Blood-brain barrier models: one route publishes every number, the other publishes none

The blood-brain barrier is the reason most central nervous system candidates never reach their target, and it is one of the very few tissue models in this catalogue where a buyer can put a real number on the purchase before speaking to a salesperson, provided they pick the right route.

Two routes exist. They are priced in completely opposite ways, and the difference between them is not cost.

The self-assembly route is publicly priced end to end. A dedicated barrier supplement carries a live list price, every microfluidic plate format carries a live list price, bundles pairing the two carry live list prices, and primary human brain microvascular endothelial cells carry a live list price. You can cost a barrier experiment to the chip before committing to anything.

The prepared route publishes nothing. Perfusable primary human brain endothelial tubules, prepared by the vendor’s own staff, supplied at 40 or 64 tissue models per plate with all reagents, medium and optimised protocols included and explicitly no cell or extracellular matrix handling required by the customer, behind “Discuss your needs”.

So the decision that ought to be made on skill availability and transporter requirements gets made, in practice, on whichever route happened to publish a number. That is the buying problem this page exists to fix.

What is purchasable today, plainly

The bundle arithmetic, checked

One of the very few places in this market where a published discount can be independently verified.

On the largest bundle: IDTX40 at $2,220.00 plus five BBB Boost vials at $190.00 each is $3,170.00, against a bundle price of $2,690.00, a saving of $480.00. Both component prices and the bundle price came from the same public store endpoint on the same date, so the arithmetic is checkable by anyone who wants to repeat it.

Two observations a buyer can act on. The saving is essentially one free supplement vial per bundle, which means the discount scales with the supplement rather than the plate. And the cheapest way to run a first barrier experiment is the idenTx 3 Chip bundle at $760.00 for 25 chips plus 2 vials of supplement, under $31 per chip including supplement, which is a genuinely low bar to establishing whether your cells form a barrier at all before committing to a plate format.

What the prepared route actually removes

It would be easy to read the priced route as the sensible one and the quote-gated route as vendor opacity. That is not the right conclusion, and the reason is worth stating carefully.

The prepared product is described as 3D perfusable primary human brain endothelial tubules with functional transporter expression, prepared by the vendor’s staff, supplied at 40 or 64 tissue models per plate with all reagents, culture medium and optimised protocols included, and explicitly no cell or ECM handling required by the customer.

What that removes is not labour. It is operator-induced variability and the seeding skill requirement. Patterning an extracellular matrix gel in a microfluidic lane by surface tension, then seeding endothelium against it so that a patent lumen forms in every chip, is a manual skill with a learning curve. A group learning it on their own plates will produce a first run with a failure rate they cannot distinguish from a biological result.

So the two routes trade differently:

The clean decision rule: if the barrier model is the object of study, buy the parts. If the barrier model is an instrument you need working so you can study something else, ask for the prepared quote, and accept that you cannot compare it on cost until it arrives.

The field that should decide it, and usually does not

Functional transporter expression. For most CNS programmes this is the entire point of using a barrier model at all, and it is the field most likely to be assumed rather than specified.

There is a distinction here that costs programmes real money: a transporter that is expressed and a transporter that functions are different claims, demonstrated by different experiments. Efflux transporter protein visible by immunostaining tells you the protein is present. Directional transport of a substrate, inhibited by a known inhibitor, tells you the transporter works and is correctly polarised. Only the second one predicts anything about brain penetration.

Force the quote to state which one it is offering, for which named transporters, measured how.

The same caution applies to barrier numbers. TEER values are not comparable across measurement methods, electrode geometry, chopstick versus endohm configuration, and area normalisation all move the number substantially. A quote that promises “high TEER” without naming the method has not agreed to anything you can hold it to.

What the published record supports

These are findings under the conditions each paper describes, offered so you can judge whether a proposal is ambitious or ordinary. They are not product specifications.

Perfused, three-cell-type barrier models at genuine throughput are established. A perfused human blood-brain barrier on-a-chip built from brain endothelial cells, astrocytes and pericytes has been reported in a two-lane or three-lane microfluidic platform harbouring 96 or 40 chips respectively in a 384-well plate format, with a perfused endothelial vessel grown against an extracellular matrix gel patterned by surface tension techniques, and with anti-transferrin receptor antibody penetration quantified against a control antibody. That is the architecture the self-assembly route sells you the parts for, and it has been executed at plate scale.

The iPSC-derived brain endothelium question is unresolved and you should know it before you specify. Derivation of blood-brain barrier endothelial cells from human pluripotent stem cells was reported in 2012 and became a widely used route because it gives very high barrier values. A 2021 paper argued that pluripotent stem cell-derived cells widely used as brain microvascular endothelium are misidentified epithelium and require ETS factors to acquire vascular fate. Independent work has validated barrier function and endothelial behaviour in iPSC-derived BBB microvessels.

We are not adjudicating that. The procurement consequence is concrete: if you specify an iPSC-derived source, expect a reviewer to raise the identity question, and decide in advance whether your programme can absorb it. Primary human brain microvascular endothelial cells at $920.00 per vial avoid the argument entirely, at the cost of donor variability and finite lifespan.

A note on distributors

At least one large distributor resells these consumables at a published uplift over direct purchase. If a quote for barrier plates arrives through a distributor rather than the manufacturer, check the line against the manufacturer’s public store price before approving it. The store endpoint is public and takes a minute to read. This is the single easiest saving available in this category and it is routinely left on the table.

What to put in the RFQ

• Route. Prepared perfused tubules, or plate-plus-supplement you seed yourself. Say which, so the two quotes are not silently different products.
• Cell source, primary human brain microvascular endothelial, iPSC-derived, or immortalised, and whether the identity question above is acceptable to your reviewers.
• Whether astrocyte and pericyte compartments are required for a triple co-culture, and in what configuration.
• Named efflux and influx transporters that must be functionally demonstrated, with the word “functionally” written down and the demonstrating experiment named.
• Barrier acceptance criterion: TEER value or permeability coefficient for a stated tracer, plus the measurement method and electrode configuration.
• Number of tubules or chips and the replicate structure, remembering that a plate here carries 40 or 64 models on the prepared route and 3, 9 or 40 chips on the self-assembly plates.
• Perfusion regime, and whether pumpless rocker-driven flow is acceptable.
• Imaging requirement and the optical properties the plate must have to support it.
• Whether a disease or donor-specific background is required, which effectively rules out the prepared route as currently published.

Our position

We broker this. Every prepared tissue delivered through us is made by a named partner and labelled as partner-manufactured.

The honest summary: this is one of the few categories where a competent lab can and probably should just buy the parts. Under $800 gets you 25 chips and two vials of supplement, which is enough to find out whether your cells form a barrier before any larger commitment. The parts are public, the bundle discounts are real and checkable, and the arithmetic above is the whole of it.

The broker’s value here is confined to two things: getting a number out of the quote-gated route so the comparison can actually be made, and forcing the functional-transporter and TEER-method wording into the quote so that what arrives is testable against what was promised. Send us the specification if you want the prepared route priced against the parts bill above.