1 Supporting Information High-throughput instant quantification ofprotein expression and puritybased on photoactive yellow protein turn off/on label (POOL) Youngmin Kim1,2, PrabhakarGanesan1,2 and Hyotcherl Ihee1,2* 1CenterforNanomaterials and Chemical Reactions, InstituteforBasic Science, Daejeon, 305-701, Republicof Korea 2Department of Chemistry, KAIST, Daejeon, 305-701, Republicof Korea Materials and Methods POOL-calmodulin cloning The PYP gene was subcloned in the pQE80L vector using EZ-Cloning (enzynomics™). Initially the PYP gene with BamHI and SalI restriction enzyme sites was amplified using PCR. Theprimers contained the BamHIand SalIrestriction enzyme sites areas follows. 5’-CACCATCACGGATCCATGGAACACGTAGCCTTCGG-3’ 5’-TGGCTGCAGGTCGACGACGCGCTTGACGAAGACCC-3’ In this PCR step, the stop codon of the end of the PYP genewas removed. The amplified PCR product with BamHI and SalI restriction enzyme sites was inserted into the pQE80L vector. Then, the calmodulin sequence from human cDNA with a thrombin protease recognition site was amplified byPCR with thefollowingprimers. 5’-AAGCGCGTCGTCGACGGCAGCCTGGTGCCGCGCGGCAGCATGGCTGATCA 2 GCTGACCGA-3’ 5’-TGGCTGCAGGTCGACCTATTTTGCAGTCATCATCTGTACGAA -3’ The amplified calmodulin gene was inserted into the pQE80L-PYP plasmid at the SalI restriction enzyme site using EZ cloning kit. The same approach was used to make a construct with adifferent vector (the pQE30-PYP-calmodulin construct) except that the SalI restriction enzyme sitewas substituted with the BglIIrestriction enzymesite. Calmodulin cloning ThePCR amplified calmodulin genewith the thrombin proteaserecognition sitewas inserted into the pQE80L vector. The primers for the calmodulin gene containing BamHI and SalI restriction enzyme sites areas follows. 5’-AAGCGCGTCGTCGACGGCAGCCTGGTGCCGCGCGGCAGCATGGCTGATCA GCTGACCGA-3’ 5’-TGGCTGCAGGTCGACCTATTTTGCAGTCATCATCTGTACGAA-3’ Theamplified thrombin calmodulin genewas subcloned into the pQE80L vector usingEZ- Cloningkit (EnzynomicsTM). Thesameapproach was also used to make aconstruct with a different vector (pQE30) except that the SalIrestriction enzyme sitewas substituted with the BglIIrestriction enzymesite. POOL-horsemyoglobin cloning Weperformed three-fragments cloningusingEZ-Cloning kit (EnzynomicsTM). ThepQE30- PYP-calmodulin was cut by both BamHI and BglII to make a linear pQE30 vector (first fragment). ThePYP gene (second fragment)was amplified without the stop codon usingPCR 3 with thefollowingprimers containingtheBamHIrestriction enzyme site. 5’-CACCATCACGGATCCGATGACGATGACAAAATGGAACACGTAGCCTTCGG-3’ 5’-GCTGCCGCGCGGCACCAGGACGCGCTTGACGAAGACCC-3’ Thereverseprimer contains the thrombin protease recognition site. The horse Mb gene (third fragment) was amplified by PCR with the following primers containing the BglII restriction enzyme recognition site. The forward primer contains the thrombin proteaserecognition site at the 5’region to overlap with theamplified PYP gene. 5’-CTGGTGCCGCGCGGCAGCGGCCTGAGCGATGGCGAATG-3’ 5’-CTCTTGATCAGATCTTTAGCCCTGAAAGCCCAGTTCTTTA-3’ SUMO-POOL cloning First, the SUMO gene was cloned in the pQE80 vector. The kpnI restriction enzyme recognition sitewas inserted next to the SUMOgeneusingsite-directed mutagenesis with the followingprimers. 5'-GGTACCGTCGACCTGCAGCCAAGCTTAATTAG-3’ 5'-CCACCAATCTGTTCTCTGTGAGCCTCAATAATATCG-3’ Then, the plasmid was treated with KpnIrestriction enzyme. ThePYP genewas amplified by PCR with thefollowingprimers. 5'-GAACAGATTGGTGGGATGGAACACGTAGCCTTCGG-3’ 5'-GCAGGTCGACGGTACCTAGACGCGCTTGACGAAGAC-3’ Finally, byusingEZ-cloningkit, the amplified PYP genewas subcloned to the vector treated with kpnIrestriction enzyme. 4 MerP-POOL cloning The MerP was inserted in front of the PYP gene by using EZchange™ Site-directed Mutagenesis kit (EnzynomicsTM)with thefollowingprimers. 5’-ACCCTGGCGGTGCCGGGCATGACCTGCGCGGCGTGCCCGATTACCGTGAAAA AAGCGGCGATGGAACACGTAGCCTTCGGTAGC-3’ 5’-TTTGTCATCGTCATCGGATCCGTG-3’ Protein expression, purification and instant quantification of the target protein Thegeneconstructs were transformed into the E. Coli BL21 (DE3)cell line. Thebacterial cultures weregrown at 37 °C with ampicillin and vigorous shaking until amid-exponential phase (OD 0.6 at 600 nm) was reached. Then in some cases temperature was lowered to 18 °C. Theexpression was induced with 1 mM or 0.2 mM IPTG. Cultures weregrown foran additional 18 h and wereharvested bycentrifugation at 6000 gfor10 minutes. Thepellet was sonicated in lysis buffer (50 mM NaPO4 buffer pH 7.4 with 0.1 M NaCl).Thecharacteristic yellow color of PYP is absent in thisstage, but can beturned on bybindingits chromophore, p-coumaric acid. Theexpression of the POOL labeled target protein was checked byadding the chromophore precursor to the cell lysate and inspecting the color change. The concentration of theexpressed target protein can bequantified bycomparingits yellow color with those of the standard referencesolutions of known concentrations. Thecell lysate was centrifuged at 15000 gfor 1 hour and the supernatant was applied to anickel affinitycolumn that was previouslyequilibrated with thelysis buffer. The protein bound to the nickel affinity column was additionally washed with a lysis buffer containing 40 mM imidazole and was subsequently eluted using a lysis buffer containing 200 mM imidazole. After dialyzing the 5 eluted protein with 20 mM NaPO4 buffer pH 7.4, the dialyzed protein was applied to ion exchangecolumn (HiTrap, GEHealthcareLifeSciences). Then the protein was eluted in a gradient method by slowly increasing the concentration of NaCl. The amount of the target protein was checked in all the eluted fractions by comparing the color with the standard referencesolutions of known concentrations. Purity test of thetarget protein forfraction selection afterpurification work The yellow colored fractions eluted from the ion-exchange column were further used to measure the concentration and purity. By using a UV-VIS spectrometer or a microplate absorption reader after transferringthe fractions to anew 24 well plates, theUV-VIS spectra of the selected fractions weremeasured to assess the exact amount of the concentration and puritywhich can bedetermined byusingEquations (1)and (2), respectively. Engineering mini-photoactiveyellow protein We used the pQE80L-SUMO-POOL plasmid for engineering the mini-photoactive yellow protein. Thedeletion mutagenesis was performed byEZchange™ Site-directed Mutagenesis kit (EnzynomicsTM). We followed the procedures according to the EZchange™ Site-directed Mutagenesis kit manual. Wemadeatotal of 12 deletion mutants; eight types of N-terminal deletion mutants (deletion of residues 1~5 (N5), deletion of residues 1~10 (N10), deletion of residues 1~18 (N18), deletion of residues 1~27 (N27), deletion of residues 1~31 (N31), deletion of residues 1~40 (N40), deletion of residues 1~42 (N42), and deletion of residues 1~46 (N46)) and three types of C-terminal deletion mutants (deletion of residues 73~125 (C53), deletion of residues 107~125 (C18), and deletion of residues 117~125 (C8)). In 6 addition, a mutant containing both N-terminal deletion (deletion of residues 1~27) and C- terminal deletion (deletion of residues 107~125)was made Theprimers used formakingthe deletion mutants areas follows. Primer name Sequence Reverse primer for all N- terminal deletionmutant 5’-CCCACCAATCTGTTCTCTGTGAGCCTCAATAATATC Forward primer for deletionof residues 1~5 5’-TTCGGTAGCGAGGACATCGAGAACACCC Forward primer for deletionof residues 1~10 5’-ATCGAGAACACCCTCGCCAAGATGGAC Forward primer for deletionof residues 1~18 5’-GACGACGGCCAGCTCGACGGC Forward primer for deletinof residues 1~27 5’-TCCGGCGCCATCCAGCTCGACG Forward primer for deletionof residues 1~31 5’-CAGCTCGACGGCGACGGCAACATC Forward primer for deletionof residues 1~40 5’-CAGTACAACGCCGCGCAGGGCG Forward primer for deletionof residues 1~42 5’-AACGCCGCGCAGGGCGACAT Forward primer for deletionof residues 1~46 5’-GGCGACATCACCGGCCGCGA Forward primer for deletionof residues 73~125 5’-GTCTAGGTACCGTCGACCTGCAGCCAAG Reverse primer for deletionof residues 73~125 5’-GCTGTCAGTGCACGGGGCCACG Forward primer for deletionof residues 107~125 5’-GTCTAGGTACCGTCGACCTGCAGCCAAG Reverse primer for deletionof residues 107~125 5’-CTTCACCTTCGTGGGCGTCATTTGGTAAT Forward primer for deletionof residues 118~125 5’-TAGGTACCGTCGACCTGCAGCCAAGC Reverse primer for deletionof residues 118~125 5’-GCTGTCGCCGGAGAGGGCCTTC 7 Cloning forapplication of POOLto eukaryotic(mammalian and insect)cells All cells used in this studywereobtained from the American TypeCultureCollection (ATCC, Manassas, VA, USA). The signal sequence and extracellular domain of human CD40 receptor (1M ~ 277Q) was cloned together with POOL into the pcDNA3.1 vector. The CD40-negative HEK 293 cells were transfected with the plasmid using PEI reagent with 1:2.5 ratio between PEIand DNA. TheHEK 293 cells for coexpressingCD40 receptor (1M ~ 277Q) were cultured in DMEM containing 5% FBS, L-glutamine, penicillin, and streptomycin (Wisent). Aftertransfection, weincubated the cells for5 days. Then the CD40 receptor was expressed. Thecells weresubsequentlyharvested and sonicated. The truncated TLR3 (toll-like receptor 3 without transmembranedomain) with POOLwere cloned into pIEX3 (EMD Biosciences, San Diego) for transient expression in insect cells. TheSpodopterafrugiperda(Sf21)insect cells weretransfected with the plasmid. Theprotein was produced by incubating Sf21 cells for 1 hour in modified Excel 401 medium (JRH Biosciences, Lenexa, KS) supplemented with 10% dialyzed, heat-inactivated fetal bovine serum (Life Technologies, Grand Island, NY) and 50 μg/ml gentamicin. The cells were subsequentlyharvested and sonicated. Recipeformaking a precursorsolution of chromophore(p-coumaricanhydride) Theprecursor, p-coumaric anhydride, of the chromophore can be synthesized based on the methodology developed by Imamoto et al. [1]. First, 1 mM DCC (N, N'- Dicyclohexylcarbodiimide) solution in DMF(Dimethylformamide) (2.7 gin 50 ml) and 0.77 mM p-coumaric acid in DMF(2.55 gin 50 ml) areprepared bydissolvingeach into DMF and stirring the resulting solution at 4 °C. After complete dissolution of each substance in 8 DMF, two solutions aremixed andstirred at 4 °Covernight. Thesolution can bescaled up if necessary, but the molar ratio of DCC versus p-coumaric acid has to be maintained at 1.3:1. At the end of this reaction, some white precipitation in light yellow solution appears. The white precipitation is removed by centrifugation at 12,000 rpm, 4 °C. The supernatant is storedat -70 °C. Typicallyforefficient bindingof the chromophoreto all the PYP molecules, an excess amount of the chromophoreprecursoris used. Weusuallyuse10 ml choromophore solution for 2 liter PYPculture. Comparison of POOLand GFPtag ThePOOL taghas the followingadvantages and disadvantages compared with the GFPtag. Theadvantages of POOL areas follows. First, the most important aspect of the POOL tagis that the visible yellow color is initiallyoff but can beturned on byadding the chromophore. This turn-on property is in stark contrast to the GFP tag which usually comes with its chromophore(generated bythe reaction of inherent amino acid residues)alreadypresent and thus its color is already in the on-state. Therefore the color cannot be easily distinguished from other potential sources of color such as contamination and the inherent color of the target protein, which essentiallymakes it difficult to usethe GFPtagforquantitative purpose. In contrast, in the case of POOL, the yellow color is turned on only when we add the chromophoreand thus can beeasilydistinguished from such otherpotential sources of color. For example, this visual colorimetricinspection scheme works even when the target protein is myoglobin, which has an inherent strong red color. For this reason, the POOL tag can be easilyused to measuretheconcentration and purity. Second, POOL allows the expression of the target protein to be checked even without using any equipment. GFP requires 9 fluorescencedetection and it is well known that the GFPsuffers from the bleachingproblem. Finally, the POOL tag is much smaller and more soluble than the GFPtag, and thus we can expect that the former interferewith theexpression of the target protein compared with the latter.Oneof disadvantages of the POOLtag compared with theGFPtagis that the cells need to be disrupted for clear color detection whereas the fluorescence of the GFP tag can be detected without cell disruption. Reference 1. Imamoto, Y., et al., Reconstitution photoactive yellow protein from apoprotein and p- coumaricacid derivatives. FEBS Lett, 1995. 374(2): p.157-60. 10 FigureS1. POOL expression vectorincluding thePYPgeneand a target protein gene. Theorder of genes in thePOOLexpression system shown in the left is as follows: poly- histidinetag, PYP, proteaserecognition site and target protein. Reverseorder arrangement of genes in thePOOLsystem is also feasibleas shown in the right. ThepQE80, pQE30 and pET15b vectors wereused as the bases for designingthe construct with thrombin, HRV C3 protease, TEV orenterokinaseas the proteaserecognition site. 11 FigureS2. Theobserved differences in theexpression rateof soluble human calmodulin with orwithout POOL. Thesamples loaded into all wells werethe same amount (total 1ug). Thefigureshows PAGE results ofthe cell lysate of calmodulin expressed in18 C. Lines 1, 3, 5 and 7 correspond to human calmodulin with POOL. Lines 2, 4, 6 and 8 correspond to human calmodulin without POOL. The band thickness can be related to the calmodulin expression rate when the size of the POOL is taken into account. The expression rate of calmodulin with POOL is eitherhigher than or thesame as that of calmodulin withoutPOOL. Theseresults indicate that the POOL system does not interferewith the expression of soluble target proteins and sometimes even improves the expression rate. 12 Figure S3. For eluted fractions during the purification steps, the visual colorimetric detection, PAGE gel and UV-VIS spectra arecompared. One can seethat the purityof the target protein (in this case, SUMOprotein) increases with each purification step. Thebrighter yellow color of the eluted fraction indicates higher purity of the target protein, which is reflected in the PAGE gel as well as the magnitudeof absorption peak at 460 nm in the UV- VIS spectra. The results indicate that the POOL system can measure the purity and concentration of the target protein in aneasierand simplermanner than byPAGE. 13 FigureS4. ApplyingPOOL to the insect and mammalian cells. Theexpression of TLR3 and CD40 fused with POOL can beeasilychecked byaddingaprecursorof chromophorein the insect cell (A) and mammalian 293 cell (B), respectively. The left and right bottles correspond to beforeand after addingaprecursorof chromophore.