Dispase

Dispase is a neutral protease isolated from culture filtrates of Paenibacillus polymyxa (described in the older literature as Bacillus polymyxa). It is catalogued as EC 3.4.24.4, CAS 9001-92-7, with a molecular mass of approximately 35,900 Da. "Dispase II" refers to the commonly supplied grade rather than a different enzyme.

Structurally it is a metalloprotease that binds one zinc ion and four calcium ions per subunit. The zinc is catalytic - remove it with a chelator and you have an inactive apoenzyme. The calcium is structural, maintaining the fold and protecting the enzyme against autolysis. Both facts have direct practical consequences that are covered below, and both are the main way dispase differs in handling from trypsin.

Its substrate profile is the reason it exists as a distinct product rather than being one more general protease. Stenn and colleagues characterised it in 1989 and showed that dispase cleaves fibronectin and type IV collagen, but not laminin, type V collagen, serum albumin or transferrin. Type I collagen is degraded only minimally, and the action on type IV collagen produces several stable degradation products rather than complete digestion.

Type IV collagen and fibronectin are basement membrane components. Laminin, which dispase leaves intact, is another. What this means in practice is that dispase attacks the layer between an epithelium and its underlying connective tissue while leaving the epithelial cells' own attachments to each other largely alone. That is a very specific and useful capability: it separates tissue at the basement membrane, and it lifts sheets rather than dissolving them into single cells.

Because it is bacterial, dispase is animal-origin-free, which matters in regulated workflows where porcine trypsin is a problem.

Dispase activity is defined by a casein digestion assay: one unit is the amount of enzyme that liberates Folin-positive amino acids equivalent to 1 micromole of tyrosine per minute at 37 degrees C, pH 7.5, using casein as substrate.

The specific activity of commercial dispase varies substantially by grade and by lot. A partially purified preparation may be specified at around 1.2 units/mg, while a higher-grade product is specified at 4 units/mg or more - a threefold or greater difference in enzyme activity for the same mass of powder.

This is the single most important practical point about dispase, and it is where protocols most often fail to transfer between laboratories. A protocol written as "1 mg/mL dispase" is not reproducible unless it also states the specific activity of the lot used. One laboratory's 1 mg/mL is another's 3 mg/mL in enzymatic terms. Where a published method gives a concentration in units per millilitre, use units; where it gives milligrams, check the certificate of analysis for the lot and convert.

The published literature reflects both conventions. Skin separation methods are often written in units - 500 to 1000 U/mL in DMEM has been used for peeling epidermal sheets from human skin after 24 hours - while stem cell and epithelial protocols are commonly written in mass terms, such as 1 mg/mL for passaging human pluripotent stem cells, often in combination with 1 mg/mL collagenase type IV.

Supplier documentation also differs on the pH optimum, with figures from 5.9-7.0 through to 8.5 appearing on different technical sheets. What is consistent across sources is that the enzyme is active and stable across a broad neutral range, roughly pH 4 to 9, which is where the name "neutral protease" comes from. In practice, running the digestion in a standard balanced salt solution at physiological pH works, and the discrepancy in quoted optima is not something you need to resolve to get a result.

Preparing a stock. A common preparation is 5 mg/mL dissolved in cell-culture-grade water, passed through a 0.2 micron filter for sterility. Ready-made sterile solutions are also sold, typically at 10 mg/mL, and avoid the weighing and filtration steps. Dispase is stable for around 12 months at 2-8 degrees C in solution; frozen stocks should be aliquoted and thawed once.

Critical: dilute in a calcium-containing buffer. Dispase needs calcium for structural stability, and the optimal calcium concentration is around 2 mM. Diluting dispase into calcium- and magnesium-free PBS or DPBS - the buffer you would reach for with trypsin-EDTA - progressively destabilises the enzyme. Use HBSS or DPBS with calcium and magnesium, or a complete medium. This is the opposite of the rule for trypsin-EDTA, and it is a common source of unexplained loss of activity.

Protocol 1 - separating epidermis from dermis. Published approaches span a wide range of concentration and time, trading one against the other:

1. Rinse the skin sample and cut it into strips of a few millimetres width.
2. Float the tissue dermis-side down on dispase solution. Two commonly used conditions are 1 mg/mL in PBS at 37 degrees C for around 30 minutes, or a higher concentration (500-1000 U/mL) with an extended incubation of up to 24 hours; overnight incubation at 4 degrees C for 15-18 hours is a third widely used option and is the condition given in some supplier instructions.
3. Transfer to a dish of buffer and peel the epidermal sheet from the dermis with fine forceps. It should lift as an intact sheet.
4. Proceed to a second enzyme if single keratinocytes are wanted - trypsin-EDTA is normally used at this point to dissociate the sheet.

Protocol 2 - releasing an epithelial monolayer intact. Where the objective is a sheet rather than a suspension:

1. Aspirate medium and rinse the monolayer with a calcium-containing balanced salt solution. Do not use an EDTA-containing wash.
2. Add dispase solution sufficient to cover the growth surface.
3. Incubate at 37 degrees C, checking every 10-15 minutes. The sheet begins to lift from the edges.
4. Collect the released sheet by gentle aspiration with a wide-bore pipette. Do not pipette vigorously; the point of using dispase is that the sheet stays whole.

Protocol 3 - dispase as a secondary enzyme. Dispase is very commonly used alongside collagenase in solid tissue dissociation, where collagenase attacks the fibrillar interstitial collagen and dispase handles the basement membrane and fibronectin. The two are usually combined in the same digestion in a calcium-containing buffer at 37 degrees C with gentle agitation, with time determined by tissue type and fragment size.

The reliable way to stop dispase is chelation. Adding EDTA or EGTA to a final concentration of around 1 mM strips the catalytic zinc and inactivates the enzyme. 1,10-phenanthroline and heavy metals also inhibit it. This is a hard chemical stop and it is dependable.

Serum is not a reliable stop, and sources disagree about it. Some published protocols quench dispase digestions by adding an equal volume of medium containing 10% serum, and report that it works. Some supplier documentation, by contrast, states that dispase is not inhibited by serum - which is presented as an advantage, since it means the enzyme can be used in complete medium. Both statements can be partly true: serum contains broad-spectrum protease inhibitors, but dispase is not a serine protease, and the serine protease inhibitors that shut down trypsin so decisively do not apply here. Soybean trypsin inhibitor, in particular, has no effect on metalloproteases.

The practical position: do not rely on serum alone to stop dispase. If you need the reaction to stop at a defined point, chelate. If you need to remove the enzyme, dilute and wash by centrifugation. Serum-containing medium is fine as the diluent for that wash, but treat the dilution rather than the serum as the thing doing the work.

The corollary catches people out in the other direction. Because chelators inactivate dispase, you cannot carry cells straight from an EDTA-containing step - trypsin-EDTA, Versene, or ACK lysis buffer, which contains EDTA - into a dispase digestion and expect the dispase to work. Residual EDTA will inhibit it, sometimes completely. Wash twice in a calcium-containing buffer between the two steps.

These three are not interchangeable, and choosing between them is mostly a question of what holds your tissue together and what you need to survive the process.

Trypsin is a serine protease with broad specificity for peptide bonds after lysine and arginine. It is fast, cheap and effective, and it is indiscriminate: it cleaves cell-surface proteins along with everything else, so surface epitopes are damaged and cell-cell junctions are destroyed. It produces single-cell suspensions, which is often exactly what you want and sometimes exactly what you do not. It is poor at digesting fibrillar collagen, so it performs badly on connective-tissue-rich solid tissue.

Collagenase, from Clostridium histolyticum, cleaves native triple-helical collagen, which almost no other protease can do. It is the enzyme for solid tissue with a substantial interstitial collagen matrix. Commercial collagenase is supplied as blends distinguished by the balance of secondary activities - clostripain, caseinase and tryptic activity - and the type designations reflect the tissues those blends suit. Type I is the general-purpose blend used for epithelium, liver, lung, fat and adrenal tissue; type II has higher clostripain activity and suits heart, bone, muscle and cartilage; type IV has low tryptic activity and is chosen when receptor integrity matters, classically for pancreatic islets and adipocytes. Like dispase, collagenase is a calcium-dependent metalloprotease and is inhibited by EDTA.

Dispase occupies the space neither covers: basement membrane. It does not touch fibrillar collagen I to any useful degree and it does not aggressively strip the cell surface, but it dismantles the fibronectin and type IV collagen anchoring an epithelium to its substrate. That is why it is the enzyme for dermal-epidermal separation and for releasing epithelial sheets whole, and why it is so often paired with collagenase rather than used alone on solid tissue.

The choice for epithelial sheet integrity is the clearest case. If you want an intact sheet, use dispase and handle it gently. If you want single cells from that sheet, use dispase first and then trypsin-EDTA. Using trypsin from the start gives you single cells but destroys the junctional proteins you might have wanted to study, and using collagenase alone on skin will chew the dermis without cleanly releasing the epidermis.

Storage. Lyophilised dispase is stable for around 12 months at 2-8 degrees C. Solutions are commonly stored frozen, protected from light, and should be aliquoted so that a working volume is thawed once rather than cycled. As with any protease, repeated freeze-thaw causes autolytic loss of activity that is invisible until a digestion mysteriously stops working.

Sterility and endotoxin. Dispase preparations are bacterial fermentation products and are a plausible route for endotoxin into a primary cell preparation. For primary immune cells, stem cells or anything heading toward a regulated application, use a preparation with a stated endotoxin specification, and dissolve powder in cell-culture-grade water rather than laboratory deionised water.

Common problems

Digestion is slow or does not work. Check for chelator carryover first - residual EDTA from a previous step is the most frequent cause. Then check whether the diluent contained calcium; dispase diluted into Ca/Mg-free PBS loses stability. Then check the lot's specific activity against the one the protocol was written for.

The epithelial sheet fragments instead of lifting whole. Usually over-digestion, mechanical handling that is too rough, or both. Check earlier and more often, and collect with a wide-bore pipette rather than a standard tip.

Cell viability is poor after dissociation. Dispase is gentle relative to trypsin but it is still a protease, and overnight digestions at 37 degrees C are a long time for cells to spend in enzyme. Where a protocol permits it, an extended digestion at 4 degrees C is kinder than the equivalent at 37 degrees C.

Yields vary between lots. This is the units problem described above. Dose in units per millilitre, record the lot and specific activity in the method, and re-titrate when the lot changes.

The downstream assay behaves oddly. Dispase leaves cell-surface proteins largely intact compared with trypsin, but "largely" is not "entirely", and the fibronectin it cleaves is part of the matrix that the cells will need to re-attach. Allow a recovery period in complete medium before assays that depend on adhesion.