September 15, 2005

Purdue scientists see biochemistry's future – with quantum physics

WEST LAFAYETTE, Ind. – Chemists who have trouble predicting how some large, complex biological molecules will react with others may soon have a solution from the world of computational quantum physics, say Purdue University researchers.

Jorge H. Rodriguez
Download photo
caption below

Using powerful supercomputers to analyze the interplay of the dozens of electrons that whirl in clouds about these molecules, a team of physicists led by Purdue's Jorge H. Rodriguez has found that the quantum property of electrons called "spin" needs to be considered to obtain a complete and fundamental picture of how many biochemical reactions take place. In particular, a class of metal-based proteins that includes hemoglobin and chlorophyll, and their reactions in plants and animals, can be better understood with the technique.

Not only will this discovery sharpen our basic knowledge of biology, Rodriguez said, but it also could help scientists with a number of practical problems – such as selecting the best potential new drug compounds from a vast group of candidates, a process that can cost pharmaceutical companies years of work and millions of dollars.

"Whereas we have had to be satisfied with observing the chemistry in living things and describing it afterward without complete understanding, we are developing computational tools that can predict what will happen between molecules before they meet in the test tube," said Rodriguez, who is an assistant professor of physics in Purdue's College of Science. "Not only does this research open up a new field of science that reveals how metalloproteins and their constituent particles interact, but the quantum theory behind it also should allow us to model and predict these behaviors accurately with computer simulation alone. It is an example of how much can be accomplished with interdisciplinary science."

Rodriguez is pioneering a new field he calls "quantum biochemistry" – a field that involves both biochemistry and particle physics, which are often cited among the more formidable subjects science students tackle. Ordinarily, the two disciplines share little common ground. Although biochemistry deals with interactions among the complex molecules that our bodies use for the fundamental processes of life, these microscopically small molecules are nonetheless gargantuan entities in comparison with the tinier subatomic particles such as protons and electrons that physicists study.

"Despite these differences, there is one point of overlap between chemistry and physics that has interested me, and that is in the elementary particles that whirl about these molecules – the electrons," Rodriguez said. "Physicists have long known that, according to the laws of quantum mechanics, there are some chemical reactions in our bodies that are 'forbidden' – such as hemoglobin's binding oxygen in our lungs when we breathe. But they do happen nonetheless. So, because these reactions involve electron spin, we decided to take a closer look at them."

Charge is a familiar property of an electron, but it is not the only one. Electrons also have another quantum property called spin, and though they are all negatively charged, they can spin in one of two opposing directions – up or down.

"Nature loves balance, and you see evidence of it in both charge and spin," Rodriguez said. "For example, electrons of opposite spin like to pair up with each other as they fly around the nucleus. This allows their spins to balance one another, just as positive and negative charges do between protons and electrons. Even when you have hundreds of electrons forming an immense cloud around a complex molecule, you still see balance in both charge and spin; we call this balance 'conservation,' and it's something we count on in both chemistry and physics to help us understand these tiny objects.

"But sometimes the electrons in metalloproteins seem to be playing a trick on us. As we see with hemoglobin, nature appears to be conserving electronic charge while sacrificing this conservation in spin."

Hemoglobin's active center contains iron, one of the so-called transition metals. These metals are noted for the way several of their electrons can fly around the nucleus unpaired.

When a red blood cell encounters oxygen in our lungs, its hemoglobin is able to grasp some of the oxygen with some of these unpaired electrons, carrying it to the rest of our body. But in the process, the cumulative spin of the system changes in a way that is not conserved, which to a physicist looks as strange as a ball hitting the water without making a splash.

"This chemistry is vital for life, but physicists wonder how it can happen," Rodriguez said. "The charge between the electrons in the bonded oxygen and hemoglobin is balanced in the end, which makes sense to chemists. But the electronic spin of the entire system is not conserved, making a physicist frown at what appears to be a formally forbidden process. Of course, we needed to learn more about nature at the microscopic level."

As many of these supposedly forbidden reactions involve biomolecules centered upon transition metals, which can flip back and forth between different spin states under certain conditions, Rodriguez theorized that it was this variability in spin state that was influencing the rate of these reactions. To explore whether this effect, which Rodriguez calls spin-dependent reactivity, was indeed the decisive factor, the team is modeling the reaction rates with a supercomputer, the only tool capable of keeping track of the motion of so many particles at once.

"Supercomputers have allowed us to check our models against our understanding of spin's effect on a reaction, and our models have been closely checked by experiment," Rodriguez said. "The results suggest that our understanding of electron behavior is sufficient to create virtual models of molecules that we can then 'react' with one another in simulations that accurately predict what will happen when they meet in the physical world."

Rodriguez said the approach, though still in its nascent stages, could provide insight into far more biologically important molecules when it is further developed.

"We are at the point where we have developed computational tools to analyze the spin-dependent processes of biomolecules and have applied them to a few important test cases," he said. "But our methods are based on approaches that are valid for any molecular system. Therefore, hundreds more metalloproteins that are of great scientific and practical interest may be studied in the future with the methods we have developed."

For example, Rodriguez is planning to study the manganese involved in photosynthesis to understand how water is broken down to produce molecular oxygen. But for now, he is happy that the four years of work his team has put into the project have produced such encouraging results.

"We are creating a new field that attempts to understand biochemical processes at the most fundamental level – that of quantum mechanics," he said. "It could be the most important step toward making biochemistry a predictive science rather than a descriptive one."

Two papers on the subject, one of which Rodriguez authored with Purdue's Teepanis Chachiyo, appear in this week's issue of the Journal of Chemical Physics. Jeffrey Long, a professor of chemistry at the University of California at Berkeley, commented on Rodriguez's work.

"Rodriguez has come up with an elegant means of evaluating excited-state electronic structures," he said. "It lends insight to the detailed mechanisms of poorly understood transformations in inorganic complexes."

This research has been supported in part by a Career Award from the National Science Foundation.

Writer: Chad Boutin, (765) 494-2081, cboutin@purdue.edu

Source: Jorge H. Rodriguez, (765) 496-2830, jhrodrig@physics.purdue.edu

Purdue News Service: (765) 494-2096; purduenews@purdue.edu

Note to Journalists: Copies of the two research papers are available by contacting Chad Boutin at cboutin@purdue.edu or (765) 494-2081.

PHOTO CAPTION:
Physicist Jorge H. Rodriguez and his Purdue research team are using powerful computers to probe the effects that spin, a quantum property of elementary particles, has on certain biochemical reactions. With further refinement, Rodriguez's efforts could permit scientists to simulate and predict the results of reactions between complex substances in a virtual environment without the need to combine them physically in the lab. This technique could reduce drug development time and costs for pharmaceutical companies and expand our knowledge of life's fundamental processes.

(Purdue News Service photo/David Umberger)

A publication-quality photograph is available at https://www.purdue.edu/uns/uns/images/+2005/rodriguez-cluster.jpg

An efficient computational method for locating minimum energy crossing points (MECPs) between potential energy surfaces in spin crossover transitions and nonadiabatic spinforbidden (bio)chemical reactions is introduced. The method has been tested on the phenyl cation and the computed MECP associated with its radiationless singlet-triplet spin crossover is in good agreement with available data. However, the convergence behavior of the present method is significantly more efficient than some alternative methods which allows us to study nonadiabatic processes in larger systems such as spin crossover in metal-containing compounds. The convergence rate of the method obeys a fast logarithmic law, which has been verified on the phenyl cation. As an application of this new methodology, the MECPs of the ferrous complex [Fe(ptz)₆](BF₄)₂, which exhibits light induced excited spin state trapping (LIESST), have been computed to identify their geometric and energetic parameters during spin crossover. Our calculations, in conjunction with spin-unrestricted density functional calculations, show that the transition from the singlet ground state to a triplet intermediate and to the quintet metastable state of [Fe(ptz)₆](BF₄)₂ is accompanied by unusually large bond length elongations of the axial ligands (0.26 Angstroms and 0.23 Angstroms, respectively). Our results are consistent with crystallographic data available for the metastable quintet but also predict new structural and energetic information about the triplet intermediate and at the MECPs, which is currently not available from experiment.

The electronic structure of [Fe(ptz)₆](BF₄)₂, a prototype of a class of complexes that display light-induced excited state spin trapping (LIESST), has been investigated by time-independent and time-dependent density functional theories. The density of states of the singlet ground state reveals that the highest occupied orbitals are metal-centered and give rise to a low spin configuration Fe₂+(3d↑x↓y3d↑x↓z3d↑y↓z) in agreement with experiment. Upon excitation with light in the 2.3 — 3.3 eV range, metal-centered spin-allowed but parity-forbidden ligand eld (LF) antibonding states are populated which, in conjunction with electron-phonon coupling, explain the experimental absorption intensities. The computed excitation energies are in excellent agreement with experiment. Contrary to simpler models we show that the LF absorption bands, which are important for LIESST, do not originate in transitions from the ground to a single excited state but from transitions to manifolds of nearly degenerate excited singlets. Consistent with crystallography, population of the LF states promotes a drastic dilation of the ligand cage surrounding the iron.

To the News Service home page