HKOR (b) and hMOR (c) with six 
histidine residues and a c-myc tag fused to their C-terminus were over-expressed in Pichia Pastoris. GPCR-enriched crude fractions obtained by differential centrifugation of GPCR-expressing yeast cell lysates were solubilized and then chromatographed on a nickel affinity column. Ni-bound GPCRs were eluted with imidazole-containing buffer. GPCR-enriched samples run in SDS-polyacrylamide gels were stained by silver nitrate (left panels) or probed with anti-c-myc monoclonal antibody after transfer onto PVDF membrane (right panels). doi:10.1371/4 IBP chemical information journal.pone.0046348.gcases, the 15900046 staining patterns of anti-GPCR BI 78D3 chemical information peptide antibodies are similar in wild-type and GPCR-deficient mice as assessed by immunohistochemistry or western-blotting [16,17,18,19,20,21,22]. A recent study, comparing the specificity of a number of commercial anti-opioid receptor antibodies, has shown that all the antibodies revealed numerous non-specific bands including a band at the expected molecular weight in both wild-type CHO cells (negative control) and GPCR-expressing CHO cells as assessed by western-blotting [23]. Given the lack of specificity of anti-GPCR peptide antibodies, it is now generally accepted that the production of relevant anti-GPCR antibodies able to recognize native proteins requires immunizing animals with receptors in native conformation, thus maintaining their ligand-binding activity [7,10,24,25]. The hydrophobic nature of GPCRs and their low natural expression make, however, the purification of high amounts of native-like functional receptors a formidable challenge [26,27]. Considering the technical skills that are needed to obtain GPCRs in native form, the peptide strategy is still used by many laboratories despite its low success level. Here, we describe a novel strategy to easily produce highly specific anti-GPCR antibodies by using purified denatured full length recombinant GPCRs, as immunogens. We show an unanticipated finding that correctly folded native-like receptors are not required to produce highly specific antibodies to GPCRs. This method successfully 
applied to the human neuropeptide FF receptor type 2 (hNPFFR2), the human k opioid receptor (hKOR) and the human m opioid receptor (hMOR) might be extended to a wide range of other GPCRs and applicable in most laboratories.their C-terminal histidine residues were eluted using an imidazole gradient buffer [28]. Purity of each GPCR-enriched preparation was assessed by comparing silver nitrate staining of SDS-PAGE gels with an anti-c-myc immunoblotting (Fig. 1). All proteins stained by silver nitrate were also revealed with anti-c-myc antibodies indicating that the three GPCR preparations used to immunize mice were virtually pure. Silver nitrate dyeing as well as anti-c-myc staining, revealed bands with apparent molecular weights of 47 kDa, 40 kDa and 47.5 kDa corresponding to the calculated size of hNPFFR2 (Fig. 1a), hKOR (Fig. 1b) and hMOR (Fig. 1c) respectively (Table 1). Bands with higher molecular weights were also stained with both silver nitrate and anti-c-myc antibodies. Thus, the preparations of recombinant GPCRs purified from Pichia Pastoris contain receptors in unglycosylated monomeric forms but also in polymeric and/or glycosylated forms as described elsewhere [29,30,31]. Receptor preparations depleted in their imidazole content (to minimize toxicity) were then either maintained in solubilized form in 0.1 SDS or dialyzed against water and lyophilized. T.HKOR (b) and hMOR (c) with six histidine residues and a c-myc tag fused to their C-terminus were over-expressed in Pichia Pastoris. GPCR-enriched crude fractions obtained by differential centrifugation of GPCR-expressing yeast cell lysates were solubilized and then chromatographed on a nickel affinity column. Ni-bound GPCRs were eluted with imidazole-containing buffer. GPCR-enriched samples run in SDS-polyacrylamide gels were stained by silver nitrate (left panels) or probed with anti-c-myc monoclonal antibody after transfer onto PVDF membrane (right panels). doi:10.1371/journal.pone.0046348.gcases, the 15900046 staining patterns of anti-GPCR peptide antibodies are similar in wild-type and GPCR-deficient mice as assessed by immunohistochemistry or western-blotting [16,17,18,19,20,21,22]. A recent study, comparing the specificity of a number of commercial anti-opioid receptor antibodies, has shown that all the antibodies revealed numerous non-specific bands including a band at the expected molecular weight in both wild-type CHO cells (negative control) and GPCR-expressing CHO cells as assessed by western-blotting [23]. Given the lack of specificity of anti-GPCR peptide antibodies, it is now generally accepted that the production of relevant anti-GPCR antibodies able to recognize native proteins requires immunizing animals with receptors in native conformation, thus maintaining their ligand-binding activity [7,10,24,25]. The hydrophobic nature of GPCRs and their low natural expression make, however, the purification of high amounts of native-like functional receptors a formidable challenge [26,27]. Considering the technical skills that are needed to obtain GPCRs in native form, the peptide strategy is still used by many laboratories despite its low success level. Here, we describe a novel strategy to easily produce highly specific anti-GPCR antibodies by using purified denatured full length recombinant GPCRs, as immunogens. We show an unanticipated finding that correctly folded native-like receptors are not required to produce highly specific antibodies to GPCRs. This method successfully applied to the human neuropeptide FF receptor type 2 (hNPFFR2), the human k opioid receptor (hKOR) and the human m opioid receptor (hMOR) might be extended to a wide range of other GPCRs and applicable in most laboratories.their C-terminal histidine residues were eluted using an imidazole gradient buffer [28]. Purity of each GPCR-enriched preparation was assessed by comparing silver nitrate staining of SDS-PAGE gels with an anti-c-myc immunoblotting (Fig. 1). All proteins stained by silver nitrate were also revealed with anti-c-myc antibodies indicating that the three GPCR preparations used to immunize mice were virtually pure. Silver nitrate dyeing as well as anti-c-myc staining, revealed bands with apparent molecular weights of 47 kDa, 40 kDa and 47.5 kDa corresponding to the calculated size of hNPFFR2 (Fig. 1a), hKOR (Fig. 1b) and hMOR (Fig. 1c) respectively (Table 1). Bands with higher molecular weights were also stained with both silver nitrate and anti-c-myc antibodies. Thus, the preparations of recombinant GPCRs purified from Pichia Pastoris contain receptors in unglycosylated monomeric forms but also in polymeric and/or glycosylated forms as described elsewhere [29,30,31]. Receptor preparations depleted in their imidazole content (to minimize toxicity) were then either maintained in solubilized form in 0.1 SDS or dialyzed against water and lyophilized. T.