Most side-by-side write-ups of these two compounds open the same way: HCG gives you LH activity, HMG gives you LH plus FSH activity, pick accordingly. That summary is half right, and the half it gets wrong is the more interesting half.
Published compositional analyses of highly purified human menopausal gonadotropin report that the great majority of its LH-receptor-mediated bioactivity comes not from luteinizing hormone but from human chorionic gonadotropin — the same hormone that constitutes the other product on this shelf. On the LH arm, these two are far closer to each other than the labels suggest. The genuine dividing line is FSH.
This is a comparison of HCG and HMG on molecular architecture, receptor pharmacology, and what the analytical literature actually reports about what is in the vial. Both are research compounds; nothing here is a protocol.
One shared subunit, two different beta chains
LH, FSH, TSH and hCG are all members of the glycoprotein hormone family, and they are all heterodimers: a common alpha subunit (92 amino acids, identical across all four in a given species) non-covalently paired with a hormone-specific beta subunit. The alpha chain alone does nothing. Specificity lives entirely in the beta partner, and the dimer must stay assembled to be active.
LH-beta and hCG-beta are the closest pair in the family. They share roughly 85% sequence identity across their first ~114 residues — near-twins by the standards of this family. Then they diverge in a way that matters enormously:
hCG-beta carries a C-terminal extension that LH-beta does not. This roughly 24-residue tail — the carboxy-terminal peptide, or CTP — brings the hCG beta chain to 145 amino acids against LH-beta's 121, and it carries four O-linked glycosylation sites clustered along it.
That tail is the single most consequential structural difference between the two hormones, and it is a pharmacokinetic one. Sialylated O-glycans on the CTP slow renal and hepatic clearance dramatically. Circulating LH has a half-life on the order of tens of minutes; hCG persists for roughly a day or more. Same receptor, same broad biological instruction, wildly different exposure curve — a clean illustration of the point made in our pharmacokinetics primer that mechanism and PK are independent axes.
The CTP is a well-validated enough half-life module that protein engineers have grafted it onto other glycoprotein hormones deliberately, which is about as strong a functional endorsement as a structural motif gets.
What HCG is
HCG is a defined heterodimer: common alpha plus hCG-beta with its CTP tail. It is produced by placental syncytiotrophoblast, and the classical manufacturing route is extraction from the urine of pregnant women — though recombinant choriogonadotropin alfa also exists, so the name alone does not disclose lineage. We covered that route-vs-name trap in peptide synthesis routes and impurity profiles.
Its target is the LHCGR (luteinizing hormone/choriogonadotropin receptor), a class A GPCR expressed on gonadal steroidogenic cells. Because hCG and LH converge on that one receptor, hCG functions as a long-acting LH-receptor agonist — which is why it sits at the gonadotropin tier of the HPG axis, downstream of GnRH-receptor compounds like gonadorelin and further downstream still of kisspeptin-10.
Worth being precise: hCG has essentially no FSH-receptor activity. It covers one of the two gonadotropin signals, not both.
What HMG is — a fraction, not a molecule
HMG (menotropin) is where the comparison stops being molecule-to-molecule. By definition it is extracted from the urine of postmenopausal women, in whom gonadotropin output is elevated by the loss of gonadal negative feedback. What comes out of that process is a mixture, standardized by biological activity rather than by defined chemical composition.
Two published analyses put numbers on that mixture.
van de Weijer et al., Reproductive BioMedicine Online 2003;7:547–557 analyzed a highly purified hMG preparation with a claimed 1:1 FSH:LH activity ratio and found three gonadotropins present — FSH, LH and hCG. hCG immunoactivity ran three-fold higher than LH immunoactivity, and because of hCG's much longer half-life, the authors concluded that about 95% of the in-vivo LH-receptor-mediated bioactivity is attributable to hCG. They also reported that non-gonadotropin protein impurities accounted for at least 30% of the material on an RP-HPLC peak-area basis, identifying leukocyte elastase inhibitor, protein C inhibitor and zinc-alpha-2-glycoprotein among them.
Capolupo et al., International Journal of Molecular Sciences 2024;25:9405 revisited the question with LC-MS/MS glycopeptide mapping across five HP-hMG samples. Their headline finding is a source assignment: the FSH and LH in the samples carried sulfated glycans, which are diagnostic of pituitary-derived glycoprotein hormones, while no sulfated glycopeptides were detected on any site of the hCG beta subunit. The hCG glycan distribution instead matched a urinary placental hCG control — pointing to a non-pituitary origin for the hCG present. Quantitatively, they detected LH-beta only in traces (0.9–1.2%) against beta-hCG at 18–47%, estimated protein impurities at 20–30%, found the alpha subunit strongly oxidized (~20% relative abundance), and identified more than 200 non-gonadotropin proteins.
Two caveats on reading that second paper. Analytical comparisons between urinary-derived and recombinant gonadotropin products are frequently authored by parties with commercial interests in one side of that comparison — check the declared competing interests before treating the impurity and oxidation figures as neutral. And a sourcing inference drawn from glycan patterns is a strong inference, not a manufacturing record.
But the direction of both papers agrees, twenty years apart and by different methods: in menotropin preparations, the LH-labeled activity is carried overwhelmingly by hCG.
So what actually separates them
Strip out the label language and the comparison resolves cleanly:
- The LH-receptor arm is not the difference. Both deliver LHCGR agonism, and in both cases the molecule doing most of that work is hCG.
- FSH is the difference. HMG supplies FSH-receptor activity; HCG supplies none. If a research question involves the FSH limb of the axis, that is the entire distinction.
- Compositional definition is the second difference. HCG is one heterodimer. HMG is a standardized urinary fraction containing multiple gonadotropins plus a substantial and characterized non-gonadotropin protein load.
That reframes the choice from "one hormone or two" to "do you need the FSH limb, and can your experiment tolerate a mixture?"
The receptor does distinguish LH from hCG
One nuance keeps the "it's all just LH activity" simplification from being fully safe. LHCGR does not treat its two ligands identically.
Work on biased agonism at this receptor — primarily Riccetti et al., Scientific Reports 2017, synthesized in Casarini et al., Endocrine Reviews 2018;39:549–592 — reports that hCG preferentially drives the cAMP/PKA arm associated with steroidogenic output, while LH is comparatively more potent at ERK1/2 and AKT phosphorylation. Same receptor, two ligands, qualitatively different downstream signatures.
Read that with the caveats from our receptor pharmacology primer attached. These are cell-model findings, and cell-model signaling profiles are sensitive to receptor reserve and overexpression; a bias measured in a transfected line is a hypothesis about the intact system, not a measurement of it. But it does mean "hCG substitutes for LH" is a statement about receptor occupancy, not a claim that the two are pharmacologically equivalent.
The sourcing angle
Both compounds sit outside the analytical framework used for synthetic peptides, for the same reason: they are glycoform mixtures, not defined molecules. Potency is assigned in International Units by bioassay against a reference standard, an HPLC purity percentage of a mixture is not a characterization number, and the absence of a "98% pure" line is not a red flag. We worked through that logic in peptide synthesis routes and impurity profiles and again in net peptide content — no need to re-argue it here.
What the two papers above add is specific: for menotropins, the meaningful analytical questions are which gonadotropins are present and in what proportion, what the non-gonadotropin protein load is, and what oxidation state the subunits are in — none of which a purity percentage answers, and all of which required LC-MS/MS glycopeptide mapping to surface. Biologically sourced material — whether extracted from human urine or expressed in a host organism — also carries microbial and endotoxin questions that no chromatogram addresses, which is why endotoxin testing sits on its own axis.
FAQ
Is HMG just HCG with FSH added? Functionally closer to that than the labels imply, but not literally. HMG is a urinary fraction containing FSH, LH, hCG and a large complement of other human proteins. What the analytical literature establishes is that its LH-receptor activity is carried mostly by hCG — not that it is a two-component formulation.
Why does hCG last so much longer than LH if they hit the same receptor? The hCG beta subunit's C-terminal peptide extension carries four O-linked glycosylation sites that LH-beta lacks entirely. Those glycans slow clearance. It is a pharmacokinetic difference written into the primary structure, with no bearing on which receptor is engaged.
Can a COA confirm what is in an HMG vial? Not with the standard synthetic-peptide panel. Identity by intact mass and purity by HPLC area both assume a single defined molecule. The published characterizations of these preparations used glycopeptide mapping by LC-MS/MS and immunoassay — see /quality/ for how the test panel should follow the manufacturing route.
For compound-by-compound references, see the research library.
This article is educational and for the laboratory research community. Trulogic Labs products are sold for laboratory and research use only and are not for human consumption.