Using the Crystal Structures of Cystathionine β-Synthase to Understand Hydrazine Drug Interactions

Author: Andrew Rice

The described article can be found here:

Recently in this course, we discussed the important of both cystathionine β-synthase (CBS) and cystathionine β-lyase in generating H2S. H2S floats around and sulfhydrates proteins, chemically altering them in important ways. Sulfhydration itself may even be as prevalent as nitrosylation, a long studied and well-understood post-translational modification. Without this gasotransmitter, our bodies would cease to maintain homeostatic function and epigenetic maintenance.

CBS itself catalyzes a reaction in which homocysteine is converted into cystathionine. This reaction is the first of two that ultimately generate cysteine for the cell to utilize in protein synthesis. As such an important enzyme, it should come as no surprise that CBS is heavily regulated. The enzyme, in eukaryotes, consists of three separate domains. The catalytic core domain binds PLP and catalyzes the reaction described above, while the C-terminal regulatory domain is activated by binding with S-adenosylmethionine (SAM) and the N-terminal heme domain is a “remarkable redox-sensitive sensor”. However, dysregulation in this enzyme is possible, and this dysregulation comes with a host of problems. Homocysteine is toxic, and so the accumulation of too much of it due an impaired CBS enzyme (Homocysteinemia) can increase one’s risk of cardiovascular damage.

CBS is incredibly well-conserved across forms of life, but some species have a much different version of the enzyme. Despite the wide variance in forms of this enzyme, the reaction is catalyzes remains the same across different organisms and so the reaction kinetics can be applied to other forms of the same enzyme. Ringe et. al. understood this connection and successfully obtained the crystal structures for various intermediates of the CBS reaction in Saccharomyces cerevisiae, or yeast.

Yeast was chosen as the organism from which to explore this enzyme because yeast’s CBS enzyme lacks the N-terminal heme domain. This redox-sensitive sensor may be remarkable, but it’s presence has been shown to hamper UV-vis spectroscopy investigations of the enzyme. CBS, when lacking both of its regulatory domains, is much less active than its full counterpart, but still active. The exploration of just the catalytic core domain of CBS, therefore, is justified.

The authors first determined the temperature of unfolding through both CD spectroscopy and a ThermoFluor assay. From this, they determined that the CBS domain has no significant effect on thermal stabilization of the enzyme. From there, the authors described in rigorous detail the specificities in the crystal structure of CBS. The catalytic CBS domain is a head-to-tail homeodimer, and no significant changes in catalytic site structure were observed during ligand binding.

The research team also experimented with two inhibitors of CBS. The first inhibitor was D-cycloserine. Cycloserine is a common inhibitor of PLP-dependent enzymes, and enters the transamination pathway to form an intermediate that cannot proceed with the remainder of the reaction. Cycloserine was added to the CBS enzyme and CD was used. Based off of the results of the absorption spectrum, the research team does not believe the predicted adduct was formed.

The second inhibitor the team was interested in was a hydrazine (N2H4) inhibitor. Considering the inhibitor is made up of a nitrogen-nitrogen bond and some hydrogens, it’s clear how this molecule can be predicted to inhibit the CBS catalytic domain. When tested, the research team found that they expected full inactivation of CBS to occur with this inhibitor. The inhibitor didn’t appear to change the structure of the full-length CBS enzyme or the catalytic core CBS enzyme, but both were found to be thermally destabilized by their interactions with the inhibitor. The authors believe this is likely the result of a lack of successful binding with PLP; interactions with PLP likely stabilize the uninhibited form of this enzyme.

What makes this study interesting is that the research team didn’t observe what they directly expected to observe when analyzing the crystal structure of the hydrazine-inhibited CBS. The predicted intermediate was a hydrazine intermediate; once the enzyme forms the hydrazine intermediate, it theoretically couldn’t continue on with its reactions. Contrary to this prediction, however, the research team observed electron density consistent with the pyridoxamine 5-phosphate (PMP) form of the enzyme. The hydrazine intermediate, therefore, must form within a couple minutes but promptly be converted into the PMP-enzyme. Given enough time, we observe the pyridoxamine form of the cofactor, implying there must have been some new mechanistic pathway used to achieve this product.

In the accompanying figure, two possible mechanisms are proposed as the pathway from which we achieve our PMP product. The PLP hydrazone product likely tautomerizes to form the azo-product. This azo product can then undergo N-N cleavage. How this second cleavage occurs is where these two proposed mechanisms differ. Mechanism A proposes that this occurs through a second tautomerization and decarboxylation, while mechanism B proposes that this occurs through hydration and disproportionation. Both mechanisms end with the formation of the PMP-enzyme product. Regardless of the mechanistic pathway used to achieve this product, the reverse reactions occur to regenerate PLP, albeit quite slowly.

This research was able to provide clear crystal structures for the intermediate steps of this well-established mechanism, which in and of itself is impressive. It was also able to demonstrate that CBS functions to catalyze off-target reactions and does so more readily without its C-terminal and N-terminal regulatory domains. This type of research is critical in the manipulation of the body’s CBS and subsequently in cystathionine and hydrogen sulfide generation.

Similarly, the paper utilizes successful methods in obtaining the crystal structures for the various reaction intermediates and the catalytic states of the enzyme. A similar system could be derived and applied to other, less well-studied enzymes, providing a new perspective from which we can rationally design drugs with the three-dimensional structure of our target in mind. The field of computational chemistry is rapidly advancing in relation to drug discovery, and systems such as these have already significantly changed the way in which we discover new drugs and drug interactions.


1. Le, Tao, Vikas Bhushan, Matthew Sochat, Kimberly Kallianos, Yash Chavda, Andrew H Zureick, and Mehboob Kalani. First Aid for the USMLE Step 1 2017: A Student-to-Student Guide, 2017.
2. Tu, Yupeng, Cheryl A. Kreinbring, Megan Hill, Cynthia Liu, Gregory A. Petsko, Christopher D. McCune, David B. Berkowitz, Dali Liu, and Dagmar Ringe. “Crystal Structures of Cystathionine β-Synthase from Saccharomyces Cerevisiae: One Enzymatic Step at a Time.” Biochemistry, April 9, 2018.
3. Watanabe, M, J Osada, Y Aratani, K Kluckman, R Reddick, M R Malinow, and N Maeda. “Mice Deficient in Cystathionine Beta-Synthase: Animal Models for Mild and Severe Homocyst(e)Inemia.” Proceedings of the National Academy of Sciences of the United States of America 92, no. 5 (February 28, 1995): 1585–89.

4. Bakker, R. C., and D. P. M. Brandjes. “Etoposide Phosphate, the Water Soluble Prodrug of Etoposide.” Pharmacy World and Science 19, no. 3 (June 1, 1997): 126–32.

5. Jorgensen, William L. “The Many Roles of Computation in Drug Discovery.” Science 303, no. 5665 (March 19, 2004): 1813–18.

Go to Source
Click the link above to make comments on the author’s site

Powered by WPeMatico