FGL (FG Loop Peptide)

A 15-residue synthetic peptide copied from the second fibronectin type III module of NCAM that binds and activates FGFR1, studied in rodents for memory, synaptic plasticity, and neuroprotection; under its clinical designation FGLL it completed a single phase 1 study in healthy volunteers.

Also referenced as: FGLL, FG loop, FG loop peptide, FGL peptide, NCAM mimetic peptide FGL, EVYVVAENQQGKSKA

Also appears in: Neuroprotection

What is FGL (FG Loop Peptide)?

FGL is a synthetic pentadecapeptide, Glu-Val-Tyr-Val-Val-Ala-Glu-Asn-Gln-Gln-Gly-Lys-Ser-Lys-Ala (EVYVVAENQQGKSKA), copied from residues 681–695 of the neural cell adhesion molecule (NCAM); the rat and human sequences are identical. The sequence forms the F–G loop of NCAM’s second fibronectin type III (F3) module, the region that the Copenhagen Protein Laboratory identified by NMR and surface plasmon resonance as the site where NCAM docks directly onto fibroblast growth factor receptor 1 (FGFR1). The peptide was designed so that a short peptide could reproduce NCAM’s heterophilic signaling through FGFR1 without requiring NCAM–NCAM adhesion. Because receptor activation needs two FGFR1 molecules brought together, published pharmacology uses multimeric constructs rather than the free monomer: the early memory studies used a dendrimer of four monomers coupled to a lysine backbone (Cambon et al., 2004), and later in vivo work used a dimer of two monomers joined at their N-termini through an iminodiacetic acid linker, with amidated C-termini (Knafo et al., 2012).

It appears in the research-peptide market on the strength of a dense rodent literature from the early 2000s reporting enhanced memory consolidation, synaptogenesis, long-term potentiation, and neuroprotection, and because the clinical form, FGLL, reached a phase 1 healthy-volunteer study. FGL should be read as a laboratory probe of the NCAM–FGFR1 axis rather than as a developed drug.

How it works

• Direct FGFR1 binding at the NCAM F3 module 2 site. The first and second F3 modules of NCAM bind Ig modules 2 and 3 of FGFR1 directly; the NMR structure of F3 module 2 located the FGFR-binding surface, which overlaps with an ATP-binding site, and the module alone induced FGFR phosphorylation and neurite outgrowth (Kiselyov et al., Structure, 2003;11(6):691–701). FGL is the peptide excised from that surface.
• FGFR agonism in neurons via MAPK and PI3K–Akt. FGL binds FGFR and induces its phosphorylation without prior homophilic NCAM binding, promoting neurite outgrowth and survival in dopaminergic, hippocampal, and cerebellar granule neurons in a manner blocked by FGFR, MEK, or PI3K inhibition (Neiiendam et al., Journal of Neurochemistry, 2004;91(4):920–935). Downstream signaling is not identical to the cognate growth factor: FGL activates FRS2alpha more weakly than FGF1 does (Chen et al., Journal of Neuroscience Research, 2010;88(9):1882–1889).
• Presynaptic enhancement and synapse formation. In hippocampal cultures FGL enhanced presynaptic function through FGFR1 activation and increased synapse number; intracerebroventricular injection immediately after training produced long-lasting improvement in fear-conditioning and water-maze memory in rats (Cambon et al., Journal of Neuroscience, 2004;24(17):4197–4204).
• Facilitated AMPA receptor synaptic delivery. Subcutaneous FGL triggered hippocampal FGFR1 phosphorylation in vivo and boosted delivery of AMPA receptors into CA1 synapses, enhancing NMDA-dependent long-term potentiation through early PKC activation followed by persistent CaMKII activity; blocking this delivery abolished the cognitive effect (Knafo et al., PLoS Biology, 2012;10(2):e1001262). FGL also facilitated induction and maintenance of dentate gyrus LTP in anesthetized and freely moving rats (Dallérac et al., Learning & Memory, 2011;18(5):306–313).
• Glial and anti-inflammatory actions. In aged rats FGL attenuated microglial activation, interleukin-1beta expression, and the age-related LTP deficit, an effect attributed to IL-4-dependent up-regulation of neuronal CD200 (Downer et al., Neurobiology of Aging, 2010;31(1):118–128).

Research status

Preclinical. The rodent record is broad for a research peptide and comes largely from the originating Copenhagen laboratory and its collaborators. Beyond the memory and plasticity studies above, FGL given intranasally to newborn rats accelerated early postnatal sensorimotor development, subcutaneous FGL prolonged social-memory retention in adults, and the peptide was detectable in blood and cerebrospinal fluid for up to five hours after either route (Secher et al., Neuroscience, 2006;141(3):1289–1299). In a rat model of amyloid pathology induced by intracerebroventricular Abeta(25–35), FGL reduced amyloid immunoreactivity, tau phosphorylation, gliosis, neuronal loss, and the short-term memory deficit whether given intracisternally, intranasally, or subcutaneously, with inhibitory phosphorylation of GSK3beta proposed as the mechanism (Klementiev et al., Neuroscience, 2007;145(1):209–224; an erratum was issued in 2014). FGL reversed depression-like behavior in NCAM-deficient mice (Aonurm-Helm et al., European Journal of Neuroscience, 2008;28(8):1618–1628) and restored FGFR phosphorylation in that model (Aonurm-Helm et al., Brain Research, 2010;1309:1–8). Three-dimensional ultrastructural work in aged rats reported altered synapse and dendritic spine morphology in the dentate gyrus after FGL (Popov et al., European Journal of Neuroscience, 2008;27(2):301–314). The most recent program-linked work is a Cologne stroke study, with ENKAM co-authors, in which subcutaneous FGL mobilized endogenous neural stem cells and promoted remyelination after middle cerebral artery occlusion in rats (Klein et al., Journal of Neuroimmune Pharmacology, 2016;11(4):708–720).

Two studies from outside the originating laboratory qualify the picture. Systemic FGL partially protected CA1 pyramidal cells against Abeta(25–35), but FGL given to young adult rats that had not received amyloid itself produced an approximately 40% reduction in dorsal CA1 pyramidal cell counts, leading the authors to call the peptide “potentially disruptive in non-pathological conditions” (Corbett et al., PLoS One, 2013;8(8):e71479; an Open University study, though with an ENKAM-affiliated co-author). In the mouse kindling model, FGL at 2 and 10 mg/kg reduced the number of stimulations needed to reach a generalized seizure, and the authors wrote that the result raised concern about promotion of a hyperexcitable network (Zellinger et al., ACS Chemical Neuroscience, 2014;5(3):185–193).

Clinical. The only published human study is of FGLL, the clinical designation used by ENKAM Pharmaceuticals, a University of Copenhagen spin-off, for the full-length 15-residue FGL peptide (the subscript L, for “long,” distinguishes it from the company’s shortened “FGLS” analogs). In an eight-day, open-label, ascending-dose phase 1 study, 24 healthy male volunteers (mean age 42) received single intranasal doses of 25, 100, or 200 mg. All doses were tolerated, with no notable ECG, vital-sign, or laboratory abnormalities; three subjects reported five adverse events, consisting of a brief nasal burning sensation in two subjects at 200 mg, transient watery eyes in one subject at 25 mg, and dizziness, vomiting, and headache in one subject beginning more than two days after dosing and judged unrelated. Plasma levels were quantifiable only at the two higher doses, with mean peak concentrations of 0.52 and 1.38 ng/mL and exposure that rose roughly with dose. The authors stated that studies in patients with Alzheimer’s disease were planned (Anand et al., Clinical Pharmacokinetics, 2007;46(4):351–358).

Those patient studies were never reported. ENKAM announced in mid-2004 that it had filed to begin a phase 1/2 trial of FGLL in the United Kingdom, and an EU Seventh Framework consortium, NEUROFGL (grant 278006, coordinated by the University of Copenhagen with ENKAM as the largest beneficiary), ran from January 2012 to December 2014 with the stated aim of accelerating clinical development of FGL and building translational tests for patient selection in Alzheimer’s disease and mild cognitive impairment. The project’s final report on CORDIS describes a single-ascending-dose intravenous study of the shortened analog FGLS, said to be well tolerated, with no participant numbers, doses, or publication; the planned multiple-ascending-dose proof-of-concept study was not completed within the project, and a planned ten-patient PET study in stroke was not started. No clinical results from the program have appeared in the peer-reviewed literature, and the last ENKAM-affiliated FGL publication is the 2016 rat stroke study. A 2019 US academic study tested FGLL and FGLS in rats and non-human primates, reporting brain penetration and antidepressant-like effects, and suggested FGLS as a candidate for development in depression (Turner et al., Neuropsychopharmacology, 2019;44(2):356–363); no human study followed. As of October 2026, ClinicalTrials.gov and the EU Clinical Trials Register contain no study of FGL, FGLL, FGLS, or any ENKAM-sponsored intervention, and no regulatory agency has approved any of them for any use. Published human exposure is limited to 24 single-dose volunteers in 2007, and there is no active clinical development.

Common dosage forms

• Lyophilized powder in vials, typically at low-milligram scale, sold as a research reagent for laboratory use only. Vendor listings rarely state whether the material is the FGL monomer or one of the multimeric constructs (the lysine-backbone dendrimer or the N-terminally linked dimer) used in nearly all published pharmacology, and that distinction matters because monomer alone is not expected to dimerize FGFR1.
• Intranasal solution is the format used in the only published human study (FGLL) and in several rodent studies; it has no counterpart in the published human record beyond single doses.
• Reconstituted solution for subcutaneous injection is the format of most adult rodent studies; there is no published human data for injected FGL or FGLL by any route.

Key considerations

• Regulatory status. FGL is not approved by the FDA or any other regulator, and it is not an approved API eligible for pharmacy compounding. Its clinical form FGLL completed one phase 1 single-dose study in 2007 and has no registered trial anywhere; the EU-funded 2012–2014 program closed with its proof-of-concept study unfinished and no published clinical output. The Research Market status reflects the absence of US approval or any development program aimed at approval.
• The human evidence is a 24-person tolerability study. No efficacy endpoint has ever been tested in humans, no repeat-dose safety data have been published, and the one published trial used single intranasal doses. Memory or nootropic claims rest entirely on rodent experiments, most of them from the laboratory that designed the peptide and holds the patents.
• Two safety signals in animals. FGL alone reduced CA1 pyramidal cell counts by about 40% in healthy young adult rats (Corbett et al., 2013) and accelerated kindling in mice (Zellinger et al., 2014). Chronic FGFR1 agonism in the hippocampus is a plausible route to such effects, and neither signal has been followed up.
• FGL, FGLL, FGLS, and FGF ligands are not interchangeable. FGL is the 15-residue NCAM fragment; FGLL is ENKAM’s designation for the full-length peptide as a clinical candidate; FGLS denotes the company’s shortened analogs of at most 13 residues, the form dosed intravenously in the unpublished NEUROFGL study and tested in the 2019 primate work; and none of them are fibroblast growth factors. The 2007 trial report does not specify whether FGLL was dosed as monomer or dimer. FGL is an NCAM fragment that engages FGFR1 and produces a signaling profile distinct from FGF1 (Chen et al., 2010); it should not be described as an FGF analog. Literature on “FGL2” (fibrinogen-like protein 2, an unrelated immune protein) is frequently mixed in with this peptide in search results.
• Delivery and construct uncertainty. Published activity depends on multimeric presentation; monomeric material of the same sequence is a different pharmacological entity. Blood–brain barrier penetration and CSF levels were documented in rats after intranasal and subcutaneous dosing (Secher et al., 2006) and for FGLS in rats and non-human primates (Turner et al., 2019), but human CNS exposure has never been measured.