| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Service of Transfusion Medicine, S. Orsola-Malpighi Hospital, Bologna, Italy;
Department of Experimental Pathology and
Institute of Haematology and Medical Oncology "Seràgnoli," University of Bologna, Bologna, Italy;
Immunogenetics Laboratory, National Cancer Research Institute and Department of Oncology, Biology, and Genetics c/o Advanced Biotechnology Center, Genoa, Italy; and
¶ Department of Internal Medicine, University of Turin, Turin, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Targeting of drugs to kill selected cell populations might be an important tool in immunotherapy and oncology. In particular, immunotoxins, conjugates containing an Ab and a toxic substance, still could represent an important weapon for this purpose, although general toxicity, alloimmunization against components, and difficulties in solid tumor targeting are limits to a general application of these tools (8).
Ribosome-inactivating proteins (RIPs) are N-glycosylases from plants that cleave one or more adenine molecules from ribosomal RNA, thus damaging ribosomes in an irreversible manner. RIPs may be divided into type 1 or type 2 according to their single- or double-chain structure (9).
A series of immunotoxins containing different Abs and type 1 RIPs was previously described (10, 11, 12, 13, 14, 15, 16). One of these immunotoxins (CD30-saporin) also gave encouraging results in a phase I clinical trial (17).
In this study, we describe the in vitro and in vivo results obtained with an immunotoxin consisting of saporin, a type 1 RIP from the seeds of Saponaria officinalis (9), chemically linked to recombinant human mAbs against CTLA-4, generated by selecting a synthetic phage single-chain fragment variable (scFv) library (18, 19). Fv fragments are the smallest functional units of Abs required to maintain the binding and the specificity of the whole Ab. Fragment variables can be produced as single chains (scFvs) by recombinant techniques. In these molecules, VH and VL domains are joined by a peptide linker to form a single fusion protein (20). In our study, two distinct scFvs were used, namely scFv 83 and 40, both recognizing an epitope on the CD152 dimer (18).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A human scFv phage library (18, 19) was used to generate rCTLA-4 Abs that were selected and characterized, as already described (18). Specificity was checked by competitive ELISA, Western blot, and immunofluorescence (18). In particular, two scFvs, namely 83 and 40, were further selected on the basis of their reactivity with T cells, both resting and activated (via CD3/CD28) (18). The positivity of the cells was checked by immunofluorescence and analyzed by flow cytometry (EPICS XL; Coulter Immunotech, Hialeah, FL): in particular for scFvs and immunotoxins, reactivity was checked by indirect immunofluorescence, as already described (11, 14, 15, 18). Appropriate negative (mouse IgG; Coulter Immunotech) and positive (CTLA-4 BN13 mAb; Coulter Immunotech) controls were used. The 83 and 40 scFvs were also conjugated to FITC and used for direct staining, following standard procedures. In particular, cell lines were stained by direct immunofluorescence technique, and CD34/scFv double staining was performed using FITC-conjugated 83 and 40 Abs and a PE-conjugated CD34 (CD34-PE, 581 clone; Coulter Immunotech).
Immunotoxins
Type 1 single-chain RIP saporin S6 (hereupon referred to as saporin) from the seeds of Saponaria officinalis (Azienda Regionale delle Foreste della Regione Emilia Romagna, Casola Valsenio, Ravenna, Italy) was purified essentially as described by Barbieri et al. (21). The RIP was labeled with 125I with the Iodogen reagent (Pierce, Rockford, IL), according to manufacturer’s instructions.
The RIP and the scFv were conjugated via a disulfide bond between chemically inserted sulfhydryl groups. ScFvs 83 and 40 were modified using N-succimidyl-3-(2-pyridyldithio) propionylate (SPDP; Sigma, St. Louis, MO) (22) at a 14:1 SPDP:scFv molar ratio. After 30 min at 28°C, the modified protein was separated from unreacted reagent by a gel filtration on a Sephadex G25 column (30 cm x 1.6 cm) (Pharmacia Biotech, Uppsala, Sweden), equilibrated, and eluted with PBS (0.14 M NaCl in 5 mM sodium phosphate buffer, pH 7.4). The number of sulfhydryl groups introduced was determined by the A343 prior to and after reduction with 20 mM 2-ME, as described by Carlsson et al. (23).
Saporin, containing a trace of 125I-labeled RIP, was dissolved in 50 mM sodium borate buffer, pH 9, at a concentration of 7 mg/ml; the RIP was subsequently modified by adding 2-iminothiolane (2-IT; Sigma) to a final concentration of 1 mM and purified following already published procedures (15). The number of sulfhydryl groups introduced was determined by the A412 prior to and after reduction with 20 mM 2-ME, as described by Ellman (24).
The modified RIP was reduced with 20 mM 2-ME, filtered as above through a Sephadex G25 column, and collected directly onto the unreduced derivatized scFv. The RIP/Ab mixture, in a 10:1 molar ratio, was allowed to react for 16 h at room temperature.
The resulting conjugate was separated from the unreacted reagents and from RIP homopolymers by affinity chromatography with HiTrap rProtein A-Sepharose fast flow column (1 ml; Pharmacia Biotech), equilibrated with PBS, washed with 0.5 M NaCl in 5 mM sodium phosphate buffer, pH 7.4, and eluted with 0.6% glycine in 20 mM HCl, pH 3. Fractions of 1 ml each were collected onto 200 µl of 1 M Tris-HCl, pH 9. The immunotoxin was separated from the unconjugated Ab by gel filtration on a Sephacryl S200 high-resolution column (100 cm x 2.5 cm; Pharmacia Biotech), equilibrated and eluted with PBS. The RIP:Ab ratio of the immunotoxin was estimated by the 125I-labeled RIP radioactivity and by the protein concentration calculated from the A280.
Protein synthesis-inhibitory activity of the pooled fractions was assayed on a rabbit reticulocyte lysate, as described below. The immunotoxins were analyzed by SDS-PAGE with a PhastGel Gradient 10-15 (Pharmacia Biotech) under nonreducing conditions, loading 1 µg of protein in 1 µl of sample buffer containing 10 mg/ml iodoacetamide, as described by Lambert et al. (25).
The conjugates were sterilized by filtering through a 0.22-µm filter, divided into aliquots, and stored in liquid nitrogen at 1 µM concentration, as RIP.
Cell-free protein synthesis inhibition assays
The inhibitory activity of immunotoxin on cell-free protein synthesis was evaluated with a rabbit reticulocyte lysate, prepared as described by Allen and Schweet (26). Immunotoxins were prior reduced with 20 mM 2-ME for 30 min at 37°C, appropriately diluted, and then added to a reaction mixture, as already described (15, 27). Each experiment was conducted in duplicate. The IC50 was calculated by linear regression analysis (15).
Target cell models
Immunotoxins were tested in different experimental models, including 1) T cells activated via CD3/CD28; 2) MLR with B lymphoblastoid cell lines as allogeneic target; 3) MLR with monocyte-derived dendritic cells (DC); and 4) human lymphoid and myeloid cell lines.
Activated T cells via CD3/CD28. T cells were isolated from PBL (obtained from healthy random donors after informed consent) either by E-rosetting or by MACS CD3 microbeads and magnetic separation columns (Miltenyi Biotec, Bergisch Gladbach, Germany). Purity was always >95%, as checked by direct immunofluorescence with FITC-CD3 and FITC-CD2 mAbs (Coulter Immunotech). T cells were then activated via CD3/CD28 (1 µg/ml each; Coulter Immunotech) in RPMI 1640 medium containing 10% FCS, and checked for surface positivity with murine mAb BN13 (anti-CTLA-4 mAb used as positive control) (28) and with FITC 83 and FITC 40. Resting T lymphocytes were used as negative controls. Cells were cultured for 72 h in the presence of immunotoxins, or separated components, mixed or as single reagents.
MLR with EBV-infected lymphoblastoid B cell line (BM14). Standard MLR was performed by using PBL from random healthy donors (after informed consent) and a lymphoblastoid EBV B cell line (BM14) irradiated with 100 Gy in a 100:1 ratio (100,000 effector cells to 1,000 target cells/well in a 96-well round-bottom plate). Cells were cultured for 6 days in a 200-µl final volume of complete medium.
MLR with monocyte-derived DC. DC were generated as follows: mononuclear cells were obtained by gradient centrifugation (Lymphoprep; 1.077 g/ml; Nycomed Pharma, Oslo, Norway) from whole blood of random healthy donors (after informed consent). Light-density cells were washed twice in PBS with 1% BSA (Sigma), and CD14+ cells were highly purified from mononuclear cell fraction by MiniMacs high-gradient magnetic separation column (Miltenyi Biotec), according to the manufacturer’s instructions. Flow cytometry reanalysis of CD14+ purified cell fractions was performed on a gated population set on scatter properties by using EPICS XL equipment, as earlier described (29). A minimum of 10,000 events was collected in list mode.
DC were obtained from peripheral blood CD14+
cells, as previously described (29), with some
modifications. Briefly, 1 x 106 purified
CD14+ cells were cultured for 5–7 days in 1 ml
of RPMI 1640 supplemented with 10% FCS (Sera Lab, Crawley Down,
Sussex, U.K.), antibiotics, L-glutamine, 50 ng/ml GM-CSF
(Sandoz, Basel, Switzerland), and 50 ng/ml IL-4 (Genzyme, Cambridge,
MA). When indicated, 25 ng/ml TNF-
(Innogenetics, Zwijndrecht,
Belgium) was added to the culture for 24–36 h to induce terminal
maturation of DC (29). Cultures were maintained at 37°C
in 5% CO2 by replacing culture medium and
cytokines at day +3. To check the percentage of DC, dual-color
immunofluorescence was performed using the following panel of mAbs: PE
or FITC anti-human CD1a (BD PharMingen, San Diego, CA); FITC
anti-human CD86 (BD PharMingen); FITC BB1/B7 (anti-CD80; BD
Biosciences, San Jose, CA); FITC anti-human HLA-DR (BD
Biosciences); FITC or PE Leu-M3 (anti-CD14; BD Biosciences); FITC
Leu-4 (anti-CD3; BD Biosciences); PE anti-human CD83 (Coulter
Immunotech); FITC anti-human CD40 (BD PharMingen); and FITC
anti-human Leu-3a (anti-CD4; BD Biosciences). Negative controls
were isotype-matched irrelevant mAbs (BD PharMingen and BD
Biosciences). Cells were incubated in the dark for 30 min at 4°C in
PBS containing 1% BSA. After washing, cells were resuspended in PBS
containing 1% paraformaldehyde and analyzed, as reported above.
To test their allogeneic stimulatory activity, DC were irradiated (30 Gy) and tested as stimulators in primary MLR.
Human cell lines. Human-derived cell lines were analyzed for surface positivity of 83 and 40 scFv, and of BN13 anti-CTLA-4 reference mAb (28). The following cell lines were evaluated by immunofluorescence: Daudi and Raji, derived from a Burkitt lymphoma; CEM, Jurkat, and Molt 4, derived from T lymphoblastic cells; L428, a B cell line derived from a Hodgkin’s lymphoma; HOM-2, a B lymphoblastoid cell line; K562, an erythroleukemia cell line; HL60, a myeloid-leukemia-derived cell line. Positive cells were used to test the toxicity of immunotoxins, as described below.
Moreover, RNA analysis of CTLA-4 extracellular domain was compared in HL60, K562, and HOM-2 and in resting and CD3/CD28-activated PBLs. The RT-PCR was performed using specific primers (forward primer, 5'-ATGCACGTGGCCCAGCCTGCTG-3'; reverse primer, 5'-TCTTAGACCCGTGCCAAGACCTA-3') (30). Total RNA was prepared using 500 µl of RNA clean reagent (Hybaid-AGS, Heidelberg, Germany) for 2 x 106 cells. The obtained RNA was quality checked by electrophoresis in a denaturing 1.5% agarose gel; cDNA was prepared from 5 µg of total RNA with a Retroscript kit (Ambion, Austin, TX), following manufacturer’s recommendations. PCR was conducted in a 50-µl volume, using 1/10 of the reverse-transcriptase mixture (500 ng of RNA). Notably, the PCR conditions were designed to maintain the reaction efficiency in logarithmic phase. The intensity of electrophoretic bands with background subtraction was measured in a Bio-Rad densitometer (type Gel Doc 1000; Bio-Rad, Hercules, CA), and the adjusted volume was calculated for each band. All specific PCR products were confirmed by direct sequence analysis with an automated sequencer (ABI PRISM 377; PerkinElmer, Santa Clara, CA).
Toxicity tests
MLRs. Effector T cells and target DC (100:1) were resuspended in RPMI 1640, 25 mM HEPES, antibiotics, and 15% AB human serum that had been inactivated at 56°C for 30 min. Allogeneic T cells (5 x 104) were mixed with stimulators (DC) in round-bottom 96-well plates for 6 days at 37°C in a 5% CO2 humidified atmosphere, as described above for MLR with EBV-infected lymphoblastoid B cells (BM14), in the presence of immunotoxins, or separated components, mixed or single.
Cells were pulsed with 1 µCi/well [3H]TdR (Pharmacia Biotech) on day 6, and after another 18 h were harvested with an automatic cell harvester (Skatron Instruments, Lier, Norway) onto glass-fiber diskettes. The radioactivity incorporated was measured with a beta counter (Beckman, Fullerton, CA). Each experiment, performed three times, was conducted in triplicates. IC50 was calculated by linear regression analysis.
Inhibition of protein synthesis in activated T lymphocytes and cell lines. The effects of the immunotoxins on CD3/CD28-activated T lymphocytes and on cell lines were evaluated from the inhibition of [3H]leucine incorporation. Cells were harvested, checked for viability, and adjusted to a concentration of 105 cells/ml in complete RPMI 1640 medium. Cells (104/well) were seeded in 96-well microtiter plates (Falcon; BD Biosciences) in a 100-µl volume, and 100 µl of immunotoxin was added to final concentrations ranging from 10-11 to 10-7 M, as RIP. Parallel samples were run with RIP alone, anti-CTLA-4 mAb alone, and a mixture of unconjugated Ab and RIP. After 72 h, 74 kBq L-[4,5-3H]leucine (Pharmacia Biotech) was added, and after another 18 h, cells were harvested with an automatic cell harvester (Skatron Instruments) onto glass-fiber diskettes. The radioactivity incorporated was measured with a beta counter (Beckman). Each experiment, performed three times, was conducted in triplicate. IC50 was calculated by linear regression analysis.
Evaluation of apoptosis by FITC-Annexin V binding and propidium iodide staining. In a series of experiments performed on CD3/CD28-stimulated lymphocytes and on a cell line (HOM-2), viable and dead cells were evaluated by a double staining via FITC Annexin V and propidium iodide with a kit purchased from Bender Medsystem (Wien, Austria), following manufacturer’s instructions. Cells incubated with different concentrations of immunotoxins, RIP, scFv Abs, or a mix of RIP and scFv Abs were cultured in complete RPMI 1640 medium for 72 h. Analysis was performed by flow cytometry, calculating both the percentage of apoptotic (annexin V+/propidium iodide-) and necrotic (annexin V+/propidium iodide+) cells. The concentration of immunotoxin causing apoptosis in 50% of the cells (AC50) was calculated by linear regression analysis.
Toxicity to hemopoietic precursors. Tests were performed using CD34+ cells purified from normal donors, as already described (31). Briefly, mononuclear cells were separated onto Ficoll-Hypaque gradient (Lymphoflot; Biotest, Dreieich, Germany). CD34+ cells were then purified with MACS CD34- microbeads and magnetic separation columns. Purity was assessed by flow cytometry with a CD34-FITC mAb (Coulter Immunotech) and propidium iodide dye exclusion for dead cells. For colony assay, 5 x 103 cells were seeded in semisolid medium consisting of 1 ml of IMDM supplemented with 24% FCS, 0.8% BSA, 10-4 M 2-ME, 2 U of human recombinant erythropoietin (Dompè Biotec, Milan, Italy), and 2 x 10-4 M bovine hemin. To measure the optimum clonogenic efficiency, 10% (v/v) of a selected batch of a PHA-lymphocyte-conditioned medium was added. ScFv 83-saporin conjugate, scFv 83, saporin, or a mixture of the two was added to the cultures at 10-8 and 10-9 M concentrations. Methylcellulose final concentration was 1.1%. Granulocyte-macrophage CFU and burst-forming unit (BFU) erythroid were scored after 14 days of incubation at 37°C in a fully humidified 5% CO2 atmosphere. The clonogenic efficiency of CD34+ cells was 7.3 ± 3.1%.
Mouse in vivo model. To preliminarily verify the in vivo availability and activity of the conjugate, we used a mouse model consisting of implanted allogeneic cells, capable of giving rise to an acute immune rejection reaction with a T lymphoid infiltrate, as described by Garlanda et al. (32). Briefly, cells from the murine H.end endothelioma, able to induce a tumor when implanted s.c., were derived from DBA/2 mice. These cells are able to elicit an allogeneic response when injected in C57BL/6 mice, with a picture comparable with an acute rejection showing a T lymphoid cell inflammatory infiltrate. The endothelioma cells are quickly killed, while an immune suppression could diminish the reaction and prolong the survival of the neoplastic cells.
To set the model, these cells were cultured in DMEM (Irvine Scientific, Santa Ana, CA), supplemented with 2 mM glutamine (Life Technologies, Paisley, U.K.), 10% FCS (Irvine Scientific), and gentamicin, before the in vivo injection in C57BL/6 mice. Moreover, C57BL/6 PHA/PMA-activated T lymphocytes were matched with FITC 83 scFv. The reagent stained the activated T lymphocytes, thus indicating that 83-saporin immunotoxin effectively could target in vivo the activated mice lymphocytes infiltrating the allogeneic tumor.
For in vivo experiments, cells were gently detached from plates with EDTA, washed with PBS, counted in a microcytometer chamber, and resuspended in saline. Cells (107) in a total volume of 150 µl were injected s.c. into the left back of 8-wk-old female C57BL/6 mice using a 1-ml syringe with a 26-gauge needle at day 0. A group of four mice was injected i.p. at days 0, 1, and 2, with 4 µg of anti-CTLA-4 immunotoxin 83-saporin, diluted in 100 µl of PBS. As a control, four mice were injected with PBS alone with the same schedule. Animals were sacrificed at day 3 and subjected to autopsy, because systemic acute toxicity usually appears within 48 h. Tissue corresponding to the site of injection was fixed in 10% formaldehyde and processed for light microscopy. Ten fields were examined at magnification x250, and infiltrating lymphoid cells were counted by two different investigators.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two immunotoxins were prepared with the anti-CTLA-4 scFvs 83
and 40, conjugated to the type 1 RIP saporin by an artificial disulfide
bond. Sulfhydryl groups were inserted in the saporin by an imidoester
reaction between 2-IT and the primary amino groups of the proteins, as
previously reported (15). One thiol group per
molecule was inserted in saporin, using a 4-fold molar excess of
2-IT (Table I
).
|
). The inhibitory activity of native and
conjugated RIP on protein synthesis by a rabbit reticulocytes lysate is
also reported in Table I. Despite a loss of activity of 3- to 5-fold,
saporin still retained a high protein synthesis-inhibitory activity
after conjugation. Reactivity of immunotoxins
To evaluate the possible change in reactivity of the
conjugates, indirect immunofluorescence was performed and the degree of
cell staining was analyzed by flow cytometry. The scFv Abs after
conjugation with saporin maintained their reactivity with T cells
activated via CD3/CD28. All tested cell lines reacted with BN13
anti-CTLA-4 reference mAb (28) and were found positive
for the 83-recognized epitope, with different degrees of reactivity
(Table II). All cell lines showed a
weaker staining in comparison with activated (via CD3/CD28) T cells. To
confirm and extend these results, we performed the RT-PCR on three
representative cell lines (HL60, K562, HOM-2) and on resting and
activated PBLs. The transcript of the CTLA-4 extracellular domain was
present both in cell lines and in activated PBLs. The intensity of
electrophoretic band was 28.0% (HL60), 26.1% (K562), and 32.6%
(HOM-2) compared with activated PBLs. Resting PBLs were negative.
|
Cytotoxicity tests
In previous studies, we demonstrated that RIP-containing
conjugates kill cells via apoptosis (15, 27). Thus, to
ascertain the effectiveness of immunotoxins in killing cells, we
evaluated the apoptotic phenomenon on target cells, using the annexin
V/propidium iodide staining. A series of tests were performed using
CD3/CD28-stimulated T lymphocytes, expressing CTLA-4 Ag. Both 83- and
40-saporin immunotoxins were specifically toxic to activated T cells.
In particular, 83-saporin had an AC50 of
2.33 x 10-12 M, as saporin, on
CD3/CD28-stimulated T lymphocytes, while 40-saporin had an
AC50 of 4.29 x 10-11
M (Fig. 1). Saporin alone or mixed with
scFvs was weakly toxic only at 10-8 M
concentration. Unstimulated T lymphocytes were not affected (Fig. 1).
The 83- saporin caused apoptosis in >90% of CD3/CD28-stimulated T
lymphocytes, at concentrations 
10-10 M. The
most effective immunotoxin (83-saporin), tested either on a model of
CD3/CD28-stimulated lymphocytes incorporating
[3H]leucine or in an MLR having an EBV-infected
B lymphoblastoid cell line as target, had an IC50
of 7.83 x 10-11 M and of 7.20 x
10-11 M as saporin, respectively, whereas a
mixture of unconjugated scFv and RIP had a limited toxicity only at
10-8 M.
|
).
|
) with an IC50
of 1.3 x 10-10 M (Fig. 3A).
These cells underwent apoptosis with an
AC50 2.79 x 10-10 M,
and were 100% apoptotic at 10-8 M immunotoxin
concentration (Fig. 3B), while controls incubated with 83 +
saporin were not affected. The toxicity of the immunotoxin, as compared
with that of the mixture of scFv and saporin, was not detectable in the
tested concentration range both on Daudi and CEM cell lines.
|
|
Toxicity of 83-saporin was analyzed on CFU obtained from normal
G-CSF-mobilized peripheral blood stem cell (Table IV). The immunotoxin appeared
slightly toxic on granulocyte-macrophage CFU and BFU erythroid only at
the highest concentration tested (10-8 M). Same
results were obtained with free saporin, alone or mixed with the Ab,
whereas 83 scFv was not toxic.
|
H.end cells implanted in allogeneic mice typically form highly
vascularized tumors characterized by massive lymphocyte infiltration
with rapid destruction of the implant, showing a picture largely
overlapping that of an acute rejection. It has been shown that this
event is strictly T cell dependent because it does not occur in athymic
mice (32). As shown in Fig. 4, A and C, a
marked infiltration of lymphocytes was observed in H.end implants in
allogeneic mice after control treatment with PBS with a mean ± SD
of 256 ± 54 infiltrating lymphoid cells. In contrast, scarce to
moderate infiltration of inflammatory cells was detectable within the
tumor of mice treated with 12 µg (total dose) of 83-saporin
immunotoxin with a mean ± SD of 69 ± 42 cells (Fig. 4, B and D). It should be highlighted that, in
treated mice, no lymphoid-infiltrating cells were detectable in many
fields, whereas allogeneic tumor cells were completely spared. At
postmortem examination, no signs of acute organ toxicity were
observed.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In a previous study, a series of CD80/86 immunotoxins were described that could kill CD80/86-positive cells, these Ags being present on APCs (15). To better control the interactions between activated T cells and APC, it could be useful to construct conjugates able to selectively target T cells recognizing APC. CTLA-4 is an Ag that is specifically expressed on the surface of T cells during APC recognition. Thus, a selective killing of activated T cells during the early phases of allogeneic stem cell transplantation or organ transplantation might help in preventing GVHD and graft rejection.
In addition, it is possible to use scFvs as carriers of toxic moieties. In fact, the use of small human Ab fragments as toxin vehicles will have the advantages of a lower immunogenicity compared with xenogeneic Abs, a better tissue penetration, and nonaccumulation in the liver. The lack of C regions would mitigate retention by Fc receptors, found in most tissues and organs, thus further reducing their possible side effects and the probability of giving rise to an immune response due to a low degree of Ag internalization in APC.
In this view, we designed and synthesized by chemical link two immunotoxins containing two different scFvs (83 and 40) against CTLA-4 and the type 1 RIP saporin, which is one of the most effective toxic moieties used to generate immunotoxins. The conjugates were able to selectively kill T cells, activated via CD3/CD28, as well as to inhibit MLRs obtained by mixing T cells and allogeneic APC (in vitro generated DC). Moreover, 83-saporin could inhibit the proliferation of T cells when EBV-positive B lymphoblastoid cells were used as target for the MLR. We studied also the capacity of the 83-saporin immunotoxin of killing cells constitutively expressing CTLA-4. Although results varied depending on the cell line, probably due to a different recycling or internalization of the recognized Ag, we demonstrated that at least six cell lines were killed in a selective fashion, in comparison with control experiments. In fact, HOM-2, K562, L428, HL60, Jurkat, and Molt-4 cells showed a significant sensitivity to the action of 83-saporin conjugate, in comparison with the unconjugated components.
Toxicity to hemopoietic precursors of 83-saporin, the most efficient immunotoxin both on activated lymphocytes and cell lines, was nearly absent, or was present only at a low degree, at the highest dose tested (10-8 M). The latter results indicate that a possible in vitro use of the immunotoxin should not prevent the engraftment of the transplanted hemopoietic stem cells, and that the in vivo administration should not be limited by bone marrow toxicity.
Fusion proteins containing a type 1 RIP have been prepared with human growth factors or hormones (36, 37, 38, 39, 40), and with a humanized CD5 (41). In the present study, for the first time conjugates containing human-derived recombinant scFvs and a type 1 RIP (saporin) are described.
As to the use of scFvs, it should be highlighted that they might be too small to sterically occupy the recognized epitope and to give any trigger signal, thus being suitable to carry modifier molecules. In this instance, we obtained cytotoxic conjugates, but it is possible to obtain conjugates with drugs, biological response modifiers, and other Abs, the latter for a targeting with bispecific Abs.
Anti-CTLA-4 Abs linked to toxic moieties might be a new tool to prevent acute GVHD and rejection. Maximal expression of CTLA-4 occurs within 48–72 h after T cell activation. A cytotoxic conjugate, which could selectively kill activated T cells involved in GVHD or in graft rejection, if given in the early phases after transplant, might ameliorate the clinical outcome. Our results showed that both conjugates efficiently prevent the proliferation of T cells activated by DC, one of the most potent stimuli to activate allogeneic T cell responders.
A preliminary study, performed in a mouse model of allogeneic reaction, indicates that the 83-saporin actually could well target and kill activated T lymphocytes, thus reducing the lymphoid infiltrate and prolonging the survival of the allogeneic cells. Although unusual, the herein described model strongly overlapped the events occurring in a transplanted organ rejection, because an allogeneic endothelial cell population elicits an immune acute rejection mediated by T lymphocytes (32). These results strongly support the view that the 83-saporin conjugate could be used to eliminate activated T cells, which are early basic events in organ rejection and in GVHD. As a consequence, the conjugate could be used in the early posttransplant follow-up within the immunosuppressive therapy schedule. Moreover, this latter experiment could open the way to a clinical approach for targeting with scFv-RIP conjugates.
The targeting of CTLA-4 with scFv-saporin conjugates might not be limited to GVHD or graft rejection, but it is possible to conceive an ex vivo or an in vivo model for the therapy of CTLA-4-bearing tumors. The presence of CTLA-4 on malignant (42) and stimulated normal B cells (43, 44) was previously demonstrated. During our tests, we found that CTLA-4 epitopes are constitutively expressed by some cell lines and myeloid and lymphoid leukemia cells. These findings could open new insights in CTLA-4 expression patterns and lead to a possible application for neoplastic cell depletion. The cytotoxicity results showed that a preclinical test should be performed anyway, because there is a great variability in the sensitivity of tumor cells to anti-CTLA-4 immunotoxins.
A possible concern might be raised about the fact that both in vivo and in vitro small percentages of T cells might escape the immunotoxin, being spared and thus being able to repopulate. It will be possible to control a T lymphocyte repopulation, after the acute activation phase, by using different immunotoxins containing scFv recognizing other epitopes or Ag, as well as protocols for immunosuppression combining the action of immunotoxins and drugs.
Finally, experimental in vivo kinetics studies will be required to
verify a possible faster clearance of scFv-containing conjugates in
comparison with conventional immunotoxins containing the whole IgG
molecule. It should be highlighted that the molecular mass of
the conjugates in the present study is 
60 kDa (compared with 66 kDa
of albumin), thus probably avoiding the excessive renal clearance of
the scFv alone, which has a molecular mass of 30 kDa.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Prof. Andrea Bolognesi, Department of Experimental Pathology, University of Bologna, Via S. Giacomo 14, I-40126 Bologna, Italy. E-mail address: bolo{at}alma.unibo.it
3 Abbreviations used in this paper: GVHD, graft-vs-host disease; RIP, ribosome-inactivating protein; AC50, concentration of immunotoxin (expressed as RIP content) causing apoptosis in 50% of the cells; DC, dendritic cell; 2-IT, 2-iminothiolane; scFv, single-chain fragment variable; SPDP, N-succimidyl-3-(2-pyridyldithio) propionylate; BFU, burst-forming unit.
Received for publication February 26, 2001. Accepted for publication August 9, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. R. Hess, C. Barnes, M. D. Woolard, M. D. L. Johnson, J. M. Cullen, E. J. Collins, and J. A. Frelinger Selective deletion of antigen-specific CD8+ T cells by MHC class I tetramers coupled to the type I ribosome-inactivating protein saporin Blood, April 15, 2007; 109(8): 3300 - 3307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Piazza, E. Cha, I. Bongarzone, S. Canevari, A. Bolognesi, L. Polito, A. Bargellesi, F. Sassi, S. Ferrini, and M. Fabbi Internalization and recycling of ALCAM/CD166 detected by a fully human single-chain recombinant antibody J. Cell Sci., April 1, 2005; 118(7): 1515 - 1525. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Pistillo, P. L. Tazzari, G. L. Palmisano, I. Pierri, A. Bolognesi, F. Ferlito, P. Capanni, L. Polito, M. Ratta, S. Pileri, et al. CTLA-4 is not restricted to the lymphoid cell lineage and can function as a target molecule for apoptosis induction of leukemic cells Blood, January 1, 2003; 101(1): 202 - 209. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), saporin (•), scFv (
), or a mixture of
saporin and Ab (
). The percentage of apoptosis in unstimulated
lymphocytes is also reported (*). Results are means of three
different experiments. SD never exceeded 10%.
).
After 72 h of incubation and another 18 h with
[3H]leucine, the radioactivity incorporated was
determined. B, Percentage of viable cells stained via
FITC Annexin V and propidium iodide. HOM-2 cells were treated for
72 h with 83-saporin (•) or a mixture of saporin and 83 scFv
(