Sermorelin is the amidated N-terminal fragment of human growth hormone-releasing hormone: 29 residues, written throughout the literature as GHRH(1–29)NH₂ or GRF(1–29)NH₂. The parent hormone is 44 residues, so sermorelin discards everything past Arg29 and caps the C-terminus as a carboxamide rather than a free acid. Its sequence is Tyr-Ala-Asp-Ala-Ile-Phe-Thr-Asn-Ser-Tyr-Arg-Lys-Val-Leu-Gly-Gln-Leu-Ser-Ala-Arg-Lys-Leu-Leu-Gln-Asp-Ile-Met-Ser-Arg; the single methionine at position 27 accounts for the one sulfur atom in the molecular formula, and the N-terminal Tyr-Ala pair is the feature that governs how the peptide behaves in the presence of aminopeptidases.

Identity and characterisation

The verified identity block carried on the Echo Peptides product page reads:

The molecular weight is an average mass computed from the formula, not a monoisotopic mass; anyone reconciling an LC-MS identity check against it should confirm which of the two their instrument reports. The formula describes the free base. Material supplied as an acetate salt will not match it on a gravimetric basis, and peptide content by weight is a separate question from purity by HPLC area.

A reading caution matters more here than for most compounds. Sermorelin appears in the primary literature under at least five names — sermorelin, sermorelin acetate, GRF(1–29)NH₂, hGRF(1–29)NH₂, GHRH(1–29)NH₂ — and a large fraction of the papers that look like sermorelin studies are actually studies of close analogs. [Nle²⁷]-hGHRH(1–29)-NH₂, [Ala¹⁵]-GRF(1–29)-NH₂ and the various lactam-bridged cyclic derivatives are different molecules with different measured properties. Where the sections below rest on an analog rather than the parent sequence, that is stated.

The distinction is not pedantic. The recurring modifications were introduced precisely because the parent sequence behaves differently at those positions: substitution at residue 2 was pursued explicitly to design analogs resistant to dipeptidyl peptidase IV cleavage [11], and alanine at residue 15 was among the single substitutions that raised in-vitro potency several-fold over the reference standard [5]. Norleucine at 27 replaces the sequence's only methionine. Reading an analog's data onto sermorelin therefore inverts the logic of the experiment that produced it — the analog exists because sermorelin does not do that.

Mechanism as studied

The receptor

The target is the growth hormone-releasing hormone receptor (GHRHR), a class B, seven-transmembrane G protein-coupled receptor. Mayo isolated a novel cDNA from rat pituitary mRNA, isolated the human homolog, and expressed it in human kidney 293 cells; membrane fractions from those cells bound human GHRH with high affinity and specificity, and GHRH stimulated intracellular cAMP production in the transfected cells [1]. The encoded protein is homologous to the receptors for secretin and vasoactive intestinal peptide, and the rat receptor mRNA was found expressed predominantly, if not exclusively, in the anterior pituitary [1]. Working independently, Lin and colleagues cloned mouse and rat cDNAs encoding a pituitary-restricted seven-transmembrane receptor that mediates increases in intracellular cAMP and cAMP-dependent gene transcription in response to GRF, and reported that its expression pattern corresponds to that of the growth hormone gene and depends on the POU-domain transcription factor Pit-1 [2].

The signaling model assembled from that period — ligand engaging a Gₛ-coupled receptor on the somatotroph, release of stored granules, cAMP accumulation, protein kinase A-dependent activation of CREB, and CREB-driven expression of Pit-1 — is set out in a review by Mayo and colleagues [3]. It should be read as a model: several of its steps, particularly the CREB and Pit-1 links, were inferred from experiments in other cell types rather than measured end-to-end in somatotrophs [3].

How the peptide is recognised

A 2.6 Å cryo-electron microscopy structure of the human GHRHR bound to its endogenous ligand and the stimulatory G protein shows the α-helical peptide making an extensive, continuous interaction network involving all of the extracellular loops, the extracellular domain, and every transmembrane helix except TM4, with the ligand N-terminus engaging a broad set of specific contacts [4]. That structure was solved with the endogenous hormone, not with sermorelin, so it constrains rather than demonstrates how the 29-residue fragment is accommodated — though the fragment contains the entire N-terminal region the structure identifies as making those contacts.

Structure-activity work on the fragment itself points the same way. A systematic alanine scan of [Nle²⁷]-hGHRH(1–29)-NH₂ found near-complete loss of in-vitro potency when position 1, 3, 5, 6, 10, 11, 13, 14 or 23 was replaced by alanine, while substitutions at 8, 9, 15, 22 and 28 gave analogs two- to six-fold more potent than the hGHRH(1–40)-OH standard in the same assay [5]. Radioligand binding to rat adenopituitary membranes, using [¹²⁵I-Tyr¹⁰]hGRF(1–44)NH₂ as tracer, gave a complementary picture: deletion of Tyr1, Ser9, Lys12, Val13, Gly15, Gln16 or Lys21 from hGRF(1–29)NH₂ dropped relative affinity below 3%, whereas removal of Gln24, Asp25, Ile26, Met27, Ser28 or Arg29 was tolerated far better, and replacing the C-terminal carboxamide with a free carboxylic acid reduced relative affinity to 57% [6]. The two datasets are not in conflict where they appear to differ — substituting alanine at position 15 is not the same experiment as deleting Gly15 — but they do measure different things, potency in a functional release assay versus affinity in a membrane binding assay, and should not be pooled.

Conformation

Circular dichroism and two-dimensional NMR of [Nle²⁷]-hGHRF(1–29)NH₂ showed an ordered structure at roughly 30% trifluoroethanol, with two distinct regions of α-helix spanning residues 6–13 and 16–29 [7]. A separate NMR and constrained molecular dynamics study of [Ala¹⁵]-GRF(1–29)-NH₂ and its cyclic derivatives found that solvent dominates the answer: in 75% methanol the linear peptide is a single helical segment with fraying at both ends, while in water at pH 3 it resolves into two helical segments, residues 7–14 and 21–28, with flexible regions between them and at the termini [8]. Both studies used substituted analogs, and both used co-solvents or low pH. The practical consequence is that a conformational measurement on reconstituted material is only interpretable against a matched solvent condition.

Model systems in which it has actually been examined

The published systems are narrower than the compound's visibility suggests. Receptor cloning and signaling work used transfected human kidney 293 cells and rodent pituitary tissue [1, 2]. Binding SAR used membranes prepared from rat adenopituitary [6]; the alanine and Aib scans used an in-vitro growth hormone release assay referenced to an hGHRH(1–40)-OH standard [5]. Conformational work used purified synthetic analogs in defined solvent mixtures [7, 8].

Two further systems are worth naming because they show how much the readout depends on the cell. An alternatively spliced human GHRHR transcript encodes a receptor truncated just before the sixth transmembrane domain; expressed in HeLa T4 cells, that variant cannot signal through cAMP, forms a complex with the full-length receptor, and acts as a dominant inhibitor — substantially reducing GHRH binding by the wild-type receptor without altering its cell-surface expression [9]. Any transfected-cell assay therefore reports the behaviour of whatever receptor complement that cell actually expresses, not of the receptor in isolation. And GHRHR is not confined to the pituitary: in rat kidney, receptor mRNA was localised to Henle's loop, a four-hour incubation of Henle's loop cell cultures with 1–100 nM rat GHRH-(1–29)NH₂ significantly reduced receptor mRNA levels, and a carboxyfluorescein-labelled, multiply-substituted hGHRH-(1–29)NH₂ analogue internalised in a time- and temperature-dependent manner that was blocked by phenylarsine oxide [10]. That work used the rat sequence and a labelled analog, not sermorelin, and it establishes localisation and regulation — not a function.

What the evidence does not establish

No paper cited here reports a binding affinity or functional EC₅₀ for unmodified sermorelin at a cloned human GHRHR under conditions that would let it be ranked against other GHRH-family peptides. The SAR series report relative potencies against an hGHRH(1–40)-OH standard [5] and relative affinities against hGRF(1–29)NH₂ itself [6], which are internally consistent but not portable between laboratories.

Selectivity is genuinely open. The GHRHR is homologous to the secretin and VIP receptors [1], and none of the work above quantifies what sermorelin does, if anything, at those related class B receptors. Treating GHRHR engagement as exclusive is an assumption, not a finding.

The cryo-EM structure is of the receptor with the full-length endogenous ligand [4]; there is no published high-resolution structure of the receptor with sermorelin bound. Likewise, the downstream chain beyond cAMP — PKA, CREB, Pit-1 — is a model with segments imported from other systems [3], not a measured cascade in a single somatotroph preparation.

Nothing in this literature speaks to outcomes in an intact organism, and this guide makes no such statement. What is described above is receptor binding, cAMP accumulation, mRNA regulation, internalisation and in-vitro release assays, in the specific systems named.

Handling and stability

The one stability property that is well characterised is proteolytic. Dipeptidyl peptidase IV cleaves the Ala2–Asp3 bond, removing the N-terminal Tyr-Ala dipeptide. With purified human placental DPP IV at pH 7.8 and 37 °C, initial cleavage rates were approximately 5 µmol min⁻¹ mg⁻¹ and were essentially the same for GRF(1–44)-NH₂, GRF(1–29)-NH₂ and GRF(1–20)-NH₂ — truncation to 29 residues does not slow the cleavage, though shorter fragments were cleaved more slowly, which the authors read as evidence for extended substrate binding beyond the scissile bond [11]. Analogs with a D-configuration at P2, P1 or P1', and desNH₂Tyr1 or N-MeTyr1 derivatives, were not cleaved at all [11] — which is why so many GHRH analogs carry exactly those modifications and why comparing them to sermorelin on stability is comparing designed resistance to its absence.

An in-vitro metabolism study of sermorelin alongside tesamorelin and CJC-1295 in fortified urine identified nineteen major metabolites across the analogs studied, synthesised and characterised them as reference materials, and used sermorelin(3–29)-NH₂ — the fragment corresponding to loss of the N-terminal dipeptide — as one of the target analytes, reaching LC-MS/MS limits of detection generally at or below 1 ng/mL [12]. Separately, synthetic fragments corresponding to sermorelin(1–11), (13–20) and (22–29) were incubated with human blood, trypsin and chymotrypsin; each fragment showed a distinct protein-binding and enzymatic-stability profile, and sermorelin(22–29) was stable enough under those treatments to serve as an in-house internal standard [13].

All of that describes behaviour in enzyme-containing matrices. It says nothing about the stability of lyophilized powder in a sealed vial, or of a reconstituted solution in a peptide-free buffer. No published study located here reports defined stability data for lyophilized sermorelin across storage temperatures and intervals. Storage guidance on the product page follows general practice for lyophilized peptides rather than a compound-specific stability study, and should be read that way. The conformational work also implies that a reconstituted sample's secondary structure is solvent-dependent [7, 8], so buffer composition is a variable to control and record, not a detail.

Open questions

Four gaps stand out. First, a direct affinity and efficacy characterisation of unmodified sermorelin at the cloned human GHRHR — binding constant, cAMP EC₅₀, and a β-arrestin readout in the same cell background — does not appear in the literature surveyed, despite the compound's age. Second, selectivity against the secretin and VIP receptor family is unquantified. Third, the functional significance of the dominant-negative GHRHR splice variant [9] and of extrapituitary receptor expression [10] remains unresolved, and both complicate the interpretation of any cell-based assay. Fourth, the compound has essentially no published formulation science: no stability-indicating assay, no forced-degradation profile, no data on aggregation or oxidation of Met27 under storage. For a molecule this well known, the gap between how often it is named and how thinly it is characterised is the most striking thing about its literature.