Supramolecular allosteric cofacial porphyrin complexes

Abstract

Nature routinely uses cooperative interactions to regulate cellular activity. For years, chemists have designed synthetic systems that aim toward harnessing the reactivity common to natural biological systems. By learning how to control these interactions in situ, one begins to allow for the preparation of man-made biomimetic systems that can efficiently mimic the interactions found in Nature. To this end, we have designed a synthetic protocol for the preparation of flexible metal-directed supramolecular cofacial porphyrin complexes which are readily obtained in greater than 90% yield through the use of new hemilabile porphyrin ligands with bifunctional ether-phosphine or thioether-phosphine substituents at the 5 and 15 positions on the porphyrin ring. The resulting architectures contain two hemilabile ligand-metal domains (RhI or CuI sites) and two cofacially aligned porphyrins (ZnII sites), offering orthogonal functionalities and allowing these multimetallic complexes to exist in two states, "condensed" or "open". Combining the ether-phosphine ligand with the appropriate RhI or CuI transition-metal precursors results in "open" macrocyclic products. In contrast, reacting the thioether-phosphine ligand with RhI or CuI precursors yields condensed structures that can be converted into their "open" macrocyclic forms via introduction of additional ancillary ligands. The change in cavity size that occurs allows these structures to function as allosteric catalysts for the acyl transfer reaction between X-pyridylcarbinol (where X = 2, 3, or 4) and 1-acetylimidazole. For 3- and 4-pyridylcarbinol, the "open" macrocycle accelerates the acyl transfer reaction more than the condensed analogue and significantly more than the porphyrin monomer. In contrast, an allosteric effect was not observed for 2-pyridylcarbinol, which is expected to be a weaker binder and is unfavorably constrained inside the macrocyclic cavity.

Figure 1

Graphical representations of the X-ray crystal structure of 8a⊂DABCO as viewed (A) from the side and (B) from the top containing a molecule of DABCO bridging both Zn atoms. Hydrogen atoms, disordered DABCO carbon atoms, and solvent molecules have been omitted for clarity. Pink = Rh, Red = O, Yellow = Cl, Green = P, Blue = N, Light Blue = Zn.

Figure 2

Graphical representations of the X-ray crystal structure of 15a⊂DABCO as viewed (A) from the side and (B) from the top containing a molecule of DABCO bridging both Zn atoms. Hydrogen atoms, disordered DABCO carbon atoms and solvent molecules have been omitted for clarity. Gray = Carbon, Pink = Rh, Red = O, Orange = S, Yellow = Cl, Green = P, Dark Blue = N, Light Blue = Zn.

Figure 3

Acyl transfer reactions catalyzed by (left) a closed macrocycle vs (right) the corresponding open macrocycle. The open macrocycle can preorganize the substrates within the cavity, thereby increasing the rate of the reaction (k₂) in comparison to that (k₁) observed in the presence of the closed macrocycle.

Figure 4

Formation of the three X-(acetoxymethyl)pyridine (X = 2, 3, or 4) isomers by an acyl transfer reaction between 1-acetylimidazole and X-pyridylcarbinol, as catalyzed by Zn–porphyrin complexes 14a and 15a. Concentration vs time plots are shown for the formation of 4-(acetoxymethyl)-pyridine (A, 4-AMP), 3-(acetoxymethyl)pyridine (B, 3-AMP), and 2-(acetoxymethyl)pyridine (C, 2-AMP). All data were corrected for background reactions (see Supporting Information). Conditions: CH₂Cl₂, rt, 9 mM X-pyridylcarbinol, 6 mM 1-acetylimidazole, 2.5 mM biphenyl (GC reference standard), and 0.3 mM supramolecular catalyst (14a and 15a). CO (1 atm) and appropriate amounts of benzyltriethylammonium chloride when indicated.

Scheme 1. Design of Allosteric Porphyrin-Based Supramolecules Whose Cavity Sizes Can Be Modified by the Binding of a Ligand L^a

^a I: Condensed Macrocycle, II: Open Macrocycle. PPh₂ = diphenylphosphine and MES = 1,3,5-trimethylbenzene.

Scheme 2. Synthesis of Ether-Based Ligand 7 and Macrocycles 8a and 8b^a

^a (i) 1-bromo-2-chloroethane, K₂CO₃, Acetone, Reflux; (ii) 1,3-propanedithiol, Y(OTf)₃ (5 mol %), CH₃CN; (iii) KPPh₂, THF; (iv) S₈, THF;(v) NaNO₂, AcCl/H₂O, CH₂Cl₂, 0 °C → rt; (vi) 5-mesityldipyrromethane, BF₃•OEt₂, DDQ, NEt₃, CHCl₃, 4 Å Molecular Sieves; (vii) Zn(OAc)₂•2H₂O, 4:1 CHCl₃/MeOH, Reflux; (viii) Cp₂ZrHCl, THF, 60 °C; (ix) [Rh(CO)₂(Cl)]₂, CH₂Cl₂/THF; (x) [Cu(CH₃CN)₄]PF₆, CH₂Cl₂/THF.

Scheme 3. Synthesis of Thioether-Based Ligand 13 and Macrocycles 14a–b, 15a–b^a

^a (i) ClCH₂CH₂PPh₂, Cs₂CO₃, CH₃CN, Reflux; (ii) S₈, THF; (iii) n-BuLi, DMF, THF, −78 °C; (iv) 5-mesityldipyrromethane, BF₃•OEt₂, DDQ, NEt₃, CHCl₃, 4 Å Molecular Sieves; (v) Zn(OAc)₂•2H₂O, 4:1 CHCl₃/MeOH, Reflux; (vi) Cp₂ZrHCl, THF, 60 °C; (vii) for 14a: [Rh(NBD)Cl]₂, AgBF₄, CH₂Cl₂/THF; (viii) for 14b: [Cu(CH₃CN)₄]PF₆, CH₂Cl₂/THF; (ix) for 15a: PPNCl/CO (1 atm); (x) for 15b: C₅D₅N.