| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Immunology |
Department of Anesthesiology and Pharmaceutical Chemistry, University of California, San Francisco, California 94110 [U. B. N., J. D. M.], and Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111 [G. P. A., L. M. W.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mr 155,000, these molecules exhibit
both a slow diffusion into tumors and a slow elimination from
circulation. The former property leads to heterogeneous delivery into
tumors, whereas the latter property results in dose-limiting
myelotoxicity. Recent advances in antibody-engineering technology has
led to the development of scFv molecules, composed of the variable
light (VL) and heavy (VH)
domains of an immunoglobulin molecule (1
, 2) . These small
scFv molecules currently represent the minimal antibody-based construct
capable of specifically interacting with antigen without excessive
cross-reactivity. In both mice and patients their size leads to a rapid
renal elimination, yielding excellent tumor:normal organ ratios and
lower nonspecific background, as compared with intact IgG antibodies
(3, 4, 5)
. In addition, the scFv penetrates more deeply into
poorly vascularized regions of tumors than do the Fab',
F(ab')2, and intact IgG (6)
.
However, the monovalent nature and rapid renal clearance of the scFv
results in the specific retention of only small quantities in the
tumors in immunodeficient mice with rarely >1%ID localized per gram
of tumor at 24 h after injection (5, 6, 7, 8)
. We have
recently examined the tumor-targeting properties of a series of scFv
mutants that vary in affinity for the same epitope of the tumor antigen
ErbB2. In this model, the 24-h tumor retention of a scFv with an
affinity of 133 nM (C6G98A) was indistinguishable
from that achieved with an irrelevant scFv. Increasing the affinity to
25 nM (C6.5) and 1 nM
(C6ML3–9) resulted in significantly greater tumor retentions of
0.8%ID/g and 1.4%ID/g, respectively (9)
.
We, and others, have investigated the use of larger, multivalent,
scFv-based constructs to improve the degree and specificity of in
vivo targeting of solid tumors (5
, 10, 11, 12, 13)
. In
general, increasing the number of antigen binding sites has led to
enhanced tumor retention, as compared with that achieved with
monovalent scFv molecules. One of the more promising scFv-based
molecules is a noncovalent dimer or diabody (14)
.
Diabodies are constructed by shortening the scFv peptide linker from 15
aa to 5 aa. The shorter linker does not permit pairing of the
VH and VL domains on the
same polypeptide chain, forcing pairing between complementary domains
of two different chains. The resulting molecule has two antigen-binding
sites at opposite ends of the molecule, separated by 65 Å
(15)
. We have previously reported on the construction of a
diabody from the C6.5 scFv, which specifically recognizes ErbB2. In
tumor-bearing mice the C6.5 diabody exhibited a >7-fold increase in
tumor retention without the loss of targeting specificity
(16)
. To date, however, the relative impact of increased
size, increased valance, and the affinity of the parental scFv
molecules on the tumor-targeting properties of scFv dimers has yet to
be elucidated.
In this study, we analyze the importance of intrinsic antibody affinity on the in vitro and in vivo targeting of ErbB2 overexpressing tumor cells using a series of diabodies constructed from the three affinity variants of the C6.5 anti-ErbB2 scFv (spanning a 133-fold range of affinity) described above. By applying SPR technology to the analysis of antigen binding, the intrinsic as well as apparent bivalent binding kinetics of the three diabodies were determined. Diabody binding to tumor cells was investigated by fluorescence cytometry using the ErbB2-positive breast cancer cell line SKOV3 to estimate equilibrium constants. Finally, biodistribution studies were performed in scid mice bearing established SKOV3 tumors.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cells were harvested by centrifugation (4000 x g, 20 min), and the pellets were resuspended on ice for 30
min in periplasmic extraction buffer [30 mM
Tris, 2 mM EDTA, and 20% sucrose (pH 8.0)]
containing 100 µg/ml DNase. Bacteria were pelleted by centrifugation
at 5000 x g for 20 min, resuspended in
osmotic shock buffer (5 mM
MgSO4), and incubated for another 20 min on ice.
Bacteria were pelleted (7000 x g, 20 min),
and supernatants from the periplasmic extraction buffer and
MgSO4 fractions combined and cleared by
centrifugation at 10,000 rpm for 30 min at 4°C. The
resulting solution was dialyzed in PBS. All molecules were purified by
immobilized metal affinity chromatography (17)
, followed
by size exclusion chromatography on a BioCAD SPRINT fast protein liquid
chromatography system (PerSeptive Biosystems) using either a
Superdex 75 (for scFv) or a Superdex 200 column (for diabodies).
Protein concentrations were determined spectrophotometrically from the
absorbance at A280 using the extinction
coefficient 
= 1.4. The C6.5 scFv-Fc fusion protein
was expressed from Pichia pastoris and purified using
protein G affinity chromatography, as described elsewhere
(20)
.
Measurement of Binding by SPR.
Association rate constants (kon) were determined
using SPR in a BIAcore1000 (BIAcore Inc.). Approximately 500 RU of the
ErbB2 ECD were coupled to a CM5 sensor chip as described previously
(17)
, and association rate constants were measured under
continuous flow of 15 µl/min using scFv and diabody concentrations
ranging from 100-1200 nM. Association rate constants were
calculated from a plot of [ln(dR/dt)]/t versus
concentration of binding sites using the BIAanalysis software (version
2.1). Apparent dissociation rate constants (koff)
were determined using the function BIGinjection. Different volumes
[600 µl, 330 µl, 100 µl, 50 µl, and 5 µl of diabody or scFv
solutions (concentration, 25 µg/ml)] were injected over the CM5
sensor chip (500 RU ECD immobilized) at a flow rate of 5 µl/min. The
dissociation rate constants of all molecules were determined at >90%
of maximal binding to the chip, with the exception of the C6G98A and
C6.5 scFv, which were measured as close to maximal binding as possible
(>50%). To determine intrinsic rate constants, diabodies were
biotinylated with NHS-LC-biotin (Pierce Chemical Co.) at a
biotin:diabody ratio of 5:1 and as described by the manufacturer.
Approximately 5000 RU avidin (Sigma Chemical Co.) was conjugated to a
CM5 sensor chip using similar conditions as described for ErbB2 ECD
(17)
. Biotinylated diabody was injected onto the surface
to yield 500 RU diabody bound to the surface. Saturating
concentrations of ErbB2 were then injected, and dissociation rate
constants were determined immediately following ErbB2 ECD dissociation.
To determine their serum stability, diabodies were incubated in 90%
human serum at a final concentration of 50 µg/ml for 3 days at
37°C. After diluting 10-fold in running buffer, the binding
concentration was determined by SPR using immobilized ErbB2 ECD as
described (18)
and compared with that of the diabody stock
stored at 4°C.
Cell Surface Binding Measurements.
Human ovarian carcinoma SKOV3 cells (HTB 77; American Type Culture
Collection) that overexpress ErbB2 were grown to 80–90% confluence in
RPMI media supplemented with 10% FCS and harvested by trypsinization.
Each scFv or diabody was incubated in triplicate with 1 x 105 cells in 96-well plates with V-shaped
wells for 2 h at the concentrations indicated. Cell binding was
performed at room temperature in PBS containing 2% FCS and 0.1%
sodium azide in a total volume of 200 µl. Sodium azide was included
in the incubation buffer to minimize receptor internalization. After
two washes with 200 µl of PBS, bound scFv or diabody was detected by
the addition of 100 µl (10 µg/ml) of FITC-labeled anti-FLAG
monoclonal antibody clone M1. After incubating 30 min at room
temperature, the cells were washed twice and resuspended in PBS
containing 4% paraformaldehyde. Fluorescence was measured by flow
cytometry in a FACSort (Becton Dickinson), and median fluorescence (F)
was calculated using Cellquest software (Becton Dickinson) and the
background fluorescence was subtracted. Equilibrium constants were
determined as described (21)
, except that values were
fitted to the equation 1/F = 1/Fmax + (KD/Fmax)(1/[scFv]) using
the software program SigmaPlot (SPSS Inc.).
Biodistribution Studies.
Diabody and scFv molecules were radiolabeled with
125I using the chloramine T method
(125I: protein ratio, 1:10), as described
previously (5)
. The quality and immunoreactivity of the
radiopharmaceuticals were evaluated by SDS-PAGE and in a live
cell-binding assay as described (5)
. CB.17 Icr
scid mice, 6–8 weeks of age, were obtained from the Fox
Chase Cancer Center Laboratory Animal Facility. SKOV3 cells
(2.5 x 106) were implanted s.c.
on the abdomen of each mouse. When the tumors had achieved a size of
50–200 mg (8 weeks), Lugol’s solution was placed in their drinking
water to block thyroid accumulation of radioiodine, and biodistribution
studies were initiated. Twenty micrograms (100 µl) of radioiodinated
diabody or scFv were administered by i.v. tail vein injection to each
mouse. Cohorts of five mice that received the
125I-diabodies or scFv were sacrificed at 1, 4, 8
(except C6.5db), 24, 48, and 72 h after injection. The mean and
SEM of retention of each radiopharmaceutical in tissue (%ID/g) and
blood (%ID/ml) were determined from decay-corrected counts, as
described (5)
. Calculations of the estimated cumulative
localization (AUC) of diabody in tumor (% h-1
g-1) and blood (% h-1
ml-1) were determined using the NCOMP program
(22)
.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In Vitro Binding Kinetics of scFv and Diabodies.
The association and dissociation rate constants of the three scFv were
remeasured using SPR, and the KD was calculated
as koff/kon. The
KD of the three scFv were comparable with values
previously reported and spanned a 133-fold range of affinities (Table 1)
. Intrinsic diabody association and dissociation rate constants were
measured to determine whether construction of diabody molecules
affected the antigen binding. The association rate constant was
determined by immobilizing the ECD on the sensor chip surface. The
intrinsic dissociation rate constants of each diabody were determined
by immobilizing biotinylated diabody on an avidin-coated sensor chip.
Because the recombinant ErbB2 ECD is monomeric in solution (results not
shown), by immobilizing the diabody, bivalent binding is not possible
and the dissociation rate constant represents that of the monovalent
binding. The diabody association rate constant and the intrinsic
dissociation rate constant were determined, and the intrinsic
equilibrium-binding constants were calculated. These were slightly
lower (2–3-fold) than the values measured for the scFv from which they
were constructed, mainly as a result of decreased association rate
constants (Table 1)
. Dissociation rate constants were similar to those
of the scFv, indicating that the diabody homodimer formation does not
significantly alter ligand binding to the individual binding site. The
dissociation rate constant for the C6.5 scFv determined by the same
approach was similar to what was determined by immobilizing ECD
(results not shown).
|
As expected, the rate of dissociation decreased with increasing
incubation time (Fig. 1A)
. This experiment was repeated for the C6.5 and C6ML3–9
scFv and diabody, and the results were plotted as dissociation rate
versus incubation time (Fig. 1, B
–D).
In these experiments, the apparent bivalent equilibrium constants of
diabodies decreased with increased association time (Fig. 1)
. For
C6ML3–9db, the diabody with the highest intrinsic affinity, the change
in dissociation rate was only minimal over the 2-h examination period
(Fig. 1D)
. However, for the C6G98Adb, which has the lowest
intrinsic affinity, the dissociation rate dropped dramatically during
the first 5 min of association and then stabilized (Fig. 1B)
. Similarly, for the C6.5 diabody, the dissociation rate
stabilized after about 70 min of association (Fig. 1C)
.
These results indicate an inverse relationship between the dissociation
rate constant of a bivalent molecule and the time required to achieve
bivalent binding.
|
The dissociation rate constants of the diabodies obtained after 2 h of association are reported in Table 2
. The diabody apparent equilibrium constants were then calculated as
kon/koff (after 2 h of
binding; Table 2
). Not surprisingly, the apparent affinity was
significantly greater than the intrinsic affinity. The magnitude of the
increase in affinity, however, was inversely proportional to the
intrinsic affinities of the molecules (Tables 1
and 2)
. For the lowest
affinity diabody, C6G98Adb, the increase in affinity mediated by
bivalent binding was 51-fold, from 409 nM to 8
nM. Similarly, for the C6.5 diabody the affinity increased
21-fold from 34 nM to 1.6 nM as a result of
bivalent binding. For the diabody with the highest affinity, C6ML3–9,
the increase in apparent affinity was only 5.6-fold, from 2.0
nM to 0.36 nM. This relationship was also
observed when comparing the increase in apparent affinity of the
diabody to the affinity of the parental scFv (17-fold, 16-fold, and
2.8-fold for C6G98A, C6.5, and C6ML3–9, respectively; Table 2
).
Differences between the increment in apparent KD
seen for scFv versus diabody are due to minor differences in
the intrinsic association and dissociation rate constants that resulted
from conversion of the scFv to diabody format (Table 1)
.
|
. Because direct fluorescent labeling of antibody fragments is often
associated with decreased affinity, diabodies were genetically tagged
with an affinity-matured version of the FLAG tag (DYKAK; Ref.
19
) and detected with FITC-labeled anti-FLAG M1. We
determined that the affinity of the anti-FLAG antibody for the FLAG
peptide tag on the diabody constructs was 2 nM (results not
shown) and, thus, was similar to what has previously been published for
its interaction with the synthetic peptide (19)
. This
high-affinity interaction ensured that the secondary antibody could be
used at saturating concentration. As described by Benedict et
al. (21)
, these conditions are required for the
measured fluorescence to be directly proportional to the amount of
antibody fragment bound.
|
and Fig. 2
). The correlation indicates that the diabodies, despite
their relatively rigid structure (15)
, are able to bind
two ligands on the cell surface simultaneously, as observed in the more
artificial SPR analysis. As with the case of equilibrium constants
measured by SPR, bivalent diabodies had significantly higher apparent
affinities than the parental scFv. Similarly, the increment in
equilibrium constant for bivalent diabody was greatest for the lowest
affinity scFv C6G98A (65-fold) and least for the highest affinity scFv
(7.7-fold).
|
and is similar to
that of the C6.5 diabody (3.9 nM
versus 3.4 nM, respectively). The
results indicate that differences in binding site flexibility and size
of the molecules for these two bivalent molecules does not
significantly affect the affinity for binding to cell surface antigens.
This result also suggests that the relationship observed between
monovalent and bivalent equilibrium constants observed for diabodies is
likely to hold for other bivalent antibody constructs.
Biodistribution of Diabodies in scid Mice Bearing
ErbB2-overexpressing Tumors.
The relevance of the in vitro observations to in
vivo tumor targeting was determined by measuring the
biodistribution of the three diabodies and the C6.5 scFv in
scid mice bearing s.c. SKOV3 tumors overexpressing the ErbB2
antigen. The tumor, blood, and organ retention of radioiodinated scFv
and diabody molecules were determined at 1, 4, 24, 48, and 72 h
after i.v. administration. As expected, the larger size (50 kDa) of the
diabody constructs resulted in a prolonged blood retention as compared
with that seen with the smaller (25 kDa) C6.5 scFv molecule (Fig. 3, B, C, and D
versus
A). This is reflected in the 4–5-fold greater blood AUC
values for the diabody molecules as compared with C6.5 scFv (Table 4)
. The calculated t1/2 
for C6.5 scFv and
diabody were 0.23 h and 0.67 h; the calculated
t1/2 ß were 5.70 and 6.42 h, respectively.
Diabodies exhibited significantly greater (2–5-fold) quantitative
tumor retention at 24 h than was achieved with the highest
affinity scFv studied (Fig. 3
and Table 4
). This likely results from a
combination of a higher apparent affinity, because of bivalent binding,
and a slower serum clearance. Calculations of the cumulative residence
of the radioiodinated diabodies and scFv, expressed as AUC, were
determined. These were also significantly greater for diabodies
compared with scFv (Table 4)
. Importantly, the difference in apparent
affinity between diabodies did not significantly alter the quantitative
tumor retention or tumor:blood ratios. In fact, the tumor AUC,
tumor:blood AUC, and the 24-h tumor retention of the two lower-affinity
diabodies (C6G98A and C6.5) were 2-fold better than for the
high-affinity C6ML39 diabody.
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mr 50,000, they are small enough to
be rapidly eliminated from the circulation via first-pass renal
clearance. Because diabody size and molecular structure is similar to
Fab fragments, it is expected that diabodies will readily penetrate
from blood vessels into solid tumors, as reported for Fabs
(6)
. The divalent nature of the interaction of diabodies
with cell surface tumor antigen is widely recognized as being important
in maintaining prolonged residence in tumors. What has been less clear
is the precise relationship between the intrinsic affinity of the
binding site, the increase in apparent affinity due to bivalent
binding, and the impact of higher-affinity binding on in
vivo tumor targeting. The studies presented here were designed to examine the effect of affinity of a series of bivalent diabodies on their antigen-binding kinetics and their in vitro and in vivo tumor-targeting properties. Because all of the constructs were the same size and recognized the same epitope of ErbB2, any observed differences likely resulted solely from the differences in binding affinity. The role of valency on the impact of affinity was readily apparent when we compared the scFv and diabody constructs. The C6.5 scFv and its affinity mutants differ from each other by only one to three amino acid residues, yet differ in affinity for the same epitope of ErbB2 by 133-fold (18) . However, when diabodies were constructed from the scFv, the resulting difference in affinity was reduced to only 22-fold. Most importantly, the greatest increment in affinity was observed for the diabody constructed from the lowest affinity scFv. Clearly, the kinetics of interaction is dependent on more than the straightforward additive impact of the individual affinities of the binding sites.
The equilibrium between a bivalent antibody and its antigen has often been depicted as a two-step reaction, involving free antibody and antigen as well as antibody, monovalently or bivalently complexed to its antigen. In this model, the association occurs in two steps. In the first reaction, the antibody monovalently binds to a single antigen before encountering a second antigen, after which the interaction can become bivalent. Whereas the rate of first reaction is determined by the association rate constant of the monovalent antibody arm, the second rate is dependent on external factors such as the density and fluidity of the antigen in the cell membrane (25) and the radius spanned by the antibody. In theoretical models of antibody interactions with cell surface antigen, it is often assumed that the antigen is in excess and the rate of bivalent binding solely depends on antigen diffusion (25 , 26) . In clinical use, however, large doses of antibody are used, potentially resulting in an excess of antibody at the binding site and the possibility of significant quantities of antigen bound monovalently.
To understand the dynamics of the binding kinetics, we studied the time dependence of bivalent binding of diabodies to ErbB2 by SPR under the conditions of high diabody concentration that might be expected in regions of tumor proximal to blood vessel. The results indicated that a large fraction of diabodies initially bind to only one antigen. Under these conditions, the bivalent dissociation rate constant decreased with increased binding time and the decrease in dissociation rate constant was inversely proportional to that of the monovalent interaction. Whereas the dissociation rate of the diabody with the lowest affinity rapidly stabilized at a 15-fold lower rate after 2 h of association, the decrease in the bivalent dissociation rate constant for the C6ML3–9 diabody was only 3-fold. One possible explanation for this result is that diabodies with lower intrinsic equilibrium constant can more rapidly achieve bivalent binding, because those bound monovalently dissociate rapidly freeing up antigen for bivalent binding by neighboring diabodies. Diabodies with higher intrinsic equilibrium constants dissociate more slowly from antigen and, thus, can prevent bivalent binding of neighboring diabodies. This effect has not been previously taken into account in theoretical models of bivalent binding.
The actual dynamics of the interaction between antibody and cell surface antigens in the tumor is, however, much more complex. The ability of IgG to extravasate and penetrate into tumor is severely limited by both the size of the antibody and the high hydrostatic pressure in the tumor resulting from a lack of draining lymphatics (27) . This results in a very uneven distribution of the antibody, ranging from a situation of antibody excess in areas adjacent to the blood vessels to antigen excess in regions distant from the vasculature. Despite their improved tumor penetration properties, similar gradients will probably result from the administration of smaller scFv and diabody molecules (6) .
In our study, the 22-fold difference in affinity (as determined by SPR) between the three diabodies did not result in greater quantitative tumor retention or tumor:blood ratio. In fact, the diabody constructed from the lowest affinity scFv (C6G98A) exhibited comparable tumor targeting to the diabody constructed from the higher-affinity C6.5 scFv and better targeting than the diabody constructed from the highest affinity C6ML3–9 scFv. Interestingly, the C6G98A scFv does not target tumor better than an irrelevant control scFv (9) . We conclude that above a threshold affinity, other factors determine the quantitative tumor delivery of a bivalent antibody fragment. This is consistent with in vivo targeting results observed for the three C6.5-based scFv (9) . This threshold affinity may be partly attributed to tumor physiology rather than simple antigen-binding kinetics. Indeed, barriers other than the antibody fragment size may exist in tumor tissue that restrict their penetration to areas distal from the blood vessels. Fujimori et al. (28) have postulated that high-affinity antibodies will not successfully penetrate deeply into tumors due to a binding site barrier effect, in which interaction with the first antigen encountered at the periphery of the tumor will block further diffusion of the antibody into the tumor. We have investigated the tumor penetration of the monovalent scFv used in this study. Whereas increasing the affinity improves the selective targeting of scFv to solid tumors (9) histochemical staining for scFv in the tumor xenografts supports the theory of Fujimori et al. (28) .4 This may explain why the C6ML3–9 diabody had significantly worse tumor-targeting properties than the other two diabodies.
In tumor-bearing immunodeficient mice, the 24-h tumor retentions of all three diabodies were superior to that of the highest affinity scFv. The differences in biodistributions of the diabodies and the scFv, thus, cannot be solely attributed to their KD. This is apparent when comparing the tumor retention at 24 h of the C6G98A diabody to that of the higher-affinity C6ML3–9 scFv (7.1 versus 1.4%ID/g tumor and KD = 5.6 nM versus 3.8 nM for the C6G98A diabody and C6ML3–9 scFv, respectively, for binding to cells). Clearly, the longer blood retention of the larger diabody molecules may account for some of the increased tumor retention because the diabodies would be expected to have more opportunities to perfuse the tumor and interact with target antigen. In addition, quantitative tumor localization may be affected by differences between kinetic versus equilibrium control of binding. The C6G98A diabody, with each binding site having a rapid dissociation rate constant, may be able to more easily dissociate from antigen and percolate through the tumor compared with a high-affinity scFv where the (single) binding site has a slower dissociation rate constant.
Differential effects of antibody fragment size, binding rates, and equilibrium constant on tumor penetration may explain the differences between our results and those of Viti et al. (29) . In those studies, biodistributions of low-affinity (KD = 41 nM) and high-affinity (KD = 0.054 nM) scFv and their diabody dimers were studied in xenografted mice whose tumors expressed the neovascular antigen fibronectin. In contrast to our results, the higher-affinity scFv exhibited greater tumor retention than the diabody constructed from the lower-affinity scFv (4-fold higher %ID/g tumor at 24 h). In their model, tumor penetration is not an issue because the antigen is in the vasculature, whereas we studied an epithelial antigen where penetration will have a dramatic effect on antibody localization. A strict comparison of results between the two systems is not possible because: (a) Viti et al. (29) did not measure the apparent affinities of the diabodies; and (b) the "diabodies" had normal length linkers and, thus, could reequilibrate to mixtures of monomer and dimer after gel filtration and before injection into mice.
On the basis of our results, it is apparent that the construction of bivalent diabodies, even from low-affinity scFv, can lead to the generation of tumor-targeting agents that are superior to those achieved through the cumbersome processes involved in affinity maturing monovalent scFv molecules. This observation may have a significant impact on the design of future multivalent antibody-based molecules for cancer therapy.
|
FOOTNOTES |
|---|
1 Supported by National Cancer Institute Grants
CA65559 and CA06927; Department of Defense Grants DAMD17-98-1-8189,
DAMD17-98-1-8307, and DAMD17-94-J-4433; an appropriation from the
Commonwealth of Pennsylvania, the Bernard A. and Rebecca S. Bernard
Foundation, the Frank Strick Foundation, and the CaPCURE Foundation. 
2 To whom requests for reprints should be
addressed, at Department of Anesthesiology and Pharmaceutical
Chemistry, San Francisco General Hospital, University of
California–San Francisco, Room 3C-38, 1001 Potrero Avenue, San
Francisco, CA 94110. E-mail: Marksj{at}anesthesia.ucsf.edu
3 The abbreviations used are: Mab, monoclonal
antibody; scFv, single-chain Fv; %ID, percentage of the injected dose;
SPR, surface plasmon resonance; RU, resonance unit; ECD, extracellular
domain; AUC, area under the curve; aa, amino acid.
4 G. P. Adams, R. Schier, A. M. McCall, H.
Simmons, E. M. Horak, R. K. Alpaugh, J. D. Marks, and L. M. Weiner.
High affinity restricts the localization and tumor penetration of
single chain Fv antibody molecules, submitted for publication.
Received 3/27/00. Accepted 9/13/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. El Emir, U. Qureshi, J. L.J. Dearling, G. M. Boxer, I. Clatworthy, A. A. Folarin, M. P. Robson, S. Nagl, M. A. Konerding, and R. B. Pedley Predicting Response to Radioimmunotherapy from the Tumor Microenvironment of Colorectal Carcinomas Cancer Res., December 15, 2007; 67(24): 11896 - 11905. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hasegawa, C. Hu, T. Nakamura, J. D. Marks, S. J. Russell, and K.-W. Peng Affinity Thresholds for Membrane Fusion Triggering by Viral Glycoproteins J. Virol., December 1, 2007; 81(23): 13149 - 13157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.-B. Kim, Z. Wang, Y. Y. Liu, S. Stavrou, A. Mathias, K.J. Goodwin, J. M. Thomas, and D. M. Neville A fold-back single-chain diabody format enhances the bioactivity of an anti-monkey CD3 recombinant diphtheria toxin-based immunotoxin Protein Eng. Des. Sel., September 1, 2007; 20(9): 425 - 432. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Shively 18F Labeling for Immuno-PET: Where Speed and Contrast Meet J. Nucl. Med., February 1, 2007; 48(2): 170 - 172. [Full Text] [PDF]
|
|
|
|
|
|
W. Cai, T. Olafsen, X. Zhang, Q. Cao, S. S. Gambhir, L. E. Williams, A. M. Wu, and X. Chen PET Imaging of Colorectal Cancer in Xenograft-Bearing Mice by Use of an 18F-Labeled T84.66 Anti-Carcinoembryonic Antigen Diabody J. Nucl. Med., February 1, 2007; 48(2): 304 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-Y. Xiong, A. Natarajan, X.-B. Shi, G. L. Denardo, and S. J. Denardo Development of tumor targeting anti-MUC-1 multimer: effects of di-scFv unpaired cysteine location on PEGylation and tumor binding Protein Eng. Des. Sel., August 1, 2006; 19(8): 359 - 367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Orlova, M. Magnusson, T. L.J. Eriksson, M. Nilsson, B. Larsson, I. Hoiden-Guthenberg, C. Widstrom, J. Carlsson, V. Tolmachev, S. Stahl, et al. Tumor Imaging Using a Picomolar Affinity HER2 Binding Affibody Molecule. Cancer Res., April 15, 2006; 66(8): 4339 - 4348. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Burvenich, S. Schoonooghe, B. Cornelissen, P. Blanckaert, E. Coene, C. Cuvelier, N. Mertens, and G. Slegers In vitro and In vivo Targeting Properties of Iodine-123- or Iodine-131-Labeled Monoclonal Antibody 14C5 in a Non-Small Cell Lung Cancer and Colon Carcinoma Model Clin. Cancer Res., October 15, 2005; 11(20): 7288 - 7296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Mosquera, K. F. Card, S. A. Price-Schiavi, H. J. Belmont, B. Liu, J. Builes, X. Zhu, P.-A. Chavaillaz, H.-i. Lee, J.-a. Jiao, et al. In Vitro and In Vivo Characterization of a Novel Antibody-Like Single-Chain TCR Human IgG1 Fusion Protein J. Immunol., April 1, 2005; 174(7): 4381 - 4388. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Tang, M. Kmet, L. Corral, S. Vartanian, A. Tobler, and J. Papkoff Testisin, a Glycosyl-Phosphatidylinositol-Linked Serine Protease, Promotes Malignant Transformation In vitro and In vivo Cancer Res., February 1, 2005; 65(3): 868 - 878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Gonzalez Trotter, R. M. Manjeshwar, M. Doss, C. Shaller, M. K. Robinson, R. Tandon, G. P. Adams, and L. P. Adler Quantitation of Small-Animal 124I Activity Distributions Using a Clinical PET/CT Scanner J. Nucl. Med., July 1, 2004; 45(7): 1237 - 1244. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Miller, G. Meng, J. Liu, A. Hurst, V. Hsei, W.-L. Wong, R. Ekert, D. Lawrence, S. Sherwood, L. DeForge, et al. Design, Construction, and In Vitro Analyses of Multivalent Antibodies J. Immunol., May 1, 2003; 170(9): 4854 - 4861. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Graff and K. D. Wittrup Theoretical Analysis of Antibody Targeting of Tumor Spheroids: Importance of Dosage for Penetration, and Affinity for Retention Cancer Res., March 15, 2003; 63(6): 1288 - 1296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kathuria, R. Sriraman, R. Nath, M. Sack, R. Pal, O. Artsaenko, G.P. Talwar, R. Fischer, and R. Finnern Efficacy of plant-produced recombinant antibodies against HCG Hum. Reprod., August 1, 2002; 17(8): 2054 - 2061. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Chames, R. A. Willemsen, G. Rojas, D. Dieckmann, L. Rem, G. Schuler, R. L. Bolhuis, and H. R. Hoogenboom TCR-Like Human Antibodies Expressed on Human CTLs Mediate Antibody Affinity-Dependent Cytolytic Activity J. Immunol., July 15, 2002; 169(2): 1110 - 1118. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Henderikx, N. Coolen-van Neer, A. Jacobs, E. van der Linden, J.-W. Arends, J. Mullberg, and H. R. Hoogenboom A Human Immunoglobulin G1 Antibody Originating from an in Vitro-Selected Fab Phage Antibody Binds Avidly to Tumor-Associated MUC1 and Is Efficiently Internalized Am. J. Pathol., May 1, 2002; 160(5): 1597 - 1608. [Abstract] [Full Text] [PDF]
|
) or scFv (
) were
determined from sensorgrams such as the one in shown in
A and plotted as a function of time. B,
C6G98A; C, C6.5; D, C6ML3–9.
),
diabodies (
), or C6.5-fusions (