Why your GLP-1 class peptide CoA might be hiding impurities

Hims & Hers, a health products company, recently announced plans to shift its business toward peptide products. This came just before a U.S. Food and Drug Administration (FDA) advisory meeting on peptide compounding (the practice of mixing custom versions of drugs outside a standard factory). The company's stock dropped sharply the same week. The business story is brief, but the chemistry behind it is not. HIMS had built much of its recent growth on compounded semaglutide, a molecule called a GLP-1 receptor agonist (a compound that binds to and activates the GLP-1 receptor, a protein studied in metabolism research). In early 2025, the FDA removed semaglutide from its drug shortage list, cutting off the legal basis for large-scale compounding. Now the company is scrambling to replace that lost income by moving into other peptide lines. This pushes commercial players into a space where research labs already operate, and that creates a sourcing quality problem worth understanding in detail.

What semaglutide actually is, structurally

Semaglutide is built from 31 amino acids (the small building blocks that make up proteins and peptides). It is a modified version of a natural signaling molecule called GLP-1(7-37). Two key changes set it apart from the natural version. First, one building block near the start of the chain has been swapped for a different one (a substitution called Aib, short for alpha-aminoisobutyric acid, at position 8). This swap blocks a protein-cutting enzyme called DPP-4 from snipping the molecule apart too quickly. Think of it like fitting a lock onto one end of a chain so a pair of scissors cannot cut there. Second, a long fatty acid chain (a C18 fatty diacid, similar to a fat molecule) is attached near the other end of the peptide. This fatty chain is the key to semaglutide's extended stability. It sticks loosely to a blood carrier protein called serum albumin, which slows the molecule's removal and protects it from enzymes. The natural GLP-1 molecule breaks down in about two minutes. These two structural changes extend semaglutide's half-life (the time it takes for half the amount to break down) to about seven days.

Related compounds in active research include tirzepatide, which activates two receptors (GIP and GLP-1) at once, and retatrutide, which activates three (GIP, GLP-1, and glucagon) at the same time. Both are structurally more complex than semaglutide. That extra complexity makes them harder to synthesize correctly and increases the chance of impurities in research-grade batches.

Where purity problems enter the picture

Peptide synthesis is done using a method called solid-phase peptide synthesis (SPPS), where the chain is built one amino acid at a time on a tiny solid bead. Protective chemical "caps" cover each building block until it is needed. For semaglutide, the step where the fatty acid chain is attached (called acylation) is where most impurities come from. If the protective caps are not fully removed (incomplete deprotection), short, broken chain fragments called truncated sequences and deletion peptides are left behind. These fragments can hide alongside the real molecule during standard purification. The fatty acid attachment step adds more possible impurities: versions that lost their fatty chain (deacylated species), versions where the chain attached to the wrong spot, and versions with the wrong chain length, all of which show up in lower-grade batches.

A Certificate of Analysis (CoA) is the document a supplier provides to report a product's purity. A CoA that says "purity: 98%" but does not state how that was measured tells you almost nothing. Some suppliers test using ultraviolet (UV) light set to a wavelength of 254 nanometers (nm). That setting is good at detecting molecules with ring-shaped chemical structures (called aromatic residues), but it misses most peptide chain fragments. The core peptide backbone absorbs UV light at a lower wavelength, between 210 and 220 nm. That is the range where broken chain fragments and related peptides clearly show up. Think of it like searching for a stain on glass with the wrong color flashlight. A sample carrying 13% related impurities can look perfectly clean on a 254 nm test.

When commercial demand spikes for a peptide, smaller manufacturers feel pressure to ship product fast. That is when analytical shortcuts appear. A thorough CoA for a GLP-1 class peptide should include:

• RP-HPLC purity with UV detection at 210-220 nm, not 254 nm alone
• Mass spectrometry confirmation of exact molecular weight, not just nominal MW from sequence
• Lot-specific data, not batch-average figures that pool multiple synthesis runs
• Identification of named related substances, including deacylated species for acylated analogues specifically

Cold storage and reconstitution for GLP-1 analogues at the bench

The fatty acid chain that extends semaglutide's stability also makes the molecule less likely to clump together (aggregate) when it is in its dry, freeze-dried form (lyophilized). That is a real advantage compared to many shorter, unmodified peptides. But the advantage is not unconditional. Freeze-dried GLP-1 class peptides should be stored at -20°C or colder, away from light and moisture. Each time a sample is thawed and refrozen (a freeze-thaw cycle), it speeds up a chemical breakdown called hydrolysis at vulnerable spots in the chain and promotes clumping, especially at higher concentrations. Plan your aliquots (small, single-use portions) before the very first thaw, so you never need to freeze and thaw the same vial repeatedly.

For vials that will be used more than once at the bench, bacteriostatic water is the right diluent (the liquid used to dissolve the dry peptide). Bacteriostatic water contains 0.9% benzyl alcohol, a preservative that keeps the solution free of microbial growth across multiple needle entries into the vial. Plain sterile water cannot maintain sterility after the septum (the rubber stopper on top) is first punctured. At the concentrations normally used in research reconstitution, the benzyl alcohol does not cause the peptide to clump.

When adding diluent to the vial, aim the stream slowly down the inner glass wall rather than directly onto the dry peptide cake. Directing the stream straight at the cake creates a pressure disturbance that breaks up the fragile structure and promotes clumping, especially in acylated peptides. After adding the liquid, swirl the vial gently and let it rest for 5 to 10 minutes at room temperature before drawing from it. Store the reconstituted solution at 2 to 8°C in the refrigerator rather than in the freezer. Freezing a reconstituted peptide solution drives clumping instead of preserving it.

Why the commercial noise matters for your sourcing decisions

The Hims & Hers situation shows a pattern that repeats across peptide categories. When commercial attention floods into a peptide category, the number of suppliers multiplies, and average quality drops as poorly documented material enters the market. Researchers who check how a CoA's purity number was measured (rather than just accepting the number), and who confirm mass spectrometry data against the expected molecular weight, consistently work with cleaner starting material. For acylated analogues like semaglutide, a deacylated impurity (a version of the molecule missing its fatty chain) at even a few percent changes the research sample in ways that carry through every downstream experiment run with it. The cost of a degraded or impure vial is not just the vial. It is everything built on top of it.

Prompted by this coverage at Google News →

Related from our lab: the pen · cartridges · bacteriostatic water · reconstitution calculators

Shared by PreppinPeppers for research, educational, and demonstration awareness only. We link to third-party coverage; we do not endorse it, and nothing here is medical advice or a recommendation to use any substance in humans or animals. Our products are sold for laboratory research use only.

Frequently asked questions

What impurities are most common in SPPS-synthesized GLP-1 peptides like semaglutide?

Truncated sequences, deletion peptides, deacylated species, wrong-site acylation products, and incorrect chain-length variants are typical, with most arising during the acylation step that attaches semaglutide's C18 fatty diacid chain.

Why does UV detection at 254 nm miss peptide chain impurities on a CoA?

254 nm UV preferentially excites aromatic ring-containing residues; the peptide backbone absorbs at 210-220 nm. Fragments lacking aromatic groups are effectively invisible at 254 nm, causing reported purity figures to be overstated.

What analytical details should a research-grade GLP-1 peptide CoA specify?

The CoA should state the HPLC detection wavelength (210-220 nm to capture peptide bond absorption) and the analytical method; a bare purity percentage with no method detail provides no meaningful verification of actual peptide identity or purity.

What the research community gets wrong about GLP-1 peptide CoAs

A Certificate of Analysis looks like a settled fact, but a purity number by itself does not tell you much. Here are the assumptions that trip up researchers when they read a CoA for a GLP-1 class peptide.

• Treating one purity number as the whole story. HPLC tells you how much of the sample is your target peptide, but it does not confirm what the other peaks actually are. Bachem describes analytical HPLC as the standard purity method, with the main peak area compared to the total peak area. That percentage is only as honest as the method behind it.
• Ignoring the detection wavelength. The peptide backbone absorbs UV light at 210 to 220 nm. A CoA that does not state the wavelength, or that reports a reading at 254 nm alone, can miss chain fragments that do not carry ring-shaped groups. Two labs can report very different numbers on the same vial just from this one setting.
• Assuming HPLC confirms identity. It does not. HPLC separates and measures; it cannot tell you the peak is really your peptide. That job needs mass spectrometry, which assigns a mass to each peak. A peak short one amino acid points to a truncation, and a shift of a set number of mass units points to oxidation or a missing fatty chain.
• Forgetting that acylated analogues have their own failure mode. Semaglutide and similar peptides carry a fatty acid chain. During synthesis the chain can end up missing, attached at the wrong spot, or the wrong length. Work on the acylated GLP-1 peptide liraglutide showed truncations that elute right after the main peak, so a method that does not separate well can bury them under the target.
• Accepting a batch-average figure as lot-specific. A number that pools several synthesis runs is not the same as data measured on the exact vial in your hand. Deletion peptides, truncated sequences, and residual reagents vary run to run, so ask for the lot-specific sheet.

From our bench: when a CoA lists a purity figure but leaves off the HPLC detection wavelength and the method, we treat that number as unconfirmed and go back to the supplier for the lot-specific sheet before we plan any work around the vial.

Sources

1. Bacteriostatic Water for Injection, USP , FDA/DailyMed label (0.9% benzyl alcohol)
2. Duerkop et al., Biotechnol J 2018 , Impact of Cavitation, High Shear Stress and Air/Liquid Interfaces on Protein Aggregation
3. Sigma-Aldrich (Merck) , Handling and Storage Guidelines for Peptides and Proteins
4. Bachem , Quality Control of Amino Acids & Peptides: A Guide
5. Bachem , Peptide Purification Process & Methods: An Overview
6. Sterling Pharma Solutions , Characterisation of a Peptide Conjugate and Impurities (liraglutide, LC-MS)