ANALYTICAL

BIOCHEMISTRY
Analytical Biochemistry 342 (2005) 300–311
www.elsevier.com/locate/yabio

SPOT synthesis: Reliability of array-based measurement

of peptide binding affinity
Armin A. Weiser a,1, Michal Or-Guil b,1, Victor Tapia b, Astrid Leichsenring c,
Johannes Schuchhardt d, Cornelius Frömmel a, Rudolf Volkmer-Engert c,*
a
Institute of Biochemistry, Charité, Universitätsmedizin Berlin, Monbijoustr. 2, 10117 Berlin, Germany
b
Institute of Theoretical Biology, Humboldt University, Invalidenstrasse 43, 10115 Berlin, Germany
c
Institute of Medical Immunology, Charité, Universitätsmedizin Berlin, Hessische Str. 3-4, 10115 Berlin, Germany
d
MicroDiscovery GmbH, Marienburger Str. 1, 10405 Berlin, Germany

Received 23 February 2005

Available online 23 May 2005

Abstract

Peptide arrays prepared by the SPOT synthesis technology have emerged as a proteomic tool to study molecular recognition and
identify biologically active peptides. However, it was previously not clear how accurately signal intensities obtained by probing pep-
tide arrays for protein binding really reflect the dissociation constants of the protein–peptide complexes. Using the monoclonal anti-
body CB4-1 as a model system, we systematically compared dissociation constants of antibody–peptide complexes with signal
intensities obtained using the SPOT technology. By analyzing a set of peptides possessing different affinities to the antibody, we
determined the strengths of the SPOT screening method. The accuracy of the measured results was improved by taking regional
trends in the membrane surface into account. A model based on the mass action law compares well with the experimental results.
Interestingly, the applied concentrations of the binding partners do not directly correspond to the effective concentrations in the
assay. We show that the SPOT technology is an accurate method for assigning the spotsÕ measured signal intensities to three different
binding affinity classes. The dissociation constants of the intermediate region were found to be between pKdis = 5 and pKdis = 7.
Altering the experimental parameters causes a directed change of this region.

2005 Elsevier Inc. All rights reserved.

Keywords: SPOT technology; Synthetic peptide arrays; Binding affinity; Dissociation constant; Effective concentration; Mass action law;
Competition

As demonstrated by DNA microarray experiments tional protein molecules—the principal components in

(‘‘DNA chips’’), the array format is a robust, reliable,
and now well-established method for large-scale analysis
of gene expression, allowing a global view of a living cell
or organismÕs transcriptome based on a single experi-
ment [1,2]. However, it became obvious that the infor-
mation obtained from DNA chips is not sufficient for
understanding life in general. Gene expression levels
do not closely correlate with the abundance of func-
*
Corresponding author. Fax: +49 30 450 524942.
E-mail address: rve@charite.de (R. Volkmer-Engert).
1
These two authors have contributed equally.
organizing and mediating cellular physiological func-
tions via protein–protein networks [3]. One of the pri-
mary goals in functional genomics is to understand
how such networks support cell physiology. Ideally,
we would like to draw a map displaying all protein inter-
actions in the cell and then trace functional pathways
along lines connecting the proteins. Recent interaction
analyses produce an intricate picture and interpretation
of the data is still inadequate [4–7]. We have to consider
that the wiring between all components of the intricate
network is dynamic. Protein complex formation is influ-
enced by all the parameters that determine association

0003-2697/$ - see front matter
2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.ab.2005.04.033
 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311
and dissociation of all the functional components.
Therefore, more or less quantitative readout for map-
ping protein–protein interaction in a high-throughput
manner is of general interest.
Synthetic peptide arrays offer the benefit of studying
asymmetric protein–protein interactions: a domain of
partner A acts as a receptor for a linear peptide in part-
ner B. In fact, a fairly large set of cellular protein inter-
actions are mediated by families of relatively small
protein interaction domains (PID),2 such as SH2, SH3,
GYF, WW, or PDZ, which act as receptors accommo-
dating short peptides in their binding pockets. The
SPOT technology is an ideal tool for preparing synthetic
peptide arrays, because it is an easy-to-use, robust, and
inexpensive method. Furthermore, the technique and
many of its applications have been reviewed extensively
[8–10]. Recently, we successfully applied SPOT technol-
ogy to studying the interaction network of PIDs [11–14].
Landgraf et al. [12] describes how we were able to dem-
onstrate for the first time that measured spot signal
intensities provide an approximation of the affinities of
the several thousand interactions analyzed simulta-
neously on a peptide array. In principle, signal intensi-
ties can be used to distinguish between several
interactions, with different affinities, mediated by the
same protein. We first showed this for peptide–antibody
interactions [15,16]. Here, we focus in more detail on the
correlation between spot signal intensities measured in
the array format and dissociation constants determined
for the appropriate interactions. The question we want
to address is the quantitative reliability of the SPOT
technique. First, we investigated the reproducibility of
signals measured on cellulose membrane-bound peptide
arrays prepared by the SPOT technology. Second, we
assessed the correlation between the measured signals
and the known dissociation constants. Third, we investi-
gated the accuracy with which measured signals can be
used to classify peptides into binding affinity classes. Fi-
nally, we discuss the experimental requirements needed
to obtain reliable binding affinity class predictions. To
study the quantitative aspects of the SPOT technology
we chose the interaction of the anti-p24 (HIV-1) human
monoclonal antibody (mAb) CB4-1 [17,18] with 38 well-
known peptides as a model system. The recognition pro-
file of mAb CB4-1 is a well-established experimental sys-
tem that has already been used to evaluate a variety of
structurally different types of peptide arrays prepared
by the SPOT technology. This includes scans of overlap-
2
Abbreviations used: AuC, area under curve; BG, background signal
intensity; FQ, membrane functionality quotient; Kdis, dissociation
constant; pKdis, negative decadic logarithm of Kdis; mAb, monoclonal
antibody; ROC, receiver operator characteristic; CM, cellulose mem-
brane; SCM, standard cellulose membrane; SI, signal intensity; PID,
protein interaction domain; MALD, matrix-assisted laser description
ionization.
301
ping peptides [10], randomly generated peptide libraries
[19], transition pathway libraries that reveal evolution-
ary connections between different peptides with similar
activities [20], and peptide arrays to elucidate the
cross-reactivity of mAb CB4-1 [21].
Materials and methods
Synthesis of the cellulose-membrane-bound peptide arrays
Cellulose-bound peptide arrays were semiautomati-
cally prepared according to standard SPOT synthesis
protocols using a SPOT synthesizer (Intavis, Koeln,
Germany) as described in detail [22]. The peptides were
synthesized on amino-functionalized cellulose mem-
branes of the ester type [10] prepared by modifying a cel-
lulose paper (Whatman 50, Whatman, Maidstone, UK)
with Fmoc-b-alanine as the first anchor residue. Load-
ing of the amino-functionalized cellulose membrane
was determined as described [22]. In the second step of
coupling the next anchor position, a mixture of different
ratios of a 0.3 M solution of Fmoc-b-alanine-OPfp and
a 0.3 M solution of N-acetyl-b-alanine-OPfp in dimethyl
sulfoxide was used (100, 50, 25, and 10 Fmoc-b-alanine-
OPfp) leading to membranes with a 100, 50, 25, and 10%
amino-functionalization quotient (FQ) at the spot posi-
tions [16]. The cellulose-bound peptide arrays were
assembled on these membranes by using 0.3 M solutions
of Fmoc-amino acid-OPfp in NMP (in case of Ser and
Thr the ODNp-esters were used). Side-chain protection
of the used Fmoc-amino acids was as follows: Glu, Asp
(OtBu); Ser, Thr, Tyr (tBu); His, Lys, Trp (Boc); Asn,
Gln, Cys (Trt); Arg (Pbf). The sequence files for the ar-
ray design were generated with the in-house tool Pool
1.0 and translated into chemistry files for the automated
synthesis using the in-house software LISA 1.571. For
synthesis quality control of the peptide library, a repre-
sentative selection of peptides was prepared (spot size
0.25 cm2) and cleaved by ammonia vapor in the dry
state [22,23]. Subsequently, identity was verified by ma-
trix-assisted laser desorption ionization mass spectrome-
try (MALDI-MS) (LaserTec BenchTop II, Applied
Biosystems, Foster City, CA, USA) and peptide quality
was tested by analytical reversed-phase HPLC (Waters
600, Waters, Eschborn, Germany) on a C-18 column
(Vydac, Hesparia, CA, USA). According to the general
protocol a standard peptide array (dimension
6 cm · 8 cm) was synthesized on cellulose membranes
containing 37 different peptides (Table 1, entries 2–38)
with an average of 13 repeats of each peptide and 120
times the peptide GATPEDLNQKLAGN (Table 1, en-
try 1). The peptides were equally distributed on the ar-
ray, resulting in a cellulose membrane that contains
607 spots. Such an array was generated on membranes
with four different FQ values (100, 50, 25, and 10%)
302 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311

Table 1
Dissociation constants obtained for mAb CB4-1/peptide complexes
No. Sequence pKdis No. Sequence pKdis
a a
1 GATPEDLNQKLAGN 9.0 20 RDFDKEWNLIEQNS 5.7
2 GATPQDLkTMLb 9.0 21 LELIQDLNQKLQDGFd 5.7
3 GATPEDLNQKLc 8.2 22 TTMEWFRITDGARIMd 5.7
4 sATPwDLkTsle 7.7 23 sAGPwDLksslb 5.3
5 FATPEDLNQKLc 7.7 24 LEMKQDLNQKLQDGFd 5.3
6 GATPQDLNTMLe 7.3 25 LEMKQDLNQMLQDGFd 5.2
7 GLKEWGGARITc 7.1 26 DYFDLTQDNIITRRLd 5.0
8 DATPEDLNAKLe 7.0 27 LEMKQDLNIKLQDGFd 5.0
9 DATPEDLNARLe 7.0 28 efslkGpllqwrsGa 4.7
10 GATPQDLkTMle 6.9 29 DYFDLTQDNIITRRNd 4.7
11 GATPEDLNAKLe 6.8 30 LEMKQDLNPMLQDGFd 4.7
12 FDKEWGGIRITc 6.8 31 sAGdwwLksslb 4.5
13 DALYEWGGARIc 6.7 32 saGdwwGksslb 4.4
14 GLYEWGGARITNTDa 6.7 33 sAGdwDLksslb 4.3
15 sATPwDLkTMle 6.6 34 saGdwwLksslb 4.2
16 DALPEWGGARIe 6.2 35 ITDGARIMd 4.0
17 GATPwDLkTMle 6.1 36 DYFDLTQDNIIERRNd 4.0
18 sATPwDLksslb 6.0 37 LEMKQDLNIMLQDGFd 4.0
19 EAWVLEGAMILWKTDc 6.0 38 EAWVLRGAMILWKTDd 3.5
a
[21].
b
[19].
c
[20].
d
[15].
e
BIAcore studies.

leading to the standard cellulose membranes SCM-100,
SCM-50, SCM-25, and SCM-10. The membrane
SCM-50 was prepared three times and each replica
was incubated with a different antibody concentration.
A membrane with the same FQ and the same peptides
as SCM-50 was prepared but with about the three- to
fourfold number of each peptide resulting in the mem-
brane CM2090-50 with 2090 spots. Furthermore, the cel-
lulose membrane CM6000-50 (FQ = 50%) was prepared
using six different peptides from Table 1 (entries: 1, 6,
14, 20, 26, 36) but with each peptide equally distributed
as 1000 reiterations on the array (size: 18 · 24.4 cm).
The peptides cover the affinity spectrum from pKdis = 9
to pKdis = 3.5.
centrations in T-TBS blocking buffer for 16 h at 4 C.
The following concentrations of mAb CB4-1 were used:
0.1, 1, and 10 lg/ml. The concentration of 1 lg/ml was
used with the membranes SCM-10, SCM-25, SCM-50,
and SCM-100. After washing three times for 10 min
with T-TBS, the second anti-mouse IgG peroxidase-la-
beled antibody (Sigma, Deisenhofen, Germany) was
added at a final concentration of 1 lg/ml in T-TBS
blocking buffer for 2 h, followed by washing three times
with T-TBS. Analysis and quantification of peptide-
bound mAb CB4-1 was carried out using a chemilumi-
nescence substrate and a Lumi-Imager (Roche Diagnos-
tics, Basel, Switzerland). All steps were carried out at
room temperature, unless stated otherwise.

Binding studies on cellulose-membrane-bound peptides Measurement of spot signal intensities

All incubation and washing steps were usually carried
out under gentle shaking. After washing the membrane
with ethanol once (10 min) and three times for 10 min
with Tween–Tris-buffered saline (T-TBS, 50 mM
Tris(hydroxymethyl)aminomethane, 137 mM NaCl,
2.7 mM KCl, adjusted to pH 8 with HCl/0.05% Tween
20), the membrane-bound peptide arrays were blocked
(3 h) with blocking buffer, i.e., blocking reagent (CRB,
Northwich, UK) diluted 1:10 in T-TBS containing 5%
(w/v) sucrose, and then washed with T-TBS
(1 · 10 min). Subsequently, the peptide arrays were
incubated with the anti-p24 (HIV-1) monoclonal anti-
body CB4-1 (mouse IgG2a/k) [17,18] at different con-
Analysis and quantification of spot signal intensities
were executed with the software Genespotter (MicroDis-
covery GmbH, Berlin, Germany). Genespotter has a
fully automatic grid finding routine resulting in repro-
ducible signal intensities. The spot signal is calculated
from a circular region around the spot center detected
on the image. The background signal for each spot is
determined with a safety margin to this circular region.
Standard solid-phase peptide synthesis
Soluble peptides were synthesized (50 l mol scale) as
amides on a multiple synthesizer AMS 422 (Abimed,
 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311 303

Langenfeld, Germany) according to the standard Fmoc

machine protocol using TentaGel S RAM resin (Rapp
Polymere, Tübingen, Germany) and PyBOP activation.
All peptides were analyzed by reversed-phase HPLC
on a Vydac C18 column using a linear gradient of
5–60% acetonitrile/water (0.05% trifluoroacetic acid)
for 20 min at 1.2 ml/min flow rate (detection at
214 nm) and by MALDI-MS using a-cyano-4-hy-
droxy-cinnamic acid as matrix and purified to >95%
purity by preparative HPLC on a Vydac C18 column,
if necessary.

Surface plasmon resonance measurements

The affinity of peptides in solution to the mAb CB4-1

was measured with regard to Kdis by monitoring adsorp-
tion-dependent surface plasmon oscillations using the
BIACOREX system (Biacore, Uppsala, Sweden). Both
the mAb CB4-1 and the anti-GST monoclonal antibody
(control antibody) were immobilized on a dextran-
coated gold surface in separate flow cells of a CM5
sensor chip (Biacore AB) using the amine coupling pro-
cedure (5 ll/min; activation with 1:1 EDC/NHS, 7 min;
immobilization with 50 lg/ml antibody, 5- to 75-ll mul-
tiple injections; deactivation of excess groups with 1 M
ethanol amine-HCl, pH 8.5, 7 min). The amount of
covalently coupled antibodies corresponded to a signal
increase of approximately 5000 resonance units (RU)
for both antibodies (flow cell 1: control antibody; flow
cell 2: mAb CB4-1). All binding experiments were per-
formed at 20 C with a flow rate of 15 ll/min (injection
volume 10 ll). Peptides were used at various concentra-
tions between 20 pM and 200 mM in HBS (10 mM
Hepes with 0.15 M NaCl, 3.4 mM EDTA, and 0.005%
surfactant P20, pH 7.4). Complete regeneration was ob-
tained after dissociation without regeneration buffer in
most cases. Different regeneration procedures (BIAco-
reX manual) were required in some cases. Transforma-
tion of data and analysis were performed with the
BIA-evaluation software, version 3.0. The control sen-
sorgram (flow cell 1) was subtracted from the sensor-
grams obtained with flow cell 2. The steady state
values of the binding equilibrium were plotted versus
the different peptide concentrations and fitted using
the implemented steady state evaluation, resulting in
the Kdis for each model peptide (Table 1).
Results and discussion
Spot signal intensities: reproducibility and improvements
Performing a binding experiment with a cellulose-
membrane-bound peptide array (see Materials and
methods) results in measurable spot signal intensities
signifying peptide–protein interactions (Fig. 1). Stan-
Fig. 1. Binding of the anti-p24 (HIV-1) monoclonal IgG antibody
CB4-1 to cellulose-bound peptides. Shown is the membrane SCM-10
(6 · 8 cm) with 607 generated spots representing 38 peptides (Table 1)
all known to interact with mAb CB4-1 but possessing different
affinities to the antibody.
dard deviation of the SIs measured for several peptide
replicas on one membrane varies in the range from 8
to 22%. We assume that inhomogeneity of the modified
cellulose membrane is largely responsible for this varia-
tion, since the synthesis quality of all peptides used in
the binding experiment is well established [15,16,19–
21]. The membrane-related inhomogeneity of measured
SIs was confirmed by analyzing their spatial distribution
(Fig. 2). For this purpose, we determined the mean val-
ues of the replica SIs for each peptide on the membrane
CM6000-50 and calculated the relation of each single SI
to its mean value, thus determining regional trends. This
revealed that peptides with similar mean signal intensi-
ties tend to have similar regional trends and peptides
with different mean signal intensities have different re-
gional trends (Figs. 2B and D). Here, regional trends de-
note the trends of spots in different local parts of the
membrane. Furthermore, signal intensity fluctuations
of the background correlate with those of weak binders
(Figs. 2C and D). After analyzing several membranes it
became obvious that regional trends are intrinsic charac-
teristics of individual membranes. This presents an
opportunity to correct measured SIs and thus reduce
standard deviation. A minimum of two references are
304 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311

Fig. 2. (A) Neighborhood of a spot (gray). The size of the neighborhood N indicates that it consists of N · N spots. The black spots represent
reference spots. (B–D) Mapping of the membrane CM6000-50 (18 cm · 24.4 cm) containing 6000 spots (67 rows · 90 columns). Light areas denote
relatively high and dark areas low signal intensity values. (B) Relative signals of the high-affinity reference peptide (about 1000 data points,
pKdis = 9), (C) background (6000 data points), (D) a weak-binding peptide (about 1000 data points, pKdis = 4). Note that maxima and minima of
mapping (B) have different locations compared to (C and D). The scale gives the relation between the local and the global mean signal.

necessary—one peptide with high-affinity and one with
low-affinity to the binding partner. The algorithm for
improving measured SIs involves the following five
steps: first, calculating the mean SI values of all high-
ðH
global Þ and low-affinity references ðL
global Þ; second,
defining a neighborhood i for each spot with N spots
around it (Fig. 2A); third, calculating the mean SI values
of the high- ðH
local;i Þ and low-affinity references ðL
local;i Þ
within each neighborhood i; fourth, calculating the vari-
ables a and b with the equations L
global ¼ ai
L
local;i þ bi
and H
global ¼ ai
H
local;i þ bi ; fifth, the improved signal
intensity for each spot in its neighborhood i (SIimproved)
can then be calculated from the measured signal intensi-
ties (SImeasured).
SIimproved ¼ ai
SImeasured þ bi . ð1Þ
improved
The standard deviation of the SI values for
replicas of the same peptide will be smaller than that
of SImeasured. As an alternative to a low-affinity reference
peptide, the background signals of each spot can also be
used as references. The resulting standard deviations are
shown for the SCM-50 membrane in Fig. 3A (stars). The
efficiency of improvement is influenced by the neighbor-
hood size N and the percentage of reference peptides
regularly distributed on the membrane. To determine
optimal parameters, standard deviations of SIimproved
were calculated for the CM6000-50 membrane using
background signals as the low-affinity reference and
considering different neighborhood sizes N and different
percentages of high-affinity reference peptide replicas
(Fig. 3B). As expected, increasing the number of high-
affinity reference peptides improves the efficiency of
the method and reduces the optimal neighborhood size
N (Fig. 3B). The results suggest that good adjustment
for regional trends can be achieved when around 4%
of the spots on a membrane are high-affinity reference
peptides and by using a neighborhood with N · N spots,
when N is between 10 and 20.
In practice, there are often no replicas available on a
membrane or no high-affinity peptide is known before-
hand. In such cases, SIimproved can be calculated by using
only the background signal around each spot according
to

global;i
L
SIimproved ¼ SImeasured

. ð2Þ
Llocal;i
The improvements achieved by applying Eqs. (1) and
(2) are compared in Fig. 3C. Note that the decrease in
standard deviation is smaller when using Eq. (2), espe-
cially for peptides with high signal intensities.
We compared SI measurements made from two iden-
tical membranes prepared with the same peptides, the
 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311 305

Fig. 3. Standard deviation of the signal intensity versus size of the neighborhood and the mean signal intensities, respectively, after applying the noise
reduction algorithm. (A) Standard deviations of the signal intensities for each of the 38 different peptides (10–15 spots each) on the SCM-10
membrane (Fig. 1) versus their mean values (circles). Gray diamonds denote the standard deviation after application of the noise reduction algorithm
using only background SI values as a reference. Black stars show the effect of the noise reduction algorithm using background SI values and high-
affinity peptide repeats (14% of total spots) as references. The standard deviation obtained for the corrected signal intensities depends on the size of
the chosen local neighborhood (N) and on the number of reference peptides in it. (B) and (C) show the standard deviation of corrected signal
intensities for two peptides, one with low and one with high binding affinity, using the background and a high-affinity reference peptide (B) or only
the background as reference (C). In (B), four cases where the numbers of high-affinity reference peptides are 14, 7, 4, and 2% of the total number of
spots on the membrane are shown. The signals are derived from membrane CM6000-50.

same number of spots, the same concentration of anti-
body, the same FQ, and the same incubation time. We
found signal intensity deviations of around a factor of
10. Nevertheless, the relations of the signals were fairly
constant for all peptides, suggesting that, to compare
SIs from different membranes, these values should first
be normalized using
SImeasured  SImin
SInorm ¼ . ð3Þ
SImax  SImin
Note that, below, this equation will be used only for bet-
ter visualization of the intermediate region (Fig. 4), with
SI being the mean signal intensity values for each recur-
rent peptide.
Correlation between signal intensities and dissociation
constants
A representative repertoire of 38 peptides was se-
lected from a set of over 100 peptides, all known to
interact with mAb CB4-1 (Table 1). All chosen peptides
are well described in the literature and dissociation con-
stants (pKdis) for the various peptide/antibody com-
plexes span a range of more than five orders of
magnitude (pKdis = 9 to pKdis = 3.5). As an example,
mean values of measured SIs obtained from a SCM-10
membrane were used to draw an SI/pKdis scatter plot
(Fig. 4A). It demonstrates that the obtained SIs corre-
late with given pKdis. To describe the SI/pKdis depen-
306 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311

Fig. 4. Mean signal intensities versus dissociation constants of all measured peptides. All signal intensities are normalized according to Eq. (3). The
gray region marks all signals that are weaker than the mean background signal. (A) Mean values of the 38 different peptides on the SCM-10 membrane
with an antibody concentration of [AT] nM. The solid curve was derived by fitting Eqs. (4)–(7) to the data. The scale of the SI axis is logarithmic. (B–D)
Fitting of the results from several experimental setups with all parameters held constant except one: (B) two membranes prepared with a different
number of peptide replicas, (C) two membranes loaded with different FQ, (D) two membranes incubated with different antibody concentrations.

dency mathematically and elucidate understand which
parameters influence the correlation, we defined a mass
action law model. It is based on the assumption that the
system is in equilibrium and that the mass action law is
the underlying effect. Furthermore, we have to take into
account that competition among the peptides for mAb
CB4-1 might take place, especially in cases when peptide
concentrations or the number of spots are high com-
pared to the antibody concentration.
Since cellulose membranes are porous, the concen-
tration of antibody within the fibers may be different
from the antibody concentration [AT] in the bulk solu-
tion. The inter-fiber spaces could be on the order of
magnitude of antibody molecules in size. As described
by Minton [24], this could lead to confinement effects
and modified diffusion lengths. Since peptide–antibody
interactions primarily occur in the inter-fiber spaces, we
define a total effective antibody concentration ½Aeff T  for
this space. Furthermore, the surface density of accessi-
ble peptides on a spot is not known. It may be influ-
enced by positive cooperativeness: cluster formation
of several peptide molecules within a spot may decrease
its concentration [25,26]. We define a total effective
peptide concentration per spot ½P effT  and assume that
it is the same for all spots. Unfortunately, it is not pos-
eff
sible to determine the parameters ½P eff
T  and ½AT  exper-
imentally. We therefore estimate these effective
parameters by the proposed mass action law model.
To formulate the model, we first define Eqs. (4)–(6).
eff
Here, ½P eff
i  and [A ] denote the concentrations of un-
bound peptide within a spot i (i = 1, . . . , n) and un-
bound antibody, respectively.
½Aeff 
½P eff
i  ¼ K disi
½AP i ; ð4Þ
 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311 307

½P eff eff
T  ¼ ½P i  þ ½AP i ;
X
n
½Aeff eff
T  ¼ ½A  þ ½AP i .
i¼1
Eq. (4) simply states the mass action law for one sin-
gle spot, Eq. (5) sums the unbound and bound fractions
of the peptide population in spot i, and Eq. (6) sums the
concentration of free antibody with the sum of all com-
plexes to the given total antibody concentration. This
describes a nonlinear system of 2n + 1 equations with
eff
the 2n + 1 variables ½P effi , [APi], and [A ] and the
eff eff
parameters ½AT  and ½P T . Furthermore, it is reasonable
to assume that the concentration of the antibody–pep-
tide complex [APi] within a spot is proportional to its
measured signal intensity and to introduce the back-
ground signal as an additive factor. Thus, we addition-
ally define Eq. (7) with the proportionality factor a
and the background parameter b:
SIi ¼ f ðpK disi Þ ¼ a
½AP i  þ b.
Eqs. (4)–(7) are used to fit the free parameters ½Aeff T ,
½P eff
T , and b to the measured experimental data using
the functions ‘‘nls’’ and ‘‘nls.lm’’ from the R language
environment [27]. We exclude parameter a from the fit-
ting procedure to avoid redundancies and due to the fact
that a differs from membrane to membrane. Instead, it
max
was estimated by the formula a ¼ SI½P 0  , where ½P 0T  de-
T
notes the sweeping start parameter of the fitting process
for ½P eff
T .
To evaluate the relevance of the model, we specifi-
cally varied several experimental parameters: the num-
ber of spots on a membrane, the FQ, and the antibody
concentration. The fitting process was accomplished
for each binding experiment. The correlation coefficients
between the mean values of the measured SI and the fit-
ted SI were found to be between 0.83 and 0.91. Table 2
summarizes all fitting results and the experimental con-
ditions used. Figs. 4B–D show the best fits to the ob-
tained experimental data after normalization with Eq.
(3).
First, we observed that the effective peptide concen-
tration increases with decreasing FQ. Increasing the
membrane FQ from 10 to 100% resulted in a decrease
ð5Þ of the fitted total effective peptide concentration from
280 to 42 pM. This result is supported by the fact that
absolute measured SIs decrease with increasing FQ also
ð6Þ
(data not shown). The decrease of effective accessible
peptide concentration with increasing FQ can be ex-
plained by a clustering effect [25,26], suggesting that
much lower FQs than those used here could also yield
good results.
Second, the fitted total effective antibody concentra-
tion ½Aeff
T  does not increase with the same order of mag-
nitude as the concentration [AT] originally applied in the
bulk solution. While [AT] increases by a factor of 100,
½Aeff
T  increases only by a factor of approximately 4. Most
likely, the total effective concentration of the antibody
saturates within the membrane and one consequence is
a nonlinear correlation between the total effective con-
centration within the spots and the concentration of
the antibody in solution.
Third, several competition effects were observed. Two
ð7Þ membranes with identical antibody concentration and
FQ and the same peptides, but with a different number
of spots, i.e., different reiteration of the peptides (see
Materials and methods), show very different signal
behavior. This can be understood if one considers that
many spots with high-affinity bind many antibody mole-
cules, thus depleting the solution of free antibodies. Few-
er antibody molecules are then available for binding to
peptides with lower affinity, so they will have lower SIs
compared to when there are only a few high-affinity spots
and therefore less competition for antibody (Fig. 4B).
Furthermore, competition for antibodies may also
possibly explain an effect described by Kramer et al.
[16]: signal intensities either increase or decrease with
the FQ of the membrane, depending on the peptide
investigated. In a simulation of this experiment applying
Eqs. (4)–(7), we compared two model results where only
the peptide concentration was varied: ½P eff T  ¼ 100 nM
eff
and ½P eff
T  ¼ 400 nM, both with ½A T  ¼ 10 lM and seven
peptides (pKdis = 9 up to pKdis = 3) each replicated 15
times. The results shown in Fig. 5 illustrate that peptides
with high binding affinity show decreasing signal inten-
sities for decreasing ½P eff
T  (and hence increasing FQ),
while the converse applies to peptides with low binding
affinity.

Table 2
Several fitted and experimental parameters for different experimental setups
No. Membrane Spots FQ (%) [AT] (nM) a b ½P eff
T  ðpMÞ ½Aeff
T  ðnMÞ Correlation
1 SCM-10 607 10 6.3 5.2e+13 2100 280 180 0.88
2 SCM-25 607 25 6.3 6.6e+13 3100 140 190 0.86
3 SCM-50 607 50 0.63 8.7e+13 1400 130 70 0.89
4 SCM-50 607 50 6.3 7.4e+13 2800 90 190 0.86
5 SCM-50 607 50 63 6.9e+13 2500 100 260 0.85
6 SCM-100 607 100 6.3 1.0e+14 2400 40 190 0.83
7 CM2090-50 2090 50 6.3 9.8e+12 300 100 100 0.91
308 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311

Fig. 5. Signal intensity versus dissociation constant calculated from

Eqs. (4)–(7) for two cases which differ only in the peptide concentra-
tion. In this theoretical investigation the signal intensities are smaller
for ½P eff eff
T  ¼ 100 nM than for ½P T  ¼ 100 nM if peptides have low-
affinity but larger if their binding affinity is high.

Detection of high-affinity binders from signal intensity

data

The most important application of the SPOT synthe-

sis technique is to detect peptides that have a strong
binding affinity to defined targets. In practice, there
are often no replicate peptides on cellulose membrane
Fig. 6. (A) Signal intensity versus dissociation constant for each
arrays so one cannot calculate mean values. Therefore, single spot. The class borders obtained by maximizing the contin-
we investigated the confidence of the SPOT synthesis gency coefficient R (Eq. (12)) for classification into three affinity
technique in detecting high-affinity binders for nonre- classes are shown as dashed lines. The gray region marks all signals
current spots. First, we had to define classes of high- that are weaker than the mean background signal. The SI axis has
and low-affinity peptides. As depicted in Fig. 6A, we logarithmic scale. (B) ROC curve generated for classification of
peptides on the SCM-25 membrane with [AT] = 6.3 nM in two
define two borders SI1 and SI2 on the y axis (for the sig- binding affinity classes. The area under the curve (AuC value) is 0.97,
nal intensities) and two borders pK 1dis and pK 2dis on the x indicating that signal intensity obtained as the output of the SPOT
axis (for the peptidesÕ affinity). The four borders are free technology is an excellent quantity for the discrimination between
parameters defining three classes of signal intensities and high- and low-affinity binders.
three classes of dissociation constants, resulting in nine
disjoint quadrants in the plot. For each of the nine
quadrants the number of occurrences of spots can be as- P3 P3 !!
signed for each membrane and is denoted as nij (i,j = 1, X
3
j¼1 nji j¼1 nji
H ðpK dis Þ ¼  P3
log 2 P3 ; ð9Þ
2, 3 are the enumerations of the SI classes and the pKdis i¼1 j;k¼1 njk j;k¼1 njk
classes, respectively).
!!
To determine how well the signal intensity class cor- X
3 X
3
nij nij
responds to the actual affinity class of the peptide, we H ðSI; pK dis Þ ¼  P3
log 2 P3 ;
calculate their mutual information (transinformation). i¼1 j¼1 k;l¼1 nkl k;l¼1 nkl

We introduce the following quantities: first, the entro- ð10Þ

pies of signal intensity and dissociation constant (Eqs.
(8)–(10)) and second, the transinformation of signal IðSI; pK dis Þ ¼ H ðSIÞ þ H ðpK dis Þ  H ðSI; pK dis Þ. ð11Þ
intensity and dissociation constant (Eq. (11)): The value for the transinformation I lies between zero
P3 P3 !! and infinity. To rank the value of the transinformation,
X 3
j¼1 nij j¼1 nij
H ðSIÞ ¼  P3
log 2 P3 ; ð8Þ we normalize it and define the contingency coefficient
i¼1 j;k¼1 njk j;k¼1 njk (R) that is bounded by zero and one (Eq. (12)):
 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311 309

Table 3
Calculated pKdis and SI borders with their corresponding contingency coefficients (R) and AuC values
No. Membrane FQ (%) [AT] (nM) R SI1 SI2 pK 1dis pK 2dis AuC
1 SCM-10 10 6.3 0.92 4500 8900 6.8 7.2 0.97
2 SCM-25 25 6.3 0.92 3300 7400 5.3 6.8 0.97
3 SCM-50 50 0.63 0.96 3700 9200 6.8 7.2 0.98
4 SCM-50 50 6.3 0.90 3700 6500 5.3 6.8 0.96
5 SCM-50 50 63 0.93 3200 7600 5.3 6.8 0.95
6 SCM-100 100 6.3 0.83 3200 4800 5.3 6.8 0.90
For calculating the AuC value the SI border of the ROC curve was the mean value of the background signals plus three times its standard deviation
(see Fig. 6B).
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1  expð2
IðSI; pK dis ÞÞ between low- and high-binding affinity classes. Fur-
RðSI; pK dis Þ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi . thermore, we determine the border for separating
1  expð2
minðH ðSIÞ; H ðpK dis ÞÞÞ
low and high signal intensities (SI1) with the criteria
ð12Þ described above (background signal plus three times
Note that, when the classifications SI and pKdis are inde- its standard deviation), finally resulting in four quad-
pendent, then H(SI) + H(pKdis) = H(SI, pKdis) and rants nij (i,j = 1,2). With these, we calculate true posi-
therefore R(SI, pKdis) = 0. The larger the value of the tive (TP) rates and false positive (FP) rates
contingency coefficient R, the more information ðTP ¼ n11nþn
11
21
; FP ¼ n12nþn
12
22
Þ. Moving the border pK 1dis
the two classifications have in common. We calculate over the range of all possible values will change TP
the R values for a sweep of the parameters pK 1dis , and FP. Plotting (TP, FP) points into a diagram
pK 2dis , SI1, and SI2 and determine the borders with the and connecting those gives a ROC curve (Fig. 6B).
best R value. The area under the ROC curve is a measure of the
This procedure was applied to several cellulose mem- quality of discrimination between the two classes.
branes that we examined and the resulting contingency Random prediction would result in a bisecting ROC
coefficients lay between 0.83 and 0.96 (Table 3) showing line, giving an AuC value of 0.5. The closer the
high correlation for the distinction of the three defined AuC is to 1.0, the better is the classification. The
classes of peptide binding affinities. SIs larger than SI2 SI1 border of background signal plus three times its
may be associated with a dissociation constant smaller standard deviation results in AuC values of up to
than pK 2dis and SIs smaller than SI1 may be associated 0.98 for the differently synthesized membranes (Table
with a dissociation constant larger than pK 1dis . The inter- 3), suggesting that signal intensity is an excellent clas-
mediate class between the two borders pK 1dis and pK 2dis sifier for high- and low-affinity binders. Note that the
isolates a small region of dissociation constants (‘‘dy- best pK 1dis value for the distinction of high and low-af-
namic range’’), so that measured signals inside the corre- finity is defined by the experimental setup and will
sponding SI class may be associated with an usually not be known by the experimenter. The exam-
approximate dissociation constant. Under the often ple ROC curve for membrane SCM-25 shown in Fig.
used condition (FQ = 50% and [AT] = 6.3 nM), the dy- 6B suggests a good pK 1dis value at 6.7: 90% of peptides
namic range of the curve is between pK 1dis ¼ 5.3 and classified as having high affinity in fact belong to the
pK 2dis ¼ 6.8. class of peptides with pK dis < pK 1dis , while only 7%
Applying the transinformation to more than the three of the peptides in this class were wrongly classified
classes described above led to singleton classes (data not as being low-affinity binders ðpK dis > pK 1dis Þ. An even
shown). This suggests that any other attempt to classify higher accuracy of classification can be achieved when
the results of the SPOT technology into more classes mean SI values of replica peptides are used.
would be meaningless. For instance, it seems to be
impossible to assign a dissociation constant to each
measured signal. This is due to the extreme variability Conclusions
of the data in the central class and the saturation taking
place far off the dynamic range. We describe a method to reduce the standard devia-
In the data investigated we found that the lower tion of signals measured with the SPOT technique. It
border for signal intensities (SI1) never exceeds the is able to approximately halve the standard deviation
background signal plus three times its standard devia- (Fig. 3A), resulting in improved signal intensities for fur-
tion. To confirm that this value is an objective border ther analysis. To make the best use of this method, we
for distinguishing between the two classes of high- and suggest that around 4% of the spots on a membrane
low-affinity peptides we applied ROC statistics to the should be homogeneously distributed repeats of a refer-
measured data. Contrary to the above, we need to de- ence peptide possessing high-affinity toward the ligand,
fine only one border pK 1dis as the border distinguishing if available. This leads to (i) optimal results in the algo-
310 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311
rithm for reducing signal noise due to regional trends
and (ii) optimal conditions for automatic grid position-
ing by the software that measures signal intensities. For
experiments where no strong binding peptides are
known beforehand, the noise reduction algorithm can
still be applied with only background signals as refer-
ences. In this case noise reduction is suboptimal (Fig.
3A).
We report good agreement between model results
based on the mass action law (Fig. 4A). The model takes
into account competition among peptides for free anti-
body molecules. It describes several observed effects,
such as a shift in the ‘‘dynamic’’ range and hence the
border separating high- and low-affinity classes when
only the number of peptide replicas on a membrane is
changed (Fig. 4B) and the apparently contradictory
behavior of different peptidesÕ SIs when experiments per-
formed with different amino-functionalization (FQ) val-
ues are compared. This observation suggests that
competition effects do indeed play a role in the usual
experimental setup and that, due to this effect, inter-
membrane SI comparisons have to be handled with care
even after normalization of the values (but note that this
effect does not disrupt the ability to classify peptides
according to their affinity when the SIs used were mea-
sured on a single membrane). It also suggests that a de-
crease in effective peptide concentration, possibly by
dramatically decreasing FQ, could help diminish this
effect.
Fitting of the model to experimental data to estimate
experimental parameters suggests that the effective con-
centration of accessible peptides on the cellulose mem-
brane decreases for increasing FQ. It is expected that
for an FQ much smaller than 10% this behavior can
be reversed. Moreover, the fitted parameters suggest
that the effective ligand concentration within the mem-
brane fibers increases sublinearly with increased ligand
concentration in the bulk solution.
Classification of peptides into two classes representing
high and low binding affinities is possible with high pre-
dictive power using the measured signal intensities as cri-
teria. The SI threshold separating both classes is best
chosen as the mean value of the SIs of the spot back-
ground signals plus three times its standard deviation. Re-
cently, the method derived here was successfully applied
to characterizing PDZ domain/ligand specificity [13,14].
The SPOT technique also allows differentiation be-
tween three classes of dissociation constants: high,
low, and intermediate. The most frequently used
experimental condition (FQ = 50% and [AT] = 6.3 nM)
yields an intermediate class whose affinities lie between
pK 1dis ¼ 5.3 and pK 2dis ¼ 6.8. This dynamic range, and
with that the pK 1dis border of classification in two clas-
ses, can be shifted by altering the effective concentra-
tions of the binding partners. There are two
possibilities for varying the peptide concentration:
changing the FQ or using a differently coated mem-
brane, e.g., CAPE, as described [23]. Furthermore,
the ligand concentration in the bulk solution can be
varied.
Acknowledgments
This work was in part supported by the Deutsche
Forschungsgemeinschaft (SFB 618 and SFB 449).
M.O.G and V.T. thank the Volkswagen Foundation
for funding. We thank W. Hoehne (Institut für Bioche-
mie, Berlin, Germany) for providing the CB4-1 anti-
body. Excellent technical assistance by I. Kretzschmar
is gratefully acknowledged. We thank N. Bruni for help-
ful criticism and comments.
References
[1] M.B. Eisen, P.O. Brown, DNA arrays for analysis of gene
expression, Methods Enzymol. 303 (1999) 179–205.
[2] E. Fehrenbach, D. Zieker, A.M. Niess, E. Moeller, S. Russwurm,
H. Northoff, Microarray technology - the future analyses tool in
exercise physiology, Exerc. Immunol. Rev. 9 (2003) 58–69.
[3] S.P. Gygi, Y. Rochon, B.R. Franza, R. Aebersold, Correlation
between protein and mRNA abundance in yeast, Mol. Cell. Biol.
19 (1999) 1720–1730.
[4] Y. Ho, A. Gruhler, A. Heilbut, et al., Systematic identification of
protein complexes in Saccharomyces cerevisiae by mass spectrom-
etry, Nature 415 (2002) 180–183.
[5] A.C. Gavin, M. Bosche, R. Krause, et al., Functional organiza-
tion of the yeast proteome by systematic analysis of protein
complexes, Nature 415 (2002) 141–147.
[6] L. Giot, J.S. Bader, C. Brouwer, et al., A protein interaction
map of Drosophila melanogaster, Science 302 (2003)
1727–1736.
[7] S. Li, C.M. Armstrong, N. Bertin, et al., A map of the
interactome network of the metazoan C. elegans, Science 303
(2004) 540–543.
[8] R. Frank, The SPOT-synthesis technique. Synthetic peptide arrays
on membrane supports—principles and applications, J. Immunol.
Methods 267 (2002) 13–26.
[9] U. Reineke, R. Volkmer-Engert, J. Schneider-Mergener, Appli-
cations of peptide arrays prepared by the SPOT-technology, Curr.
Opin. Biotechnol. 12 (2001) 59–64.
[10] H. Wenschuh, R. Volkmer-Engert, M. Schmidt, M. Schulz, J.
Schneider-Mergener, U. Reineke, Coherent membrane supports
for parallel microsynthesis and screening of bioactive peptides,
Biopolymers 55 (2000) 188–206.
[11] L. Otte, U. Wiedemann, B. Schlegel, J.R. Pires, M. Beyermann,
P. Schmieder, G. Krause, R. Volkmer-Engert, J. Schneider-
Mergener, H. Oschkinat, WW domain sequence activity relation-
ships identified using ligand recognition propensities of 42 WW
domains, Protein Sci. 12 (2003) 491–500.
[12] C. Landgraf, S. Panni, L. Montecchi-Palazzi, L. Castagnoli, J.
Schneider-Mergener, R. Volkmer-Engert, G. Cesareni, Protein
interaction networks by proteome peptide scanning, PLoS Biol. 2
(2004) 94–103 (Epub 2004 Jan 20).
[13] P. Boisguerin, R. Leben, B. Ay, G. Radziwill, K. Moelling, L.
Dong, R. Volkmer-Engert, An improved method for the synthesis
of cellulose membrane-bound peptides with free C termini is
useful for PDZ domain binding studies, Chem. Biol. 11 (2004)
449–459.
 Array-based measurement of peptides / A.A. Weiser et al. / Anal. Biochem. 342 (2005) 300–311
[14] U. Wiedemann, P. Boisguerin, R. Leben, D. Leitner, G. Krause,
K. Moelling, R. Volkmer-Engert, H. Oschkinat, Quantification
of PDZ domain specificity, prediction of ligand affinity and
rational design of super-binding peptides, J. Mol. Biol. 343 (2004)
703–718.
[15] R. Volkmer-Engert, B. Ehrhard, J. Hellwig, A. Kramer, W.
Höhne, J. Schneider-Mergener, Preparation, analysis and anti-
body binding studies of simultaneously synthesized soluble and
cellulose-bound HIV-1 p24 peptide epitope libraries, LIPS 1
(1994) 243–253.
[16] A. Kramer, U. Reineke, L. Dong, B. Hoffmann, U. Hoffmüller,
D. Winkler, R. Volkmer-Engert, J. Schneider-Mergener, Spot
synthesis: observation and optimizations, J. Peptide Res. 54 (1999)
319–327.
[17] R. Grunow, R. Giese, T. Porstmann, H. Döpel, K. Händel, R.
Von Baehr, Developement and biological testing of human and
murine monoclonal antibodies against HIV antigens, Z. Klin.
Med. 45 (1990) 367–369.
[18] W.E. Höhne, G. Kuttner, S. Kiessig, G. Hausdorf, R. Grunow,
K. Winkler, H. Wessner, E. Giessmann, R. Stigler, J. Schneider-
Mergener, R. von Baehr, D. Schomburg, Structural base of the
interaction of a monoclonal antibody against p24 of HIV-1 with
its peptide epitope, Mol. Immunol. 30 (1993) 1213–1221.
[19] U. Reineke, C. Ivascu, M. Schlief, C. Landgraf, S. Gerike, G.
Zahn, H. Herzel, R. Volkmer-Engert, J. Schneider-Mergener,
Identification of antibody epitopes and mimotopes applying a
small single peptide array of 5520 randomly generated sequences,
J. Immunol. Methods 267 (2002) 37–51.
311
[20] U. Hoffmüller, T. Knaute, M. Hahn, W. Höhne, J. Schneider-
Mergener, A. Kramer, Evolutionary transition pathways for
changing peptide ligand specificity and structure, EMBO J. 19
(2000) 4866–4874.
[21] A. Kramer, T. Keitel, K. Winkler, W. Stöcklein, W. Höhne, J.
Schneider-Mergener, Molecular basis for the binding promiscuity
of an anti-p24 (HIV-1) monoclonal antibody, Cell 91 (1997) 799–
809.
[22] H. Wenschuh, H. Gausepohl, L. Germeroth, M. Ulbricht, H.
Matuschewski, A. Kramer, R. Volkmer-Engert, N. Heine, T.
Ast, D. Scharn, J. Schneider-Mergener, Positionally addressable
parallel synthesis on continuous membranes, in: H. Fenniri (Ed.),
Combinatorial Chemistry: A Practical Approach, Oxford Univer-
sity Press, Oxford, UK, 2000, pp. 95–116.
[23] R. Volkmer-Engert, B. Hoffmann, J. Schneider-Mergener, Stable
attachment of the HMB-linker to continuous cellulose membranes
for parallel solid phase spot synthesis, Tetrahedron Lett. 38 (1997)
1029–1032.
[24] A.P. Minton, The influence of macromolecular crowding and
macromolecular confinement on biochemical reactions in physi-
ological media, J. Biol. Chem. 276 (2001) 10577–10580.
[25] A.P. Minton, Effects of excluded surface area and adsorbate
clustering on surface adsorption of proteins I. Equilibrium
models, Biophys. Chem. 86 (2000) 239–247.
[26] P. Minton, Effects of excluded surface area and adsorbate
clustering on surface adsorption of proteins. II. Kinetic models,
Biophys. J. 80 (2001) 1641–1648.
[27] http://www.r-project.org.