Tetracysteine-tagged prion protein allows discrimination between the native and converted forms Jernej Gas ˇ pers ˇ ic ˇ 1 , Iva Hafner-Bratkovic ˇ 1 , Michel Stephan 2 , Peter Veranic ˇ 3 , Mojca Benc ˇ ina 1 , Ina Vorberg 4 and Roman Jerala 1,5 1 Department of Biotechnology, National Institute of Chemistry, Ljubljana, Slovenia 2 Department of Organic and Medicinal Chemistry, National Institute of Chemistry, Ljubljana, Slovenia 3 Faculty of Medicine, Institute of Cell Biology, Ljubljana, Slovenia 4 Institute of Virology, TU Munich, Germany 5 Faculty of Chemistry and Chemical Technology, University of Ljubljana, Slovenia Introduction Prion diseases belong to a group of conformational diseases characterized by the structural conversion of native protein to alternative conformations [1]. The protein-only hypothesis states that prions are com- posed predominantly of abnormally folded prion pro- tein (PrP), the scrapie pathogenic form of PrP (PrP Sc ) [2]. This form of PrP forms amyloid, which can be detected by compounds that bind to these types of ordered protein aggregate. Molecules that bind specifi- cally to amyloids include thioflavin T (ThT) [3], Congo Keywords biarsenical; conversion; fibril; prion; tetracysteine Correspondence R. Jerala, Department of biotechnology, National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia Fax: +386 1 476 0300 Tel: +386 1 476 0335 E-mail: roman.jerala@ki.si (Received 5 October 2009, revised 19 January 2010, accepted 17 February 2010) doi:10.1111/j.1742-4658.2010.07619.x The conformational conversion of prion protein (PrP) from a native con- formation to the amyloid form is a hallmark of transmissible spongiform encephalopathies. Conversion is usually monitored by fluorescent dyes, which bind generic amyloids and are less suited for living cell imaging. We report a new method for the synthesis of membrane-permeable and mem- brane-impermeable biarsenical reagents, which are then used to monitor murine PrP (mPrP) misfolding. We introduced tetracysteine (TC) tags into three different positions of mPrP, which folded into a native-like structure. Whereas mPrPs with a TC tag inserted at the N-terminus or C-terminus supported fibril formation, insertion into the helix 2–helix 3 loop inhibited conversion. We devised a quantitative protease-free method to determine the fraction of converted PrP, based on the ability of the fluorescein arseni- cal helix binder reagent to differentiate between the monomeric and fibril- ized form of TC-tagged PrP, and showed that TC-tagged mPrP could be detected on transfected cells, thereby expanding the potential use of this method for the detection and study of conformational diseases. Structured digital abstract l MINT-7709757: Prp (uniprotkb:P04925) and Prp (uniprotkb:P04925) bind (MI:0407)byelec- tron microscopy ( MI:0040) l MINT-7709744: Prp (uniprotkb:P04925) and Prp (uniprotkb:P04925) bind (MI:0407)by circular dichroism ( MI:0016) l MINT-7709730: Prp (uniprotkb:P04925) and Prp (uniprotkb:P04925) bind (MI:0407)byfluor- escence technology ( MI:0051) Abbreviations BME, b-mercaptoethanol; CrAsH, carboxy fluorescein arsenical helix binder; EDT, ethane dithiol; FlAsH, fluorescein arsenical helix binder; GPI, glycosylphosphatidylinositol; H, helix; mPrP, murine prion protein; PrP, prion protein; PrP Sc, scrapie pathogenic form of prion protein; TC, tetracysteine; TCEP, Tris(2-carboxyethyl)phosphine; ThT, thioflavin T. 2038 FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS red [4], 2-[1-(6-[(2-fluoroethyl)(methyl)amino]-2-naph- thyl)ethylidene]malononitrile [5], curcumin [6], and luminescent conjugated polymers [7]. Of these, ThT is the most frequently used dye for detecting the forma- tion of prion fibrils in vitro [8,9]. Localization of different proteins within living cells became possible with the fusion of a fluorescent protein, e.g. green fluorescent protein, to the protein of interest [10]. Fluorescent protein domains may interfere with protein folding or ligand binding and, more particularly, with protein packing into ordered arrays. Although the green fluorescent protein fusion does not influence PrP trafficking [11], it prevents prion infection [12], similarly to the PrP–Fc fusion protein [13]. A tetracysteine (TC) tag that specifically binds biarsenic fluorescent dyes rep- resents an alternative fluorescent labeling technique. A TC tag is a short peptide motif defined by the sequence pattern CCXXCC, where C is a cysteine and X any amino acid except cysteine [14]. Fluorescein arsenical helix binder (FlAsH) is an organoarsenical compound that covalently binds to the TC tag at subnanomolar concentrations [14]. FlAsH is based on fluorescein with two arsenic (III) atoms at the 4 ¢-posi- tion and 5¢-position, and is membrane-permeable (Fig. 1). This compound by itself is nonfluorescent [14]. Rigid spacing between both arsenic atoms in FlAsH enables it to bind with high affinity and specificity to the TC motif introduced into a variety of different proteins [15]. The binding of FlAsH reagent and also the quan- tum yield of the formed FlAsH-peptide adduct depends on the conformation of the peptide backbone adopted by the tetracysteine tag [16]. Tagging a protein with a TC motif has been used for the detection of proteins of interest in vivo [17,18], in purification [19], and for detection on gel electrophoresis [15]. We decided to investigate the potential of introduc- ing TC peptide tags at different positions on PrP in order to monitor conformational changes in PrP. We show that insertion of a TC tag into the N-terminal and C-terminal segments of a protein does not inter- fere with fibril formation. Fluorescent biarsenical reagents do not label the converted forms of PrP, but rather the native and denatured forms. This property is at the core of our new quantitative fluorescent con- version assay for PrP. Results Synthesis of FlAsH and carboxy FlAsH (CrAsH) The fluorescent reagents FlAsH and CrAsH, which bind to the TC peptide tag, have been previously reported on. We modified the synthetic method described by Griffin et al. [20] (Fig. 1). The new practi- cal procedure limits the amounts of toxic reagents needed, and reduces the formation of byproducts by minimizing the number of manipulation steps during work-up. Thus, starting from 4¢,5¢-bis(trifluoroacetoxy- mercuri)fluorescein, analytically pure FlAsH-ethane dithiol (EDT) 2 was obtained in 29% yield (> 99% purity) after trituration with a minimal amount of CH 2 Cl 2 . In addition, following the simplified proce- dure and starting from an approximately 1 : 1 mixture of 4¢,5¢-bis(trifluoroacetoxymercuri)-5-carboxy-fluores- cein and 4¢,5¢-bis(trifluoroacetoxymercuri)-6-carboxy- fluorescein [21], an analytically pure mixture of 5-CrAsH-EDT 2 and 6-CrAsH-EDT 2 was obtained in 21% yield (> 99% purity). Both reagents were active and bound effectively to peptides containing the TC tag, as described below. Production, secondary structure analysis and thermal stability of the TC tag-modified protein Insertion of even a small peptide tag could affect the ability of PrP to convert to PrP Sc . We wanted to Fig. 1. Scheme of improved method of chemical synthesis of FlAsH and CrAsH reagents. NMP, N-methyl-pyrolidinone; RT, room temperature. J. Gas ˇ pers ˇ ic ˇ et al. PrP conversion monitored by biarsenical reagent FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS 2039 investigate which positions within the tertiary structure of PrP allow the introduction of a peptide tag without compromising protein folding and conversion into the fibrillar structure. We selected three positions: one at the disordered N-terminal segment of PrP (TCN), another at the C-terminal segment, which connects PrP to the glycosylphos- phatidylinositol (GPI) anchor (TCC), and a third at a position within the PrP structure between helix (H)2 and H3 that is surface-exposed, provides the geometry that would allow binding of the biarsenical reagent via the thiol groups, and requires the most conservative amino acid replacements (TCL) (Fig. 2A). All recombinant proteins were produced in Escherichia coli in the form of inclusion bodies, and successfully refolded into the native-like conformation as judged by the far-UV CD spectra. Native murine PrP (mPrP) has a high content of a-helical secondary structure, with characteristic minima at 210 nm and 222 nm in the far-UV CD spectrum. The spectra of TCC, TCN and TCL overlap the spectra of mPrP, indicating that the secondary structure is conserved in all TC-tagged mPrPs (Fig. 2B, left). The thermal stabilities of TC-tagged mPrPs determined by CD spectroscopy, when we introduced a TC tag at the C-terminus (64 ± 0.5 °C), at the N-terminus before octare- peats (67 ± 0.5 °C), or in the loop between H1 and H2 (63 ± 0.5 °C), were not substantially different from the thermal stability of mPrP alone (65 ± 0.5 °C) (Fig. 2B, right). Conformational dependence of the fluorescence of the FlAsH–TC-tagged mPrPs The fluorescence of the FlAsH reagent increased upon binding to the TC-tagged mPrP. Addition of the native PrP without a TC tag did not cause an increase in FlAsH fluorescence, demonstrating that this increase is due to specific binding to the TC tag. In prion disease, the native a-form of PrP is converted to oligomers and amyloid fibrils with a high b-sheet content [22,23]. We investigated whether the introduction of a TC tag into PrP allows discrimination between the conformational states of PrP. The fluorescence of the FlAsH–TC- tagged mPrP complex increased with increasing con- centration of urea for TCN (Fig. 3A), TCL (Fig. 3B), and TCC (Fig. 3C), suggesting that the fluorescence of the protein–FlAsH adduct is stronger when it is in the denatured conformation. This was in contrast to the measurements on the nonstructured TC-containing peptide, which retained approximately the same 121 Octarepeats TCN 23 A B 230 TCC α2 α1 α3 β3 β1 TCL Fig. 2. TC tag insertion does not signifi- cantly affect protein secondary structure or stability. (A) Structural representation of TC tags inserted into mPrP based on the solu- tion structure (Protein Data Bank code: 1XYX [62]), with the unstructured domain represented by dots. (B) Left: comparison of far-UV CD spectra of mPrP and its TC-tagged counterparts shows that the secondary structure is conserved in TC-tagged mPrPs. Right: the thermal stability of TC-tagged mPrPs demonstrates similar melting temperatures. Scans were obtained at protein concentrations of 0.1 mgÆmL )1 in MilliQ water, with a temperature scan rate of 1 °CÆmin )1 , monitored by the ellipticity at 215 nm. PrP conversion monitored by biarsenical reagent J. Gas ˇ pers ˇ ic ˇ et al. 2040 FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS fluorescence regardless of the concentration of the dena- turing agent, ruling out the solvent effect of the denatur- ing agent on the FlAsH fluorescence (Fig. 3D). Several studies have shown transient interactions between the flexible N-terminus and globular domain [24–26] and ordering of the N-terminal octarepeats above pH 6.5 [27], so the increase in fluorescence of the FlAsH–TCN adduct with increasing denaturant concentration could result from the change in the local chemical environment of the PrP. As a negative con- trol, FlAsH fluorescence was measured in the presence of increasing concentrations of denaturants with (Fig. 3E) or without (Fig. 3F) mPrP. One possible explanation for the increase in FlAsH fluorescence with increasing concentration of denaturant might be that FlAsH binds more efficiently to the denatured PrP than to the folded protein and that the higher flu- orescence occurs because of additional binding under denaturing conditions. However, we obtained the same effect when FlAsH was initially bound to the protein under denaturing conditions, and a lower final concentration of denaturing agent was obtained by dilution (Fig. 3A–C). This demonstrates that TC tags on PrP and FlAsH represent a conformation-sensitive probe. A TC tag at the N-terminus or C-terminus of PrP does not prevent conversion We investigated whether the TC tag interferes with PrP conversion. Fibril formation under mildly denatur- ing conditions [28] was monitored with the amyloid- specific dye ThT. We showed that TCC and TCN formed fibrils, whereas TCL, which also folded into a A B C D E F Fig. 3. FlAsH fluorescence of TC-tagged mPrP depends on the conformational state of the protein. The fluorescence of TCN (A), TCL (B) and TCC (C) in the presence of FlAsH was measured as a function of con- centration of urea (s). In parallel, conjugates of PrPs with FlAsH were formed in 8 M urea, and diluted to the final concentration of urea as indicated ( ). The fluorescence of FlAsH bound to the short TC peptide DDCCPGCCDD did not depend on the pres- ence of denaturant (D). Control reactions with mPrP without TC tag in the presence of FlAsH (E) and FlAsH itself (F) do not exhibit fluorescence. J. Gas ˇ pers ˇ ic ˇ et al. PrP conversion monitored by biarsenical reagent FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS 2041 native-like conformation, did not form fibrils within 4 days (Fig. 4A). mPrP, TCN and TCC changed into the b-structured conformation, whereas TCL remained in the a-monomeric conformation (Fig. 4B). TCN and TCC fibril formation were also confirmed by transmis- sion electron microscopy (Fig. 4C), which showed fibrils similar in size and morphology to those of wild- type mPrP. Detection of TC-tagged mPrPs in cell culture We introduced TC-tagged and non-TC-tagged mPrP into the PrP-deficient cell line HpL3-4. Introduced mPrPs additionally harbored mutations L108M and V111M, enabling recognition by antibody 3F4. Although both TCC (Fig. 5Ab) and TCN (Fig. 5Ac) were expressed at the cell surface, we had difficulties in obtaining specific labeling with FlAsH using previously published protocols [16,20,29–33]. However, with a modified protocol by Taguchi [34], the specific labeling of TCC (Fig. 5Bb) and TCN (Fig. 5Bc) with FlAsH was achieved. We also used the CrAsH, which is similar to FlAsH but contains an additional charged group and is not membrane-permeable, to label only surface-exposed TCC (Fig. 5Cb) and TCN (Fig. 5Cc) [35]. No surface FlAsH or CrAsH labeling was observed in HpL3-4 transduced with PrP without a TC tag (Fig. 5Ba,Ca). FlAsH fluorescence discriminates native from fibrillar mPrP As we had demonstrated that TCC and TCN have the ability to be converted into fibrils, we investigated FlAsH fluorescence in combination with the native or fibrillar form of TC-tagged PrP. FlAsH binding and its fluorescence was followed in parallel with ThT fluo- rescence during fibril formation. We found that FlAsH selectively bound to the native forms of TCN and TCC and exhibited fluorescence. In the presence of TCN and TCC fibrils, however, FlAsH exhibited no fluorescence (Fig. 6A,C). The fluorescence of ThT was inversely proportional to the fluorescence of FlAsH. When amyloid started to form, ThT fluorescence increased, and at the same time FlAsH fluorescence decreased along with the amount of monomeric PrP. In the case of TCL, which does not form fibrils, both ThT and FlAsH fluorescence remained unchanged (Fig. 6B). This phenomenon could be explained in two ways: either FlAsH does not bind to the fibrillar form of TCN and TCC, because the TC tag becomes inac- cessible to the reagent; or the cysteines of the TC tag become oxidized during fibril formation. However, the addition of a reducing agent to PrP fibrils did not result in enhanced FlAsH fluorescence intensity (data not shown), and the addition of FlAsH to nonconvert- ing TCL resulted in fluorescence under the same reac- tion conditions. Thus, we conclude that the TC tag in fibrils becomes inaccessible to FlAsH. This is also sup- ported by the finding that fibrils made from TCC pre- labeled with FlAsH under native conditions retained their fluorescence (Fig. 6D). Quantitative fluorescent PrP conversion assay We showed that TC-tagged PrP and FlAsH represent a sensor for PrP conversion. This phenomenon could A C B Fig. 4. Conversion of TC-tagged PrPs. (A) The kinetics of fibril formation of mPrP ( ), TCC ( ), TCN (•) and TCL (.) in microtiter plate assays were monitored by the fluores- cence of ThT. TCC and TCN formed amy- loid, whereas TCL did not undergo conversion. (B) CD spectra of converted PrP show that mPrP, TCC and TCN were converted to the b-form, whereas TCL remained in the a-conformation. (C) Formation of PrP fibrils of TCC, TCN and mPrP was confirmed by transmission electron microscopy. PrP conversion monitored by biarsenical reagent J. Gas ˇ pers ˇ ic ˇ et al. 2042 FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS be used to quantify PrP conversion without the need to perform proteolytic digestion of converted PrP, which is the basic principle of most conversion assays. We have devised an assay to determine the fraction of converted PrP, based on the difference in fluorescence between the monomeric and fibrillar forms of TCC. FlAsH detects only the amount of nonconverted TC-tagged PrP. In order to obtain an accurate result for the conversion efficiency, we needed to normalize the reading of the nonconverted TC-tagged PrP against the total amount of TC-tagged PrP in the sam- ple. This can be determined by solubilizing all, i.e. con- verted and nonconverted, TC-tagged PrP into the soluble denatured form by using 6 m guanidine hydro- chloride and 10 mm dithiothreitol. The denatured form was fluorescent, owing to the binding of FlAsH, and was used to determine the amount of total TC-tagged PrP from the calibration curve for the unfolded pro- tein (Fig. 7C). We showed that mixtures of different ratios of converted and native TC-tagged PrP showed a linear response over the whole range (Fig. 7D), thus providing a method for detecting TC-tagged PrP conversion without the need to employ proteolytic digestion. Discussion Different fluorescent tags have revolutionized protein science and cell biology. However, large and slow-fold- ing fluorescent proteins may influence the folding of their fusion partner, and, in the case of prion disease and other conformational diseases, they may interfere with conformational conversion, as demonstrated for PrP [11,12]. Therefore, small, genetically encoded tags may prove more useful in prion research, particularly with cell culture assays for testing prion infectivity [36,37]. We introduced a TC tag, enabling the selective incorporation of biarsenical fluorophores into different positions on PrP. Ideally, tags should not affect 0µm 25 0µm25 0µm 25 0µm25 0µm25 0µm25 A B C Fig. 5. Fluorescent biarsenical compounds specifically label TC-tagged PrPs expressed in the PrP-deficient cell line HpL3-4. (A) mPrP (a), TCC (b) or TCN (c) are expressed at the cell surface, as judged from flow cytometry analysis using 3F4 as primary antibody (white). Cells stained only with Cy2-conjugated secondary antibodies were used as controls (gray). Cells were stained with FlAsH (B) or CrAsH (C), and imaged under the confocal microscope. In cells that express TCC (Bb and Cb) and TCN (Bc and Cc), FlAsH and CrAsH selectively stain the cell surface, whereas there is no staining of the cells expressing mPrP (Ba and Ca) protein. J. Gas ˇ pers ˇ ic ˇ et al. PrP conversion monitored by biarsenical reagent FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS 2043 protein stability or trafficking, and should not interfere with prion misfolding. We avoided introducing TC tags into segments around the residues that had been previously shown to contribute to the species barrier [38] or that show a different susceptibility to prion strains, with the aim of developing a test with broad applicability. None of the three introduced TC tags substantially affected protein secondary structure. mPrPs with TC tags inserted at the N-terminus or C-terminus were able to undergo conversion, whereas introduction of a TC tag into the loop between H2 and H3 prevented fibril formation. TCN and TCC seem to undergo fibrillization even faster than mPrP, and we are currently investigating the mechanism of this phenomenon. Although antibody studies [39] and some PrP Sc models [40,41] do not predict that the seg- ment in the loop between H2 and H3 will undergo a conformational change, some human pathogenic muta- tions, such as E196K [42], indicate that this segment might influence PrP misfolding. We speculate that this segment forms contacts with other monomers in the core of the fibrils, although PrP has been previously shown to be sensitive to point mutations throughout its compact domain. Biarsenical reagents have great potential as molecu- lar sensors, particularly as improved cell-staining pro- cedures allow a high signal-to-background noise ratio. The new simplified synthetic procedure reported here may allow researchers to prepare reagents to stain either all cellular TC-tagged proteins or only a subset exposed at the cell surface. We used a modified cell- labeling method [34] to show that FlAsH and CrAsH specifically label TC-tagged PrPs. CrAsH is particu- larly suitable for following the trafficking of PrPs and other TC-tagged extracellular proteins, as it does not cross the cell membrane, owing to the presence of charged groups, and its fluorescence is more stable at physiological pH [35]. FlAsH fluorescence increased with protein denatur- ation, showing that FlAsH in the context of PrP is a conformation-sensitive probe. This is similar to what was found in a previous study, where FlAsH fluores- cence increased upon unfolding of TC-tagged cellular retinoic acid-binding protein I [16]. On the other hand, under PrP conversion conditions, FlAsH fluorescence decreased with an increase in ThT fluorescence, which marks the formation of amyloid fibrils. As the pre- sence of reducent did not reconstitute FlAsH binding to fibrils (data not shown), the best explanation is that the TC tag in fibrils of TCN and TCC is not accessible to FlAsH, and that both the N-terminal and C-termi- nal regions are buried within the formed fibrils. This is in agreement with the results of antibody-binding stud- ies [43,44]. Additionally, a GPI anchor attached to the C-terminus of PrP Sc cannot be cleaved by phospholi- pase C, indicating that this PrP segment is protected after formation of the protease-resistant form of PrP [45], which is in accordance with our results. An MS study showed that the flexible N-terminus is highly protected in fibrils in comparison to native PrP [46]. 0µm25 AB DC Fig. 6. FlAsH follows the conversion of TCC and TCN in an inversely proportional manner relative to ThT. Structural conversion of TCN (A), TCL (B) and TCC (C) was followed by ThT ( ) and FlAsH fluorescence (s). Aliquots of the fibril formation mixture were taken at the indicated times. Samples were incubated with FlAsH for 2 h at room tem- perature, with 1 m M TCEP, 1 mM BLE and 50 m M Hepes (pH 7.5) prior to fluorescence measurement. The fluorescence intensity at emission maxima (k FlAsH = 528 nm, k ThT = 474 nm) is plotted against conversion time. The FlAsH–TCC adduct was prepared prior to fibril formation, and left to undergo conversion for 48 h. Fluorescent aggregates were visualized by confocal microscopy (D). PrP conversion monitored by biarsenical reagent J. Gas ˇ pers ˇ ic ˇ et al. 2044 FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS Recently Coleman et al. [47] observed an increase in FlAsH fluorescence when it was bound to oligomeric PrP as compared with a-structured PrP. In their con- struct, a TC tag was introduced into the hydrophobic domain of PrP, N-terminally to the structured domain. This region has been shown to participate in the amy- loid core [41,48–50], so it is possible that their result was a consequence of observing b-oligomers, as the presence of mature fibrils has not been demonstrated. We used the selectivity of FlAsH fluorescence between the native and converted forms of TC-tagged PrP as the basis of a new quantitative fluorescent PrP conversion assay (Fig. 7B). Most assays for PrP conver- sion are based on the proteolytic degradation of PrP by proteinase K, which is time-consuming and, above all, has to be carefully calibrated for the amount of the active protease and the duration and reaction condi- tions of digestion, which can significantly affect the result [36,51]. In our assay, we can quantify the amount of nonconverted TC-tagged PrP by measuring the fluo- rescence after the addition of FlAsH. The total initial amount of TC-tagged PrP in the sample required for the normalization and comparison between different samples can be determined from the FlAsH fluores- cence of the solubilized and unfolded PrP (Fig. 7A). The principle of our described assay is similar to that of the conformation-dependent immunoassay [52], where one set of antibodies is used to detect the amount of 23 23 A BC D Fig. 7. Quantitative fluorescent PrP conver- sion assay. (A) Schematic illustration of the principle of the assay: under native condi- tions, only native PrP exhibits fluorescence (indicated by a star), whereas after the addition of denaturant, all forms of PrP are solubilized and exhibit FlAsH fluorescence. (B) Fluorescence emission spectra of the native and converted form of TC-tagged PrP (TCC) in the presence of FlAsH. (C) Fluores- cence of converted and nonconverted PrP in the presence of 6 M guanidine hydrochloride and 10 m M dithiothreitol. (D) Samples with different fractions of converted PrP were prepared by mixing converted and native TCC. FlAsH fluorescence was determined under native conditions, and normalized by the fluorescence under the denaturing con- ditions of each sample divided by the fluo- rescence of native PrP under the denaturing conditions, displaying a linear response. J. Gas ˇ pers ˇ ic ˇ et al. PrP conversion monitored by biarsenical reagent FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS 2045 nonconverted PrP, and the total amount of PrP, unfolded in the presence of denaturing agent, is deter- mined using the second set of antibodies. The described method could be particularly useful in research requir- ing fast results, such as in kinetic and structural studies of PrP intermediates, for in vitro conversion, such as in an amyloid seeding assay [9] and protein misfolding cyclic amplification [53,54], and potentially also in diag- nostics in cell-based assays based on the reporter cell line producing the TC-tagged PrP [55]. Experimental procedures Materials The 3F4 antibodies were purchased from Dako (Glostrup, Denmark), Cy2-conjugated anti-mouse IgG from Dianova (Hamburg, Germany), pRSET A plasmid from Invitrogen (Madison, WI, USA), the Quikchange kit from Stratagene (La Jolla, CA, USA), and Ni 2+ –nitrilotriacetic acid resin from Qiagen (Hilden, Germany). Guanidine hydrochloride and urea were purchased from Fluka (Buchs, Switzerland). All other chemicals were purchased from Sigma (St Louis, MO, USA). Preparation of TC-tagged PrP constructs TC tags were introduced into three positions on 3F4-tagged murine PrP (L108M ⁄ V111M). In TCC, CCPGCC was inserted after Ser231. Mutations in TCL were T192C, T193C, K194P, E196C, and N197C. Mutations in TCN were R37C, Y38C, Q41C, and G42C (Fig. 2A). Insertions were introduced into the 3F4-tagged mPrP ORF (from 23 to 230 amino acids) cloned into plasmid pRSET A (Invitrogen) with a Quikchange kit (Stratagene), using specific sense and complementary antisense oligonu- cleotides, as follows: for TCC sense, 5¢-GGG CGT CGT TCC AGC TGT T GT CCG GGT TGT TGT TAA GAA TTC GAA GC-3¢; for TCC antisense, 5¢-GC TTC GAA TTC TTA ACA ACA ACC CGG ACA ACA GCT GGA ACG ACG CCC-3¢; for TCN sense, 5¢-AAC ACC GGT GGA AGC TGT TGT CCT GGT GTT GTA GCC CTG GAG GCA AC-3¢; for TCN antisense, 5¢-GT TGC CTC CAG GGC TAC AAC ACC AGG ACA ACA GCT TCC ACC GGT GTT-3¢; for TCL sense, 5¢-CAC ACG GTC ACC ACC TGC TGC CCG GGG TGC TGC TTC ACC GAG ACC GAT-3¢; and for TCL antisense, 5¢-ATC GGT CTC GGT GAA GCA GCA CCC CGG GCA GCA GGT GGT GAC CGT GTG-3¢. Protein expression, purification, and refolding Plasmid pRSET A, encoding mPrP (or TC-tagged mPrP: TCN, TCL, or TCC), was transformed into competent E. coli BL21(DE3) pLysS, and mutated PrPs were expressed in the form of inclusion bodies. The protein was purified and refolded on an Ni 2+ –nitrilotriacetic acid column, using a similar procedure to one previously described [6,56]. The pur- ity of the isolated protein was checked by SDS ⁄ PAGE. Conversion to the fibrillar form of PrP Proteins were denatured overnight in 6 m guanidine hydro- chloride at 4 °C. The amyloid form of mPrP was produced by diluting denatured mPrP and TC-tagged mPrP with 1 m guanidine hydrochloride, 3 m urea and NaCl ⁄ P i (pH 6.8) at protein concentrations of 20 lm, and shaking at 37 °C [28]. The amyloid form of PrP in 96-well microtiter plates was produced by an identical protocol, and three 3 ⁄ 32 teflon balls were also added to each well for better shaking [57]. FlAsH cannot bind to the formed fibrils; however, labeled PrP fibrils can be prepared by conversion of the FlAsH-labeled PrP (TCC), as the label does not hinder con- version. FlAsH was added to native TCC at a 2 : 1 ratio, in a reaction mixture containing 50 mm Hepes, 1 mm Tris(2-carboxyethyl)phosphine (TCEP), and 1 mm b-mer- captoethanol (BLE). The reaction solution was incubated at room temperature for 2 h, protected from light. Fibrils made from FlAsH-labeled TCC were made using an identi- cal protocol to the one described above. Electron microscopy Holey formvar carbon-coated copper grids (SPI Supplies) were coated with 0.1% poly(l-lysine) and placed on one drop of the protein sample for 3 min. Samples on grids were negatively stained with 1% (w ⁄ v) aqueous uranyl ace- tate, and observed on a Jeol 100CX electron microscope operating at 80 keV. CD spectroscopy CD spectra were recorded on an Applied Photophysics Chirascan spectropolarimeter under nitrogen flow. Far-UV CD spectra for protein secondary structure determination were recorded between 190 nm and 260 nm in a 0.1 cm pathlength cuvette at a protein concentration of 0.1 mgÆmL )1 , using steps of 0.5 nm with 1 s per point. The temperature stability of proteins was recorded in a 0.1 cm pathlength cuvette (300 lL) at a protein concentra- tion of 0.1 mgÆmL )1 with a temperature scan rate of 1 °CÆmin )1 at 215 nm. Spectra were smoothed by software supplied with the instrument. Fluorescence spectroscopy For fluorescence measurements, a Perkin Elmer LS55 fluo- rimeter was used. ThT emission (460–535 nm) was tracked PrP conversion monitored by biarsenical reagent J. Gas ˇ pers ˇ ic ˇ et al. 2046 FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS by excitation at 442 nm with a protein concentration of 1 lm and 10 lm ThT. The fluorescence of FlAsH was monitored between 510 nm and 560 nm with excitation at 508 ± 5 nm in a 0.3 cm pathlength cuvette. FlAsH (5 lm) was incubated at room temperature for 2 h with 5 l m protein or peptide in the presence of 1 mm BME and 1 mm TCEP in 50 mm Hepes (pH 7.5) [16]. Conversion calculation The amount of converted PrP was calculated from the fluo- rescence intensity values of the sample after the 1 h incuba- tion period in the presence of 10 lm FlAsH, 1 mm TCEP, 1mm BME, and 50 mm Hepes (pH 7.5), and after reading of the same sample after the addition of 6 m guanidine hydrochloride, 10 mm dithiothreitol, 50 mm Hepes (pH 7.5), and 10 lm FlAsH. The amounts of native and denatured PrP were read from the calibration curve obtained with different amounts of PrP under native and denaturing conditions in the presence of FlAsH, in both cases with subtracted fluorescence of sample with FlAsH without TC-tagged prion protein. The percentage of converted PrP was calculated using the equation fraction converted ¼ðPrP tot À PrP nat Þ=PrP tot where PrP nat represents the nonconverted amount and PrP tot the total amount (converted plus nonconverted) of PrP. Construction of retroviral plasmids, production of retrovirions and transduction of the HpL3-4 cell line, and flow cytometry TC tags were introduced into the 3F4-tagged mPrP ORF in pcDNA3.1zeo(+) by Quikchange mutagenesis (Stratagene). Whereas the N-terminal TC tag was prepared using the same oligonucleotides as given above, sense (5¢ -TCC CAG GCC TAT TAC TGT TGT CCA GGA TGT TGT GAC GGG AGA AGA TCC-3¢) and antisense (5¢-GGA TCT TCT CCC GTC ACA ACA TCC TGG ACA ACA GTA ATA GGC CTG GGA-3¢) oligonucleotides were used to insert the TC tag before the GPI attachment signal. Mutant mPrP ORFs were subcloned into the retroviral expression vector pSFF [58–60]. The pSFF vectors were transfected into a coculture of packaging cell lines w2 and PA317. When cells were more than 80% positive for PrP, retroviral supernatants were harvested and cleared by cen- trifugation (120 g,4°C, 10 min). HpL3-4 cells (3 · 10 5 per well) [61] were plated into six-well microtiter plates 1 day before transduction. Cells were incubated with polybrene (4 lgÆmL )1 ) 2 h before addition of the retrovirions. One milliliter of retroviral supernatant was incubated with the cells for 2 days, after which the cells were transferred to a 6 cm or 10 cm culture plate. We used a flow cytometry protocol adapted from Maas et al. [37] to check whether TC-tagged proteins are expressed at the cell surface of HpL3-4 cells similarly to wild-type PrP. Cells (5 · 10 5 per tube) were first incubated with FACS buffer (2.5% fetal bovine serum in NaCl ⁄ P i ) for 10 min at 4 °C. One hundred microliters of 3F4 anti- body (5 lgÆmL )1 ; Dako) was added to the cells and incu- bated for 45 min at 4 °C. After washing, the cells were incubated with Cy2-conjugated anti-mouse IgG as second- ary antibodies (Dianova) for 45 min at 4 °C in the dark. Rinsed cells were analyzed by flow cytometry. Cell labeling with biarsenical compounds Cells were stained with FlAsH (CrAsH) using an adapted protocol [34]. Cells were incubated in l-Slide eight-well slides (iBidi) overnight. Prior to staining, FlAsH (CrAsH) was preincubated with 0.8 m dithiothreitol and 5 mm EDT for 5 min at room temperature. Cells were washed with HBSS and stained for 1 min with 1.3 lm FlAsH (CrAsH) and 15 mm dithiothreitol in HBSS. Unbound FlAsH (CrAsH) was rinsed off with HBSS. Cells were fixed either with 4% paraformaldehyde or ice cold ()20 °C) methanol for at least 5 min. Confocal microscopy Images were obtained on a Leica TCS SP5 laser scanning microscope mounted on a Leica DMI 6000 CS inverted microscope (Leica Microsystems, Germany) with an HCX plan apo ·63 oil (numerical aperture: 1.4) oil immersion objective. For excitation, the 514 nm line of a 25 mW argon laser was used. As laser power of 5% was used for the argon laser. Fluorescence emission was detected at 530–560 nm. Acknowledgements We would like to thank D. Oven and R. Rost for excellent technical assistance. We are grateful to A. Aguzzi for the plasmids for mPrP expression, S. A. Priola and B. Chesebro for providing the w2 and PA317 cells and the vector pSFF, and T. Onodera for the HpL3-4 cells. We would like to thank C. Taft for careful reading of the manuscript. The authors acknowledge financial support from the state budget by the Slovenian Research Agency. This project was supported by the 6th framework EU project, TSEUR. References 1 Stefani M (2004) Protein misfolding and aggregation: new examples in medicine and biology of the dark J. Gas ˇ pers ˇ ic ˇ et al. PrP conversion monitored by biarsenical reagent FEBS Journal 277 (2010) 2038–2050 ª 2010 The Authors Journal compilation ª 2010 FEBS 2047 [...]... prion protein proceeds via a molten globule intermediate J Biol Chem 282, 6300–6307 Baskakov IV, Legname G, Gryczynski Z & Prusiner SB (2004) The peculiar nature of unfolding of the human prion protein Protein Sci 13, 586–595 Donne DG, Viles JH, Groth D, Mehlhorn I, James TL, Cohen FE, Prusiner SB, Wright PE & Dyson HJ (1997) Structure of the recombinant full-length hamster prion protein PrP(29–231): the. .. epitope in the C terminus of normal prion protein whose expression is modulated by binding events in the N terminus J Mol Biol 301, 567– 573 Zahn R, Liu A, Luhrs T, Riek R, von Schroetter C, Lopez Garcia F, Billeter M, Calzolai L, Wider G & Wuthrich K (2000) NMR solution structure of the human prion protein Proc Natl Acad Sci USA 97, 145–150 Zahn R (2003) The octapeptide repeats in mammalian prion protein. .. fluorescent protein as a marker for gene expression Science 263, 802–805 Barmada S, Piccardo P, Yamaguchi K, Ghetti B & Harris DA (2004) GFP-tagged prion protein is correctly localized and functionally active in the brains of transgenic mice Neurobiol Dis 16, 527–537 Barmada SJ & Harris DA (2005) Visualization of prion infection in transgenic mice expressing green fluorescent protein- tagged prion protein. .. (2003) Soluble dimeric prion protein binds PrP(Sc) in vivo and antagonizes prion disease Cell 113, 49–60 Griffin BA, Adams SR & Tsien RY (1998) Specific covalent labeling of recombinant protein molecules inside live cells Science 281, 269–272 Adams SR, Campbell RE, Gross LA, Martin BR, Walkup GK, Yao Y, Llopis J & Tsien RY (2002) New biarsenical ligands and tetracysteine motifs for protein 2048 16 17 18... et al (2002) Localization of neurofibrillary tangles and beta-amyloid plaques in the brains of living patients with Alzheimer disease Am J Geriatr Psychiatry 10, 24–35 Hafner-Bratkovic I, Gaspersic J, Smid LM, Bresjanac M & Jerala R (2008) Curcumin binds to the alphahelical intermediate and to the amyloid form of prion protein – a new mechanism for the inhibition of PrP(Sc) accumulation J Neurochem 104,... 11 12 13 14 15 side of the protein world Biochim Biophys Acta 1739, 5–25 Prusiner SB (1998) Prions Proc Natl Acad Sci USA 95, 13363–13383 Hobbs JR & Morgan AD (1963) Fluorescence microscopy with thioflavine-T in the diagnosis of amyloid J Pathol Bacteriol 86, 437–442 Johansson GA & Pfeiffer HH (1954) On the amyloid– Congo red complex in histological sections and the genesis of the amyloid substance Acta... SR & Tsien RY (2005) Mammalian cell-based optimization of the biarsenical-binding tetracysteine motif for improved fluorescence and affinity Nat Biotechnol 23, 1308–1314 34 Taguchi Y, Shi ZD, Ruddy B, Dorward DW, Greene L & Baron GS (2008) Specific biarsenical labeling of cell surface proteins allows fluorescent- and biotin-tagging of APP and prion proteins Mol Biol Cell 24, 233–244 35 Cao H, Chen B, Squier... for mouse scrapie prions Proc Natl Acad Sci USA 100, 11666–11671 37 Maas E, Geissen M, Groschup MH, Rost R, Onodera T, Schatzl H & Vorberg IM (2007) Scrapie infection of prion protein- deficient cell line upon ectopic expression of mutant prion proteins J Biol Chem 282, 18702–18710 38 Vorberg I, Groschup MH, Pfaff E & Priola SA (2003) Multiple amino acid residues within the rabbit prion protein inhibit... K et al (2008) The POM monoclonals: a comprehensive set of antibodies to nonoverlapping prion protein epitopes PLoS ONE 3, e3872, doi:10.1371/journal.pone.0003872 Peretz D, Williamson RA, Matsunaga Y, Serban H, Pinilla C, Bastidas RB, Rozenshteyn R, James TL, Houghten RA, Cohen FE et al (1997) A conformational transition at the N terminus of the prion protein features in formation of the scrapie isoform... LL, Chesebro B & Race RE (1990) Normal and scrapieassociated forms of prion protein differ in their sensitivities to phospholipase and proteases in intact neuroblastoma cells J Virol 64, 1093–1101 Nazabal A, Hornemann S, Aguzzi A & Zenobi R (2009) Hydrogen ⁄ deuterium exchange mass spectrometry identifies two highly protected regions in recombinant full-length prion protein amyloid fibrils J Mass Spectrom . Tetracysteine-tagged prion protein allows discrimination between the native and converted forms Jernej Gas ˇ pers ˇ ic ˇ 1 ,. fluorescent proteins may influence the folding of their fusion partner, and, in the case of prion disease and other conformational diseases, they may interfere with