The synthesis of biologically relevant conjugates of Re(CO)₃ using pyridine-2-carboxyaldehyde

Abstract

The new pyridine-2-carboxaldehyde adduct, Re(CO)₃(NC₆H₅C(O)H)Cl 1, and previously reported complex Re(CO)₃(NC₆H₅C(O)H)Br 2 react with aniline derivatives sulfanilamide or 4-aminofluorescein in methanol giving Schiff base conjugates Re(CO)₃(pyca-R)X (pyca = pyridinecarbaldehyde imine, X = Cl, Br), 3–6. Pre-isolation of compounds 1 and 2 provides a convenient method for preparing conjugate complexes in addition to the known methods of ligand synthesis and one-pot reactions. All new compounds were completely characterized, and compound 1 and the sulfanilamide derivatives 3 and 4 were structurally elucidated by X-ray crystallography.

1. Introduction

The biological chemistry of organometallic rhenium compounds has attracted research interest in recent years [1–3]. Initially, the goal was to examine cold analogs of technetium-99m, which is the most commonly used nuclide in nuclear medicine with 50,000 clinical applications per day in the US [4]. The creation of a kit for the preparation of ^99mTc(CO)₃(H₂O)⁺ [5], opened up the possibility of ^99mTc(CO)₃⁺ radiopharmaceuticals [6]. As technetium has no stable nuclides, exploring the similar chemistry of its higher congener is a convenient method of testing new targeting strategies for researchers searching for target specific radiopharmaceuticals. More recently, the field has been expanded by the development of a carrier-free source of rhenium-188 (half-life of 17 hours with β⁻ = 2.1 MeV) and the production of a clinically appropriate method for the preparation of ¹⁸⁸Re(CO)₃(H₂O)₃⁺ [7–9]. The possibility of creating a matched pair of radiopharmaceuticals—one for diagnosis and one for therapy—that would target the same location is an attractive goal for clinical medicine [10]. Similarly, d⁶ rhenium compounds have been tested as in vivo luminescence agents. Many of the compounds examined for this purpose contain polypyridyl units or dyes as ligands [11–13].

We and others have frequently made use of Schiff base condensation reactions to create d⁶ rhenium compounds with diimine-based bioconjugates formed by the reaction of pyridine-2-carboxaldehyde with amino acid esters [14, 15], amino acids [16], peptides [17] or amino alcohols [18]. Reactions of Re(CO)₃⁺ precursors and pyridine-2-carboxaldehyde with amino acid esters yielded discrete mononuclear complexes, Re(CO)₃(pyca-Xxx-OR)Cl (pyca = pyridinecarbaldehyde imine, Xxx = amino acid), while amino acids or peptides yielded C₂-symmetric dimers, [Re(CO)₃(pyca-Xxx-O)]₂ or complex structures with extended hydrogen bonding networks. After the application of a strong base, amino alcohol conjugates produced a dimetallacycle with a proton strongly bonded between the two alkoxo oxygen atoms.

These metal Schiff base compounds have typically been made in one-pot reactions [16], as the iminopyridine ligands are not easy to isolate and use in further reactions. With the report on the synthesis of 2, a bidentate adduct of pyridine-2-carboxaldehyde, we began looking to explore this reagent to expand the scope of amines used to create d⁶ rhenium Schiff base complexes. We chose to focus on examining the reactivity of 2 and its chloride analog with substituted anilines, inspired by the work of Miguel and co-workers [19].

There has been a significant amount of work on the synthesis and characterization of N-phenyl-2-pyridylimino Re(CO)₃ complexes. Since the early 1980’s, both the chloride and the bromide variants of the unsubstituted N-phenyl-2-pyridylimino compounds have been prepared, typically by first separately synthesizing the pyridyl imine followed by chelation to the metal ion [20–22]. Alternatively, one-pot methods can be used; we employed such a method to generate Re(CO)₃ complexes incorporating pyca-amino acid conjugates. Using the older method of pre-generating the pyridyl imine ligand, a variety of functional groups can be introduced on the phenyl ring, including alkyl groups [21, 23, 24], hydroxides [25], halides [26, 27], additional phenyls [23, 28], and crown ethers [29, 30]. Miguel used compound 2 to not only generate the phenyl species, but also analogs with carboxylic acids and alkynes substituted on the phenyl ring [19, 31].

For this study, we chose to prepare complexes prepared from sulfanilamide and 4-aminofluorescein. Sulfonamide compounds have found use as inhibitors of carbonic anhydrase, and thus have been used as therapeutics for glaucoma, hypertension, and for use as diuretics [32]. The discovery of isozymes of carbonic anhydrase that are overexpressed in certain tumors has also encouraged researchers to create new sulfonamide compounds with applications such as targeting by luminescence or γ-emission [33, 34]. The publication of the crystal structure of carbonic anhydrase with the metal complex, Cp^RRe (CO)₃ (R = arylsulfonamide) bound to the zinc ion at the active site highlights the interest in this area [35, 36]. Although 4-aminofluorescein is not fluorescent, the parent fluorescein molecule finds heavy use in cellular imaging, and the development of Re(CO)₃⁺ based luminescent molecules has been a goal of many research groups [11–13]. We were curious whether incorporating the exocyclic amine of 4-aminofluorescein into a metal-bound diimine ligand would return the fluorescent properties of this chromophore.

In this report we present the reaction of sulfanilamide and 4-aminofluorescein with Re(CO)₃(C₅H₄C(O)H)X; X = Cl, Br (1 and previously prepared 2, respectively) to form new compounds 3–6 (see Scheme 1 for numbering of compounds). The structures of both sulfanilamide products, 3 and 4, as well as 1, were elucidated by X-ray crystallography. Compounds 5 and 6 each regained fluorescence intensity upon incorporation of the exocyclic amine into the rhenium-bound diimine. The new pyridine aldehyde complex 1 and all of the new pyridyl imine compounds are completely characterized.

Scheme 1.

General reaction scheme and numbering system for compounds 1–6.

2. Results and Discussion

2.1. Syntheses

The synthesis of compound 2 using an extended reflux of the aldehyde and one equivalent of Re(CO)₅Br in hexane was first reported by Miguel and coworkers [15]. We found that the synthesis also works well in toluene, producing pure product that can be isolated and stored as a reagent for later reactions. We also prepared the new compound 1 by this same method. Both complexes were prepared in yields of greater than 80%. Compounds 3–6 were prepared as orange solids by methanol reflux of the appropriate aldehyde complex with one equivalent of the amine. Sulfonamide derivatives 3 and 4 were obtained as pure compounds by slow evaporation of the reaction solution. Fluorescein conjugate compounds 5 and 6 were insoluble. After precipitating from the solution during the reaction, they were washed and dried. Single crystals of 1, 3 and 4 were grown using either diffusion of hexane into a CH₂Cl₂ solution in the case of 1 or by slow evaporation of a methanol solution in the cases of 3 and 4.

Preparations of all reported compounds are insensitive to the presence of atmospheric oxygen and water vapor. No special precautions were taken in these syntheses.

2.2. Description of Crystal Structures

The structures of compounds 1, 3 and 4 were elucidated by single crystal X-ray diffraction. The molecular structures are shown in Figures 1 and 2. Crystal data and experimental details are listed in Table 1. Table 2 lists some of the key bond lengths and angles in these complexes. All of the complexes have slightly distorted octahedral geometries, with the three carbonyls arranged in a facial configuration as expected for d⁶ Re(CO)₃ compounds.

Figure 1.

The structure of compound 1 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity.

Figure 2.

The structures of 3 (left) and 4 (right) with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity.

Table 1.

X-ray data collection and structure parameters for compounds 1, 3, and 4. Measurements were made at 100(2)K using Mo K_α radiation (λ = 0.71073 Å).

Compound	1	3	4
Empirical formula	C9H5ClNO4Re	C16H10ClN3O6ReS	C15H11BrN3O5ReS
Formula weight	412.79	593.98	611.44
Crystal system	Monoclinic	Monoclinic	Rhombohedral
Space group	P2(1)/c	C2/c	R-3
Unit cell dimensions	a = 10.533(5) Å α= 90°	a = 25.649(13) Å α= 90°	a = 36.910(5) Å α= 90°
	b = 8.880(4) Å β= 91.130(7)°	b = 10.666(6) Å β= 91.639(6)°	b = 36.910(5) Å β= 90°
	c = 11.506(5) Å γ = 90°	c = 14.356(8) Å γ = 90°	c = 8.0828(10) Å γ = 120°
Volume	1076.0(8) Å3	3926(4) Å3	9536(2) Å3
Z	4	8	18
Density (calculated)	2.548 mg/m3	2.010 mg/m3	1.916 mg/m3
Absorption coefficient	11.537 mm−1	6.471 mm−1	7.745 mm−1
F(000)	760	2264	5184
Crystal size	0.17 x 0.14 x 0.09 mm3	0.10 x 0.10 x 0.03 mm3	0.30 x 0.08 x 0.04 mm3
Theta range for data collection	1.93° to 26.99°	1.59° to 27.57°	1.91° to 27.52°
Index ranges	−13<=h<=13, −11<=k<=11, −14<=l<=14	−32<=h<=33, −13<=k<=13, −18<=l<=18	−47<=h<=47, −47<=k<=47, −10<=l<=10
Reflections collected	8689	15759	26228
Independent reflections	2352 [R(int) = 0.0582]	4418 [R(int) = 0.0755]	4851 [R(int) = 0.0550]
Completeness to theta	99.9%	97.1%	99.1%
Absorption correction	SADABS	SADABS	SADABS
Max. and min. transmission	0.4233 and 0.2445	0.8295 and 0.5639	0.7470 and 0.2047
Refinement method	Full-matrix least-squares on F2	Full-matrix least-squares on F2	Full-matrix least-squares on F2
Data / restraints / parameters	2352 / 0 / 145	4418 / 0 / 256	4851 / 0 / 236
Goodness-of-fit on F2	0.91	1.001	0.923
Final R indices [I>2sigma(I)]	R1 = 0.0294, wR2 = 0.0574	R1 = 0.0461, wR2 = 0.1012	R1 = 0.0402, wR2 = 0.1189
R indices (all data)	R1 = 0.0395, wR2 = 0.0585	R1 = 0.0614, wR2 = 0.1107	R1 = 0.0496, wR2 = 0.1274
Largest diff. peak and hole	2.231 and −0.935 e.Å−3	1.863 and −2.189 e.Å−3	1.653 and −1.334 e.Å−3

Table 2.

Selected bond lengths (Å) and bond angles (°) for 1, 3, and 4.

	1	3	4
Bond lengths
Re(1)-C(1)	1.896(7)	1.916(8)	1.926(7)
Re(1)-C(2)	1.903(6)	1.909(8)	1.922(6)
Re(1)-C(3)	1.908(7)	1.935(8)	1.918(6)
Re(1)-N(1)	2.177(5)	2.187(6)	2.170(5)
Re(1)-Cl(1)	2.4792(17)	2.499(2)
Re(1)-Br(1)			2.6294(7)
Re(1)-O(4)	2.172(4)
Re(1)-N(2)		2.161(6)	2.187(6)
C(9)-O(4)	1.241(7)
C(9)-N(2)		1.278(10)	1.303(9)
Bond angles
C(1)-Re(1)-C(3)	90.1(3)	90.1(3)	89.1(3)
C(1)-Re(1)-C(2)	89.7(2)	88.1(3)	86.9(3)
C(3)-Re(1)-C(2)	87.5(2)	88.9(3)	90.3(3)
C(3)-Re(1)-O(4)	97.6(2)
C(2)-Re(1)-O(4)	91.52(19)
C(3)-Re(1)-N(2)		96.7(3)	91.7(2)
C(2)-Re(1)-N(2)		97.2(3)	100.8(2)
C(2)-Re(1)-N(1)	92.4(2)	94.2(3)	94.3(2)
C(1)-Re(1)-N(1)	97.4(2)	98.1(3)	97.5(2)
O(4)-Re(1)-N(1)	74.91(17)
N(2)-Re(1)-N(1)		74.9(2)	74.8(2)
C(1)-Re(1)-Cl(1)	95.40(18)	92.4(2)
C(3)-Re(1)-Cl(1)	94.31(18)	93.1(2)
O(4)-Re(1)-Cl(1)	83.17(11)
N(1)-Re(1)-Cl(1)	85.15(12)	83.76(16)
N(2)-Re(1)-Cl(1)		82.08(17)
C(1)-Re(1)-Br(1)			93.74(19)
C(3)-Re(1)-Br(1)			92.39(19)
N(1)-Re(1)-Br(1)			82.77(14)
N(2)-Re(1)-Br(1)			85.19(14)

Compound 1 is nearly isostructural to 2. The aldehyde and pyridine units form a planar bidentate chelate to the metal ion with a Re-N distance of 2.177(5) Å and a Re-O distance of 2.172(4) Å. The carbonyls adopt the expected facial geometry and exhibit standard bond lengths and angles for Re(CO)₃ compounds. The Re-Cl bond length (2.4792(17) Å) is typical for the Re(CO)₃Cl compounds, such as those that we previously observed in the similarly structured oxime and dioxime complexes [37]. The C-O bond length of the aldehyde (1.241(7) Å) is slightly lengthened versus that seen in free pyridine aldehydes, indicative of back bonding to the carbonyl oxygen. Miguel and co-workers see a similar C-O bond length of 1.223(8) Å for the bromide derivative [15].

Complexes 3 and 4 were each characterized by single crystal X-ray crystallography; the molecular structures both of these compounds are shown in Figure 2. Compounds 3 and 4 maintain a planar bidentate ligand, a facial set of carbonyls and a Re-X bond perpendicular to the pyridyl imine chelate. The bond distances and angles around the rhenium metal centers in these two compounds are as expected for pyridylimine Re(CO)₃ compounds. The imine bonds in 3 and 4 are longer (1.284(10) and 1.303(9) Å, respectively) than the corresponding aldehyde carbonyl bonds in 1 and 2, due to the increased π backbonding to the diimine unit. The two compounds also exhibit different orientations of the phenyl ring of the sulfanilamide group; in compound 4, the plane of the ring phenyl ring is out of the plane of the pyridyl imine chelate by ~41°, whereas in compound 3 the two planes are more orthogonal, with an angle of ~76°. Compounds 5 and 6 did not produce single crystals despite numerous attempts and were characterized by spectroscopic methods. All CO bond lengths fall within the normal range as expected for Re(CO)₃ compounds.

In compound 3, we were able to refine on a solvent methanol in the crystal structure. This methanol engages in a hydrogen bonding interaction with the sulfonamide nitrogen with a O···N distance of ~2.85 Å. In addition, there is a second hydrogen bond observed between the sulfonamide NH₂ group and the rhenium bound chloride of a neighbouring compound. This second interaction has a N···Cl distance of 3.26 Å. For compound 4, the solvent methanol was modeled as a diffuse contribution to the difference map without assigning specific atom positions using the SQUEEZE program, so we were not able to discern any solid state hydrogen bonding interactions between 4 and solvent. However, due to the alternate packing structure in 4, no hydrogen bonding between the sulfonamide NH₂ group and the metal bound bromide is observed. We speculate that the differences in packing between the two structures (and the observation of two different space groups for essentially isostructural molecules) may result from the presence of the NH-Cl hydrogen bond in 3 and the absence of a analogous NH-Br interaction in 4.

2.3. Spectroscopic Characterization

IR spectra of the new compounds displayed two metal carbonyl bands between about 2010 and 1840 cm⁻¹. This is consistent with vibrations analogous to a₁ and e modes resulting from a pseudo-C_3v symmetry for the compounds. Compound 1 also exhibits a carbonyl stretch from the metal bound aldehyde unit at 1662 cm⁻¹. This value is significantly altered from that of free pyridine-2-carboxyaldehyde, which appears at 1730 cm⁻¹ [38].

The NMR spectra of each compound were consistent with the observed structures for these complexes. The spectrum of compound 1 provided important information confirming its identity. The ¹H NMR spectrum showed a singlet at 9.98 ppm that is diagnostic of the aldehyde proton, and pyridine protons show up in expected locations. The ¹³C NMR spectrum showed a peak at 192.6 ppm—appearing in the middle of the metal carbonyl resonances—due to the aldehyde carbon atom. As expected, the ¹H and ¹³C spectra of 1 closely match the spectra of 2. The ¹H and ¹³C NMR spectra of compounds 3 and 4 displayed peaks due to the amide and phenyl groups in the regions expected. Spectra for each compound, show peaks for the pyridine and imine functional groups in the expected regions. As typically observed for similar metal complexes, the protons closest to the metal are shifted downfield more than their neighbors.

As the structures of compounds 5 and 6 could not be elucidated by X-ray crystallography, NMR spectra were carefully analysed and mass spectrometry was employed to verify their identities. In addition to the expected peaks for the pyridine and imine groups in both the ¹H and ¹³C NMR spectra, the compounds showed the appropriate number of peaks in proper locations, consistent with a modified 4-aminofluorescein moiety. In addition, ¹³C NMR data established the formation of compound 5 and 6 as observed by the conversion of the aldehyde (C(H)=O) to an imine via the reaction with 4-fluoresceinamine, as shown in Figure 3. The aldehyde carbon atom shifts from 192.6 ppm (top spectrum) in 1 to 159.7 ppm in 5. Compound 6 shows a similar resonance for its imine carbon at 160.2 ppm. Comparison of the actual sample isotope pattern recorded for the mass spectra for 5 and 6 to the simulated (expected) isotope pattern (Figure 4) confirms the identity of these compounds.

Figure 3.

Top: Predicted (left) and observed (right) ESI MS of 5. Bottom: Predicted (left) and observed (right) ESI MS of 6.

Figure 4.

¹³C NMR spectra for 1 (top), 5 (middle) and 6 (bottom). The arrow shows the shift of the aldehyde carbon resonance for 1 relative to the imine carbon resonances for 5 and 6.

The UV-visible spectra for 5 and 6 were measured, and as expected the spectra are dominated by fluorescein-based transitions. The spectra of both compounds along with that of 4-aminoflurescein are shown in Figure 5. We determined the extinction coefficients for the primary absorption peak to be 7.96 x 10⁴ M⁻¹cm⁻¹ at 500 nm for compound 5 and 6.54 x 10⁴ M⁻¹cm⁻¹ at 497 nm for compound 6. These molar absorptivities are in the same range as that seen in unmodified 5-aminoflurescein. Compounds 5 and 6 each fluoresce when the absorption maxima at 500 nm and 497 nm, respectively, are irradiated. Figure 6 shows the excitation and emission spectra of the two fluorescein conjugate complexes. The emission resembles that of normal fluorescein in both profile and Stokes shift. When compared to normal fluorescein, which has a quantum yield of 0.79 [39], the quantum yield of emission is significantly reduced, measuring ~0.038 for both compounds 5 and 6. The reduction in quantum yield can be directly attributed to the heavy atom effect due to the close proximity of the rhenium and the halide. However, the emission of these two compounds is increased relative to that of 4-aminofluorescein, which is non-emissive. The formation of the Schiff base and metal coordination of the amine nitrogen of 4-aminofluorescein restores the fluorescence; the restoration of fluorescence upon metal binding of exocyclic amines is a well-known property of fluorescein-based dyes [40–43].

Figure 5.

UV-visible absorption spectra for 5, 6, fluorescein, and 4-aminofluorescein.

Figure 6.

Excitation and emission spectra for 5, 6, and fluorescein.

2.4. Conclusions

We have found that pyridine aldehyde-bound rhenium(I) carbonyls can be used to readily generate biologically relevant conjugate complexes. The metal-bound aldehyde readily reacts with amines such as anilines to afford imines. In the current study, we have successfully generated conjugates incorporating sulfanilamide and 4-aminofluorescein, producing the corresponding Re(CO)₃(pyca-R)X compound with the pyridylimine covalently linked to the desired biological molecule. Our work continues with these compounds and similar biologically relevant conjugates of Re(CO)₃.

3. Experimental

3.1. Materials and Methods

All reagents were purchased from Strem, Acros Organics or Sigma-Aldrich and used as received. NMR spectroscopy was performed with Varian VXR 300 MHz and Varian 500 MHz NMR instruments. Elemental analyses were carried out at the School of Chemical Sciences Microanalytical Laboratory at the University of Illinois at Urbana-Champaign. Mass spectrometric analyses were carried out at the Mass Spectrometry and Proteomics Facility at the Ohio State University in Columbus, OH or at the University of Akron in Akron, OH. IR spectra were recorded on a Nicolet Series II Magna-IR 750 spectrometer. UV-visible spectroscopy was carried out on a Hitachi U-3010 spectrometer. Fluorescence measurements were made on a Horiba Jobin Yvon FluoroMax-4 spectrofluorometer. Fluorescence quantum yield measurements were made relative to normal fluorescein as a standard [44].

3.2. Synthesis of 1 and 2

Compound 1 and previously reported 2 [15] were prepared by similar methods. No special precautions were taken as the synthetic procedure and the resulting products are not air or moisture sensitive. In a typical reaction, Re(CO)₅Cl (0.200 g, 0.552 mmol) was mixed with one equivalent of pyridine-2-carboxaldehyde (0.0590 g, 0.552 mmol) and 15 mL of toluene. The mixture was then heated to reflux for 24 h, slowly causing the solution to turn orange-red. Upon completion of the reaction, a red product was collected by filtration. The product was washed with diethyl ether and dried. Crystals suitable for X-ray diffraction were grown by vapor diffusion of pentane into a CH₂Cl₂ solution of the compound.

Compound 1: Yield 132 mg, 58%. ¹H NMR (d₆-DMSO, δ ppm): 9.98 (s, 1 H, (CHO), 8.83 (d, J = 4.2 Hz, 1H, H on py), 8.06 (t, J = 7.5 Hz, 1H, H on py), 7.94 (d, J = 7.5 Hz, 1H, H on py), 7.71(m, 1H, H on py). ¹³C NMR (d₆-DMSO, δ ppm) 196.3, 193.7, 192.6, 152.30, 150.3, 137.7, 128.4, 121.68. ESI MS (positive ion) [M-Cl]⁺: 377.9 M/Z Calc. 377.35 M/Z; CHN Anal. Calc. for ReC₉H₅O₄NCl: C, 26.19; H, 1.22; N, 3.39. Found: C, 25.87; H, 0.99; N, 3.20. IR (CO stretch, cm⁻¹): 2023(s), 1892(s), 1662(m).

3.3. Synthesis of 3 and 4

Preparation of 3: 1 (0.050 g, 0.12 mmol) was mixed with one equivalent of sulfanilamide (0.020 g, 0.12 mmol) and 10 mL methanol was then added to the flask. The mixture was heated to reflux. During the 6-hour reflux, the red solution turned to a dark orange-red color. The solution volume was decreased by slow evaporation and an orange solid was collected by filtration. The product was washed with diethyl ether and dried. Crystals suitable for characterization by X-ray diffraction were formed by slow evaporation of methanol.

Compound 3: Yield 56 mg, 82%. ¹H NMR (d₆-DMSO, δ ppm): 9.41 (s, 1H, N=CH), 9.10 (d, J = 5.5 Hz, 1H, H on py), 8.39 (d, J = 3.5 Hz, 1H, H on py), 8.03 (m, 2H, H on C₆H₄), 7.88 (m, 2H, H on py), 7.72 (d, J = 9.0 Hz, 2H, H on C₆H₄), 7.51 (s, 2H, NH₂). ¹³C NMR ( d₆-DMSO, δ ppm) 197.9, 197.3, 187.7, 171.9, 155.3, 153.7, 153.1, 144.7, 141.1, 131.3, 130.7, 127.6, 123.4, 112.9, 105.0. ESI MS (positive ion) [M+Na]⁺: 589.9 M/z Calc. 589.9 M/z; CHN Anal. Calc. for ReC₁₅H₁₁O₅N₃SCl: C, 31.78; H, 1.96; N, 7.41. Found: C, 31.74; H, 1.99; N, 7.29. IR (CO stretch, cm⁻¹): 2021(m), 1895(s).

Preparation of 4: This compound was synthesized via an identical procedure as 3. Crystals for structural analysis were grown by slow evaporation from methanol.

Compound 4: Yield 57 mg, 85%. ¹H NMR (d₆-DMSO, δ ppm): 9.39 (s, 1H, N=CH), 9.13 (d, J = 5.5 Hz, 1H, H on py), 8.39 (m, 1H, H on py), 8.03 (d, J = 7 Hz, 2H, H on C₆H₄), 7.87 (m, 2H, H on py), 7.73 (d, J = 5 Hz, 2H, H on C₆H₄), 7.51 (s, 2H, NH₂). ¹³C NMR (d₆-DMSO, δ ppm) 197.4, 196.9, 187.1, 171.8, 155.3, 153.9, 153.21, 144.7, 141.0, 131.4, 130.6, 127,8, 127.6, 123.4, 112.9 ESI MS (positive ion): [M+Na]⁺: 633.9 M/z Calc. 634.4 M/z; CHN Anal. Calc. for ReC₁₅H₁₁O₅N₃SBr: C, 29.46; H, 1.81; N, 6.87. Found: C, 29.01; H, 1.64; N, 6.66. IR (CO stretch, cm⁻¹): 2023(m), 1897(s).

3.4. Synthesis of 5 and 6

Preparation of 5: Compound 1 (50 mg, 0.12 mmol) was mixed with one equivalent of 4-aminofluorescein (42 mg, 0.12 mmol) and 25 mL methanol was added to the flask. The mixture was heated to reflux. During the 6-hour reflux, an orange precipitate formed which was collected by filtration. The product was washed with cold methanol and dried.

Compound 5: Yield 81 mg, 91%. ¹H NMR (d₆-DMSO, δ ppm): 10.23 (br s, 2H OH), 9.47 (s, 1H, N=C-H), 9.10 (d, J = 5.6 Hz, 1H, H on py), 8.39 (m, 2H, H on Ar), 8.14 (d, 2.0 Hz, 1H, H on Ar), 7.89 (m, 1H H on py), 7.72 (m, 1H, H on py), 7.51 (d, J = 8.1 Hz, 1H, H on Ar), 6.70 (m, 2H, H on Ar), 6.64 (d, J = 8.1 Hz, 1H, H on Ar), 6.63 (m, 3H, H on Ar). ¹³C NMR (d₆-DMSO, δ ppm) 197.8, 197.4, 187.8, 172.7, 168.1, 160.2, 155.3, 153.7, 152.8, 152.3, 141.2, 131.4, 130.8, 130.6, 129.5, 129.4, 127.9, 126.0, 118.2, 113.2, 109.6, 102.8, 83.6. ESI MS (positive ion) (M+ H): 743.0 M/z Calc. 743.1 M/z; CHN Anal. Calc. for ReC₂₉H₁₆O₈N₂Cl·0.5H₂O: C, 46.37; H, 2.18; N, 3.73. Found: C, 46.16; H, 1.96; N, 3.70. IR (CO stretch, cm⁻¹): 2017(m), 1920(s), 1862(s).

Preparation of 6: This compound was synthesized via an identical procedure as 5.

Compound 6: Yield 76 mg, 89%. ¹H NMR (d₆-DMSO, δ ppm): 10.19 (br s, 2H OH), 9.47 (s, 1H, N=C-H), 9.12 (d, J = 5.6 Hz, 1H, H on py), 8.38 (m, 2H, H on Ar), 8.14 (d, 2.0 Hz, 1H, H on Ar), 8.00 (m, 1H H on py), 7.85 (m, 1H, H on py), 7.53 (d, J = 8.1 Hz, 1H, H on Ar), 6.67 (m, 2H, H on Ar), 6.63 (d, J = 8.1 Hz, 1H, H on Ar), 6.60 (m, 3H, H on Ar). ¹³C NMR (d₆-DMSO, δ ppm) 196. 9, 196.10, 186.7, 172.2, 167.7, 159.7, 154.8, 153.4, 152.3, 152.0, 151.8, 140.6, 131.0, 130.2, 129.1, 128.9, 127.4, 125.5, 117.8, 112.8, 109.0, 102.4, 83.6. ESI MS (positive ion) (M+ H): 787.0 M/z Calc. 787.5 M/z; CHN Anal. Calc. for ReC₂₉H₁₆O₈N₂Br·0.5H₂O: C, 43.78; H, 2.15; N, 3.52. Found: C, 43.75; H, 1.84; N, 3.46; IR (CO stretch, cm⁻¹): 2018(m), 1927(s), 1868(s).

3.5. X-ray structure determination of 1, 3 and 4

X-ray intensity data were measured at 100 K (Bruker KYRO-FLEX) on a Bruker SMART APEX CCD-based X-ray diffractometer system equipped with a Mo-target X-ray tube (λ = 0.71073 Å) operated at 2000 W power. The crystals were mounted on a cryoloop using Paratone N-Exxon oil and placed under a stream of nitrogen at 100 K. The detector was placed at a distance of 5.009 cm from the crystals. The data were corrected for absorption with the SADABS program. The structures were refined using the Bruker SHELXTL Software Package (Version 6.1), and were solved using direct methods until the final anisotropic full-matrix, least squares refinement of F² converged [45]. Crystal data and structure refinement parameters are shown in Table 1.

Highlights.

• We report the preparation of a new Re(I) pyridine-aldehyde

• We relate the preparation of four new bioconjugates

• Three crystal structures are reported

• The fluorescein derivatives are luminescent; quantitative spectra were recorded.

• The sulfonamide derivatives mimic derivatives used in carbonic anhydrase targeting.

Appendix A. Supplementary material

CCDC 895955 – 895957 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.