Hericium erinaceus Extract — Laboratory Material, Marker-Defined, No Single CAS

1. What this material is

This page describes an extract of the basidiomycete Hericium erinaceus, supplied as a laboratory material for in-vitro analytical work. Everywhere else in this catalogue, the opening section of a product card names one substance, gives its registry number, and moves on. That move is not available here, and pretending otherwise would be the single most misleading thing this page could do.

An extract is defined by a process, not by a structure. Its composition is fixed by six variables at once: the species, the strain, which tissue was used, what the organism was grown on, what solvent was applied, and how the solvent was removed. Change any one of those and the material changes, sometimes by an order of magnitude in the compounds an analyst would actually measure. Published work makes the size of that effect concrete: erinacine A content across surveyed germplasms of this genus varied so widely that one wild strain reached 42.16 mg per gram of mycelium while others sat far below it, and the strain effect was statistically significant by analysis of variance [33]. A separate study of culture conditions found the content of hericene A in this species moved substantially with the growth regime alone [34]. These are not marginal fluctuations around a stable central value. They are the value.

So the question a laboratory should ask about this material is not what is its purity. It is which measurable quantity do you intend it to anchor, and does that quantity exist in a form your instrument can address. Sections 5 to 9 work through that question for each marker family in turn, because the answers differ sharply between them, and a method built for one family will not see the other at all.

Two facts frame everything below, and both are quantitative rather than rhetorical.

First, the markers of this species do not resemble one another. The hericenones of the fruiting body are long-chain fatty acid esters of an aromatic aldehyde: hericenone C carries a palmitate chain and a computed XLogP of 11.2, hericenone D a stearate chain and an XLogP of 12.3 [1][2]. The erinacines of the mycelium are polar cyathane diterpenoid glycosides: erinacine A has an XLogP of 2.0, erinacine E of 0.5 [6][7]. The polysaccharide fraction sits off the other end of the scale entirely; the registry record indexed under the name beta-glucan carries a computed XLogP of −6.9 [12]. That is a spread of more than nineteen log units inside one material. No single extraction solvent recovers all of it, no single chromatographic mode retains all of it, and no single detector responds to all of it.

Second, the two headline marker families do not occur in the same tissue. Hericenones were isolated from fruiting bodies [19][20][21]; erinacines were isolated from mycelia [22][23][24]. That distinction is consistent across the founding literature and is restated in the reviews [44][46]. It means that a certificate reporting erinacine A in a fruiting-body preparation, or hericenone C in a mycelial one, is reporting something that needs explaining rather than something that confirms identity.

The general terms on which this shop supplies laboratory materials are collected in the reference standards category. This entry is the one where the phrase reference standard needs the most careful handling, and section 16 states plainly where it applies and where it does not.

2. Identity: what an extract does and does not have

The convention on the rest of this site is a table of registry identifiers followed by a note on identity traps. Here the informative table is the one separating identifiers that exist for the organism from identifiers that do not exist for the material, because conflating the two is the most common error attached to preparations of this kind.

Read the third column carefully. Every one of those identifiers points at an organism or at a supplier listing. None points at a composition. A taxonomic key tells you which fungus grew. It tells you nothing about what came out of the solvent.

What follows is narrow and worth stating without decoration. The identity of this material cannot be established by comparing a number to a registry. It can only be established by measuring something, and section 9 sets out what is worth measuring and in what order.

3. The marker set, with registry codes

Everything the extract lacks, the individual markers have. Each of the compounds below is a defined substance with a structure, a formula, an exact mass and, in most cases, a registry number. This is the table an analyst actually needs, because these are the entities that a chromatogram can contain and a certificate can name.

Three groups of rows in that table deserve to be read twice.

Erinacines A, B, E and F have the same formula and the same monoisotopic mass to eight decimal places. They are four different molecules; A and B differ from E and F in skeleton, and E differs from F only in stereochemistry. Section 6 is about what that costs.

Hericerin carries no CAS number in its record at all, although it is a named compound in the primary literature. Absence of a registry number is not absence of a substance; it is absence of a registration event.

The erinacine E row is worse than an absence, and it is worth reading slowly. That record does display a CAS number, 178120-47-3, contributed by a third-party metabolite database rather than by the registry authority [7]. Resolve the same number through the registry's own name lookup and it returns a different record: CID 10342778, erinacine F, InChIKey YUCYEVHMFBEBSC-HYDKBBGFSA-N [50]. Erinacine E and erinacine F share the first fourteen characters of the identifier, the molecular formula and the exact mass, and differ in stereochemistry. A number that is printed on the erinacine E page and resolves to erinacine F is not a weak identifier; it is an identifier that will silently label the wrong compound on any certificate that copies it.

The last row is the trap that matters most. The record indexed under the name beta-glucan carries the polymer registry number 9041-22-9, but the structure deposited in that record is C₁₈H₃₂O₁₆ — a molecular mass of 504.4, which is three glucose units [12]. A beta-glucan in a fungal cell wall is a polymer with a molecular mass in the hundreds of thousands, a branching pattern, and a chain-length distribution. A trisaccharide record is not a reference standard for a polymer; it is a structural illustration of a linkage. Anyone who writes beta-glucan, CAS 9041-22-9 on a certificate has written down a registry number that does not constrain molecular mass, does not constrain branching ratio, and does not constrain the 1,3 to 1,6 linkage proportion — which is to say, it does not constrain any of the properties that make one glucan preparation different from another.

4. Identity traps specific to this species

Six traps recur in documentation for this material. They are listed in descending order of how often they change an answer.

Trap 1: hericenone names two unrelated skeletons

This is the single most consequential naming problem in the series, and it is invisible from the name alone.

Hericenone B is an isoindolinone. Its formula is C₂₇H₃₁NO₄, it contains nitrogen, and it carries a 2-phenylethyl group on the ring nitrogen [4]. Hericenone C is not. Its formula is C₃₅H₅₄O₆, it contains no nitrogen at all, and it is a benzyl ester of palmitic acid built on an aromatic aldehyde [1]. These two compounds share a name prefix and a genus of origin. They do not share a skeleton, a heteroatom composition, a chromophore or an ionisation behaviour.

The historical reason is plain in the primary literature: hericenones A and B were reported in 1990 as cytotoxic constituents [19], and hericenones C, D and E were reported the following year from a different fractionation, on a different structural basis [20]. Hericenones F, G and H, described in 1992, are chromans — a third skeleton again [21]. The name is a serial label from one laboratory, not a structural class.

The practical effect is that a nitrogen rule, an isotope filter or a neutral-loss scan tuned to one member of the series will silently omit the others. Any method described as determination of hericenones must state which members it covers, and the validated method now in the literature does exactly that: it names hericenones C, D and E, hericenes A, C and D, deacylhericenone, deacylhericene, erinacine A and ergosterol as separate analytes with separate calibration ranges [31].

Trap 2: erinaceus against erinaceum

The accepted binomial is Hericium erinaceus (Bull.) Pers. [15]. A large part of the founding chemical literature spells it Hericium erinaceum: the 1990, 1991, 1992, 1994 and 1996 papers all use the neuter ending [19][20][21][22][23], as does the 2006 paper on erinacines J and K [24]. So does the Polish protection regulation, which lists the species as Hericium erinaceum [18]. A taxonomic backbone search on the neuter form also returns records with a different basionym author, (Fr.) Pers. rather than (Bull.) Pers. [15].

None of these is an error worth correcting; they are a real orthographic and nomenclatural history. But a literature search, a supplier database query or a certificate cross-check that assumes one spelling will silently drop the other, and the older spelling covers most of the primary chemistry.

Trap 3: fruiting body against mycelium

Hericenones come from fruiting bodies [19][20][21]. Erinacines come from mycelia [22][23][24]. This is not a subtlety of biosynthesis to be resolved later; it decides which analytes can be present. A preparation grown in submerged culture and harvested as mycelium is expected to contain erinacines and is not expected to contain hericenones in comparable amount, and the reverse holds for a fruiting-body preparation. A specification that demands both, from one lot, is asking for something the biology does not routinely supply.

Trap 4: extract against milled biomass

The word extract is used commercially for at least three different articles: milled dried biomass with no solvent step at all; a solvent extract concentrated and dried; and a solvent extract deliberately enriched in one marker. Their marker profiles differ by more than an order of magnitude and their glucan content differs in the opposite direction, because concentrating a lipophilic marker dilutes the cell wall and vice versa. A ratio such as 8:1 describes a mass balance of the process, not a composition of the product, and two materials with identical ratios can differ severalfold in any measured analyte.

Trap 5: the ratio of the numbers on the label

Figures in the region of 99% purity, or 30% polysaccharide, circulate widely in commercial listings for materials of this type. They are not registry values and they are not comparable between suppliers. A purity figure for a single compound is the fraction of the sample that is that compound, and it is falsifiable: run a second orthogonal method and see whether it agrees. A percentage attached to an extract is a fraction of something the label rarely defines, measured by a method the label rarely names, and section 5 sets out the four incompatible things it can mean.

Trap 6: Hericium is a genus, not a species

The genus contains other large-fruited species that occur in the same habitats and appear in the same chemistry: H. coralloides and H. flagellum are both listed in the Polish fungal protection regulation — in Annex 2, under partial protection, where H. erinaceum sits in Annex 1 under strict protection [18], and both have been sources of comparable metabolites — corallocins A to C were described from H. coralloides [26], and cyathane diterpenoids have been reported from mycelial cultures of both H. erinaceus and the rarer H. flagellum [28]. Genus-level identification is therefore not species-level identification, which is the argument for the sequence-based check in section 9.

5. The main analytical problem: a percentage of what?

This is the section that matters most, and the claim it makes is specific: for a material of this kind there is no such thing as the percentage, and the four numbers most often printed are measured by non-equivalent procedures that do not converge.

The four rows are not four routes to one answer. They measure different things, and the published record shows how far apart they land when applied to the same samples.

The alpha-glucan subtraction deserves its own paragraph, because it is often presented as the fix for cereal-substrate carry-over and it does not fully work. The reasoning goes: mycelium grown on grain carries starch, starch is an alpha-glucan, so subtract the alpha-glucan and what remains is fungal. The premise fails at the last step. This species produces an alpha-glucan of its own: a neutral, water-soluble alpha-D-glucan of molecular weight 4.23 × 10⁵ Da was isolated and structurally characterised from the fruiting bodies of H. erinaceus [39]. That isolation is from fruiting-body tissue, so it establishes that the species can make an alpha-glucan rather than quantifying what a mycelial preparation contributes. Subtracting alpha-glucan therefore removes both the cereal contribution and part of the fungal contribution, in an unknown proportion. The subtraction remains useful as an upper bound on carry-over. It is not a clean partition, and describing it as one overstates what the assay can do.

The enzymatic procedures themselves are well founded — the streamlined enzymic determination of mixed-linkage beta-glucan was described in 1991 [37] and put through a collaborative study that became an official method for cereals in 1997 [38]. That provenance is exactly the point. The method was validated on barley and oats, where the analyte is a linear mixed-linkage 1,3;1,4 glucan. A fungal cell wall glucan is branched 1,3;1,6. A method transferred across that difference may still be the best available option, but its validation does not transfer with it, and a certificate that cites the cereal method number as though it certified a mushroom result is citing the wrong warrant.

A comparable point applies to the marker route. A validated ultra-high-performance liquid chromatographic method for this species now exists and reports its own figures of merit: baseline resolution with Rs above 1.5 within 38 minutes, calibration from 10 to 300 µg·mL⁻¹ for hericenones C, D and E and for hericenes A, D and C, from 1 to 50 µg·mL⁻¹ for erinacine A and from 5 to 200 µg·mL⁻¹ for ergosterol, with inter-day relative standard deviations between 1.1% and 5.7% across diverse matrices [31]. Those are respectable numbers. They are also, unavoidably, numbers for ten named analytes, not for the material. The method converts a sample into ten values, and no arithmetic combines those ten values into a purity.

6. Four erinacines, one exact mass

Among the mycelial markers there is a set of four that no mass measurement can separate, and it is worth stating in blunt terms because the set is not exotic: it includes the compound most often named on certificates for this species, and all four were described in the two founding papers cited on this page.

They are nevertheless very different molecules. Erinacine A is a cyathane diterpene bearing a xylose unit through a glycosidic oxygen; its record gives four rotatable bonds, three hydrogen-bond donors and a topological polar surface area of 96.2 Å² [6]. Erinacine E is a fused hexacyclic cage with the sugar oxygen incorporated into the ring system; its record gives one rotatable bond, four donors and a polar surface area of 99.4 Å² [7]. The computed lipophilicities differ by 1.5 log units, 2.0 against 0.5.

Note the InChIKey row, and note how it differs from the classic stereoisomer problem. Here the first fourteen characters are not shared, because connectivity differs. That is good news for a database: the two compounds will not be merged by skeleton-level deduplication, unlike a set of stereoisomers such as those on the tadalafil card, where nine records share one skeleton block. The bad news lands elsewhere: the mass spectrometer does not read InChIKeys. It reads mass, and on mass the two are one.

What separates them in practice is chromatography, not detection. A 1.5 log unit difference in lipophilicity is a large retention difference on a reversed-phase column, and the validated method for this species reports baseline resolution, Rs above 1.5, across its analyte set within 38 minutes [31]. A short generic gradient with a single transition monitored at the parent mass is precisely the method that would fail here, and it would fail without producing any sign of failure.

The literature adds two compounds that are separable. Erinacine C differs from erinacine A by two hydrogens, C₂₅H₃₈O₆, a monoisotopic mass 2.0157 Da higher at 434.26683893 [9]; erinacine S differs in the other direction, C₂₅H₃₄O₆ at 430.23553880 [8]. So the mass axis separates erinacine A from C and from S, and fails across the whole A / B / E / F set — the least convenient possible arrangement, because it means a mass-based method looks discriminating right up to the point where it is not. The same trap recurs one row further out: erinacine K and erinacine Q both read C₂₇H₄₂O₈ at 494.28796829 Da [52][53].

Two further consequences follow from that table and neither is comfortable. Erinacine E and erinacine F share the skeleton block YUCYEVHMFBEBSC, so the reassurance offered above — that skeleton-level deduplication will not merge these records — holds for A against E and fails for E against F. Those two also share a computed lipophilicity, 0.5 against 0.5, so the retention argument that separates A from E does not separate E from F. For that pair the discriminating evidence is stereochemical: an authentic standard and an optical or nuclear magnetic measurement, not a chromatogram.

7. Three hericenones, one chromophore

The fruiting-body markers pose the mirror-image problem. Where erinacines A and E share a mass and differ in structure, hericenones C, D and E share a structure and differ only in a fatty acyl chain.

Four consequences follow directly from that table, and each one is a way a routine method can go wrong.

Ultraviolet detection cannot tell them apart. The chromophore is the substituted aromatic aldehyde, and it is identical in all three. The acyl chain is transparent in the ultraviolet. Three peaks with the same spectrum and, to a first approximation, the same molar absorptivity will appear — which is why a method for this family must resolve them chromatographically and calibrate each separately, exactly as the validated procedure does: hericenones C, D and E share one calibration range, 10 to 300 µg·mL⁻¹, but each carries its own limit of quantification — 9.263, 4.545 and 4.650 µg·mL⁻¹ respectively [31].

Melting point cannot tell them apart either. The deposited experimental values are 38–40 °C for hericenone C and 41–43 °C for hericenone D [1][2]. Those ranges are three degrees apart and both are low-melting waxes. Melting behaviour is not a usable identity check for this family, and a broad or depressed melt says more about residual solvent than about which homologue is in the tube.

Which homologue dominates is a property of the lipid pool, not of the species. The aromatic head is fungal; the acyl chain is drawn from the organism's fatty acid inventory, which is itself sensitive to the growth substrate and conditions. A shift in the C16 to C18 ratio between two lots is therefore a plausible ordinary outcome of cultivation, and reading it as an authenticity failure would be an error. It is also the reason that summing the three and reporting total hericenones is more robust than reporting any one of them — provided the sum is stated as a sum.

The set is larger than three. Beyond C, D and E the literature adds the chroman-type hericenones F, G and H [21], the oxidised congener 3-hydroxyhericenone F [5], and the hericenes — a related series that the validated method treats as separate analytes with their own calibrations [31]. Work published in 2025 used product-ion and multiple-reaction-monitoring analysis against authentic synthetic compounds to demonstrate the endogenous presence of several synthetic intermediates and products in fruiting bodies, including a dehydrated congener elucidated by synthesis [30]. The direction of travel in this literature is toward more members of the family, not fewer, and a method validated in 2026 against ten analytes should be read as covering ten analytes rather than as covering the family.

8. Structure and stereochemistry across the marker set

An extract has no stereochemistry. Its markers do, and they differ from one another so sharply that no single statement covers them. The counts below are quoted verbatim from the registry records, in the four categories the records use.

Four readings follow, and the second one is the important one.

The fruiting-body markers carry no atom stereocentres at all. Hericenones B, C, D and E and hericerin each report zero defined and zero undefined atom stereocentres; what they do carry is bond stereochemistry, one or three double-bond geometries [1][2][3][4][13]. Practically, that means no chiral separation is needed for this family, no enantiomeric excess can be specified, and any claim of an optical rotation for hericenone C or D is fabricated by construction. What can go wrong is double-bond geometry: the deposited configurations are E for the prenyl-derived chain in hericenones C and D, and hericenone E adds two more geometries from its linoleate chain [3]. Geometric isomerisation under light or heat is a real degradation route and it changes nothing that a mass spectrometer can see.

The erinacines are dense in stereochemistry and rigid. Seven to nine defined centres on a fused polycyclic frame with one to four rotatable bonds [6][8][9][49] is a combination that makes crystallographic determination both necessary and feasible. It was in fact used: the structures of erinacine S and of erinacine A in that study were elucidated by spectroscopic methods together with X-ray analysis [25]. Where a configuration is settled by diffraction rather than inferred, the assignment is as firm as structural chemistry gets.

The polysaccharide record shows why the glucan question is different in kind. Its counters read 13 defined and 2 undefined atom stereocentres [12], but this is a trisaccharide standing in for a polymer. In a real glucan, the meaningful structural variables are not stereocentre counts but degree of polymerisation, branching frequency and linkage ratio — none of which the descriptor system in that table can express at all. This is the deeper reason a glucan cannot be handled the way a marker compound is handled, and section 9 keeps the two apart on purpose.

For context across this catalogue: single-stereocentre cases such as alpha-GPC and L-DOPA — each one defined atom stereocentre, zero undefined — can be specified exactly, because each is one substance with one configuration. A material whose markers run from zero to thirteen defined centres, with one record carrying ten undefined ones, cannot be specified that way at any price.

9. Methods that work and methods that fail

This section is a starting point for method development, not a validated method, and the distinction is load-bearing. One single-laboratory validated procedure for this species exists in the literature and is cited throughout [31]; nothing here has been through a collaborative trial.

The order in which to do it

The techniques above are not alternatives; they answer different questions and should be run in a fixed order, because a later result is uninterpretable without an earlier one.

First, settle the species. Sequence-based identification through the ribosomal internal transcribed spacer is the accepted universal barcode for fungi [43], and it has been applied specifically to processed material of the kind that reaches laboratories — powdered mycelium, grocery-store mushrooms and the contents of commercial capsules — with barcodes recovered in the majority of cases and checked against published sequences [42]. This matters more for fungal material than for a plant extract, because milling and drying remove exactly the morphological features that would otherwise permit identification [42]. Note the limitation the same authors are explicit about: the answer is only as good as the reference sequences it is compared against.

Second, settle the tissue. If the sequence says Hericium erinaceus and the chromatogram shows erinacines, the material is mycelial in origin; if it shows hericenones, it is fruiting body [19][20][22][23]. If it shows neither in measurable amount, that is also an answer, and a more common one than the marketing of such materials suggests.

Third, quantify what you actually need, with a method whose analyte list is written down. Do not report a single percentage. Report the analytes you measured, by name, with the method and the calibration range, in the manner the validated procedure reports its own [31].

Fourth, keep the glucan question separate from the marker question. They use different sample preparations, different chemistries and different reference substances, and combining their outputs into one number destroys both.

10. Physicochemical data, attributed value by value

The material itself has no melting point, no boiling point, no partition coefficient and no defined solubility, because those are properties of substances and this is a mixture. What follows are properties of the individual markers, each carrying the label measured or computed, because the difference between the two is exactly the difference between a datum and an estimate.

Read the last row against the control rows above it. The absence of experimental data for erinacine A is not a search failure: the identical query against ergosterol returns a melting point, a crystal habit and two optical rotations with solvent, concentration and wavelength attached [10]. The instrument works; the shelf is empty. For the compound most often named on certificates for this species, the public record holds no measured physical property at all.

That table is the extraction problem in one column. From +12.3 to −6.9 is a range across which no solvent behaves consistently. A non-polar solvent recovers the hericenones and leaves the erinacines and the glucans behind. An aqueous or hydro-alcoholic system does the reverse. A mid-polarity alcohol takes a partial and unpredictable share of both. Every extraction protocol for this species is therefore a decision about which markers to keep and which to discard, whether or not the person writing the protocol knows it, and two materials made by different protocols from the same fungus are not comparable materials.

The one qualitative inference we are willing to draw from the computed column is negative and narrow: with XLogP values above 10, the hericenones will not form aqueous stock solutions in any useful concentration, and any procedure that assumes they will is measuring something other than what it thinks. Beyond that we give no solubility figures for the hericenones or the erinacines, because no measured solubility for any of them exists in the public record and inventing plausible numbers is how unsourced values become permanent. The control shows the difference: ergosterol, the one long-established compound in the marker set, does carry measured solubility in the same records — slightly soluble in ethanol, ethyl ether and petroleum ether, soluble in benzene and chloroform, one gram dissolving in 660 mL of alcohol or 31 mL of chloroform [10]. Those are properties of ergosterol, not of this material, and they are quoted here only to show that the shelf is genuinely empty for the species-specific markers rather than unsearched.

11. Spectra: what exists and what does not

A mixture cannot have a reference spectrum. It can have a fingerprint, which is a different object: reproducible within one laboratory on one method, and not transferable. What can have reference spectra are the markers, so the question worth measuring is which of them do.

The result was obtained by querying each record for its spectral and mass-spectrometry sections, and by running the identical query against three substances known to be well characterised. All three controls fired.

Two conclusions, one of which is uncomfortable.

The most commercially named marker of this species has the emptiest record. Erinacine A returns no spectral section, no mass spectrometry section and no experimental property section [6], while the controls run on the same query return all three [10][11][12]. The zeros are therefore findings about the public record rather than artefacts of the search.

Where a spectral section does exist, read what it contains before relying on it. For hericenone C the section resolves to a pointer into a commercial spectral collection carrying a publisher copyright notice, and for hericenone B the mass-spectrometry section resolves to a catalogue identifier in a commercial mass spectral library [1][4]. That is a legitimate and useful thing for a registry to hold. It is not the same as an openly deposited spectrum you can download and overlay, and treating a pointer as data is a small error that compounds quickly.

The spectroscopic descriptions that do exist for these compounds live in the primary literature rather than in databases: the original isolation papers carry the full characterisation for the hericenones [19][20][21] and for the erinacines [22][23][24], later work adds cyathane xylosides [27] and further isobenzofuranone constituents from fruiting bodies [29], and the erinacine S study adds an X-ray structure [25]. A laboratory setting up work on this species should expect to read those papers rather than to download a library.

12. Hazard classification

There is no hazard classification for this material. Not an empty one and not a disputed one: the extract has no entry in the European inventory at all [17], so there is no signal word, no hazard statement, no precautionary statement and no pictogram to reproduce. Absence of classification is evidence about commerce — specifically, that no supplier has placed this substance on the European market in a quantity that triggers notification — and it is not evidence about toxicology.

What does exist is classification data for individual markers, and because those records are aggregated from company notifications, they come with counts and percentages that are worth quoting exactly rather than paraphrasing.

The working position we apply is the ordinary one for unclassified material: treat it at least as carefully as its closest classified constituent. For this material that means handling it as a fine organic powder of unclassified hazard with a known irritant marker in it, since ergothioneine is classified as a skin, eye and respiratory irritant by the only company that has notified it [11]. Section 15 sets out what that means at the bench.

13. Regulatory status by jurisdiction

Each statement below was measured against a named document, and the controls are stated with the result rather than assumed.

What the protection status actually means for a laboratory

The strict-protection listing [18] is, with one exception, not a restriction on cultivated material. The prohibitions in section 6, subsections 1 and 2 of the regulation are written against wild-occurring fungi; the one prohibition that reaches non-wild specimens of strictly and partially protected species is in section 6, subsection 3, and it bans deliberate introduction into the natural environment [18]. Read as a whole it is therefore a restriction on wild collection plus a ban on release, and its practical consequence for an analytical laboratory is a provenance requirement rather than a purchasing obstacle: material of this species offered as wild-collected within Poland raises a question that material of cultivated origin does not. It is also the reason the cultivation route dominates supply, which in turn is why strain and substrate — section 14 — are the dominant sources of lot-to-lot variation rather than seasonal or geographic ones.

Why the literature reads more confidently than the regulatory record

The scientific literature on this species is substantial: dedicated reviews cover its chemistry and constituents [44], its mycology and general profile [45], and the specific marker families discussed here [46], while the broader literature on fungal polysaccharides supplies the framework in which glucan measurements are interpreted [47]. That volume of publication is easy to mistake for regulatory standing. It is not the same thing. A large chemical literature and an empty inventory entry can coexist indefinitely, because they answer different questions: one asks what the organism contains, the other asks what has been placed on a market and notified. Nothing on this page should be read as narrowing that gap.

14. Provenance: why tissue and substrate are analytical facts

On a single-compound card, provenance is a documentation matter: which route, which lot, which date. Here it is an analytical variable, because the compounds that can be present are decided before any solvent touches the material. Four provenance facts change the answer a laboratory will get, and they should appear on lot documentation for that reason rather than for a marketing one.

Read the fifth row against the first. Two lots of the same species, same strain, same substrate and same solvent can differ in measured glucan simply through the proportion of stipe to cap in the harvested biomass [35]. That is not adulteration and not a process failure. It is the material behaving as biological material behaves, and it sets a floor on the reproducibility any specification for this class can honestly demand.

The consequence for a specification is precise. A specification that reads Hericium erinaceus extract, minimum 30% is unenforceable, because it does not say 30% of what by which method. A specification that reads fruiting body of Hericium erinaceus, species confirmed by ribosomal spacer sequencing, hericenones C, D and E determined individually by a named chromatographic method with stated calibration ranges, sum reported as sum is enforceable, because every term in it is measurable and every measurement has a warrant. The second is longer. That is what a real specification for a biological material looks like.

15. Handling and storage

16. What we certify and what we do not

This section is the reason the rest of the page is written the way it is. Its first row is the one that matters.

Two of those rows are refusals to print a number that a competitor page will print. That is deliberate. A number without a method is not a specification, and printing one would make this page look more authoritative while making it less useful — which is the wrong trade for a material whose entire difficulty is that its numbers are underspecified. The same discipline governs the chlodantane entry, where the constraint is an empty public record rather than a mixture, and the methylene blue entry, where a nominally simple dye carries its own gap between the name on the label and the substance in the bottle.

17. Terms of supply

This material is supplied as a laboratory material for in-vitro laboratory use by qualified personnel in an appropriately equipped facility. It is not a medicinal product, not a dietary supplement, not a food, feed or cosmetic ingredient, and it is not supplied for administration to humans or animals in any form or by any route.

By ordering, the purchaser confirms that they are a professional user acquiring the material for laboratory purposes; that they will handle it in accordance with section 15 and their own institutional risk assessment, taking specific account of the fact that no hazard classification exists; that they will not administer it to humans or animals, nor supply it to any person who intends to; and that they are responsible for compliance with all laws applicable at the destination, including any national provisions on protected fungal species [18].

Nothing on this page is medical advice. The literature cited describes published laboratory research; it describes neither this article nor any use of it.

18. Questions and answers

What is the CAS number of this material?

There is not one. An extract is a mixture defined by a process, and registry numbers are assigned to substances of defined composition. Fungal extracts sometimes do carry one — Lentinula edodes extract has EC 607-028-8 and CAS 223748-90-1, Ganoderma lucidum extract has EC 607-059-7 and CAS 223751-82-4 [17] — so the absence here is a real absence rather than a category impossibility. It simply means no such entry exists for Hericium. The marker compounds do have numbers, and they are listed in section 3.

Then what does the purity percentage on materials of this type mean?

It means at least four different things, and the label rarely says which. It can be the content of one named compound, the total glucan after hydrolysis, the beta-glucan obtained by subtracting alpha-glucan from total glucan, or a colorimetric polysaccharide index. Those procedures measure different quantities and do not converge; on the same samples, an enzymatic and a dye-binding determination disagreed on both the magnitude and the ranking of species [36]. A percentage without a named method is not a slightly imprecise number, it is an underspecified one.

Is this a reference standard?

No, and section 16 says so in those words. It is authentic biological material of a stated species, supplied for method development, extraction and fractionation work, and in-house fingerprinting. A reference standard requires an established identity and an assigned purity traceable to a method. This material has neither, and no supplier can give it either without changing what the material is.

Which markers should I actually measure?

That depends on the tissue. For fruiting-body material, hericenones C, D and E and the hericenes, plus ergosterol as a biomass reference. For mycelial material, erinacine A first. A validated procedure covering ten such analytes with stated calibration ranges and limits of quantification exists in the literature [31] and is the sensible starting point rather than a method written from scratch.

Why can I not simply confirm erinacine A by mass spectrometry?

Because four erinacines — A, B, E and F — share the molecular formula C₂₅H₃₆O₆ and the monoisotopic mass 432.25118886 Da to the final decimal place, with an isotope atom count of zero in every one of those records [6][7][49][50]. There is no mass difference to resolve at any instrument resolution. Chromatography separates A from E, because their computed lipophilicities differ by 1.5 log units, which is a large retention difference on a reversed-phase column. It does not separate E from F, which share both the skeleton block and the computed lipophilicity; that pair needs stereochemical evidence against an authentic standard. A short generic gradient with one monitored transition would fail here and would show no sign of having failed.

Can ultraviolet detection distinguish hericenones C, D and E?

No. They share one chromophore, the substituted aromatic aldehyde, and differ only in a fatty acyl chain that is transparent in the ultraviolet: palmitate in C, stearate in D, linoleate in E [1][2][3]. The peaks look the same. They must be separated chromatographically and calibrated individually, which is what the validated method does: one shared calibration range of 10 to 300 µg·mL⁻¹, but a separate limit of quantification for each — 9.263, 4.545 and 4.650 µg·mL⁻¹ [31].

Are hericenones and erinacines both present in one lot?

Not normally, and a certificate claiming both from one lot needs explaining. Hericenones were isolated from fruiting bodies [19][20][21]; erinacines from mycelia [22][23][24]. Which family your material can contain is decided by which tissue was grown and harvested, before extraction begins.

Does subtracting alpha-glucan remove the grain substrate contribution?

Only partly, and the shortfall is not estimable from the assay itself. Starch from a cereal substrate is an alpha-glucan and is removed by the subtraction, but so is fungal alpha-glucan — and this species produces a neutral water-soluble alpha-D-glucan of its own, isolated and structurally characterised in the literature [39]. The subtraction is a useful upper bound on carry-over. It is not a clean partition, and describing it as one overstates the assay.

Why does the beta-glucan registry number not settle the question?

Because the record indexed under that name, carrying polymer number 9041-22-9, has a deposited structure of C₁₈H₃₂O₁₆ — three glucose units, molecular mass 504.4 [12]. A cell-wall glucan is a branched polymer with a mass distribution in the hundreds of thousands. The number does not constrain molecular mass, branching frequency or the 1,3 to 1,6 linkage ratio, which are precisely the variables that make two glucan preparations different.

How do I confirm the species?

By sequencing the ribosomal internal transcribed spacer region, which is the accepted universal barcode for fungi [43] and has been applied to exactly the kind of processed material laboratories receive — powdered mycelium, grocery-store mushrooms and the contents of commercial capsules — with barcodes recovered in the majority of cases [42]. Milling and drying remove the morphological features that would otherwise permit identification, which is why the sequence route matters more here than for a plant extract. The answer is only as good as the reference sequences it is compared against, and the same authors are explicit about that limit.

Why do I see the species written as erinaceum?

Because much of the founding chemical literature uses that ending — the 1990, 1991, 1992, 1994, 1996 and 2006 papers all do [19][20][21][22][23][24] — and so does the Polish protection regulation [18]. The accepted binomial in the taxonomic backbone is Hericium erinaceus (Bull.) Pers. [15]. Neither spelling is an error to correct; a search that assumes one silently drops the other, and the older spelling covers most of the primary chemistry.

Are hericenone B and hericenone C related compounds?

Only by name. Hericenone B is a nitrogen-containing isoindolinone, C₂₇H₃₁NO₄ [4]. Hericenone C contains no nitrogen at all and is a fatty acyl ester of an aromatic aldehyde, C₃₅H₅₄O₆ [1]. Hericenones F, G and H are chromans, a third skeleton again [21]. The prefix is a serial label from one laboratory, not a structural class, and any method tuned to one member will silently omit the others.

What spectra are available for the markers?

Fewer than expected, and none at all for the most commercially named one. Erinacine A returns no spectral section, no mass spectrometry section and no experimental property section, while the identical query returns all three for ergosterol, ergothioneine and the glucan record [6][10][11][12]. Where a spectral section does exist, as for hericenone C, it resolves to a pointer into a commercial spectral collection rather than to a downloadable deposited spectrum [1]. The real characterisation lives in the original isolation papers.

What hazard classification applies?

None exists for this material; it has no entry in the European inventory [17]. That reflects the absence of commercial notification, not a finding of safety. Individual markers do carry classifications with counts worth reading: ergosterol is aggregated from 47 company reports arising from just 3 notifications, with H300 at 12.8% and H413 at 85.1%, and one of the 47 reports states it does not meet the criteria at all [10]. Handle the material as an unclassified fine organic powder and record in your risk assessment that you are proceeding by analogy.

Is the species protected in Poland?

Wild specimens are. Hericium erinaceum, Polish name soplówka jeżowata, appears in Annex 1 to the fungal species protection regulation, the annex of species under strict protection, and carries the flag (1), which the annex footnote defines as a species to which the derogation in section 7, point 1 does not apply [18]. That is a restriction on wild collection, not on cultivated material, and it is one reason cultivation dominates supply.

Why can you not just isolate the marker and sell that?

It is done, and the published accounting shows what it takes: a two-dimensional chromatographic isolation reported 19.4 mg of erinacine A from approximately 130 g of mushroom material at 97.4% chromatographic purity, needing normal-phase flash chromatography in one dimension and semi-preparative reversed-phase in the other specifically to remove structural isomers and analogues [32]. That is a yield near 0.015% by mass. The authors cite limited commercial availability and high price as their motivation, which tells you most of what you need to know about the economics of single-marker certification for this species.

What should I write on my own certificate for material of this class?

The species and how it was confirmed; the tissue; the process, including solvent and ratio; the analytes measured by name; the method with its calibration range; and each analyte value separately, with sums reported as sums. Do not write a single percentage. A specification with six terms that are each measurable is enforceable; a specification with one term that is not measurable is decoration. The same reasoning applied to defined single compounds gives short certificates — see coluracetam or CDP-choline, where one substance, one assay and one number suffice. The length of a certificate is set by the material, not by the diligence of the writer.

References

Registry records for the marker compounds

1. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 15658905, Hericenone C." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/15658905.
2. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 15658906, Hericenone D." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/15658906.
3. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 15658907, Hericenone E." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/15658907.
4. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 14482559, Hericenone B." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/14482559.
5. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 44588861, 3-Hydroxyhericenone F." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/44588861.
6. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 10410568, Erinacine A." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/10410568.
7. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 73082498, Erinacine E." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/73082498.
8. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 127047879, Erinacine S." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/127047879.
9. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 10252378, Erinacine C." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/10252378.
10. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 444679, Ergosterol." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/444679.
11. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 5351619, Ergothioneine." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/5351619.
12. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 439262, beta-Glucan." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/439262.
13. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 102195678, Hericerin." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/102195678.
14. National Center for Biotechnology Information. 2026. "PubChem Substance Record SID 530257490, Hericium Erinaceus Extract." PubChem. https://pubchem.ncbi.nlm.nih.gov/substance/530257490.

Taxonomic and inventory registers

15. GBIF Secretariat. 2026. "Hericium erinaceus (Bull.) Pers. — GBIF Backbone Taxonomy, usage key 5248508." Global Biodiversity Information Facility. https://www.gbif.org/species/5248508.
16. National Center for Biotechnology Information. 2026. "Taxonomy Browser: Hericium erinaceus, taxonomy ID 91752." NCBI Taxonomy. https://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?id=91752.
17. European Chemicals Agency. 2026. "Search for Chemicals — Substance Inventory." ECHA. https://chem.echa.europa.eu/.
18. Minister Środowiska. 2014. "Rozporządzenie Ministra Środowiska z dnia 9 października 2014 r. w sprawie ochrony gatunkowej grzybów." Dziennik Ustaw Rzeczypospolitej Polskiej, poz. 1408. https://api.sejm.gov.pl/eli/acts/DU/2014/1408/text.pdf.

Isolation and structure of the marker compounds

19. Kawagishi, Hirokazu, Motoharu Ando, and Takashi Mizuno. 1990. "Hericenone A and B as Cytotoxic Principles from the Mushroom Hericium erinaceum." Tetrahedron Letters 31 (3): 373–376. https://doi.org/10.1016/S0040-4039(00)94558-1.
20. Kawagishi, Hirokazu, Motoharu Ando, Hideki Sakamoto, Satoshi Yoshida, Fumihiro Ojima, Yukio Ishiguro, Nobuo Ukai, and Shoei Furukawa. 1991. "Hericenones C, D and E, Stimulators of Nerve Growth Factor (NGF)-Synthesis, from the Mushroom Hericium erinaceum." Tetrahedron Letters 32 (35): 4561–4564. https://doi.org/10.1016/0040-4039(91)80039-9.
21. Kawagishi, Hirokazu, Motoharu Ando, Kayoko Shinba, Hideki Sakamoto, Satoshi Yoshida, Fumihiro Ojima, Yukio Ishiguro, Nobuo Ukai, and Shoei Furukawa. 1992. "Chromans, Hericenones F, G and H from the Mushroom Hericium erinaceum." Phytochemistry 32 (1): 175–178. https://doi.org/10.1016/0031-9422(92)80127-Z.
22. Kawagishi, Hirokazu, Atsushi Shimada, Ryoko Shirai, Kenji Okamoto, Fumihiro Ojima, Hideki Sakamoto, Yukio Ishiguro, and Shoei Furukawa. 1994. "Erinacines A, B and C, Strong Stimulators of Nerve Growth Factor (NGF)-Synthesis, from the Mycelia of Hericium erinaceum." Tetrahedron Letters 35 (10): 1569–1572. https://doi.org/10.1016/S0040-4039(00)76760-8.
23. Kawagishi, Hirokazu, Atsushi Shimada, Satoshi Hosokawa, Hironobu Mori, Hideki Sakamoto, Yukio Ishiguro, Shiichi Sakemi, Jon Bordner, Nakao Kojima, and Shoei Furukawa. 1996. "Erinacines E, F, and G, Stimulators of Nerve Growth Factor (NGF)-Synthesis, from the Mycelia of Hericium erinaceum." Tetrahedron Letters 37 (41): 7399–7402. https://doi.org/10.1016/0040-4039(96)01687-5.
24. Kawagishi, Hirokazu, Ayano Masui, Shinji Tokuyama, and Tomoyuki Nakamura. 2006. "Erinacines J and K from the Mycelia of Hericium erinaceum." Tetrahedron 62 (36): 8463–8466. https://doi.org/10.1016/j.tet.2006.06.091.
25. Chen, Chien-Chih, Tsai-Teng Tzeng, Chin-Chu Chen, Ching-Li Ni, Li-Ya Lee, Wan-Ping Chen, Young-Ji Shiao, and Chien-Chang Shen. 2016. "Erinacine S, a Rare Sesterterpene from the Mycelia of Hericium erinaceus." Journal of Natural Products 79 (2): 438–441. https://doi.org/10.1021/acs.jnatprod.5b00474.
26. Wittstein, Kathrin, Monique Rascher, Zeljka Rupcic, Eduard Löwen, Barbara Winter, Reinhard W. Köster, and Marc Stadler. 2016. "Corallocins A–C, Nerve Growth and Brain-Derived Neurotrophic Factor Inducing Metabolites from the Mushroom Hericium coralloides." Journal of Natural Products 79 (9): 2264–2269. https://doi.org/10.1021/acs.jnatprod.6b00371.
27. Ma, Bing-Ji, Yan Zhou, Lian-Zhen Li, He-Min Li, Zhi-Ming Gao, and Yuan Ruan. 2008. "A New Cyathane-Xyloside from the Mycelia of Hericium erinaceum." Zeitschrift für Naturforschung B 63 (10): 1241–1242. https://doi.org/10.1515/znb-2008-1017.
28. Rupcic, Zeljka, Monique Rascher, Sae Kanaki, Reinhard Köster, Marc Stadler, and Kathrin Wittstein. 2018. "Two New Cyathane Diterpenoids from Mycelial Cultures of the Medicinal Mushroom Hericium erinaceus and the Rare Species Hericium flagellum." International Journal of Molecular Sciences 19 (3): 740. https://doi.org/10.3390/ijms19030740.
29. Li, Jing, Xu-Li Wang, Guang Li, Ping-Sheng Xu, Kang-Ping Xu, and Gui-Shan Tan. 2017. "Two New Isobenzofuranone Derivatives from the Fruiting Bodies of Hericium erinaceus." Journal of Asian Natural Products Research 19 (11): 1108–1113. https://doi.org/10.1080/10286020.2017.1307185.
30. Wang, Junhong, Jing Wu, Ryo Yamaguchi, Kaoru Nagai, Chengwei Liu, Jae-Hoon Choi, Hirofumi Hirai, Xiaonan Xie, Shoji Kobayashi, and Hirokazu Kawagishi. 2025. "Uncovering Hericenones from the Fruiting Bodies of Hericium erinaceus through Interdisciplinary Collaboration." Journal of Natural Products 88 (1): 80–85. https://doi.org/10.1021/acs.jnatprod.4c01018.

Quantification, isolation and method validation

31. Tang, Yijin, Ozan Kahraman, Anthony J. Goos, and Christine Fields. 2026. "Simultaneous UHPLC-UV Determination of Hericenones, Hericenes, Erinacines and Ergosterol in Hericium erinaceus Raw Materials or Products." Molecules 31 (3): 569. https://doi.org/10.3390/molecules31030569.
32. Naumoska, Katerina, Andrej Gregori, and Alen Albreht. 2025. "Two-Dimensional Chromatographic Isolation of High Purity Erinacine A from Hericium erinaceus." Journal of Fungi 11 (2): 150. https://doi.org/10.3390/jof11020150.
33. Liu, Mengchen, Liangliang Liu, Xiaoya Song, Yingjun Zhou, Yuande Peng, Chunliang Xie, and Wenbing Gong. 2024. "Isolation and Evaluation of Erinacine A Contents in Mycelia of Hericium erinaceus Strains." Foods 13 (11): 1649. https://doi.org/10.3390/foods13111649.
34. Turk, Ayman, Sang Won Yeon, Se Hwan Ryu, Sung Min Ko, Beom Seok Kim, Bang Yeon Hwang, and Mi Kyeong Lee. 2021. "Effect of Culture Conditions on the Content of Hericene A, an α-Glucosidase Inhibitory Constituent of Hericium erinaceus." Scientia Horticulturae 288: 110407. https://doi.org/10.1016/j.scienta.2021.110407.
35. Sari, Miriam, Alexander Prange, Jan I. Lelley, and Reinhard Hambitzer. 2017. "Screening of Beta-Glucan Contents in Commercially Cultivated and Wild Growing Mushrooms." Food Chemistry 216: 45–51. https://doi.org/10.1016/j.foodchem.2016.08.010.
36. Mirończuk-Chodakowska, Iwona, and Anna Maria Witkowska. 2020. "Evaluation of Polish Wild Mushrooms as Beta-Glucan Sources." International Journal of Environmental Research and Public Health 17 (19): 7299. https://doi.org/10.3390/ijerph17197299.
37. McCleary, Barry V., and Rachel Codd. 1991. "Measurement of (1→3),(1→4)-β-D-Glucan in Barley and Oats: A Streamlined Enzymic Procedure." Journal of the Science of Food and Agriculture 55 (2): 303–312. https://doi.org/10.1002/jsfa.2740550215.
38. McCleary, Barry V., David C. Mugford, et al. 1997. "Determination of β-Glucan in Barley and Oats by Streamlined Enzymatic Method: Summary of Collaborative Study." Journal of AOAC International 80 (3): 580–583. https://doi.org/10.1093/jaoac/80.3.580.
39. Zhang, Anqiang, Yinglin Deng, Peilong Sun, Xianghe Meng, and Jingsong Zhang. 2011. "Structural Elucidation of a Neutral Water-Soluble α-D-Glucan from the Fungus of Hericium erinaceus." Journal of Food Biochemistry 35 (6): 1680–1685. https://doi.org/10.1111/j.1745-4514.2010.00492.x.
40. Nguyen, The Han, Anupam Giri, and Toshiaki Ohshima. 2012. "A Rapid HPLC Post-Column Reaction Analysis for the Quantification of Ergothioneine in Edible Mushrooms and in Animals Fed a Diet Supplemented with Extracts from the Processing Waste of Cultivated Mushrooms." Food Chemistry 133 (2): 585–591. https://doi.org/10.1016/j.foodchem.2012.01.061.
41. Jasinghe, Viraj J., and Conrad O. Perera. 2005. "Distribution of Ergosterol in Different Tissues of Mushrooms and Its Effect on the Conversion of Ergosterol to Vitamin D2 by UV Irradiation." Food Chemistry 92 (3): 541–546. https://doi.org/10.1016/j.foodchem.2004.08.022.
42. Raja, Huzefa A., Timothy R. Baker, Jason G. Little, and Nicholas H. Oberlies. 2017. "DNA Barcoding for Identification of Consumer-Relevant Mushrooms: A Partial Solution for Product Certification?" Food Chemistry 214: 383–392. https://doi.org/10.1016/j.foodchem.2016.07.052.
43. Schoch, Conrad L., Keith A. Seifert, Sabine Huhndorf, Vincent Robert, John L. Spouge, C. André Levesque, et al. 2012. "Nuclear Ribosomal Internal Transcribed Spacer (ITS) Region as a Universal DNA Barcode Marker for Fungi." Proceedings of the National Academy of Sciences 109 (16): 6241–6246. https://doi.org/10.1073/pnas.1117018109.

Reviews and comparative context

44. Friedman, Mendel. 2015. "Chemistry, Nutrition, and Health-Promoting Properties of Hericium erinaceus (Lion's Mane) Mushroom Fruiting Bodies and Mycelia and Their Bioactive Compounds." Journal of Agricultural and Food Chemistry 63 (32): 7108–7123. https://doi.org/10.1021/acs.jafc.5b02914.
45. Thongbai, Benjarong, Sylvie Rapior, Kevin D. Hyde, Kathrin Wittstein, and Marc Stadler. 2015. "Hericium erinaceus, an Amazing Medicinal Mushroom." Mycological Progress 14 (10): article 91. https://doi.org/10.1007/s11557-015-1105-4.
46. Ma, Bing-Ji, Jin-Wen Shen, Hai-You Yu, Yuan Ruan, Ting-Ting Wu, and Xu Zhao. 2010. "Hericenones and Erinacines: Stimulators of Nerve Growth Factor (NGF) Biosynthesis in Hericium erinaceus." Mycology 1 (2): 92–98. https://doi.org/10.1080/21501201003735556.
47. Wasser, Solomon P. 2002. "Medicinal Mushrooms as a Source of Antitumor and Immunomodulating Polysaccharides." Applied Microbiology and Biotechnology 60 (3): 258–274. https://doi.org/10.1007/s00253-002-1076-7.
48. Li, I-Chen, Yen-Lien Chen, Wan-Ping Chen, Li-Ya Lee, Yueh-Ting Tsai, Chin-Chu Chen, and Chin-Shuh Chen. 2014. "Genotoxicity Profile of Erinacine A-Enriched Hericium erinaceus Mycelium." Toxicology Reports 1: 1195–1201. https://doi.org/10.1016/j.toxrep.2014.11.009.

Additional registry records cited in sections 2, 3, 6, 8, 9, 10 and 11

49. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 9980261, Erinacine B." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/9980261.
50. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 10342778, Erinacine F." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/10342778.
51. National Center for Advancing Translational Sciences. 2026. "Global Substance Registration System: Hericium erinaceus whole, UNII Y62T8P9AAP; Hericium erinaceus fruiting body, UNII BU6Y4J4C2E." GSRS. https://gsrs.ncats.nih.gov/ginas/app/beta/substances/Y62T8P9AAP.
52. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 16085268, Erinacine K." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/16085268.
53. National Center for Biotechnology Information. 2026. "PubChem Compound Summary for CID 10206893, Erinacine Q." PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/10206893.