Reversal of Protein Chemical Aging Through Engineered Enzymatic Deglycation

The field of geroscience has long operated under the assumption that certain forms of molecular damage, particularly the chemical modifications that accumulate on long-lived proteins, are effectively permanent. However, a landmark study published in Nature Communications by a collaborative team of researchers from Calico Life Sciences, Revel Pharmaceuticals, and the University of Colorado has challenged this paradigm. By engineering a specialized enzyme capable of stripping away advanced glycation end products (AGEs) from human proteins, the team has provided the first concrete evidence that the chemical "aging" of the extracellular matrix can be enzymatically reversed. This breakthrough offers a potential roadmap for treating a variety of age-related conditions, from arterial stiffness and skin degradation to chronic inflammatory diseases.

The Biochemical Architecture of Protein Aging

To understand the significance of this discovery, one must first look at the lifespan of proteins within the human body. While many intracellular proteins are recycled within hours or days, the proteins that make up the extracellular matrix (ECM)—such as collagen and elastin—are remarkably durable. In tissues like the skin, heart, and joints, these proteins can persist for years, or even decades. This longevity, however, comes at a cost: exposure to the body’s internal chemical environment leads to the spontaneous, non-enzymatic attachment of sugars to amino acids, a process known as glycation.

Glycation begins with the Maillard reaction, the same chemical process responsible for the browning of food. Over time, these initial sugar attachments undergo complex rearrangements to become chemically stable, irreversible structures known as Advanced Glycation End Products (AGEs). The accumulation of AGEs is a hallmark of biological aging. These modifications act as "molecular grit," forming crosslinks that bind neighboring protein fibers together. This crosslinking is the primary driver of tissue stiffening, which manifests clinically as systolic hypertension, loss of skin elasticity, and reduced joint mobility. Furthermore, certain AGEs act as ligands for the Receptor for Advanced Glycation Endproducts (RAGE), triggering a cascade of pro-inflammatory signaling that contributes to the systemic "inflammaging" characteristic of the elderly.

Identifying the Target: N-epsilon-carboxymethyl-lysine (CML)

The research team focused their efforts on a specific, prevalent AGE known as $N^epsilon$-carboxymethyl-lysine (CML). CML is formed when the amino acid lysine is modified by glyoxal or other sugar-derived precursors. It is one of the most widely studied AGEs because it serves as a reliable biomarker for oxidative stress and long-term glycemic exposure.

In clinical settings, elevated levels of CML are associated with diabetic complications, neurodegenerative diseases, and cardiovascular stiffening. Despite its prevalence, CML has been historically viewed as a terminal modification—a "dead end" for the protein. The human body lacks an endogenous enzyme capable of efficiently removing CML once it has integrated into a folded protein structure. The challenge for the researchers was to find or create a biological catalyst that could recognize CML and convert it back into functional lysine without damaging the underlying protein backbone.

The Search for a Molecular Catalyst

The investigation began with a massive bioinformatic search. The researchers hypothesized that certain bacteria might possess enzymes capable of processing CML as a nutrient source. They initially looked at glycine oxidases—enzymes that break down glycine, a molecule that bears a strong chemical resemblance to the carboxymethyl-lysine group.

Early tests with a glycine oxidase from Bacillus subtilis showed promise, as it was able to process "free" CML (individual molecules not attached to a protein). However, the enzyme failed when presented with CML embedded in a peptide chain. This indicated a fundamental structural hurdle: the enzyme’s active site was physically blocked from accessing the CML when it was part of a larger, more complex protein structure.

To overcome this, the team utilized the AlphaFold protein structure database to analyze over 44,000 potential enzyme candidates. They were searching for a variant that lacked a specific structural element known as the $alpha$9 helix, which they identified as the primary obstruction preventing the enzyme from docking with peptide-bound CML. This computational screening eventually identified a rare enzyme variant that showed a faint, but detectable, ability to act on CML within a peptide model.

Weaponized Evolution: The CMLase Development

With a baseline enzyme identified, the researchers turned to directed evolution—a laboratory technique that mimics natural selection to "breed" better enzymes. They created a massive library of millions of mutated enzyme variants and introduced them into a specialized strain of E. coli.

This E. coli strain was engineered to be "lysine auxotrophic," meaning it could not produce the essential amino acid lysine on its own. To survive, the bacteria had to "harvest" lysine from CML provided in their growth medium. This created an intense evolutionary pressure: only the bacteria carrying an enzyme variant capable of efficiently converting CML back into lysine would survive and multiply.

Through five grueling rounds of mutation and selection, the team successfully "weaponized" the enzyme’s evolution. Each round selected for variants that were more stable, had a more open active site, and were less dependent on the specific surrounding amino acid sequence of the target protein. The result was a highly optimized enzyme the researchers named "CMLase."

Experimental Validation and Tissue Results

The efficacy of CMLase was first tested on bovine serum albumin (BSA) that had been artificially glycated in the lab. CMLase successfully removed the CML modifications from the BSA without causing any degradation to the protein itself. Encouraged by these results, the team moved to naturally aged human tissues, which represent a much more complex "real-world" challenge.

The results were significant across several tissue types:

• Human Lens Proteins: The researchers treated soluble lens proteins from a 64-year-old donor. Lens crystallins are among the longest-lived proteins in the body, accumulating damage over decades. CMLase successfully reduced the endogenous CML burden in these proteins.
• Human Skin: In skin samples, CMLase treatment reduced CML levels by 55%. Histological staining revealed that after treatment, the CML density in the aged skin was lower than that typically observed in a 31-year-old.
• Human Arteries: Perhaps most significantly, the enzyme reduced the CML burden in aged human arterial tissue by 70%. Given that arterial stiffening is a major cause of cardiovascular disease in the elderly, this result highlights the potential therapeutic value of the enzyme.

Chronology of the Discovery

The path to CMLase represents a multi-year effort that reflects the changing landscape of longevity research:

• 2010s: Theoretical frameworks for "Damage Repair" (SENS) gain traction, identifying AGE-cleaving as a primary goal for rejuvenation biotechnology.
• 2020-2022: Calico and Revel Pharmaceuticals begin a deep-dive into bioinformatic screening for deglycation enzymes.
• 2023: Identification of the structural role of the $alpha$9 helix in obstructing enzyme access to glycated peptides.
• 2024: Completion of the five-round directed evolution process in E. coli.
• 2025: Successful testing on human donor tissues (skin, lens, and artery).
• 2026: Formal publication of the findings in Nature Communications.

Analysis of Implications and Future Challenges

While the creation of CMLase is a historic milestone, the research team and the broader scientific community acknowledge that several hurdles remain before this can be translated into a clinical therapy.

One primary concern is immunogenicity. Because CMLase is derived from bacterial origins, introducing it into the human body could trigger an immune response. Future iterations may require "humanizing" the enzyme or developing delivery systems that shield it from the immune system. Furthermore, the study was conducted on thin tissue sections in a laboratory setting. Delivering a large enzyme molecule deep into the intact, living tissues of a human patient—such as into the walls of the aorta or the deep layers of the dermis—remains a significant pharmacological challenge.

Moreover, CML is only one type of AGE. To fully rejuvenate the extracellular matrix, scientists will likely need to develop a "cocktail" of enzymes. The next major target is expected to be glucosepane, a complex crosslink that is much more difficult to break than CML and is believed to be the primary cause of age-related tissue rigidity.

Expert Reactions and Industry Impact

The study has sent ripples through the biotechnology sector. "This is the first time we’ve seen a designed biological tool capable of undoing the chemical wear-and-tear that we’ve long considered a one-way street," noted one independent researcher in the field of proteostasis.

The involvement of Calico Life Sciences, which is funded by Alphabet (Google), underscores the high-stakes nature of this research. Revel Pharmaceuticals, a startup specifically focused on AGE-cleavage, has signaled that this proof-of-concept validates their broader mission to treat aging as a repairable problem of molecular biology.

The successful engineering of CMLase shifts the conversation in geroscience from "slowing down" the clock to "turning it back." If the stiffness of the cardiovascular system and the inflammation of the skin can be reversed at the molecular level, the implications for human healthspan are profound. This research suggests that the "irreversible" damage of time may simply be a matter of finding the right biological key to unlock the chemical bonds of aging.