Characterization of in vitro migratory properties of anti-CD19 chimeric ...

Experimental Hematology 34 (2006) 1218–1228

Characterization of in vitro migratory properties of anti-CD19 chimeric receptor-redirected CIK cells for their potential use in B-ALL immunotherapy Virna Marina, Erica Dandera, Ettore Biagia, Martino Introna , Grazia Fazioa, Andrea Biondia,*, and Giovanna D’Amicoa b

a

b

Centro Ricerca ‘‘M. Tettamanti,’’ Clinica Pediatrica Universita` Milano-Bicocca, Ospedale San Gerardo, Monza, Italy; Laboratorio di Terapia Cellulare e Genica ‘‘G. Lanzani,’’ Divisione di Ematologia, Ospedale Riuniti di Bergamo, Bergamo, Italy (Received 29 November 2005; revised 7 April 2006; accepted 3 May 2006)

Objective. Cytokine-induced killer (CIK) cells are ex vivo expanded cells enriched in CD3+CD56+ natural killer T (NKT) cells with major histocompatibility-unrestricted cytotoxicity against several tumoral targets, except B-lineage acute lymphoblastic leukemia (B-ALL). We redirected CIK cells cytotoxicity toward B-ALL with a chimeric receptor specific for the CD19 antigen and then explored if modified-CIK cells maintain the same chemotactic properties of freshly isolated NKT cells, whose trafficking machinery reflects their preferential localization into the sites of B-ALL infiltration. Material and Methods. CIK cells were expanded ex vivo for 21 days and analyzed for expression of adhesion molecules and chemokine receptors regulating adhesion and homing toward leukemia-infiltrated tissues. CIK cells were then transduced with the anti-CD19-z-internal ribosomal entry site-green fluorescent protein retroviral vector and characterized for their cytotoxicity against B-ALL cells in a 51Cr-release assay and for their trafficking properties, including chemotactic activity, adhesion and transendothelial migration, and metalloproteases-dependent invasion of Matrigel. Results. Similarly to freshly isolated NKT cells, CD49d and CD11a were highly expressed on CIK cells. Moreover, CIK cells expressed CXCR4, CCR6, and CCR7 (mean expression 72%, 60%, and 32%, respectively), presenting chemotactic activity toward their respective ligands. Anti-CD19 chimeric receptor-modified CIK cells became cytotoxic against B-ALL cells (mean lysis, 60%) and showed, after exposure to a CXCL12 gradient, high capacity to adhere and transmigrate through endothelial cells and to invade Matrigel. Conclusion. The potential capacity to localize into leukemia-infiltrated tissues of anti-CD19 chimeric receptor-redirected CIK cells, together with their ability to efficiently kill B-ALL cells, suggests that modified-CIK cells represent a valuable tool for leukemia immunotherapy. Ó 2006 International Society for Experimental Hematology. Published by Elsevier Inc.

Cytokine-induced killer (CIK) cells are a recently discovered population of immune-effector cells that can be expanded from peripheral blood mononuclear cells up to 200 to 1000 fold in 14 to 21 days of culture after an initial priming with interferon (IFN)-g and OKT3, followed by repeated stimulation with interleukin (IL)-2 [1–3]. CIK cells show potent cytotoxic activity against a number of tumor Offprint requests to: Giovanna D’Amico, Ph.D., Centro Ricerca ‘‘M. Tettamanti,’’ Clinica Pediatrica Universita` Milano-Bicocca, Ospedale San Gerardo, Via Donizetti, 106 20052 Monza (MI), Italy; E-mail: giovanna. [email protected] *Andrea Biondi and Giovanna D’Amico shared senior authorship.

cell lines or freshly isolated tumor samples, including acute myeloid leukemia, chronic myeloid leukemia and B lymphoma cells [3–7]. B-lineage acute lymphoblastic leukemia (B-ALL) and B-chronic lymphocytic leukemia cells appear to be more resistant to CIK-mediated lysis [7,8]. At the same time, CIK cells show no cytolytic activity against normal bone marrow or spleen cells in vitro, and limited inhibition of colony-forming capacity, suggesting specificity for tumor cells [9,10]. CIK cells are a heterogeneous population, enriched in the CD3þCD56þCD8þ NKT subset. NKT cells are immune cells that share phenotypic and functional properties with

0301-472X/06 $–see front matter. Copyright Ó 2006 International Society for Experimental Hematology. Published by Elsevier Inc. doi: 10.1016/j.exphem.2006.05.004

V. Marin et al./ Experimental Hematology 34 (2006) 1218–1228

both NK cells and effector T lymphocytes and that coexpress the CD3 and CD56 antigens. They represent a rare population detectable in peripheral blood (from 1% to 5% in uncultured peripheral blood mononuclear cells). NKT cells are involved in various immune responses, including control of microbial infections, tumor killing, tolerance induction, and suppression of graft-vs-host disease after human stem cell transplantation (HSCT) [11,12]. A characteristic of these cells is their capacity to exert a cytolytic activity in a major histocompatibility complex nonrestricted manner and to produce different immunoregulatory cytokines (mainly IFN-g and IL-4) [13–15]. Two populations of NKT cells have been described. One population is either CD4þ or CD4/CD8/ [16,17]. These NKT cells recognize glycolipids associated with CD1d and b2-microglobulin. A second population expresses a variable T-cell receptor repertoire and is not dependent on CD1d for maturation and development [18,19]. In addition, these CD1d-unrestricted NKT cells express mainly CD8 [18–20] and it has been demonstrated that the CD3þCD56þ fraction of CIK cells represent an expanded population deriving from the CD1dunrestricted CD8þ NKT subset [20]. Recently, it has been shown that this subpopulation of CD1d-unrestricted CD8þ NKT cells secrete different chemokines (CCL3, CCL4, and CCL5) when activated by Fas ligand [21]. These chemokines can amplify the antitumor response of Fas-Ligand positive tumor cells by recruiting into the tumor-rejection site various immune-effector cells implicated in the innate response, such as granulocytes, monocytes, and NK cells. NKT cells are widely distributed in the body, including bone marrow, liver, thymus, lymph nodes, spleen, and lung [22]. In particular, bone marrow, liver, and spleen are enriched in CD1d-unrestricted NKT cells [18,23]. These latter organs represent the main sites of acute leukemia cell infiltration and proliferation [24–26]. Therefore, as CIK cells represent an expanded population of CD1d-unrestricted NKT cells, which can physiologically reach tissues where leukemia cells infiltrate, these cells could potentially represent a useful tool for antileukemia immunotherapy. Unfortunately, as previously indicated, CIK cells lack the capacity to kill B-ALL [7]. It has been shown that genetic modification of T lymphocytes and NK cells, by inducing the expression of anti-CD19 chimeric receptors, results in powerful cytotoxicity against CD19þ ALL cells. Chimeric receptor binding initiates a signaling cascade, leading to cell activation, cytokine secretion, and cytotoxicity [27,28]. Here we demonstrated that CD3þCD56þ CIK cells maintain the original trafficking properties of NKT cells, after analyzing their pattern of expression of adhesion molecules, chemokine receptors, and in vitro chemotactic activity. Furthermore, we showed that, after transduction with a retroviral vector encoding a chimeric receptor-specific for the CD19 antigen, CIK cells not only become capable to efficiently kill B-ALL cells, but also exhibit in vitro

1219

migratory properties that indicate their potential capacity to reach leukemia-infiltrated tissues, rendering them an interesting tool for B-ALL immunotherapy.

Materials and methods Cells Bone marrow cells were collected from children with B-ALL at diagnosis. Flow cytometry analysis showed that, in all tested cases, blast infiltration was O80% and O90% of the blasts expressed the CD19 antigen and lacked surface Ig (sIg). All leukemia samples were cryopreserved and subsequently thawed to be used in the experiments. The Institutional Review Board approved this study and informed consent was obtained from patients and their guardians. The CD19þ human B-lineage ALL cell line (REH) and the myeloid cell line K562 were both purchased from American Type Culture Collection (ATCC) and were maintained in RPMI1640 supplemented with 10% fetal calf serum, L-glutamine, and antibiotics (complete RPMI medium). The mouse embryonic NIH-3T3 cell line and the Phoenix-Ampho packaging cell line were both purchased from ATCC and were maintained in Dulbecco’s Modified Eagle’s Medium (Biowest, Nuaille`, France), supplemented with 10% fetal calf serum, L-glutamine and antibiotics. Generation of CIK cells CIK cells were prepared as described previously [7]. Briefly, peripheral blood mononuclear cells of healthy subjects were obtained after centrifugation of fresh blood on a density gradient using Ficoll-Hypaque (Pharmacia LKB, Uppsala, Sweden). Cells were then resuspended in complete RPMI medium. At the beginning of the culture, IFN-g (Gammakine; Dompe` Biotec S.p.A, Italy) was added at 1000 U/mL. The next day, IL-2 (Proleukin; Chiron B.V, Emeryville, CA, USA) and OKT-3 (Janssen-Cilag S.p.A., Italy) were added at 500 U/mL and at 50 ng/mL, respectively. Cells were cultured at the concentration of 3 106 cells/mL. Fresh medium and IL-2 (500 U/mL) were added weekly during culture and the cell concentration was maintained around 0.5 106 cells/mL. FACS analysis Aliquots of cells were analyzed for expression of various surface markers using Peridinin-chlorophyll-protein Complex-anti-CD3 (Becton Dickinson, San Jose, CA, USA), phycoerythrin (PE)-antiCD56 (IQ product, Groningen, The Netherlands), with isotypematched antibodies (Becton Dickinson), as controls. Surface expression of adhesion molecules was analyzed using PE-antiCD49d, FITC-anti-CD62L (Chemicon International, Temecula, CA, USA) and FITC-anti-CD11a (Becton Dickinson, Bedford, MA). Surface expression of chemokine receptors was studied by standard indirect flow cytometry staining using the following monoclonal antibodies: anti-CCR6 (Pharmingen, San Diego, CA, USA), antiCCR7 (Pharmingen), anti-CXCR4 (Pharmingen), followed by incubation with anti-mouse biotin (Jackson Immunoresearch, West Grove, PA, USA) and PE-conjugated streptavidin (Pharmingen) for nontransduced cells or TriColor-conjugated streptavidin (Caltag Laboratories, Burlingame, CA, USA) for transduced cells. To determine the surface expression of the chimeric anti-CD19 receptor, transduced CIK cells were stained with goat anti-mouse

1220


(Fab)2 polyclonal antibody conjugated with biotin (Jackson Immunoresearch), followed by incubation with tricolor (TC)conjugated streptavidin. A FACScan flow cytometer device was used (Becton Dickinson) to analyze the samples. 51

Cr release cytotoxicity assay Cytotoxicity was measured in a standard 4 hours 51Cr release cytotoxicity assay. K562 cell line or B-ALL cells were labeled with 25 mCi of 51Cr for 1 hour and washed three times. CIK cells were cultured in triplicate with 5 103 target cells at effector-to-target (E:T) ratios of 30:1, 10:1, and 1:1. After 4 hours, 30 mL supernatant were collected, 170 mL scintillation liquid (PerkinElmer Life Science, Boston, MA, USA) was added to the supernatant and the radioactivity was then detected by b-scintillation counter (PerkinElmer Life Science), as counts per minutes. The percentage of specific lysis was calculated as described previously [29]. Chemotaxis assays Chemotactic migration assays were performed as previously described [30]. Briefly, CIK cells were labeled with 25 mCi of 51 Cr for 1 hour and washed. 0.5 106 CIK cells were subsequently placed in the upper chamber of Transwell insert (5-mm pore size; Corning Costar, Corning, NY). Inserts were placed in wells containing medium alone (basal) or medium plus chemokine. CXCL12, CCL20, CCL19, and CCL21 were purchased from PeproTech (Rocky Hill, NJ). Chemokines were used at optimal concentration determined by titration. In selected experiments, supernatant was collected from the co-culture of donor-derived human bone-marrow mesenchymal cells (HBMC) [31] and BALL primary blasts and subsequently used as chemoattractant. Where indicated, cells were preincubated with a blocking antiCXCR4 antibody, AMD3100 (1 mg/mL; Sigma, St. Louis, MO, USA). Radioactivity in bottom wells (migrated cells) was evaluated after 1 hour with a b-scintillation counter. Results are expressed as the migration index of CIK cells in response to the chemokine versus the basal condition. Transendothelial migration assay Human umbilical vein endothelial cells were kindly provided by Dr. Allavena (Istituto Clinico Humanitas, Milano, Italy). Cells were maintained in 199 medium with 20% bovine serum, supplemented with endothelial cell growth supplement (100 mg/mL; Collaborative Research, Inc., Lexington, MA, USA) and heparin (100 mg/mL; Sigma). The transendothelial migration assay was done as described previously [32]. In brief, 51Cr-labeled CIK cells were seeded in the upper compartment and co-incubated with endothelial cells monolayers for 1 hour at 37 C. Nonadherent cells were gently washed away and adherent cells were scraped with a cotton swab. The radioactivity in the lower compartment referred to transmigrated cells. The adherent cells were considered to comprise both cells bound to endothelial cells, as well as those that had transmigrated. Gelatinase activity assay CIK cells were stimulated 24 hours with 300 ng/mL CXCL12. Supernatants were analyzed using a MMP Gelatinase Activity Assay kit (Chemicon International), according to manufacturer’s instructions.

Trans-Matrigel migration Trans-Matrigel migration was performed as described previously [33]. Briefly, the upper chamber of polycarbonate Transwell insert was coated with 10 mL (0.5 mg/mL) of reconstituted basement membrane (Matrigel; Becton Dickinson) and incubated at 37 C for 30 minutes to allow Matrigel polymerization. Subsequent steps were performed as for the chemotaxis assay. Plasmids and retrovirus production The anti-CD19-z-internal ribosomal entry site-green fluorescent protein (IRES-GFP), anti-CD19-truncated-IRES-GFP (control) and pMSCV-IRES-GFP (vector) constructs were kindly provided by Dr. D. Campana (St. Jude Children’ Research Hospital, Memphis, TN) and have been described previously [27,28]. The Phoenix Ampho packaging cell line was seeded and transfected as reported previously [34]. Fresh supernatant was collected 48 hours after transfection and frozen at 80 C. NIH-3T3 cells were used to titrate virus concentration. Retroviral transduction of CIK cells On day 3 of CIK culture, coating with the recombinant fibronectin fragment FN CH-296 (RetroNectin; TaKaRa BioEurope, Gennevilliers, France) was performed in 24-well plates. CIK cells were resuspended in 1 mL fresh viral supernatant on retronectin-coated well. CIK cells were then spin-infected in the presence of Polybrene (8 mg/mL; Sigma) and IL-2 (500 U/mL). This procedure was repeated 8 hours later and the following day. The multiplicity of infection used in each experiment was maintained between 4 and 6. Immunomagnetic selection of CD56þ CIK cells CD56þ transduced CIK cells were immunoselected by passage through a separation column (MS MiniMacs, Miltenyi Biotec, Bergisch Gladbach, Germany) with anti-CD56 microbeads (Miltenyi Biotec) according to the manufacturer instructions. Both the positive and negative fractions were used for the cytotoxicity assay. Statistical analysis Results were compared using the paired Student’s t-test. A p value !0.05 was considered to be significant.

Results Generation and characterization of CIK cells CIK cells were generated according to the protocol previously described [7]. After 21 days of culture, mean fold increase of CD3þCD56þ was 458.0 (range, 117.3859.9; n 5 10). Mean percentage of CD3þCD56þ cells was 44.2% (range, 17.3%-69.0%; n 5 14; Fig.1A); most CD3þCD56þ cells expressed CD8 (mean, 74.0%; range, 58.4%-93.1%; n 5 6; Fig.1A). The remaining cells were CD3þCD56- cells, mostly CD8þ, as previously reported [5]. The capacity of CIK cells to recognize and kill the standard target K562 cell line was determined, by using a standard 51Cr-release assay. CIK cells showed a mean cytotoxic activity against K562 of 36.1% at an E:T ratio of 30:1 (range, 13.585.2%; n 5 11; Fig. 1B). As expected,


1221

Figure 1. Phenotypic and functional characterization of cytokine-induced killer (CIK) cells. (A) Expression of CD56 along with CD3 and CD8 on the surface of CIK cells was evaluated at the beginning of and after 21 days of culture by flow cytometry. One representative experiment out of 14 is shown. (B) Cytotoxic activity of CIK cells was determined after 21 days of culture with a standard 4 hours 51Cr-release assay. CIK cells efficiently killed the K562 cell line, but had a limited cytotoxic activity against B-lineage acute lymphoblastic leukemia (B-ALL) cells. One representative experiment out of 11 is shown. Data are mean SD of three replicates (**p ! 0.005 and *p ! 0.02 vs REH and primary B-ALL cells).

cytotoxicity against B-ALL blasts was limited (mean lysis !15% at E:T ratio of 30:1, for both the REH cell line and primary leukemia cells; range, 5.419.2%; n 5 5; Fig. 1B).

Expression of adhesion molecules, chemokine receptors and chemotactic activity of CIK cells In vivo NKT cells accumulate at high levels in the liver, bone marrow, thymus, and spleen [22]. Because tissue distribution is likely to be regulated by a series of adhesive and activation steps mediated by adhesion molecules and chemokines, we examined the expression of the adhesion molecules implicated in homing to extra-lymphoid tissues such as liver (CD11a) [23], bone marrow (CD49d) [35,36], and lymph nodes (CD62L) [37] on CD3þCD56þ CIK cells. A small, but distinct subset of CD62LþCD3þ CD56þ (510%; n 5 3, Fig. 2A) was detectable within CIK populations, whereas CD49d and CD11a were highly expressed on the cell surface of O98% of CD3þCD56þ CIK cells. Mean levels of expression on the CD3þCD56 fraction of CD62L, CD49d, and CD11a were 8.1% (range, 2.015.3%; n53), 98.5% (range, 97.199%; n 5 3) and 100%, respectively. To investigate the chemotactic capacity of CIK cells, we analyzed the expression of chemokine receptors specifically involved in the process of trafficking into different tissue sites. Most CD3þCD56þ CIK cells expressed CXCR4 [38,39], a chemokine receptor implicated in the recruitment of cells to the bone marrow (mean expression, 72.2%; range, 56.884.1%; n 5 4; Fig. 2A). Moreover, a mean of 60% of CD3þCD56þ cells expressed high levels of CCR6 (range, 55.368.5%; n 5 4 on CD3þCD56þ; Fig. 2A), whose ligand (CCL20) [40] is expressed in liver and among various nonlymphoid tissues. As reported by others [41,42] for NKT freshly isolated cells, the lymphoid

homing receptor CCR7 [43,44] was expressed only by a minority of CD3þCD56þ CIK cells (mean, 31.7%; range, 29.837.6%; n 5 4; Fig. 2A). CXCR4 was expressed at similar levels (mean, 80.4%, range, 77.786.9%; n 5 4) on CD3þCD56 CIK subpopulation, while CCR6 (mean, 17.3%, range, 12.024.0%; n 5 4) and CCR7 (mean, 56.0%, range, 49.462.5%; n 5 4), were significantly (p ! 0.05) more expressed on the CD3þCD56 fraction. We next evaluated the chemotactic response of CIK cells to chemokines. CXCL12 [45,46], the CXCR4 ligand, induced robust chemotaxis of CIK cells (Fig. 2B), with the CD3þCD56þ and CD3þCD56 subpopulations being represented at similar levels in the fraction of migrated cells (Fig. 2C, n 5 4). CCL20 induced lower, but still significant, migration of CIK cells with no major differences in the CD3þCD56þ and CD3þCD56 subpopulations recovered in the migrated fraction (Fig. 2C). CIK cells were also attracted at considerable levels by the CCR7 ligands, CCL19 [47,48], and CCL21 [49,50] (Fig. 2B), with a significant (p ! 0.05) predominance of the CD3þCD56 subpopulation in the migrated fraction, probably because of their higher expression of CCR7 (Fig. 2C). To demonstrate the role of CXCL12 in determining CIK accumulation in leukemia-infiltrated bone marrow, which represents the major site of leukemia infiltration and proliferation [24,26], we analyzed the capacity of CIK cells to migrate in response to the supernatant collected from the co-culture of donor-derived HBMC and primary B-ALL blasts. We demonstrated that CXCL12 could be found in this supernatant at 470 pg/mL. Indeed, CIK cells showed to be able to consistently migrate after exposure to the supernatant of HBMCB-ALL co-culture and this migratory activity was completely abrogated when cells were preincubated with the anti-CXCR4 blocking antibody (AMD3100) (Fig. 2D).

1222


Figure 2. Adhesion molecules, chemokine receptor expression, and chemotactic activity of cytokine-induced killer (CIK) cells. (A) Expression of adhesion molecules (upper quadrants) and chemokine receptors (lower quadrants) on CIK cells was evaluated by flow cytometry. One representative experiment is shown. (B) The migratory activity of CIK cells in response to chemokines was determined by an in vitro chemotactic assay (four donors were evaluated). The horizontal line at the migration index 1.0 indicates lack of chemotaxis. (C) The phenotypic analysis of the migrated populations (CD3þCD56þ and CD3þCD56) was evaluated by flow cytometry. Both the CD3þCD56þ and the CD3þCD56 subsets were attracted at different levels by the chemokines analyzed, according to the surface expression of the respective chemokine receptors (*p ! 0.05 vs CD3þCD56). The bars represent mean SD of the four donors shown in (B). (D) Pretreatment of CIK cells with AMD3100 (1 mg/mL) abrogated CIK cells migration in response to the supernatant collected from human bone-marrow mesenchymal cells (HBMC)B-lineage acute lymphoblastic leukemia (B-ALL) cells co-culture (*p 5 0.05 vs HBMCB-ALL co-culture supernatant without blocking antibody). The bars represent mean SD of three independent experiments.

Retroviral transduction of CIK cells with the anti-CD19 chimeric receptors and evaluation of their cytotoxic and migratory activity Migratory properties of CIK cells represent a characteristic that renders them an attractive tool for leukemia immunotherapy. Unfortunately, CIK cells do not show any significant antitumor activity toward B-ALL. Therefore, we analyzed if their cytotoxic activity could be enhanced and redirected towards B-ALL, and we extensively characterized the in vitro migratory properties of transduced CIK cells in response to chemokines responsible for the homing to leukemia-infiltrated tissues. We transduced CIK cells with retroviral vectors containing the anti-CD19 chimeric receptors [27,28] and GFP, and analyzed their phenotypic characteristics (CD3/CD56 and chimeric receptor expression) and cytotoxic activity against B-ALL cells in a 51Cr-release assay. Percentage of CD3þ/ CD56þ was not altered by transduction (data not shown).

CIK cells were transduced with the retroviral vectors with a mean expression of GFP/chimeric receptor 5 15% for both the vectors used (n 5 8; Fig. 3A and B). Furthermore, only CIK cells expressing anti-CD19-z showed a strong and significant cytotoxic activity against the REH B-ALL cell line (mean lysis, 59.9% at an E:T ratio 5 30:1, range, 45.769.5%; n 5 3; Fig. 4A) and against primary B-ALL blasts (mean lysis, 60.2% at E:T ratio of 30:1, range, 39.772.8%; n 5 3; Fig. 4B). As CIK cells are a heterogeneous population of cells, constituted by CD3þCD56þ and CD3þCD56 cells, we have investigated the susceptibility to retroviral infection and the cytotoxic activity of transduced cells of both subpopulations. CD3þCD56þ cells showed a significantly higher percentage (n 5 6; Fig. 5A) of cells expressing GFP/chimeric receptor than CD3þCD56 cells. We also analyzed the cytotoxic activity of transduced positive and negative fractions after CD56 immunoselection in two


1223

Figure 3. Expression of green fluorescent protein (GFP)/chimeric receptor on transduced cytokine-induced killer (CIK) cells. (A) Mean expression of GFP/ anti-CD19 chimeric receptors on CIK cells after retroviral transduction. Data shown are mean SD of eight separate experiments. (B) Surface receptor expression was detected by flow cytometry after staining with a goat anti-mouse (Fab)2 polyclonal antibody conjugated with biotin followed by TC-conjugated streptavidin (Y axes); expression of GFP is also shown (X axes). A representative phenotype of anti-CD19 chimeric receptor expression is shown.

donors, and we observed that CD3þCD56þ cells were more cytotoxic against B-ALL blasts than CD3þCD56 (Fig. 5B). It has been shown by others [51] that certain chemokines enhance the cytotolytic activity of NK cells. We tested this hypothesis with transduced CIK cells for all the set of chemokines used in the chemotaxis assay. In no case the B-ALL killing was improved by the presence of the chemokines (data not shown).

To verify the potential exploitability of chimeric-receptor-modified CIK cells in B-ALL immunotherapy, we extensively characterized their trafficking machinery. For this purpose, we analyzed if the transduction altered their chemotactic response to CXCL12 and CCL19. Moreover, we investigated if transduced CIK cells were able to adhere and transmigrate through an endothelium layer and to migrate through reconstituted basement membrane (Matrigel) in response to CXCL12.

Figure 4. Cytotoxic activity of anti-CD19 chimeric receptor-modified cytokine-induced killer (CIK) cells. Cytotoxic activity of transduced CIK cells was evaluated by a standard 4 hours 51Cr release assay after 21 days of culture. Only CIK cells expressing the anti-CD19-z chimeric receptor showed a significant cytotoxic activity against the REH B-lineage acute lymphoblastic leukemia (B-ALL) cell line (A) and primary B-ALL blasts (B). Data are mean SD of three replicates of three independent experiments (**p ! 0.01 and *p ! 0.05 vs vector and anti-CD19-trunc).

1224


Figure 5. Chimeric receptor expression and functional characterization of transduced CD3þCD56þ and CD3þCD56 cytokine-induced killer (CIK) cells. (A) Mean expression of green fluorescent protein (GFP)/anti-CD19 chimeric receptor on gated CD3þCD56þ and CD3þCD56 transduced CIK cells. Data are mean SD of six donors. CD3þCD56þ CIK cells express significantly (*p ! 0.05) higher levels of anti-CD19 chimeric receptor than CD3þCD56 CIK cells. (B) Anti-CD19 chimeric receptor-modified CD3þCD56þ showed higher and, in one donor out of two, significant (**p ! 0.01) cytotoxicity than CD3þCD56 cells (two donors were evaluated). Data are mean SD of three replicates.

CIK cells were significantly attracted by CXCL12 and CCL19, with levels similar to nontransduced cells (Fig. 6A). These data were confirmed after evaluation of the percentage of chimeric receptor-expressing cells in the migrated population (Fig. 6B), which was the same as in the population before migration for both vectors and for both CXCL12 and CCL19. Furthermore, CIK cells showed substantial adhesion and transmigration through an endothelium layer in response to CXCL12 (Fig. 6C), and they were capable to migrate through Matrigel (Fig. 6D). This property is related to the significant stimulation of gelatinase (matrix metalloproteinases MMP-2 and MMP-9) activity induced by CXCL12 in transduced cells (Fig. 6E), with no additional increase of secretion of the MMP-2 and MMP-9 inhibitors, tissue inhibitor of MMP (TIMP)-1 and TIMP-2, when compared with the basal condition without chemokine stimulation (data not shown).

Discussion To be optimal for in vivo killing of the specific tumor target, immune-effector cells must have the property to reach the tissues and organs where the tumor infiltrates. In this study, we demonstrated that CD3þCD56þ CIK cells have chemotactic properties analogous to those reported for freshly isolated NKT cells, indicating their potentiality to reach leukemia-infiltrated tissues. On the basis of these findings, we further and more extensively characterized the trafficking machinery of CIK cells that have been genetically modified with an anti-CD19 chimeric receptor able to confer them a strong cytotoxic activity against otherwise resistant B-ALL cells.

The recruitment of leukocytes into tissues is dependent on a series of adhesive and activation steps mediated by adhesion molecules. It has been demonstrated that the migration of NKT cells into bone marrow is mediated by the interaction of CD49d and vascular adhesion molecule-1 [52], constitutively expressed on bone marrow endothelium, whereas the migration and/or retention of the majority of NKT cells in the liver is CD11a-dependent [23]. Furthermore, CD62L is required for extravasation of a small population of NKT cells into lymph nodes [53,54]. In line with these observations, we found that CD3þCD56þ CIK cells expressed high levels of CD49d and CD11a and low levels of CD62L, thus implying their potential capacity to adhere to the bone marrow and nonlymphoid endothelium. The trafficking of immune cells is also regulated by chemokines and their corresponding cell-surface receptors [55]. It has been shown that both CXCR4 and CCR6 are expressed on CD1d-restricted and CD1d-unrestricted NKT cells [41,42,56] and are crucial to guide their homing into the bone marrow and extralymphoid tissues. CCR7, although found on a minority of NKT cells (around 20%) [41], is relevant in the process of migration into the spleen and lymph nodes [44]. We found that CD3þCD56þ CIK cells, similarly to NKT cells, present high expression of CXCR4. We also found that CCR6 [40] is expressed at significant levels on CD3þCD56þ CIK cells. The in vitro response to chemokines confirmed the pattern of expression of the specific chemokine receptors. CD3þCD56þ CIK cells migrated robustly when stimulated with CXCL12, CCL20, CCL19, and CCL21. The CD3þCD56 fraction showed similar pattern of expression of adhesion molecules and of the chemokine receptor CXCR4. On the contrary, CCR7 was highly


1225

Figure 6. Chemotactic activity of anti-CD19 chimeric receptor-modified cytokine-induced killer (CIK) cells. (A) Anti-CD19 chimeric receptor-modified CIK cells maintain their ability to migrate in vitro in response to CXCL12 and CCL19. The bars represent mean SD of four separate experiments. NT 5 nontransduced. (B) Percentage of anti-CD19 chimeric receptor-modified CIK cells in the migrated population was the same as in the population before migration (INPUT), for both CXCL12 and CCL19. (C) Anti-CD19 chimeric-receptormodified CIK cells strongly adhere and transmigrate through endothelial cells after stimulation with CXCL12. Data are mean SD of three independent experiments. (D) CXCL12 stimulates migration of anti-CD19 chimeric receptor-modified CIK cells through reconstituted basement membrane (Matrigel). Trans-Matrigel migration was assessed after 3 hours of incubation in the presence of CXCL12 (300 ng/mL) in the lower compartment. Data are mean SD of three independent experiments. (E) CXCL12 enhances the gelatinase activity of matrix metalloproteinase (MMP)-2 and MMP-9 metalloproteases in the supernatant collected from chemokine-stimulated transduced CIK cells. Human monocytes were used as positive control. Data are mean SD of three independent experiments (*p ! 0.05 vs control).

expressed and this could represent the explanation of the higher migration of the CD56 negative fraction in response to CCL19 and CCL21. Although the trafficking machinery makes CIK cells a potential optimal candidate for leukemia cell therapy, CIK cells do not exert any killing activity toward B-ALL cells. Transduction of chimeric receptors can be used to

redirect the cytotoxicity of immune cells [57,58]. We found that this strategy is also applicable to CIK cells. In this article we have explored the applicability of this approach, choosing the CD19 antigen, which is widely expressed on B-ALL cells. CIK cells are susceptible to retroviral transduction and, although the expression of the anti-CD19 receptor is lower than described in other cell types, it

1226


nevertheless triggers considerable antileukemia cytotoxicity. We noted differences among the transduced-CD56 positive and negative fractions. The former appear to be more susceptible to transduction and expressed higher levels of the anti-CD19 receptor, perhaps due to the higher proliferation rate [9]. The stronger expression of the receptor could also be responsible for the greater cytotoxic activity of the CD56þ fraction when compared to that of CD56L cells. To verify if chimeric-receptormodified CIK cells are suitable for B-ALL immunotherapy, we extensively characterized their in vitro trafficking machinery, to confirm the high migratory capacities toward leukemia-infiltrated tissues previously observed for unmanipulated CIK cells. Firstly, we showed that transduction did not alter their chemotactic response to CXCL12 and CCL19, which are involved in cell recruitment to tissue sites where B-ALL cells proliferate and infiltrate mostly [24–26]. Secondly, our experiments demonstrated that transduced CIK cells, after stimulation with CXCL12, were able to adhere and transmigrate through an endothelium layer and to migrate through a reconstituted basement membrane, which is the first essential step for any effector cell to reach the specific extravascular tumor target. Indeed, the gelatinase activity was significant, indicating that transduced CIK cells produce the metalloproteases MMP-2 and MMP-9, whose role is relevant in the process of extracellular matrix digestion, implicated in tissue invasion [59]. Moreover, no induction of secretion of MMP-2 and MMP-9metalloproteases inhibitors, TIMP-1 and TIMP-2 [59], was detected. Anti-CD19 chimeric receptorexpressing CIK cells can therefore potentially be able in vivo to migrate into tissues were leukemia cells preferentially localize, and then be able to kill the tumor target. It is relevant that CD1d-unrestricted NKT cells secrete substantial amount of CC and CXC chemokines active toward various immune-effector cells upon Fas engagement [21]. Because Fas ligand is highly expressed on B-ALL cells [60,61], contact between anti-CD19 chimeric receptor-modified CIK cells and leukemic blasts could result not only in direct leukemia cell killing, but also in the activation of CIK cells, chemokines secretion and recruitment of different immune-effector cells, further amplifying the antileukemia response. From a clinical application standpoint, CIK cells are attractive because of the reproducible and straightforward method for their generation and expansion, which only requires good manufacture practice-grade cytokines [62]. Thus, a large numbers of cells can be rapidly expanded in a close system with minimal manipulation. For this particular reason, use of CIK cells has been already investigated in phase I trials, with the aim of exploring their toxicity and clinical efficacy [63–65]. Although our findings need to be finally tested in an in vivo model, the presented data suggest that anti-CD19 chimeric receptor-modified CIK cells, besides acquiring

the capacity to powerfully kill B-ALL cells, show a trafficking machinery that indicate their potentiality to reach in vivo sites of leukemia infiltration, therefore, representing an interesting tool for B-ALL immunotherapy.

Acknowledgments This work was supported in part by grants from the Associazione Italiana Ricerca sul Cancro (AIRC), the Fondazione ‘‘M. Tettamanti’’ and the Comitato ‘‘S. Verri.’’ V. Marin is a fellow of the Vita-Salute San Raffaele University doctoral program in molecular medicine. We thank Dr. Dario Campana (St. Jude Children’ Research Hospital, Memphis, TN) for providing us with the antiCD19-z-IRES-GFP, anti-CD19-truncated-IRES-GFP and pMSCVIRES-GFP constructs and for helpful comments on our article.

References 1. Schmidt-Wolf IGH, Negrin RS, Kiem H, Blume KG, Weissman IL. Use of a SCID mouse/human lymphoma model to evaluate cytokine-induced killer cells with potent antitumor cell activity. J Exp Med. 1991;174:139–149. 2. Verneris MR, Baker J, Edinger M, Negrin RS. Studies of ex-vivo activated and expanded CD8þ NKT cells in humans and mice. J Clin Immunol. 2002;22:131–136. 3. Alvarnas JC, Linn YC, Hope EG, Negrin RS. Expansion of cytotoxic CD3þCD56þ cells from peripheral blood progenitor cells of patients undergoing autologous haematopoietic cell transplantation. Biol Blood Marrow Transplant. 2001;7:216–222. 4. Lu PH, Negrin RS. A novel population of expanded human CD3þCD56þ cells derived from T cells with potent in vivo antitumor activity against leukemia and lymphoma cells by reverse antibody-dependent cellular cytotoxicity. J Immunol. 1994;153:1687–1696. 5. Hoyle C, Bangs CD, Chang P, Kamel O, Mehta B, Negrin RS. Expansion of Philadelphia chromosome-negative CD3þCD56þ cytotoxic cells from chronic myeloid leukemia patients: in vitro and in vivo efficacy in severe combined immunodeficiency disease mice. Blood. 1998;92:3318–3327. 6. Schmidt J, Eisold S, Bu¨chler MW, Ma¨rten A. Dendritic cells reduce number and function of CD4þCD25þ cells in cytokine-induced killer cells derived from patients with pancreatic carcinoma. Cancer Immunol Immunother. 2004;53:1018–1026. 7. Linn YC, Lau LC, Hui KM. Generation of cytokine-induced killer cells from leukemic samples with in vitro cytotoxicity against autologous and allogeneic leukaemic blasts. Br J Haematol. 2002;116:78–86. 8. Lefterova P, Schakowski F, Buttgereit P, Scheffold C, Huhn D, Schmidt-Wolf IG. Expansion of CD3þCD56þ cytotoxic cells from patients with chronic lymphocytic leukemia: in vitro efficacy. Haematologica. 2000;85:1108–1109. 9. Schmidt-Wolf IGH, Lefterova P, Mehta BA, et al. Phenotypic characterization and identification of effector cells involved in tumor cell recognition of cytokine-induced killer cells. Exp Hematol. 1993;21: 1673–1679. 10. Verneris MR, Ito M, Baker J, Arshi A, Negrin RS, Shizuru JA. Engineering hematopoietic grafts: purified allogeneic haematopoietic stem cells plus expanded CD8þ NKT cells in the treatment of lymphoma. Biol Blood Marrow Transpl. 2001;7:532–542. 11. Godfrey DI, Hammond KJ, Poulton LD, Smyth MJ, Baxter AG. NKT cells: facts, functions and fallacies. Immunol Today. 2000;21:573–583. 12. Bendelac A, Rivera MN, Park SH, Roark JH. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu Rev Immunol. 1997;15:535–562.

V. Marin et al./ Experimental Hematology 34 (2006) 1218–1228 13. Nicol A, Nieda M, Koezuka Y, et al. Human invariant Va24þ natural killer T cells activated by a a-galactosylceramide (KRN7000) have cytotoxic anti-tumor activity through mechanisms distinct from T cells and natural killer cells. Immunology. 2000;99:229–234. 14. Prussin C, Foster B. TCR Va24 and Vb11 coexpression defines a human NK1 T cell analog containing a unique Th0 subpopulation. J Immunol. 1997;159:5862–5870. 15. Yoshimoto T, Paul WE. CD4 pos, NK1.1 pos T cells promptly produce interleukin 4 in response to in vivo challenge with anti-CD3. J Exp Med. 1994;179:1285–1295. 16. Hammond KJ, Pelikan SB, Crowe NY, et al. NKT cells are phenotypically and functionally diverse. Eur J Immunol. 1999;29:3768–3781. 17. Exley M, Porcelli S, Furman M, Garcia J, Balk S. CD161 (NKR-P1A) costimulation of CD1d-dependent activation of human T cells expressing invariant Va24JaQ T cell receptor a chains. J Exp Med. 1998;188: 867–876. 18. Eberl G, Lees R, Smiley ST, Taniguchi M, Grusby MJ, MacDonald HR. Tissue-specific segregation of CD1d-dependent and CD1d-independent NK T cells. J Immunol. 1999;162:6410–6419. 19. Stremmel C, Exley M, Balk S, Hohenberger W. Kuchroo VK Characterization of the phenotype and function of CD8(þ) alpha/beta (þ) NKT cells from tumor-bearing mice that show a natural killer cell activity and lyse multiple targets. Eur J Immunol. 2001;31:2818–2828. 20. Baker J, Verneris MR, Ito M, Shizuru JA, Negrin RS. Expansion of cytolytic CD8(þ) natural killer T cells with limited capacity for graft-versus-host disease induction due to interferon gamma production. Blood. 2001;97:2923–2931. 21. Giroux M, Denis F. CD1d-unrestricted human NKT cells release chemokines upon Fas engagement. Blood. 2005;105:703–710. 22. van der Vliet HJ, Molling JW, von Blomberg BM, et al. The immunoregulatory role of CD1d-restricted natural killer T cells in disease. Clin Immunol. 2004;112:8–23. 23. Emoto M, Kauffmann SHE. Liver NKT cells: an account of heterogeneity. Trends Immunol. 2003;24:364–369. 24. Watanabe S, Shimosato Y, Kameya T, et al. Leukemic distribution of a human acute lymphocytic leukemia cell line (Ichikawa strain) in nude mice conditioned with whole body irradiation. Cancer Res. 1978;38:3494–3498. 25. Crazzolara R, Kreczy A, Mann G, et al. High expression of the chemokine receptor CXCR4 predicts extramedullary organ infiltration in childhood acute lymphoblastic leukemia. Br J Haematol. 2001;115:545–553. 26. Lock RB, Liem N, Farnsworth ML, et al. The nonobese diabetic/sever combined immunodeficient (NOD/SCID) mouse model of childhood acute lymphoblastic leukemia reveals intrinsic differences in biological characteristics at diagnosis and relapse. Blood. 2002;99:4100– 4108. 27. Imai C, Mihara K, Andreansky M, et al. Chimeric receptors with 4-1BB signaling capacity provoke potent cytotoxicity against acute lymphoblastic leukemia. Leukemia. 2004;14:676–684. 28. Imai C, Iwamoto S, Campana D. Genetic modification of primary natural killer cells overcomes inhibitory signals and induces specific killing of leukemic cells. Blood. 2005;106:376–383. 29. Montagna D, Maccario R, Locatelli F, et al. Ex vivo priming for longterm maintenance of antileukemia human cytotoxic T cells suggests a general procedure for adoptive immunotherapy. Blood. 2001;98: 3359–3366. 30. Del Prete A, Vermi W, Dander E, et al. Defective dendritic cell migration and activation of adoptive immunity in PI3Kgamma-deficient mice. EMBO J. 2004;23:3505–3515. 31. Manabe A, Coustan-Smith E, Behm FG, et al. Bone marrow-derived stromal cells prevent apoptotic cell death in B-lineage acute lymphoblastic leukemia. Blood. 2002;79:2370–2377. 32. Bianchi G, Sironi M, Ghibaudi E, et al. Migration of natural killer cells across endothelial cell monolayers. J Immunol. 1993;151: 5135–5144.

1227

33. Zheng Y, Kong Y, Goetzl EJ. Lysophosphatidic acid-receptor-selective effects on Jurkat T cell migration through a Matrigel model basement membrane. J Immunol. 2001;166:2317–2322. 34. Introna M, Barbui AM, Golay J, et al. Rapid retroviral infection of human haemopoietic cells of different lineages: efficient transfer in fresh T cells. Br J Haematol. 1998;103:449–461. 35. Rose DM, Han J, Ginsberg MH. Alpha4 integrins and the immune response. Immunol Rev. 2002;186:118–124. 36. Frenette PS, Subbarao S, Mazo IB, von Adrian UH, Wagner DD. Endothelial selectins and vascular adhesion-molecule-1 promote hematopoietic progenitor homing to bone marrow. Proc Natl Acad Sci. 1998; 95:14423–14428. 37. Stamenkovic I. The L-selectin adhesion system. Curr Opin Hematol. 1995;2:68–75. 38. Nagasawa T. A chemokine, SDF-1/PBSF, and its receptor, CXC chemokine receptor 4, as mediators of hematopoiesis. Int J Hematol. 2000;72:408–411. 39. Lapidot T. Mechanism of human stem cell migration and repopulation of NOD/SCID and B2mnull NOD/SCID mice. The role of SDF-1/CXCR4 interactions. Ann N Y Acad Sci. 2001;938:83–95. 40. Schutyser E, Struyf S, Van Damme J. The CC chemokine CCL20 and its receptor CCR6. Cytokine Growth Factor Rev. 2003;14: 409–426. 41. Kim CH, Johnston B, Butcher EC. Trafficking machinery of NKT cells: shared and differential chemokine receptor expression among Va24þb11þ NKT cell subsets with distinct cytokine-producing capacity. Blood. 2002;100:11–16. 42. Campbell JJ, Qin S, Unutmaz D, et al. Unique subpopulation of CD56þ NK and NK-T peripheral blood lymphocytes identified by chemokine receptor expression repertoire. J. Immunol. 2001;166: 6477–6482. 43. Birkenbach M, Josefsen K, Yalamanchili R, Lenoir G, Kieff E. Epstein-Barr virus-induced genes: first lymphocyte-specific G proteincoupled peptide receptors. J Virol. 1993;67:2209–2220. 44. Muller G, Lipp M. Shaping up adaptive immunity: the impact of CCR7 and CXCR5 on lymphocyte trafficking. Microcirculation. 2003;10:325–334. 45. Peled A, Kollet O, Ponomaryov T, et al. The chemokine SDF-1 activates the integrins LFA-1, VLA-4 and VLA-5 on immature human CD34(þ) cells: role in transendothelial/stromal migration and engraftment of NOD/SCID mice. Blood. 2000;95:3289–3296. 46. Ponomaryov T, Peled A, Petit I, et al. Induction of the chemokine stromal-derived factor-1 following DNA damage improves human stem cell function. J Clin Invest. 2000;106:1331–1339. 47. Yoshida R, Imai T, Hieshima K, et al. Molecular cloning of a novel human CC chemokine EBI1-ligand chemokine that is a specific functional ligand for EBI1, CCR7. J Biol Chem. 1997;272:13803– 13809. 48. Kim CH, Pelus LM, White JR, Applebaum E, Johanson K, Broxmeyer HE. CK beta-11/macrophage inflammatory protein-3 beta/EBI1ligand chemokine is an efficacious chemoattractant for T and B cells. J Immunol. 1998;160:2418–2224. 49. Gunn MD, Tangemann K, Tam C, Cyster JG, Rosen SD, Williams LT. A chemokine expressed in lymphoid high endothelial venules promotes the adhesion and chemotaxis of naı¨ve T lymphocytes. Proc Natl Acad U S A. 1998;95:258–263. 50. Hedrick JA, Zlotnik A. Identification and characterization of a novel beta chemokine containing six conserved cysteines. J Immunol. 1997;159:1589–1593. 51. Robertson MJ. Role of chemokines in the biology of natural killer cells. J Leukoc Biol. 2002;71:173–183. 52. Franitza S, Grabovsky V, Wald O, et al. Differential usage of VLA-4 and CXCR4 by CD3þCD56þ NKT cells and CD56þCD16þ NK cells regulates their interaction with endothelial cells. Eur J Immunol. 2004; 34:1333–1341.

1228


53. Spertini O, Luscinskas FW, Kansas GS, et al. Leukocyte adhesion molecule-1 (LAM-1, L-selectin) interacts with an inducible endothelial cell ligand to support leukocyte adhesion. J Immunol. 1991;147:2565–2573. 54. Kansas GS, Ley K, Munro JM, Tedder TF. Regulation of leukocyte rolling and adhesion to high endothelial venules through the cytoplasmic domain of L-selectin. J Exp Med. 1993;177:833–838. 55. Mantovani A. The chemokine system: redundancy for robust outputs. Immunol Today. 1999;20:254–257. 56. Kim CH, Butcher EC, Johnston B. Distinct subsets of human Va24-invariant NKT cells: cytokine responses and chemokine receptor expression. Trends Immunol. 2002;11:516–519. 57. Schumacher TN. T-cell-receptor gene therapy. Nat Rev Immunol. 2002;2:512–519. 58. Sadelain M, Riviere I, Bretjens R. Targeting tumours with genetically enhanced T lymphocytes. Nat Rev Cancer. 2003;3:35–45. 59. Janowska-Wieczorek A, Marquez LA, Dobrowsky A, et al. Differential MMP and TIMP production by human marrow and peripheral blood CD34þ cells in response to chemokines. Exp Hematol. 2000;28:1274–1285. 60. Volm M, Zintl F, Sauerbrey A, Koomagi R. Expression of Fas ligand in newly diagnosed childhood acute lymphoblastic leukemia. Anticancer Res. 1999;19:3399–3402.

61. Lickliter JD, Kratzke RA, Nguyen PL, Niehans GA, Miller JS. Fas ligand is highly expressed in acute leukemia and during the transformation of chronic myeloid leukemia to blast crisis. Exp Hematol. 1999;27:1519–1527. 62. Introna M, Borleri GM, Conti E. Ex vivo generation and characterization of cytokine induced killer cells in GMP conditions. Paper presented at the Annual International Society of Cell Therapy Meeting, Vancouver, Canada, May 4, 2005. 63. Schmidt-Wolf IG, Finke S, Trojaneck B, et al. Phase I clinical study applying autologous immunological effector cells transfected with the interleukin-2 gene in patients with metastatic renal cancer, colorectal cancer and lymphoma. Br J Cancer. 1999;81:1009– 1016. 64. Shi M, Zhang B, Tang ZR, et al. Autologous cytokine-induced killer cell therapy in clinical trial phase I is safe in patients with primary hepatocellular carcinoma. World J Gastroenterol. 2004;10:1146– 1151. 65. Leemhuis T, Wells S, Scheffold C, Edinger M, Negrin RS. A phase I trial of autologous cytokine-induced killer cells for the treatment of relapsed Hodgkin disease and non-Hodgkin lymphoma. Biol Blood Marrow Transplant. 2005;11:181–187.