My laboratory is interested in the molecular mechanisms that underly physiological processes.   We are investigating the relationship between protein structure and function in enzyme catalysis and inhibitor action and applying this knowledge to understand the effects of mutations in vivo.  This work involves a wide array of experimental techniques, including enzyme kinetics, site directed mutagenesis, organic synthesis, protein purification and various methods of monitoring protein structure. We choose problems of pharmacological interest and/or paradigms of enzyme action.  Our ongoing projects include:

IMP dehydrogenase: mechanism, drug design and retinitis pigmentosa
Acireductone dioxygenase
Serine proteases: serpins, engineering specificity, zymogen activation, streptokinase and PSA
Aptamers
                                  

IMP dehydrogenase (IMPDH) 

Mechanism IMPDH catalyzes the pivotal step in guanine nucleotide biosynthesis: the conversion of IMP to XMP with concomitant reduction of NAD.  IMPDH controls the guanine nucleotide pool, which in turn controls proliferation and many other important cellular processes.  IMPDH is a target for immunosuppressive, cancer and antiviral chemotherapy.  We have

discovered that IMPDH undergoes a large conformational change in mid-catalytic cycle that converts the enzyme from a dehydrogenase to a hydrolase, and thus present s a unique opportunity to investigate the role of conformational dynamics in catalysis.  This conformational change also appears to be a major determinant of drug selectivity.  Lastly, IMPDH is activated by monovalent cations in what may be another dynamic process. The dehydrogenase step of the IMPDH reaction produces NADH and a covalent E-XMP* intermediate.  After NADH departs, a large conformational change transforms the enzyme into a hydrolase that converts E-XMP* to XMP.  This conformational change positions a conserved Arg-Tyr dyad in the NADH site adjacent to the E-XMP*.  Surprisingly, the Arg residue appears to act as the general base.  We are investigating the mechanism of water activation and the function of the conserved Arg-Tyr dyad using an array of kinetic and spectroscopic techniques.  We have delineated a complete kinetic mechanism for T. foetus IMPDH that greatly facilitates these studies.

Mycophenolic acid (MPA) is a potent inhibitor of human IMPDH but a poor inhibitor of the T. foetus enzyme.  MPA competes with the flap for the NADH site, blocking the closed conformation and hydrolysis of E-XMP*.  Thus the equilibrium between “open” and “closed” conformations may determine drug selectivity as well as catalysis.  Clearly, this is a critical issue for drug design.  We are initiating a series of structure/function experiments to correlate the conformational equilibrium with drug selectivity in the human and T. foetus enzymes.

Monovalent cation activation is a widespread but poorly understood phenomenon.  K⁺ is generally believed to have a passive role, simply stabilizing enzyme structure.  However, in the case of IMPDH, K⁺ appears to associate/dissociate during the IMPDH catalytic cycle, permitting the enzyme to transiently access different conformations for each catalytic step.  We are testing this model by determining the effect of K⁺ on individual steps in the reaction.

Drug design Cryptosporidium parvum  is an important AIDS pathogen and a potential agent for bioterrorism.  No drugs or vacines currently exist to treat C. parvum infections.  We are collaborating with Boris Striepen of the University of Georgia to determine if IMPDH and other enzymes of the purine nucleotide salvage pathways are valid targets for chemotherapy.  Our preliminary experiments suggest that C. parvum IMPDH is significantly different from the human isozymes, and we are currently trying to develop selective inhibitors of the parasite enzyme.

Retinitis pigmentosa
Retinitis pigmentosa (RP) is the most common hereditary blindness.  RP can often be attributed to mutations in photoreceptor specific proteins; defects in such proteins can easily be expected to compromise photoreceptor function leading to apoptosis.  However, autosomal dominant RP also results from mutations in several ubiquitously expressed proteins.  One such protein is inosine monophosphate dehydrogenase type 1 (IMPDH1).  The RP-causing mutations are within a subdomain of unknown function and do not effect enzymatic activity.  We have made the surprising discovery that all IMPDHs bind nucleic acids with high affinity, and that the nucleic acid binding site is the subdomain.  This is the first and only function assigned to the subdomain.  Since the physiological role of this nucleic acid binding activity has not been elucidated, the effects of these mutations on photoreceptor function cannot be understood.  We are currently trying to define the physiological consequences of IMPDH's nucleic acid binding activity, and describe the perturbations resulting from the three RP-associated mutations of IMPDH1. 

Selected Publications:

Gan, Lu; Seyedsayamdost, Mohammad R.; Shuto, Satoshi; Matsuda, Akira; Petsko, Gregory A. & Hedstrom, Lizbeth.  The immunosuppressive agent mizoribine monophosphate forms a transition state analog complex with IMP dehydrogenase. Biochemistry 42, 857-863 (2003).  [abstract] [pdf]

Gan, Lu; Petsko, Gregory & Hedstrom, Lizbeth. Crystal structure of a ternary complex of Tritrichomonas foetus inosine 5’-monophosphate dehydrogenase:  NAD orients the active site loop for catalysis. Biochemistry 41, 13309-1317 (2002). [abstract] [pdf]

Digits, Jennifer A. & Hedstrom, Lizbeth. Drug selectivity is determined by coupling across the NAD site of IMP dehydrogenase. Biochemistry, 39, 1771-1777. [abstract]  [pdf]

Digits, Jennifer A. & Hedstrom, Lizbeth.  The kinetic mechanism of Tritrichomonas foetus inosine 5’-monophosphate dehydrogenase.  Biochemistry 38, 2295-2306 (1999). [abstract]  [pdf]

Acireductone dioxygenase We are characterizing sipL, the human acireductone dioxygenase (ARD) in collaboration with Tom Pochapsky.  Klebsiella ARD is an Fe-containing enzyme that catalyzes an oxidative cleavage.  This reaction is the penultimate step in the methionine salvage pathway.  However, Klebsiella ARD can also contain Ni; this form catalyzes a different oxidative cleavage reaction that produces CO.  The human ARD homolog plays a role in cancer and viral infection.  Selected Publications: Pochapsky, T.C., Pochapsky, S.S., Ju, T., Mo, H., Al-Mjeni, F., Maroney, M.J. Modeling and experiment yields the structure of acireductone dioxygenase from Klebsiella pneumoniae.  Nat Struct Biol. 9, 966-972 (2002). [abstract] [pdf]

Serine proteases

Mechanism of serpin action

Engineering specificity  

We are investigating the structural basis of enzyme specificity in the trypsin family of serine proteases. The initial goal of this project was to convert trypsin, a protease with primary specificity for positively charged residues, into a protease with primary specificity for hydrophobic residues (i.e., a chymotrypsin-like protease). We succeeded in engineering a trypsin variant with 10% of the activity of chymotrypsin on hydrophobic substrates, involving a 1010-fold change in substrate preference.   This transformation required the substitution of two surface loops which do not contact the substrate in addition to the residues at  the substrate bindng site.  Thus specificity is determined by a network of interactions which extend beyond the enzyme-substrate interface.  Our best mutant catalyzes the hydrolysis of enzyme-bound substrate at  rates comparable to chymotrypsin.  However, the structure of the re-engineered primary substrate binding site is deformed which causes a defect in substrate binding.

We also attempted to convert trypsin into an elastase-like protease, i.e., a protease which cleaves after small aliphatic residues in order to determine if the residues identified in the trypsin-to-chymotrypsin experiment are a general set of specificity determinants.  Substitution of trypsin with the analogous residues of elastase did not create an elastase-like protease, although it did produce an elastase-like esterase.  These experiments demonstrate that unique structural solutions are required for each different specificity.

Legend:  The S1 sites of trypsin (blue) and chymotrypsin (purple).  A fragment of bovine pancreatic trypsin inhibitor corresponding to a peptide substrate is shown in red.  Loops 1 and 2 are structural determinants of specificity although they do not contact the substrate.

We extended this work to change the specificity of the secondary  sites. We changed the specificity of the S1' site of trypsin from a preference for hydrophobic residues to a preference for positively charged residues, creating a protease that cleaves specifically between two Arg residues.

Selected Publications:

Kurth, Torsten; Grahn, Sibylla; Thormann, Michael; Ullman, Dirk; Hofman, Hans-Jorg; Jakubke, Hans-Dieter and Hedstrom, Lizbeth. Engineering the S1'-subsite of trypsin: design of a protease which cleaves between dibasic residues. Biochemistry 37, 11434-11440 (1998). [abstract]

Hung, Su-Hwi & Hedstrom, Lizbeth. Converting trypsin to elastase: Substitution of the S1 site and adjacent loops reconstitutes esterase specificity but not amidase activity. Protein Engineering, 11, 669-673 (1998). [abstract]

Perona, John J.; Hedstrom, Lizbeth; Rutter, William J. & Fletterick, R.J. Structural Origins of Substrate Discrimination in Trypsin and Chymotrypsin. Biochemistry 34, 1489-1499 (1995). [abstract]

Hedstrom, Lizbeth; Perona, John J. & Rutter, William J. Converting Trypsin to Chymotrypsin: Residue 172 is a Specificity Determinant. Biochemistry 33, 8757-8763 (1994).  [abstract]

Hedstrom, Lizbeth; Szilagyi, Laszlo & Rutter, William J. Converting Trypsin to Chymotrypsin. Science 255, 1249-1253 (1992).  [abstract]

Understanding zymogen activation

The complex interplay between protein conformation, dynamics and electrostatics is dramatically illustrated in the conformational change that activates serine proteases. Trypsin is synthesized as an inactive precursor, trypsinogen.  This zymogen is activated by removal of a peptide from the N-terminus. The new N-terminus, Ile16, forms a salt bridge with Asp194. This interaction triggers a conformational change involving four disordered segments of trypsinogen.  These segments form the rigid S1 substrate binding site and oxyanion hole. Thus the conformational change is essentially an example of protein folding, albeit on a smaller and more confined scale. Trypsinogen can be considered to exist in two conformations, the inactive zymogen conformation and the active trypsin-like conformation. These two conformations are in equilibrium, with K_eq = 10⁹ in favor of the inactive conformation.  We are studying the effect of mutations on both the active trypsin conformation and the inactive trypsinogen conformation. Our work has focused on the contribution of a key salt bridge between Asp194 and the N-terminus, Ile16. We have also identified several substitutions which "activate" trypsinogen in the absence of the salt bridge interaction.  Most importantly, we have demonstrated that BPTI affinity correlates with activity.  This correlation allows us to use the structures of the trypsinogen-BPTI complexes to identify the structural features which confer activity.  In addition, characterization of trypsin(ogen) variants which do not obey this correlation will allow us to probe fundamental relationships between enzyme catalysis and protein conformation, dynamics, and electrostatics. 

Selected Publications:

Pasternak, Annette; White, Andre; Jeffrey, Constance; Medina, Nivia; Cahoon, Marguerite;
Ringe, Dagmar & Hedstrom, Lizbeth.  The energetic cost of induced fit catalysis: crystal structures of trypsinogen mutants with enhanced activity and inhibitor binding.  Protein Science 10, 1331-1342 (2001). [abstract]

Mechanism of streptokinase action

Zymogens can also be activated without proteolytic processing by forming coplexes with protein cofactors.   The activation of plasminogen by streptokinase is an example of this phenomenon.  The conversion of plasminogen to plasmin is the final step in the fibrinolytic pathway; plasmin is the protease responsible for the lysis of blood clots and plasminogen activators are used to clinically to treat heart attacks and strokes.  Usually this activation involves proteolysis and formation of a salt bridge in a manner analogous to the activation of trypsinogen. However, unlike typical plasminogen activators, streptokinase is not a protease (nor is it a "kinase").  Rather, it forms an active complex with plasminogen (SK•Plgn*); this complex can proteolytically activate additional plasminogen molecules. We hypothesized that the N-terminus of streptokinase forms a salt bridge in the SK•Plgn* complex.  We have shown that deletion of Ile1 of streptokinase prevents formation of SK•Plgn*, consistent with this hypothesis.  In addition, we have shown that  fibrin can restore the activity of some inactive strepokinase mutants. These fibrin-dependent variants should have improved therapeutic properties.  This work is a collaboration with Guy Reed at Harvard School of Public Health.

Reed, Guy L.; Houng, Aiilyan K.; Liu, Lin; Parhami-Seren, Behnaz; Matsueda, Lee H., Wang, Shunguang and Hedstrom, Lizbeth. A catalytic switch and the conversion of streptokinase to a fibrin-targeted plasminogen activator. Proc. Natl. Acad. Sci USA 96, 8879-8883 (1999). [abstract]

Wang, Shunguang; Reed, Guy L. & Hedstrom, Lizbeth. Deletion of Ile1 of streptokinase impairs plasminogen activation: evidence for the molecular sexuality hypothesis. Biochemistry 38, 5232-5240 (1999).  [abstract]

Pasternak, Annette; Liu, Xiaolin & Hedstrom, Lizbeth. Activating a zymogen without proteolytic processing: mutation of Lys15 and Asp194 activate trypsinogen. Biochemistry 37, 16201-16210 (1998).  [abstract]

PSA
 PSA is a chymotrypsin-like serine protease, albeit a very poor one, with ~10⁴ less activity than chymotrypsin.  PSA has only one known physiological function: to degrade the seminal coagulum, releasing sperm.  We have discovered that PSA is activated by anions, with the highest activity observed in the presence of citrate.  This salt activation is accompanied by a conformational change.  Interestingly, the ionic strength of seminal fluid is 0.5 N, composed primarily of citrate, which is more than 3 times greater than the ionic strength of the cytosol.  This suggests that anion concentrations may regulate PSA activity in vivo.

Elucidation of the role of PSA in prostate cancer biology has been frustrated by a plethora of apparently contradictory observations.  With recombinant enzyme in hand, we are now investigating the effects of PSA on the progression of prostate cancer.

Selected Publications:

Huang, Xinyi ;  Knoell, Christopher T.;  Frey, Gary;  Hazegh-Azam, Maryam; Tashjian, Jr.,  Armen H.;  Hedstrom, Lizbeth & Abeles, Robert H. Recombinant human prostate-specific antigen: anion activation and azapeptide inhibitors.  Appendix:  thermodynamic interpretation of the activation by concentrated salts by S. N. Timasheff. Biochemistry 40, 11734-41 (2001). [abstract]

Looking for the aptamer project?
This project has moved to industry:  Dr. Martin Stanton has started Archemix to exploit aptamers in the treatment of chronic and acute disease.

updated: July 2003