A variant of the MP2-R12 class of theories is introduced using an arbitrary geminal function in the place of r12. Integrals are derived for the case where the geminal is expanded in a basis of Gaussian functions in the interelectronic distance. Recurrence relations are derived that do not depend on the exponents of the Gaussian geminals, allowing much of the integration work to be performed after summations over the geminal expansion. Sample calculations at various levels of explicitly correlated MP2 theory are presented for He, Ne, and water.

The recently introduced MP2-R122*A(loc) and LMP2-R122*A(loc) methods are modified to use a short-range correlation factor expanded as a fixed linear combination of Gaussian geminals. Density fitting is used to reduce the effort for integral evaluation, and local approximations are introduced to improve the scaling of the computational resources with molecular size. The MP2-F122*A(loc) correlation energies converge very rapidly with respect to the atomic orbital basis set size. Already with the aug-cc-pVTZ basis the correlation energies computed for a set of 21 small molecules are found to be within 0.5% of the MP2 basis set limit. Furthermore the short-range correlation factor leads to an improved convergence of the resolution of the identity, and eliminates problems with long-range errors in density fitting caused by the linear r12 factor. The DF-LMP2-F122*A(loc) method is applied to compute second-order correlation energies for molecules with up to 49 atoms and more than 1600 basis functions.

The explicitly correlated second order Møller-Plesset (MP2-R12) methods perform well in reproducing the last detail of the correlation cusp, allowing higher accuracy than can be accessed through conventional means. Nevertheless in basis sets that are practical for calculations on larger systems (i.e., around triple- or perhaps quadruple-zeta) MP2-R12 fails to bridge the divide between conventional MP2 and the MP2 basis set limit. In this contribution we analyse the sources of error in MP2-R12 calculations in such basis sets. We conclude that the main source of error is the choice of the correlation factor r12. Sources of error that must be avoided for accurate quantum chemistry include the neglect of exchange commutators and the extended Brillouin condition. The generalized Brillouin condition is found not to lead to significant errors.

The crystal structure of a ternary complex of pig muscle phosphoglycerate kinase (PGK) containing 3-phosphoglycerate (3-PG) and manganese adenylylimidodiphosphate (Mn AMP-PNP) has been determined and refined at 2.0 A resolution. The complex differs from the true substrate ternary complex only in the presence of an imido- rather than an oxylink between beta- and gamma-phosphates of the bound nucleotide. The 3-PG is bound in a similar manner to that observed in binary complexes. The nucleotide is bound in a similar manner to Mg ADP except that the metal ion is coordinated by all three alpha-, beta-, and gamma-phosphates, but not by the protein. The gamma-phosphate, which is transferred in the reaction, is not bound by the protein. One further characteristic of the ternary complex is that Arg-38 moves to a position where its guanidinium group makes a triple interaction with the N-terminal domain, the C-terminal domain, and the 1-carboxyl group of the bound 3-PG. Although a hinge-bending conformation change is seen in the ternary complex, it is no larger than that observed in the 3-PG binary complex. To reduce that distance between two bound substrates to a value consistent with the direct in-line transfer known to occur in PGK, we modeled the closure of a pronounced cleft in the protein structure situated between the bound substrates. This closure suggested a mechanism of catalysis that involves the "capture" of the gamma-phosphate by Arg-38 and the N-terminus of helix-14, which has a conserved Gly-Gly-Gly phosphate binding motif. We propose that nucleophilic attack by the 1-carboxyl group of the 3-PG on the gamma-phosphorus follows the capture of the gamma-phosphate, leading to a pentacoordinate transition state that may be stabilized by hydrogen bonds donated by the NH groups in the N-terminus of helix 14 and the guanidinium group of Arg-38. During the course of the reaction the metal ion is proposed to migrate to a position coordinating the alpha- and beta-phosphates and the carboxyl group of Asp-374. The mechanism is consistent with the structural information from binary and ternary substrate complexes and much solution data, and gives a major catalytic role to Arg-38, as indicated by site-directed mutagenesis.

In noncardiac chest pain (NCCP), altered visceral perception may result from abnormal cerebral processing of sensory input rather than abnormalities of afferent pathways. However, the interactions between symptoms, autonomic function and oesophageal stimuli are poorly studied. Oesophageal stimulation elicits reproducible cortical evoked potentials [CEP] and modulates heart rate variability via vagal pathways, as visible on power spectrum analysis of heart rate variability [PS-HRV]. These methods are increasingly used to study the function of visceral afferent neural pathways in human. The aim of this study was to compare EP and PS-HRV during oesophageal stimuli in NCCP and controls. Twelve healthy volunteers (one female, 11 male; aged 24-51 years; mean 32 +/- 8 years), and eight NCCP patients (three female, five male; age range 26-58, mean 40.5 +/- 10 years) were studied. Electrical oesophageal stimulation (EOS; 200 microseconds, 0.2 Hz, 25 stimuli) was applied to the oesophageal wall 5 cm above the lower oesophageal sphincter (LOS), and perception thresholds (measured in mA) determined. EP responses were recorded using 22 standard electroencephalogram scalp electrodes. Autonomic activity was assessed using PS-HRV, before, during, and after oesophageal stimulation. Measured PS-HRV indices included high frequency (HF; 0. 15-0.5 Hz) and low frequency (LF; 0.06-0.15 Hz) power, respectively, assessing vagal and sympathetic activity, and the LF/HF ratio. EOS perception occurred at lower thresholds in NCCP than in controls (3. 6 +/- 1 vs. 7.8 +/- 2 mA, P < 0.05). EP amplitude was greater (13 +/- 2 vs. 6 +/- 1 microV, P < 0.0001), and latency longer in controls vs. NCCP (191 +/- 7 ms vs. 219 +/- 6 ms, P < 0.001). In NCCP, EOS decreased sympathetic outflow (low frequency peak on PS-HRV) and increased cardiovagal activity (high frequency peak, P < 0.02) to a significantly higher degree in comparison with controls. During EOS, heart rate decreased in NCCP from 68 vs. 62 beats min-1 (P < 0.003) but not in controls. In NCCP patients, EOS was perceived at lower intensities and was associated with a greater cardiovagal reflex response. EP responses associated with EOS were smaller in NCCP than in controls, suggesting that an increased perception of oesophageal stimuli results from an enhanced cerebral processing of visceral sensory input in NCCP, rather than from hyperalgesic responses in visceral afferent pathways.