Comparative Density Functional Study of Models for the Reaction

Feb 19, 2008 - Faculdade de Ciências da Saúde, Universidade Fernando Pessoa, Rua Carlos da Maia, 296, 4200-150 Porto, Portugal, and REQUIMTE, ...
0 downloads 0 Views 292KB Size
3144

J. Phys. Chem. B 2008, 112, 3144-3148

Comparative Density Functional Study of Models for the Reaction Mechanism of Uroporphyrinogen III Synthase Pedro J. Silva† and Maria Joa˜ o Ramos*,‡ Faculdade de Cieˆ ncias da Sau´ de, UniVersidade Fernando Pessoa, Rua Carlos da Maia, 296, 4200-150 Porto, Portugal, and REQUIMTE, Faculdade de Cieˆ ncias do Porto, Rua do Campo Alegre, 687, 4169-007 Porto, Portugal ReceiVed: August 3, 2007; In Final Form: NoVember 29, 2007

The asymmetric cyclic tetrapyrrole uroporphyrinogen III is the common precursor of heme, chlorophyll, siroheme, and other biological tetrapyrroles. In vivo, it is synthesized from a linear symmetric precursor (hydroxymethylbilane) by uroporphyrinogen III synthase, which catalyzes the inversion of one of the four heterocyclic rings present in the substrate. Two mechanisms have been proposed to explain this puzzling ring inversion, either through sigmatropic shifts or through the direct formation of a spirocyclic pyrrolenine intermediate. We performed the first high-level quantum mechanical calculations on model systems of this enzyme to analyze these contrasting reaction mechanisms. The results allow us to discard the sigmatropic shift mechanism and suggest that the D-ring of the hydroxymethylbilane substrate binds to the enzyme in a conformation that shields its terminal portion from reacting with ring A and prevents the formation of the biologically useless uroporphyrinogen I, whose accumulation (in individuals lacking functional uroporphyrinogen III synthase) leads to severe cutaneous dermatosis.

1. Introduction All biological tetrapyrroles are ultimately derived from the condensation of four molecules of porphobilinogen. Abiotic condensation of this precursor yields a mixture of four uroporphyrinogen isomers, of which isomer III is the dominant form.1 Probably for this reason, uroporphyrinogen III has been selected by evolution as the precursor for the wide variety of naturally occurring tetrapyrroles (heme, chlorophyll, siroheme, cobalamin, and factor F430). However, in contrast to the abiotic case, biological condensation of porphobilinogen molecules occurs through a sequential, nonrandom mechanism yielding a single product: hydroxymethylbilane (Figure 1). In solution, hydroxymethylbilane spontaneously cyclizes into the symmetric uroporphyrinogen I, which is not a substrate for the subsequent enzymes in the tetrapyrrole biosynthesis pathways, and therefore, synthesis of uroporphyrinogen III from hydroxymethylbilane requires the rearrangement of the last pyrrole ring in hydroxymethylbilane.2 This reaction is catalyzed by uroporphyrinogen III synthase (also called uroporphyrinogen III cosynthase, EC 4.2.1.75), a bi-lobed monomeric enzyme that has been crystallized recently.3 Impairment of this enzyme causes the accumulation of type I porphyrins and leads to severe cutaneous dermatosis, a characteristic of congenital erythropoietic porphyria.4 Uroporphyrinogen synthase (UroS) has been proposed5,6 to catalyze the removal of the hydroxyl group from C20 in hydroxymethylbilane, yielding a reactive azafulvenium cation (Figure 2). Electrophilic attack at C16 would yield a spirocyclic pyrrolenine intermediate, which might break in the opposite direction, forming a new azafulvenium cation. Rotation of the D ring around the C20-C16 bond would bring C19 close to * Corresponding author. E-mail: [email protected]. † Universidade Fernando Pessoa. ‡ REQUIMTE.

the reactive carbocation at C15, allowing closure of the macrocycle. On the other hand, semiempirical calculations7 have suggested that direct formation of the spirocyclic intermediate through electrophilic attack on C16 is very unfavorable and that attack at C19, followed by 1,5-sigmatropic shifts along the pyrrole, would be a more favorable way of generating the spiro intermediate. The existence of this intermediate is not in doubt since a synthetic spirolactam differing from the proposed spirocyclic pyrrolenine only by having an amide instead of an imine is a very effective inhibitor of uroporphyrinogen synthase.8,9 To test these hypotheses, we carried out DFT calculations of both reaction mechanisms. Our calculations confirmed the direct formation of the spirocyclic pyrrolenine, identified the ratelimiting step, and showed that the substrate must bind the enzyme in a conformation that prevents C19 from reacting with C20. 2. Materials and Methods All calculations were performed at the Becke3LYP level of theory.10-12 Autogenerated delocalized coordinates13 were used for geometry optimizations, using a medium-sized basis set, 6-31G(d), since it is well-known that larger basis sets give very small additional corrections to the geometries, and their use for this end is hence considered to be unnecessary from a computational point of view.14-16 More accurate energies of the optimized geometries were calculated with a triple-ζ quality basis set, 6-311+G(d,p). Zero-point energy (ZPE) and thermal effects (T ) 298.15 K and P ) 1 bar) were evaluated using a scaling factor of 0.9804 for the computed frequencies. The full substrate was modeled as a linear tetrapyrrole with all acetate and propionate substituents replaced by methyl groups. Since the binding site of hydroxymethylbilane to UroS is not known, environmental contributions to the stationary

10.1021/jp076235f CCC: $40.75 © 2008 American Chemical Society Published on Web 02/19/2008

Reaction Mechanism of Uroporphyrinogen III Synthase

J. Phys. Chem. B, Vol. 112, No. 10, 2008 3145

Figure 1. Tetrapyrroles involved in the reaction catalyzed by uroporphyrinogen synthase. Uroporphyrinogen I and III differ only in the relative arrangements of the substituents present on ring D.

Figure 2. Proposed reaction mechanisms for uroporphyrinogen synthase. (A) Direct formation of a spiro intermediate (Mathewson-Corwin mechanism). (B) Indirect formation of a spiro intermediate by means of sigmatropic shifts (Tietze-Geissler mechanism). A: CH2-COO- and P: CH2-CH2-COO-.

points and transition states were computed with the polarizable conductor model,17-19 as implemented in PcGamess,20 with several dielectric constants ( ) 4, 20, and 78.3). Unless otherwise noted, all energies mentioned in the text were computed by applying the PCM model ( ) 4) on gas-phase optimized geometries. Dispersion and repulsion effects were evaluated as described by Amovilli and Mennucci.21 Atomic charge and spin density distributions were calculated with a Mulliken population analysis22 based on symmetrically orthogonalized orbitals.23 All calculations were performed with the PcGamess 7.0 software package. 3. Results Although a key tyrosine residue was implicated in substrate binding or catalysis,24 the lack of precise information regarding the binding mode of hydroxymethylbilane to uroporphyrinogen synthase does not allow the computational study of the initial steps of hydroxymethylbilane protonation and hydroxide elimination from C20. However, since these steps are common to both mechanisms, this information is not needed for a comparative analysis between them. Our study therefore starts with the reactive azafulvenium cation arising from hydroxide elimination.

Contrary to our expectation, there was considerable charge delocalization in the azafulvenium cation: ring A (and its methylene C20) carries 0.72 au charge, rings B and C each carry ca. -0.1 au charge, and each methylene carbon between pyrrole rings carries 0.15 au charge. The azafulvenium cation is extremely flexible, and several energetic minima are available at similar energies. The search for the initial minima was therefore performed by following the Intrinisic Reaction Coordinate (IRC) from the transition state for the initial electrophilic attack. 3.1. Direct Formation of the Spiro Intermediate (Mathewson-Corwin Mechanism). The transition state for electrophilic attack of C20 on C16 lies 14.3 kcal mol-1 above the reacting conformation of the azafulvenium cation. In this state, C20 lies only 2.152 Å from C16, and 0.32 au charge has been transferred to ring D (Figure 3A). Completion of this modestly endergonic (7.7 kcal mol-1) reaction step drives C20 to a final distance of 1.578 Å from C16. In the C16 spirocyclic intermediate, 0.58 au positive charge is localized in ring D, 0.13 in ring A, and 0.09 in ring C. Since in the substrate model we used both acetate and propionate substituents of the pyrrole rings in uroporphyrino-

3146 J. Phys. Chem. B, Vol. 112, No. 10, 2008

Silva and Ramos

Figure 3. Structures of important transition states. (A) Transition state for ring A addition to C16. (TS1 in the Mathewson-Corwin mechanism. TS2 in this mechanism is the mirror image of this. This transition state also is the fifth step in the Tietze-Geissler mechanism.) (B) Transition state for ring A addition to C19 in the Tietze-Geissler mechanism). (Its mirror image, i.e., addition of ring C to C15, occurs as the last step in the Mathewson-Corwin mechanism.) (C) Transition state for ring A transfer from C19 to C18 (Tietze-Geissler mechanism).

gen have been replaced by methyl groups, azafulvenium cations I and II are modeled by the same structure in our calculations. Therefore, the transformation of the C16-spiro intermediate into azafulvenium II will simply be the mirror image of the formation of C16-spiro from azafulvenium I (i.e., a modestly exergonic (-7.7 kcal mol-1) reaction with an activation free energy of 14.3-7.7 ) 6.7 kcal mol-1). The final reaction step consists of the electrophilic attack of C19 by the methylene on ring C. This is a very fast step, with an activation free energy of 4.6 kcal mol-1. In the transition state (Figure 3B), the distance between methylene ring C and C19 has shortened from the reactants’ distance of 2.987 to 2.343 Å, and a modest charge (0.22 au) has been transferred from ring C to ring D. As the C15-C19 distance decreases to its final value (1.58 Å), ring D acquires an additional 0.42 au positive charge. The product of this reaction is uroporphyrinogen III protonated at C19. This step is exergonic by 9.5 kcal mol-1 (Figure 4 and Table 1). 3.2. Indirect Formation of the Spiro Intermediate (TietzeGeissler Mechanism). The Tietze-Geissler mechanism postulates an initial electrophilic attack at C19, followed by sigmatropic shifts to C18, C17, and C16. This initial step is the mirror image of the ring C and C19 attack described in the previous section. The product of this reaction is uroporphyrinogen I protonated at C19, instead of protonated uroporphyrinogen III. This step is exergonic by 9.5 kcal mol-1 and very fast (activation ∆G ) 4.6 kcal mol-1). The shift of ring A from C19 to C18, however, is not as favorable: its transition state lies 24.9 kcal mol-1 above its reactant. This transition state is

Figure 4. Overall reaction energy profile of the Mathewson-Corwin mechanism at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) level (s-:  ) 4; - - -:  ) 20; and ----:  ) 78.3).

TABLE 1: Relative Electronic Energies (kcal mol-1) of Intermediates in the Mathewson-Corwin Model, Calculated at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) Level azafulvenium I TS 1 C16-spiro TS 2 azafulvenium II TS 3 product

)4

 ) 20

 ) 78.3

0.0 14.3 7.7 14.3 0.0 4.6 -9.5

0.0 15.6 8.0 15.6 0.0 5.0 -10.5

0.0 16.0 8.2 16.0 0.0 5.1 -10.7

quite asymmetric: the (breaking) C19-C20 bond is 2.625 Å long, whereas the (forming) C18-C20 bond is much shorter, at 1.981 Å (Figure 3C). This observation seems to be more

Reaction Mechanism of Uroporphyrinogen III Synthase

J. Phys. Chem. B, Vol. 112, No. 10, 2008 3147

Figure 5. Two-dimensional potential energy surfaces of ring A transfer from C19 to C18 (left panel), C18 to C17 (center panel), and C17 to C16 (right panel). Computations performed at the B3LYP/3-21G level.

(15.4 kcal mol-1, computed from the value reported by Omata et al.25). 4. Conclusion

Figure 6. Overall reaction energy profile of the Tietze-Geissler mechanism at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) level (s-:  ) 4; - - -:  ) 20; and ----:  ) 78.3).

TABLE 2: Relative Electronic Energies (kcal mol-1) of Intermediates in the Tietze-Geissler Model, Calculated at the B3LYP/6-311+G(d,p)//B3LYP/6-31G(d) Level azafulvenium I TS1 C19-bound intermediate (protonated uroporphyrinogen I) TS2 C18-bound intermediate C17-bound intermediate C16-spiro TS 5 azafulvenium cation II TS 1 products

)4

 ) 20

 ) 78.3

0.0 4.6 -9.5

0.0 5.0 -10.5

0.0 5.1 -10.7

15.4 13.1 7.9 8.0 14.5 0.2 4.8 -9.3

15.5 12.2 7.5 8.2 15.8 0.2 5.2 -10.4

15.5 12.0 7.5 8.2 16.0 0.0 5.1 -10.8

consistent with bond breaking followed by new bond formation than with the occurrence of a sigmatropic shift. The C18 bound intermediate formed in this step (Supporting Information Figure S1) lies 13.1 kcal mol-1 above the initial azafulvenium cation (i.e., 22.6 kcal mol-1 above the protonated uroporphyrinogen I). The formation of subsequent intermediates occurs through similarly asymmetric transition states, albeit with lower energetic barriers (Figure 5). After the formation of the C16-spirocyclic intermediate, the reaction is postulated to occur as in the previous (Mathewson-Corwin) mechanism. The full reaction profile (Figure 6 and Table 2) shows that indirect formation of the C16-spiro intermediate from the azafulvenium cation is not consistent with experimental observations since it has an activation energy (24.9 kcal mol-1) far above the measured value

Our calculations clearly show that, although C20 addition to C19 is thermodynamically much more favorable than addition to C16 (∆G ) -9.5 kcal mol-1 vs 7.7 kcal mol-1), the C19 reaction product is a kinetic dead-end: the activation energy for its transformation into the experimentally suggested C16spirocyclic pyrrolenine through a sigmatropic arrangement is far too large (24.9 kcal mol-1). The thermodynamically disfavored addition of C20 to C16, on the other hand, occurs with a modest activation energy of 14.3 kcal mol-1. The noninclusion of (unknown) active site amino acids in our calculations should not alter dramatically the relative activation energies of both mechanisms since the occurrence of C16-spiro as an intermediate in both mechanisms implies that any active site characteristics affecting its energy would equally affect either of them. An analysis of the overall reaction profiles of the Tietze-Geissler mechanism further shows that even if the active site destabilizes the initial, very stable, C19 bound intermediate and prevents it from becoming a kinetic dead-end, the similarity of the energies of the transition states for C19C18 rearrangement (TS2) and C20-C16 bond breaking (TS5) prevents the Tietze-Geissler mechanism from becoming noticeably more favorable than the simpler Mathewson-Corwin mechanism. Therefore, and although our calculations do not address the initial dehydration of hydroxymethylbilane, which prevents direct comparison of our computed values with the experimental value (15.4 kcal mol-1), it seems very unlikely that the reaction mechanism will proceed through the series of sigmatropic arrangements proposed by Tietze and Geissler. It can therefore be concluded that in order for uroporphyrinogen III to be synthesized, the enzyme must bind the hydroxymethylbilane substrate in a conformation than prevents the thermodynamically favored reaction of C20 with C19, thereby allowing reaction with C15, in full agreement with the Mathewson-Corwin mechanism. Preliminary attempts to characterize this binding site through substrate docking to the crystal structure of the enzyme (UroS) were not successful: the subtrate binding landscape was found to be extremely shallow, and many putative binding modes with similar energies were found in dispersed regions of the large crevice separating the two domains of UroS. Supporting Information Available: Geometries of reactants, transition states, and products of all chemical reactions. This material is available free of charge via the Internet at http:// pubs.acs.org.

3148 J. Phys. Chem. B, Vol. 112, No. 10, 2008 References and Notes (1) Mauzerall, D. J. Am. Chem. Soc. 1960, 82, 2605-2609. (2) Shoolingin-Jordan, P. M. The biosynthesis of coproporphyrinogen III. In The Porphyrin Handbook. Volume 12: The Iron and Cobalt Pigments: Biosynthesis, Structure, and Degradation; Kadish, K. M., Smith, K. M., Guilard, M., Eds.; Elsevier: Amsterdam, 2003. (3) Matthews, M. A. A.; Schubert, H. L.; Whitby, F. G.; Alexander, K. J.; Schadick, K.; Bergonia, H. A.; Phillips, J. D.; Hill, C. P. EMBO J. 2001, 20, 5832-5839. (4) Sassa, S.; Kappas, A. J. Intern. Med. 2000, 247, 169-178. (5) Mathewson, J. H.; Corwin, A. H. J. Am. Chem. Soc. 1961, 83, 135137. (6) Leeper, F. J. Ciba Found. Symp. 1994, 180, 111-123. (7) Tietze, L. F.; Geissler, H. Angew. Chem., Int. Ed. Engl. 1993, 32, 1040-1042. (8) Stark, W. M.; Hart, G. J.; Battersby, A. R. J. Chem. Soc., Chem. Commun. 1986, 465. (9) Cassidy, M. A.; Crockett, N.; Leeper, F. J.; Battersby, A. R. J. Chem. Soc., Chem. Commun. 1991, 6, 384-396. (10) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (11) Lee, C.; Yang, W.; Parr, R. J. Phys. ReV. B: Condens. Matter Mater. Phys. 1998, 37, 785. (12) Hertwig, R. W.; Koch, W. J. Comput. Chem. 1995, 16, 576. (13) Baker, J.; Kessi, A.; Delley, B. J. Chem. Phys. 1996, 105, 192212.

Silva and Ramos (14) Fernandes, P. A.; Ramos, M. J. Chemistry 2004, 10, 257-266. (15) Fernandes, P. A.; Ramos, M. J. J. Am. Chem. Soc. 2003, 125, 63116322. (16) Riley, K. E.; Op’t Holt, B. T.; Merz, K. M., Jr. J. Chem. Theory Comput. 2007, 3, 407-433. (17) Tomasi, J.; Persico, M. Chem. ReV. 1994, 94, 2027-2094. (18) Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 106, 5151-5158. (19) Cossi, M.; Mennucci, B.; Pitarch, J.; Tomasi, J. J. Comput. Chem. 1998, 19, 833-846. (20) (a) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen, J. J.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993, 14, 1347-1363. (b). Granovsky, A. A. PC GAMESS, version 7.0; http://classic.chem.msu.su/gran/gamess/index.html. (21) Amovilli, C.; Mennucci, B. J. Phys. Chem. B 1997, 101, 10511057. (22) Mulliken, R. S. J. Chem. Phys. 1995, 23, 1833. (23) Lowdin, P.-O. AdV. Chem. Phys. 1970, 5, 185-199. (24) Roessner, C. A.; Ponnamperuma, K.; Scott, A. I. FEBS Lett. 2002, 525, 25-28. (25) Omata, Y.; Sakamoto, H.; Higashimoto, Y.; Hayashi, S.; Noguchi, M. J. Biochem. (Tokyo, Jpn.) 2004, 136, 211-220.