| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Medicine, Division of Immunology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA 02215; and
Leukosite Inc., Cambridge, MA 02142
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain of the IL-2R, as well as IL-2 fusion
toxins, have been used in animals and humans to achieve selective
immunosuppression. Here we report on the development of a novel IL-2R
targeting agent, a cytolytic chimeric IL-2/Fc fusion protein. This
immunoligand binds specifically and with high affinity to IL-2R and is
structurally capable of recruiting host Ab-dependent cell-mediated
cytotoxicity and complement-dependent cytotoxicity activities. The Ig
component ensures an extended circulating
t1/2 of 25 h following systemic
administration. To subsequently explore the mechanisms of the
antidiabetogenic effects of IL-2/Fc, we have mutated the FcR binding
and complement C1q binding (Fc-/-) domains of the Fc
fragment to render the Fc unable to direct Ab-dependent cell-mediated
cytotoxicity and complement-dependent cytotoxicity activities. In a
model of passive transfer of diabetes in nonobese diabetic mice, lytic
IL-2/Fc, but not nonlytic IL-2/Fc-/-, exhibited striking
antidiabetogenic effects. Together with the negligible potential of
IL-2/Fc for immunogenicity, this finding forecasts that cytolytic
IL-2/Fc may offer a new therapeutic approach for selective targeting of
auto and alloimmune T cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain in mice, rats, and humans, as well as a
chimeric IL-2 diphtheria toxin-related fusion protein, have proven to
be effective in preventing and, in some instances, treating acute
allograft rejection (4, 5). Moreover, treatment with the
IL-2 diphtheria toxin fusion protein aborted diabetogenic autoimmunity
in the nonobese diabetic
(NOD)4 model
(6). However, humanized mAbs and recombinant fusion toxin
proteins have not completely circumvented problems of immunogenicity,
short circulating t1/2, or diminished
affinity compared with IL-2 for IL-2R (6). The humanized
mAbs, although possessing low immunogenicity and a long circulating
t1/2, do not match the high affinity
of IL-2 for the IL-2R, and IL-2 toxin-related fusion proteins are
highly immunogenic and short-lived. Moreover, the IL-2 moiety loses
some affinity for IL-2 due to its fusion with toxin (6).
We now hypothesize that the hurdles of immunogenicity, suboptimal
affinity, short circulating t1/2, and
inability to direct host cytolytic immune effector mechanisms against
target cells may all be addressed in large measure by the creation of a
fusion between host-species IL-2 and host-species cytolytic
Fc.
To develop a novel high affinity, nonimmunogenic, and long-circulating
IL-2R-targeting reagent, we have genetically fused murine IL-2 (mIL-2)
to murine Fc
2a and have produced a "cytolytic" IL-2/Fc
immunoligand with the potential to kill target cells via the activation
of complement and FcR of leukocytes. We now demonstrate that this
immunoligand binds with high affinity and specificity to IL-2R and to
FcRs; in addition, it directs complement-dependent cytolysis. The Ig
component ensures an extended circulating
t1/2 to 25 h following systemic
administration. To explore the mechanisms of the antidiabetogenic
immunosuppressive effects of lytic IL-2/Fc, we have mutated the FcR
binding and complement C1q binding domains of the Fc fragment to render
it incapable of directing Ab-dependent cell-mediated cytotoxicity
(ADCC) and complement-dependent cytotoxicity (CDC) activity. Herein, we
report on the construction and use of these IL-2/Fc fusion proteins in
a model of passive transfer of diabetes in NOD mice.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
mIL-2 cDNA, including 36 bp of 5' untranslated sequence and its
native leader sequence, was amplified by PCR from the plasmid ATCC
37553 using synthetic oligonucleotide primers. The sense primer was
designed to append a unique NotI site to the 5' end. The
antisense primer eliminated the termination codon, substituted Ser for
the unpaired Cys residue, and changed the codon usage for the terminal
Gln residue from CAA to CAG to provide the first base of a unique
BamHI site appended to the 3' end. The murine Fc2a mRNA
sequence was generated from an IgG2a-secreting hybridoma (HB129,
American Type Culture Collection (ATCC), Manassas, VA) and was
converted to cDNA using a standard technique with reverse transcriptase
Moloney murine leukemia virus (Life Technologies, Grand Island, NY)
with a synthetic oligo(dT)(12, 13, 14, 15, 16, 17, 18)
oligonucleotide (Life Technologies). To construct a nonlytic IL-2/Fc
construct, oligonucleotide site-directed mutagenesis was used to
replace the C1q binding motif Glu318,
Lys320, Lys322 with Ala
residues (7). Similarly, Leu235 was
replaced with Glu to inactivate the FcRI binding site as described
previously (Fig. 1
) (8, 9).
Ligation of IL-2 and Fc2a components in the correct translational
reading frame at the unique BamHI site yields a 1236-bp long
open reading frame encoding a single 411-aa polypeptide (including the
18-aa IL-2 signal peptide) with a total of 13 cysteine residues (Fig. 1). The mature secreted homodimeric IL-2/Fc is predicted to possess up
to eight intramolecular and three interheavy chain disulfide linkages
and a molecular mass of 90.1 kDa exclusive of glycosylation.
|
The proper genetic construction of IL-2/Fc (carrying the
wild-type Fc2a sequence) and IL-2/Fc-/-
(bearing mutated complement C1q and FcRI binding domains) sequences
was confirmed by DNA sequence analysis after cloning of the fusion
genes as NotI-XbaI cassettes into the eukaryotic
expression plasmid pRc/CMV (Invitrogen, San Diego, CA). The plasmid was
transfected into Chinese hamster ovary (CHO) cells and selected by
G418. High-yield clones were selected and cultured in a serum-free
medium. IL-2/Fc fusion proteins were then purified by protein
A-Sepharose affinity chromatography, followed by dialysis against PBS
and 0.22-µm filter sterilization. Purified proteins were stored at
-20°C before use (9).
Confirmation of size, IL-2, and Fc content specificity
Western blot analysis following SDS-PAGE under reducing (with
DTT) and nonreducing (without DTT) conditions was performed using
anti-mIL-2 mAb (PharMingen, San Diego, CA) as well as polyclonal
anti-mouse Fc primary Abs (Pierce, Rockford, IL).
Standardization of the biological activity of rIL-2 and IL-2/Fc
Using a standard curve based on commercially supplied rIL-2 (PharMingen), IL-2/Fc concentrations were determined by both ELISA and bioassay. Unit activity based on ELISA corresponded with that obtained in a standard IL-2 bioassay, using a mIL-2-dependent cell line (CTLL-2, ATCC).
In vitro characterization of IL-2/Fc cytolytic activity
IL-2/Fc cytolytic activity was assessed by two independent
assays. First, the ability of IL-2/Fc to bind FcRI was analyzed
using CHO-K1 cells transfected with human FcRI cDNA. The murine
FcRI- and FcRII-negative, IL-2R- CHO cells
were transfected by electroporation with 20 µg of PvuI
linearized pRc/CMV carrying cDNA for human FcRI (a gift of Brian
Seed, Massachusetts General Hospital, Boston, MA). CHO/FcRI cells
(5 x 105) were washed twice with FCM buffer
(PBS containing 0.1% FCS (BioWhittaker, Walkersville, MD) and 0.1%
sodium azide (Sigma, St. Louis, MO)) and subsequently incubated with 10
µg/ml of murine IgG2a (mIgG2a) (Cappel, West Chester, PA), IL-2/Fc,
or IL-2/Fc-/-. After incubation for 60 min on
ice, the cells were harvested and washed in FCM buffer and subsequently
incubated with fluorescein-conjugated polyclonal goat anti-mIgG Fc
Ab (Pierce) for 60 min in the dark. The cells were washed and stored in
1% formalin/PBS solution at 4°C and then analyzed for fluorescent
staining with a FACStar (Becton Dickinson, San Jose, CA).
Complement-dependent cell lysis was assayed by
51Cr release using IL-2R+
CTLL-2 cells as targets. Cells were stripped of bound unlabeled IL-2 by
exposure for 20 s to RPMI 1640 medium with 10% FCS (pH 3),
washed, and incubated with 51Cr for 60 min at
37°C (DuPont, Boston, MA). The radiolabeled cells were washed and
plated at 5 x 104 cells/well in a
flat-bottom, 96-well microtiter plate, followed by incubation for 60
min on ice in the presence of various concentrations of IL-2/Fc,
IL-2/Fc-/-, or mIgG2a (Cappel). Low Tox-R
complement (Cedarlane, Hornby, Ontario, Canada) at a 1/15 dilution or
1% Nonidet P-40 was then added before a final 60-min incubation at
37°C with gentle agitation. After centrifugation at 200 rpm for 10
min, 140 µl of supernatant from individual wells was transferred to
glass tubes for counting. The percentage of specific lysis was
calculated using the following formula: % specific lysis =
([experimental release with complement - release with complement
alone]/[release with Nonidet P-40 - release with complement
alone]) x 100%.
Determination of IL-2/Fc circulating t1/2
The concentration of free IL-2/Fc in the serum was determined
over time after a single bolus i.v. injection of the fusion protein to
three 10-wk-old BALB/c mice (The Jackson Laboratory, Bar Harbor, ME).
Serial 100-µl retro-orbital blood samples were obtained at 5 min,
1 h, 5 h, 8 h, 24 h, 48 h, 72 h, and
96 h postinjection. A sandwich ELISA was employed. A rat
anti-mIL-2 mAb was used as the capture Ab, and HRP-conjugated rat
anti-mouse Fc2a mAb was used as the detection Ab (PharMingen),
thus assuring that this assay was specific for only the IL-2/Fc and not
for IL-2 or mIgG2a.
Determination of the immunogenicity of IL-2/Fc
To probe the immunogenicity of IL-2/Fc, three BALB/c male mice were given IL-2/Fc i.p. at a dose of 10 µg/mouse on the first day and 5 µg/day for the next 2 wk. Serum samples were taken from these mice at 2, 3, and 4 wk after the cessation of treatment and were tested by ELISA to detect Abs against IL-2/Fc. ELISA plates were coated with serially diluted serum samples obtained from mice receiving IL-2/Fc treatment. As a positive control, some ELISA plates were coated with serially diluted anti-IL-2 mAb or anti-IgG2a Fc mAb (PharMingen). Biotinylated IL-2/Fc was used to detect the Abs against IL-2/Fc. The IL-2/Fc fusion protein was biotinylated following a protocol for biotinylation of proteins in solution (10).
Determination of competence of Ab-dependent target cell lysis in NOD mice
We treated 8- to 12-wk-old NOD and BALB/c mice with depleting GK1.5 anti-CD4 mAb (TB207, ATCC) at a dose of 0.2 mg/mouse/day i.p. for 3 consecutive days. Serial 100-µl retro-orbital blood samples were obtained at 1, 2, and 3 wk after the initiation of treatment. Erythrocytes were depleted by treatment with ACK lysing buffer (BioWhittaker). Leukocytes were stained with the PE-conjugated anti-CD4 mAb RM4-4 (PharMingen), which does not compete with GK1.5 for binding to T cells. The cells were analyzed for cell surface phenotype by flow cytometry using CellQuest software (Becton Dickinson).
Adoptive transfer of diabetes in NOD mice
We obtained 8- to 12-wk-old NOD/Lt male mice and 5-wk-old NOD/Lt female mice from the Jackson Laboratory. The incidence of diabetes in 25-wk-old NOD/Lt control female mice was 81% (n = 17); the incidence in control males was 25% (n = 12). The incidence of diabetes in irradiated male NOD/Lt mice by 25 wk of age was 33% (n = 12). Monodispersed spleen cells were depleted of erythrocytes by treatment with ACK lysing buffer (BioWhittaker). Each of a series of 8- to 12-wk-old irradiated (700-rad) NOD male recipients were injected i.v. with aliquots of 20 x 106 splenic leukocytes obtained from acutely diabetic female NOD mice (hyperglycemia at <2 wk). Treatment was initiated on the day of adoptive transfer with 10 µg of IL-2/Fc, IL-2/Fc-/-, or control IgG2a i.p. for the first dose and then 5 µg every day for 4 wk.
Blood glucose levels (BGLs) were tested weekly using Chemstrip bG strips and an Accucheck III glucose monitor (Boehringer Mannheim Biochemicals, Indianapolis, IN). Elevated BGLs were followed up on the following day by a repeat test. Diabetes was diagnosed when the BGL was >16.5 mmol/L on any single measurement or >13.8 mmol/L on 3 consecutive days.
Antibodies
Isotype-matched control rat mAbs (IgG1, IgG2a, IgG2b) and rat
mAbs to mouse cell surface and cytokine Ags were purchased from
PharMingen; the mAbs employed were directed against mouse T cells (CD5,
53-7.3) or subsets (CD4, H129.19; CD8a, 53-6.7), B cells (CD45R/B220,
RA3-6B2), mononuclear phagocytes (CD11b, M1/70, Mac-3, M3/84), and
IL-2R (CD25, 3C7). The secondary Abs employed were mIg-absorbed goat
anti-rat Ig (Sigma), rabbit anti-goat Ig, and goat
peroxidase-antiperoxidase (PAP) (Dako, Carpinteria, CA).
Histology and immunohistochemistry
Pancreatic samples were harvested from NOD mice at 10 wk after the adoptive transfer of splenic leukocytes from acutely diabetic NOD donors. Samples were embedded in optimal cutting temperature compound (Tissue TCK, Miles, Elkhart, IN), snap frozen in liquid nitrogen, and stored at -70°C until sectioning; two to three samples/group/timepoint were analyzed. For morphological evaluation, cryostat sections were fixed in methanol and stained with hematoxylin and eosin. For immunohistochemistry staining, cryostat sections were fixed (10 min, 4°C) in paraformaldehyde-lysine-periodate and stained by a four-layer PAP method involving overnight incubation with rat anti-mouse mAb (4°C) followed by goat anti-rat Ig (5 µg/ml, 30 min), a methanol/hydrogen peroxide block (10 min), rabbit anti-goat Ig (1:50, 30 min), goat PAP complexes (1:50, 30 min), and diaminobenzidine substrate (11, 12). Sections were then washed, counterstained in hematoxylin, and mounted. The samples were evaluated in a blinded fashion, using two to three different levels of sectioning/sample.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To confirm the molecular size and the cytokine/isotype specificity
of the IL-2/Fc fusion protein, the affinity-purified fusion protein was
characterized by Western blot following SDS-PAGE. As shown in Fig. 2, the IL-2/Fc fusion proteins migrate
under reducing (with DTT) conditions as a single species at the
expected molecular size of 45 kDa. Under nonreducing (without DTT)
conditions, each IL-2/Fc runs as a single species at a molecular size
of 90 kDa, which indicates that the fusion proteins are expressed as
homodimers. Moreover, the IL-2/Fc fusion proteins are bound by both
anti-mIL-2 mAb (Fig. 2B) and anti-mIgG heavy chain
polyclonal Abs (Fig. 2A), confirming the cytokine and
isotype specificity of the IL-2 moiety and Fc2a domain,
respectively. Supernatants of the transfected CHO cells yielded 
0.5
µg/ml of IL-2/Fc protein.
|
). In a radiolabeled IL-2 displacement
assay, IL-2/Fc was essentially equivalent to rmIL-2 in displacing
[125I]-labeled human rIL-2 (rhIL-2) from
IL-2R+ CTLL-2 cells (Fig. 4). IL-2/Fc-/-
possesses biological functions of the IL-2 moiety that are identical
with those of IL-2/Fc (data not shown). The function of the Fc fragment
of the fusion protein was then assessed for its ability to direct CDC
and to bind to FcRI. Specific lysis of CTLL-2 cells in the presence
of lytic IL-2/Fc and complement was as high as 60% at 0.5 µg/ml; the
nonlytic (IL-2/Fc-/-) fusion protein did not
direct CDC (Fig. 5). FcR binding was
assessed using CHO cells transfected with human FcRI cDNA. This
approach was necessary to eliminate the binding of immunoligand by
IL-2Rs on cells normally bearing FcRI on their surface. FACS
analysis demonstrated that the lytic IL-2/Fc fusion protein bound to
FcRI-bearing CHO cells like mIgG2a, whereas
IL-2/Fc-/- did not bind to FcRI (Fig. 6). Taken together, these results
demonstrate that the IL-2 and Fc moieties of the fusion molecule
possess their expected physical and functional attributes in vitro.
|
|
|
|
The circulating t1/2 of IL-2/Fc
after single i.v. bolus administration was determined to be biphasic,
with a rapid initial clearance of 3.5 h (Fig. 7A) followed by a slower
terminal component of 25 h (Fig. 7B). The circulating
IL-2/Fc concentration was determined to be >50 ng/ml after IL-2/Fc
administration at a dose of 5 µg/mouse i.p. daily (data not
shown).
|
Administration of depleting GK1.5 anti-CD4 mAb diminished CD4+ cells in NOD mice
As shown in Table I, in NOD mice
anti-CD4 mAb GK1.5 treatment decreased the frequency of
CD4+ cells of peripheral blood to 11.45 ±
1.25% of gated lymphocytes after 1 wk, which recovered to 18.45
± 1.35% after 2 wk and 22.26 ± 1.91% after 3 wk. In
comparison, there were 43.08 ± 4.18% CD4+
cells in the gated lymphocytes from the peripheral blood of untreated
NOD mice. However, the Ab-dependent target cell lysis is somewhat
impaired. In BALB/c mice, only 2.09 ± 1.87% of gated peripheral
lymphocytes were CD4+ cells after 1 wk of GK1.5
treatment, and the frequency of CD4+ cells at 2
and 3 wk was 8.20 ± 1.23% and 13.87 ± 2.46,
respectively.
|
The effects of IL-2/Fc were tested in a model of passive transfer
of diabetes in NOD mice. As shown in Fig. 8, 100% of irradiated male NOD
recipients in both untreated (n = 10) and control
Ig-treated (n = 9) groups developed diabetes at 24 wk
after the transfer of 20 x 106 splenic
leukocytes from acutely diabetic female NOD mice. In contrast, only
30% of recipients receiving IL-2/Fc treatment (n = 10)
developed diabetes (Fig. 8). It is notable that one of the
IL-2/Fc-treated NOD mice developed hyperglycemia at 2 wk after the
adoptive transfer of diabetogenic leukocytes. While the treatment
continued, the BGL declined and returned to normal 2 wk later (Fig. 9).
|
|
). Immunopathology
Sections of pancreatic samples from untreated NOD diabetic mice at
10 wk after the adoptive transfer of splenic leukocytes from acutely
diabetic NOD donors showed dense insulitis (Fig. 10A). Infiltrating
mononuclear cells consisted primarily of CD4+ T
cells, plus small numbers of macrophages and CD8+
T cells (data not shown); evidence of immune activation was
demonstrated by IL-2R expression by 25% of intraislet mononuclear
leukocytes (Fig. 10B). Sections from mice receiving adoptive
transfer of splenic leukocytes from untreated diabetic mice and therapy
with nonlytic IL-2/Fc also showed dense insulitis that was comparable
with that of control diabetic mice (Fig. 10C), and prominent
IL-2R expression by islet-infiltrating mononuclear cells (Fig. 10D). By contrast, pancreatic samples obtained from mice at
10 wk after adoptive transfer and treatment with lytic IL-2/Fc showed
only peri-islet accumulation of IL-2R-
mononuclear leukocytes (Fig. 10, E and F).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4–8 wk
after the cessation of treatment (15). In addition, the
short circulating t1/2 of DAB486-IL-2
fusion toxin together with the high immunogenicity of its diphtheria
toxin-related domain are not ideal (20). We have largely
overcome these obstacles, short circulating
t1/2 and immunogenicity, by
genetically fusing mIL-2 to a murine Fc2a to produce an IL-2/Fc2a
immunoligand.
This IL-2/Fc fusion protein possesses the biological functions of both
IL-2 and Fc2a. The IL-2 moiety binds to the IL-2R with essentially
the same affinity as rmIL-2 (Fig. 4) and vigorously supports CTLL-2
cell growth (Fig. 3). The Fc2a fragment of IL-2/Fc fusion protein
retains the property of the IgG2a murine isotype to bind effectively to
cells expressing high affinity FcRI (Fig. 6) and to activate
complement (Fig. 7). Thus, IL-2/Fc is able to facilitate ADCC and
CDC activities. Moreover, the Fc fragment ensures a prolonged
circulating t1/2 of 25 h,
200-fold longer than the t1/2 of rIL-2
(t1/2 = 2–9 min (21, 22)). Hence, cytolytic IL-2/Fc can be maintained at constant
circulating levels without resorting to continuous administration.
Furthermore, the murine IL-2/Fc has the promise of being minimally to
negligibly immunogenic, because the fusion protein consists of murine
protein sequences. In fact, we did not detect anti-IL-2/Fc Ab in
the serum of mice receiving multiple doses of IL-2/Fc (data not shown).
When the therapeutic application calls for a nonlytic IL-2-related
immunoligand, the amino acid residues necessary for FcR and C1q binding
to the Fc fragment can be substituted to yield long-lived molecules
essentially devoid of ADCC or CDC potential (Figs. 5 and 6).
The effects of IL-2/Fc were tested in a model of passive transfer of
diabetes in NOD mice. Because NOD mice have been reported to be
partially defective in ADCC and CDC immune effector mechanisms
(23, 24), the depleting anti-CD4 mAb GK1.5 was used to
assay Ab-dependent target cell lysis in NOD mice. As shown in Table I,
administration of GK1.5 reduced the level of CD4+
T cells from an initial 43 ± 4.18% of gated peripheral
lymphocytes to 11.45% ± 1.25 at 1 wk posttreatment. This finding is
consistent with the report of Wang et al. (25) that
administration of GK1.5 mAb diminishes peripheral
CD4+ cells and prolongs cultured islet allograft
in NOD recipients. However, the ADCC and CDC immune effector mechanisms
in NOD mice were impaired in comparison with BALB/c mice, in which the
same GK1.5 treatment reduced the peripheral CD4+
T cells to <4% of the PBLs at 1 wk posttreatment.
A short course of cytolytic IL-2/Fc treatment protects 70% of
irradiated NOD recipients from diabetes at 24 wk after the transfer
of 20 x 106 splenic leukocytes from
acutely diabetic female NOD mice. In contrast, 100% of control Ig and
untreated male NOD recipients developed diabetes 24 wk after the
transfer of splenic leukocytes from acutely diabetic female NOD mice
(Fig. 8). Moreover, NOD recipients treated with lytic IL-2/Fc remain
diabetes free for 
52 wk after adoptive transfer. These results appear
to be superior to those noted for DAB486-IL-2, which conferred
temporary (4–12 wk) protection in NOD recipients receiving adoptive
transfer of diabetogenic splenic leukocytes (15).
The adoptive transfer protocol dramatically accelerates T
cell-dependent diabetogenic autoimmunity. At 2 wk after an injection
with diabetogenic spleen cells, irradiated NOD mice become diabetic
(18, 26). Preventing the aggressive diabetogenic
autoimmune state induced by adoptive transfer of diabetic leukocytes
into adult irradiated recipients is considerably more difficult than
controlling the spontaneous disorder. For example, anti-MHC class
II mAb treatment, which can prevent spontaneous diabetes in NOD mice,
failed to alter the expression of IDDM in the passive transfer model.
In our past experiences, we have not observed a single spontaneous
remission of diabetes in NOD mice once hyperglycemia is evident.
It is notable that one of the IL-2/Fc-treated NOD mice developed
hyperglycemia 2 wk after the adoptive transfer of diabetogenic
leukocytes. While the treatment continued, the BGL declined and
returned to normal 2 wk later (Fig. 9). This observation hints that
lytic IL-2/Fc, as an antidiabetogenic therapy, might abolish the
diabetes if it is administered at an early stage of the disease
process. Clearly, further studies are warranted.
To explore the mechanisms of the antidiabetogenic effect of IL-2/Fc, we
have mutated the FcR binding and complement C1q binding domains of the
Fc fragment to render Fc incapable of directing ADCC and/or CDC
activity. Interestingly, 100% of irradiated male NOD recipients
treated with IL-2/Fc-/- developed diabetes at
16 wk after the transfer 20 x 106 splenic
leukocytes from acutely diabetic female NOD mice (Fig. 8). The
immunohistological results revealed that treatment with cytolytic
IL-2/Fc diminished IL-2R+ mononuclear leukocytes
within the peri-islet region in NOD mice compared with massive
infiltrates in which 25% of the infiltrating cells are
IL-2R+ in control and nonlytic IL-2/Fc-treated
NOD mice. These data strongly suggest that cytolytic IL-2/Fc blocks
diabetogenic autoimmunity in NOD mice by recruiting host immune
effector mechanisms to destroy IL-2R+ leukocytes
and not through competitive occupation of IL-2R by the IL-2 component
in the absence of IL-2R+ cell deletion. We have
reported previously that the IL-2R+ CD4 T cell
rich population is evidence of the "insulitis" of autoimmune NOD
female mice (11, 26). At an age of 5 wk, sparse
adventitial and peri-islet infiltrating IL-2R+
CD4 T cells were observed in one-third of untreated mice. By week 10,
all islets in all samples from untreated mice showed dense accumulation
of CD4 T cells, and the IL-2R+ cells accounted
for 5–10% of peri- and intraislet infiltrating leukocytes. At an age
of 20 wk, a massive leukocytic invasion obliterated the islets, with
10–20% of infiltrating cells expressing IL-2R (11, 26).
Thus, administration of this novel IL-2R-targeting reagent, cytolytic
IL-2/Fc, at an early stage of insulitis may be useful in preventing the
onset of diabetes in NOD mice. Because the "control" nonlytic
IL-2/Fc fusion protein has genetically engineered deletions in both the
complement C1q and FcRI binding sites, we were unable to distinguish
the individual contributions of direct complement-mediated lysis and
ADCC in this study. As NOD mice are genetically defective in both FcR
and complement C5 (23, 24), it will be worthwhile to
produce mutant IL-2/Fc proteins deficient in either complement C1q or
FcRI alone in future studies.
High affinity IL-2R is present on recently activated T cells but not on
resting or memory T cells (27). The selective targeting of
T cells bearing high affinity IL-2R is an attractive therapy for many T
cell-dependent disease processes. A variety of rodent mAbs directed
against the -chain of the IL-2R, as well as IL-2 fusion toxins, have
been used in animals and humans to achieve selective immunosuppression.
Here we report that we have developed a novel IL-2R-targeting agent, a
cytolytic chimeric IL-2/Fc fusion protein. This immunoligand binds
specifically and with high affinity to IL-2Rs, and it is capable of
recruiting host ADCC and CDC activities. The Ig component ensures an
extended circulating t1/2 of 25 h
following systemic administration. The striking antidiabetic effects,
together with its negligible potential for immunogenicity, forecast
that cytolytic IL-2/Fc may offer a new therapeutic approach for the
selective targeting of auto and alloimmune T cells.
|
Footnotes |
|---|
2 X.X.Z. and A.W.S. share co-first authorship.
3 Address correspondence and reprint requests to Dr. Terry B. Strom, Department of Medicine, Division of Immunology, Beth Israel Deaconess Medical Center, Research North, P.O. Box 15707, Boston, MA 02215. E-mail address:
4 Abbreviations used in this paper: NOD, nonobese diabetic; IL-2/Fc, cytolytic IL-2/Fc fusion protein; IL-2/Fc-/-, noncytolytic IL-2/Fc fusion protein; IDDM, insulin-dependent diabetes mellitus; ADCC, Ab-dependent cell-mediated cytotoxicity; CDC, complement-dependent cytotoxicity; BGL, blood glucose level; PAP, peroxidase-antiperoxidase; CHO, Chinese hamster ovary; mIL, murine IL; hIL, human IL.
Received for publication April 28, 1999. Accepted for publication July 22, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in long-surviving mouse heart allografts after brief CD4-monoclonal antibody therapy. Transplantation 59:559.[Medline]
This article has been cited by other articles:
|
|
 |
|
  C. S. Brandt, M. Baratin, E. C. Yi, J. Kennedy, Z. Gao, B. Fox, B. Haldeman, C. D. Ostrander, T. Kaifu, C. Chabannon, et al. The B7 family member B7-H6 is a tumor cell ligand for the activating natural killer cell receptor NKp30 in humans J. Exp. Med., July 6, 2009; 206(7): 1495 - 1503. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Benson, S. R. Dillon, E. Castigli, R. S. Geha, S. Xu, K.-P. Lam, and R. J. Noelle Cutting Edge: The Dependence of Plasma Cells and Independence of Memory B Cells on BAFF and APRIL J. Immunol., March 15, 2008; 180(6): 3655 - 3659. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Burden, P. Snoddy, R. J. Buick, J. A. Johnston, B. Walker, and C. J. Scott Recombinant cathepsin S propeptide attenuates cell invasion by inhibition of cathepsin L-like proteases in tumor microenvironment Mol. Cancer Ther., March 1, 2008; 7(3): 538 - 547. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Koulmanda, E. Budo, S. Bonner-Weir, A. Qipo, P. Putheti, N. Degauque, H. Shi, Z. Fan, J. S. Flier, H. Auchincloss Jr., et al. Modification of adverse inflammation is required to cure new-onset type 1 diabetic hosts PNAS, August 7, 2007; 104(32): 13074 - 13079. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Sanchez-Fueyo, M. Weber, C. Domenig, T. B. Strom, and X. X. Zheng Tracking the Immunoregulatory Mechanisms Active During Allograft Tolerance J. Immunol., March 1, 2002; 168(5): 2274 - 2281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. C. Li, A. Ima, Y. Li, X. X. Zheng, T. R. Malek, and T. B. Strom Blocking the Common {gamma}-Chain of Cytokine Receptors Induces T Cell Apoptosis and Long-Term Islet Allograft Survival J. Immunol., February 1, 2000; 164(3): 1193 - 1199. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
