← Back to Research Library

MGF and PEG-MGF: What the Research Shows

MGF (mechano growth factor) is the name given to a form of IGF-1, called IGF-1Ec in people and IGF-1Eb in rodents, that the IGF-1 gene produces by alternative splicing and that a London group found in rabbit muscle subjected to stretch. The research peptide called MGF is a synthetic 24-amino-acid copy of the end of that form, its E domain, and PEG-MGF is the same kind of peptide with a polyethylene glycol (PEG) chain attached.

Below is what the studies looked at, in what kind of model, and what they reported. This is a summary of published research, not a description of what the compound does in a person.

Quick facts

• Names: MGF, mechano growth factor, IGF-1Ec (the human form), IGF-1Eb (the rodent form), MGF E peptide, PEG-MGF, pegylated MGF. The human and rodent forms carry a 49-base and a 52-base insert in the gene’s message, which is why the letters differ (Kandalla et al., 2011).
• Sequence: the peptide used in a 2013 heart study is YQPPSTNKNTKSQRRKGSTFEEHK, 24 amino acids, described as the unique end of the human form (Mavrommatis et al., 2013). The UniProt reference entry for human IGF-1 lists this form ending in EERK rather than EEHK (UniProt P05019, isoform 4), so published sequences do not all match.
• Modified forms: one research team stabilized the peptide by amidating its end and switching the two arginines at positions 14 and 15 to the D form (Mavrommatis et al., 2013). PEG-MGF adds a PEG chain. We did not find a published structure that fixes the PEG size or where it attaches.
• Formula and molecular weight: these depend on which sequence and which modifications a given synthetic version uses, so this page does not print one. PubChem holds no standard compound record for MGF or PEG-MGF.
• Class: a peptide fragment of an IGF-1 splice form. It is not mature IGF-1, and not a growth hormone secretagogue. MGF shares the mature IGF-1 region and differs in the E domain at the end (Carpenter et al., 2008). The mature hormone is covered in the IGF-1 research summary.
• Approval status: neither MGF nor PEG-MGF is an approved drug in the United States, and the FDA’s Drugs@FDA database lists no product containing either.

What the research found

The record splits into three parts: work on the natural IGF-1 form in muscle, cell and animal studies of the synthetic peptide, and later work that questioned the peptide. Results are listed in date order, as the authors reported them.

The form in stretched muscle (Yang, Journal of Muscle Research and Cell Motility, 1996). A London group led by Geoffrey Goldspink compared rabbit muscle subjected to active stretch with resting muscle. Resting muscle carried one IGF-1 form, IGF-1Ea. Stretched muscle also carried a second, IGF-1Eb, whose end differs because a 52-base insert shifts the reading frame. With primers for both species, they detected this form (IGF-1Ec in humans) after 2 hours to 6 days of altered activity, and very little of it in muscle without stretch or extra activity. Rabbits. The primers were built to detect the human form too. PMID 8884603.

The name (McKoy, Journal of Physiology, 1999). The same group called the second form mechano growth factor. In rabbit leg muscle, stretch alone raised both IGF-1 forms within 4 days, stretch with electrical stimulation raised them more, and stimulation alone did not raise them significantly over sham-operated controls. A structural protein gene rose alongside, which the authors read as a sign that IGF-1 production may begin as a response to local damage. Rabbits. PMID 10087355.

The synthetic peptide in muscle cells (Yang and Goldspink, FEBS Letters, 2002). In a muscle cell culture model, the MGF E domain peptide increased myoblast proliferation and inhibited their final differentiation, the opposite of mature IGF-1 on the second point. Blocking the IGF-1 receptor with an antibody did not stop it, which the authors read as action through a different receptor. Cells. PMID 12095637.

The natural form in people (Hameed, Journal of Physiology, 2003). Eight young adults (25 to 36) and seven older adults (70 to 82) did one bout of heavy single-leg knee-extension exercise, and muscle samples were taken from both legs 2.5 hours later. MGF messenger RNA rose in the exercised leg of the young group but not the older group. The other form, IGF-1Ea, did not change in either. At rest, MGF levels were about 100 times lower than IGF-1Ea. No peptide was given. This measured the body’s own MGF. PMID 12562960.

Nerve cells after interrupted blood flow (Dluzniewska, FASEB Journal, 2005). In gerbils with a brief, induced interruption of blood flow to the brain, the synthetic MGF end peptide protected vulnerable neurons, which the authors described as very significant protection. In brain slices kept in culture, it was as potent as full-length IGF-1 and its effect lasted longer. Its action did not depend on the IGF-1 receptor. Gerbils and brain tissue in a dish. PMID 16144956.

Transplanted muscle cells (Mills, American Journal of Transplantation, 2007). A Quebec group tested the 24-amino-acid peptide in a cell-transplant approach being explored for Duchenne muscular dystrophy. In mice, the peptide increased how many transplanted human muscle precursor cells took hold. In culture, it increased the cells’ proliferation and delayed their differentiation without changing survival, through a route other than the IGF-1 receptor. Mice and human cells in culture. PMID 17845560.

Heart tissue in sheep (Carpenter, Heart, Lung and Circulation, 2008). A New Zealand group induced heart attacks in sheep with microspheres, then compared vehicle with mature IGF-1, the MGF E domain peptide and full-length MGF, and in a second experiment vehicle with the E domain peptide alone. Over eight days, heart function was better preserved in the E domain groups, less heart muscle was compromised, and a marker of cell death was absent in the E domain hearts. Sheep. PMID 17581790.

Human muscle cells by donor age (Kandalla, Mechanisms of Ageing and Development, 2011). A Paris group added the 24-amino-acid peptide to muscle stem cells from donors of different ages. It lengthened how long cells from newborn and young adult donors kept dividing in culture, but not cells from older adult donors. In all cultures, the authors reported larger fused cells and fewer reserve cells. Human cells in culture. PMID 21354439.

Heart cells and mice (Mavrommatis, Molecular and Cellular Biochemistry, 2013). A Chicago group used the stabilized peptide. In a rat heart cell line, it entered cells and moved to the nucleus without activating the IGF-1 receptor, and it blocked a cell-death pathway under chemical stress. In mice given the peptide at the time of an induced heart attack, heart contractility two weeks later was better preserved than in untreated mice, with fewer dying cells in the surviving heart tissue. Cells and mice. PMID 23712705.

The isolation problem (Matheny, Endocrinology, 2010). A review of the field noted that the synthetic peptide matches the last 24 amino acids of the rodent form, but that no matching peptide had been identified in or isolated from cultured cells, their growth medium, animal tissues or body fluids. The authors separated what synthetic MGF does in experiments from anything the IGF-1 gene is known to make. Review. PMID 20130113.

What muscle actually holds (Durzyńska, Endocrinology, 2013). A Pennsylvania group measured the forms of IGF-1 in mouse muscle and found that most of it was pro-IGF-1, with the E peptide still attached, rather than mature IGF-1. Mice. PMID 23407451.

The failed replication (Fornaro, American Journal of Physiology, Endocrinology and Metabolism, 2014). Researchers at two pharmaceutical companies tried to reproduce the reported cell effects. The MGF peptide did not increase proliferation of a mouse muscle cell line or of primary human muscle cells at the concentrations tested, while mature IGF-1 and full-length IGF-1Eb did. It did not block differentiation, had no significant effect on primary mouse muscle stem cells, and neither the native nor a stabilized version produced the signaling response reported earlier in heart cells. The authors concluded the results call into question whether MGF has a physiological role. Cells. PMID 24253050.

PEG-MGF itself (Dominikowski, Frontiers in Endocrinology, 2026). An endocrinology review from Poznan searched the literature through January 2026 and listed no peer-reviewed human studies of PEG-MGF. The authors wrote that adding PEG changes how a peptide moves through the body and reaches tissue, so results with the natural form or the unmodified E domain peptide cannot be assumed to apply to the PEG version, and that the evidence base is mostly preclinical. Review. PMID 42395176.

How it works

The IGF-1 gene is read in more than one way. Its forms share the same mature IGF-1 region and differ in the E domain at the end, and MGF is the form whose E domain is changed by an extra stretch of genetic code (Yang et al., 1996, Carpenter et al., 2008). The human form was also described in liver in 1995 (Chew et al., 1995). The synthetic peptide copies only that changed end.

Several of the groups above reported that the peptide acted without the IGF-1 receptor (Yang and Goldspink, 2002, Dluzniewska et al., 2005, Mavrommatis et al., 2013). No other receptor for it is named in those studies. Two open questions sit underneath: a 2010 review found that no matching peptide had been isolated from the body (Matheny et al., 2010), and in mouse muscle most IGF-1 keeps its E peptide attached (Durzyńska et al., 2013).

What the research does not establish

• We found no published study that gave the synthetic MGF peptide or PEG-MGF to people. The human work measured the body’s own MGF in muscle samples.
• One attempt to reproduce the cell results failed. In 2014, researchers at two pharmaceutical companies found no effect on proliferation or differentiation.
• A 2010 review found that no matching peptide had been isolated from cells, tissues or body fluids, so the synthetic peptide may not match anything the body makes.
• Synthetic versions differ. Published sequences differ at the end, some versions are amidated or carry D-form amino acids, and PEG-MGF adds a PEG chain of unstated size. A result with one version may not apply to another.
• The heart and brain findings come from sheep, mice and gerbils. They describe what happened in a research model, not what would happen in a person.
• Much of the early work comes from one London group and its collaborators.
• Nothing here is a safety conclusion, and this is not a claim that the compound is safe to use.

Common questions

What is MGF?

MGF, or mechano growth factor, is a form of IGF-1 made by alternative splicing of the IGF-1 gene, which a 1996 study found in rabbit muscle subjected to stretch (Yang et al., 1996). The research peptide called MGF is a synthetic 24-amino-acid piece of the end of that form. It is not mature IGF-1.

What is the difference between MGF and PEG-MGF?

PEG-MGF is an MGF peptide with a polyethylene glycol chain attached. A 2026 review listed no peer-reviewed human studies of PEG-MGF and cautioned that results with the unmodified peptide cannot be assumed to apply to the PEG version (Dominikowski et al., 2026).

Has MGF been studied in people?

We did not find a published study that gave the synthetic MGF peptide or PEG-MGF to people. Human studies have measured the body’s own MGF in muscle samples, for example after one bout of exercise in young and older adults (Hameed et al., 2003).

Do the MGF studies agree?

The MGF studies do not agree. Earlier groups reported that the peptide changed muscle cell behavior in culture (Yang and Goldspink, 2002), and a 2014 study by researchers at two pharmaceutical companies could not reproduce those results (Fornaro et al., 2014).

Is MGF in the Etched Research shop?

No. MGF and PEG-MGF are not part of the Etched Research retail catalog. This page is in the research library because MGF is part of growth hormone and IGF-1 research, and the library covers more compounds than the shop does.

More in this research area: Growth Hormone Signaling Research | Research Library

Last reviewed 2026-10-05.

New reports, new research.

We send a note when a new lab report is posted or a compound comes back in stock, plus a short list of new research pages in any month we publish some. Never more than one email a week, and nothing else.

Please enable JavaScript in your browser to complete this form.

Address Email

Email Address *

Subscribe

For research use only. Not for human consumption.