| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Surgery, University of Florida College of Medicine, Gainesville, Florida 32610; 2 Amgen Incorporated, Thousand Oaks, California 91320-1799; and 3 Amgen Colorado, Boulder, Colorado 80301-2549
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Pharmacokinetics and immunogenicity of six different recombinant
human soluble p55 tumor necrosis factor (TNF) receptor I (sTNFR-I)
constructs were evaluated in juvenile baboons. The constructs included either an sTNFR-I IgG1 immunoadhesin (p55 sTNFR-I Fc) or five
different sTNFR-I constructs covalently linked to polyethylene glycol.
The constructs were administered intravenously three times, and
pharmacokinetics and immunogenicity were examined over 63 days. All of
the constructs were immunogenic, with the exception of a 2.6-domain
monomeric sTNFR-I. To evaluate whether the nonimmunogenic 2.6-domain monomeric construct could protect baboons against
TNF-
-induced mortality, baboons were pretreated with 1, 5, or 10 mg/kg body wt and were compared with baboons receiving either placebo
or 1 mg/kg body wt of the dimeric 4.0-domain sTNFR-I construct
(n = 3 each) before lethal Escherichia coli
bacteremia. The monomeric construct protected baboons and neutralized
TNF bioactivity, although greater quantities were required compared
with the dimeric 4.0-domain sTNFR-I construct. We conclude that
E. coli-recombinant-derived human sTNFR-I constructs can be
generated with minimal immunogenicity on repeated administration and
still protect against the consequences of exaggerated TNF- production.
immunoadhesin; sepsis; pharmacokinetics; inflammation
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
PROTEIN-BASED THERAPIES
AIMED at inhibiting the actions of tumor necrosis factor
(TNF)- currently include monoclonal antibodies and soluble TNF
receptor (sTNFR) constructs. These inhibitors have been or
are presently undergoing clinical evaluation in patients with a variety
of acute and chronic inflammatory diseases, such as sepsis syndromes,
rheumatoid arthritis, inflammatory bowel disease, multiple sclerosis,
and congestive heart failure. Although anti-TNF- therapies have not
proven successful in acute systemic inflammatory syndromes (1,
11), these same therapeutic approaches have shown efficacy in
patients with rheumatoid arthritis, inflammatory bowel disease, and
congestive heart failure (7, 17, 23) and are approved for
clinical use in rheumatoid arthritis and inflammatory bowel disease.
As the potential application for anti-TNF- therapies has shifted
from acute to chronic inflammatory diseases, however, concerns have
arisen regarding the antigenicity, safety, and altered pharmacokinetics of these constructs when administered repeatedly over extended periods
of time. A common approach for inhibiting the actions of TNF- has
been to employ protein constructs composed of the extracellular domains
of either the p55 type I (sTNFR-I) or p75 type II sTNFR (sTNFR-II).
Several years ago, our laboratory reported that infusion of the human
sTNFR-I could bind TNF- in vivo and attenuate the inflammatory
response to a lethal bacterial challenge (24). However,
use of native sTNFR-I was limited by the short half-life of the
monomeric form and the relative instability of the sTNFR-I/TNF-
plasma complexes that were formed.
Two different approaches have been employed to increase the biological
half-life of sTNFR-I and its capacity to neutralize homotrimeric
TNF-. By fusing the extracellular domains of either sTNFR-I or
sTNFR-II to the hinge and Fc region of a human immunoglobulin, novel immunoadhesins have been constructed with specificity for TNF-
and the pharmacokinetics of an immunoglobulin (2). An alternative approach has been to modify the amino acid sequence of the
sTNFR-I through site-directed mutagenesis and to covalently bind the
soluble receptor to polyethylene glycol (PEG). Our laboratory has
recently demonstrated that constructs composed of two sTNFR-I covalently linked to PEG [dimeric sTNFR-I or TNF binding protein (TNF-bp)] have extended plasma half-lives (20-30 h vs. 20 min) and increased TNF- neutralization activity compared with native, unPEGylated sTNFR-I (9, 20).
Although these initial efforts have resulted in constructs with much
longer plasma half-lives and increased biological activity, immunogenicity has remained a problem. In clinical trials with the
administration of an humanized monoclonal antibody against TNF-,
patients frequently developed antibodies; the appearance of antibodies
was associated with increased plasma clearance on repeated
administration (10). Similarly, an sTNFR-I immunoadhesin was withdrawn from clinical trials in rheumatoid arthritis in part
because of concerns regarding its immunogenicity (5, 15), although a similar sTNFR-II immunoadhesin has been shown to be only
modestly immunogenic (17). A dimeric, PEGylated sTNFR-I has also been evaluated in patients with rheumatoid arthritis, and,
although the construct has shown efficacy, immunogenicity has remained
a significant problem (18).
In a previous report, our laboratory examined the plasma half-life and
immunogenicity of three different PEGylated dimeric sTNFR-I constructs
that differed only in the number of functional domains of the
sTNFR-I present (20). Native sTNFR-I contains four
functional domains, and recombinant proteins were generated that
contained 4, 3, or 2.6 functional domains (20). All three constructs were equally effective in neutralizing TNF- and
protecting the juvenile baboon from lethality in bacteremic shock.
Although all of the dimeric constructs were immunogenic, the degree of immunogenicity increased with the number of sTNFR-I functional domains.
Furthermore, plasma clearance of the constructs was most rapid in those
animals that developed antibodies from the previous injection
(20).
The present study was a continuation of this investigational approach.
In the present report, we have evaluated the pharmacokinetics and
immunogenicity of six third-generation sTNFR-I constructs in the
healthy baboon at clinically relevant doses, in an effort to minimize
immunogenicity and maintain their in vivo capacity to neutralize
TNF-. Five of the constructs were PEGylated monomeric or dimeric
sTNFR-I, which varied in their number of functional domains
(2.6 or 4.0). These PEGylated sTNFR-I were compared with a 4.0-domain
sTNFR-I immunoadhesin, similar to one that our laboratory previously
reported (25). In the second phase of the study, a
PEGylated monomeric sTNFR-I construct shown to be nonimmunogenic in the
first phase was compared with a dimeric sTNFR-I construct for its
ability to neutralize TNF- at doses previously shown to protect
against lethal bacteremic shock in the naive baboon. Finally, the
baboon sTNFR-I was cloned, and its sequence was compared with human
sTNFR-I in an effort to extrapolate these results from the nonhuman
primate to the clinical setting.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Development of the sTNFR-I constructs.
A total of six sTNFR-I constructs were evaluated in this study (Table
1). Each was composed of defined regions
of the extracellular domain of the human sTNFR-I either fused to the Fc
and hinge region of a human immunoglobulin or covalently linked to PEG.
Construct I was a sTNFR-I IgG1 fusion protein (p55 sTNFR-I
Fc) that contained the complete extracellular domain of the human
sTNFR-I (4.0 domain) and the hinge and downstream constant domains of
the human IgG1 heavy chains placed carboxy-terminal to the sTNFR-I. The
structure of the immunoadhesin was similar to the sTNFR-I immunoadhesin that our laboratory previously examined in baboons (25),
although it was constructed independently.
|
Molecular cloning of primate TNFR-I soluble domain. Peripheral blood from baboons (Biomedical Resources Foundation, Houston, TX) was collected into heparinized tubes and stored at 4°C. The lymphocyte fraction of whole blood was isolated by centrifugation over density gradients (Ficoll-Paque Plus, Sigma Chemical, St. Louis, MO), following the manufacturer's recommended procedures. The resulting nucleated cells were washed in sterile phosphate-buffered saline and then lysed in guanidinium thiocyanate (Pharmacia, Piscataway, NJ). Total cellular RNA was isolated by layering the cell lysates onto cesium trifluoroacetate density gradients (Pharmacia) and then centrifuging them at 30,000 rpm for 24 h at 15°C. The RNA pellets were resuspended in RNase-free distilled water and then analyzed by gel electrophoresis to check integrity.
The protein coding region spanning the extracellular domain of baboon TNFR-I was amplified from peripheral blood total RNA by RT-PCR using the following primer pairs derived from the human TNFR-I sequence
|
|
|
|
In vitro TNF- neutralization activity of the individual
construct.
The ability of each of the six different sTNFR-I constructs to
neutralize TNF- bioactivity was initially determined in a cell-based
culture system. Very briefly, WEHI 164 clone 13 cells were incubated in
96-well microtiter plates with RPMI-1640 medium with 10%
heat-inactivated fetal calf serum and 1 µg/ml streptomycin at 37°C
with 5% CO2. Added to each well were increasing quantities of one of the six sTNFR-I constructs, ranging in concentration from 10 pg/ml to 100 µg/ml. Ten minutes later, 600 pg/ml of recombinant human
TNF- were added to each well, and cells were incubated an additional
18 h. Over the last 6 h, cell viability was determined by the
addition of 6 mg/ml of the tetrazolium salt,
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide.
After removal of the medium, cells were lysed with isopropanol and
diluted with distilled water. Absorbance at 570 and 690 nm was used to
estimate the number of viable cells remaining.
Experimental study 1. The initial baboon study was aimed at determining the immunogenicity and pharmacokinetics of the six different sTNFR-I constructs when repeatedly administered to healthy, juvenile baboons at doses that approximated their potential use in patients. Eighteen juvenile female baboons (Papio anubis; 2.0-2.9 kg) were purchased from Biomedical Research Foundation (Houston, TX). All animals were quarantined for a period of at least 4 wk at the Animal Resource Center of the University of Florida College of Medicine to confirm their good health and lack of transmissible disease. The studies were approved by the Institutional Animal Care and Use Committee of the University of Florida.
On day 0, after an overnight fast, all of the animals were anesthetized and instrumented, as previously described (20). During a 1-h equilibration period, hemodynamic parameters and blood samples were obtained. The animals were randomly assigned to one of six treatment groups, wherein each baboon received a bolus (30 s) intravenous infusion of 0.2 mg/kg of one of the six sTNFR-I constructs. The dose of the sTNFR-I constructs was chosen to approximate the quantities administered to patients [0.03, 0.10, and 0.30 mg/kg body wt (BW)] as a therapy for rheumatoid arthritis (18). In this clinical study, construct V had been administered to patients with refractory rheumatoid arthritis every 3 wk, and, although some efficacy was seen, significant immunogenicity was also observed. The present study was designed to mimic the clinical study and to determine whether any of the constructs had altered pharmacokinetics and immunogenicity. After dosing on day 0, the animals were observed for 8 h, during which time they received 0.9% sodium chloride (3 ml · kg
1 · h1) as
maintenance fluid. At the end of the 8-h period, all catheters were
removed, and the animals were returned to their cages after receiving analgesia.
On days 21 and 42, the baboons were
reanesthetized, and, after collection of a baseline venous blood
sample, each animal was administered the same dose (0.2 mg/kg) of the
same protein as on day 0.
Blood sampling for pharmacokinetics.
Venous blood samples were obtained predose, at 15 min, and on
days 1, 2, 3, 5,
8, 11, 16, and 21 after
each of the three intravenous injections (given on days 0,
21, and 42). All samples were anticoagulated with
EDTA or heparin and cooled on ice immediately after drawing. The plasma
fraction was separated by centrifugation and stored at 
80°C until assay.
Plasma sTNFR-I dimer ELISA analysis. Determination of plasma levels of each construct was performed by using an antigen capture ELISA developed using affinity-purified polyclonal antibodies to the 4.0-domain construct. The antibody was raised against recombinant human sTNFR-I in the goat and was affinity purified using dimeric 4.0-domain sTNFR-I. For each construct, plasma samples were run against a standard curve of the same test article. Ninety-six-well plates were coated with 4.0 µg/ml of goat anti-4.0-domain sTNFR-I for 2 h at 37°C. The plates were blocked with 2% BSA for 2 h at room temperature. After washing, the samples and the respective sTNFR-I construct standards (diluted in 20% goat serum in Dulbecco's phosphate-buffered saline, containing 0.05% Tween 20) were placed on the plate. After washing, the biotinylated primary antibody (goat anti-human dimeric sTNFR-I) was added to the plate. After washing again, streptavidin-horseradish peroxidase was added. The substrate solution contained 2.5 mg/ml 2,2-azino-bis(3-ethylbenzylthiazoline-6-sulfonic acid) and 0.05% H2O2. The plates were read at 405-490 nm on a Molecular Devices Vmax ELISA plate reader and fit to a four-parameter equation by Softmax software. The parameters from the equation were used to calculate the respective construct sTNFR-I concentrations in plasma samples. The lowest limit of detection for all sTNFR-I constructs in the assay, when corrected for dilution, was 0.490 ng/ml. All samples were baseline corrected using the prestudy value for each individual to account for endogenous levels of sTNFR-I. To ensure accuracy in the reported plasma levels of each construct in antibody-positive animals, plasma samples were analyzed at a variety of dilutions to ensure linearity.
Anti-sTNFR-I IgG antibody analysis.
The plasma samples were evaluated for the presence of anti-sTNFR-I
antibodies using an antibody-capture enzyme-linked immunoassay technique. Briefly, 96-well plates were coated with capture antigen (1 µg/ml) by incubating overnight at 4°C. The capture antigen for each
construct is summarized in Table 2.
|
Noncompartmental analysis. The parameters defining the plasma concentration time profile characteristics were determined by noncompartmental analysis using WinNonlin (version 1.1, Scientific Consulting, Lexington, KY).
Experimental study 2.
The second study was aimed at determining the relative efficacy of one
of the third-generation sTNFR-I constructs to neutralize TNF- in
vivo and prevent lethality against E. coli-induced
bacteremic shock, compared with a control dimeric construct
(construct V) that had previously been shown to be effective
at this dose (9, 20). Because of the high levels of
TNF- produced in response to the intravenously administered E. coli, the quantities of sTNFR-I required to protect the animals
are much greater than are those required to treat disease progression
in rheumatoid arthritis. Fifteen juvenile baboons were studied. Animals
were quarantined and instrumented as described in Experimental
study 1. After the animals were equilibrated for 1 h, they
were randomized to one of the five treatment groups to receive either
placebo (n = 3); 1 mg/kg BW of a dimeric 4.0-domain
sTNFR-I (construct V) (n = 3); or 1, 5, or
10 mg/kg BW of a monomeric 2.6-domain sTNFR-I (construct IV)
(n = 3 each).
1 · h1) were administered via the
antecubital fossa to all animals for 8 h. In addition, E. coli- septic animals received crystalloid resuscitation during the
first 8 h with 10 ml/kg BW of physiological saline administered
every 15 min, if they met two of the following three criteria:
1) a 30% decrease in mean arterial pressure; 2) a 30% increase in heart rate; and 3) a urine output <1.0
ml · kg1 · h1. At 8 h,
fluid resuscitation was withdrawn from all animals, the venous
catheters were removed, the percutaneous wounds were closed with
sterile sutures, and the animals were allowed to emerge from anesthesia.
Blood samples were obtained from the baboons at specified times (0, 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 7, and 8 h for cytokines, pharmacokinetics, and total and differential white blood cell counts).
Before the animals were returned to their cages, the baboons received
intramuscular 0.01 mg/kg BW of buprenorphine as an analgesic. In
addition, the animals received at this time and at 24 h, the intramuscular injection of 30 mg/kg BW of cephtriaxon. The animals were
returned to their cages, and subsequent survival to the lethal bacterial challenge was evaluated for 14 days. During this period, the
animals received appropriate analgesics (buprenorphine, 0.02 mg/kg BW
im every 24 h) to eliminate or minimize discomfort. At 24 and
48 h, and on days 3, 5,
8, 11, and 14, the animals were briefly anesthetized with ketamine HCl (10 mg/kg BW im), and venous blood samples were obtained.
Multisystem organ failure developed in some of these animals,
necessitating a premature euthanasia, as determined by the
Institutional Animal Care and Use Committee, University of Florida.
Euthanasia was performed in any animal that was deemed by the clinical
veterinarians (blinded to the treatment groups) to be suffering
excessive discomfort and unlikely to recover. Excessive discomfort that
required the termination of the study included any of the following:
1) failure to maintain a sitting or upright position over
the previous 12 h; 2) failure to take food or water
within the previous 12 h; 3) uncontrollable bleeding
from catheter sites; and 4) unresponsiveness to external
stimuli. Animals that met any of these criteria were promptly killed by
a pentobarbital sodium overdose (150 mg/kg BW iv). Animals euthanized
for animal welfare reasons were deemed to be nonsurvivors.
Animals surviving 14 days were anesthetized with 10 mg/kg ketamine HCl,
a venous blood sample was obtained, and baboons were euthanized as
described above.
Analytic methods.
Plasma TNF- activity was determined by both ELISA and plasma-based
bioassay. The TNF- sandwich ELISA employs a monoclonal antibody as
the capture and a polyclonal rabbit anti-TNF antiserum as the secondary
antibody. The ELISA can recognize both free TNF- and TNF- bound
to either of its sTNFRs (24), although the affinity for
TNF- bound to its shed receptor is reduced. The decreased affinity
was estimated by analyzing known quantities of TNF- in the ELISA
with concentrations of the different sTNFR-I construct at levels found
in the baboon plasma.
bioactivity was assessed by using the WEHI 164 clone 13 cytotoxicity assay (21), which detects only bioactive
protein. Interleukin (IL)-1
and IL-6 concentrations were measured by
ELISA as previously described (25).
Hematology and biochemical measurements. Complete blood counts were performed at defined intervals, and the number of circulating leukocytes per cubic millimeter was determined by Coulter counter (Coulter Electronics, Hialeah, FL). Platelet counts were determined in the same manner. Differential counts were obtained by examining at least 100 cells on a Wright stained peripheral blood smear.
Statistical analyses. Data are presented as means ± SD or SE. Differences in pharmacokinetics among the treatment groups at the same time points were determined by one-way analysis of variance. Changes in the pharmacokinetics with repeated injections of the constructs were analyzed by paired t-test. Differences in hemodynamics in response to E. coli and the different constructs were determined by two-way analysis of variance (time vs. treatment). Post hoc comparisons were performed by using the Student-Newman-Keuls multiple-range test. Significance was determined at the 0.05 level of confidence.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
In vitro neutralization studies.
All of the constructs effectively neutralized TNF- bioactivity as
determined in a cell-based bioassay (Fig.
1). The dimeric constructs
(constructs I, V, and VI), including
the dimeric immunoadhesin, were most effective at neutralizing TNF-.
Approximately 50% neutralization of TNF- bioactivity (to 600 pg/ml
of TNF-) was achieved with 5-10 ng/ml of each of the
constructs. In contrast, the monomeric constructs (constructs
II, III, and IV) were ~20-fold less
effective at neutralizing TNF- bioactivity on a weight basis, with
concentrations on the order of 100-200 ng/ml required to produce
50% neutralization of TNF-.
|
Study 1: Physiological analyses. Administration of all six sTNFR-I constructs to the healthy naive primates at a dose of 0.2 mg/kg BW was without any acute hemodynamic or hematological effect, as evaluated over the initial 8 h of constant monitoring. Mean arterial blood pressure, heart rate, core temperature, and urine output were unaffected by treatment with any of the constructs (data not shown). Furthermore, on days 21 and 42, when the animals were reinjected with additional quantities of the same constructs as administered on day 0, no adverse physiological responses were noted. Thus the constructs appeared to be safe in the otherwise healthy, anesthetized baboon.
Study 1: Pharmacokinetic analyses.
The parameters describing the plasma half-life of the six groups are
shown in Table 3. The elimination
half-life did not differ significantly among any of the PEGylated
sTNFR-I groups (constructs II-VI) after the first
injection, although it did tend to be longer for the monomeric
constructs. The half-life of the sTNFR-I immunoadhesin was
significantly longer than any of the PEGylated constructs
(P < 0.05).
|
Study 1: Antibody responses. All of the constructs were immunogenic in the baboon with the exception of the 2.6-domain monomeric sTNFR-I (construct IV) (Table 3). The strongest antibody responses were seen in the two groups of baboons that received either the 4.0-domain sTNFR-I immunoadhesin (construct I) or the PEGylated 4.0-domain dimeric sTNFR-I construct (construct V). In contrast, the most modest immunological responses were seen in baboons receiving the monomeric sTNFR-I constructs. In general, when antibody responses developed, they tended to become stronger after the second and third injection.
Study 2: Efficacy study.
Based on the results of the initial study, we concluded that the
PEGylated 2.6-domain monomeric sTNFR-I construct was nonimmunogenic in
the baboon when administered three times. To determine whether this
nonimmunogenic construct was as effective in blocking TNF--mediated injury as a dimeric 4.0-domain construct in naive animals, baboons were
pretreated with either the 4.0-domain dimeric sTNFR-I construct (construct V) at 1.0 mg/kg BW, or the monomeric sTNFR-I
construct (construct IV) at 1.0, 5.0, or 10 mg/kg BW
(n = 3 each) 1 h before the administration of
lethal quantities of E. coli. A fifth group received only
placebo injections before E. coli infusions. The choice of
the dimeric construct and the quantities employed were based on our
laboratory's prior studies (9, 20), which demonstrated that this dose of dimeric 4.0-domain sTNFR-I construct was protective in the baboons. The range of doses of the administered monomeric sTNFR-I construct was subsequently based on the relative neutralization capacity of two different constructs under in vitro conditions in the
WEHI 164 clone 13 cell-based assay.
Survival.
All three of the baboons pretreated only with placebo died or required
euthanasia (for animal welfare reasons) within 5-30 h after
E. coli challenge (Fig. 2). In
contrast, all three of the baboons treated with the 4.0-domain dimeric
sTNFR-I construct survived 14 days and were euthanized in good health
at the end of the study. Survival in baboons treated with the monomeric
2.6-domain sTNFR-I construct was dose dependent. One of three baboons
treated with 1.0 mg/kg BW survived, whereas two of the three baboons
treated with 5.0 mg/kg BW and all three of the three baboons treated
with 10 mg/kg BW of the monomeric 2.6-domain construct survived.
|
|
Plasma TNF- immunoactivity and bioactivity.
Administration of E. coli to the placebo-treated baboons
resulted in the rapid appearance of TNF- immunoactivity and
cytotoxicity that peaked within 90 min and declined thereafter (Fig.
4). By 4 h, all TNF- activity had
returned to baseline. In baboons treated with either the monomeric or
dimeric sTNFR-I constructs, TNF- immunoactivity was markedly
prolonged, and plasma TNF- immunoactivity remained elevated for at
least 8 h. However, it appears that the majority of the TNF-
immunoactivity was bound and inactive by the sTNFR-I because the
dimeric sTNFR-I construct reduced TNF bioactivity (cytotoxicity) and
the monomeric construct reduced TNF- bioactivity in a dose-dependent
fashion at the higher doses (5 and 10 mg/kg BW). At 14 days, none of
the surviving baboons that were administered the monomeric sTNFR-I
construct (construct IV) had developed antibodies to the
administered protein (data not shown).
|
Other cytokines.
Treatment with both the dimeric 4.0 domain (1 mg/kg BW) and the higher
doses (5 and 10 mg/kg BW) of the monomeric sTNFR-I constructs
significantly attenuated both the peak IL-1 and IL-6 concentrations
in the baboons, and the response to the monomeric sTNFR-I construct was
dose dependent (Table 4). At 10 mg/kg BW, the monomeric construct was as effective as the 4.0-domain dimeric construct. IL-8 concentrations were not affected by the treatments (data not shown).
|
Hematology responses. E. coli administration in the placebo-treated baboon produced a profound leukopenia that was sustained until the animal expired. Pretreatment with the dimeric 4.0-domain sTNFR-I construct did not prevent the neutropenia or lymphopenia that accompanied the response but significantly shortened the duration of neutropenia (Table 4). Similar results were seen in the surviving baboons treated with the monomeric 2.6-domain sTNFR-I construct.
Cloning of the baboon TNFR-I.
To help interpret the immunogenicity data from systemic administration
of various human sTNFR-I constructs into baboons, we obtained the cDNA
sequence of the baboon TNFR-I extracellular domain for comparing amino
acid identities. As shown in Fig. 5, the
baboon sTNFR-I domains were very highly conserved in these higher
primates. The cognate baboon amino acid sequence was ~97.4% identical, without gaps, with five relatively conservative changes located in domains 1, 3, and 4. This
same sequence was obtained either from individual clones or from
sequencing the purified PCR product. Furthermore, cDNA clones obtained
from a baboon tissue cDNA library produced an identical sequence (data
not shown), which confirms the accuracy of the deduced protein sequence
shown. For comparison, the amino acid homology between the mouse and human TNFR-I across this same region of the protein is only ~63%.
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The present study confirms that sTNFR-I constructs can be
developed that are safe, have extended biological half-lives compared with native sTNFR-I, have reduced nonimmunogenicity, and can block the
pathological effects of a systemic TNF- response. In fact, we have
identified one sTNFR-I construct that was nonimmunogenic in the baboon
when administered repeatedly over 2 mo at clinically relevant doses. In
addition to identifying an optimal compound for further clinical
investigation, the studies also provide important information about
factors determining immunogenicity and efficacy of these constructs in vivo.
The six different constructs employed in the present study differed in three major ways: 1) the sTNFR-I was covalently bound either to a human immunoglobulin (IgG1; immunoadhesin) or to PEG, 2) the number of functional domains of the extracellular region of the sTNFR-I was either 2.6 or 4.0, and 3) the PEG was covalently linked to either monomeric or dimeric sTNFR-I. Although a random block design was not employed, there were sufficient numbers of groups to ascertain the relative contribution of each variable to the clearance and immunogenicity in the baboon.
We can speculate that the use of PEG to extend the biological half-life
of the constructs was probably not responsible for the immunogenicity
of the constructs, as the immunogenicity between a dimeric 4.0-domain
immunoadhesin (IgG1) (construct I) and an identical dimeric
PEGylated construct were similar (construct V) (Table 3).
Rather, the data appear to suggest that increasing the number of
functional domains of the sTNFR-I and its presence in dimeric forms
were the primary contributors to immunogenicity. The most antigenic
forms of the construct were dimers composed of 4.0-domain sTNFR-I,
whereas the one construct that was nonimmunogenic was a monomeric
2.6-domain sTNFR-I. In a previous report (20), our
laboratory demonstrated that reducing the number of functional domains
from 4.0 to 2.6 also decreased the immunogenicity of a dimeric
construct, suggesting that some of the immunogenicity is directed
against this region of the molecule. Similarly, in patients receiving a
sTNFR-I immunoadhesin for rheumatoid arthritis, several of the
antibodies that developed against the immunoadhesin recognized epitopes
in the fourth functional domain (5). In a previous report,
our laboratory demonstrated that removal of the fourth and part of the
third domain had no effect on the construct's neutralization capacity
of TNF-, either under in vivo or in vitro conditions
(20). Similar results are reported here (Fig. 1). Crystal
structure of the sTNFR-I binding to homotrimeric TNF- confirms that
the major sites of interaction between ligand and receptor are in the
first and second functional domains (3).
Similarly, dimeric forms of sTNFR-I appeared to be more immunogenic as
a group than monomeric forms. This reduced immunogenicity to monomeric
sTNFR-I appears to be achieved with some loss of biological activity.
In the in vitro studies reported here, approximately 20 times as much
monomeric sTNFR-I was required to neutralize TNF- as the dimeric
construct. Similarly, in this acute, lethal model of bacteremic shock,
10 times as much monomeric sTNFR-I was required to protect the baboons
than a comparable dimeric construct.
TNF- circulates in a homotrimeric form, and receptor signaling
appears to involve simultaneous binding and complex formation of more
than one receptor (6). Neutralization of homotrimeric TNF- may be better accomplished by a single dimeric sTNFR-I
construct that could prevent multiple simultaneous receptor attachment. Early studies by Haak Frendscho et al. (12) demonstrated
that the TNF- neutralization capacity of a dimeric sTNFR-I
immunoadhesin was ~2 logs greater than that of the native sTNFR-I. In
our laboratory's earlier report (20), we observed that
dimeric sTNFR-I constructs were ~5- to 10-fold more effective than
monomeric sTNFR-I constructs at neutralizing
lipopolysaccharide-stimulated RAW 264.7 cell TNF-.
The in vivo results presented in experimental study 2 confirm these in vitro findings. Although the monomeric sTNFR-I construct was nonimmunogenic, doses on the order of 5- to 10-fold greater were required to produce a similar degree of protection against lethal bacteremia, as seen with the 4.0-domain dimeric sTNFR-I. Baboons treated with 10 mg/kg BW of the monomeric construct had comparable survival to that seen with baboons treated with 1 mg/kg BW of the dimeric sTNFR-I construct. Similarly, the capacity to reduce the hemodynamic changes, the proinflammatory cytokine levels (IL-1 and IL-6), and the hematological responses were similar between baboons treated with the dimeric sTNFR-I construct at 1.0 mg/kg BW and the monomeric sTNFR-I construct at 5.0 and 10 mg/kg BW. Doses of the monomeric sTNFR-I construct that were 50 times higher than in the initial study were also nonimmunogenic in the surviving E. coli-treated baboons.
The plasma TNF- response was surprisingly different among the
treatment groups (Fig. 4). Although both constructs significantly prolonged the appearance of immunological TNF- in the circulation, baboons treated with the dimeric 4.0-domain sTNFR-I construct had
essentially no detectable free TNF- bioactivity. In contrast, treatment with the monomeric 2.6-domain sTNFR-I construct appeared to
reduce the free TNF- activity in a dose-dependent fashion, albeit
not completely. Interestingly, low doses of the monomeric sTNFR-I
construct (1 mg/kg BW) appeared not to neutralize TNF- bioactivity,
and the plasma IL-1 and IL-6 responses were also unaffected. At
these low concentrations, the monomeric sTNFR-I construct may have
prolonged the life and bioactivity of the TNF-. All of the
physiological parameters evaluated (survival, leukocyte kinetics,
plasma IL-6 and IL-1), however, are consistent with a reduction in
TNF- pathology in vivo associated with the neutralization of TNF-
bioactivity as the dose of monomeric sTNFR-I was increased.
It should also be noted that this differential dose requirement between the dimeric 4.0-domain and monomeric 2.6-domain sTNFR-I constructs in the baboon has not been seen in rodent models of chronic inflammation. Bendele and colleagues (4, 16) have observed that, in a rat model of adjuvant arthritis, 4.0-domain sTNFR-I immunoadhesins and dimeric and monomeric PEGylated sTNFR-I constructs were equally effective in blocking the disease progression, and monomeric 2.6-domain sTNFR-I constructs were nonimmunogenic.
The question that naturally arises is whether these findings in primates can be extended to humans. We did not evaluate immunogenicity to repeated administration of the monomeric sTNFR-I construct at doses equivalent in neutralization activity to the dimeric construct. In contrast, we chose to evaluate immunogenicity at doses that are likely to be used in humans with rheumatoid arthritis (18). Nevertheless, evaluating immunogenicity with human proteins in primates is still limited by species specificity. To that end, we cloned the baboon sTNFR-I and compared the predicted amino acid sequence from the transcribed region of the TNFR-I cDNA of humans. The transcribed region for the extracellular domain of the baboon TNFR-I differed by only 5 amino acids out of 193 (97% homology). Because the human 2.6-domain monomeric sTNFR-I was not immunogenic in the baboon after three injections, even with two amino acid deviations, we believe that this strong homology argues that the same construct will also have reduced immunogenicity in humans. We cannot, however, conclude that the 4.0-domain constructs, which have greater immunogenicity in the baboon, will be similarly immunogenic in humans. However, significant immunogenicity has been seen with the 4.0-domain sTNFR-I immunoadhesins administered to patients with rheumatoid arthritis and multiple sclerosis (5). Of course, future clinical trials are required to determine whether this monomeric 2.6-domain construct will be suitable for the treatment of human disease.
| |
ACKNOWLEDGEMENTS |
|---|
This study was supported in part by National Institute of General Medical Sciences Grant GM-40586-13, by United States Public Health Service, and by a contract from Amgen, Inc.
| |
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: L. L. Moldawer, Dept. of Surgery, Univ. of Florida College of Medicine, PO Box 100286, Gainesville, FL 32610-0286 (E-mail: moldawer{at}surgery.ufl.edu).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 6 February 2001; accepted in final form 6 June 2001.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Abraham, E,
Glauser MP,
Butler T,
Garbino J,
Gelmont D,
Laterre PF,
Kudsk K,
Bruining HA,
Otto C,
Tobin E,
Zwingelstein C,
Lesslauer W,
and
Leighton A.
p55 Tumor necrosis factor receptor fusion protein in the treatment of patients with severe sepsis and septic shock. A randomized controlled multicenter trial. Ro 45-2081 Study Group.
JAMA
277:
1531-1538,
1997[Abstract].
2.
Ashkenazi, A,
Marsters SA,
Capon DJ,
Chamow SM,
Figari IS,
Pennica D,
Goeddel DV,
Palladino MA,
and
Smith DH.
Protection against endotoxic shock by a tumor necrosis factor receptor immunoadhesin.
Proc Natl Acad Sci USA
88:
10535-10539,
1991
3.
Banner, DW,
D'Arcy A,
Janes W,
Gentz R,
Schoenfeld HJ,
Broger C,
Loetscher H,
and
Lesslauer W.
Crystal structure of the soluble human 55 kd TNF receptor-human TNF beta complex: implications for TNF receptor activation.
Cell
73:
431-445,
1993[ISI][Medline].
4.
Bendele, AM,
McComb J,
Gould T,
Chilipala L,
Frazier J,
Colagiovanni D,
and
Edwards CK, III.
Comparative efficacy of sTNFR-I, a novel monomeric recombinant soluble TNF type I receptor to dimeric sTNFR-I and sTNFR-II IgG Fc fusion proteins in animal models of rheumatoid arthritis (Abstract).
J Interferon Cytokine Res
18:
89A,
1998.
5.
Christen, U,
Thuerkauf R,
Stevens R,
and
Lesslauer W.
Immune response to a recombinant human TNFR55-IgG1 fusion protein: auto-antibodies in rheumatoid arthritis (RA) and multiple sclerosis (MS) patients have neither neutralizing nor agonist activities.
Hum Immunol
60:
774-790,
1999[ISI][Medline].
6.
Darnay, BG,
and
Aggarwal BB.
Early events in TNF signaling: a story of associations and dissociations.
J Leukoc Biol
61:
559-566,
1997[Abstract].
7.
Deswal, A,
Bozkurt B,
Seta Y,
Parilti-Eiswirth S,
Hayes FA,
Blosch C,
and
Mann DL.
Safety and efficacy of a soluble P75 tumor necrosis factor receptor (Enbrel, etanercept) in patients with advanced heart failure.
Circulation
99:
3224-3226,
1999
8.
Ellison, JW,
Berson BJ,
and
Hood LE.
The nucleotide sequence of a human immunoglobulin C gamma 1 gene.
Nucleic Acids Res
10:
4071-4079,
1982
9.
Espat, NJ,
Cendan JC,
Beierle EA,
Auffenberg TA,
Rosenberg J,
Russell D,
Kenney JS,
Fischer E,
Montegut W,
Lowry SF,
Copeland EM, III,
and
Moldawer LL.
PEG-BP-30 monotherapy attenuates the cytokine-mediated inflammatory cascade in baboon Escherichia coli septic shock.
J Surg Res
59:
153-158,
1995[ISI][Medline].
10.
Feldmann, M,
Brennan F,
Paleoplog E,
Taylor P,
and
Maini R.
AntiTNF antibody therapy in rheumatoid arthritis: clinical results and mechanism of action (Abstract).
J Interferon Cytokine Res
18:
116A,
1998.
11.
Fisher, CJ, Jr,
Agosti JM,
Opal SM,
Lowry SF,
Balk RA,
Sadoff JC,
Abraham E,
Schein RM,
and
Benjamin E.
Treatment of septic shock with the tumor necrosis factor receptor: Fc fusion protein.
N Engl J Med
334:
1697-1702,
1996
12.
Haak Frendscho, M,
Marsters SA,
Mordenti J,
Brady S,
Gillett NA,
Chen SA,
and
Ashkenazi A.
Inhibition of TNF by a TNF receptor immunoadhesin. Comparison to an anti-TNF monoclonal antibody.
J Immunol
152:
1347-1353,
1994[Abstract].
13.
Hale, KK,
Smith CG,
Baker SL,
Vanderslice RW,
Squires CH,
Gleason TM,
Tucker KK,
Kohno T,
and
Russell DA.
Multifunctional regulation of the biological effects of TNF-alpha by the soluble type I and type II TNF receptors.
Cytokine
7:
26-38,
1995[ISI][Medline].
14.
Kohno, T,
Brewer MT,
Baker SL,
Schwartz PE,
King MW,
Hale KK,
Squires CH,
Thompson RC,
and
Vannice JL.
A second tumor necrosis factor receptor gene product can shed a naturally occurring tumor necrosis factor inhibitor.
Proc Natl Acad Sci USA
87:
8331-8335,
1990
15.
Lesslauer, W.
Recombinant TNF receptor fusion proteins. Design principles and clinical experiences with TNF neutralizing strategy (Abstract).
J Interferon Cytokine Res
18:
116A,
1998.
16.
McComb, J,
Gould T,
Chlipala E,
Sennelo G,
Frazier J,
Kieft G,
Seely J,
Edwards CK,
and
Bendele A.
Antiarthritic activity of soluble tumor necrosis factor receptor type I forms in adjuvant arthritis: correlation of plasma levels with efficacy.
J Rheumatol
26:
1347-1351,
1999[ISI][Medline].
17.
Moreland, LW,
Baumgartner SW,
Schiff MH,
Tindall EA,
Fleischmann RM,
Weaver AL,
Ettlinger RE,
Cohen S,
Koopman WJ,
Mohler K,
Widmer MB,
and
Blosch CM.
Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein.
N Engl J Med
337:
141-147,
1997
18.
Moreland, LW,
McCabe DP,
Caldwell JR,
Sack M,
Weisman M,
Henry G,
Seely JE,
Martin SW,
Yee CL,
Bendele AM,
Frazier JL,
Kohno T,
Cosenza ME,
Lyons SA,
Dayer JM,
Cohen AM,
and
Edwards CK.
Phase I/II trial of recombinant methionyl human tumor necrosis factor binding protein PEGylated dimer in patients with active refractory rheumatoid arthritis.
J Rheumatol
27:
601-609,
2000[ISI][Medline].
19.
Simonet, WS,
Lacey DL,
Dunstan CR,
Kelley M,
Chang MS,
Luthy R,
Nguyen HQ,
Wooden S,
Bennett L,
Boone T,
Shimamoto G,
DeRose M,
Elliott R,
Colombero A,
Tan HL,
Trail G,
Sullivan J,
Davy E,
Bucay N,
Renshaw-Gegg L,
Hughes TM,
Hill D,
Pattison W,
Campbell P,
and
Boyle WJ.
Osteoprotegerin: a novel secreted protein involved in the regulation of bone density.
Cell
89:
309-319,
1997[ISI][Medline].
20.
Solorzano, CC,
Kaibara A,
Hess PJ,
Edwards PD,
Ksontini R,
Abouhamze A,
McDaniel S,
Frazier J,
Trujillo D,
Kieft G,
Seely J,
Kohno T,
Cosenza ME,
Clare-Salzler M,
MacKay SD,
Martin SW,
Moldawer LL,
and
Edwards CK, III.
Pharmacokinetics, immunogenicity, and efficacy of dimeric TNFR binding proteins in healthy and bacteremic baboon.
J Appl Physiol
84:
1119-1130,
1998
21.
Solorzano, CC,
Ksontini R,
Pruitt JH,
Hess PJ,
Edwards PD,
Kaibara A,
Abouhamze A,
Auffenberg T,
Galardy RE,
Vauthey JN,
Copeland EM,
Edwards CK, III,
Lauwers GY,
Clare-Salzler M,
MacKay SL,
Moldawer LL,
and
Lazarus DD.
Involvement of 26-kDa cell-associated TNF-alpha in experimental hepatitis and exacerbation of liver injury with a matrix metalloproteinase inhibitor.
J Immunol
158:
414-419,
1997[Abstract].
22.
Su, X,
Zhou T,
Yang P,
Edwards CK, III,
and
Mountz JD.
Reduction of arthritis and pneumonitis in moth-eaten mice by soluble tumor necrosis factor receptor.
Arthritis Rheum
41:
139-149,
1998[ISI][Medline].
23.
Targan, SR,
Hanauer SB,
van Deventer DS,
Mayer L,
Present DH,
Braakman T,
DeWoody KL,
Schaible TF,
and
Rutgeerts PJ.
A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn's disease. Crohn's Disease cA2 Study Group.
N Engl J Med
337:
1029-1035,
1997
24.
Van Zee, KJ,
Kohno T,
Fischer E,
Rock CS,
Moldawer LL,
and
Lowry SF.
Tumor necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumor necrosis factor alpha in vitro and in vivo.
Proc Natl Acad Sci USA
89:
4845-4849,
1992
25.
Van Zee, KJ,
Moldawer LL,
Oldenburg HS,
Thompson WA,
Stackpole SA,
Montegut WJ,
Rogy MA,
Meschter C,
Gallati H,
Schiller CD,
Richter WF,
Loetscher H,
Ashkenazi A,
Chamow SM,
Wurm F,
Calvano SE,
Lowry SF,
and
Lesslauer W.
Protection against lethal Escherichia coli bacteremia in baboons (Papio anubis) by pretreatment with a 55-kDa TNF receptor (CD120a)-Ig fusion protein, Ro 45-2081.
J Immunol
156:
2221-2230,
1996[Abstract].
| ||||||||||||||||||||||||||||||||||||||||||||||
, Group I (4.0-domain immunoadhesin);
, group II (4.0 domain, monomeric 30-kDa
N105); 
, group III (4.0-domain monomeric
C105); ×, group IV (2.6 domain, monomeric N105); *,
group V (4.0 domain, dimeric C105); 
,
group VI (2.6 domain, dimeric C106). sTNFR, soluble TNF
receptor.
