Introduction
Mitogen activated protein kinases (MAPKs) are key factors in cellular signal transduction and are involved in a whole range of diseases, including cancer and inflammation 1. The investigation of specific MAPKs is challenging, as these enzymes are highly homologous, with highly conserved active sites. This problem becomes even more pronounced for distinguishing members within the different subfamilies. For example, the c-Jun NH2-terminal kinase (JNK) subfamily (Kallunki et
al., 1994) consists of three members, JNK1, JNK2 and JNK3, each with different splice variants. While JNK1 and JNK2 are expressed ubiquitously, JNK3 expression is mostly restricted to the brain, heart and testis 2. These kinases are responsible for the activation of cellular responses to extracellular signals (Davis, 2000). JNK-dependent signal transduction is important in a wide range of different processes, including cellular proliferation, oncogenic transformation or programmed cell death and many others. While JNK specific functions are known, assigning them to the different JNK isoforms is still difficult. For this purpose binding molecules which can reliably differentiate between these isoforms, are desired. Antibodies, the most commonly used binding molecules, are not ideal to target MAPKs, as these enzyme are located within the cell and antibodies tend not to sustain such a reducing environment, as their stability relies on disulfide bonds 3. While different successful attempts have been reported to select antibodies, which are functional inside the cell, the number of selected and characterized binders is usually limited 4; 5. These results might be explained by the selection pressure, which is not only specific binding but also stability under reducing conditions. Although considerable scientific effort has focused on the construction of stable antibody libraries, which are adapted to reducing selection conditions 6, the intracellular selection of specific antibodies still remains challenging. To overcome the limitations of antibodies and their derivatives, a variety of novel scaffolds for the generation of antibody-like binding molecules, possessing
favorable biophysical properties, has been
developed 7. Designed ankyrin repeat (AR)
proteins represent such a novel scaffold 8; 9.
These proteins are built from single repeat modules of 33 amino acids with fixed framework residues (27 amino acids) and randomized potential interaction residues (6 amino acids), which stack together to form the AR protein. This modular architecture is ideal for the assembly of very large combinatorial AR protein libraries. The library used for the selection experiments described here was built from three randomized AR modules, flanked by an N- and a C-terminal capping repeat shielding the hydrophobic core (N3C). While the theoretical diversity of such a library is 3.79 x 1023, the experimentally sampled one had a diversity of greater than 1010. As the number of modules in one designed AR protein can be adjusted 9, the target binging surface may be adapted as desired and is not fixed in dimensions as it is for other scaffolds, including antibodies.
It has been shown that the designed AR proteins match all requirements for functional intracellular applications: these molecules are very well expressed in the bacterial cytoplasm, they are stable and cysteine-free 10. Moreover, specific binders with high affinity and specificity can be selected from designed AR protein libraries 11. While the selection of specific binders from other novel scaffold libraries has also been reported 12, their intracellular applications are scarce 13. In contrast, specific binders and inhibitors of a prokaryotic kinase, which are fully functional inside the cell, have been selected from the AR protein libraries 14.
We here report and analyze the selection of JNK2-specific AR proteins. The combination of ribosome display, a complete in vitro selection technology, and protein fragment complementation assay (PCA), an in vivo selection technology, allowed the rapid selection of large pools of intracellularly active JNK2 specific binders.
Results
The goal of this project was the selection of AR proteins, which are able to discriminate between JNK2 and JNK1, to monitor the selection process and to test the functionality of the selected AR proteins in vitro as well as in vivo. For initial selections of binding molecules from large AR protein libraries, ribosome display was applied. As this complete in vitro selection technology circumvents any transformation step which would otherwise limit the applicable library size, it is ideally suited to handle large libraries (diversity t
1010). PCA, a bacterial intracellular selection
technology, based on the reconstitution of DHFR
activity by two binding partners 15, was
subsequently applied to analyze the ribosome- display generated pools of binders (Fig. 1), to further enrich these binders and to demonstrate their activity under reducing conditions. Selected AR proteins were further examined for their target binding properties with respect to affinity and specificity in ELISA and SPR experiments.
In vitro and in vivo selection of JNK2-binding AR proteins
For the selection of JNK2-specific binders from an N3C AR protein library, four standard ribosome-display selection rounds against
immobilized JNK2 were performed 16. A pool of
DNA fragments encoding AR proteins
naïve starting library and after each ribosome- display selection round was inserted into the PCA selection plasmid as mDHFRI fusion. In parallel experiments, co-transformation of these pools with the JNK2-mDHFRII-bait-plasmid allowed determining the bacterial survival under selective conditions in dependence of the respective ribosome-display selection round (Fig. 2). As expected, nearly no bacterial growth resulted from co-transformation of the bait-plasmid with the naïve AR protein library. The pool after the first round of ribosome display also gave almost no growth under selective PCA conditions. In contrast, co-transformations of the AR protein pools corresponding to ribosome-display selection rounds 2, 3, and 4 gave rise to a large number of bacterial survivors, which increased with the selection round number (Fig. 2). As a control, the individual pools were also plated under non-selective conditions, where no difference in bacterial cell growth was observed, indicating that all pools had been treated equally (data not shown). These results suggest that two rounds of ribosome display are sufficient for selection of binders.
ELISA screen to identify JNK2-specific AR proteins
To elucidate whether the increased bacterial growth, observed in the in vivo PCA selection, did indeed result from an enrichment of JNK2- specific AR proteins, the binding of the selected AR proteins to JNK2 was tested in ELISA screens. Seventeen clones from each of the selection rounds 1, 2, and 4, obtained either with or without additional PCA selection were randomly picked. Crude E. coli extracts from small-scale expression cultures were used to compare the binding of the selected clones either to JNK2 or, as a control, to the bacteriophage lambda protein D (pD) in ELISA experiments (Fig. 3).
Figure 1. Schematic representation of the selection and analysis strategy.