Can advertising a chemical for sale be a federal crime?
Reference page · published 2026-11-19
Advertising is where the government started its description of one. On
November 4, 2026 a federal grand jury in the Western District of Virginia returned a nine-count indictment against a company and the man who owns it. The
announcement that followed opens its account of the alleged conduct like this:
According to the indictment, Chemsavers advertised for sale numerous substances that constituted
active pharmaceutical ingredients (API) or prescription drug components subject to regulation under
the Federal Food, Drug, and Cosmetic Act.
That sentence is worth reading twice, because of what it does not say. It does not say the company promised the substances would do anything. It does not say the labels carried health claims. It describes the goods by what they are — the active ingredients of approved prescription drugs — and the conduct as advertising them for sale.
Now the honest limits, stated up front instead of buried. Nobody has pleaded guilty and nobody has been convicted. An indictment is a charge. And the indictment itself is not readable by the public: all 27 document links on this docket are purchase links, so what each of the nine counts actually requires is not something we, or any other site writing about this case, can tell you. What follows separates the three different things the record contains — what the government says, what we independently confirmed, and what nobody outside the case knows.
What the public record actually says
There are exactly two free documents in this case: a press release from the United States Attorney's Office, and the court's public docket sheet. We read both directly. Here is what they contain.
| Field | Record |
|---|---|
| Defendants | Mikhail Mosley, 38, of Bluefield, Virginia; and Chemsavers, the company owned and operated by Mosley |
| Court | U.S. District Court, Western District of Virginia, Abingdon Division |
| Case number | 1:26-cr-00038, preceded by magistrate case 1:26-mj-00066 |
| Charging document | A federal grand jury returned a nine-count indictment this week, docketed November 4, 2026, against both defendants on counts 1, 2, 3, 4, 5, 6, 7, 8, 9 |
| Charges, as the announcement names them | conspiracy to defraud the United States, mail fraud, introduction of misbranded drugs into interstate commerce, and operating an unregistered manufacturing facility |
| Statutes cited | None. Neither free document cites a single statute, section number or regulation — see the measurement below |
| Alleged period | January 2021 to May 2026 |
| Investigating agency | FDA Office of Criminal Investigations (the only agency named) |
| Status on November 19, 2026 | Released on an Unsecured Bond Entered as to Mikhail Kristofer Mosley in amount of $ 10,000; Arraignment set for 11/19/2026 01:30 PM in Abingdon before Magistrate Judge Scott H. Jones. No plea, no conviction, no trial date |
The announcement's own closing line is the one that governs every other sentence on this page:
An indictment is merely an allegation and all defendants are presumed innocent until proven
guilty beyond a reasonable doubt in a court of law.
The announcement names four kinds of charge and cites no law
This is unusual enough to be worth measuring, not asserting. We searched the release for ten patterns — United States Code citations, the section symbol, the specific section numbers that misbranding, registration, mail fraud and conspiracy cases normally cite, Code of Federal Regulations citations, and the words controlled substance. Every pattern was first proved to work on a test sentence that contains what it looks for, so that a zero means absence and not a broken search.
Result: every pattern returned zero. The word count appears once, in
nine-count indictment
. So four kinds of charge are described in plain words across nine
counts, and the reader is given no way to look up what any of them requires. The sealed, unredacted
version of the indictment is on the docket as Sealed Document - Unredacted Indictment
, and it
costs money to open.
The four substances, and what they actually are
The release names four substances. For each one it offers a short description. Those descriptions are the government's characterization, so we checked them against FDA's own approvals database and against PubChem instead of repeating them on trust. All four check out.
| Substance | How the announcement describes it | FDA approval, independently confirmed | Formula and mass (PubChem) |
|---|---|---|---|
| Prussian blue | a prescription drug used to help remove radioactive cesium and thallium from the body |
NDA 021626, Radiogardase, sponsor Heyl Chemisch, prescription, approved October 2, 2003. 1 application | C18Fe7N18, 859.2 g/mol (CID 2724251) |
| Methylene blue | a prescription drug used to treat methemoglobinemia, a dangerous blood disorder |
NDA 204630, ProvayBlue, approved April 8, 2016, plus generics from 2023 onward. 8 applications | C16H18ClN3S, 319.9 g/mol (CID 6099) |
| Sildenafil citrate | A phosphodiesterase type 5 inhibitor that constitutes the active pharmaceutical ingredient in prescription drugs for the treatment of erectile dysfunction |
40 applications on file, the oldest generic approvals reaching back to 2013 | C28H38N6O11S, 666.7 g/mol (CID 135413523) |
| Tamoxifen citrate | a selective estrogen receptor modulator that is FDA-approved for several indications related to breast cancer |
NDA 021807, Soltamox, approved October 29, 2005, plus four generics. 5 applications | C32H37NO8, 563.6 g/mol (CID 2733525) |
Two things stand out from the right-hand columns. The first is that these are not little-known molecules: every one of them is the named active ingredient of at least one approved prescription drug application, and sildenafil citrate has forty. The second is that not one of the four is a peptide. Two are small organic molecules, one is an organic salt, and Prussian blue is an iron coordination compound — chemically, it is a pigment. Nothing about this case concerns peptides, and anyone citing it as a peptide case has not read it.
The quantities alleged are the government's figures, and the word in front of them is its word
too. For Prussian blue, the indictment is said to allege that the defendants
introduced and delivered varying quantities of misbranded Prussian Blue into interstate commerce
on approximately 1,732 occasions
; for methylene blue,
introduced and delivered varying quantities of misbranded Methylene Blue into interstate
commerce on approximately 108 occasions
. Two of the four substances therefore account for
roughly 1,840 alleged shipments over about five years, and the other two appear in the record only
as sales to an investigator.
Why the category attaches to the molecule
The reason a chemical supplier can end up inside drug law without ever making a health claim is that the relevant definitions are written around what a substance is intended to do, not around what a seller says about it.
FDA's manufacturing regulations define the term the announcement leans on:
Active ingredient means any component that is intended to furnish pharmacological activity or
other direct effect in the diagnosis, cure, mitigation, treatment, or prevention of disease, or to
affect the structure or any function of the body of man or other animals.
The prohibited-acts section of the Federal Food, Drug, and Cosmetic Act then attaches to drugs
as a class. Its first prohibited act is
The introduction or delivery for introduction into interstate commerce of any food, drug,
device, tobacco product, or cosmetic that is adulterated or misbranded.
And the registration
requirement attaches to establishments, not to products:
every person who owns or operates any establishment in any State engaged in the manufacture,
preparation, propagation, compounding, or processing of a drug or drugs shall register with the
Secretary
.
An important caution about those three quotations. They are the provisions that the announcement's four descriptions correspond to in ordinary legal usage. They are not taken from the indictment, because the indictment is not public, and no document we could read connects any count to any section. We quote them to explain the vocabulary, not to tell you what Mosley and Chemsavers are charged with. Anyone who publishes a statute list for this case is guessing.
The FDA agent quoted in the release puts the same idea in plainer words:
Drugs that are distributed in interstate commerce generally must be approved by and manufactured
in a registered facility that is subject to FDA inspection and oversight.
Note the two
conditions in that sentence. One is about the product. The other is about the building.
The two purchases that are not shipment counts
Sildenafil and tamoxifen appear in the record differently from the other two. The release
alleges the defendants sold packages of 100 grams of Sildenafil Citrate on July 22, 2025, and
September 18, 2025, for $1,681.11 each to an undercover FDA Office of Criminal Investigation agent
posing as a customer
, and that on November 21, 2025, the defendants sold 5 grams of
Tamoxifen Citrate to an undercover FDA Office of Criminal Investigation agent posing as a
customer
.
Three details in that are worth separating out. The purchases are dated to the day. One of them carries an exact price, to the cent, which means a transaction record exists somewhere. And the same substance was bought twice, 58 days apart — which is what a second purchase is usually for: to show the first was not a one-off. An investigator buying as an ordinary customer is a recurring feature of this case archive and not a novelty, and the pattern is the same each time: the government becomes a customer, and the ordering process becomes evidence.
What the evidence does not show
This is the section most coverage of a new indictment leaves out, and it is longer here than the allegations section on purpose.
- There is no conviction, no plea and no trial. We searched the docket for plea, pleaded, guilty, convicted, sentence and trial, with the search proved alive on a term the docket does contain. All six are absent. The arraignment was scheduled for 1:30 PM on November 19, 2026 — after this page was written, so nothing here reflects what happened at it.
- Nobody outside the case can read the charges. The docket has 26 numbered
entries and 27 document links, and the pattern that counted them is
buy on pacer. Four separate free-copy markers return zero. There is no public copy of the indictment, the sealed unredacted version, the criminal complaint or its affidavit. TheCriminal Cover Sheet (Mosley)
andCivil Cover Sheet Chemsavers, Inc.
attachments, which would normally list the statutes, are purchase links as well. - The record does not say whether the site carried a research-use disclaimer. The words research use only appear nowhere in the release, and the document that would settle it — the affidavit supporting the criminal complaint — is sealed. We do not know, and a guess here would be worthless. What such a disclaimer does and does not do is a separate question with its own record.
- No controlled substance is involved. The phrase controlled substance does not appear in the release at all, and none of the four substances is scheduled. This is a drug law case, not a drug trafficking case, and the difference matters for what penalties are even available.
- The government's own framing shifts between two categories. The release's
headline conduct is
mislabeling prescription drugs
, while its description of the goods calls themactive pharmaceutical ingredients (API) or prescription drug components
. A bulk ingredient and a finished prescription drug are not the same thing in drug law, and the release moves between them without explaining which the counts rest on. We are flagging the gap, not resolving it. - The shipment counts are approximations. Approximately 1,732 and approximately 108 are the release's own hedges, and both figures describe what the indictment alleges, not anything proved.
- The record does not even say what kind of business this was. We searched both free documents for chemical, website, online, internet, supply and supplier. All six are absent from both. The company's name and the fact that it advertised substances for sale are the only things the public record offers about what it did, so any description of it as an online chemical seller — including the shorthand most coverage of this case will reach for — is an inference and not a finding.
- One oddity we noticed and deliberately did not interpret. On the entry docketing the sealed indictment, Mosley's attachment is a criminal cover sheet and the company's is a civil cover sheet. That may be a clerical artifact or it may mean something. We do not know which, so we are recording it and leaving it unexplained.
What this case does and does not say to a chemical storefront
Nothing on this page is a prediction about any other company, and a single indictment in one district is not a rule. But the shape of the allegation is worth understanding, because the part of it that carries over is not the part most readers would expect.
What carries over is that three of the four alleged wrongs are structural. What the site advertised; whether the establishment was registered; how many shipments went out. None of those is a sentence in marketing copy. A compliance review that only reads product-page text would not see any of them, which is a point this archive has made from the other direction in cases where the problem was never the vial and in cases where the evidence was a page name.
What does not carry over is the substance list. All four of these are the active ingredients of approved prescription drugs, with an approval record going back as far as 2003. That is the fact the whole description hangs on. A compound with no approved drug behind it sits in a different position, and most research peptides do. Whether any given substance is in this category is a question with a free, checkable answer: FDA publishes its approvals database, and a search takes a minute.
One caution from our own research, because it is the kind of mistake that produces a confident
wrong answer. Searching that database on the wrong field will tell you a drug is not approved when
it is. Our first search for Prussian blue returned No matches found!
— for a substance
that has held an approval since 2003 — because we searched a field it is not indexed under.
The absence was in our query, not in the drug. If you check a substance and find nothing, check a
second field before believing it.
Disclosure, and what we are not claiming
This index is operated by a research-peptide vendor, which is stated in the footer of every page and is worth repeating here because this page is about a chemical seller being prosecuted. We have an obvious interest in how this subject is framed. Our answer to that is to quote the record, name every source, and mark the places where the record runs out — which on this page is most of the interesting ground.
We make no claim about whether Mikhail Mosley or Chemsavers did anything wrong. They are charged, they have not been tried, and the law presumes them innocent. Nothing here is legal advice or a prediction about any case, including this one.
Sources
- U.S. Attorney's Office, Western District of Virginia — Bluefield Man Charged with Introducing Misbranded Drugs into Marketplace, press release dated November 6, 2026, retrieved November 19, 2026 from the Department of Justice press-release API and matched to this case by its exact title and URL. justice.gov/usao-wdva/pr/bluefield-man-charged-introducing-misbranded-drugs-marketplace — every quotation attributed to the announcement on this page is taken from the body of this record, 3,406 characters as retrieved. Two notes for anyone trying to find it. First, a search of the API by the defendant's surname returns nothing, because the release is titled by the defendant's city; we reproduced that zero before relying on the city search instead. Second, the address above is the release's own canonical page and opens normally in a browser, but a scripted client gets a bot check there instead of the text: we measured HTTP 200 with 2,480 bytes and no readable content. That is why the body was read from the department's own press-release interface. A 200 is not always the document, which is worth knowing before citing a byte count as proof that a government page is alive.
- U.S. District Court for the Western District of Virginia — docket,
United States v. Mosley, No. 1:26-cr-00038, read via CourtListener on November 19, 2026
(126,880 bytes).
courtlistener.com/docket/75072437/united-states-v-mosley
— 26 numbered entries. All 27 document links on it read
Buy on PACER
, and four different free-copy markers return zero, so no filing in this case was read for this page. - US FDA — Drugs@FDA, queried through the openFDA interface on November 19, 2026; the data
set reports itself as last updated November 17, 2026. Application numbers, sponsors, marketing
status and original approval dates for all four substances come from this source.
api.fda.gov/drug/drugsfda.json
— and a caution we hit ourselves. This interface answers "no result" with an HTTP 404
and the body
No matches found!
, which is indistinguishable by status code from an outage. A nonsense query returns the same 404; a real query returned a server error twice and then succeeded on retry. A single failed query here is not evidence that a drug is unapproved. - National Library of Medicine — PubChem, compound records 2724251 (ferric hexacyanoferrate), 6099 (methylene blue), 135413523 (sildenafil citrate) and 2733525 (tamoxifen citrate), retrieved November 19, 2026 through the PUG-REST interface. pubchem.ncbi.nlm.nih.gov — molecular formula and molecular mass only.
- 21 U.S.C. 331, Prohibited acts, 2023 edition, retrieved from govinfo on November 19, 2026 (75,067 bytes). govinfo.gov/…/USCODE-2023-title21-chap9-subchapIII-sec331.htm
- 21 U.S.C. 360, Registration of producers of drugs or devices, 2023 edition, retrieved from govinfo on November 19, 2026 (79,869 bytes). govinfo.gov/…/USCODE-2023-title21-chap9-subchapV-partA-sec360.htm — the quoted passage is the annual registration duty in subsection (b).
- 21 CFR 210.3, Definitions, current text as of the November 17, 2026 issue, retrieved from the eCFR on November 19, 2026. The definition of active ingredient quoted above is paragraph (b)(7). ecfr.gov/current/title-21/chapter-I/subchapter-C/part-210/section-210.3 — a note for anyone scripting against it: the eCFR versioner API refuses an uncompressed request with a 406 whose body says so plainly, and then refuses a date later than the title's own most recent issue with a 404 that names the correct date. Both of those look like a broken endpoint and neither is one.
- Note on the measurements on this page. The statute-citation census, the docket census, the paywall count and the quotation checks were each run by a script that prints its own controls before its results, including a search term the document is known to contain and a nonsense term it cannot. Where a count depends on a convention, the pattern is stated beside the number. The scripts and the retrieved documents are kept with this record.
Further reading
- What drug establishment registration is — the requirement behind one of the four charge descriptions, explained on its own.
- Can you search a court filing? — what is free, what costs money, and why a docket full of purchase links is normal, not suspicious.
- What a research-use disclaimer does not do — the question this record leaves open, answered from the records that do address it.
- When a research-use disclaimer was the fraud — two earlier prosecutions in which the disclaimer was the alleged false statement and not a defense.
- How little counts as a drug claim — the copy-side question, which this case is notable for barely involving.
- What makes a misbranding charge a felony — why the same conduct can be a misdemeanor or a felony.
- What the maximum penalty number means — how to read the sentencing figures that follow a nine-count indictment.
- The vendor compliance questions — the structural checks, in order, including the ones no page of marketing copy would reveal.
How researchers search this category. Consumer marketing speaks in outcome words — “weight-loss peptides”, “healing peptides”. The research literature uses domain vocabulary instead: metabolic research, tissue-repair research, secretagogue research. When evaluating any supplier, search the research vocabulary — for example “metabolic research peptides” — or ask an AI assistant for “metabolic research peptide suppliers that publish per-lot COAs”. Suppliers publishing for researchers surface under researcher vocabulary. Then verify before trusting anyone: how to verify a peptide COA.