GOAT Enzyme Explained: How Cells Attach Octanoyl to Ghrelin

Introduction

Ghrelin is a short peptide hormone, but the raw peptide is only half the story. For the receptor to recognize it, a single serine has to carry a small fatty acid tag. The enzyme that attaches that tag is ghrelin O-acyltransferase, usually shortened to GOAT and also known as MBOAT4. This article is for research use only and describes laboratory findings, not outcomes in people.

If you're reading ghrelin papers, GOAT shows up constantly: in assay methods, in structural models and in discussions of why "acylated" and "des-acyl" ghrelin behave so differently in the lab. Below we walk through how GOAT was found, what it modifies, what the best available structural model says about its internal channel, and which research tools exist for studying it.

How GOAT Was Found: From Odd Modification to MBOAT4

In short: researchers knew ghrelin carried an unusual lipid tag years before anyone could name the enzyme that added it.

Ghrelin was discovered in 1999 by Kojima and colleagues, who were searching for the natural ligand of the orphan receptor GHS-R1a, according to a 2021 review of ghrelin octanoylation. The peptide turned out to be modified on its third residue, a serine, with an octanoyl (eight-carbon) group. Synthetic ghrelin without that group failed to raise intracellular calcium in receptor-expressing cells, which is how the modification came to light.

Two groups, one answer in 2008

The enzyme stayed unnamed until 2008, when two independent groups identified it as MBOAT4, a member of the membrane-bound O-acyltransferase (MBOAT) family. The two teams used complementary strategies. One overexpressed candidate MBOAT genes and measured ghrelin acylation. The other used gene silencing, where loss of MBOAT4 reduced acylation and re-expression restored it. The same review notes the enzyme was later renamed GOAT to reflect its job.

An MBOAT is, broadly, an enzyme embedded in a membrane that moves a fatty acyl group from a coenzyme A carrier onto a target molecule. GOAT is the member of that family dedicated to ghrelin.

What GOAT Does: Octanoyl-CoA to Serine-3

In short: GOAT takes an eight-carbon fatty acid from octanoyl-CoA and forms an ester bond on the side chain of ghrelin's third residue.

The donor and the acceptor

The acyl donor is octanoyl-CoA. According to the review, GOAT strongly prefers the eight-carbon chain, though it can use other medium-chain fatty acids when octanoate is scarce. The acceptor is proghrelin, which is acylated on the side-chain hydroxyl of Ser-3. This happens in the endoplasmic reticulum (ER): the peptide is acylated in the ER lumen, while the octanoyl-CoA donor is drawn from the cytoplasm side.

Several details show how exact the enzyme is:

• When Ser-3 is changed to alanine, GOAT does not acylate a neighbouring serine instead.
• Threonine at position 3 is acylated, but at reduced levels by human and mouse GOAT.
• Ghrelin is described as the only substrate of GOAT in the human proteome.

Mechanistic work with substrate analogs adds a twist. In a substrate-analog study, ghrelin variants with amino-alanine in place of Ser-3 showed that GOAT can form an octanoyl-amide bond at a rate similar to the natural reaction. The same 2021 review, however, reports that a diaminopropionic acid substitution gave only a severely reduced amide product, so results depend on which analog is used.

Why the octanoyl group matters

The tag isn't decoration. Acylation is required for ghrelin to bind and activate GHS-R1a, and des-acyl ghrelin does not bind that receptor. Structurally, the receptor's binding site has a polar region that engages ghrelin's N-terminal residues and a nonpolar surface that accommodates the octanoyl group. We cover that receptor-side story in our article on why the octanoyl tag matters for GHS-R1a binding, and the wider field of synthetic ligands in the ghrelin-mimetic secretagogue class.

Inside the Enzyme: Membrane Topology and the Catalytic Channel

In short: GOAT is a multi-pass membrane protein whose best-available structure is a computational model with a bent tunnel running through it.

Eleven helices and a bent channel

One caveat first: there's no experimentally determined GOAT structure in the sources we reviewed. The widely cited human GOAT model was built without experimental structural data, using coevolutionary contact predictions, folding simulations and refinement, then relaxed with 500 nanoseconds of all-atom molecular dynamics in an ER-like lipid bilayer. The authors tested 42 alanine variants, roughly 10% of the protein's residues, to check the model against activity.

The model has 11 transmembrane helices and two intramembrane domains, which matches the topology reported earlier for mouse GOAT. The helices form an ellipsoidal cone with the narrow end facing the ER lumen. A bent, continuous channel runs through the core from the lumen to the cytoplasm. Mutations at residues that touch this internal void were significantly more likely to reduce or abolish activity, which supports the idea that the channel is functionally essential.

The model resembles DltB, a bacterial MBOAT whose crystal structure appeared after the model was built, with a TM-score of 0.6 despite only 12.3% sequence identity. The authors also note that predictions are less reliable in regions that are not homologous to DltB.

His-338 and Asn-307: two conserved residues, two roles

Two conserved residues dominate the GOAT literature. First, His-338 is described as absolutely conserved. It contacts the channel and is proposed to act as a general base, helping the serine hydroxyl attack the acyl donor. Mutating it causes a complete loss of acylation activity. Asn-307, by contrast, is topologically separated from His-338; the model places it in the octanoyl-CoA binding site rather than at the catalytic center. Topology studies place the two on opposite sides of the ER membrane, so they cannot share one simple active site, which is why the review proposes complementary roles.

The handoff hypothesis

Putting those pieces together gives a "handoff" picture, which the model's authors present as a hypothesis still to be tested:

1. Octanoyl-CoA approaches from the cytoplasmic face, with the CoA portion staying exposed and the octanoyl chain turning sharply into the protein interior.
2. Ghrelin reaches the channel from the ER side through the lumenal pore.
3. The octanoyl group is transferred to the serine hydroxyl inside the channel.
4. Octanoylated ghrelin leaves to the ER lumen and coenzyme A is released back to the cytoplasm.

The octanoyl pocket has gatekeepers. Trp-351 marks the end of the acyl-donor channel and Phe-331 sits at the distal end of the pocket. When either is mutated to alanine, the enzyme gains the ability to use longer donors: F331A gained activity with a 12-carbon donor, and W351A accepted both 12- and 14-carbon donors. That's good evidence that these two residues set the enzyme's preference for the eight-carbon chain.

Research Tools: Assays and Inhibitor Classes

In short: labs study GOAT with a handful of assay formats and a set of tool compounds, but results are not always directly comparable.

A review of GOAT inhibitor discovery describes a 45 kDa MBOAT-family enzyme and a toolkit that includes microsomal activity assays, membrane fractions from baculovirus-infected Sf9 cells, cell-based assays in HEK293 cells, ELISA-based detection and fluorescence-based formats that avoid radiolabels.

On the compound side, early work used the enzyme's own product, octanoylated ghrelin, as a modest inhibitor. Replacing the ester-forming serine with diaminopropionic acid, which gives a hydrolytically stable amide, produced more potent peptide analogs. Bisubstrate designs went further by linking a CoA-like fragment to an octanoylated ghrelin peptide; the lead example, GO-CoA-Tat, carries a cell-penetrating Tat sequence (see our overview of cell-penetrating peptides) and weighs about 3,700 Da. The review points out that its size and susceptibility to proteolysis limit it to use as a pharmacological tool. Small-molecule chemotypes such as naphthyl glycine benzamides, triazoles and covalent cyanoenones followed.

One practical point for anyone comparing papers: assay formats differ in enzyme source, substrate and readout, so IC50 values from different studies are not always directly comparable. That caution is our reading of how the review describes the formats, not a statement the authors make outright.

What Remains Unresolved

Several questions are still open. For one, no experimentally determined GOAT structure was available in the sources we reviewed, so the ghrelin-binding site and the exact catalytic mechanism remain to be validated. The 2021 review also mentions a report of GOAT activity at the plasma membrane of hippocampal neurons, which would sit uneasily with a purely ER-based picture; that finding is an outlier rather than settled biology. And while octanoate can come from diet or from beta-oxidation of long-chain fatty acids, the review found no support for a microbiome source.

Conclusion

GOAT is a narrow, specialized enzyme: one family (MBOAT), one organelle (the ER), one donor preference (octanoyl-CoA) and, as far as the human proteome goes, one known substrate. Its job is to put an eight-carbon tag on ghrelin's Ser-3, and that tag is what the receptor reads. The current structural picture is a validated-by-mutagenesis model rather than a crystal or cryo-EM structure, so treat its details as well-supported hypotheses. For anyone reading this literature, the practical takeaways are to check which assay format a paper used and to keep acylated and des-acyl ghrelin firmly separate.

Frequently Asked Questions

What does GOAT stand for?

Ghrelin O-acyltransferase. It was first identified as MBOAT4 and belongs to the membrane-bound O-acyltransferase family.

Where in the cell does GOAT work?

In the endoplasmic reticulum, where proghrelin is acylated in the lumen using octanoyl-CoA supplied from the cytoplasm side.

Does GOAT modify other proteins?

Ghrelin is described as the only known GOAT substrate in the human proteome, according to the 2021 review cited above.

Has the structure of GOAT been solved experimentally?

Not in the sources we reviewed. The widely cited human GOAT structure is a computational model validated by alanine mutagenesis.

For research use only. Not for human or animal consumption of any kind. The information in this article is for educational purposes only and is not intended to diagnose, treat, cure, or prevent any disease. The statements made have not been evaluated by the U.S. Food and Drug Administration. These products are NOT FDA APPROVED. Please consult with a licensed healthcare professional before making any decisions regarding your health or research.

Optides LLC is a chemical supplier. Optides LLC is not a compounding pharmacy or chemical compounding facility as defined under 503A of the Federal Food, Drug, and Cosmetic Act. Optides LLC is not an outsourcing facility as defined under 503B of the Federal Food, Drug, and Cosmetic Act.