Sugar molecules in our bodies, derived primarily from food, can spontaneously adhere to various proteins, a process called glycation. Glycation can form dangerous Advanced Glycation End Products (AGEs) that lead to various pathologies like Alzheimer’s disease and diabetes, but it can also disable proteins that help cancer cells proliferate. In the early 2000s, scientists discovered that an enzyme called fructosamine-3-Kinase (FN3K) reverses protein glycation. That has made FN3K a valuable target for drug developers hoping to control when and where glycation occurs.
The data needed for such work has been lacking. But a new study published in Nature Communications involving high-resolution structures determined from data collected at the Advanced Photon Source (APS), a U.S. Department of Energy (DOE) Office of Science user facility at DOE’s Argonne National Laboratory, reveals how FN3K deglycates a protein. These findings can serve as the basis for structure-based and in silico drug design targeting FN3K.
Glycation normally occurs in the bloodstream, but it can happen quickly and spontaneously wherever sugar levels are high. One place would be a tumor microenvironment, where cancer cells require the energy from sugar to proliferate.
Proliferation of cancer cells is also aided by a protein called NRF2 (nuclear factor erythroid 2-related factor 2). This transcription factor regulates genes involved in cell growth and survival. It can both suppress and promote tumors, depending on the type of cancer and what stage it’s in. Early on, it’s thought to suppress tumors; later, it’s thought to promote them.
When NRF2 is glycated, it loses its stability and is rendered ineffective at protecting cancer cells. But NRF2 can regain this detrimental function when the sugar is removed—deglycated—by the enzyme FN3K, according to a 2019 study. This study opened two new ways of thinking about how to limit cancer cell proliferation: targeting NRF2 directly or modulating its activity through FN3K—a back door approach that would prevent the enzyme from deglycating NRF2.
With the 2019 study in mind, the scientists behind the current research set out to explore the therapeutic potential of FN3K. They determined a series of crystal structures of human FN3K (HsFN3K) in its unbound, or apo, form. Some variation of a deglycating enzyme occurs in nearly every form of life; the human form is the only one to feature the amino acid tryptophan near its core catalytic site.
The scientists also determined crystal structures of HsFN3K bound to an analog of glycated NRF2 and the nucleotide ATP triggering different catalytic states—ATP prior to phosphorylation, ADP following phosphorylation, and AMPPNP, an ATP analog, for comparison. The X-ray diffraction data were collected at the Northeast Collaborative Access Team (NE-CAT) beamline at 24-ID-E of the APS.
With careful scrutiny of such high-resolution structures, the team deciphered what no one had seen before. In the pre-catalytic state, the tryptophan recognized the ATP, then flipped 180 degrees. That caused a conformational change in the sugar moiety on the NRF2 analog that made it receptive to phosphorylation; addition of the phosphate destabilized the sugar half and removed it from the protein. Had the resolution been any lower, scientists would not have perceived and recognized the importance of the tryptophan flip. Analyzing their structures, they found that it did not occur with ADP or AMPPNP; it only happened with ATP.
What about tryptophan? To investigate its role, the team substituted a different amino acid to see if deglycation would still take place. It didn’t, leading the team to hypothesize that tryptophan may function as an ATP sensor, promoting HsFN3K’s kinase activity, even though it’s not part of the usual kinase classical active site players. But one tryptophan substitution gave the opposite result—changing it to a histidine, present in several versions of this enzyme from other species, made HsFN3K unusually hyperactive.
These findings have broad implications for advancing scientific knowledge and conducting basic research. The scientists hypothesize that in humans, evolution enhanced cellular homeostasis by slowing down HsFN3K’s glycating activity through tryptophan, creating an advantageous baseline level of glycation.
As for basic research, many scientists use a derivative of ATP that doesn’t trigger phosphorylation (AMPPNP) to study how ATP interacts with proteins without bothering with catalytic reaction. Only one atom distinguishes ATP from AMPPNP, but that one atom makes all the difference in the world, according to the authors of this study. They found that AMPPNP sits slightly differently from ATP in HsFN3K, preventing the tryptophan from flipping. While they don’t believe that their findings invalidate studies using AMPPNP, they do believe that in some cases, scientists should be careful how they interpret findings.
Read more on Argonne website
Image: A representation of the crystal structure of HsFN3K in complex with ATP and the sugar mimic substrate DMF. The typical kinase fold is shown with its N-lobe in blue and C-lobe in green. DMF is positioned next to ATP in the catalytic site, both shown in stick representation. Tryptophan W219, shown with electron density shown as a mesh, is in a flipped conformation induced by the ATP binding (top inset), while in the presence of ADP it does not flip (bottom inset). The sugar (DMF) conformation is also different in the two states, underscoring the precise placement of the molecules involved for a productive reaction and providing important mechanistic insights into glycation by FN3K.