| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Chair of Genetics [M. P., H. K., D. S., S. J. Z., G. H. F.], Institute of Biochemistry [B. S.], Division of Hematology/Oncology, Department of Medicine III [M. G.], University of Erlangen-Nuremberg, D-91054 Erlangen, Germany, and Department of Pediatric Oncology, Children’s Hospital, University of Tübingen, D-72076 Tübingen, Germany [J. G.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One of the prerequisites for a successful immunotherapy of T-cell neoplasias is the selection of an appropriate target antigen, which ideally should be T cell specific and expressed on most T-cell lymphomas and leukemias but absent on at least a portion of normal T lymphocytes. The CD7 antigen meets these requirements. In addition, a substantial portion of AML of M1/2 type is also CD7 positive (7, 8, 9, 10)
. CD7 is a cell surface glycoprotein of 
40 kDa and a member of the immunoglobulin superfamily (11)
. Galectin I is a ligand for the extracellular domain of CD7, and interaction of T-lymphoid cells with this ligand promotes apoptosis (12)
. Therefore, CD7 may be viewed as a novel type of "death receptor," and other specific functions of CD7 are still unknown. The protein participates in signaling processes to the inside of cells, demonstrated by its association with phosphoinositide 3'-kinase after binding of CD7-specific antibodies (13)
. On activation of T cells, the surface density of CD7 is increased, and CD7 has been proposed to participate in the activation and surface adhesion of mature T cells and natural killer cells (14
, 15)
. CD7 is a marker for very early stages of T-cell maturation and is already present on lineage-committed hematopoietic progenitors in the fetal liver and on pluripotent progenitors of T cells in the thymus, bone marrow, and cord blood (16
, 17)
. CD7 is further expressed on a majority of human thymocytes and a large subset (85%) of peripheral blood T cells and natural killer cells (11
, 18, 19, 20, 21, 22, 23)
. The remaining subset of CD7-negative peripheral T cells maintains immune functions needed for the prevention of opportunistic infections and the engraftment of hematopoietic stem cells. Therefore, this subset may become relevant for therapeutic purposes, because it may serve to repopulate the T-cell compartment at least in part after a CD7-directed therapy.
A key property of CD7 for therapeutic applications is its rapid internalization after binding by an antibody, even after binding by monovalent antibody fragments (24) . This property makes the antigen well suited for targeting by immunotoxins, which are internalized together with the antigen and subsequently poison the cell from inside. Here, the CD7 hybridoma antibody TH-69 (25) , generated by one of the authors (M. G.; Ref. 26 ), was used for the construction of a CD7-specific immunotoxin, because of its high binding affinity (7.6 x 109 Mol-1). Binding of this antibody to the human T-cell lines CEM, Jurkat, and MOLT-16, derived from ALL, did not affect the proliferation of these cells in culture and did not induce apoptosis to a measurable extent. However, in athymic ("nude," nu-/-) and SCID mice xenotransplanted with CEM or MOLT-16 cells, the same antibody produced significant antitumor effects (26) . Binding of the antibody caused a rapid down-regulation ("modulation") of the antigen, and the Fc portion of the antibody was essential for the antitumor effect in vivo. Because of these promising results obtained with the unmodified antibody, the antibody has been used in pilot studies with patients with advanced T-cell tumors.4
Several investigators have targeted CD7, because the rapid internalization of the antigen after antibody binding makes this molecule attractive for the design of immunotoxins (27, 28, 29, 30) . The DA7 immunotoxin, consisting of the ricin A toxin chemically linked to a CD7-specific antibody, showed clinical efficacy in Phase I trials (30 , 31) . The use of this toxin, however, was limited by its instability and associated vascular toxicity. Therefore, current efforts are directed at linking the ricin A toxin by recombinant DNA procedures to an scFv fragment of a CD7-specific antibody, but results of this effort are still unknown (27) . The toxin saporin was linked via a bifunctional cross-linker and a disulfide bridge (32) , and pokeweed antiviral protein was chemically linked to CD7-specific antibodies (28) . In previous attempts, our group has linked cis-platinum to the TH-69 antibody, and this combination caused an inhibition of proliferation and a direct cytotoxic effect in the T-cell line CEM (33) . Therefore, CD7 meets the requirements for a potentially useful target for immunotoxins. However, until now, most CD7-directed immunotoxins were generated by conventional chemical coupling and suffered from instability and accompanying side effects, such as vascular toxicity.
To overcome some of these shortcomings, two key steps were taken in the present study. One was the generation of a stable peptide bond between the scFv antibody fragment and the toxin by recombinant genetics. The other was the use of a truncated toxin, which was unable to enter human cells on its own, in case it should be cleaved from the antibody moiety. This truncated toxin was devoid of its own binding domain for receptors on mammalian cells and, therefore, was expected to generate significantly fewer side effects. The recombinant toxin created here can only be absorbed by CD7-positive cells by virtue of its scFv portion and the CD7 internalization mechanisms of these cells.
The toxin used in our approach was Pseudomonas ETA3
, specifically the truncated version lacking domain I and containing only domains II and III (34)
. Domain I is the binding domain for a cell surface receptor present on most mammalian cells. The intact toxin enters mammalian cells via binding of domain I to this receptor. Domains II and III are required for intracellular transport and carry the active center of the toxin, respectively, which causes apoptosis by inhibiting protein synthesis via a block of the translation elongation factor-2 (35)
. Consequently, the truncated version of ETA lacking domain I is not toxic as long as it remains in the extracellular space. In addition, the truncated form of Pseudomonas ETA is expected to have fewer side effects on vascular endothelial cells, because it has an 1000-fold lower relative affinity to the LDV receptor on human vascular endothelial cells than, e.g., ricin A (36)
. Once coupled to an antibody directed against an antigen capable of internalizing, the truncated ETA becomes a potent immunotoxin, e.g., CD22- and CD25-specific scFvs linked to truncated ETA were remarkably efficient immunotoxins for the treatment of hairy cell leukemias and CD25-positive hematological malignancies, respectively (37
, 38) . Here, we asked whether scFv-ETA immunotoxins may also be effective against CD7-positive cells derived from T-ALLs, and this turned out to be the case. In this report, we describe the construction of the immunotoxin and its initial functional characterization in vitro with leukemic cell lines and long-term cultures of primary leukemic cells.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Culture of Eukaryotic Cells.
Leukemia-derived SEM cells (40)
, CEM cells (41)
, Jurkat cells (42)
, REH cells (DSMZ; German Collection of Microorganisms and Cell Lines, Braunschweig, Germany), Nalm-6 cells (DSMZ), MV4;11 cells (American Type Culture Collection, Rockville, MD), and the two hybridomas TH-69 (26)
and 9.2.27 (anti-MCSP; Ref. 43
) were cultured in RPMI 1640-Glutamax-I (Invitrogen, Karlsruhe, Germany) containing 10% FCS and penicillin and streptomycin at 100 units/ml and 100 µg/ml, respectively. COS monkey cells (DMSZ) were maintained in DMEM-Glutamax-I medium containing 10% FCS, penicillin, and streptomycin. The LLM cell line was established from bone marrow lymphoblasts of a 6-month-old female infant diagnosed with ALL.5
LLM cells (44)
were cultured in Iscove’s Modified Dulbecco Medium (Life Technologies, Inc.), supplemented with human transferrin and linoleic acid at 3 µg/ml, BSA at 10 mg/ml, 
-thioglycerol at 50 mg/ml, bathocuproin-disulfonic acid at 20 nM, insulin at 1 µg/ml, and recombinant human interleukin-7 (Strathmann, Inc., Hamburg, Germany) at 10 ng/ml.
Patient-derived Leukemic Cells.
HOL cells were derived from a 10-year-old patient, diagnosed with precursor T-ALL (CD1a+, CD2+, CD4+, CD7+, CD8+, and CD3-) at the Children's Hospital of the University of Erlangen. He was treated according to ALL-BFM-86 protocol and reached remission at day 33 of protocol I. The culture was established from peripheral blood, and the cells proliferated with a doubling time of 3 days. They were cultured in the same media as described for the cultivation of LLM cells. After 60 generations, the cells entered a proliferative crisis, and the majority of the cells died. Sufficient aliquots were frozen from early passages (generation 30) to allow reproducible work with cells of a defined number of generations in culture. The cultured cells continued to express high levels of CD7. In addition, fresh cells from a 27-year-old male patient with pre-T-ALL were tested. The immunophenotype of bone marrow blasts (80% marrow infiltration) initially was CD7+/CD5+/cytoplasmic CD3+/CD34+/CD1-/CD3-. Cells showed Terminal desoxynucleotidyl transferase positivity. Tests were performed after the patient was referred to our university hospital after initially being treated elsewhere with one course of cyclophosphamide-Adriamycin-vincristine-prednisone chemotherapy for aggressive lymphoma.
Preparation of scFv Phage Display Libraries.
Total RNA was prepared with Trizol (Invitrogen) from the CD7 antibody producing hybridoma, clone TH-69, and the hybridoma 9.2.27, producing a control antibody against MCSP. First-strand cDNA was prepared from 10–15 µg of total RNA (39)
. PCR amplification of immunoglobulin variable region cDNAs and cloning into the phagemid vector pAK100 was performed as described (39
, 45)
. Propagation of combinatorial scFv libraries and filamentous phages was performed after published procedures (45)
.
Panning of Phage Display Libraries with Intact Cells.
Panning of phage display libraries with intact cells was carried out using CD7-positive CEM cells and COS cells positive for the MCSP antigen. For this purpose, 5 x 106 cells were incubated in PBS (Invitrogen, Groningen, the Netherlands) containing 2% nonfat dry milk (NM-PBS) to block nonspecific binding sites and then incubated with 500 µl of the scFv phage library for 1.5 h at 25°C under slow agitation. Cells were washed 10 times with 2% NM-PBS and twice with PBS. Bound phages were eluted by adding 50 µg of the parental antibody as competitor. After incubation for 10 min at 25°C, cells were sedimented by centrifugation, and the supernatant was used to infect 10 ml of exponentially growing E. coli TG1 cells. A total of 20 ml of 2 x YT medium [16 grams/liter tryptone/peptone (Roth, Karlsruhe, Germany), 10 grams/liter yeast extract (Roth), and 5 grams/liter NaCl] containing 1% glucose and chloramphenicol at 30 µg/ml was added, and the cells were incubated for 2 h at 37°C under vigorous shaking (250 rpm). The cells were then superinfected with helper phage. A total of 70 ml of 2 x YT medium was then added, and isopropylthiogalactoside was added at a final concentration of 0.5 mM. Two h after infection, kanamycin was added to a final concentration of 25 µg/ml, and the culture was grown overnight at 30°C. On the next day, phages were prepared as described (45)
. After two rounds of panning, individual phages were purified, and the inserts were sequenced (46)
using an Applied Biosystems automated DNA sequencer (ABI Prism 310 Genetic Analyzer; Perkin-Elmer, Ueberlingen, Germany).
Bacterial Expression and Purification of Soluble scFv Fragments.
For the soluble expression of antibody fragments, cDNAs coding for CD7- and MCSP-specific scFvs were subcloned into the expression vector pAK400, and the plasmids were propagated in E. coli HB2151 (from Dr. G. Winter; MRC, Cambridge, United Kingdom). Expression and purification of CD7- and MCSP-specific scFv fragments were performed as described (45)
.
Construction and Expression of scFv-ETA Fusion Proteins.
For periplasmic expression of scFv fragments fused to truncated ETA under the control of the inducible T7 promoter, the plasmid pet27b(+) (Novagen, Inc.) was modified as follows. To introduce a 6 x His-tag at the NH2 terminus and a SfiI-site, two oligonucleotides, oligo 1: 5'-cc cat cac cat cac cat cac ggg gcc cag ccg gcc g-3' and oligo 2: 5'-ga tcc ggc cgg ctg ggc ccc gtg atg gtg atg gtg atg gg-3' (MWG-Biotech, Munich, Germany) were hybridized and ligated into pet27b(+) digested with MscI and BamHI. In a second step, a DNA fragment coding for a 20 amino acid linker with a second SfiI site was inserted into HindIII/NotI-digested modified pet27b(+) using oligonucleotides 3 and 4 (oligo 3: 5'-agc ttg gcc tcg ggg gcc ggt ggt ggc ggc agt ggt ggt ggc ggc agt ggt ggt ggc ggc agt ggt ggt ggc ggc agt gc-3'; oligo 4: 5'-ggc cgc act gcc gcc acc acc act gcc gcc acc acc act gcc gcc acc acc act gcc gcc acc acc ggc ccc cga ggc ca-3'). This procedure resulted in the construct pet27b(+)-L3HS9. Sequences coding for CD7- and MCSP-specific scFvs were excised from the pAK400-anti CD7 and pAK400-anti MCSP expression constructs harboring the corresponding scFv fragments and cloned as SfiI-cassettes into pet27b(+)-L3HS9. The resulting plasmids were digested with NotI/Cel II, and a DNA fragment encoding the truncated ETA was amplified by PCR from plasmid pSW202 (47)
using the primers oligo 5: 5'-gat cgc ggc cgc gct aga ggg cgg cag cct gg –3' and oligo 6: 5'-cac tag gct cag cgc ggc agt tac ttc agg tcc tcg–3'). The amplified DNA was then ligated into the vector and sequenced.
The scFv-ETA fusion proteins were expressed under osmotic stress as described (48) . Induced cultures were harvested 16–20 h after induction. The bacterial pellet from a 1 liter culture was resuspended in 10 ml of lysis-buffer [50 mM NaH2PO4, 300 mM NaCl, and 10 mM Imidazol (pH 8.0)]. Lysozyme was added at a final concentration of 1 mg/ml, and the suspension was incubated for 30 min on ice. Cells were disrupted three times for 1 min at 120 W in a sonicator/cell disrupter (B. Braun Biotech, Melsungen, Germany). The scFv-ETA fusion proteins were enriched by affinity chromatography using nickel-nitrilotriacetic agarose Beads (Qiagen, Hilden, Germany) according to the manufacturer’s instructions.
Flow Cytometric Analysis.
The binding of scFvs to cells was analyzed using a FACSCalibur FACS instrument and CellQuest software (Becton Dickinson, Mountain View, CA). Cells were stained with scFv fragments as described (45)
. Ten thousand events were collected for each sample, and analyses of whole cells were performed using appropriate scatter gates to exclude cellular debris and aggregates. To monitor binding of scFv-ETA fusion proteins, 5 x 105 cells were incubated for 30 min on ice with 20 µl of the immunotoxin at a concentration of 1 µg/ml. A nonrelated immunotoxin served as a control for background staining. The cells were washed with PBA buffer [containing PBS, 0.1% BSA, and 7 mM Na-azide] and then incubated with 50 µl of a polyclonal rabbit anti-Pseudomonas ETA serum (Sigma, Deisenhofen, Germany) diluted 1:250 in PBA buffer. Cells were washed and incubated with fluorescein-iso-thiocyanate (FITC)-conjugated pig antirabbit-IgG (DAKO Diagnostica GmbH, Hamburg, Germany) for 30 min. After a final wash, the cells were analyzed by FACS.
Measurement of Cytotoxic Effects of Immunotoxins.
Cells were seeded at 2.5 x 105/ml in 24-well plates, and immunotoxins were added at varying concentrations. Cell death was measured by staining nuclei with a hypotonic solution of PI as described (49
, 50)
. The extent of cell death was determined by measuring the fraction of nuclei with subdiploid DNA content. Fifteen thousand events were collected for each sample and analyzed for subdiploid nuclear DNA content. To determine whether cell death was attributable to apoptosis, whole cells were stained with FITC-conjugated Annexin V (PharMingen, Heidelberg, Germany; Ref. 51
) and PI in PBS according to the manufacturer’s protocol. For blocking experiments, cells were seeded at 2.5 x 105/ml in 24-well plates, and a 100-fold molar excess of the parental antibody or a nonrelated antibody of the same isotype was added to the culture 1 h before adding the immunotoxin (100 ng/ml). Viable cell counts were determined by trypan blue staining. Cell death was quantitated by staining with Annexin V and PI in PBS as described above.
SDS-PAGE and Western Blot Analysis.
SDS-PAGE was performed by standard procedures (52)
. Western blots were performed with secondary antibodies coupled to horseradish peroxidase (Dianova, Hamburg, Germany; Ref. 53
). Enhanced chemiluminescence reagents (Amersham Pharmacia, Freiburg, Germany) were used for detection. ScFvs were detected with a penta-His antibody (Qiagen). ScFv-ETA fusion proteins were detected using an anti-Pseudomonas ETA polyclonal antibody (Sigma). Full-length PARP and its specific cleavage product were detected using a mouse antihuman PARP antibody (PharMingen).
Statistical Analysis.
All statistical analyses were performed with Microsoft EXCEL software. Ps were obtained using two-tailed paired t tests with a confidence interval of 95% for evaluation of the significance of differences between treatment groups.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
but failed to bind to the CD7-negative pro-B leukemia cell line SEM (Fig. 1B)
. The MCSP-specific control scFv subcloned from the hybridoma cell line 9.2.27 specifically bound to MCSP-positive COS cells (Fig. 1C)
but failed to bind to the MCSP-negative cell line MV4–11 (Fig. 1D)
. Thus, the process of subcloning the hybridoma antibodies to scFv fragments did not detectably alter the binding specificity of the antigen-combining sites.
|
. The protein was purified by affinity chromatography to >90% purity, and 100 µg of purified protein were routinely obtained from 1 liter of bacterial culture. The protein apparently retained its native folding after expression in bacteria, because it was not necessary to denature and renature it to obtain efficient specific binding after purification. The protein also did not show a noticeable tendency to form intracellular precipitates. Coupling of the scFv coding sequences to the truncated ETA coding sequences via sequences coding for a polypeptide linker also did not affect the binding specificity of the antigen-combining site. The recombinant CD7-specific toxin still bound to CEM (Fig. 3A)
but not to CD7-negative SEM cells (Fig. 3B)
. An additional control was performed to exclude that any recombinant scFv-ETA protein may have these binding properties because of their common design rather than their particular scFv-moiety. To this effect, the hybridoma 9.2.27-derived antibody specific for the MCSP antigen (43)
was also subcloned into an scFv fragment and converted into an anti-MCSP-ETA scFv immunotoxin by analogous procedures. This molecule was similarly expressed in bacteria and purified. It failed to bind to MCSP-negative CEM cells (Fig. 3D)
but bound to LLM cells, a leukemia-derived pro-B cell line with chromosomal translocation t(4;11) (Fig. 3C)
that expresses MCSP on the cell surface. Thus, the specificity of binding of the scFv-ETA immunotoxins was the result of their particular scFv moieties and not the result of their common design.
|
|
. Next, the extent of cell death was measured as a function of time and the dose of the agent by staining of nuclei with PI and FACS. The fraction of cells with subdiploid DNA content was equated with the fraction of dead cells. The immunotoxin was added to the cultures in a single dose, and culture aliquots were analyzed for cell death at different times after addition. The toxin promoted death of CEM cells in a time- and dose-dependent manner (Fig. 5A)
but failed to affect SEM cells at the same concentrations (Fig. 5B)
. Significant cell death was already obtained at concentrations of 1 ng/ml, compared with untreated cells (P = 0.002), and maximum death was reached at 100 ng/ml, corresponding to a concentration of 1.3 nM. Thus, the agent was capable of killing CD7 antigen-positive cells with high efficiency within 96 h after the addition of a single dose in the nanomolar concentration range. The immunotoxin also killed cells from a second leukemia-derived T-ALL line, Jurkat, at similar concentrations. At concentrations causing maximum death of Jurkat cells, the B-leukemic cell lines REH, Nalm-6 (Fig. 6)
, and SEM remained unaffected (Fig. 5B)
. Thus, it can be excluded that the toxin killed nonspecifically, and the data support the conclusion that the cytotoxic effect was antigen-specific. The fraction of dead cells measured by this procedure did not reach 100%, although microscopic analysis suggested near complete lysis. This apparent discrepancy was resolved and found to be the result of this particular method of quantitating cell death. When a different method was used, the fraction of dead cells exceeded 95% under identical conditions (see below).
|
|
|
, bottom right quadrant). As a control, the recombinant anti-MCSP-ETA scFv immunotoxin of similar design failed to induce Annexin V staining and cell death. As an additional sign of death by apoptosis, the CD7-ETA toxin induced cleavage of PARP from its intact form (116 kDa) to its characteristic 85 kDa proteolytic fragment in CEM cells. Treatment of these cells with equivalent concentrations of the anti-MCSP-ETA toxin failed to induce PARP cleavage (Fig. 7B)
. Thus, by two independent criteria, the CD7-specific immunotoxin was shown to induce cell death via apoptosis for both the CEM and Jurkat lines (Fig. 7, A and B)
.
|
) or by staining with PI (PI uptake) and Annexin V (method 2; Fig. 8B
). Method 1 showed 80% of dead cells (subdiploid DNA content), whereas measurement by method 2 showed a significantly higher number of dead cells (
97%; P = 0.04). Thus, measurement by method 1 underestimated the number of dead cells relative to determination by microscopic counting and method 2 (Fig. 8, A and B)
. Apoptotic death induced by the toxin and evidenced by method 2 was inhibited specifically by pretreatment with an excess of the parental monoclonal TH-69 antibody (Fig. 9
, middle two panels). Treatment with a comparable excess of an isotype control antibody (Fig. 9
, bottom two panels) showed cell death to the same extent as cells treated with the CD7-ETA toxin alone (Fig. 9
, top two panels). These results confirm the conclusion that killing was antigen specific and that induction of apoptosis was attributable to receptor-mediated uptake of the immunotoxin.
|
|
. Compared with the MCSP-ETA-treated controls, 20% specific death (Annexin V positive; P = 0.007) was observed 96 h after administration of an initial dose of 100 ng/ml and a second dose after 48 h, given with a renewal of the culture medium. The effect was specific, because it was prevented by pretreatment with an excess of the parental TH-69 antibody but not by pretreatment with a comparable excess of a control antibody of matching isotype (P = 0.03). Pretreatment with the parental antibody did not completely block CD7-ETA-induced killing, but the difference in cell death at time point 96 h between MCSP-ETA and CD7-ETA blocked by TH-69 did not reach statistical significance (P = 0.08). The anti-MCSP-ETA scFv immunotoxin had no effect on HOL cells, which are MCSP negative. In conclusion, the CD7-ETA immunotoxin was able to induce cell death in short-term cultures of primary leukemic cells but not to the same extent as in established cell lines, although the toxin was added twice.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Contrasted with this background, the results of the present study offer several new elements of knowledge: (a) the construction principle of recombinant scFv-ETA molecules used here works not only for the CD22 and CD25 antigens (37 , 38) but also for CD7; it may also work for a number of other cell surface antigens expressed in similar concentrations and internalized with similar kinetics after binding of an antibody; (b) several of these scFv-ETA toxins induce efficient lysis in the concentration range of ng/ml (37 , 47) . This is true both for the CD22-ETA scFv toxin (37) and the CD7-ETA chimeras reported here, as well as other published constructs (27 , 47) . Although different methods were used to quantitate cell death, and the results are not strictly comparable, it is probably still fair to state that several of these scFv-ETA toxins work with similar potency on human leukemic cells and are effective in the concentration range of <100 ng/ml. A precise comparison of the various agents can only be achieved in a side-by-side evaluation using the same assay for cellular toxicity; (c) the attachment of a truncated ETA domain did not alter the binding specificity of the scFv portion in agreement with the results reported by others (37 , 38 , 47) ; in our construct, a different linker was used than in other published constructs, and, thus, a new element of information is that the precise sequence of the linker does not appear to be critical; (d) cell death occurred by apoptosis as opposed to other mechanisms; this was shown in our study by Annexin V staining and PARP cleavage; and (e) our CD7-specific toxin worked not only for the rapidly dividing T-ALL cell lines CEM and Jurkat but also for long-term cultures of leukemia-derived HOL cells, although with lower efficiency. The reasons for the reduced efficiency are unknown, but probably the difference is attributable to the fact that stable lines differ from primary cells by the gain of additional mutations. This often leads to the selection of cells with a shorter generation time that may also be more susceptible to the agent.
The finding that 20% of specific lysis was obtained in long-term primary cultures and cultures of fresh biopsy material as opposed to >95% in stable lines is not a disappointment. In other studies published to date, the effect of immunotoxins was evaluated primarily by using stable lines and xenograft models, in which human leukemic cells were transplanted into immunocompromised mice and were then treated with the agent (26)
. Studies describing the efficacy of scFv immunotoxins for primary patient cells (54)
are limited. Therefore, other authors may also have experienced difficulties in evaluating such agents in primary cells. In view of these difficulties, it is encouraging that 20% of cellular lysis was reproducibly obtained in the present study in long-term primary cultures. The observation that only 20% of lysis was obtained in a 96-h period does not preclude the potential future usefulness of this agent in vivo, because in vivo it may be administered in several successive doses, and, thus, a higher percentage of killing may be achieved. The usefulness of this agent in vivo may further depend on a number of parameters that have not been evaluated in our studies with cultured cells and that can only be assessed by in vivo experiments.
In summary, in the present study, a CD7-ETA scFv immunotoxin was constructed with a design aiming at circumventing some of the problems encountered with earlier CD7-directed immunotoxins generated by chemical coupling of the toxin to the antibody. The new agent bound to CD7-positive T-lymphoid cells and killed them by the induction of apoptosis. In view of the known fact that CD7 internalizes rapidly after ligand binding and the unpublished observation that clinical application of the native unmodified TH-69 antibody leads to antigen modulation resulting in CD7-negative cells,4 a CD7-specific immunotoxin appears promising against T-cell malignancies. This expectation is further supported by the recent findings that other immunotoxins constructed by fusions to a truncated ETA produced encouraging clinical results. Treating patients with hairy cell leukemia with the immunotoxin BL22 (CD22-ETA), Kreitman et al. did not observe any pulmonary edema, a complication encountered after treatment with corresponding immunotoxins containing a deglycosylated ricin A chain (30 , 31) . The CD7-ETA scFv immunotoxin reported here may therefore circumvent some of the problems seen with older immunotoxins.
Although a CD7-positive subpopulation of CD34+,CD38- cord blood cells with differentiation capabilities beyond the T-cell lineage has been described (16 , 17) and the expression of CD7 on immature AML (7, 8, 9, 10) could even be interpreted as an indication for progenitors to myeloid and T lineage, no significant CD7 expression is found on most hematopoietic progenitor cells (16) . Therefore, an immunotoxin directed against CD7 has a promising range of prospective applications against T-cell neoplasms. Although a greater fraction of T-cell leukemias and some cases of AML express CD7 on their surface than CD3, CD4, CD5, or CD25, the CD7 construct still avoids eliminating too many normal T cells important for the maintenance of relevant immune functions, because it spares the CD7-negative subset of normal T-lymphoid cells (23) . Elimination of a too large subset of normal T cells is a problem accompanying the clinical use of the CD52 antibody Campath-1H. Therefore, the CD7 immunotoxin reported here deserves further testing in in vivo models, and such experiments are under way.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by Research Grant 1996.047.3 from the Wilhelm Sander Stiftung, Neustadt, Germany (to G. H. F. and S. J. Z.). 
2 To whom requests for reprints should be addressed, at Chair of Genetics, University of Erlangen Nuremberg, Staudtstrasse 5, D-91058 Erlangen, Germany. Phone: 49-9131-852-8493; Fax: 49-9131-852-8526; E-mail: gfey{at}biologie.uni-erlangen.de
3 The abbreviations used are: T-ALL, acute T-cell leukemia; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; ETA, exotoxin A; FACS, fluorescence-activated cell sorter; MCSP, melanoma-associated chondroitin sulfate proteoglycan; PARP, poly(ADP-ribose) polymerase; PI, propidium iodide; scFv, single-chain fragment from the variable region.
4 M. Gramatzki, unpublished data.
Received 11/ 7/01. Accepted 3/14/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Bremer, B. t. Cate, D. F. Samplonius, L. F. M. H. de Leij, and W. Helfrich CD7-restricted activation of Fas-mediated apoptosis: a novel therapeutic approach for acute T-cell leukemia Blood, April 1, 2006; 107(7): 2863 - 2870. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Bremer, D. F. Samplonius, M. Peipp, L. van Genne, B.-J. Kroesen, G. H. Fey, M. Gramatzki, L. F.M.H. de Leij, and W. Helfrich Target Cell-Restricted Apoptosis Induction of Acute Leukemic T Cells by a Recombinant Tumor Necrosis Factor-Related Apoptosis-Inducing Ligand Fusion Protein with Specificity for Human CD7 Cancer Res., April 15, 2005; 65(8): 3380 - 3388. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Lang, K. Barbin, T. Feuchtinger, J. Greil, M. Peipp, S. J. Zunino, M. Pfeiffer, R. Handgretinger, D. Niethammer, and G. H. Fey Chimeric CD19 antibody mediates cytotoxic activity against leukemic blasts with effector cells from pediatric patients who received T-cell-depleted allografts Blood, May 15, 2004; 103(10): 3982 - 3985. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Weidmann, M. Gramatzki, M. Wilhelm, and P. S. Mitrou Diagnosis and actual therapy strategies in peripheral T-cell lymphomas: summary of an international meeting Ann. Onc., March 1, 2004; 15(3): 369 - 374. [Full Text] [PDF]
|
|
|
|
|
|
M. K. Tur, M. Huhn, T. Thepen, M. Stocker, R. Krohn, S. Vogel, E. Jost, R. Osieka, J. G. van de Winkel, R. Fischer, et al. Recombinant CD64-Specific Single Chain Immunotoxin Exhibits Specific Cytotoxicity against Acute Myeloid Leukemia Cells Cancer Res., December 1, 2003; 63(23): 8414 - 8419. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) or were treated with 100 ng/ml CD7-ETA (
), 100 ng/ml CD7-ETA + TH-69 (
), 100 ng/ml CD7-ETA + isotype control antibody (
), TH-69 (
), and isotype antibody (
). At given time points, viable cells were counted by trypan blue staining. Each time point was measured in triplicate, and SDs are indicated by error bars. The data are representative of four separate experiments.
; P = 0.002), 5 ng/ml (
; P = 0.003), 100 ng/ml (
, treated with the CD7-ETA toxin alone; x, treated with the CD7-ETA toxin plus a 100-fold molar excess of an isotype-matched control antibody;