Polishing the Key to Muscle Repair: Kyushu’s Trisulfide Makes HGF Stronger

In aging muscle, the repair signal HGF does not necessarily disappear. Two points on the molecular key can chemically “rust,” leaving it unable to turn the c-Met lock and wake muscle stem cells. Kyushu University researchers report that a small, three-sulfur molecule can protect that key—and, in laboratory tests, make it work better than before.

Between every skeletal-muscle fiber and the thin basement membrane wrapped around it lies a flattened cell waiting in reserve. Walking, lifting and merely holding the body upright produce microscopic damage. The tissue survives this wear because satellite cells—muscle’s resident stem cells—can wake, divide, become myoblasts and donate fresh nuclei to a wounded fiber.

One of their wake-up calls is hepatocyte growth factor, or HGF. Its name comes from the liver, but in muscle the active protein is stored in the extracellular matrix. Stretch or injury helps release it. HGF then docks with the c-Met receptor on a satellite cell and helps push that cell out of quiescence and into the cell cycle. It is useful to imagine HGF as a key that opens the first door of repair.

A team led by Professor Ryuichi Tatsumi at Kyushu University’s Faculty of Agriculture has found that treating HGF with lipoic acid trisulfide, abbreviated LASSS, made the protein bind c-Met more than twice as strongly as untreated HGF in an in-vitro assay. The effect remained after free LASSS was removed by ultrafiltration. Treated HGF also resisted nitration at two vulnerable amino acids, Y198 and Y250. In primary rat satellite cells, one nanogram per milliliter of LASSS-treated HGF produced maximal activation comparable to five nanograms per milliliter of ordinary HGF.

The study, published in Scientific Reports on July 24, 2026, suggests a new strategy: repair a growth signal damaged by aging instead of simply adding more of it. Yet no older person became stronger, no aged animal regained lost muscle, and no injured muscle was shown to heal faster. The in-vivo portion involved groups of three young male mice, three days of pretreatment and five days of tail suspension. It showed less HGF nitration during disuse—not restored strength. The molecule made HGF stronger at receptor binding; the study did not demonstrate stronger muscles.

1961Satellite cells first described by electron microscopy

1984Japanese researchers report the activity that led to HGF

Y198 & Y250HGF’s c-Met-binding sites vulnerable to age-linked nitration

>2×c-Met binding by LASSS-treated HGF in an in-vitro assay

1 : 8,000HGF-to-LASSS molar ratio at the standout effect

3 mice/groupYoung males in the disuse experiment—not an aging study

A fiber with thousands of nuclei

Skeletal muscle is an unusual tissue. During development, myoblasts fuse into long fibers that contain many nuclei inside one shared cell membrane. The mature nuclei manage enormous volumes of contractile machinery, but they do not simply divide and patch a major tear. New myogenic cells must come from outside the fiber proper.

After injury, immune cells clear debris and satellite cells leave quiescence. Their descendants activate a myogenic program involving MyoD, proliferate, then differentiate and fuse with an existing fiber or with one another. A minority retain the identity factor Pax7 and return to rest, preserving the stem-cell pool. Good regeneration depends on the sequence: activation, expansion, differentiation, fusion and self-renewal must happen in the right proportions and at the right times.

That is why a stronger wake-up signal is not automatically better healing. In 2000, experiments that delivered extra HGF to injured muscle increased myoblast numbers but could inhibit regeneration, depending on dose and timing, because HGF also delayed differentiation. HGF is an initiator, not a conductor of the entire repair orchestra. That history matters when reading the university’s evocative phrase “Super HGF.” Greater receptor affinity is a measured biochemical result; superior tissue repair remains an unanswered question.

The “satellite” seen at the edge of a microscope

In February 1961, Alexander Mauro of the Rockefeller Institute published a three-page electron-microscopy report on a mononuclear cell “wedged” between the plasma membrane and basement membrane of frog muscle fibers. The same month, Bernard Katz described a comparable cell in muscle spindles. Its peripheral position beside the giant fiber inspired the name satellite cell.

Mauro already suspected that the cell might be a dormant myoblast called into action during regeneration. Radioactive-thymidine tracing in the 1960s and 1970s showed that these quiet cells divided after damage and contributed nuclei to new muscle. In 2000, Michael Rudnicki’s group reported that mice lacking the transcription factor Pax7 essentially lacked satellite cells, moving the field from anatomy to molecular identity.

Selective satellite-cell depletion has since confirmed that major muscle regeneration depends on them. Their part in routine maintenance, exercise adaptation and the different stages of aging is more complicated. An old muscle is not merely a young muscle with fewer stem cells. Nerves, blood vessels, immune signals, mitochondria, extracellular matrix and the satellite cells’ own internal state all change. The Kyushu work addresses one part of that aging niche: the condition of a signal waiting outside the cell.

A growth factor born in Japan’s study of the liver

The HGF story began with one of biology’s old marvels: the liver’s ability to regrow mass after partial removal. In 1984, Toshikazu Nakamura, Katsuhiko Nawa and Akira Ichihara separated from the serum of partially hepatectomized rats an activity that promoted growth of mature hepatocytes in culture. They called it hepatotropin. Its activity rose about fivefold in rat serum 24 hours after surgery.

In 1987, Nakamura and colleagues purified the factor to homogeneity from platelets collected from 1,000 rats. Under reducing conditions it separated into 69- and 34-kilodalton subunits, revealing a new heterodimeric growth factor. In 1989, cloning of human HGF cDNA showed a 728-amino-acid precursor that is cut into the active two-chain form.

A parallel story was unfolding in Britain. In 1987, Michael Stoker and colleagues named a fibroblast-derived protein “scatter factor” because it broke up sheets of epithelial cells and made them move. In 1991, researchers learned that the liver mitogen and epithelial motility factor were the same molecule. That year, HGF’s receptor was identified as c-Met, the product of a proto-oncogene first discovered in 1984.

The convergence explained HGF’s broad reach. It can act as a mitogen that encourages cell division, a motogen that encourages movement and a morphogen that helps tissues organize in three dimensions. Research expanded from liver to kidney, lung, nerves, blood vessels and muscle. It also revealed a fundamental tension: the same HGF–MET program that repairs normal tissues can be hijacked by cancer cells for growth, invasion and metastasis.

How a stretching muscle becomes chemistry

The connection between HGF and muscle satellite cells sharpened in the 1990s. In 1995, Richard Allen and colleagues showed that HGF caused quiescent adult-rat satellite cells to enter the cell cycle earlier in culture. In 1998, Tatsumi—then at Hokkaido University—and collaborators found HGF in the extracellular matrix of uninjured adult muscle and c-Met on putative satellite cells. Antibodies against HGF abolished the satellite-cell-activating activity of crushed-muscle extract.

The next question was mechanical: how does stretch become a chemical command? Tatsumi’s work showed that stretch prompts nitric oxide, or NO, production and rapid release of matrix-bound HGF. In 2002, the group reported that blocking NO synthesis abolished stretch-induced satellite-cell activation, while adding HGF restored it. NO lay upstream, linking physical perturbation to the release of the growth factor.

But nitric oxide has a second face. When it reacts rapidly with superoxide, it forms peroxynitrite, a short-lived and highly reactive species. Peroxynitrite can add a nitro group to selected tyrosines in proteins. The long-running investigation of muscle mechanobiology eventually raised a troubling possibility: could chemistry that helps ring the repair alarm also damage that alarm over decades?

HGF need not vanish from aging muscle. It can remain in place while the surface that meets its receptor is chemically altered—a key whose teeth no longer fit the lock.

Two flecks of chemical rust

In 2024, Tatsumi’s team compared rat muscles at two, 10 and 20 months of age. Nitration of extracellular-matrix-bound HGF increased with age, especially around fast type IIa and IIx fibers. That pattern is provocative because fast fibers responsible for quick, powerful movement are preferentially affected in human aging. It does not prove that nitrated HGF causes human weakness, but it provides a testable molecular link.

The vulnerable residues are Y198 and Y250. “Y” is the one-letter code for tyrosine; the numbers locate the residues in HGF’s sequence. They sit in the K1 and K2 domains of the alpha chain, within the molecular region that binds c-Met. Peroxynitrite-driven nitration reduced receptor binding and erased HGF’s ability to activate cultured satellite cells. Under the same conditions, FGF2, IGF1 and TGF-β3 did not show the same nitration behavior, suggesting an unusual structural vulnerability in HGF.

Nitration is not a mutation of DNA. It is a nonenzymatic chemical modification after the protein is made. That makes it attractive as an intervention point: perhaps HGF function can be preserved without rewriting a gene. It also creates hard questions. How can one stop a damaging reaction without disrupting normal redox signaling, and what happens if a reactive protective compound modifies many other proteins?

The experiment with two three-sulfur compounds

The researchers tested glutathione trisulfide, GSSSG, and lipoic acid trisulfide, LASSS. Both contain three consecutively linked sulfur atoms. GSSSG is a linear molecule with a formula weight around 645. LASSS is a much smaller cyclic molecule, 238.39. Polysulfides and persulfides have emerged as “reactive sulfur species” that do more than quench oxidants; their unusual electrophilic and nucleophilic chemistry can regulate proteins and cellular redox systems.

In the first experiments, HGF was mixed with up to a 4,000-fold molar amount of either trisulfide before exposure to a 2,000-fold molar amount of peroxynitrite. Both compounds reduced nitration at Y198 and Y250. Receptor binding recovered to about 80 percent of untreated HGF. In primary satellite cells from adult male rats, each compound preserved the BrdU-incorporation response to five nanograms per milliliter of HGF that nitration would otherwise have erased.

The surprise appeared at an HGF-to-trisulfide ratio of 1:8,000. HGF mixed with LASSS bound c-Met more than twice as strongly as untreated HGF even when no peroxynitrite was present. The increase was dose-dependent above 1:4,000. GSSSG did not produce that enhancement. Ordinary alpha-lipoic acid—with a two-sulfur disulfide ring rather than LASSS’s three-sulfur ring—did not reproduce it either.

To rule out free LASSS acting directly in the receptor assay, the team washed small molecules away through a 10-kilodalton ultrafiltration membrane. The enhancement persisted with the retained HGF. LASSS by itself neither activated c-Met nor awakened satellite cells. LASSS treatment did not simply degrade HGF into a more active fragment. Together, these controls support an interaction or reaction with the HGF molecule itself.

The experimental staircase

Protect: At 1:4,000, both GSSSG and LASSS reduced HGF nitration and preserved receptor binding and satellite-cell activation.
Enhance: At 1:8,000, only LASSS more than doubled HGF–c-Met binding without nitration.
Wash: Enhancement and nitration resistance remained after unreacted LASSS was removed.
Control: GSSSG, ordinary alpha-lipoic acid and LASSS alone did not reproduce the effect.
Test in cells: One nanogram per milliliter of LASSS-treated HGF activated satellite cells as strongly as five nanograms per milliliter of control HGF.

The structural hypothesis behind “Super HGF”

HGF is a large protein folded and stabilized by disulfide bonds. The researchers point to two bonds—Cys149–Cys189 and Cys177–Cys201—just behind Y198 in a structural model of the HGF–MET complex. LASSS might reach these internal bonds and remodel them, perhaps inserting sulfur to create a trisulfide bond or a related modification. A small regional shift could expose a better receptor-binding shape while shielding Y198 from nitration.

LASSS is roughly one-third the formula weight of GSSSG, so it may reach sites inaccessible to the larger molecule. Its cyclic trisulfide may also react differently from a linear trisulfide. The stronger protection of Y198 than Y250 is consistent with distance from the proposed target bonds. That combination of enhanced affinity and nitration resistance inspired the university release’s phrase “Super HGF.”

It remains a working model. The study did not directly identify a modified residue, bond or three-dimensional structure. Mass spectrometry and high-resolution structural work must show what actually happened. LASSS-treated bovine serum albumin also became nitration-resistant, indicating that the chemistry may not be selective for HGF. Broad protein reactivity could ultimately become a safety liability rather than a virtue.

What a tail-suspension experiment can—and cannot—say

For an in-vivo clue, the team used tail suspension, part of the hindlimb-unloading tradition developed at NASA Ames Research Center beginning in the mid-1970s. Elevating a rodent’s hindquarters prevents weight-bearing and rapidly affects antigravity muscles such as the soleus. The model is useful for studying disuse and some consequences of microgravity; it is not a complete replica of bed rest, spaceflight or natural aging.

Eight-week-old male C57BL/6J mice entered four groups: untreated control; five days of suspension; three days of LASSS pretreatment followed by five days of suspension without LASSS; and the same design with GSSSG. Each group contained three mice. Pretreatment was delivered in drinking water at about 50 micrograms per gram of body weight per day.

Suspension increased Y198-nitrated HGF around fibers in the gastrocnemius, plantaris and soleus. LASSS pretreatment prevented that rise; GSSSG pretreatment did not. The persistence after dosing stopped is compatible with a lasting modification of HGF or another in-vivo mechanism, but the experiment did not establish pharmacokinetics or prove the target engagement chemistry.

The soleus showed only a trend toward lower weight after disuse, with p=0.084, and shank-muscle weight did not change. The researchers did not report restored muscle mass, force, gait, injury repair, fibrosis or fat infiltration. The mice were young, all male and few in number. This was a mechanistic test of HGF nitration during short disuse, not a treatment trial for atrophy.

Evidence level	What the study found	What remains unknown
Recombinant HGF	More than doubled c-Met binding after LASSS treatment; resistance at Y198/Y250.	Exact chemical bond, modified residue and three-dimensional mechanism.
Primary rat satellite cells	LASSS-treated HGF promoted BrdU incorporation and worked at a lower HGF concentration.	Human cells, long-term differentiation, fusion and stem-cell self-renewal.
Young male mice	Three-day pretreatment prevented HGF nitration during five-day disuse.	Muscle mass, strength, repair, toxicity, females, aged animals and chronic dosing.
Humans	Not studied.	Dose, absorption, distribution, efficacy and long-term safety, including cancer risk.

Sarcopenia is no longer a diagnosis of mass alone

After a 1988 conference, geriatrician Irwin Rosenberg proposed a name built from Greek roots for flesh and poverty: sarcopenia. Published in 1989, the term gave age-related loss of lean tissue a memorable identity. Modern diagnosis has moved beyond the amount of muscle. The European EWGSOP2 consensus places low muscle strength at the center, confirms the condition with low muscle quantity or quality, and treats poor physical performance as a sign of severity.

The Asian Working Group for Sarcopenia’s 2019 update similarly combines muscle mass with grip strength and performance measures such as walking speed or chair stands. Sarcopenia is not the same as soreness after exercise or a few days of unloading. Aging, inactivity, chronic disease, nutrition, motor-neuron loss, endocrine changes and inflammation can all contribute. Nor is it synonymous with the cytokine-driven wasting of cachexia, though the conditions can coexist.

The HGF hypothesis may explain one strand of this web: preferential nitration in fast-fiber niches, failed satellite-cell activation, poorer repair, and eventual fibrosis or fat infiltration. It is a plausible chain to test, not a clinical result. The paper explicitly states that it provides no direct evidence that oral LASSS mitigates sarcopenia or has therapeutic value. Aging experiments are still required.

LASSS is not an alpha-lipoic-acid supplement

Alpha-lipoic acid was isolated from liver in 1951 by Lester Reed and colleagues and later established as an enzyme-bound cofactor in mitochondrial energy metabolism. It has a five-membered disulfide ring containing two sulfur atoms and is widely sold as a supplement. LASSS adds a third sulfur to the ring. The formulas, structures and chemical reactivities are not interchangeable.

Ordinary lipoic acid failed to reproduce LASSS’s effect on HGF in the Kyushu experiments. The study provides no evidence that swallowing commercial alpha-lipoic acid generates enough LASSS in the body to modify HGF. LASSS remains an experimental compound with unresolved questions of stability, water solubility, metabolism and formulation. Treating the similarity of names as permission for self-medication would invert the study’s result.

The third sulfur is not a minor change in a product label. It creates the distinctive protein chemistry that ordinary alpha-lipoic acid failed to produce.

Five gates between a molecule and a medicine

The first gate is structural chemistry. Researchers must determine where LASSS binds, what fraction of HGF is modified, whether the bond is covalent and how long it persists. Mass spectrometry, nuclear magnetic resonance and cryo-electron microscopy could test the proposed disulfide remodeling.

The second is selectivity. Effects on albumin suggest that other proteins may be modified. A development program would need a broad search for changes in clotting proteins, immune signals, liver and kidney functions, and unintended targets. The third gate is pharmacology: absorption, blood chemistry, distribution into muscle extracellular matrix, active metabolites, repeated-dose toxicity, sex differences, age differences and drug interactions.

The fourth is function. Aged animals undergoing natural muscle decline should be tested for grip strength, endurance, fiber cross-sectional area, satellite-cell self-renewal, repair after real injury, fibrosis and fat infiltration. Studies must distinguish LASSS alone, LASSS plus HGF, and interactions with exercise. Timing is crucial because HGF can promote activation while delaying differentiation.

The fifth gate is intrinsic to the pathway. c-Met is a receptor tyrosine kinase essential for normal development and repair, but hyperactive HGF–MET signaling and MET mutations are implicated in multiple cancers. A brief, local repair pulse is biologically different from chronically strengthening the pathway throughout the body. The Kyushu study did not show carcinogenicity, but tumor surveillance and long-duration safety would be non-negotiable in development.

When 65 years of satellites meet 42 years of HGF

Sixty-five years after Mauro noticed a cell at the margin of an electron micrograph, and 42 years after a growth activity emerged from the serum of hepatectomized rats, those histories meet in Fukuoka around a molecule with three sulfur atoms. The approach does not transplant a new stem cell or replace a missing gene. It attempts to preserve—and perhaps tune—a repair signal that muscle already stores.

The scientific beauty lies in the comparisons. GSSSG versus LASSS. Two sulfurs versus three. A 4,000-fold molar ratio versus 8,000. Measurements before and after washing away free compound. A simple antioxidant expectation gave way to evidence for direct protein remodeling. That is how a surprising observation becomes a mechanism worth pursuing.

What comes next should be less dramatic in language and more demanding in evidence: identify the chemistry, follow the compound through an animal, test old muscles, measure actual function and examine safety over time. LASSS may or may not become a drug. Even if it does not, the design principle could endure—the idea that an age-damaged growth factor can be shielded from nitration and its receptor-binding surface deliberately tuned.

Muscle regeneration begins with a sleeping cell and a molecular key. Kyushu University’s work does not yet show that the key can reopen aging human muscle. It does show, with unusual chemical specificity, why the key may tarnish and how a third sulfur might polish it. That is a smaller claim than rejuvenation—and a far more compelling place for the science to begin.

Sources and references

This article draws primarily on Kyushu University’s release and the open-access Scientific Reports paper, supplemented by primary studies and consensus documents on satellite cells, HGF–c-Met, sarcopenia, reactive sulfur and hindlimb unloading. LASSS is not an approved medicine, and these data do not establish efficacy or safety for human muscle loss.