TaqManR Proximity ligation technology for the detection of heterodimeric adhesion receptors on lymphocytes

TaqManR Proximity ligation technology for the detection of heterodimeric adhesion receptors on lymphocytes

Journal of Immunological Methods 404 (2014) 81–86 Contents lists available at ScienceDirect Journal of Immunological Methods journal homepage: www.e...

654KB Sizes 0 Downloads 20 Views

Journal of Immunological Methods 404 (2014) 81–86

Contents lists available at ScienceDirect

Journal of Immunological Methods journal homepage: www.elsevier.com/locate/jim

Technical note

TaqManR Proximity ligation technology for the detection of heterodimeric adhesion receptors on lymphocytes Renate Gehwolf a,e, Elisabeth Band b, Andrea Trost c, Bernhard Iglseder a, Eugen Trinka b, Elisabeth Haschke-Becher d, Jörg Kraus b, Andrea Harrer b,⁎ a

Department of Geriatric Medicine, Christian-Doppler-Klinik, Paracelsus Medical University, Ignaz-Harrer-Str. 79, 5020 Salzburg, Austria Department of Neurology, Christian-Doppler-Klinik, Paracelsus Medical University, Ignaz-Harrer-Str. 79, 5020 Salzburg, Austria c Department of Ophthalmology, St. Johanns Spital, Paracelsus Medical University, Müllner Hauptstrasse 48, 5020 Salzburg, Austria d Central Laboratory, Christian-Doppler-Klinik, Paracelsus Medical University, Ignaz-Harrer-Str. 79, 5020 Salzburg, Austria e Institute of Tendon and Bone Regeneration, Spinal Cord Injury and Tissue Regeneration Center Salzburg, Paracelsus Medical University, Strubergasse 22, 5020 Salzburg, Austria b

a r t i c l e

i n f o

Article history: Received 29 April 2013 Received in revised form 27 October 2013 Accepted 20 November 2013 Available online 1 December 2013 Keywords: Immuno-PCR Proximity ligation technology Lymphocytes Receptor heterodimers Alpha-4/beta-1 integrin

a b s t r a c t TaqManR proximity ligation technology (TaqManR PLA) is an innovative advancement of immuno PCR. It allows a fast and quantitative detection of vicinal proteins or protein–protein interactions from cell lysates by combining antibody–antigen binding with a real-time PCR detection. We tested if TaqManR PLA also was applicable to investigate and relatively quantitate adhesion receptor heterodimers such as the alpha-4/beta-1 integrin on the surface of intact cells. Both, alpha-4, beta-1 and the alpha-4/beta-1 heterodimer were detected on the surface of lymphocytes by TaqManR PLA. Results were specific, reproducible and comparable to flow cytometric data. However, preciseness of reactions varied dependent on the antibody pairs used. Co-detection of proximate identical subunits suggested clusters of alpha-4 and/or beta-1 on the cell surface which we confirmed by microscopy. We conclude that real-time PCR-based TaqManR PLA is of limited applicability to investigate heterodimeric receptor molecules such as the alpha-4/beta-1 integrin. Determination of an abundance ratio of alpha-4/beta-1 in relation to total alpha-4 or beta-1 was not possible and real-time detection did not allow conclusions on the surface distribution of molecules. The related in situ PLA developed for microscopy allows visualizing proximate protein interactions and might be an interesting alternative for research into receptor heterodimers and their surface distribution on immune cells. © 2013 Elsevier B.V. All rights reserved.

1. Introduction The alpha-4/beta-1 (syn. very late activation antigen-4, VLA-4) integrin is a heterodimeric adhesion molecule on ⁎ Corresponding author at: Department of Neurology, Christian-Doppler-Klinik, Ignaz-Harrer-Str. 79, 5020 Salzburg, Austria. Tel.: + 43 662 4483 56033; fax: + 43 662 4483 3004. E-mail addresses: [email protected] (R. Gehwolf), [email protected] (E. Band), [email protected] (A. Trost), [email protected] (B. Iglseder), [email protected] (E. Trinka), [email protected] (E. Haschke-Becher), [email protected] (J. Kraus), [email protected] (A. Harrer). 0022-1759/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jim.2013.11.024

peripheral blood mononuclear cells (PBMC) and crucial for immune cell transmigration across the blood–brain barrier into the central nervous system (Engelhardt and Ransohoff, 2005; Yednock et al., 1992). As cell surface receptors integrins are easily accessible for monoclonal antibody therapeutics, and blocking the alpha-4 subunit by natalizumab is an effective treatment in multiple sclerosis (Polman et al., 2006). Integrin-mediated adhesion and their contribution to intercellular interactions, however, are complex and far from being understood. Severe side effects such as progressive multifocal leukoencephalopathy in patients receiving monoclonal antibody therapies against integrins demonstrate the importance of ongoing efforts to

82

R. Gehwolf et al. / Journal of Immunological Methods 404 (2014) 81–86

investigate these heterodimeric receptors on immune cells (Clifford et al., 2010;Kappos et al., 2011;Schwab et al., 2012). Flow cytometry is a frequently used method for investigating the expression of adhesion molecules on the surface of immune cells. Antibodies, however, typically do not recognize the receptor heterodimer but single subunits thereof. We therefore searched for a method allowing both, detecting heterodimeric receptors and detecting their individual subunits. The proximity ligation assay (PLA) technology uses dual target recognition to detect protein–protein interactions, posttranslational modifications, and vicinal proteins (Soderberg et al., 2007). The conversion of protein to a DNA template for PCR is enabled by a pair of biotinylated detection antibodies which, when labeled with streptavidin-coupled oligonucleotides, constitute “proximity probes” with specificities for nearby epitopes. Upon binding to proximate epitopes (b 40 nm) the oligonucleotide tails are linked in an enzymatic ligation reaction thus forming the template for the subsequent PCR. In TaqManR PLA (Life Technologies Corporation, Applied Biosystems, Austria) the linkage product is amplified by real-time PCR allowing a relative quantification of proximity reaction products (Swartzman et al., 2010). This real-time PCR-based PLA method appeared applicable for investigating heterodimeric surface receptors and technically feasible to perform routine measurements with patient samples later on. However, TaqManR PLA was optimized for studying proximity reactions from cell lysates. The principal questions therefore were (i) whether this technology was suitable for detecting receptor heterodimers such as the alpha4/beta-1 integrin on the surface of intact cells and (ii) if it was possible to quantitatively discriminate receptor heterodimers from the total of individual monomeric subunits thereof (Fig. 1A–C).

biotinylated using the Biotin-XX-Microscale Protein L Kit (Life Technologies, Invitrogen, Austria). Biotinylated antibodies (200 nM) were labeled one-to-one with 200 nM streptavidinlinked 5′ or 3′ oligonucleotide tails (5′ or 3′ Prox-Oligo) according to the TaqManR Protein Assays Probe Development Protocol (Life Technologies, Applied Biosystems, Austria). 2.3. Real-time based PLA Lymphocytes or buffer only (recommended background control by the TaqManR protein Expression Assay Protocol) were incubated with pairs of complementary 3′ and 5′ proximity probes (10 nM, each) under varying conditions (temperature, incubation times). The ligation reaction and PCR were performed using the TaqManR Protein Expression Assay reagent kit according to the instructions. Delta (Δ) Cts were calculated by subtracting the Ct of the sample reaction from the Ct of the background control. 2.4. Comparison of PLA and flow cytometry Surface expression of alpha-4 and beta-1 was investigated on CD4 + and CD8+ T cells by PLA with proximity probe pairs consisting of clones 7.2R and HP2/1 (detection of alpha-4) and clone 4B4 and polyclonal anti-beta-1 (detection of beta-1). Two-color flow cytometry was performed by staining the same sorted CD4 + and CD8+ T cell preparations either with FITC-labeled anti-alpha-4 (clone HP2/1) or FITC-labeled anti-beta-1 (clone 4B4) in combination with anti-CD4-PC7 or anti-CD8-PC5 detection antibodies. 3. Results 3.1. Real-time PCR-based PLA and the detection of proximate surface molecules on intact cells

2. Materials and methods 2.1. Lymphocytes Venous blood from healthy donors (n = 4) was collected in CPT tubes (Becton Dickinson, Basel, Switzerland) for enrichment of peripheral blood mononuclear cells (PBMC). Monocytes were depleted by plastic adherence and purity of lymphocytes was controlled by a haematology analyser (Lab Sysmex 1800, Vienna, Austria). Untouched CD4 + and CD8 + T cells were purified by indirect magnetic labelling with the CD4 + T cell Isolation Kit II and CD8+ T cell Isolation Kit (Miltenyi Biotec GmBH, Bergisch Gladbach, Germany). Monocyte-depleted lymphocytes and the two sorted T cell subpopulation were aliquoted and cryo-stored (−80 °C) until analysis. 2.2. Proximity probes Biotinylated mouse anti-human alpha-4 (CD49d; 7.2R) and polyclonal goat anti-human beta-1 (immunogen: CHO-derived recombinant human integrin beta-1 extracellular domain) antibodies were purchased from Biomedica (Vienna, Austria). Mouse anti-human alpha-4 (HP2/1), mouse anti-human beta-1 (4B4, both Beckman Coulter, Austria), and mouse anti-X (antibody devoid of specificity to human immune cells) were

Detection of single alpha-4 or beta-1 subunits and the alpha-4/beta-1 heterodimeric integrin was successful after an overnight binding reaction on ice using 5*104 lymphocytes and correspondent complementary proximity probe pairs (10 nM each) (Fig. 2A). Results could be reproduced in separate runs (n = 5) and, as long as the same monoclonal antibody pair was used, there was no difference whether 5′-alpha-4 (clone HP2/1)/3′-beta-1 (clone 4B4) (ΔCt: 7.43, SD ± 0.55) or vice-versa labeled 5′-beta-1 (clone 4B4)/ 3′-alpha-4 (clone HP2/1) (ΔCt: 7.83, SD ± 0.56) were combined (Fig. 2A). A monoclonal antibody devoid of specificity for human immune cells (5′-anti-x/3′-anti-x) served as specificity control for the proximity reaction on intact cells (Fig. 1D) and was pairwise tested with complementary oligo-labeled anti-alpha-4 (clone 7.2R) or anti-beta-1 (clone 4B4) probes, respectively. The resulting Ct values did not differ from background levels which proved specificity of the proximity reaction (Fig. 2C, grey bars). The ΔCt for anti-x/alpha-4 was 0.4 (SD ± 0.14) and for anti-x/beta-1 0.48 (SD ± 0.35). 3.2. Preciseness of real-time PCR based PLA in the detection of surface molecules The preciseness of the PLA reaction, however, depended on the antibody pairs used and polyclonal anti-beta-1 performed

R. Gehwolf et al. / Journal of Immunological Methods 404 (2014) 81–86

83

Fig. 1. Proximity ligation assay and the detection of the heterodimeric alpha-4/beta-1 integrin on the surface of immune cells. Proximity ligation probes were generated by linking streptavidin-conjugated 5´- and 3´- oligonucleotide tails to biotinylated anti-alpha-4 and anti-beta-1 antibodies. (A, B) Detection of single alpha-4 and beta-1 subunits by use of complementary 5´- and 3´-probe pairs with epitope specificities for the alpha-4 and beta-1 integrin subunit, respectively. (C) Detection of heterodimeric alpha-4/beta-1 integrin with complementary 5´-anti-alpha-4 and 3’-anti-beta-1 probes. (D) Specificity control with an oligo-labeled antibody without epitope specificity for immune cells (anti-x) as complementary probe. Abbreviations: α4, alpha-4; β1, beta-1; B, biotin; L, linker DNA (added during ligation reaction); SA, streptavidin.

suboptimal. Detection of the receptor heterodimer with the clones 7.2R (alpha-4) and 4B4 (beta-1) resulted in a ΔCt of 9.38 (SD ± 0.92). Use of clone HP2/1 (alpha-4) and polyclonal beta-1 resulted in a ΔCt of only 5.2 (SD ± 0.62) which means a difference in ΔCt of minus 4.06 (SD ± 1.41) (Fig. 2A, 1st and 2nd orange bars). Moreover, detection of total beta-1 with either two complementary-labeled polyclonal anti-beta-1

probes or with one polyclonal anti-beta-1 probe paired with the monoclonal probe of clone 4B4 resulted in lower ΔCts (3.97 ± 0.23 and ΔCt: 6.08 ± 0.89) than obtained for the alpha-4/beta-1 receptor heterodimer which was implausible (Fig. 2A, cut-off line). The quality of the polyclonal anti-beta1 antibody was controlled in a dot-blot experiment using recombinant alpha-4/beta-1 as antigen (Biomedica). Both,

84

R. Gehwolf et al. / Journal of Immunological Methods 404 (2014) 81–86

Fig. 2. Heterodimeric alpha-4/beta-1 integrin and the individual subunits detected by PLA. (A) Results (mean ΔCt values, n = 5) from different probe combinations for detecting the beta-1 subunit (green), and alpha-4/beta-1 heterodimers (orange). Preciseness of reactions depended on the antibodies used. The dashed line highlights suboptimal detection results from reactions involving polyclonal (p) anti-beta-1 probes. The maximal difference in ΔΔCt (4.06; ΔCt calculated by subtracting from background control) was observed when polyclonal (p) beta-1 was involved. Results were comparable with vice-versa labeled complementary probes for the detection of heterodimeric alpha-4/beta-1. (B) Dot-blot showing equal detection properties of anti-alpha-4 and polyclonal anti-beta-1 for recombinant human alpha-4/beta-1 integrin. (C) Specificity of reactions was controlled by pairing anti-alpha-4 (red) and anti-beta-1 (green) probes with a complementary probe of irrelevant epitope specificity for immune cells (3′ anti-x). The resulting ΔCt values did not exceed background levels (grey bars). Potential co-detection of vicinal integrin subunits was controlled by use of complementary monoclonal probe pairs with identical epitope specificities (alpha-4: clone HP2/1; reddish bar, beta-1: clone 4B4, greenish bar). ΔCt values approximated the detection of single alpha-4 subunits (red) or beta-1 subunits (green) with probe pairs of different epitope specificities indicating involvement of vicinal integrin subunits in proximity reactions. (D) Surface distribution of alpha-4 and beta-1 on immune cells visualized by confocal immunofluorescence microscopy (LSM710, Zeiss): Upper panel showing the surface distribution of beta-1 (green spots) and alpha-4 (red spots) integrin subunits, lower panel showing the bright field image and the merged image with clusters of alpha-4 only (red spots and arrow), clusters of beta-1 only (green spots and arrow), and colocalization of alpha-4 and beta-1 (yellow spots and arrow). Abbreviations: α4, alpha-4; β1, beta-1; biot, biotinylated; BSA, bovine serum albumin; p, polyclonal; SD, standard deviation; *cut-off line to illustrate that ΔCt values for the detection of alpha-4/beta-1 heterodimers exceed ΔCt values for the detection of beta-1 subunits.

polyclonal anti-beta-1 and anti-alpha-4 (7.2R) detected the recombinant protein with comparable efficiency (Fig. 2B). 3.3. Co-detection of vicinal alpha-4 and beta-1 subunits Probe pairs with identical epitope specificity (5′-HP2.1/3′HP2.1 for alpha-4, 5′-4B4/3′-4B4 for beta-1) were assayed to test if neighboring integrin subunits were involved in the proximity reaction. The resulting ΔCts (alpha-4: 9.05 ± 0.41; beta-1: 6.78 ± 0.66; Fig. 2C) suggested detection also of subunits of vicinal integrins. This suggestion was supported by subsequent immunofluorescence microscopy in which clusters of either alpha-4 or beta-1 and clusters of colocalized alpha-4 and beta-1 were observed on the surface of lymphocytes (Fig. 2D). 3.4. PLA versus flow cytometry PLA and flow cytometric data were compared using purified CD4+ and CD8+ T cells from different donors. Results obtained by PLA showed higher surface levels of alpha-4 on CD8+

compared to CD4+ T cells in all four donors. Analysing beta-1 levels on the two T cell subpopulations by PLA revealed different expression patterns comparing the individuals. Donors 1 and 4 showed higher beta-1 levels on CD8+ compared to CD4+ T cells whereas donor 2 displayed higher beta-1 levels on CD4+ compared to CD8+ T cells. Equal beta-1 levels were detected on both T cell subpopulations in donor 3. Comparing PLA data with flow cytometry showed congruent results in all four donors concerning alpha-4 expression (Fig. 3A + B). Beta-1 expression levels were confirmed in donors 2, 3, and 4 but discrepant in donor 1 (Fig. 3C + D). 4. Discussion After optimizing cell numbers, temperature, and incubation time we were successful in detecting alpha-4, beta-1 and alpha-4/beta-1 heterodimers on the surface of lymphocytes. Specificity of the proximity reaction was controlled by an antibody devoid of specificity to immune cells (anti-x), and, thus incapable of generating a proximity-based PCR product. By use of the anti-x antibody we were able to show that

R. Gehwolf et al. / Journal of Immunological Methods 404 (2014) 81–86

85

Fig. 3. Comparison of PLA (left panel) and flow cytometry (right panel). Upper panel: Detection of alpha-4 by PLA (A) and flow cytometry (B). Lower panel: Detection of beta-1 by PLA (C) and flow cytometry (D). Bars represent standard deviations of the mean (triplicates). CD4, CD4+ T cells; CD8, CD8+ T cells; CT, cycle threshold; MFI, median fluorescence intensity.

lymphocytes binding just one probe do not serve as “solid phase” (in analogy to immunobead assays) and capture the complementary probe from the solution via oligonucleotidetail hybridization despite a lack of spatial proximity. So far the methodology appeared promising. However, the necessity to use more than 10 to the 4 immune cells in order to be able to detect alpha-4/beta-1 heterodimers was not convincing in regard to sensitivity. Moreover, preciseness of the reaction differed dependent on the antibody pair used. Especially the polyconal anti-beta-1 probes performed suboptimal in the detection of both, total beta-1 and alpha-4/beta-1 heterodimers. The resulting lower ΔCts for total beta-1 compared to alpha-4/beta-1 heterodimers were implausible and rendered our intention to determine an abundance ratio between subunit heterodimers and monomers impossible. This was unexpected, since the polyclonal anti-beta-1 antibody used in this study had passed a screening TaqMan Protein Assay performed by the company (TaqManR Protein Assays Probe Development Protocol, PN4499282A). Apparently, this antibody worked in cell lysate but was less qualified to access the receptor complex on the surface of intact cells. We confirmed this assumption by dot-blot experiments showing that polyclonal anti-beta-1 and monoclonal antialpha-4 equally well detected recombinant alpha-4/beta-1 spotted onto nitrocellulose. Another important question was whether single monomeric subunits and individual receptor heterodimers were detected or if vicinal integrins were close enough to be involved in proximity ligation reactions. This question was approached by assaying pairs of identical monoclonal probes because the monoclonal antibodies only bind one single

epitope per subunit. Detection of alpha-4 and beta-1 with two monoclonal probes therefore demonstrated involvement of vicinal integrins, respectively their subunits, in the proximity reactions. This was interesting since co-detection of neighboring integrin subunits suggested either presence of homodimers or a tightly clustered distribution of subunit monomers on the cell surface. We actually observed clusters of alpha-4 or beta-1 and clusters in which both subunits appeared colocalized on the surface of lymphocytes by immunofluorescence microscopy — an issue for further research. Last we investigated comparability of PLA data with flow cytometric data. Both methods matched perfectly in detecting the differential expression of alpha-4 on CD4+ and CD8+ T cells from the four donors. The detection of differential beta-1 levels, however, was consistent in only three of the four cases. Considering the different detection strategies of PLA and flow cytometry, the observed suboptimal performance of the polyclonal anti-beta-1 probes in PLA, and the yet unknown impact of surface clusters of either alpha-4 or beta-1 on the proximity-based detection system, we conclude an overall good comparability of results. Summing up — alpha-4, beta-1, and receptor heterodimers were detected on the surface of lymphocytes by TaqManR PLA and results were comparable with flow cytometry. A relative quantitation was not possible due to differential antibody binding properties, and monoclonal antibodies were preferable to polyclonals. Vicinal integrins were also detected but the real-time PCR-based detection method did not allow drawing conclusions on the surface distribution of the molecules. TaqManR PLA hence proved to be of limited applicability for investigating receptor heterodimers on immune cells. In situ PLA, which has been developed for

86

R. Gehwolf et al. / Journal of Immunological Methods 404 (2014) 81–86

microscopy and recently also for flow cytometry (Leuchowius, et al., 2011) allows visualizing the product of the proximity ligation reaction and might be an interesting alternative for research into heterodimeric surface receptors. Acknowledgments We thank Dr. Martin Himly of the Paris-Lodron University of Salzburg for kindly providing the mouse anti-X antibody which served as specificity control for the proximity reaction in our experiments. This study was funded by the PMU-FFF Project R-10/05/019-HAR. References Clifford, D.B., De Luca, A., Simpson, D.M., Arendt, G., Giovannoni, G., Nath, A., 2010. Natalizumab-associated progressive multifocal leukoencephalopathy in patients with multiple sclerosis: lessons from 28 cases. Lancet Neurol. 9, 438. Engelhardt, B., Ransohoff, R.M., 2005. The ins and outs of T-lymphocyte trafficking to the CNS: anatomical sites and molecular mechanisms. Trends Immunol. 26, 485. Kappos, L., Bates, D., Edan, G., Eraksoy, M., Garcia-Merino, A., Grigoriadis, N., Hartung, H.P., Havrdova, E., Hillert, J., Hohlfeld, R., Kremenchutzky, M., Lyon-Caen, O., Miller, A., Pozzilli, C., Ravnborg, M., Saida, T., Sindic,

C., Vass, K., Clifford, D.B., Hauser, S., Major, E.O., O'Connor, P.W., Weiner, H.L., Clanet, M., Gold, R., Hirsch, H.H., Radu, E.W., Sorensen, P.S., King, J., 2011. Natalizumab treatment for multiple sclerosis: updated recommendations for patient selection and monitoring. Lancet Neurol. 10, 745. Leuchowius, K.J., Weibrecht, I., Soderberg, O., 2011. In situ proximity ligation assay for microscopy and flow cytometry. Curr Protoc Cytom 56, 9.36.1–9.36.15. Polman, C.H., O'Connor, P.W., Havrdova, E., Hutchinson, M., Kappos, L., Miller, D.H., Phillips, J.T., Lublin, F.D., Giovannoni, G., Wajgt, A., Toal, M., Lynn, F., Panzara, M.A., Sandrock, A.W., 2006. A randomized, placebocontrolled trial of natalizumab for relapsing multiple sclerosis. N. Engl. J. Med. 354, 899. Schwab, N., Ulzheimer, J.C., Fox, R.J., Schneider-Hohendorf, T., Kieseier, B.C., Monoranu, C.M., Staugaitis, S.M., Welch, W., Jilek, S., Du Pasquier, R.A., Bruck, W., Toyka, K.V., Ransohoff, R.M., Wiendl, H., 2012. Fatal PML associated with efalizumab therapy: insights into integrin alphaLbeta2 in JC virus control. Neurology 78, 458. Soderberg, O., Leuchowius, K.J., Kamali-Moghaddam, M., Jarvius, M., Gustafsdottir, S., Schallmeiner, E., Gullberg, M., Jarvius, J., Landegren, U., 2007. Proximity ligation: a specific and versatile tool for the proteomic era. Genet. Eng. 28, 85 (N. Y.). Swartzman, E., Shannon, M., Lieu, P., Chen, S.M., Mooney, C., Wei, E., Kuykendall, J., Tan, R., Settineri, T., Egry, L., Ruff, D., 2010. Expanding applications of protein analysis using proximity ligation and qPCR. Methods 50, S23. Yednock, T.A., Cannon, C., Fritz, L.C., Sanchez-Madrid, F., Steinman, L., Karin, N., 1992. Prevention of experimental autoimmune encephalomyelitis by antibodies against alpha 4 beta 1 integrin. Nature 356, 63.