Scanning the human proteome for calmodulin-binding proteins

Shen et al. 10.1073/pnas.0407928102.

Supporting Information

Files in this Data Supplement:

Supporting Materials and Methods
Supporting Sequence Information
Supporting Figure 5
Supporting Figure 6
Supporting Figure 7
Supporting Figure 8
Supporting Figure 9
Supporting Figure 10
Supporting Figure 11
Supporting Figure 12
Supporting Figure 13
Supporting Figure 14
Supporting Figure 15
Supporting Figure 16
Supporting Table 3




Supporting Figure 5

Fig. 5. Elution profile of the first two rounds of calmodulin (CaM)-binding selection. The percentage eluted is relative to the input of [³⁵S]methionine-labeled mRNA-protein fusion molecules. Fraction 1, flowthrough; fractions 2-7, wash in the presence of Ca²⁺; fractions 8-12, elution with EGTA. Fractions from round 1 are shown in yellow; fractions from round 2 are shown in red.

Supporting Figure 6

Fig. 6. In vitro binding analysis of selected proteins with biotin-CaM. Protein fragments were generated by a transcription<949>translation (TNT) reaction. An aliquot of the TNT mixture was incubated with an appropriate amount of biotinylated CaM in the presence of 1 mM CaCl₂, followed by mixing with streptavidin-agarose beads. The beads were washed, and the bound molecules were eluted by using a buffer containing 2 mM EGTA. Aliquots of the TNT expressed protein (C), the last wash (W), and the eluent (E) were loaded onto an SDS/PAGE gel for binding analysis. The names of the proteins are listed in Table 1 according to their IDs. The amount of fraction C loaded was not the same as that loaded in fraction E, but they were proportional. To put the results of all samples together in one figure, the wash fraction is not shown, and different proteins are not aligned according to their molecular weights.

Supporting Figure 7

Fig. 7. Alignment of 14 selected protein fragments of nonerythrocytic a spectrin (SPTAN1, a -fodrin). The parental protein sequence (AAH53521) is shown below from residue 1061E to residue 1320D, highlighted in pink. The previously reported Ca²⁺/CaM binding motif is marked in red. The shortest overlapping region is underlined.

Supporting Figure 8

Fig. 8. Alignment of six selected protein fragments of Ca²⁺/CaM-kinase II a (CAMKII_A). The parental protein sequence (NP_057065) is shown below from residue 261T to residue 373S, highlighted in pink. P1D_E01 is from transcript variant NP_057065 (with the KRKSSSSVQLM sequence) and the other five fragments are from transcript variant NP_741960 (without the KRKSSSSVQLM sequence). The previously reported Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 9

Fig. 9. Alignment of three selected protein fragments of the TBC1 domain family, member 4 (TBC1D4). The parental protein sequence (XP_375032) is shown below from residue 1053S to residue 1215G, highlighted in pink. This is a previously uncharacterized Ca²⁺/CaM-binding protein and the predicted Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 10

Fig. 10. Alignment of 6 selected protein fragments of calcineurin A-a (CALNA_A). The parental protein sequence (NP_000935) is shown below from residue 335N to residue 521Q, highlighted in pink. P1C_F03, P1E_G09, and P1D_E03 are from transcript variant AAH25714 (without the TVEAIEADEA sequence) and the other three fragments are from transcript variant NP_000935 (with the TVEAIEADEA sequence). The previously reported Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 11

Fig. 11. Alignment of five selected protein fragments of calcineurin A-b (CALNA_B). The parental protein sequence (NP_066955) is shown below from residue 346S to residue 524Q, highlighted in pink. P1A_H06 is from transcript variant CAD32694 (without the TVEAIEAEKA sequence) and the other four fragments are from transcript variant NP_066955 (with the TVEAIEAEKA sequence). The previously reported Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 12

Fig. 12. Alignment of one selected protein fragment of calcineurin A-g (CALNA_C) with the parental protein sequence (NP_005596), shown below from residue 360M to residue 477M, highlighted in pink. The previously reported Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 13

Fig. 13. Alignment of two selected protein fragments from heat shock protein 90-a (HSP90_A) with the parental protein sequence (NP_005339), shown below from residue 417V to residue 625M, highlighted in pink. The previously reported Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 14

Fig. 14. Alignment of one selected protein fragment of heat shock protein 90-b (HSP90_B) with the parental protein sequence (AAH09206), shown below from residue 300L to residue 455D, highlighted in pink. The previously reported Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 15

Fig. 15. Alignment of one selected protein fragment of DEAD/H (Asp-Glu-Ala-Asp/His) box polypeptide 5 (DDX5, or p68 RNA helicase) with the parental protein sequence (NP_004387), shown below from residue 216R to residue 346V, highlighted in pink. This protein is a known Ca²⁺/CaM-binding protein, but the reported Ca²⁺/CaM-binding motif is different from the one we identified. Our predicted Ca²⁺/CaM binding motif is marked in red.

Supporting Figure 16

Fig. 16. Alignment of one selected protein fragment of proteasome 26S subunit, non-ATPase, 12 (PSMD12) with the parental protein sequence (NP_0028077), shown below from residue 78K to residue 187L, highlighted in pink. This Ca²⁺/CaM-binding protein is previously uncharacterized. The predicted Ca²⁺/CaM binding motif is marked in red.

 

Supporting Materials and Methods

Construction of cDNA Library. Poly(A)⁺ mRNAs from human brain, heart, spleen, thymus, and muscle (Stratagene) were mixed as a pool. Three rounds of oligo(dT) purification and a stringent DNase treatment were applied to minimize the presence of ribosomal RNA and genomic DNA. First-strand cDNA synthesis was performed by reverse transcription by using a degenerate primer TTN₆ (Novagen), which allowed nonbiased coverage of all genes and regions of the ORFs. Second-strand synthesis was performed by DNA polymerase I in the presence of RNase H. 5-methyl dCTP was used for both first- and second-strand synthesis to protect ORF regions from subsequent restriction digestion. The resulting cDNA was then treated with T4 DNA polymerase to generate blunt ends. The directional EcoRI<949>HindIII linker (GCTTGAATTCAAGC, Novagen) was ligated to both ends of the double-stranded DNA (dsDNA) and then restricted with EcoRI and HindIII. The ligated dsDNA has different restriction sites at its 5' and 3' ends because of the presence of TT at only the 3' end. This approach allowed the introduction of different left and right arms by restriction followed by ligation. The left arm contains a T7 promoter and a deletion mutant of the TMV 5'-UTR for efficient in vitro transcription and translation, respectively. The right arm contains a short sequence for hybridizing with the puromycin-containing oligo linker (23, 25). Sequences that code for an E-tag in the left arm and a FLAG/His_×6 tag in the right arm were also included to facilitate the purification of the mRNA-displayed proteome library. The ligated dsDNA was PCR-amplified, followed by fractionation on an agarose gel to generate a cDNA library with length distribution in the range of 0.5-2 kb.

Generation of an mRNA-Displayed Proteome Library. The dsDNA library was in vitro transcribed with T7 RNA polymerase (Ambion, Austin, TX) to generate mRNAs. Puromycin was added to the 3' end by psoralen-mediated crosslinking to the oligonucleotide psoralen-(ATAGCCGGTG)2'-OMe-dA₁₅-CC-puromycin (1). In vitro translation was performed by using rabbit reticulocyte lysate (Novagen), and mRNA-protein fusion formation was accomplished under optimized conditions in ref. 2. RNA templates and RNA-protein fusions were then purified from the lysate by using an oligo(dT) column, taking advantage of oligo(dA) residues in the puromycin-containing DNA linker. To remove RNA secondary structures that may interfere with the selection step, the fusion molecules were converted to DNA/RNA hybrids by reverse transcription. The resulting mRNA-displayed proteome library was then successively purified on the basis of each of the affinity tags at the N and C termini.

Biotinylation of Calmodulin. Approximately 2.25 mg of bovine brain calmodulin (Calbiochem) was dissolved in 0.9 ml of PBS buffer with 1 mM CaCl₂. Two milligrams of NHS-PEO₄-Biotin (Pierce) was dissolved in 200 m l of deionized water immediately before use. The CaM solution was mixed with 60 m l of NHS-PEO₄-Biotin solution and incubated on ice for 2 h. To minimize possible interference with the function of CaM due to overbiotinylation, the number of moles of biotin per mole of CaM was carefully controlled at » 1:1. The reaction was stopped by adding 100 m l of 1 M Tris·HCl (pH 8.0), followed by incubation for 20 min at room temperature. Unreacted biotinylation reagent was removed by passing the mixture through a NAP-10 column (Amersham Pharmacia). Biotinylated CaM was eluted by using 1.5 ml of PBS buffer and stored at 4°C until ready for use. The number of moles of biotin per mole of CaM was measured by using the avidin-HABA reagent (Pierce) according to the manufacturer’s instructions.

1. Kurz, M., Gu, K. & Lohse, P. A. (2000) Nucleic Acids Res. 28, E83.

2. Liu, R., Barrick, J. E., Szostak, J. W. & Roberts, R. W. (2000) Methods Enzymol. 318, 268-293.

 

Supporting Sequence Information

Sequences Isolated from the Selection and Mapping of Minimal Binding Regions. Because of the large amount of data, only 10 examples are listed here. The shortest overlapping region is underlined, and the predicted or previously reported Ca²⁺/CaM binding motif is marked in red. If the protein is a previously uncharacterized Ca²⁺/CaM-binding protein, the CaM-binding motif was predicted by inputting the shortest overlapping region to a Web-based program provided from Mitsu Ikura’s laboratory at the Ontario Cancer Institute (1).

The complete sequence information for all fragments isolated from each gene and the mapped shortest overlap region is available upon request.

1. Yap, K. L., Kim, J., Truong, K., Sherman, M., Yuan, T. & Ikura, M. (2000) J. Struct. Funct. Genomics 1, 8-14.

 

Table 3. Known CaM-binding proteins isolated from the first round of selection

 

• Early Edition
• Archives
• Online Submission

• Feature Articles

  (Expanded research articles of exceptional breadth)

• Commentaries

  (Expert commentary on leading research)

• Letters

  (Online-only letters to the editor)

• Inaugural Articles

  (Articles by recently elected NAS members)

• PNAS Profiles

  (Biographical profiles of Academy members)

• Sustainability Science

  (Interactions of human and environmental systems)

• Collected Papers

  (Editorials, Perspectives, Reviews, Colloquia, etc.)

• Special Features

  (Solicited articles on exceptional research topics)

• Sackler Colloquia

  (Scientific reports held under NAS auspices)