Could the medicine of the future do more than simply slow aging—and actually repair molecules that have already been damaged? Much of modern longevity medicine is built around one central idea: slowing the accumulation of damage. – Control blood glucose. Reduce chronic inflammation. Support mitochondrial function. Improve autophagy. Preserve muscle mass. Reduce oxidative stress.

But in July 2026, a study published in the multidisciplinary scientific journal Nature Communications proposed a fundamentally different approach: rather than preventing damage, researchers attempted to repair a protein that had already undergone chemical aging.

Researchers at Revel Pharmaceuticals, working with scientists from Calico Life Sciences and the University of Colorado, developed an engineered enzyme called CMLase. The enzyme can remove one of the most common forms of glycation-related protein damage—Nε-carboxymethyl-lysine, or CML—and restore the original lysine amino acid residue.

If this technology can eventually be made to work inside a living organism, it could open an entirely new field of regenerative medicine: molecular tissue repair. Not stimulating cells, not replacing damaged cells with stem cells, not suppressing inflammatory signaling, but directly correcting chemical defects in old proteins.

What Actually Ages Inside Our Tissues?

Proteins throughout the body are constantly undergoing chemical changes. Proteins that remain in the body for long periods are especially vulnerable. These include components of the extracellular matrix, such as collagen and other structural proteins found in the skin, blood vessels, cartilage, eyes, and connective tissues. Some of these molecules can remain in the body for years—or even decades. During that time, sugars and reactive carbonyl compounds derived from them can spontaneously interact with amino acids in proteins. This process is known as nonenzymatic glycation.

Over time, it produces compounds known as advanced glycation end products, or AGEs. AGEs are not a single molecule. They represent an entire family of chemically distinct forms of molecular damage.

One of the most common is Nε-carboxymethyl-lysine, or CML. CML forms when the side chain of the amino acid lysine undergoes chemical modification. It accumulates in long-lived proteins with age. Classic studies of human tissues have also shown increased CML accumulation during aging and in diabetes.

CML is not merely a convenient laboratory biomarker. CML-modified proteins can interact with RAGE—the Receptor for Advanced Glycation End Products. Experimental studies have shown that this interaction can activate intracellular signaling pathways associated with NF-κB and inflammatory responses. This creates an especially interesting situation from the perspective of aging biology. First, chemical damage appears within a protein. Then, the damaged protein may alter the mechanical properties of the tissue or influence how cells interact with the extracellular matrix. Finally, the same molecular modification can potentially become a source of biological signaling. In other words: the chemical aging of tissue structures can begin to influence the behavior of living cells.

The Central Problem with AGEs: The Body Is Better at Preventing Them Than Repairing Them

Cells possess powerful defense systems against reactive molecules. For example, the glyoxalase system helps neutralize methylglyoxal, a highly reactive precursor involved in the formation of certain AGEs. But there is a fundamental problem. Once a stable AGE modification has formed directly on a protein, the body has far fewer mechanisms capable of returning that molecule precisely to its original state. The researchers behind CMLase describe CML as a chemically stable modification that has long been considered essentially irreversible.

Think of a metal structure. You can protect the metal from corrosion. You can reduce humidity.

You can apply a protective coating. But once corrosion has already developed, preventing additional corrosion does not restore the damaged area. In longevity medicine, the distinction between these two approaches is fundamental. Prevention: reduce the rate at which damage forms. Repair: remove damage that has already accumulated. CMLase belongs to the second category.

What Is CMLase, Really?

Calling CMLase an “artificial enzyme” may suggest that scientists created an entirely new molecule from scratch. That is not quite what happened. It is more accurate to describe CMLase as an engineered enzyme. The researchers began with a naturally occurring family of enzymes known as glycine oxidases. The idea emerged from an intriguing chemical similarity. Part of the molecular structure of CML resembles glycine. The scientists therefore hypothesized that the active site of certain glycine oxidases might be capable of recognizing CML—at least weakly. And that is exactly what they found.

A glycine oxidase from Bacillus subtilis could interact with free CML, although its activity was relatively low and it was almost incapable of processing CML embedded within peptides. The enzyme needed to be redesigned. And that is where one of the most fascinating parts of the study begins.

Directed Evolution: 500 Million Enzyme Variants

Instead of trying to manually predict the ideal protein structure, the researchers used directed evolution. The principle resembles an accelerated version of natural selection. Scientists generate enormous numbers of protein variants. Mutations are introduced into their amino acid sequences. Researchers then select the variants that perform the desired function more effectively. The best-performing versions are mutated again. The cycle is repeated.

In the CMLase project, the researchers screened more than 500 million enzyme variants. After five rounds of directed evolution, they produced a version known as CrGO-897, which became CMLase. Compared with the original enzyme, the engineered version contained 15 amino acid substitutions and a two-amino-acid deletion. Its catalytic efficiency against CML within peptides increased by more than tenfold. More importantly, the enzyme acquired the ability to act on CML located within full-length proteins. The result was essentially a molecular machine performing a function that does not appear to exist naturally in this form.

How CMLase Repairs a Protein

This is perhaps the most remarkable aspect of the technology. CMLase is not designed to destroy the entire damaged protein. Instead, the enzyme recognizes a CML-modified residue and catalyzes its oxidation. The reaction produces: the original lysine + glyoxylate + hydrogen peroxide.

In other words, the process does more than simply remove the damaged portion of the molecule. It restores the amino acid that existed in the protein before the modification occurred. This is why the researchers use the term: protein repair.

The concept is fundamentally different from proteolysis, in which the body simply breaks down an old protein and synthesizes a new one. Here, the goal is closer to performing: molecular editing on an existing protein.

What Happened in the Experiments?

The researchers first created CML modifications artificially on several different proteins. CMLase was tested on: albumin, casein, hemoglobin, collagen, and protein extracts from eye tissue. After overnight treatment with 5 μM CMLase, detectable CML levels decreased by approximately 52% to 97%, depending on the protein.

Importantly, the researchers also tested whether the apparent reduction in CML might simply have resulted from protein degradation. SDS-PAGE analysis showed no significant protein fragmentation after enzyme treatment. In other words, CMLase appeared to remove the chemical modification rather than simply destroying the damaged protein.

The researchers then conducted more detailed proteomic analysis. In albumin, they identified 33 lysine sites modified by CML. After treatment with CMLase, reductions in CML were observed at 30 of the 33 sites. At 21 sites, the reduction exceeded 50%. At seven sites, it exceeded 90%. The enzyme also demonstrated substantial selectivity. The researchers did not observe significant oxidation of other standard amino acids, nor did they detect meaningful activity against the closely related modification carboxymethyl-arginine.

Then the Researchers Tested Human Tissues

This part of the study attracted particular attention. The scientists examined proteins from the eye lens of a 64-year-old donor. Lens proteins are extremely long-lived, making them an excellent model for studying the accumulation of molecular damage over time. After CMLase treatment, CML levels fell by approximately 45% according to LC-MS/MS analysis.

A second method, ELISA, showed a reduction of approximately 78%. According to the researchers, the difference between the two methods may be explained by the fact that antibodies preferentially detect more accessible CML modifications located on protein surfaces.

The researchers then treated sections of abdominal aorta from a 75-year-old donor. After overnight incubation with CMLase, the intensity of CML staining decreased by more than 70%.

Finally, in skin samples from an older donor, the CML signal decreased by more than 55%. After treatment, the immunohistochemical CML signal was even lower than the level observed in skin samples from a 31-year-old individual. And this is where an extremely important distinction must be made. This does not mean that CMLase transformed 75-year-old tissue into 31-year-old tissue. It altered one specific chemical marker: CML. The biological age of a tissue depends on thousands of other factors, including: collagen and elastin integrity, epigenetics, cellular senescence, mitochondrial function, inflammation, accumulated mutations, proteostasis, and many other processes. Therefore, saying that “the enzyme rejuvenated skin by several decades” would be scientifically inaccurate.

A much more accurate statement is: An engineered enzyme removed a substantial portion of one accumulated age-associated form of chemical damage from proteins in human tissue. That alone is a remarkable result.

Why CMLase Is Especially Interesting for Biohacking

Modern biohacking is gradually moving beyond broad lifestyle optimization toward increasingly precise molecular interventions.

Longevity strategies can be viewed as operating across several levels:

• reducing the formation of damage;
• enhancing the body’s natural cellular repair mechanisms;
• removing damaged cells or molecules;
• replacing lost cells;
• directly correcting accumulated molecular defects.

CMLase belongs to the final category. And this is why its significance may eventually extend far beyond a single molecule such as CML. If the principle proves broadly applicable, we could imagine an entire class of repair enzymes designed for molecular maintenance. One enzyme could remove a specific AGE modification. Another could target a different AGE. A third could repair a particular form of oxidative damage. A fourth could break pathological cross-links within the extracellular matrix.

The authors themselves describe CMLase as a proof-of-concept platform that could potentially be adapted to target other forms of chemical protein damage.

An Especially Interesting Next Target: Glucosepane

CML is far from the only problem affecting the aging extracellular matrix. One of the most intriguing AGE modifications is glucosepane. Unlike CML, glucosepane can form true cross-links between amino acid residues in proteins. Research examining the human extracellular matrix has identified glucosepane as one of the major age-related protein cross-links.

These cross-links are particularly interesting in longevity science because they may contribute to the increasing stiffness of long-lived protein structures. The CMLase researchers themselves point to glucosepane as an example of another form of molecular damage for which specialized enzymes might eventually be developed.

If bioengineering can systematically create enzymes against different AGEs, it could lead to an entirely new therapeutic concept: Enzymatic Rejuvenation of the Extracellular Matrix.

For now, this remains a hypothesis. But CMLase makes the concept considerably less speculative.

Why CMLase Cannot Yet Be Considered an Anti-Aging Therapy

Despite the striking results, there is an enormous distance between a laboratory experiment and a clinically useful therapy. The study’s authors clearly describe these limitations. The experiments were performed in vitro and ex vivo. Researchers worked with isolated proteins, tissue homogenates, or thin fixed tissue sections. The situation inside a living organism is entirely different.

An enzyme would need to: reach the bloodstream or target tissue, remain stable, penetrate the extracellular matrix, and reach chemically modified regions of proteins. The aging extracellular matrix itself may be dense and heavily cross-linked, potentially making it difficult for a relatively large protein enzyme to penetrate.

There is also the issue of the immune system. CMLase was engineered from an enzyme of bacterial origin. The human immune system could potentially recognize such a protein as foreign. The authors therefore explicitly identify immunogenicity as one of the challenges that would need to be solved before clinical use.

Another question concerns the products of the reaction.

CMLase generates glyoxylate and hydrogen peroxide. The researchers suggest that because the absolute amount of CML in tissues is relatively small, exposure to these reaction products may also be small compared with normal metabolic fluxes. However, the safety of this mechanism inside living organisms still requires investigation. And perhaps the most important question is: Will removing CML actually improve tissue function?

The study demonstrated chemical repair. It did not yet demonstrate that: an artery becomes more elastic, skin becomes functionally younger, or an organ begins functioning better. The authors acknowledge that improvements in tissue biomechanics and changes in pathological RAGE signaling still need to be demonstrated in vivo.

But the Idea Behind This Study Goes Much Deeper Than CML

Aging biology has long contained an implicit distinction. Cellular processes often appear potentially reversible. Gene expression can be changed. Signaling pathways can be modified. Autophagy can be stimulated. The epigenetic state of cells can potentially be reprogrammed. Old structural proteins, however, have often been viewed as something closer to a passive archive of accumulated chemical damage.

CMLase challenges that assumption. For at least one AGE modification, it demonstrates that: an old form of chemical protein damage does not necessarily have to remain permanent. It can be recognized. It can be catalytically removed. And the modified region of the protein can be returned closer to its original chemical state.

This changes the question itself. Instead of asking: How can we make the body accumulate damage more slowly?; we can begin asking: Which forms of aging-related damage can we learn to repair directly?

CMLase and the Future of Longevity

For now, CMLase is probably best viewed not as a ready-made anti-aging treatment, but as a technological demonstration of a new class of interventions. Before real medical applications become possible, several major steps will be required: confirming efficacy in living organisms, determining how the enzyme distributes throughout tissues, reducing or eliminating immunogenicity, establishing safety, demonstrating functional tissue recovery, and only then progressing toward full clinical translation. The published study does not yet answer these questions.

Furthermore, CML represents only one component of the enormous molecular landscape of aging. Removing a single AGE modification will not stop cellular senescence. It will not repair damaged DNA. It will not correct mitochondrial defects. And it will not eliminate other forms of extracellular cross-linking. CMLase is therefore unlikely ever to become an “enzyme against aging.”

But the more interesting possibility is something different. It could become one of the first representatives of an entirely new philosophy in longevity: not only protect—but repair. If enzymes targeting glucosepane and other persistent age-related molecular modifications eventually follow CMLase, bioengineering could begin approaching something that until recently seemed nearly impossible: programmable molecular maintenance of aging tissues.

From Preventive Biohacking to Molecular Engineering of the Human Body

The first era of biohacking focused primarily on behavior: sleep, nutrition, physical activity, and stress management. The second increasingly focuses on biomarkers: glucose, lipids, hormones, inflammation, and metabolomics. The third is beginning to intervene directly in biological mechanisms through: cell therapy, gene therapy, epigenetic reprogramming, and regenerative medicine.

CMLase hints at a possible fourth era: Molecular Repair In this model, aging is viewed not only as a process that should be slowed, but also as the accumulation of specific chemical defects—each of which could potentially become a target for a specialized biological tool. This is not yet a technology for rejuvenating humans. But the experiment demonstrates something fundamentally important: a molecule that accumulated in human proteins over decades and was considered an essentially irreversible mark of time proved accessible to targeted enzymatic repair.

That is why CMLase may ultimately matter to longevity science far more than the three letters “CML” might suggest. We may be witnessing one of the first steps in a transition from the concept of: “slowing aging” to a far more ambitious objective: “learning how to repair its molecular consequences.”

Scientific Basis
The primary study is: Narisa Trabosh et al., “Reversal of Protein Chemical Aging by Enzymatic Deglycation,” Nature Communications, Volume 17, Article 5926, published July 14, 2026. The research involved Revel Pharmaceuticals, Calico Life Sciences, and the University of Colorado Anschutz Medical Campus and was supported by grants from the National Institute on Aging. Revel Pharmaceuticals filed a patent application covering the technology described in the study, and two of the study’s authors are listed as inventors. This article is intended for scientific and educational purposes only. CMLase remains an experimental technology and should not be considered an available treatment or rejuvenation therapy.