|Home | About | Journals | Submit | Contact Us | Français|
We describe a protocol for the global identification of the in vitro substrates targeted by protein kinases using protein microarray technology. Large numbers of fusion proteins TAP-tagged at their carboxy-termini are purified in 96-well format and spotted in duplicate onto amino-silane coated slides in a spatially addressable manner. These arrays are incubated in the presence of purified kinase and radiolabeled ATP, and then washed, dried, and analyzed by autoradiography. The extent of phosphorylation of each spot is quantified and normalized, and proteins that are reproducibly phosphorylated in the presence of the active kinase relative to control slides are scored as positive substrates. This approach enables the rapid determination of kinase-substrate relationships on a proteome-wide scale, and although developed for yeast, has since been adapted to higher eukaryotic systems.
Often regarded as one of the most important posttranslational modifications, protein phosphorylation has been shown to be involved in virtually every basic cellular process. With nearly one-third of the eukaryotic proteome estimated to be phosphorylated1,2, much effort has been dedicated to unraveling the phosphoproteome, or network of kinase signaling pathways. However, methods to systematically and globally examine these networks have been lacking until as of late. In the recent years, several approaches have emerged which have enabled phosphorylation to be analyzed on a global scale. Among these is the use of protein microarrays as substrates in enzymatic assays.
Proteome microarrays are typically generated using whole genome overexpression collections. In yeast, there are two such collections: an N-terminally tagged GST collection3,4 and a C-terminally tagged TAP-tag (6xHis-HA-3C protease cleavage site-ZZ domain of protein A) collection5. Both have successfully been used to generate protein microarrays. We describe here the generation of protein microarrays using the TAP-tag collection as this collection is a second generation collection boasting greater coverage, higher quality (more thoroughly sequenced), and updates of more recently annotated or unannotated genes. Furthermore, C-terminal tags likely circumvent issues of improper targeting of proteins destined for the secretory pathway, an issue that may arise with the use of an N-terminal tag.
To generate the arrays, TAP-tag fusion proteins are first batch purified in 96-well format using IgG sepharose and eluted by cleaving the ZZ-domain from the rest of the fusion protein using 3C protease. The purified proteins are then printed in duplicate in 48 blocks on an amino-silane surface coated glass slide (Figure 1). Other slide surfaces have been tested, including nitrocellulose and nickel-coated surfaces, however, the amino-silane surface appears to offer the lowest background in radioactive assays. Once protein microarrays have been generated, radioactive kinase assays can be performed using the immobilized fusion proteins as substrates. For each kinase, three arrays are used; the assay is performed in duplicate and a negative control (no kinase) assay is included to determine any autophosphorylating proteins.
This approach allows the examination of all potential phosphorlyation reactions in a single direct assay. However, there are limitations. First, the kinase-substrate relationships are identified in vitro and may not represent bona fide in vivo relationships. Further validation is thus necessary using traditional genetic and biochemical techniques. Second, substrates may be missed (or occur) due to lack of appropriate modifications or the presence of adaptor/scaffold proteins. Finally, creating a collection of expression ORFs and developing methods to overexpress large numbers of proteins can require considerable effort. Many expression collections have nonetheless been generated to date 6–10, and improved high-throughput protein production has led to the construction of several high content arrays including Arabidopsis proteome arrays and human proteome arrays10–12. Despite these limitations, it is clear that the use of protein microarrays in enzymatic assays is readily amenable for many eukaryotes.
Below we describe protocols for protein purification and protein kinase assay using microarrays for yeast. These can be easily adapted for any organism.
The first part of the protocol describes the growth and induction of the >5,500 yeast clones. The second part describes the purification of TAP-tagged fusion proteins in 96-well format. The third part describes briefly the printing of the purified proteins on the modified glass slides for protein attachment. More detailed instruction on the actual printing is beyond the scope of this protocol. Please refer to http://robinsonlab.stanford.edu/microarrays/index.htm for documentation on planning a plate for printing, printing proteins onto a slide surface, and analyzing the quality of the printed slides. Finally, the last part of the protocol describes the in vitro kinase assay on the printed slides.
**PAUSE POINT The cell pellets can be stored by sealing the 96-deep well box with a fitted polypropylene seal and placing the box at −80°C. When ready to proceed to purification, thaw cell pellets on ice.
**PAUSE POINT Purified proteins can be snap frozen in an ethanol/dry ice slurry and stored at −80°C.
TROUBLESHOOTING? Single pins that do not touch the surface may be due to water or debris in the pin-holding block preventing the pins from going down all the way. Carefully clean pins as necessary using a cotton swab.
**PAUSE POINT Printed slides can be stored in a slide box at −20°C. (Slides at this point can also be probed using an anti-HA antibody followed by a fluorescently labeled secondary antibody to visualize protein spotting on the array.)
TROUBLESHOOTING? If the signal is weak or absent, add more kinase to the reaction or let the reaction go longer (up to six hours). If the signal is too strong and little selectivity is observed, reduce the amount of kinase in the kinase mastermix by ten-fold.
This assay produces an array of spots with varying intensities representing the extent of phosphorylation of each protein. Autoradiography films can then be scanned and the spots quantified using any number of array analysis softwares. We routinely use GenePix 2.2 to align an array file containing each protein’s ID onto our scanned array, calculate the average intensity value of duplicate spots, and normalize this intensity value to the background. The data is then filtered by subtracting out autophosphorylating proteins, as identified from the negative control assay. We found that kinases exhibit a wide range of specificities, with some kinases targeting hundreds of substrates and others targeting only a single substrate.
This work was supported by a grant from the NIH. The assays shown in Figure 1 were performed by Dr. Jason Ptacek and Dr. Geeta Devgan. We thank Dr. Rui Chen and Dr. Vincent Bruno for helpful comments.