Professional Documents
Culture Documents
Research paper
a r t i c l e i n f o a b s t r a c t
Article history: Antibodies are important tools for the study of G-protein-coupled receptors, key proteins in
Received 27 December 2010 cellular signaling. Due to their large hydrophobic membrane spanning regions and often very
Received in revised form 5 May 2011 short loops exposed on the surface of the cells, generation of antibodies able to recognize the
Accepted 5 May 2011
receptors in the endogenous environment has been difficult. Here, we describe an antigen-
Available online 12 May 2011
design method where the extracellular loops and N-terminus are combined to a single antigen for
generation of antibodies specific to three selected GPCRs: NPY5R, B2ARN and GLP1R. The design
Keywords: strategy enabled straightforward antigen production and antibody generation. Binding of the
GPCR
antibodies to intact receptors was analyzed using flow cytometry and immunofluorescence based
Antibody
confocal microscopy on A-431 cells overexpressing the respective GPCR. The antibody–antigen
Antigen design
interactions were characterized using epitope mapping, and the antibodies were applied in
immunohistochemical staining of human tissues. Most of the antibodies showed specific binding
to their respective overexpressing cell line but not to the non-transfected cells, thus indicating
binding to their respective target receptor. The epitope mapping showed that sub-populations
within the purified antibody pool recognized different regions of the antigen. Hence, the genetic
combination of several different epitopes enables efficient generation of specific antibodies with
potential use in several applications for the study of endogenous receptors.
© 2011 Elsevier B.V. All rights reserved.
0022-1759/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jim.2011.05.001
K. Larsson et al. / Journal of Immunological Methods 370 (2011) 14–23 15
2.1. Materials
concentrations were determined using a BCA™ Protein Assay Phusion High-Fidelity DNA polymerase (Finnzymes) and
Kit (Pierce). The dual affinity handle protein, His6ABP, was furthermore cloned into the lentivector 2K7bsd (Suter et al.,
also expressed and purified from the pAff8c vector as 2006) using Gateway® cloning (Invitrogen). The 2K7bsd-GPCR
previously described (Agaton et al., 2004). The three GPCR vectors were thereafter transfected to HEK293-T cells together
antigens were immunized in two New Zealand white rabbits with packaging and envelope plasmids using a Calcium chloride
each (Young In Frontier, South Korea) in accordance with the transfection procedure, as previously described (Pear et al.,
national guidelines. The primary immunization was per- 1993; Vernet et al., 2009). The produced viral particles were in
formed using 200 μg of antigen in Freund's complete turn used to infect A-431 cells in the presence of Polybrene
adjuvant, followed by three booster injections of 100 μg (8 μg/ml).
each in Freund's incomplete adjuvant with four week in-
tervals. The rabbit serum was collected two weeks after the
2.6. Flow cytometry analysis
final booster injection.
All generated antibodies as well as several commercial
2.4. Antibody purification and characterization
antibodies were analyzed for binding to the full-length
protein using the overexpressing cell lines and non-trans-
Antisera from immunized animals were purified by a two-
fected controls on a flow cytometer. Briefly, the cells were
step immunoaffinity protocol (Agaton et al., 2004). Initially,
trypsinated, washed with PBS and resuspended in PBS
12 ml serum was filtrated and buffered with PBS. Subse-
containing 1% bovine serum albumin (PBSB). 200,000 cells
quently, antibodies specific for the tag protein were depleted
were used for staining with each antibody. Primary anti-
from the sample using vector protein, His6ABP, coupled to
bodies were diluted with the PBS containing BSA to a final
NHS-activated Sepharose™ 4 Fast Flow matrix (Amersham
concentration of 5 μg/ml and secondary antibodies to
Biosciences). The depleted sera samples were additionally
10 μg/ml. The cells were incubated with 70 μl of primary
affinity purified on prepacked NHS-activated one ml HiTrap
antibody for 1 h at RT followed by washing with PBSB. After
columns (GE Healthcare), coupled with the respective antigen,
incubation with 70 μl of secondary antibody (Alexa Fluor 488
using the ÄKTAxpress chromatography system (GE Healthcare).
goat anti-rabbit or goat anti-mouse, Molecular probes) and
After subsequent washing of the columns, antibodies were
subsequent wash with PBSB the cells were analyzed on a FACS
eluted with a low pH buffer (0.2 M glycine, 1 mM EGTA, pH 2.5),
Vantage SE (BD Biosciences). Analyzed samples included: no
followed by a buffer exchange to 1× PBS. Finally, glycerol and
antibodies on normal A-431 cells, secondary antibodies on
NaN3 were added to a final concentration of 50% and 0.02%,
normal A-431 control cells, primary and secondary antibodies
respectively. Binding to respective antigen and lack of cross-
on normal A-431 cells and primary and secondary antibodies
reactive binding to tag sequence and linker region was confirmed
on GPCR overexpressing A-431 cells.
by Western blot. Briefly, the respective antigen and one of the
other non-related antigens were separated on SDS-PAGE
gradient gels (10–20% Criterion, Bio-Rad Laboratories) and 2.7. Immunofluorescence and immunohistochemistry analysis
transferred to PVDF membranes (Criterion Gel™ Blotting
Sandwiches, Bio-Rad Laboratories). Subsequent to blocking of To confirm that the antibodies bind a protein on the cell
the membrane in blocking buffer (5% dry milk, 0.5% Tween20, 1× surface, confocal fluorescence microscopy was utilized, essen-
TBS; 0.1 M Tris–HCl, 0.5 M NaCl), the anti-GPCR antibodies, tially as described previously (Barbe et al., 2008). Briefly, glass
diluted 1/300–1/600, were added and the membrane was bottom plates (Whatman) were coated with fibronectin
incubated for 1 h. Secondary HRP conjugated goat-anti rabbit (Sigma) and seeded with 10,000 cells per well, grown for 5 h
antibody (DakoCytomation), diluted 1/3000, was added and and fixed with ice-cold 4% paraformaldehyde (Sigma). After
following incubation for 1 h, the membranes were soaked in permeabilization with 0.1% Triton X-100 (Sigma) the cells were
substrate solution (Immobilon Western Chemiluminescent HRP incubated with the primary antibody as well as a mouse anti-
Substrate, Millipore Corporation), and binding events were tubulin antibody (Abcam) overnight at 4 °C. Moreover, the cells
visualized in a Chemidoc CCD camera system (Bio-Rad were incubated for 1.5 h with secondary antibodies, goat anti-
Laboratories). rabbit IgG conjugated with Alexa Fluor 488 and rabbit anti-
mouse IgG Alexa Fluor 555 (Molecular Probes) and counter-
2.5. Generation of GPCR-expressing cell lines stained with DAPI, a nuclear probe. Image acquisition was
performed with a LSM 510 Meta confocal laser-scanning
The HEK 293T cell line was obtained from LGC Promochem microscope (Carl Zeiss GmbH). For the immunofluorescence
and the A431 cell line from DSMZ. Cells were cultivated in 5% staining of human brain tissue, unfixed cryostat sections were
CO2 at 37 °C in Dulbecco's modified Eagle's media (DMEM, incubated with antibodies NPY5R:1 and NPY5R:2 against the
Invitrogen) supplemented with 10% fetal bovine serum (FBS) NPY5R (dilution 1/1000) and processed according to the TSA+
and 1% Antibiotic–Antimycotic solution. For selection purposes, method (Adams, 1992) using a commercial kit and with green
the medium of infected cell lines contained blasticidin fluorescein isothiocyanate (FITC) as marker, as described in
(20 μg/ml media). B2AR, GLP1R and NPY5R were overexpressed detail elsewhere (Mulder et al., 2009). Immunohistochemical
in the human epidermoid carcinoma cell line, A-431, using a staining of human pancreatic islets with the GLP1R-specific
lentiviral gene delivery system (Suter et al., 2006). Full-length antibodies (GLP1R:1, GLP1R:2, MAB28141, GA1940 and
cDNA clones of B2AR, GLP1R and NPY5R were purchased from AB9433) was performed using formalin fixed, paraffin embed-
the Mammalian Gene Collection (Genservice and imaGenes, ded tissue samples. The staining was performed according to
GmbH). The gene fragments were amplified with PCR using previously described methods (Nilsson et al., 2005).
K. Larsson et al. / Journal of Immunological Methods 370 (2011) 14–23 17
A
B2AR 1002 bp
NPY5R 781 bp
GLP1R 848 bp
+ -
β-actin 70 bp
31
AR
1R
5R
A4
B2
LP
PY
G
N
100
0
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
120
Cell count
0
1 2 3 4 1 2 3 4 1 2 3 4
10 10 10 10 10 10 10 10 10 10 10 10
100
0
1 2 3 4 1 2 3 4 1 2 3 4
10 10 10 10 10 10 10 10 10 10 10 10
GA1940 MAB28141
100
Secondary antibody + A-431 cells
Primary/secondary antibody + A-431 cells
Primary/secondary antibody + over expressing A-431 cells
0
1 2 3 4 1 2 3 4
10 10 10 10 10 10 10 10
Fluorescence intensity
Fig. 2. (A) Detection of GPCR mRNA in the cell lines was performed by RT-PCR of extracted mRNA from overexpressing (+) as well as non-transfected A-431 cells
(−) using specific primers from NPY5R, B2AR and GLP1R respectively. Primers specific for the housekeeping gene β-actin were used to verify equal amount of
sample loading. (B) Flow cytometric analysis of antibodies specific for NPY5R, (C) B2AR and (D) GLP1R. Each histogram shows data obtained from three samples:
non-transfected A-431 cells incubated solely with secondary antibody (thin lines), non-transfected A-431 cells incubated with both primary and secondary
antibodies (dashed lines) and over expressing cells immunostained with the corresponding primary antibody and secondary antibody (thick lines).
20 K. Larsson et al. / Journal of Immunological Methods 370 (2011) 14–23
A B
C D
E F
Fig. 3. Representative immunofluorescent staining of GPCR-overexpressing as well as non-transfected A-431 cells. Antibody NPY5R:2 staining of (A) NPY5R
overexpressing A-431 cells and (B) A-431 cells, B2AR:1 staining of (C) B2AR overexpressing cells and (D) A-431 cells and finally GLP1R:2 staining of (E) GLP1R
overexpressing cells and (F) A-431 cells. The cells were counterstained with the nuclear probe DAPI (blue). Antibody binding is depicted in green. Bar indicates
10 μm.
produced toward the non-redundant set of human proteins, Validation of antibodies in general, and GPCR-specific
suggest that immunization with antigens shorter than 50 amino antibodies in particular, is of great importance and seldom an
acids generate significantly lower antibody titers as well as easy task. The specificity of antibodies targeting GPCRs is a
lower success rate in subsequent applications (unpublished well-known issue and has recently been discussed by Michel
data). We chose to assemble the antigen using the extracellular et al.(2009) Since most GPCRs are endogenously expressed at
parts of the GPCRs in order to detect the receptor protein on very low levels (Sarkar et al., 2008) validation of them is
intact cells, thus allowing research on endogenous receptor in its difficult, and therefore we generated three different cell lines,
native membrane bound form on living cells, but the same design overexpressing NPY5R, B2AR and GLP1R respectively. Over-
strategy could also be envisioned for the intracellular parts. expression of the specific G-protein-coupled receptors was
K. Larsson et al. / Journal of Immunological Methods 370 (2011) 14–23 21
Fig. 4. Immunofluorescence staining of human brain tissue and immunohistochemical staining of human pancreas. (A) Immunofluorescence micrograph taken from the
dorsal part of the periaqueductal gray in the human mesencephalon after incubation of an unfixed cryostat section with NPY5R:2 antibody. A large number of
fluorescent processes are seen, but no cell bodies. Bar indicates 1 μm. (B) Immunohisotochemical staining of human formalin fixed paraffin embedded pancreatic tissues
using GLP1R antibodies. GLP1R:1 gives a clear staining of plasma membranes (brown) in the pancreatic islets with no staining observed in exocrine glandular cells. No
specific immunohistochemical staining pattern could be detected after incubation with GA1940. The nuclei of the cells are stained with hematoxylin (blue). Bar indicates
100 μm.
successfully examined by RT-PCR, showing high expression in cell line (data not shown), indicating unspecific binding in the
the transfected cells and no detectable expression in the non- absence of the intended target protein. The antibodies
transfected cells. This permitted validation of the obtained generated against NPY5R and GLP1R were also used for
antibodies with flow cytometry and immunofluorescence- immunohistochemical and immunofluorescent staining of
based confocal microscopy. According to the flow cytometric human tissues expected to express the target proteins. In
analysis, most of the antibodies showed specific binding, i.e. both cases the antibodies showed staining that also supports
increased fluorescence intensity, to their respective GPCR- the earlier acquired validation data, indicating correct and
overexpressing cell line but not to non-transfected control selective binding of the antibodies.
cells. However, when using B2AR:2 and ab13300 no increased The epitope mapping showed that sub-populations of
fluorescence signals could be detected. This could possibly be antibodies within the purified antibody pools recognized
explained by low affinity of the antibodies to their target different regions of the respective antigen. Hence, there is a
receptor or that epitopes are hidden or conformational large potential in this novel strategy. Through the genetic fusion
different on the native receptor compared to the antigen. of many different epitopes, and by avoiding family specific
Immunofluorescence staining and analysis by confocal mi- regions, specific antibodies can potentially be raised with high
croscopy of overexpressing cells showed that many of the success rate. For large-scale production of antibodies toward
used antibodies gave staining of the plasma membrane, hence GPCRs using synthetic peptides, B-cell epitope prediction would
indicating binding to the membrane-bound target protein. be a requirement in order to reduce the number of experiments
Moreover, no or very weak staining could be detected in non- needed to achieve a good antibody. However, prediction of B-cell
transfected A-431 cells. The antibodies NPY5R:1 and GA1940 epitopes has recently been demonstrated to be very difficult
showed nuclear staining in non-transfected A-431 cells that (Blythe and Flower, 2005). The results from the epitope mapping
could not be observed in the corresponding overexpressing showed that a majority of the response was directed to the N-
22 K. Larsson et al. / Journal of Immunological Methods 370 (2011) 14–23
Fig. 5. Epitope mapping of GPCR antigens using synthetic peptides and Luminex technology. Intensity plots showing binding of (A) NPY5R:1 and NPY5R:2, (B)
B2AR:1 and B2AR:2 and (C) GLP1R:1 and GLP1R:2 to the N-terminus and extracellular loops of the respective antigen. Intensity is measured as median fluorescent
intensity (MFI). For sequences longer than 20 amino acids, overlapping peptides were used. The N-terminal of the GLP1R was represented as a 118 amino acids
long peptide.
terminus, probably due to the longer amino acid sequences, technical assistance. C. Stadler is acknowledged for assistance
especially for GLP1R. However, the B2AR:2 antibody almost with the confocal microscopy imaging, Dr. David Suter for
exclusively recognized the second and third extracellular loop. providing the 2K7-plasmid and Dr. Erwan LeMaitre for taking
Therefore our method to include all or selected extracellular parts immunofluorescence micrographs. The Knut and Alice Wal-
of the receptor is a promising approach for GPCR-specific lenberg Foundation financially supported this work. The
antibody generation. Since there are some inherited limitations support of the Swedish Research Council (04X-2888) is
with polyclonal antibodies our strategy to use these combined gratefully acknowledged.
antigens could also be envisioned for production of monoclonal
antibodies. The antigen could then be used for immunization, in
References
order to have a broader range of B-cells to choose from when
producing the hybridoma cells. Adams, J.C., 1992. Biotin amplification of biotin and horseradish peroxidase signals
Supplementary materials related to this article can be in histochemical stains. J. Histochem. Cytochem. 40 (10), 1457.
found online at doi:10.1016/j.jim.2011.05.001 Agaton, C., et al., 2003. Affinity proteomics for systematic protein profiling of
chromosome 21 gene products in human tissues. Mol. Cell. Proteomics 2
(6), 405.
Acknowledgements Agaton, C., et al., 2004. Selective enrichment of monospecific polyclonal antibodies
for antibody-based proteomics efforts. J. Chromatogr. A 1043 (1), 33.
Barbe, L., et al., 2008. Toward a confocal subcellular atlas of the human
The authors would like to thank Dr. J. Galli, Dr. J. Feldwisch proteome. Mol. Cell. Proteomics 7 (3), 499.
and K. Johansson (Affibody AB) for fruitful discussions and Berman, H.M., et al., 2000. The Protein Data Bank. Nucleic Acids Res. 28 (1), 235.
K. Larsson et al. / Journal of Immunological Methods 370 (2011) 14–23 23
Blythe, M.J., Flower, D.R., 2005. Benchmarking B cell epitope prediction: Pear, W.S., et al., 1993. Production of high-titer helper-free retroviruses by
underperformance of existing methods. Protein Sci. 14 (1), 246. transient transfection. Proc. Natl. Acad. Sci. U. S. A. 90 (18), 8392.
Castro, M., et al., 2005. Turn-on switch in parathyroid hormone receptor by a Sarkar, C.A., et al., 2008. Directed evolution of a G protein-coupled receptor
two-step parathyroid hormone binding mechanism. Proc. Natl. Acad. Sci. for expression, stability, and binding selectivity. Proc. Natl. Acad. Sci. U. S. A.
U. S. A. 102 (44), 16084. 105 (39), 14808.
Gupta, A., Devi, L.A., 2006. The use of receptor-specific antibodies to study G- Sjölander, A., et al., 1997. The serum albumin-binding region of streptococcal
protein-coupled receptors. Mt. Sinai J. Med. 73 (4), 673. protein G: a bacterial fusion partner with carrier-related properties.
Jongsma, M., et al., 2007. Validation of a rapid, non-radioactive method to J. Immunol. Methods 201 (1), 115.
quantify internalisation of G-protein coupled receptors. Naunyn- Ståhl, S., et al., 1993. Solid-phase gene assembly of constructs derived from
Schmiedebergs Arch. Pharmacol. 375 (5), 329. the Plasmodium falciparum malaria blood-stage antigen Ag332. Bio-
Larsson, M., et al., 2000. High-throughput protein expression of cDNA techniques 14 (3), 424.
products as a tool in functional genomics. J. Biotechnol. 80 (2), 143. Steen, J., et al., 2006. High-throughput protein purification using an
Larsson, K., et al., 2009. Characterization of PrEST-based antibodies towards automated set-up for high-yield affinity chromatography. Protein Expr.
human Cytokeratin-17. J. Immunol. Methods 342 (1–2), 20. Purif. 46 (2), 173.
Lundström, K., 2005. Structural genomics of GPCRs. Trends Biotechnol. 23 (2), 103. Suter, D.M., et al., 2006. Rapid generation of stable transgenic embryonic
Mackrill, J.J., 2004. Generation, use, and validation of receptor-selective stem cell lines using modular lentivectors. Stem Cells 24 (3), 615.
antibodies. Methods Mol. Biol. 259, 47. Thorens, B., 1992. Expression cloning of the pancreatic beta cell receptor for
Michel, M.C., Wieland, T., Tsujimoto, G., 2009. How reliable are G-protein- the gluco-incretin hormone glucagon-like peptide 1. Proc. Natl. Acad. Sci.
coupled receptor antibodies? Naunyn-Schmiedebergs Arch. Pharmacol. U. S. A. 89 (18), 8641.
379 (4), 385. Vernet, E., et al., 2009. Affibody-mediated retention of the epidermal growth
Mulder, J., et al., 2009. Tissue profiling of the mammalian central nervous system factor receptor in the secretory compartments leads to inhibition of
using human antibody-based proteomics. Mol. Cell. Proteomics 8 (7), 1612. phosphorylation in the kinase domain. N. Biotechnol.
Nguyen, T.N., Uhlén, M., Ståhl, S., 1994. De novo gene assembly using a Wess, J., 1997. G-protein-coupled receptors: molecular mechanisms involved
paramagnetic solid support. In: Uhlén, M., Hornes, E., Olsvik, Ö. (Eds.), in receptor activation and selectivity of G-protein recognition. FASEB J.
Advances in Biomagnetic Separation. Eaton Publishing Co., Natick, p. 73. 11 (5), 346.
Nilsson, P., et al., 2005. Towards a human proteome atlas: high-throughput Zhang, Y., et al., 2004. Selection of active ScFv to G-protein-coupled receptor
generation of mono-specific antibodies for tissue profiling. Proteomics 5 CCR5 using surface antigen-mimicking peptides. Biochemistry 43 (39),
(17), 4327. 12575.