. Hajati, and S. M. Moazzeni, The efficient generationof immunocompetent dendritic cells from leukemic blasts inacute myeloid leukemia: A local experience, Pathology andOncology Research, 15, no. 2, 257–267, (2009).

[22] J. M. . Austyn, Dendritic cells, Curr Opin Hematol, 5, 3–15,(1998).

[23] S. P. Robinson, N. English, R. Jaju, L. Kearney, S. C.Knight, and C. D. L. Reid, The in-vitro generation of dendriticcells from blast cells in acute leukaemia, British Journal ofHaematology, 103, no. 3, 763–771, (1998).

[24] R. E. Brouwer, M. Van Der Hoorn, H. C. Kluin-Nelemans, S.Van Zelderen-Bhola, R.Willemze, and J. H. F. Falkenburg, Thegeneration of dendritic-like cells with increased allostimulatoryfunction from acute myeloid leukemia cells of various FABsubclasses, Human Immunology, 61, no. 6, 565–574, (2000).

[25] B. D. Harrison, J. A. Adams, M. Briggs, M. L. Brereton,and J. A. Liu Yin, Stimulation of autologous proliferativeand cytotoxic T-cell responses by ”leukemic dendritic cells”derived from blast cells in acute myeloid leukemia, Blood, 97,no. 9, 2764–2771, (2001).

[26] B. D. Harrison, J. A. Adams, M. Briggs, M. L. Brereton,and J. A. Liu Yin, Stimulation of autologous proliferativeand cytotoxic T-cell responses by ”leukemic dendritic cells”derived from blast cells in acute myeloid leukemia, Blood, 97,no. 9, 2764–2771, (2001).

[27] V. L. Crotzer and J. S. Blum, Autophagy and its role in MHC-mediated antigen presentation, Journal of Immunology, 182,no. 6, 3335–3341, (2009).

[28] F. Sallusto, J. Geginat, and A. Lanzavecchia, Central memoryand effector memory T cell subsets: Function, generation, andmaintenance, Annual Review of Immunology, 22, 745–763,(2004).

AgiAlPublishing House | http://www.agialpress.com/