Protein Microarrays for Identification of Novel Extracellular Protein‐Protein Interactions

Irene Tom1, Nicholas Lewin‐Koh2, Sree R. Ramani1, Lino C. Gonzalez1

1 Department of Protein Chemistry, Genentech, South San Francisco, California, 2 Department of Nonclinical Biostatistics, Genentech, South San Francisco, California
Publication Name:  Current Protocols in Protein Science
Unit Number:  Unit 27.3
DOI:  10.1002/0471140864.ps2703s72
Online Posting Date:  April, 2013
GO TO THE FULL TEXT: PDF or HTML at Wiley Online Library


Functional protein microarrays offer the capability for high‐throughput protein interaction analysis and have long promised to be a powerful tool for understanding protein interactions at the proteome scale. Although popular techniques for protein‐protein interaction mapping like yeast‐two‐hybrid and affinity‐purification mass spectrometry have performed well for identifying intracellular protein‐protein interactions, the study of interactions between extracellular proteins has remained challenging for these methods. Instead, the use of protein microarrays appears to be a robust and efficient method for the identification of interactions among the members of this class of protein. This unit describes methods for extracellular protein microarray production, screening, and analysis. A protocol is described for enhanced detection of low‐affinity interactions by generating multivalent complexes using Fc‐fusion bait proteins and protein A microbeads, along with a statistical method for hit scoring and identification of nonspecific interactions. Curr. Protoc. Protein Sci. 72:27.3.1‐27.3.24. © 2013 by John Wiley & Sons, Inc.

Keywords: protein microarray; microarray analysis; extracellular protein; multivalent binding; protein‐protein interaction

PDF or HTML at Wiley Online Library

Table of Contents

  • Introduction
  • Basic Protocol 1: Production of Extracellular Protein Microarrays
  • Basic Protocol 2: Preparation of Labeled Bait Protein and Multivalent Microbead Complexes
  • Alternate Protocol 1: Formation of Cy5‐Labeled IgG and Fc‐Fusion Bait Protein Multivalent Complexes on Protein A Microbeads
  • Basic Protocol 3: Protein Microarray Screening and Processing
  • Basic Protocol 4: Protein Microarray Data Analysis
  • Reagents and Solutions
  • Commentary
  • Literature Cited
  • Figures
  • Tables
PDF or HTML at Wiley Online Library


Basic Protocol 1: Production of Extracellular Protein Microarrays

  • Purified proteins
  • PBS, pH 7.2 (see recipe)
  • 80% (v/v) glycerol (Ultrapure MB grade; USB corporation, cat. no. 56‐81‐5) in PBS (see recipe for PBS)
  • 40% and 50% (v/v) glycerol in PBS
  • 20% (v/v) ethanol
  • ZeptoMARK blocking buffer BB1 (Zeptosens, cat. no. 9040)
  • Bovine serum albumin fraction V, heat shock, fatty‐acid free (Roche Applied Science, cat. no. 03‐117‐057‐001; optional)
  • 5% milk in PBST (see recipe)
  • PBST (see recipe)
  • Low‐protein‐binding 0.2‐µm membrane filters (PALL Life Sciences)
  • Polypropylene cryogenic vials, 1.2 ml (Corning, cat. no. 430658)
  • Polypropylene, conical 96‐well plates (Greiner Bio‐One, cat. no. 82050‐678)
  • Adhesive aluminum foil seal (AlumaSeal 384; Excel Scientific, cat. no. F‐384‐100)
  • 8‐channel pipettor (20‐µl capacity; various vendors)
  • Polypropylene, conical 384‐well plates (Arrayit, cat. no. MMP384)
  • NanoPrint LM60 (Arrayit) or a similar 48‐pin contact microarrayer
  • Micro Spotting Pins (Arrayit, cat. no. 946MP3)
  • Epoxysilane‐coated glass slides (Nexterion Slide E; Schott, cat. no. 1064016)
  • Gloves (powder‐free chloroprene; Microflex, cat. no. NEC‐288)
  • ZeptoFOG blocking station (Zeptosens, cat. no. 1210)
  • Glass holder and slide rack set (Wheaton, cat. no. 900303)
  • Additional reagents and equipment for preparing labeled protein ( protocol 2)

Basic Protocol 2: Preparation of Labeled Bait Protein and Multivalent Microbead Complexes

  • Cy5 monoreactive dye pack (GE Healthcare, cat. no. PA25001)
  • PBS (see recipe)
  • Bait protein: can be any extracellular protein of interest to the investigator, e.g., an orphan receptor that functions in cellular communication and regulation but whose ligand(s) are unknown
  • 5% milk in PBST (see recipe)
  • Protein A MicroBeads (Miltenyi Biotec, cat. no. 120‐000‐396)
  • End‐over‐end rotator
  • Pro‐Spin columns (Princeton Separations, cat. no. CS‐800)
  • Tabletop ultracentrifuge (Beckman Coulter, model TL‐100)
  • Nanodrop spectrophotometer (Thermo Fisher Scientific; also see unit 3.10)
  • Octet protein A sensors (ForteBio, cat. no. 18‐5010;
  • Octet biolayer inferometer (ForteBio;
  • 96‐well black polypropylene plates (Greiner Bio‐One, cat. no. 655209)

Alternate Protocol 1: Formation of Cy5‐Labeled IgG and Fc‐Fusion Bait Protein Multivalent Complexes on Protein A Microbeads

  • Human IgG (Jackson ImmunoResearch, cat. no. 009‐000‐003)

Basic Protocol 3: Protein Microarray Screening and Processing

  • Printed protein microarray slides ( protocol 1)
  • PBST (see recipe)
  • Protein A (Sigma‐Aldrich, cat. no. P7837)
  • 5% milk in PBST (see recipe)
  • PBS (see recipe)
  • Forceps
  • Hybridization station (Miltenyi Biotec, a‐Hyb or similar)
  • 50‐ml conical tubes (e.g., BD Falcon)
  • Swinging‐bucket centrifuge
  • GenePix 4000B microarray scanner (Molecular Devices) or equivalent
  • GenePix Pro 6.0 software (Molecular Devices) or equivalent data extraction software

Basic Protocol 4: Protein Microarray Data Analysis

  • Mac/PC or Unix/Linux server running R
  • Ellipse, Limma and gpscreen R packages (links provided below)
PDF or HTML at Wiley Online Library


  •   FigureFigure 27.3.1 Secretome protein library organization. (A) One box of 48 secretome protein master stock vials for storage at −20°C is shown. (B) Samples from two master stock boxes are transferred to one 96‐well stock plate as shown. (C) Samples from two 96‐well plates are transferred to one 384‐well printing plate. Note that the two 96‐well plates constitute the content of the left two quadrants of the 384‐well plate. The two right quadrants contain Cy3‐BSA. As samples are arrayed, the 48‐pin printhead should start at the top left and proceed left to right and top to bottom. Therefore, the first sample pickup will encompass rows A to D and columns 1 to 12. The second sample pickup will consist of entirely Cy3‐BSA (rows A to D and columns 13 to 24). The pattern is the same for the remainder of the plate with alternating sample and Cy3‐BSA pickup.
  •   FigureFigure 27.3.2 Protein microarray blocking. A ZeptoFog device (ultrasonic mister) used to block slides after printing is shown with 8 slides loaded on the platform surface with the lid off. An air pump (blue device) is connected to circulate the mist from the ultrasonic mister immersed in blocking solution below the perforated platform.
  •   FigureFigure 27.3.3 Multivalent bait microbead complexes. (A) Schematic model of a protein A‐microbead complex with extracellular domain (ECD)‐Fc fusion protein. Protein A is shown attached to the microbead and binding the Fc (blue) domain of the captured bait protein. (B) Schematic model representing Cy5‐labeled IgG (red) and unlabeled Fc‐fusion bait proteins (blue) co‐captured on the protein A microbeads. (C) Results from a test of different molar ratios of Fc‐fusion bait (CD200‐Fc) protein to IgG‐Cy5 mixed and complexed with protein A microbeads followed by binding to a slide with immobilized CD200R. Cy5 fluorescence signal minus the background signal is shown (F635‐B635). A 1:1 ratio provides a sufficiently strong signal.
  •   FigureFigure 27.3.4 Identification of optimal protein A microbead to Fc‐fusion protein ratios. A representative ForteBio Octet sensorgram using protein A sensors is shown. The association curves represent a titration of protein A microbeads (as indicated) against a constant amount of Fc‐fusion protein (4 µg). The minimal bead concentration where no free Fc‐fusion remains is selected as optimal (40 µl, the black curve in this case). This figure is reproduced with permission from Ramani et al. ()
  •   FigureFigure 27.3.5 Example QC plots generated by the gpscreen package. (A) Example QC plots are shown for a screen of BTLA on duplicate slides. The pixels in each plot represent spot intensities (left), background (middle), and background corrected (right) data as a function of slide location. Each slide is represented as 48 subgrids. (B) An enlargement of the plots showing background corrected data. Note that spatial effects are largely removed except for potential hits. A small amount of carryover can also be detected visually.
  •   FigureFigure 27.3.6 Statistical methodology for hit determination. (A) Histograms of scores for two array replicates with the fitted normal distribution and the 0.0001 probability cut‐off (vertical red line) indicated. Hits are represented by asterisks plotted above the x axis. (B) A representative intersection plot for hit identification is shown. The histograms for array 1 and 2 shown in (A) can be considered 1‐dimensional projections along the x and y axis of the intersection plot, respectively. The dashed, diagonal line represents equality. The horizontal and vertical lines are the individual 0.0001 probability cut‐offs. There are no array 1‐only hits. Blue circles are array 2‐only hits. Purple triangles represent hits against nonspecific binders. Black circles are intersection hits. This figure is reproduced with permission from Ramani et al. ()
  •   FigureFigure 27.3.7 Protein microarray layout and mask fitting. (A) Scanned image from a protein microarray slide (Cy3—green; Cy5—red). The yellow boxes outline the first four subgrids out of 48 that comprise the entire microarray. Several hits are shown outlined in white boxes. Enlargements of subgrids containing hits are displayed on the right. The alternating pattern of protein (blank) and Cy3‐BSA samples printed in duplicate is clearly evident. (B) A close‐up is shown demonstrating mask fitting and spot alignment for spot integration by the data extraction software.


Literature Cited

Literature Cited
   Bushell, K.M., Sollner, C., Schuster‐Boeckler, B., Bateman, A., and Wright, G.J. 2008. Large‐scale screening for novel low‐affinity extracellular protein interactions. Genome Res. 18:622‐630.
   Clark, H.F., Gurney, A.L., Abaya, E., Baker, K., Baldwin, D., Brush, J., Chen, J., Chow, B., Chui, C., Crowley, C., Currell, B., Deuel, B., Dowd, P., Eaton, D., Foster, J., Grimaldi, C., Gu, Q., Hass, P.E., Heldens, S., Huang, A., Kim, H.S., Klimowski, L., Jin, Y., Johnson, S., Lee, J., Lewis, L., Liao, D., Mark, M., Robbie, E., Sanchez, C., Schoenfeld, J., Seshagiri, S., Simmons, L., Singh, J., Smith, V., Stinson, J., Vagts, A., Vandlen, R., Wantanabe, C., Wiend, D., Woods, K., Xie, M.H., Yansura, D., Yi, S., Yu, G., Yuan, J., Zhang, M., Zhang, Z., Gaddard, A., Wood, W.I., Godowski, P., and Gray, A. 2003. The secreted protein discovery initiative (SPDI), a large‐scale effort to identify novel human secreted and transmembrane proteins: A bioinformatics assessment. Genome Res. 13:2265‐2270.
   Letarte, M., Voulgaraki, D., Hatherley, D., Foster‐Cuevas, M., Saunders, N.J., and Barclay, A.N. 2005. Analysis of leukocyte membrane protein interactions using protein microarrays. BMC Biochem. 6:2.
   Ramani, S.R., Tom, I., Lewin‐Koh, N., Wranik, B., DePalatis, L., Zhang, J., Eaton, D., and Gonzalez, L.C. 2012. A secreted protein microarray platform for extracellular protein interaction discovery. Anal. Biochem. 420:127‐138.
   Silver, J., Ritchie, M. E., and Smyth, G. K. 2009. Microarray background correction: Maximum likelihood estimation for the normal‐exponential convolution model. Biostatistics 10:352‐363.
   Smyth, G. K. 2005. Limma: Linear models for microarray data. In Bioinformatics and Computational Biology Solutions using R and Bioconductor (R. Gentleman, V. Carey, S. Dudoit, R. Irizarry, and W. Huber, eds.) pp. 397‐420. Springer, New York.
   Sun, Y., Gallagher‐Jones, M., Barker, C., and Wright, G.J. 2012. A benchmarked protein microarray‐based platform for the identification of novel low‐affinity extracellular protein interactions. Anal. Biochem. 424:45‐53.
   Voulgaraki, D., Mitnacht‐Kras, R., Letarte, M., Foster‐Cuevas, M., Brown, M.H., and Barclay, A.N. 2005. Multivalent recombinant proteins for probing functions of leukocyte surface proteins such as the CD200 receptor. Immunology 115:337‐346.
   Wright, G.J. 2009. Signal initiation I biological systems: the properties and detection of transient extracellular protein interactions. Mol. Biosyst. 5:1405‐1412.
   Wright, G.J., Martin, S., Bushell, K.M., and Sollner, C. 2010. High‐throughput identification of transient extracellular protein interactions. Biochem. Soc. Trans. 38:919‐922.
PDF or HTML at Wiley Online Library