WO2008145609A1 - Method of making covalent conjugates with his-tagged proteins - Google Patents

Abstract

The present invention refers to novel agents for covalently conjugating His-tagged proteins, to covalent conjugates of His-tagged proteins and to methods of making such conjugates.

Description

Method of Making Covalent Conjugates with His-tagged Proteins

Description

The present invention refers to novel agents for covalently conjugating His- tagged proteins, to covalent conjugates of His-tagged proteins and to methods of making such conjugates.

Background of the Invention

During the last 20 years an increasing repertoire of fluorescence detection techniques was developed to study protein function and interactions in vitro, in cellular systems and in vivo. Many recent detection techniques function at the single molecule and single cell level. A prerequisite to use all these techniques are generic fluorescence labeling technologies for biological systems. Fluorescent probes allow detection of molecular interactions, mobility and conformational changes of proteins with high temporal and spatial resolution by using fluorescence spectroscopy such as steady-state, time-resolved, anisotropy, correlation spectroscopy (FCS), fluorescence resonance energy transfer (FRET), single-molecule detection, lifetime imaging microscopy and surface-sensitive fluorescence detection (for review of spectroscopic techniques see (Hovius, et al 2000)).

The detection technology and instrumental development was accompanied by strong efforts to increase the number and stability properties of fluorescent probes. However, techniques for site-specific labeling of proteins with small synthetic molecules, although important tools for the development of functional assays, screening approaches, mechanistic molecular studies and for investigating protein interaction networks, were scarce. Of the variety of labeling strategies available, only a few are applicable to living cells. The possibility to fluorescently label a cell surface receptor with a laser stable small dye, in a color and site-specific way, is the prerequisite for experimental investigation of receptor mechanisms at the single molecule and single cell level. It might also open new opportunities for cell-based affinity selection screening and medium throughput screening under controlled conformational constraints. Ideally, chemical tools should be easily accessible, not toxic, and should interact with very high affinity with the target protein, both in vitro and on or in a cell.

In the following, a brief summary of some of the most important site-specific protein labeling methods described in the literature is presented. These techniques comprise genetic modification of proteins by the introduction of modified unnatural amino acids allowing site-specific labeling or the incorporation of encoded dyes, such as the green fluorescent protein (GFP) (Tsien 1998) and other autofluorescent proteins (YFP, CFP, etc) (Harms, et al 2001 ). Other methods involve post-translational modification of proteins by using intein-mediated ligation strategies (Wood, et al 2004) or enzymatic reaction to specifically introduce labels into proteins (e.g. using the acyl carrier protein (ACP) (Meyer, et al 2006), or the DNA repairing enzyme ((f- alkylguanine-DNA alkyltransferase, hAGT) (Keppler, et al 2003), known as SNAPtag). A third strategy, the affinity labeling, makes use of the strong interactions between a ligand and its receptor, for instance the biotin- streptavidin system (Cronan 1990) or the haloalkane dehalogenase protein and its ligand known as Halo-tag (Zhang, et al 2006). Finally, metal-ligand interactions are also used for selective labeling of proteins. Most prominently, the FIAsH-tag (biarsenical fluorescein dye) is based on the high affinity of tetracysteine motifs for arsenic (Griffin, Adams and Tsien 1998; Adams, et al 2002).

A cheap, stable and simply applicable His-tag binding molecule with a free choice of dye for confocal fluctuation analysis applications would be of interest for a variety of research projects including imaging, assay development for high throughput screening (HTS), target identification or analysis and optimization of protein expression. In 1975 (Porath, et al 1975) immobilized metal affinity chromatography (IMAC) was developed to purify proteins using the interaction between a metal and two chelating ligands, iminodiacetic acid (IDA) and histidines present in proteins. However, its application was limited due to the low affinity of the IDA-Ni²⁺/protein system, the IDA-metal complex leaving only one free coordination site for binding the protein. Higher stability was gained by designing nitrilo triacetate (NTA). The NTA occupies four of the six ligand binding sites in the octahedral coordination sphere of a Ni²⁺ ion and therefore leaves two sites for selective protein interactions via the oligohistidine. Since that time, this improved ligand has been widely used for metal-chelate chromatography (Hochuli, Dobeli and Schacher 1987), especially for the purification of histidine-tagged recombinant proteins.

This methodology was recently used for reversible labeling of (HiS)₆ tagged proteins on living cells (Guignet, Hovius and Vogel 2004) via metal (Ni²⁺) chelation, of the NTA moiety bearing a fluorophore and the histidine sequence. In order to increase the stability of the system and having in mind that a hexahistidine sequence potentially binds three NTAs, a multivalent chelator head carrying three NTA moieties was designed (Lata, et al 2005; Lata, et al 2006). These probes enable site-specific and reversible tagging of

(His)β proteins.

Despite high affinities and stabilities of a ths-NTA/(His)6 complex, the determination of quantitative biophysical interaction constants needs absolutely stable biological systems. For resolving cellular mechanisms as function of time and concentration of target receptors, changes of local ratios between particle numbers and number of dye molecules due to complex dissociation upon dilution and washing steps need to be avoided.

WO 01/72458 describes a heterofunctional crosslinking agent which may be covalently conjugated to proteins, including to His-tag proteins. The method of making such crosslinking agents described therein, however, suffers from drawbacks such as an incomplete description of synthetic procedures to synthesize the claimed tagging molecules. Furthermore, no protocol for an application of named tagging molecules for in-vitro labeling of proteins and for labeling of proteins on living cells is provided.

Thus, it was an object of the present invention to provide novel agents for covalent conjugation to proteins and to methods for making such agents, wherein the prior art drawbacks are at least partially avoided.

Description of the Invention

The present invention describes the synthesis of new agents for irreversible site-specific labeling of His-tagged proteins using the directing effect of the nitrilo triacetate (NTA) moiety towards histidine, in combination with covalent linkage groups, e.g. photoreactive functionalities such azides or electrophilic groups such as acryloyl groups. The new prototype compounds were shown to represent irreversible labeling reagents for His-tagged proteins in-vitro and on living cells. Furthermore, a convenient synthesis strategy and a simple labeling protocol is provided for these agents.

A first aspect of the present invention refers to a compound of formula (I)

wherein L₁ is a linker, L₂ and L₃ are independently chemical bonds or linkers,

F is a reporter moiety,

C is a covalent coupling moiety,

NTA is a nitrilo triacetic moiety and n is from 1 -3, or a salt, chelate or derivative thereof.

A salt of the compound (I) may be formed by replacing acidic protons, e.g. of the NTA moiety with organic or inorganic cations and/or by protonating basic groups in the presence of organic or inorganic anions. A chelate of the compound (I) is formed by the NTA moiety and a chelating cation, e.g. a transition metal cation such as Ni²⁺. A derivative of the compound (I) may be a protected derivative which comprises at least one protection group at a reactive moiety. Acid protection groups may be present, e.g. at the NTA moiety and/or base protection groups at amino moieties. Acid protection groups such as t-butyl (tBu) are well known in the art. Base protection groups, e.g. amino protection groups such as butoxycarbonyl (Boc) or benzoyl are well known in the art. A derivative may also be a synthesis intermediate of the compound (I) which comprises amino groups or protected amino groups instead of the groups -NHF and/or -NHC.

The compound (I) is based on a trifunctional core which has two amino functions and one carboxy function. The amino functions may be linked via chemical bonds and/or linker groups to the core. To one of the amino functions a detectable moiety F is bound. To the other amino function a covalent coupling moiety C is bound. To the carboxy function one or several nitrilo thacetic (NTA) moieties are bound via a linker l_i . Preferably, the carboxy moiety of the core is bound via an amide or an ester group to l_i, more preferably via an amide group.

The linker Li preferably has a length of at least two atoms, more preferably of at least 4 atoms and preferably up to 20, more preferably up to 14 atoms, wherein the atoms are selected from carbon atoms and optionally heteroatoms such as nitrogen and/or oxygen atoms. Preferably, the linker L₁ has a structure -l_i'-CO-NH-l_r, wherein Li¹ and l_r are a chemical bond or a linker group with a length of at least 1 atom, more preferably of at least 2 atoms and preferably up to 6 atoms, wherein at least one of Li¹ and l_r is a linker group. More preferably, the linker L₁ has the structure -(CH₂)_n1-CO-NH- (CH₂)n2-, wherein ni and n₂ are independently from 1 -6.

L₂ and L₃ are independently chemical bonds or linkers which may have a length of at least one atom, preferably at least two atoms and preferably up to 12 atoms, more preferably up to 8 atoms, wherein the atoms are selected from carbon atoms and optionally heteroatoms such as nitrogen and/or oxygen atoms. In an especially preferred embodiment, L₂ is a chemical bond and L₃ is a linker. In a further especially preferred embodiment, L₃ is a chemical bond and L₂ is a linker. The linkers L₂ and L₃ may comprise the structures -(CH₂)_n3- and -(CH₂)_n4- respectively wherein n₃ and n₄ are independently 0-6. More preferably, one of n₃ and n₄ is 0 and the other from 2-4.

The compound of formula (I) may comprise 1 , 2 or 3 nitrilo triacetic groups, i.e. n may be 1 , 2 or 3. If n is 2 or 3, the NTA groups are preferably coupled to a cyclic group, e.g. a azamacrocycle such as cyclam which is bound to the linker L₁. Preferably, n is 1.

The reporter moiety F is a moiety the presence of which can be detected. For example, the reporter moiety may be an optically detectable label such as a fluorescence group. Preferred fluorescent groups are rhodamine groups such as TMR or rhodamine B. Further preferred optically detectable moieties are direct labels such as gold or latex particles or enzyme groups. In a further preferred embodiment, F may be a quencher group, i.e. a group the presence of which quenches the fluorescence of adjacent fluorescent groups. An example of a quencher group is QSY7.

The covalent coupling moiety C is a group, which is capable of forming covalent groups with a protein, more particularly with reactive groups of a protein such as carboxy, amino, thio and/or hydroxy groups. In a first embodiment, C is a photoactivatable group, i.e. a group which upon illumination forms a covalent linkage to a complementary group present in a protein. Photoactivatable groups are known in the art and comprise e.g. arylketones, diazo groups, diazirene groups or azide groups. Preferably, C is an arylazide group, more preferably a hydroxy arylazide group, such as azido salicylic acid, which reacts upon illumination (preferably at a wavelength < 380 nm) with a complementary group via an azacycloheptatetraene intermediate formed by a rapid intramolecular rearrangement of a singlet nitrene (Brunner 1993). Hydroxyarylazides are highly electrophilic, but less reactive than other nitrene groups and do not insert into non-activated C-H bonds.

In a further preferred embodiment, C may also be a non-photoactivatable group. This non-photoactivatable group is a reactive group which may form a covalent bond with a complementary reactive group, wherein the reaction is accelerated by the proximity of a receptor ligand complex. High local concentrations of the reactive group of compound (I) and complementary reactive groups in the protein due to the immobilization via the receptor- ligand-complex allow the reaction to take place. In the present case, the immobilization is effected by the Ni²⁺/NTA-poly-His complex. The reactive group of compound (I) may be an electrophilic moiety such as an acryloyl moiety, which may react with nucleophilic moieties present in the protein, e.g. NH, OH, SH groups, when this reaction is accelerated by high local concentration of the respective groups.

In an especially preferred embodiment, the inventive compound is of formula (Ia):

wherein ni and n₂ are independently from 1 -6, n₃ and n₄ are independently from 0-6, and F and C are as defined above, or a salt, or derivative thereof. In compound (Ia), n-i and n₂ are preferably 2-4. Further, it is preferred that one of n₃ and n₄ is O and the other from 2-4.

Further, the present invention refers to a covalent conjugate of a compound of formula (I) or (Ia) as described above with a protein. The protein may be any protein which has a group capable of forming a receptor-ligand-complex with a nitrilo triacetic acid moiety, preferably in the presence of a heavy metal ion such as Ni²⁺. More preferably, the protein comprises polyhistidine sequence, wherein the number of histidine residues is at least 2, preferably at least 3 and more preferably at least 4. Most preferably, the protein comprises a hexahistidine sequence. The polyhistidine sequence is preferably present at the N- and/or C-terminus of the protein.

The protein may be any prokaryotic or eukaryotic protein. In a preferred embodiment, the protein comprises a domain which is localized at the surface of a cell, preferably incorporated into a cell membrane. Examples of such molecules are receptors, e.g. cytokine receptors, such as interleukin receptors, or growth-factor receptors. In a further preferred embodiment, the protein comprises a domain which is capable of interacting with a complementary protein present on the surface of a cell. Examples thereof are ligands capable of binding to a receptor present on the surface of a cell, e.g. cytokines. In an especially preferred embodiment, the protein is a fusion protein comprising a biologically functional domain, e.g. a receptor or ligand as indicated above and a fluorescent domain such as GFP or a variant thereof.

Preferred applications for the compounds and conjugates of the invention are the labeling of biological components, e.g. cell components, cells, viruses, etc. Especially preferred is the labeling of living cells. The labelled biological components may be detected via the reporter group F. Due to the covalent association of the reporter group to the protein, a very sensitive detection is possible, e.g., single-molecule detection. The detection preferably involves optical methods, more preferably a fluorescence detection, e.g. by Fluorescence Correlation Spectroscopy (FCS) or by other fluorescence detection methods as described above.

The compounds (I) are particularly suitable in screening assays, e.g. in assays wherein the effect of a candidate agent on a biological system, e.g. a living cell, is tested. The compounds and conjugates of the invention allow a fast and reliable detection of interactions between protein components of the conjugates and biological systems and may thus be used in High Throughput Screening assays, e.g. for the detection of an effect of a candidate agent on a biological system, particularly on a cell surface protein.

The compounds of the invention are preferably synthesized by a procedure which comprises the steps:

(i) reacting a compound of formula (II):

(pNTA)_n - Li' - COOH

wherein pNTA is a protected nitrilo triacetic group, n is 1 -3, and Li is a linker, with a compound of formula (III):

wherein P₁ is an amino protection group, and L₁" is a linker, and subsequently removing the amino protection group P₁, to obtain a compound of formula (IV): O (pNTA)n L₁ ' U NH L₁ " — NH₂ wherein pNTA, n, LV and l_r are as defined above, (ii) reacting the compound of formula (IV) with a compound of formula (V)

wherein L₂ and L₃ are independently chemical bonds or linkers and P₂ and P3 are amino protection groups, wherein P₂ is different from P₃, to obtain a compound of formula (Vl):

wherein pNTA, n, LΛ L₂", L₂, L₃, P₂ and P₃ are as defined above, and

(iii) (a) functionalizing the group -NHP₂ with a reporter moiety F to obtain the group -NHF, and

(b) functionalizing the group -NHP₃ with a covalent coupling moiety C to obtain the group -NHC, wherein a compound of formula (I) is obtained.

Still a further aspect of the invention refers to a method of irreversibly conjugating a compound of formula (I) to a protein, comprising the steps: (i) providing a chelate of a compound of formula (I), (ii) reversibly binding said chelate to a protein,

(iii) irreversibly coupling said chelate via its covalent coupling moiety to the protein. Further, the present invention is explained by the following figures and examples:

Figure legends

Figure 1 shows a Ni²⁺-NTA chelate.

Figure 1 b shows a non-covalent complex of a Ni²⁺-NTA chelate and a hexahistidine sequence present in a protein.

Figure 2 shows a non-covalent complex of a tridentate chelator, e.g. a Ni²⁺- NTA₃ chelate and a hexahistidine sequence in a protein.

Figure 3 shows the structure of a tris-Ni²⁺-NTA chelate having a functional amino group to which e.g. a dye R is bound.

Figure 4 shows a scheme for the synthesis of a tris-Ni²⁺-NTA chelate linked to a dye.

Figure 5 shows a scheme for the synthesis of a mono-Ni²⁺-NTA chelate linked to a dye and a covalent coupling moiety.

Figure 6 shows structures of the quencher compound QSY7 and the fluorescent dyes TMR and RhB. The indicated absorption wavelength and ε- values are as determined in MeOH.

Figure 7 shows a schematic representation of the irreversible covalent coupling of an inventive compound to a protein. The NTA moiety directs the probe to the labeling site, e.g. a hexahistidine sequence. Thereby, the reactive group (covalent coupling moiety) can bind selectively to the protein to form an irreversible covalent bond. Fiqure 8 shows the structure of the preferred photo-crossl inker moiety N- hydroxysuccinimide (NHS)-ASA, an example of a hydroxyarylazide moiety.

Figure 9 shows a scheme for the proximity-accelerated nucleophile- electrophile reaction triggered by the interaction between an NTA group and a polyhistidine sequence.

Figure 10 shows fluorescence spectra before and after addition of non- fluorescent ths-Ni²⁺-NTA quencher compound H to a GFP-C-His protein.

Figure 11 shows a fluorescence resonance energy transfer (FRET) between a tris-Ni²⁺-NTA TMR compound 10 non-covalently coupled to a GFP-C-his protein.

Figure 12 shows visualization of cell surface receptors with a ths-Ni²⁺-THR compound 10, targeted to an IL-4R alpha chain-GFP-C-His fusion protein.

Figure 13 shows a covalent cross-linking of NTA-ASA derivatives in solution, a) The emitted GFP-C-His fluorescence (15 nM, solid dark line) is quenched by addition of non-fluorescent NTA-ASA-QSY7 35 (10 μM, line annotated as "2") due to FRET. Removal of the complexating Ni²⁺ cations with a 250-fold excess of EDTA after crosslinking (120 μJ) leads to a recovery of a residual of 10% of the donor fluorescence (line annotated as "1 ") compared to a non- irradiated control (dotted line), b) Kinetics of EDTA unquenching for NTA- ASA-TMR 34 (dark line) and -QSY7 35 (light grey line) under similar conditions, c) Determination of the equilibrium dissociation constant for NTA- ASA-TMR 34 by applying the Hill equation in the form y= (1 -E/(1 +(Kd/(c^*1 E- 6))^Λh))*100 with FRET efficiency E, NTA-concentration c, dissociation constant Kd and Hill coefficient h (Guignet, Hovius and Vogel 2004). d) SDS- PAGE of GFP-C-His (1.25 μM) incubated with NTA-ASA-TMR (5 μM). The reaction was crosslinked with 120 μJ (lanes 1 and 3), 0 μJ (lane 2), or 240 μJ (lane 4). Covalent linkage in lane 3 was prevented by a 100-fold excess of EDTA (0.5 μM). Figure 14 shows covalent cross-linking of NTA-ASA-TMR at the cell surface. a)-c) and d)-f) two cells expressing NHis-IL-4Rac-GFP, g)-i) a cell expressing IL-4Rac-GFP as a negative control. Confocal cross sections in the GFP channel (a, d, g) show that the receptors are expressed in endogenous membrane systems (EM) associated with the nucleus (N) as well as at the surface plasma membrane (PM). After cross-linking, NTA-ASA-TMR 34 is found at the plasma membrane of His-tag expressing cells (b, f) but not for a negative control of similar expression level (h, compare concentration values in Table 1 ). The plasma membrane was selected as a region of interest (ROI) exemplarily shown for the TMR- (c, i) and the GFP-channel (e) of the three cells, respectively. Panel i) illustrates how the membrane of neighboring non- transfected cells (ROI 1 ) was discriminated from the membrane of a receptor expressing cell (ROI 2) in order to generate the values in Table 1.

Figure 15 shows autocorrelation curves for e-GFP-C-His and NTA-ASA-TMR 34.

Figure 16 shows calibration curves for e-GFP-C-His and NTA-ASA-TMR 34.

Figure 17 shows Kd values of NTA-I and NTA-II vs eGFP-C-His. The concentration dependent fluorescence quenching of eGFP-C-His by NTA-I and NTA-II was fitted with the Hill-equation, as previously published. Kd values of 0.25 μM and 3.5 μM were obtained for NTA-I and NTA-II, respectively.

Figure 18 shows the binding of His-tagged receptors by NTA-ASA-TMR. Examples

1 Materials and Methods

HCLH-GIU(OBZI)-OBU was purchased from Senn Chemicals.

The succinimidyl ester of TMR was purchased from Fluka Biochemika, the succinimidyl esters of QSY 7 and of Rhodamine Red X from Invitrogen.

Boc-Lys(Z)-OH was purchased from Novabiochem.

The photocrossl inkers Sulfo-SAND (Sulfosuccinimidyl 2[m-azido-o- nitrobenzamido]-ethyl-1 ,3^'-dithiopropionate), NHS-ASA (N-

Hydroxysuccinimidyl-4-azidosalicylic acid), Sulfo-SFAD (Sulfosuccinimidyl- [perfluoroazidobenzamido]ethyl-1 ,3^'-dithiopropionate) and SANPAH (N- Succinimidyl-6-[4^'-azido-2^'-nitrophenylamino]hexanoate) were purchased from Pierce.

6-(Boc-amino)caproic acid, benzyl N-(2-aminoethyl) carbamate hydrochloride, benzylethane-1 ,2-diamine and 1 ,4,8,11 -tetraaza- cyclotetradodecane were purchased from either Fluka or Sigma Aldrich.

All solvents used in this study were from Fluka.

1.1 Analytical tools

1.1.1 MS

The mass spectra were recorded on a Waters MS-70.4000 micromass spectrometer, fragment ions are given in m/z.

1.1.2 HPLC

1.1.2.1 Analytical HPLC of non-complexed compounds

Analytical reversed phase (RP) HPLC: Agilent 1100 Series System (Quat pump G1311A, degasser G1322A, multiwavelengths and fluorescence detector DAD G1315B and FLD G1321A). Column: Zorbax Ci₈ (4.6 x 150 mm, 3.5 μm).

Analyses were performed using a linear gradient of A (95% H₂O, 5% MeCN, 1 % trifluoroacetic acid (TFA)) and B (95% MeCN, 5% H₂O, 1 % TFA).

The following gradient was used (unless something else is stated):

at a flow rate of 1 ml/min with UV detection at 220 nm, and/or 555 nm for the TMR or 560 nm for the quencher QSY7 or 570 nm for the Rhodamine Red X.

1.1.2.2 Analytical HPLC of Ni 2+ complexed compounds

N ;i2+ complexed compounds were analyzed using the previous gradient but with A: ammonium acetate (AcONH₄) in water (400 mg/l) and B: MeCN.

Retention times (Rt) are given in minutes.

1.1.2.3 Preparative HPLC of non-complexed compounds

Preparative reversed phase (RP) HPLC: Agilent 1100 Series preparative system (Prep pump G1361A, multiwavelengths detector MWD G1365B). Column: Agilent prep Ci₈ (21.2 x 150 mm, 10 μm).

Crude products were purified using the following linear gradient of A (95% H₂O, 5% MeCN, 1 % TFA) and B (95% MeCN, 5% H₂O, 1 % TFA) at a flow rate of 20 ml/min with UV detection at the corresponding wavelengths.

1.1.2.4 Preparative HPLC of Ni complexed compounds

N ;i2+ complexed compounds were purified using the same gradient, flow, and UV detection with A: ammonium acetate (AcONH₄) in H₂O (400 mg/l) and B: MeCN.

Retention times (R_t) are given in minutes.

1.1.3 NMR spectra

¹H-NMR and ¹³C-NMR spectra were recorded on a Bruker Avance 500 (500MHz) and Avance 400 (400 MHz). Chemical shifts δ in ppm downfield from internal standard Me₄Si (δ = 0 ppm); J values in Hz. The multiplicities are reported by the following symbols: s (singlet), d (doublet), t (triplet), q (quadruplet), m (multiplet), dt (dedoubled triplet), dd (dedoubled doublet).

1.2 General synthesis procedures

1.2.1 Boc and ferf-butyl deprotection

For the Boc and/or fe/t-butyl deprotection, compounds were treated with a solution of TFA/triisopropyl-silane (TIS)/H₂O (95:2.5:2.5) to a final concentration of 50 mg/ml. The mixture was stirred for 1 h at room temperature (RT). The TFA was then removed under reduced pressure.

1.2.2 Coupling reactions of the dyes and/or photo-crosslinker performed in organic solvent The compound containing the free amine was dissolved in dry dimethyl formamide (DMF) to a final concentration of 7-40 mg/ml. N₁N- diisopropylethylamine (DIPEA) (6 eq) was then added to the solution followed by addition of the dye or photo-crossl inker (1 eq) dissolved in DMF. The reaction mixture was then stirred for 2 h at RT. After removing the solvents, the labeled compound was purified by preparative HPLC and analyzed by MS.

1.2.3 Coupling reactions of the dyes and/or photo-crosslinker performed in organic solvent

The compound containing the free amine was dissolved in NaHCO₃ buffer (pH 8.4) to a final concentration of 5-10 mg/ml. The dye NHS ester dissolved in dry DMF (5-10 mg/200 μl) was then added and the reaction mixture was stirred for 2 h at RT. The labeled compound was purified by preparative HPLC and analyzed by MS.

1.2.4 Complexation with Ni 2+

The compound foreseen to be complexed was dissolved in H₂O to a final concentration of 5-10 mg/ml and treated with a 0.1 M solution of NiCI₂ (1 eq). The mixture was stirred for 2 h at RT. Finally, the Ni²⁺ complex was purified by preparative HPLC and analyzed by MS.

1.3 Synthesis of the tris-NTA derivative

1.3.1 2-(bis-tert-butoxycarbonylmethyl-amino)-pentandioic acid 5- benzyl ester 1-tert-butyl ester (GIu-NTA(OfBu)₃OBzI) (2) HCLH-GIu(OBzI)-OtBu (1 ) (3.3 g, 10 mmol) was dissolved in DMF (75 ml), te/t-butyl bromoacetate (5.9 ml, 40 mmol) and DIPEA (8.6 ml, 50 mmol) were added. The mixture was heated at 65° C overnight with continuous stirring under argon. Then NH₄CI saturated solution (30 ml), Et₂O (50 ml) and H₂O (50 ml) were added. The aqueous phase was washed with Et₂O (4 x 60 ml). All organic phases were combined and dried (MgSO₄) and then the volatiles were removed under reduced pressure. FC (cyhex/AcOEt 9:1 ) yielded (2) (4.031 g, 77%). fl_f = 0.54 (cyhex/AcOEt 9:1 ). MS (ESI, C₂₈H₄₃NO₈): m/z 544.22 [M+Na]⁺. ¹H-NMR (500 MHz, CD₃OD); δ 1.44 (s, 18H, /Bu); 1.47 (s, 9H, Bu); 1.9 (m, 2H, CH₂); 2.63 (m, 2H, CH₂CO); 3.37 (m, 1 H, NCH); 3.4 (s, 4H, CH₂N); 5.11 (s, 2H, (PhCH₂-); 7.32 (m, 5H, 5 arom. H). ¹³C-NMR DEPT 135 (500 MHz, CDCI₃): δ 26.3 (CH₂); 28.4 (CH₃); 31.3 (CH₂); 54.9 (CH₂); 65.7 (CH); 67.2 (CH₂); 82.0 (CH₂); 82.6 (CO); 129.1 (CH); 172.2 (CO); 173.3 (CO); 174.9 (CO).

1.3.2 2-(bis-tert-butoxycarbonylmethyl-amino)-pentandioic acid 1- tert-butyl ester (GIu-NTA(OfBu)₃-OH) (3)

Pd/C (80 mg) was added to a solution of (2) (3.987 g, 7.64 mmol) in MeOH (200 ml) after purging with argon. The reaction mixture was stirred for 6 h under H₂ atmosphere. Pd/C was then filtered over celite and the volatiles removed under reduced pressure yielded (3) (3.21 g, 97%). R_f= 0.4 (cychex/AcOEt 3:2). MS (ESI, C₂iH₃₇NO₈): m/z 454.19 [M+Na]⁺. ¹H-NMR (400 MHz, CD₃OD); δ 1.47 (m, 27H, 3x /Bu); 1.88 (m, 2H, CH₂); 2.55 (m, 2H, CH₂COOH); 3.37 (dd, ²J=9.57, 1 H, NCH); 3.46 (s, 4H, (CH₂)₂N). ¹³C-NMR DEPT 135 (400 MHz, CD₃OD); δ 26.6 (CH₂); 28.2 (CH₃); 28.49 (CH₃); 31.22 (CH₂); 55.053 (CH₂); 65.981 (NCH); 82.05 (C); 82.573 (C); 172.238 (CO); 173.344 (CO); 177.081 (COOH).

1.3.3 5-[8,11 -Bis-^-Cbis-tert-butoxycarbonylmethyl-aminoH-tert- butoxycarbonyl-butyryll-^θ-tert-butoxycarbonylamino- hexanoyl)-1 ,4,8,11 tetraaza-cyclotetradec-1 -yl]-2-(bis-tert- butoxycarbonylmethyl-amino)-5-oxo-pentanoic acid tert- butyl ester (TrJs-NTA(OfBu)₉) (4)

GIu-NTA(OtBu)₃-OH (3) (1 eq, 1.43 mmol, 617 mg) was dissolved in dry dichloromethane (DCM) (55 ml), followed by addition of 1 ,4,8,11 tetraaza- cyclotetradecane (0.33 eq, 0.472 mmol, 94.5 mg) at 0° C. 2-(1 H-7- azabenzotriazol-1 -yl) 1 ,1 ,3,3-tetramethyl uranium hexafluorophosphate methanaminium (HATU) (1 eq, 1.43 mmol, 543.6 mg) and DIPEA (2 eq, 2.85 mmol, 490 μl) were added successively and the solution was stirred under argon atmosphere at RT. After 5 h, NaHCO₃ saturated solution (30 ml) was added and stirring was continued for 10 min. After addition of H₂O (30 ml), the aqueous phase was washed with AcOEt (2 x 30 ml) and the organic phase was dried (MgSO₄). The volatiles were removed under reduced pressure. FC (AcOEt containing 1 % NH₃) yielded (4) (532 mg, 78%). f?_f = 0.11 (AcOEt containing 1 % NH₃). MS (ESI, C₇₃Hi₂₉N₇O₂I): m/z 743.16 [M+2Na]²⁺/2; 1462.75 [M+Na]⁺.

1.3.4 5-[8,11 -Bis-[4-(bis-tert-butoxycarbonylmethyl-amino)-4-tert- butoxycarbonyl-butyryl]-4-(6-tert-butoxycarbonylamino- hexanoyl)-1 ,4,8,11 tetraaza-cyclotetradec-1 -yl]-2-(bis-tert- butoxycarbonylmethyl-amino)-5-oxo-pentanoic acid tert- butyl ester ((Boc)-amino caproic-tris-NTA(OtBu)g) (5)

Compound (4) (1 eq, 0.35 mmol, 511 mg) was dissolved in dry DCM (20 ml), followed by addition of 6-(Boc-amino)caproic acid (1 eq, 0.35 mmol, 82 mg) at 0° C. HATU (1.2 eq, 0.42 mmol, 162 mg) and DIPEA (3 eq, 1.06 mmol, 182 μl) were added to the solution. After continuous stirring for 3 h at RT under Ar atmosphere, NaHCO₃ saturated solution (20 ml) was added to the mixture, and stirring was continued for 10 min. After addition of H₂O (10 ml), the aqueous phase was washed with AcOEt (2 x 25 ml) and the combined organic phases were dried (MgSO₄). The volatiles were removed under reduced pressure. FC (AcOEt) yielded (5) (515.7 mg, 88%). f?_f = 0.45 (AcOEt). MS (ESI, C₈₄H₁₄₈N₈O₂₄): m/z 849.91 [M+2Na]²⁺/2; 1676.96 [M+Na]⁺. 1.3.5 5-{4-(6-Amino-hexanoyl)-8,11 -bis-[4-(bis-carboxymethyl- amino)-4-carboxy-butyryl]-1 ,4,8,11 -tetraaza-cyclotetradec-1 - ylJ^-fbis-carboxymethyl-aminoJ-S-oxo-pentanoic acid (amino caproic acid-Tris-NTA) (6)

Compound (5) (515 mg, 0.311 mmol) was dissolved in a solution of TFA/TIS/H₂O (95:2.5:2.5) (10 ml). The reaction mixture was stirred for 1 h at RT. The volatiles were then removed under reduced pressure. The product (6) was precipitated by addition of cold Et₂O (10 ml). Yield: 249 mg, 76%. MS (ESI, C₄₃H₆₉N₈O₂₂): m/z 525.43 [M+2H]²⁺/2; 1049.39 [IvRH]⁺. ¹H-NMR (500 MHz, DMSO): δ 1.27 (m, 2H, CH₂); 1.49 (m, 4H, 2 x CH₂); 1.85 (m, 1OH, 5 x CH₂); 2.29 (m, 2H, NCOCH₂); 2.48 (m, 3H, NCH); 2.74 (m, 2H, CH₂NH₂); 3.44 (m, 34H, 17 x CH₂).

1.3.6 Conjugation of the tris-NTA with dyes

1.3.6.1 Conjugation with TMR

The TMR-NHS ester was coupled to the amino caproic acid tris-NTA derivative (6) following the general method described in chapter 1.2.3. Preparative HPLC (using the general method described in 1.1.2.3) yielded (7), a purple solid (1.4 mg, 20%) f?_t =14 (analytical HPLC as described in 1.1.2.1 ). MS (ESI, C₆₈H₈₈Ni₀O₂₆): m/z 731.59 [M+2H]²⁺/2.

1.3.6.2 Conjugation with QSY7

The quencher QSY7-NHS ester was coupled to the amino caproic acid tris- NTA derivative (6) following the general method described in chapter 1.2.3. Preparative HPLC (using the general method described in 1.1.2.3) yielded (8), a mauve solid (3.5 mg, 33%). f?_t=22 (analytical HPLC as described in 1.1.2.1 ), MS (ESI, C₈₂HiO₂NIiO₂₆S): m/z 563.77 [M+3H]³⁺/3; 845.21 [M+2H]²⁺/2.

1.3.6.3 Conjugation with RhB Red X

The Rhodamine Red X NHS ester was coupled to the amino caproic acid tris- NTA derivative (6) following the general method described in chapter 1.2.3. Preparative HPLC (using the general method described in 1.1.2.3) yielded (9), a purple solid (2.9 mg, 57%). f?_t=23.2 (analytical HPLC as described in 1.1.2.1 ). MS (ESI, C₇₆HiO₇NIiO₂₉S): m/z 568.48 [M+3H]³⁺/3; 852.27 [M+2H]²⁺/2.

1.3.7 Complexation of the tris-NTA chromophores with Ni 2+

The tris-NTA-TMR (7), tris-NTA-QSY7 (8) and tris-NTA-RhB (9) were complexed with Ni²⁺ following the general method described in chapter 1.2.4. Preparative HPLC (using the general method described in 1.1.2.4) yielded tris-NTA-TMR-Ni²⁺ (10) as purple solid (2 mg, 24%). R_{ = 9.5 (analytical HPLC as described in 1.1.2.2). MS (ESI, C₆₈H₈₂Ni₀Ni₃O₂₆): m/z 544.59 [M+3H]³⁺/3; 837.15 [M+2Na]²⁺/2.

tris-NTA-QSY7-Ni²⁺ (11 ) as mauve solid (17.5 mg, 45%). fl_t = 17.8 (analytical HPLC as described in 1.1.2.2). MS (ESI, C₈₂H₉₆NnNi₃O₂₆S): m/z 620.31 [M+3H]³⁺/3; 930.29 [M+2H]²⁺/2.

tris-NTA-RhB-Ni²⁺ (12) as purple solid (2.9 mg, 22%). R_{ = 14.5 (analytical HPLC as described in 1.1.2.2). MS (ESI, C₇₆HioiNnNi₃O₂₉S₂): m/z 625.35 [M+3H]³⁺/3; 958.77 [M+2Na]²⁺/2.

1.4 Synthesis of mono-NTA derivatives

1.4.1 3-{[2-(2-Benzyloxycarbonylamino-ethylcarbamoyl)-ethyl]- tert-butoxycarbonyl-aminoj-pentanedioic acid di-tert-butyl ester (13)

Compound (3) (1 eq, 4.63 mmol, 2 g) was dissolved in dry DMF (150 ml) at 0° C, followed by addition of (2-amino-ethyl)carbamic acid benzyl ester (1.5 eq, 6.95 mmol, 1.6 g). HATU (1 eq, 1.762 g, 4.63 mmol) and DIPEA (6 eq, 27.8 mmol, 4.76 ml) were added to the solution so obtained. This mixture was stirred overnight at RT under Ar atmosphere. NaHCO3 saturated solution (40 ml) was added to the mixture, which was stirred for 10 min. Then AcOEt (60 ml) and H₂O (40 ml) were added. The aqueous phase was washed with AcOEt (2 x 50 ml). The combined organic phases were dried (MgSO₄) and concentrated under reduced pressure. FC (cyhex/AcOEt 2:3) yielded (13) (2.33 g, 83%). R₁= 0.4 (cyhex/AcOEt 2:3). MS (ESI, C_3IH₄₉N₃O₉): m/z 630.37 [M+Na]⁺.

1.4.2 3-{[2-(2-Amino-ethylcarbamoyl)-ethyl]-tert-butoxycarbonyl- amino}-pentanedioic acid di-tert-butyl ester (14)

Pd/C (20 mg) was added to a solution of (13) (2.3 g, 3.78 mmol) in MeOH (100 ml). The reaction mixture was vigorously stirred overnight under H₂ atmosphere at RT. Pd/C was then filtered off over celite and the volatiles were removed under reduced pressure yielding (14) (768 mg, 43%). MS (ESI, C₂₃H₄₃N₃O₇): m/z 474.28 [IvRH]⁺; 496.25 [IvRNa]⁺. ¹H-NMR (500 MHz, CD₃OD): δ 1.45(s, 18H, Bu); 1.47 (s, 9H, Bu); 1.91 (m, 3H, CH₂); 2.41 (m, 2H, CH₂CONH); 2.72 (t, ²J = 6.29, 2H, CH₂NH₂); 3.25 (td, ²J = 6.3, ³J = 1.98, 2H, CH₂CH₂NH₂); 3.33 (t, J = 4.9, 1 H, NCH); 3.58 (s, 4H, N(CH₂J₂). ¹³C-NMR DEPT 135 (500 MHz, CDCI₃): δ 27.2 (CH₂); 28.4 (CH₃); 28.4 (CH₃); 33.3 (CH₂); 43.0 (CH₂); 54.94 (CH₂); 66.1 (CH); 82.1 (CO); 82.6(CO); 172.4 (CO); 173.2 (CO); 175.9 (CO).

1.4.3 3-({2-[2-(6-Benzyloxycarbonylamino-2-tert- butoxycarbonylamino-hexanoylamino)-ethylcarbamoyl]- ethyl}-tert-butoxycarbonyl-amino)-pentanedioic acid di-tert- butyl ester (15)

Compound (14) (1 eq, 1.37 mmol, 652.2 mg) was dissolved in dry DMF (40 ml) at 0° C, followed by addition of Boc-Lys(Z)-OH (1.5 eq, 2.06 mmol, 785.7 mg). HATU (1 eq, 1.37 mmol, 523.5 mg) and DIPEA (6 eq, 8.26 mmol, 1.41 ml) were added. This mixture was stirred overnight at RT under Ar atmosphere. NaHCO₃ saturated solution (25 ml) was added to the mixture, which was stirred for 10 minutes. The aqueous phase was extracted with DCM (2 x 20 ml). The combined organic phases were then washed with NH₄CI saturated solution (20 ml), and NaCI saturated solution. The organic phase was dried (MgSO₄) and concentrated under reduced pressure. FC (cyhex/AcOEt 1 :4) yielded (15) (831.3, 72%). R, = 0.3 (cyhex/AcOEt 1 :4). MS (ESI, C₄₂H₆₉N₅Oi₂): m/z 859.58 [M+Na]⁺. ¹H-NMR (500 MHz, (CD₃)2O): δ 1.36 (m, 2H, NHCHCH₂CH₂); 1.44 (s, 9H, Bu); 1.47 (s, 9H, Bu); 1.45 (s, 18H, 2 x Bu); 1.52 (m, 2H, CH); 1.65 (m, 2H, NHCHCH); 1.89 (m, 2H, NHCOCH₂); 2.39 (td, ²J = 7.44, ³J = 1.34, 2H, COCH); 3.13 (t, ²J = 6.82, 2H, CH₂NHCO); 3.28 (m, 4H, NHCH₂CH₂NH); 3.33 (t, ²J = 5.89, NHCHCH₂); 3.44 (m, 4H, 2x NCH₂CO); 3.94 (m, 1 H, NHCH); 5.06 (s, 2H, OCH₂Ph); 7.31 (m, 5H, 5 arom.H). ¹³C-NMR DEPT 135 (500 MHz, CDCI₃): δ 24.1 (CH₂); 27.6 (CH₂); 28.4 (CH₃); 28.49 (CH₃); 28.7 (CH₃); 30.5 (CH₂); 32.9 (CH₂); 33.3 (CH₂); 39.9 (CH₂); 40.0 (CH₂); 54.9 (CH₂); 56.2 (CH); 66.1 (CH); 67.3 (OCH₂Ph); 80.6 (OC); 82.1 (CO); 82.6 (CO); 128.763 (PhCH); 138.45 (Ph); 157.92 (CO); 158.94 (CO); 172.394 (CO); 173.2 (CO); 175.5 (CO); 175.9 (CO).

1.4.4 3-({2-[2-(6-Amino-2-tert-butoxycarbonylamino- hexanoylamino)-ethylcarbamoyl]-ethyl}-tert- butoxycarbonyl-amino)-pentanedioic acid di-tert-butyl ester (16)

Pd/C (70 mg) was added to a solution of (15) (793.6 mg, 0.95 mmol) in MeOH (25 ml). The reaction mixture was vigorously stirred 3 h under H₂ atmosphere at RT. Pd/C was then filtered off over celite and the volatiles were removed under reduced pressure yielding (16) (653 mg, 98%). MS (ESI, C₃₄H₆₃N₅Oi₀): m/z 702.54 [M+H]⁺; 725.50 [M+Na]⁺. ¹H NMR (400 MHz, CD₃OD): δ 1.37(m, 2H, CH₂); 1.47(s, 27H, 3x tBu); 1.6 (m, 2H, CH₂); 1.89 (m, 2H, ₂CH₂); 2.39 (t, ²J = 7.51 , 2H, CH₂); 2.65 (t, ²J = 7.03, 2H CH₂); 3.24 (m, 2H, 2H, CH₂); 3.33 (t, ²J = 3.43, 1 H, CH); 3.44 (d, J = 6.02, 2H, CH₂COO); 3.94 (m, 1 H, NHCH). ¹³C-NMR DEPT 135 (400 MHz, CD₃OD): δ 24.6 (CH₂); 27.6 (CH₂); 28.9 (CH₃); 33.2 (CH₂); 33.5 (CH₂); 33.8 (CH₂); 40.3 (CH₂); 40.5 (CH₂); 42.4 (CH₂); 55.3 (CH₂); 56.6 (CH); 66.5 (CH); 82.6 (C); 83.1 (C); 172.8 (CO); 173.6 (CO); 175.5 (CO); 176.3 (CO).

1.4.5 Conjugation of mono-NTA with the dyes and the photo- crosslinker or the acryloyl, and complexation with Ni²⁺

1.4.5.1 Mono-NTA-sulfo-SAND (24)

The compound (16) was coupled with sulfo-SAND following the general method described in chapter 1.2.2, leading to compound (18). MS (ESI, C₄₆H₇₄Ni₀Oi₄S₂): m/z 1077.64 [M+Na]⁺.

The protecting groups of the compound (18) were removed proceeding as described in chapter 1.2.1 , the purification by preparative HPLC (using the method described in 1.1.2.3) yielded (24) (3 mg, 41 %). R_t = 8 (using the method described in 1.1.2.2). MS (ESI, C₂₉H₄₂Ni₀Oi₂S₂): m/z 787.31 [M+H]⁺.

1.4.5.2 Mono-NTA-sulfo-SFAD (25)

The compound (16) was coupled with sulfo-SFAD following the general method described in chapter 1.2.2, leading to compound (19). MS (ESI, C₄₆H_7IF₄N₉Oi₂S₂): m/z 1104.58 [M+Na]⁺.

The protecting groups of the compound (19) were removed proceeding as described in chapter 1.2.1 , the purification by preparative HPLC (using the method described in 1.1.2.3) yielded (25) (4.2 mg, 59%). R₁ = 10 (using the method described in 1.1.2.2). MS (ESI, C₂₉H₃₉F₄N₉Oi₂S₂): 814.27 [M+H]⁺.

1.4.5.3 Ni-Mono-NTA-ASA-TMR (34)

Compound (16) was coupled with NHS-ASA following the general method described in chapter 1.2.2, yielding (17). MS (ESI, C₄iH₆6N₈Oi₂): m/z 885.59 [M+Na]⁺.

The protecting groups of (17) were removed proceeding as described in chapter 1.2.1 , and purified by preparative HPLC (using the method described in 1.1.2.3) yielding (23) (9.4 mg, 56%). R_t = 10.5 (using the method described in 1.1.2.1 ). MS (ESI, C₂₄H₃₄N₈Oi₀): m/z 595.27 [M+H]⁺.

TMR was coupled to (23) proceeding as described in chapter 1.2.2. Preparative HPLC (using the method described in 1.1.2.3) yielded (29) (4.33 mg, 65%). fl_t = 18.5. MS (ESI, C₄₉H₅₄Ni₀Oi₄): m/z 1007.42 [M+H]⁺.

Compound (29) was complexed with Ni²⁺ following the method described in chapter 1.2.4. Preparative HPLC yielded (34) (1.2 mg, 17%). fl_t = 9.8 (A (400 mg AcONH₄ in 1 I H₂O) and B (MeCN); 25% to 45% B in 20 min). MS (ESI, C₄₉H₅₂Ni₀NiOi₄): m/z 555.31 [M+2Na]²⁺/2.

1.4.5.4 Mono-NTA- NM 2+⁺-ASA-QSY7 (35)

QSY7 was coupled to (23) proceeding as described in chapter 1.2.2. Preparative HPLC yielded (30) (6.2 mg, 80%). R₁ = 10.5 (with A (95% H₂O, 5% MeCN, 1 % TFA) and B (95% MeCN, 5% H₂O, 1 % TFA); 30% to 95% B within 20 min). MS (ESI, C₆₃H₆₈NnOi₄S): m/z 618.08 [M+2H]²⁺/2; 1234.60 [M+H]⁺.

Compound (30) was complexed with Ni²⁺ following the method described in chapter 1.2.4, using a mixture of MeOH/H₂O (1 :1 ). Purification by preparative HPLC (using the method described in 1.1.2.3) yielded (35) (2.7 mg, 32%). R_{ = 23 (using the method described in 1.1.2.2). MS (ESI, C_6SH₆₆NnNiOi₄S): m/z 658.82 [M+Na+H]²⁺/2.

1.4.5.5 Mono-NTA -Ni 2+ -TMR-Acryloyl (36)

The compound (16) was coupled with TMR proceeding as described in chapter 1.2.2, but using dry DCM instead of dry DMF, leading to the compound (20). MS (ESI, C₅₉H₈₃N₇Oi₄ ): 580.13 [M+2Na]²⁺/2; 1136.76 [M+Na]⁺.

The protecting groups of compound (20) were removed proceeding as described in chapter 1.2.1 , the purification by preparative HPLC (using the method described in 1.1.2.3) yielded (26). (2.2 mg, 52%). R_t = 11.2 (using the method described in 1.1.2.1 ). MS (ESI, C₄₂H₅I N₇Oi₂): m/z 423.92 [M+2H]²⁺/2.

Compound (26) (1 eq, 2 mg, 2.3 μmol) was dissolved in 1 M NaOH (1 ml), and acryloyl chloride (100 eq, 230 μmol, 20 μl) was added progressively by following the reaction progress by analytical HPLC. 0.1 M HCI (1 ml) was added to the mixture in order to neutralize it. Purification by preparative HPLC (using the method described in 1.1.2.3) yielded (31 ). R₁ = 13.2 (using the method described in 1.2.2.1 ). MS (ESI, C₄₅H₅₃N₇Oi₃): m/z 461.86 [M+Na+H]²⁺/2; 900.34 [M+H]⁺.

The compound (31 ) so obtained was complexed following the method described in chapter 1.2.4. Purification by preparative HPLC (using the method described in 1.1.2.4) yielded (36). (2.2 mg, 96%). R₁ = 12.8 (using the method described in 1.1.2.2). MS (ESI, C₄₅H_5! N₇NiOi₃): m/z 478.87 [M+2H]⁺/2. 1.4.5.6 Mono-NTA-Ni²⁺-QSY7-Acryloyl (37)

Compound (16) was coupled with QSY7 proceeding as described in chapter 1.2.2, using dry DCM instead of dry DMF, leading to compound (21 ). MS (ESI, C₇₃H₉₇N₈Oi₄S): m/z 682.72 [M+2Na]²⁺/2.

The protecting groups of compound (21 ) were removed proceeding as described in chapter 1.2.1. Purification by preparative HPLC (using the method described in 1.1.2.3) yielded (27). R_t = 20.4. MS (ESI, C₅₆H₆₅N₈Oi₂S ): m/z 537.54 [M+2H]²⁺/2; 1073.46 [M+H]⁺.

Compound (27) (1 eq, 4.6 μmol, 5 mg) was dissolved in 1 M NaOH (2 ml), and acryloyl chloride (50 eq, 232 μmol, 19 μl) was added dropwise, and the reaction was followed by analytical HPLC. After completion of the reaction, 0.1 M HCI (2 ml) was added to the mixture and the product was purified by preparative HPLC (using the method described in 1.1.2.3) yielding (32). (1.75 mg, 33%). f?_t=22.8 (using the method described in 1.1.2.1 ). MS (ESI, C₅₉H₆₇N₈Oi₃S): m/z 564.53 [M+2H]²⁺/2; 1127.51 [M+H]⁺.

Compound (32) was complexed with Ni²⁺ following the method described in chapter 1.2.4, yielding (37) after purification by preparative HPLC (using the method described in 1.1.2.4). (2.8 mg, 96%). R₁ = 21.8 (using the method described in 1.1.2.1 ). MS (ESI, C₅₉H₆₅N₈NiOi₃S): m/z 614.46 [M+2Na]²⁺/2; 1207.33 [M+Na]⁺.

1.4.5.7 Mono-NTA- Ni²⁺-Acryloyl-TMR (38)

Compound (16) (1 eq, 0.14 mmol, 100 mg) and Et₃N (2 eq, 0.28 mmol, 39 μl) were dissolved in dry DCM (15 ml) and cooled to 0° C. Acryloyl chloride (4 eq, 0.568 mmol, 46 μl) was added dropwise. The mixture was stirred for 1.5 h at RT and then washed with H₂O, saturated solution NaCI and dried (MgSO₄), leading to the compound (22). MS (ESI, C37H65N5O11 ): m/z 756.54 [M+H]⁺.

Protecting groups of compound (22) were removed proceeding as described in chapter 1.3.1 , leading after purification by preparative HPLC (with A (95% H₂O, 5% MeCN, 1 % TFA) and B (95% MeCN, 5% H₂O, 1 % TFA); 1 % to 8% B in 20 min) to (28). (42 mg, 72%). R_t = 4.2. MS (ESI, C₂₀H₃₃N₅O₉ ): m/z 488.28 [M+H]⁺ .

TMR was coupled to the acryloyl amide compound (28) following the method described in chapter 1.3.3. Purification by preparative HPLC (using the method described in 1.1.2.3) yielded (33). (9.17 mg, 37%). fl_t = 13.2 (using the method described in 1.1.2.1 ). MS (ESI, C₄₅H₅₃N₇Oi₃): m/z 450.82 [M+2H]²⁺/2; 900.42 [M+H]⁺.

Compound (33) was complexed with Ni²⁺ proceeding as described in chapter 1.2.4. Purification by preparative HPLC (using the method described in 1.1.2.4) yielded (38). (2.7 mg, 26%). R_t = 11.8 (using the method described in 1.1.2.2). MS (ESI, C₄₅H₅IN₇NiOi₃) m/z 478.93 [M+2H]²⁺/2; 956.41 [M+H]⁺.

1.5 Crosslinking

For all crosslinking experiments UV irradiation was applied with a Strata/inker 1800 (Stratagene, USA) by placing the sample in 10 cm distance to 365 nm UV tubes. An automatic routine deposits a dose of 12O mJ within 40 seconds. The dosage is calibrated internally by a photodiode. Crosslinking of solutions was performed in a droplet placed on aluminium foil from which even μl volumes could be recovered with a pipette. This treatment increased the x-link efficiency 4-fold compared to crosslinking in 1.5 ml polyethylene vials. Crosslinking of cells was performed by placing the NUNC-chambered coverglass slides (Nunc LAB-TEK Cat. No. 155409) in 10 cm distance to the UV bulbs on aluminium foil.

1.6 Fluorescence spectroscopy

Emission spectra and kinetic binding data were acquired on a Fluorolog τ-3 spectrofluorometer (Jobin Yvon / Horiba, Germany). GFP-C-His and NTA- ASA-QSY7 (35) in Figure 13 a,b were measured in PBS supplemented with 0.005% Tween-20 using 600 μl quartz cuvettes (115-F QS, Hellma, Germany). The time course and Kd of NTA-ASA-TMR (34) binding to the GFP labeled target (Figure 13 b,c) were determined in 50 mM NaPO₄, 300 mM NaCI, supplemented with 0.005% Tween-20 (according to (Guignet, Hovius and Vogel 2004) using 60 μl cuvettes (26-50-F Q, Starna, UK). Slits confining the light path were opened to 2-3 nm bandwidth. Polarizers in magic angle setting were not used since stray light contributions in this wavelength range were negligible. For emission, GFP was excited at 480 nm and detected in the range between 490-700 nm, for kinetic experiments exclusively at 510 nm. The Raman contribution from a buffer measurement (S/N > 20) was subtracted. All measurements were performed at 22° C

1.7 GeIs

GFP-C-His was displayed with SDS-PAGE using 4-12% Bis-Ths Novex precast gels (Invitrogen) run in IxMES buffer for 1 hour at constant 150 V. The gel was imaged either directly with UV excitation (365 nm) or white light after Coomassie staining (Invitrogen) using a BioSpectrumAC imaging system with integrated quantification software (UVP Cambridge UK). 1.8 Cloning

Mammalian expression vectors were produced integrating the coding sequence of human IL-4Rα into pEGFP-N1 from Clontech via directional cloning. The hexahistidine-tag was inserted by site directed mutagenesis {GeneTailor, Invitrogen). The expressed polypeptide NHis-IL-4Rac-GFP comprises the extracellular His-tagged IL-4Rα (residues 33-297 correspond to amino acids 26-292 Swiss-Prot P24394 with 6 histidine residues inserted after the leader sequence at positions 27-32), a short linker GSTGRH at positions 298-303, and eGFP amino acids 304-542 corresponding to GeneBank Accession No. U55762 (amino acids in one-letter code, numbering from N to C-terminus). As a control IL-4Rac-GFP, the receptor described above lacking the hexahistidine sequence, was used. 3-4 days after transfection, the receptor constructs were properly expressed and transported to the surface plasma membrane.

1.9 Cell maintenance

HEK cells were grown in DMEM H21 (13.38 g/l DMEM Powder (Gibco), 44 mM NaHCO₃ {Gibco), 50,000 IE Penicillin (30 mg) (BC), 50 mg Streptomycin (Gibco), 2 mM glutamine, 10% complement inactivated fetal calf serum). For passaging, the cells where washed twice with PBS (137 mM NaCI, 1.5 mM KH₂PO₄, 2.7 mM KCI, 8 mM Na₂HPO₄ ^*2H₂O), harvested with Cell Dissociation Solution (CDS, Sigma, Cat. No. C5914), diluted by PBS in a 50 ml Falcon tube, and pelleted at 300xg for 5 min at room temperature. New flasks were seeded 1 :10 and 1 :20 two times a week and maintained at 37° C under 5% CO₂ in a humidified atmosphere. 1.10 Transfection and preparation of the cells

Confluent HEK 293 cells were transfected into a 12-well (4 cm² surface area per well, Falcon) via Lipofectamine 2000 {Invitrogen, Cat. No. 11668-019) according to the manufacturer's recommendation. After 12 hours, the cells were transferred into Fibronectin-coated chamber slides (Nunc LAB-TEK 8- well, Cat. No. 155409). Coating was performed by covering the glass with 100 μg/ml Fibronectin (BD, Cat. No. 356008), followed by incubation for 30 min at 37° C. After aspiration, the cells were directly seeded at a density of 5,000 to 10,000 cells per well (1 cm² surface area per well). Within the time range of 48 hours the cells were subconfluent and sufficiently adherent to apply the staining and crosslinking procedure.

1.11 Confocal microscopy

For imaging and FCS, a proprietary Novartis PS03 microspectroscope was used. In a joint development effort with Evotec Technologies (Hamburg, Germany), the IX70 microscope had been modified to provide high-resolution confocal laser scanning functionalities and single-photon-sensitive signal acquisition via two fluorescence emission channels. The piezo-based 3D scanning enhancement is commercially available under the proprietary name Insight Cell {Evotec Technologies).

In our setup, either argon ion (488 nm) or helium/neon (543 nm) laser light was fiber-coupled into the microscope (IX70, Olympus). The excitation beam passes the beam splitting plate (reflection 10%) and is directed onto the back aperture of a high-numerical objective (Olympus Uplan 6Ow NA1.2). The generated fluorescence is collected by the same objective, passes the beam splitting plate (transmission 90%) and focused by the tube lens onto the confocal pinhole (40 μm). In the detection path, a color splitting dichroic mirror (550 DLRP, Omega Optical, Brattleboro, Vt) splits the fluorescence into the GFP- (band pass 515DF30) and TMR-specific (band pass 585DF20) color channels, which are separately detected by single-photon sensitive avalanche photodiodes (SPMC-AQR-13-FC, Perkin Elmer). For imaging, a laser power between 10 and 20 μW (488 nm) and 40-60 μW (543 nm) was applied.

1.12 Image evaluation

The binary image files generated by the MIPS software (Evotec Technologies, Hamburg, Germany) contain the number of photon counts accumulated within the residence time of the laser focus (pixel time, PT) at a certain position in the sample. The binary files are a proprietary image format and have been converted into TIF images by a script (Acapella, Evotec Technologies). The PTs used for imaging were 0.5 or 1 ms. Thus, dividing the counts for each pixel by PT in ms, converts the absolute number of photons to kHz intensities. For converting these kHz intensities into absolute particle numbers the molecular brightness (kHz per particle) for free eGFP and NTA-ASA-TMR (34) was determined by fluorescence correlation spectroscopy (FCS) in medium (Figure 15 a, b, Figure 16 a). The molecular brightness of the eGFP tagged receptors and free eGFP is the same (determined by FCS measurements at the membrane and in the cytoplasm of transfected cells (not shown). Therefore, dividing the kHz intensities of each pixel by the molecular brightnesses of eGFP and NTA-ASA-TMR (34) yielded absolute particle numbers for each pixel. A background due to autofluorescence in the medium was taken into account, according to N_corr = G(0)*(1 +U/F)^"2, where G(O) is the intercept of the correlation curve, U the background intensity for a certain laser power and F the average background subtracted signal (Figure 16 b). Dividing the intensity images by the molecular brightness returns particles per pixel. A typical volume of 0.25 femtoliter was assumed as a good estimate for the effective observation volume based on FCS of a fluorescent standard: AlexaFluor488 in PBS showed a translational diffusion time of id = 40 μs; using the diffusion coefficient for Fluorescein D_t ~ 2.6^*10^"10 m²/s, axis ratio of K = 5 and V_eff = K (4 π τd Dt)^3/2. The confocal volume of the red shifted 543 nm laser line was assumed to scale with a factor of 1.37 = (543/488)³.

The average concentrations at the surface membrane were determined with a proprietary image calculation script (MatLab Version 7.2.0.232 (R2006a), MathWorks Inc., USA). The region of interest (ROI) was segmented by threshold settings and manual corrections; a global background fluorescence from the medium was subtracted.

2 Results

2.1 Synthesis of the tris- and mono-NTA derivatives

The syntheses of ths-NTA and mono-NTA derivatives were carried out as described in Figures 4 and 5.

Briefly, the synthesis of the ths-NTA consists of coupling 3 carboxy functionalized NTA moieties, prepared by alkylation of the amino function of a protected glutamate derivative, to the amino groups of a cyclam scaffold. To the last amino group, an amino caproic spacer was attached before the coupling of the different dyes (the QSY7, the TMR and the Rhodamine Red

X) was performed. After complexation with Ni²⁺, the compounds were purified by reparative HPLC and analyzed by MS.

The mono-NTA was obtained starting from the te/t-butyl protected NTA (3)

(described in Figure 4). After an amidation with a benzylethane-1 ,2-diamine, the amine was deprotected by hydrogenolysis and coupled to a Z-protected lysine. After hydrogenolysis, the dye or the photo-crosslinker were introduced using succinimidyl activated ester, while the acryloyl functionality was introduced using the acyl chloride form. Following the removal of all the protecting groups with TFA, the last free amine was coupled either with a dye or photo-crossl inker, or with acryloyl and purified by preparative HPLC. Finally, the mono-NTA moiety was complexed with Ni²⁺, purified by preparative HPLC and analyzed by MS.

2.2 Selection of the dyes

Among the wide variety of amine reactive fluorescent dyes and quenchers, QSY7, TMR and RhB were chosen (Figure 6). QSY7 is a non-fluorescent acceptor dye and TMR and RhB are fluorescent rhodamine derivatives with similar absorption characteristics.

The non-fluorescent quencher that strongly absorbs light in the visible spectrum is convenient to assess binding to a His-tagged recombinant eGFP (GFP-C-HiS₆) via FRET (Lakowicz 1999) mediated donor quenching. TMR was chosen to investigate the possibilities to label His-tagged proteins in vitro or on cells. Rhodamine B was chosen to test a cellular assay system addressing the conformational change of a cell surface expressed lnterleukin-4 receptor (IL-4R). The brightness of Rhodamine B is known to be sensitive to the polarity of the local environment. Moreover, protein conjugates of the Rhodamine Red-X dye are frequently brighter (than those of Lissamine Rhodamine B) and are less likely to precipitate during storage.

2.3 Synthetic strategy for irreversible labeling of proteins

The combination of a metal-ion-chelating moiety (NTA) exerting a directing effect through reversible binding with a photoreactive functionality (e.g. azides) or a weak electrophile (e.g. an acryloyl) should allow a covalent linkage between the probe and the (His)β-tagged protein (Figure 7). A hydroxy-aryl azide, namely azido salicyclic acid (ASA) (Figure 8) was chosen as photo-crossl inker. Upon illumination (< 380 nm), aryl azides generate radicals (very reactive intermediates) that can react with nucleophiles present in the protein sequence (e.g OH, NH, SH) to form a covalent. Hydroxy aryl azides do not react via the short lifetime singlet nitrenes but through an azacycloheptatetrane intermediate formed by very rapid intramolecular rearrangement of the nitrenes (Brunner 1993) (Figure 8).

This cycle is highly electrophilic but less reactive than a typical nitrene and does not insert non-activated C-H bonds.

Besides of the photoreactive probe, which was shown to be successful as crosslinking reagent, a new approach based on proximity acceleration of reactions was investigated.

It was shown (Chmura, Orton and Meares 2001 ; Levitsky, Ciolli and Belshaw

2003) that by engineering receptor ligands with complementary reactive groups a covalent bond can be formed via proximity accelerated reaction through the formation of the receptor-ligand complex (Figure 9). Unwanted reactions with other molecules are avoided due to the low reactivity of these groups. Local concentrations of the complementary reactive groups in the receptor/I igand complex are huge (>10⁵ M) therefore allowing the reaction to take place. This coupling technique should be applicable to the present invention by using the NTA-(His)β complex as "receptor-ligand" and an acryloyl functionality as reactive group.

The acryloyl group is a weak electrophile (Michael acceptor), unreactive toward the nucleophiles present in the protein (e.g NH, OH, SH) unless high local concentrations of reagents accelerate this reaction.

In addition to the reaction-accelerating effect, the NTA moiety permits to localize the covalent bond on the protein terminus where the histidine sequence is located. Especially for the photo-crossl inkers this property is important as otherwise covalent bonds might be formed somewhere on the peptidic chain during illumination. This would lead to a loss in site selectivity of labeling.

3 Discussion

3.1 Chemical synthesis

The synthesis of the core structure of the tris- and mono-NTA worked without major problems. The efficiency of the coupling reaction of 3 carboxy functionalized NTA moieties to the cyclam scaffold bearing 4 free amines was surprising. Generally, the coupling of secondary amines can be critical due to steric hindrance. However, in our reaction, the expected product was obtained with a high yield and without major side products of mono, bis or tetra substituted derivatives.

The labeling reaction required a purification step which is caused by the presence of a large excess of dye and to the fact that labeled compounds are often not detected by MS in the reaction medium. However, all the desired compounds were produced and characterized.

3.2 Spectroscopic characterization of high affinity tags

In the following, the spectroscopic characterization of the produced prototype His-tag binding probes is only described exemplarically to demonstrate their functionality. The binding characteristics of the ths-NTA probes (10, 11 , 12) to (His)6-tagged proteins was characterized by ensemble averaging fluorescence spectroscopy and confocal fluorescence microscopy. The activity of the dye was tested in vitro with a FRET-based assay as published (Guignet, Hovius and Vogel 2004). In brief, a recombinant GFP carrying a hexahistidine stretch at the C-terminus (GFP-C-His) was mixed with the tris- NTA probe. Binding of the compound was observed via GFP-fluorescence (donor quenching) or an increase in anisotropy for the TMR conjugated probe (10). From both methods a K_d = 7±3 nM was derived. Due to the cooperativity of the binding reaction, the kinetics can be slow compared to a purely diffusion controlled reaction. Also the kinetics varied strongly among the ths-NTA derivatives (10, 11 , 12).

Competition experiments with EDTA confirmed that the GFP-C-His - tris-NTA probe complex was stable with a small off-rate allowing separation steps for the conjugated protein to be performed for several hours.

Figure 10 shows fluorescence spectra before (black) and after (green) addition of non-fluorescent ths-NTA-Ni²⁺-QSY7 (11). Within a few seconds, the probe quenches the GFP-C-His fluorescence to ~ 15%. Binding was reversed by a large excess of imidazole, the functional moiety of histidine (red).

To prove the FRET-origin of the observed GFP-quenching, the identical experiment was performed with the fluorescent tris-NTA-TMR tag (10). Figure 11 shows an orthogonal setting in which compound (acceptor) is added to the GFP-C-His (donor) and vice versa leading to the same concentrations in the mixtures. The spectra of the two mixtures show a good overlay and reflect both donor quenching and sensitized emission. The data rule out non-specific quenching effects and prove the binding of the compound.

To demonstrate the functionality of the tris-NTA-Ni²⁺-TMR reagent (10) for visualization of cell surface receptors, 500 nM of the tracer was applied to the supernatant of HEK293 cells expressing a His-tagged and GFP conjugated IL-4R alpha chain construct (Figure 12). Confocal images were taken at 488 nm excitation (green) and at 543 nm excitation (red) nicely demonstrating specific binding and the cell impermeability of the membrane toward NTA probes.

3.3 The irreversible X-linking probe Tris-NTA probes bind the His-tag with higher affinities compared to the mono-NTA tags and are characterized by slow dissociation processes in the time range of minutes to hours. Mono-NTA tags, however, bear the advantage of reduced binding confirmation possibilities and "cleaner" binding events. Therefore, in an effort to expand the spectrum of His-tag directed probes, the reversible mono-NTA tracer was used to add a photo-crosslink functionality (illumination at 365 nm). Covalent conjugation extends the application range for His-tag binding probes to long term observations of cellular processes. It might also provide an alternative protein labeling strategy for proteins in vitro. Labeling can be performed in a broad range of buffers followed by purification steps. Moreover, covalent conjugation is a prerequisite for single molecule spectroscopy applications in very low concentrations (pM to nM). The system ensures a homogeneous 1 :1 labeling stoichiometry of observed proteins which facilitates the calculation of quantitative data.

3.4 In vitro labeling results

Binding and fluorescence properties of the new labeling reagent were investigated by conventional fluorescence spectroscopy. We used a fluorescence resonance energy transfer (FRET) based assay system as described by (Guignet, Hovius and Vogel 2004). A purified recombinant GFP with a C-terminal His-Tag, GFP-C-His, served as a FRET donor for NTA- ASA-TMR (34) or NTA-ASA-QSY7 (35) as acceptors. Quenching of the donor due to energy transfer directly reflects the fraction of complexed GFP- C-His (Figure 13 a). The affinity was determined by measuring the degree of donor quenching under equilibrium conditions for increasing ligand concentrations (Figure 13 c). NTA-ASA-TMR (34) bound to GFP-C-His with an equilibrium dissociation constant of K_d = 1 ± 0.07 μM. For comparison, the Kd values of previously described compounds NTA-I were also determined (Figure 17). Without photoactivation, binding of probes 34 and 35 to oligohistidine-tagged proteins is reversible and mediated by a d⁸ coordinated Ni². Thus, EDTA, added in excess (> 250-fold), competes for free Ni²⁺-ions, and binding is successively reversed. The kinetics of this competition reaction critically depended on the substituents of the NTA. For example, unquenching was complete for NTA-I after 5 min, for NTA-ASA-TMR (34) after 10 min, and the release of NTA-ASA-QSY7 (35) took more than one hour under similar conditions (Figure 13 b). The amount of fluorescence recovered by EDTA was about 90% for NTA-I, whereas for the NTA-ASA derivatives only 60-70% was recovered. The data suggest that a large excess of NTA-ASA probes 34 and 35 vs. GFP-C-His in this assay (1 ,000- fold) promotes residual His-tag independent binding. In contrast to dissociation reaction, the binding of the compounds is completed within several seconds. To finally demonstrate the covalent linkage, we included a photo-activation step between binding and EDTA-mediated release of the NTA-ASA probes 34, 35. As expected, the UV-irradiated samples showed only 10% EDTA-mediated fluorescence recovery as compared to non- irradiated controls (Figure 13 a). It is noteworthy that mock photo-activation (120 mJ) of NTA-ASA-TMR 5 in the absence of GFP-C-His protein reduces the fluorescence to about 55%, whereas free TMR-COOH control showed no bleaching. We therefore assume that the activated arylazide may, to a certain extent, self-react and thereby destroy the integrity of the conjugated π- system.

To prove that photo-activation establishes a stable covalent linkage, SDS- PAGE was used to remove non-conjugated TMR-NTA-ASA (34) under harsh conditions. GFP-C-His (1.25 μM) was incubated with a 4-fold excess of NTA- ASA-TMR (34), photo-activated under several conditions and loaded on the gel (Figure 13 d). The gel clearly shows that fluorescent compound 34 associates with the GFP band. A non-activated control, as well as an EDTA quenched sample, showed only weak fluorescence. Bound and free compound was quantified by digital image analysis. The fractional fluorescence intensity associated with the GFP-band versus free dye is 9% for a dosage of 120 mJ and 15% for 240 mJ. Further increase of the dosage had no effect on increase in complexation. Neglecting the small background of protein impurities running at higher molecular weights, we recalculated these values with respect to the fraction of bound ligand, which resulted in a crosslink efficiency of 45% (120 mJ) and 75% (240 mJ) at the binding site.

3.5 Cellular labeling

To test covalent His-tag mediated labeling of a membrane receptor via NTA- ASA-TMR (34) at the cellular surface, we transiently expressed a His-tagged lnterleukin-4 receptor (IL-4R) in human embryonic kidney cells (HEK 293 T). The receptor construct (NHis-IL-4Rac-GFP) comprises a hexahistidine stretch at the N-terminus, followed by the extracellular and transmembrane domains of the lnterleukin-4 receptor α chain. The cytoplasmic tail of the receptor was replaced by eGFP. In this configuration, the His-tag is expressed extra-cellularly, while the GFP is located in the cytosol. The double-tagged receptors allow estimating crosslinking efficiencies, provided that receptor and ligand densities are observed in orthogonal color channels.

For receptor labeling, the cells were washed, incubated with NTA-ASA-TMR (34) (0.5-2.6 μM in PBS⁺⁺), UV-irradiated for 40 sec applying 120 mJ, and washed several times with culture medium. Figure 14 a-f illustrates confocal images of a transfected cell in a layer of non-transfected neighbors. In the GFP-channel only the transfected cell is visible. The fluorescent receptors are clearly localized in the membrane. Typical for high expression levels, the unprocessed receptors accumulate in intracellular membrane systems. The TMR color channel nicely shows that staining of transfected cells with NTA- ASA-TMR (34) is restricted to the surface membrane suggesting that the labeling reagent is not cell penetrating. On cells expressing a control vector lacking the His-tag, IL-4Rac-eGFP did not accumulate TMR-fluorescence over background (Figure 14 g-i).

To assess cellular IL-4R labeling quantitatively, we calibrated the confocal images using a previously described method (Weidemann, et al 2003). In brief, we determined the molecular brightness of eGFP and NTA-ASA-TMR (34) with fluorescence correlation spectroscopy (FCS) in the same optical setup as used for confocal imaging. With the numbers obtained by FCS, the pixel intensities were converted into particle concentrations. From the concentration maps, average values for eGFP and NTA-ASA-TMR (34) in the surface membrane were extracted with a proprietary script (MatLab). According to Table 1 , the transfected cells express lower micromolar concentrations of receptor. Comparison with the TMR channel shows that 10.4% (cell 1 ) and 4.2% (cell 2) of the receptors were labeled. With 100-110 nM concentration the background staining of the neighboring cells is not negligible. In both single cell experiments, a signal-to-noise ratio of 5 was achieved. Assuming a one-to-one binding model with Kd = 1 μM, we calculated the fraction of complexed receptor under equilibrium conditions to be 9.2% and 4.6% for the receptor concentrations determined at the cell surface of cell 1 and cell 2, respectively (Figure 18). These values are in remarkable agreement compared to what was determined from the images. In conclusion, labeling reached a significant S/N level for cells expressing micromolar concentrations of His-tagged receptor. Lower surface expression did not accumulate surface staining above background

Table 1 Concentrations of NHis-IL4Rac-eGFP and NTA-ASA-TMR on the surface of transfected and control cells cell # NHis-IL4Rac-eGFP NTA-ASA-TMR NTA-ASA-TMR S/N at the cell surface [μM] at the cell surface [μM] at the surface of non- transfected cells [μM]

1 ^a 4.3 0.56 0.11 5.0

2 ^a 9.8 0.51 0.10 5.2

3 ^b 6.6 0.58 0.48 1.2

Incubation with 0.5 μM NTA-ASA-TMR (34); b Incubation with 2.6 μM NTA-ASA-TMR (34)

References

Adams SR, Campbell RE, Gross LA, et al (2002) New biarsenical ligands and tetracysteine motifs for protein labeling in vitro and in vivo: Synthesis and biological applications. J.Am.Chem.Soα; 124 (21 ):6063-76.

Brunner J (1993) New photolabeling and crosslinking methods. Annu.Rev.Biochem.; 62:483-514.

Chmura AJ, Orton MS, Meares CF (2001 ) Antibodies with infinite affinity. Proc.Natl.Acad.Sci. U.S.A.; 98 (15):8480-4.

Cronan JE Jr (1990) Biotination of proteins in vivo. A post-translational modification to label, purify, and study proteins. J.Biol.Chem.; 265 (18):10327-33.

Griffin BA, Adams SR, Tsien RY (1998) Specific covalent labeling of recombinant protein molecules inside live cells. Science; 281 (5374):269-72. Guignet E, Hovius R, Vogel H (2004) Reversible site-selective labeling of membrane proteins in live cells. Nat.Biotechnol.; 22 (4):440-4.

Harms GS, Cognet L, Lommersee PH, et al (2001 ) Autofluorescent proteins in single-molecule research:applications to live cell imaging microscopy. Biophys.J.; 80 (5):2396-408.

Hochuli E, Dobeli H, Schacher A (1987) New metal chelate adsorbent selective for proteins and peptides containing neighbouring histidine residues. J.Chromatogr.; 411 :177-84.

Hovius R, Vallotton P, Wohland T, et al (2000) Fluorescence techniques: shedding light on ligand-receptor interactions. Trends Pharmacol. Sci.; 21 :266-73.

Keppler A, Gendreizig S, Gronemeyer T, et al (2003) A general method for the covalent labelling of fusion proteins with small molecules in vivo. Nat.Biotechnol.; 21 (1 ):86-9.

Lakowicz JR (1999) Principles of Fluorescence Spectroscopy. 2nd ed; New York: Plenum Publishers, p. 367-94.

Lata S, Gavutis M, Tampe R, et al (2006) Specific and stable fluorescence labeling of histidine-tagged proteins for dissecting multi-protein complex formation. J.Am.Chem.Soα; 128 (7):2365-72.

Lata S, Reichel A, Brock R, et al (2005) High-affinity adaptors for switchable recognition of histidine-tagged proteins. J.Am.Chem.Soc; 127 )29):10205- 15.

Levitsky K, Ciolli CJ, Belshaw PJ (2003) Selective inhibition of engineered receptors via proximity-accelerated alkylation. Org.Lett; 5 (5):693-6. Meyer BH, Martinez KL, Segura JM, et al (2006) Covalent labeling of cell- surface proteins for in-vivo FRET studies. FEBS Lett.; 580 (6):1654-8.

Porath J, Carlsson J, Olsson I, et al (1975) Metal chelate affinity chromatography, a new approach to protein fractionation. Nature; 258 (5536):598-9.

Tsien RY (1998) The green fluorescent protein. Annu.Rev.Biochem.; 67:509- 44.

Weidemann T, Wachsmuth M, Knoch TA, et al (2003) Counting nucleosomes in living cells with a combination of fluorescence correlation spectroscopy and confocal imaging. J. MoI. Biol.; 334 (29):229-40.

Wood RJ, Pascoe DD, Brown ZK, et al (2004) Optimized conjugation of a fluorescent label to proteins via intein-mediated activation and ligation. Bioconjug.Chem.; 15 (2):366-72.

Zhang Y, So MK, Loening AM, et al (2006) HaloTag protein-mediated site- specific conjugation of bioluminescent proteins to quantum dots. Angew.Chem.lnt.Ed.; 45 (30):4936-40.

Claims

Claims

1. A compound of formula (I)

wherein L1 is a linker,

L₂ and L₃ are independently chemical bonds or linkers, F is a reporter moiety, C is a covalent coupling moiety,

NTA is a nitrilo thacetic moiety and n is from 1 -3, or a salt, chelate, or derivative thereof.

2. The compound of claim 1 , wherein F is a fluorescent group.

3. The compound of claim 1 , wherein F is a quencher group.

4. The compound of any one of claims 1 -3, wherein C is a photoactivatable group.

5. The compound of claim 4, wherein C is an aryl ketone, diazo, diazirene or azide group.

6. The compound of claim 5, wherein C is a hydroxy aryl azide group.

7. The compound of any one of claims 1 -3, wherein C is a non-photoactivatable group.

8. The compound of any one of claim 7, wherein C is an acryloyl group.

9. The compound of any one of claims 1 -8 of formula (Ia):

wherein ni and n₂ are independently from 1 -6, n₃ and n₄ are independently from 0-6, and F and C are as defined in claims 1-8, or a salt, chelate, or derivative thereof.

10. A covalent conjugate of a compound of any one of claims 1 -9 with a protein.

11. A method of manufacturing a compound of formula (I)

wherein Li is a linker, L₂ and L₃ are independently chemical bonds or linkers,

F is a reporter moiety, C is a covalent coupling moiety, NTA is a nitrilo thacetic moiety and n is from 1 -3, or a salt, chelate protected derivative or synthesis intermediate thereof, comprising the steps:

(i) reacting a compound of formula (II):

(PNTA)_n - L₁ ¹ - COOH

wherein pNTA is a protected nitrilo triacetic group, n is 1 -3, and Li is a linker, with a compound of formula (III):

wherein Pi is an amino protection group, and Li" is a linker, and subsequently removing the amino protection group Pi, to obtain a compound of formula (IV):

O

(pNTA)n L₁ ' U NH L₁ "-NH₂

wherein pNTA, n, L₁' and L₁" are as defined above,

(ii) reacting the compound of formula (IV) with a compound of formula (V)

wherein L₂ and L₃ are independently chemical bonds or linkers and P₂ and P₃ are amino protection groups, wherein P₂ is different from P₃, to obtain a compound of formula (Vl):

wherein pNTA, n, L₁', L₂", L₂, L₃, P₂ and P₃ are as defined above, and

(iii) (a) functionalizing the group -NHP₂ with a reporter moiety F to obtain the group -NHF, and

(b) functionalizing the group -NHP₃ with a covalent coupling moiety C to obtain the group -NHC, wherein a compound of formula (I) is obtained.

12. A method of irreversibly conjugating a compound of formula (I) to a protein, comprising the steps:

(i) providing a chelate of a compound of formula (I),

(ii) reversibly binding said chelate to a protein,

(iii) irreversibly coupling said chelate via its covalent coupling moiety to the protein.

13. The method of claim 12 wherein said protein comprises a poly-histidine sequence.

14. The method of claim 12 or 13 wherein said irreversible coupling comprises an illumination step.