WO1980000374A1

WO1980000374A1 - Improvements relating to displacement electrophoresis

Info

Publication number: WO1980000374A1
Application number: PCT/GB1979/000124
Authority: WO
Inventors: A Martin; F Hampson
Original assignee: A Martin; F Hampson
Priority date: 1978-07-25
Filing date: 1979-07-24
Publication date: 1980-03-06
Also published as: JPS55500521A; GB2026546B; EP0016121A1; GB2026546A; GB2057504A

Abstract

Displacement electrophoresis is carried out on a gel (11) having the form of a tube along which substances are separated in the axial direction. The method makes use of an agarose or polyacrylamide gel (11) formed on and supported by an inner tube (12) which also acts as a conduit for coolant. Very thin gels are used to provide a short path for heat removal e.g. up to 3 mm. The gel is supported on one side only and the other side is in contact with electrolytes used in the process i.e. leading, displacing, and terminating electrolytes. To avoid drying out of the thin gel it is maintained in contact with an inert water-immiscible liquid where it is not in contact with electrolyte. The method is of especial importance for the separation of ampholyte molecules such as proteins or peptides. To provide an indication of the extent of separation fluorescent ampholytes are used as marker substances in very small quantities.

Description

IMPROVEMENTS RELATING TO DISPLACEMENT ELECTROPHORESIS This invention relates to displacement electrophoresis which is a method of separating charged species characterised by giving all ions under observation the same migration velocity so that they become separated into a number of consecutive zones in contact with each other and arranged in the order of their mobilities. One such method is disclosed in U.K. Specification No. 1,301,065. Displacement electrophoresis is attractive for preparative work because the amount of material trans¬ ported per coulomb is higher than inother methods. It is widely applicable to the separation of charged molecules e.g. proteins, peptides, amino acids, and nucleic acids.

The method of displacement eletrophoresis, also called isotaehophoresis, has been very successful for small molecules in the hands of Everaerts and his colleagues, and other investi- gators, using a long thin tube containing a more or less viscous liquid. However, attempts to separate proteins in the same type of apparatus have met with little success. The difficulties arise because of absorption on the walls of the tube, because the ratio of surface to volume is very large and because the protein solution can differ considerably in density from the solution at the end of a zone, thus making the zone front unstable. Hjerten by rotating a horizontal tube about its axis has lessened this latter difficulty but at the expense of undesirable complications of the apparatus.

OMPI _. The use of a gel instead of a viscous liquid is impractical in a narrow tube because if the- gel is very weak and dilute, electroendosmosis causes enough movement to crack the gel, which results in the simultaneous presence of gel and synerezed liquid in which the proteins run irregularly. If on the other hand the gel is strong enough to prevent syneresis, bubbles containing only water vapour are formed which results in excessive heating below the bubbles. If a thin slab of gel is used uncovered, it can shrink as the result of electroendosmosis without necessarily causing trouble. However if the potential gradient is high enough to give good resolution, evaporation from the wet surface eventually leads to dry, non conducting regions, and before complete dryness occur the resistance becomes so high that excessive heat is generated and the protein is denatured. If evaporation is prevented by laying on top of the gel a plastic or glass cover, the gel synereses at the point of contact with the cover and liquid flows in an irregular pattern and causes distorted zones. Also since the cooling is always better at the edges than in the middle of a thin slab, the temperature difference leads to different speeds of movement of the fronts and also causes different rates of electroendosmosis. Syneresis then occurs where there is swelling, and the synerezed liquid flows to the place where the gel is shrinking. If the lid extends beyond the sides of the gel synerezed liquid may well run outside the gel area into the space between the lid and the side supports.

We have avoided the irregularity of temperature at right angles to the direction of the current by employing a vertical tube surrounded by a thin annulus of gel. We have also avoided evaporation by immersing the gel, where it is not making contact with the rest of the electrical circuit, in a chemically inert water-immiscible insulating liquid, whose density is sufficiently different from water to cause the meniscus to be well defined. Under these conditions we have found a surpris¬ ingly complete absence of syneresis, and have had no difficulty in getting very regular zones.

It is desirable to have the zones migrating downwards, so that the hotter displacer is above the separating proteins. Convection in the insulating liquid carriesthe heat generated in the displacer upwards away from the proteins. This can be clearly seen as an upward moving schlieren pattern in the insulating liquid.

If separation has to be made between tvo proteins of closely similar mobility then necessarily the zones have to be run many times the combined length of the zones to obtain separation, and hence there is a big potential drop over the region which has become occupied by the displacing electro¬ lyte (terminating). We have found that lengths of more than a few centimetres of displacer are unstable, probably because the upper part becomes excessively hot due to irregular

^■- TEAtr

OMPI /,„ WIPO _{< <} current distribution, and that ultimately the protein zones are also affected. It is however easy to reduce the level of the inert liquid, replacing it with a solution of displacer. The current can then be carried by a large area of electrolyte with a correspondingly low resistance. This, results in much less heat being produced and little change in potential is required to maintain the current at a constant value. Indeed a constant voltage power pack can be used, and the rate of addition of electrolyte and removal of inert liquid adjusted to maintain a constant current. The conductivity of the lead zone can always be made so high that its length is unimportant.

According to the present invention a device for use in displacement electrophoresis comprises a gel on which sub¬ stances are to be? separated having the form of a tube supported on one side only by means of a supporting member. The tubular gel is required to have a very low wall thickness in order to provide the shortest feasible path for dissipation of heat generated in the gel at a given potential difference between one end of the tube and the other. To this end the thickness of th gel should.in most cases not be above about 3 millimetres. The

lower limit of gel thickness is governed mainly by considerations of fragility in handling and the rapidity of evaporation from the thinnest layers. In practice, a convenient range is from 0.1 to 3 πim preferably 1 - 2 mm. These figures are largely independent of the size of the apparatus. It will of course be appreciated that tubular gels of the dimensions contemplated

•-^■■ SLE

O r_b IP are not self-supporting and require the support of another member preferably internal to the gel tube, such as a glass tube, which provides a path for coolant to remove heat from the gel as electrophoresis proceeds. Such an arrangement provides support for the gel on one side only and this is a feature of considerable practical importance and overcomes difficulties experienced when the gel is supported on both sides.

Typical gel materials for the purposes of the present invention are polyacrylamide and agarose and these may be produced in tubular form on the support by polymerisation in situ or by casting from solution as- appropriate. In the polymerisation of a polyacrylamide gel on to a tubular support it has been found important forthe tube to be accurately dimensioned and in practice this is conveniently achieved by using a thick walled glass tube which is accurately ground to a uniform external diameter. The gel is then formed on the tube by placing the latter in a split mould and introducing the polymer forming materials into the space between the tube and mould surface.

An improved method for the production of a supported tubular gel which is particularly suitable for use with agarose is by dipping the tube to be coated into a column of molten agarose or equivalent gel forming material and establish- ing a continuous flow of the latter in the axial direction past the tube- whilst providing internal cooling of the tube. In operating the last mentioned method it is necessary to pay due regard to the properties of the commercial agarose used and to determine by trial and experiment the optimum concentra- tion of the agarose solution and the cooling rate and also the speed at which the agarose is caused to flow past the tube.

By the use of either of the two methods described above a supported tubular electrophoresis gel may be produced at the desired time of use or may be manufactured as an article for* sale, storage and disposal after use. In the latter case the tube will require to be sheathed in some water impermeable outer covering to prevent evaporation from the- gel.

The present invention also comprises a method of perform¬ ing displacement electrophoresis on a gel characterised in that the gel is in the form of a tube and the displacement occurs along the length of the tube. When operating this method of separating substances by displacement electro¬ phoresis along the length of a tube of gel impregnated with the leading electrolyte, evaporation from the gel over the effective separation zone can be prevented by contact of the gel with an inert liquid immiscible with the displacer (terminating) electrolyte. Preferably the gel is cooled internally during operation of the method. Desirably, the arrangement is such as to permit the position of the interface between the immiscible non- electrically conducting liquid and the displacing electrolyte

WI to be varied so that only part of the annular gel, preferably a small fraction of the total length, containing the displacer electrolyte carries theelectric current at any one time during operation whereby the generation of heat and the potential drop required are substantially reduced or minimised.

As indicated above a preferred method for producing a supported tubular gel in accordance with this invention, which is particularly suitable for use with gels which form upon cooling of a gel-forming solution or mixture, comprises causing a stream of the gel-forming liquid to flow in the axial direction along one surface of the tubular supporting member whilst cooling is applied on its other surface. This method has been -carefully studied for the production of agarose gels and the conditions under which it may be applied for optimum results with this material will now be discussed. Commercially available agarose is a mixture of polysaccharrides derived by the fractionation of agar agar and shows considerable variation from batch to batch in properties which are important for the purposes of electrophoresis. For example some batches of material require treatment to reduce the amount of electroendosmosis obtained with gels made from them and this is usually achieved by hydrolysis with aqueous sodium hydroxide in the presence of sodium borohydride. Gels of good strength for the purposes of this invention have been obtained using agarose at a concentration from about 1 to about 1-5% w/v. The properties of agarose are also highly temperature dependent and it has been found preferable* when operating with concentrations of the order of 1% as mentioned above to work with an initial temperature of the gel solution .between 65 and 80 C at which its properties remain unchanged for at least several hours.

Cooling of the gel from this temperature to cause setting of a thin layer around the supporting member gives very good results.

The rate of deposition of agarose and the evenness of the tubular gel so formed are dependent also upon the velocity of flow in the axial direction. Optimum flow conditions will also depend upon the geometry of the apparatus used and are best determined by trial and experiment in each case.

Displacement electrophoresis depends on the establishment of a series of zones comprising a solution containing a fast moving ion, the leading ion, and a solution containing a slow moving ion, the terminating ion, and an intermediate zone containing the mixture of substances to be separated. As electrophoresis proceeds, resolution of the components of the intermediate zone occurs to an extent depending on the differences in mobility of the individual components of the mixture and the duration of the experiment. If the mobility differences are so small that the electricalforces sorting the molecules are comparable in strength with diffusional forces very little separation will occur. To apply the technique successfully it is important both to have

- uR

O WI information about the course of the particular separation and a means of identifying the divisions between the various zones of material at the end of the process in order to aid recovery of the desired component. It has been proposed in U.K. specification No. 1,301,065 to modify displacement electrophoresis for protein separation by setting up a mobility and pH gradient consisting of ampholytes between the leading and terminating electrolyte and allowing the proteins to separate themselves in this gradient. This proposal however has the disadvantage of adding further compositional complexity to an already complicated system' and is not a satisfactory solution of the difficulty.

In accordance with the present invention fluorescent ampho¬ lytes are used as markers of discontinuity in the mobility profile and hence of the junctions between separated proteins. Fluorescent ampholytes are new substances obtained by associating ampholytes with compounds having fluorescent properties. The association may be by physical means but is preferably achieved by chemical attachment of fluorescent substances for example by the use of fluorescent compounds containing chemical functional groups which react with the basic or acidic groups in the ampholyte. It is most convenient to operate with functional groups which react with amino groups in the ampholyte, examples of such functional groups being acid halide groups, cyanate, and

OMPI thiocyanate groups. The preparation of fluorescent ampholytes will be further described hereinafter.

In carrying out the displacement electrophoresis procedure of the present invention use is made of a mixture of fluorescent molecules having a range of mobilities and fluorescent.enough to be detected at concentrations which are tiny compared with the concentration of the proteins, so that they do not add significantly to the total mass of the mixture. Consequently they do not disturb the mobility profile of the separating mixture or themselves set up a mobility gradient. If a mixture of two easily separated proteins contains a tiny amount of mixture- of fluorescent markers with mobilities which vary continuously or almost continuously from that of the lead to that of the displacer, then as the separation proceeds first the markers most different in mobility from the mean mobilityof the mixture of proteins will form fluorescent ends to the protein band. Then as the proteins themselves begin to separate, a fluorescent band will form at the emerging junctions, being at first very fuzzy where there are fronts between mixed zones and gradually becoming sharper and stronger as the separation proceeds. When the separation is completed each front between the zones will have some fluorescent marker concentrated in it. In the case of incomplete separation the fuzzyness of the bands gives an indication of which zones are pure and which are still mixed. Furthermore, it is possible that the distribution of residual fluorescence between the fluorescent delineating zones will allow one to deduce whether or not the particular zone is intrinsically mixed.

We have produced typical fluorescent markers by reacting 1KB carrier ampholytes which are mixtures of polyam nopoly- carboxylic acids with either fluorescein isothiocyanate or Rhodamine B200 sulphonyl chloride.

Fl-NCS + RNH ^ Fl-NH-C - NHR

Rd-SO Cl RNH_ -^■^Rd-SO - NHR 2

Other carrier ampholytes containing sulphonic acids or phos— phonic acids in place of carboxylic acids can be used.. In each case the reaction is accompanied by the loss of. an ionisable group so that at a given pH it is more

negatively charged or less positively charged, than the parent molecule. However the mobility can be adjusted by reacting with an amino epoxide as shown below.

0

R.2 NCH 2 -Clfi 2 ₊ NH -* R _→ ^R2^NCH2-^CH°^H-^CH2^NHR

Other amine containing reagents could also be used and the mobilities could be altered in the opposite sense by reacting with acidic reagents such as glycidic acid ; chloracetic acid ,

O PI Λ. WIPO - chlorosulphonic acid etc. All these modifications have the advantage of greatly increasing the numbers of different species of different mobility in the mixture. After reaction mobility markers with a particular range of mobility can be separated out, if desired, by displacement electrophoresis* between lead and displacer of chosen mobility. This- is not usually necessary however since if the crude mixture is added to the protein then the markers which are too mobile will travel zone electrophoretically in the leading electrolyte and the markers which are too slow or which have the wrong charge will migrate- zone electrophoretically in the displacing electro¬ lyte either in the same direction as the protein or in the opposite direction. In any event they will travel distinctly from it and not detract from either the separation or* the observation of it.

Synthesis of fluorescent mobility markers which are- anionlc at around pH 8.

LKB carrier ampholyte pH range 9 - 11 concentration 20% w..v (0.1 ml) was mixed with water 0.7 ml and with carbonate-bicarbonate buffer pH 9_> 2.02M, (0.2 ml) to this was added fluorescein isothiocyanate 2 mg in dioxane 0.25 ml. The mixture was left at room temperature overnight and extracted twice with 1 ml of chloroform. The aqueous layer was then warmed under water pump vacuum to remove traces of chloroform and used

__O without further treatment.. 0.2 (,1 of the mixture being ample to produce clearly visible- fluorescent markings. Synthesis of fluorescent markers which are cationic at pH 8. 1 ml of anionic markers produced as above was-■treated " with Sin 2.3 epoxypropyldiethylamine, the mixture- was allowed to react overnight. Next day it was treated with another 5 X of 2.3 epoxypropyldiethylamine and allowed to react overnight. The product, which was slightly less fluorescent than the anionic markers, was used without further purification, l tl being used per separation.

The invention is illustrated in the accompanying drawings of whichs-

Figure 1 is a vertical sectional view of a split mould containing a tube to be coated with gel, Figure 2 is a similar view of the assembled apparatus,

Figure- 3 shows the pattern of separation of zones obtained by displacement electrophoresis,

Figure 4 is a diagrammatic sectional view of apparatus for forming a supported tubular gel by the drip and axial flow method.

The mould consists of two pieces 1, 2 of stainless steel or PTFE having corresponding semi-circular grooves machined in them so that when the two halves are bolted together through holes 3 the fit of the two halves ensures that the mould is watertight without recourse to jointing compounds. This is achieved for stainless steel by firstβur ace^*grinding the- faces of the two halves and then by gently hand-lapping them together. If necessary the mould surfaces may be coated with paraffin wax to prevent adhesion of the gel to the steel. When the mould is assembled for use the glass tube* 4 to be coated is threaded into it and held concentric by means of a PTFE centring cone- 5 at the top and the centring sleeve of a PTFE bottom piece 6. The uniformity of coating of the column depends upon the accuracy of these centring devices and upon the linearity and circularity of both the glass column and the metal mould. To this end the glass columns were constructed from thick-walled glass and the surface machined to accuracy in the lathe, using a high speed diamond, wheel.. In typical experiments the diameter^* of the- glass support may be 20.0 mm and the diameter of the hole in the mould is 22.4 mm, giving a gel thickness of 1..2 mm.

The paraffin wax treated mould is assembled by means of bolts. The bottom piece 5 is attached with an ^*0' ring 7 between it and the main body of the mould, ensuring liquid tightness. The tube 4 to be coated, which already has at the top end its centering cone, is threaded into the vertically mounted mould until its bottom end engages in the centring sleeve. The centring cone is then gently pressed down until it holds the column firmly in the mould. Then a solution of monomer accelerator and catalyst in degassed buffer solution is introduced into the space between the mould and the glass tube.

O WI This can more easily be achieved by siphoning the- liquid from a mixing flask through a small PVC tube attached to the hole in the bottom piece. The flow of liquid can be regulated by slowly raising the level of the mixing flask until the liquid starts to flow up through the slit in thecentring cone-. All this must be done within 3 - 4 minutes of adding the catalyst to the monomer solution. The mould and column are then set aside for at least half an hour to polymerize. After poly¬ merization the mould is disassembled and the polyacrylamide- coated tube is removed. The irregularly shaped lump of gel, which represents the internal volume of the bottom piece is removed from the tip of the gel, which is then ready, after washing with water and soaking in appropriate leading electrolyte, for use- in displacement electrophoresis. The- mould can also be used for the preparation of agarose columns. The assembled mould and glass column are o heated to approximately 50 C in an oven, after which the molten agarose is siphoned into the apparatus, which is then allowed to cool to room temperature, before the removal of the gel.

Referring to Figure 2, an annular gel 11 is cast on to an accurately ground cylindrical glass tube 12 which acts as an inner cooling jacket for the gel 11. The tube 12 has an internal cooling water inlet 13, leading in from the top and extending down to near the bottom of tube 12, and an

- REA T

OMPI outlet 14 at the top. The sample to be separated is impregnated in a paper strip 15 secured to the gel near its upper- end.

The tube 12 is mounted in a cylindrical glass container 16 and secured at its lower end in agarose gel plug 17 containing lead electrolyte.. The tubel6 has an inlet 18 for displacing electrolyte at the top and an outlet 19 for liquid at the bottom.

The system includes an upper electrode 20 which dips into displacing electrolyte and a lower electrode 21, the latter being disposed in an electrolyte compartment22 through which electrical contact is made to gel plug 17 via a membrane,23 supported by a screw cap and 0-ring arrangement 24..

The interface between dichlorobenzene and displacer electrolyte in container 16 is at 25 at the start of separation and at 26 at the end. The arrow 27 indicates the extent of separation of the components at the end.

METHOD OF RUNNING

Prior to use the gel is washed with 2 litres of leading electrolyte overnight. The sample solution up to 0.4 ml but more usually 0.2 ml. in volume, containing appropriate markers if desired, is applied to a Whatman No. 1 filter paper strip 20 mm wide with a length corresponding exactly tothe circumference of the gel which is about 75 mm. The paper with the sample on it is wrapped carefully round the circumference of the gel about 7-5 mm below its top and tied on with rayon thread.

A pool of molten agarose containing concentrated leading electrolyte is placed in the bottom of the wide tube on to the membrane fixed at end with o rings. The end of the annular gel is dipped into the agar and then clamped vertically. Cooling water is then circulated through the central tube. The end of the tube hangs in the bottom electrode vessel.

When the agarose has congealed the wide tube is filled with o-dichlorobenzene, until the top of the sampler paper is 5 mm below the o-dichorobenzene. Then terminating electrolyte is poured gently on top of the o-dichlorobenzene to a depth of about 100 mm. The top electrode is immersed in this solution and a potential difference of 500 v. or less is applied until the sample has all migrated into the gel and is a few mm below the paper. Then o-dichlorobenzene is run out of the apparatus until its interface with the terminating electrolyte is below the paper. The potential is then increased to give the desired running current and o-dichlorobenzene is run out of the apparatus during the separation at such a rate as to keep its interface with the terminating electrolyte to 10 mm above the top of the protein zone. At the same time enough terminating electrolyte is added to the top of the apparatus to keep the cathode immersed. When the protein is near the agarose the inner tube is pulled out of the wide tube and mounted in a lathe. The protein zone is then divided into rings with a scalpel while the tube js rotating using the appropriate markers as a guide. One cut is then made parallel to the axis and each ring is lifted away from the surface of the supporting tube by wrapping a narrow strip of filter paper round it to which the gel adheres so that it can be peeled from the glass and the adjacent rings with assistance from a scalpel blade if necessary.

The rings of gel are pressed through a stainless steel gauze of 50 mesh. The broken gel is packed into a 5 mm column and the protein eluted from it with a buffer solution. When anionic proteins are being separated the cathode is at the top and the anode at the bottom. For cationic proteins the polarity is reversed. EXAMPLE

Materials

1 The gel was crosslinked polyacrylamide, with N N Methy- lene-bisacrylamide as the crosslinking agent. In the usual nomenclature the total gel concentration T was 3•4% w.v. and theproportion of crosslinking agent C was 2.9% w.v. The gel was made up in Trishydroxymethylaminomethane-phosphate buffer solution pH 8.1 containing 0.1 M Trishydroxymethylamino- methane and 0.04 M phosphoric acid. The solution also contained

1 1 NNN N - Tetramethyl~1.2-diaminoethane 0.002 M as accelerator, the polymerization catalyst was ammonium persulphate 0.14% w.v.

O . WWII The Terminator Solution contained 0.23 6 aminohexanoic acid and 0.064 M Trishydroxymethylaminomethane.

The agarose containing lead electrolyte used to make electrical contact at the anode end of the gel column was a- 2% w..v. solution of agarose in 1.0 M Trishydroxymethylamino¬ methane and 0.4 M phosphoric acid.

Lead electrolyte was 0.05 M Trishydroxymethylamino¬ methane and 0.02 M phosphoric acid pH 8.1.

The inert coolant was o-dichlorobenzene and was prepared for use by drying it over anhydrous sodium sulphate and then passing it through a column of 120 mesh chromatographic o alumina grade H which had been activated at 350 for three hours.

The sample used was bovine serum which had been dialyzed against a buffer solution containing 0.0125 M Trishydroxy¬ methylaminomethane and 0.005 phosphoric acid, the euglobulin precipitate which formed was removed by centrifugation and the material was freeze dried. EXAMPLE Separation of bovine serum

200 mg of dialyzed bovine serum in 0.4 mis of leading electrolyte and containing 2 μX of a suitable fluorescent ampholyte marker solution which is anionic at about pH 8 was applied to the sample paper. Under these conditions, most of the proteins are anionic, and the top electrode was the cathode and the bottom the anode. The sample was electrophorezed from the paper on to the gel using a current of 5 mA which required a potential drop of 300 V and took 50 minutes. The current was then raised to 20 mA, the potential difference required being 2500 V. Current passed until the displacer had moved 100 mm below the paper. The length of the protein zone was then 50 mm and it was divided into rings using the markers as indicators.

The pattern of the cuts is shown in Figure 2. Rings 1, 2 and 3 were as small as could conveniently be cut. The cut between Rings 4 and 5 was arbitrary as there was only a continuous light background.. The division between 4- and __>' was made approximately midway between 3 and 6 in order to test the homogeneity of this region. The division between 6 and 7 was^made as closely as possible to the lower^* bright line. The- fractions were tested by analytical zone electrophoresis in polyacrylamide. Fraction 1 contains at least three zones one of them just detectable infrac ion 2 while the others were not. Fraction 2 contained one major zone which was not detectable in fraction 1 but was probably the same as a stronger zone in fraction 3 which contained essentially one zone. Fraction 4 has two major zones both of which appear in fraction 5 in very different proportions.

One of these zones in fraction 5 appears at the same position as that one of the zones in fraction _t but it is

• RE

OMPI WIPO probably a different substance since the intensity of this zone is less in fraction 4 than in either fraction 3 or fraction 5» Fractions 5 and 6 are very similar and one zone in both these fractions is very much stronger than the other and is probably bovine serum albumen. Fraction ^~~ contains three zones, two very weak, and they may all be the same as the zones in fraction j6. There is reason to think that though multiple zones in the zone analytical tubes are very different in mobility in a constricting gel, they are almost certainly nearly identical in mobility in the original displacement fractions. A comparison with a much longer time of separation and correspondingly greater movement would show the substances were still in the- process of separation.. The separation already shown compares favourably with a much longer run by zone electrophoresis. With zone- electrophoresis the smaller zones would be so dilute that they might well be missed by the conventional staining technique.

Referring now to Figure 4, the apparatus is constructed from glass. It consists of two vertical glass tubes 30, 31 one of which 30 is 33 - 35 mm in diameter and into this the column 2 to fee coated is dipped,, The second tube 31 s 1 cm in diameter and is connected to the wider tube 30 at the top bya 1 cm diameter connecting tube. The bottom of the narrower tube 3 s widened to about 40 mm diameter to become the body of a crude centrifugal pump _\_W which circulates the gel forming

«gT5lEA

OMPI ° liquid in the direction of the arrows.. The outlet from the- pump is connected to the bottom of the wider tube in such a ay that the flow from the pump is distributed uniformly. The whole apparatus is constructed as compactly as possible because it must in use be mounted inside a tall narrow beaker (not shown) through which hot water is circulated to maintain the temperature of the molten agarose. The beaker stands on a laboratory magnetic stirring apparatus, which is aligned to drive the magnetic stirring bar in the pump.. The column to be coated stands in the wider tube, and its bottom end is held concentric with the outer tube by means of a split centering ring 34. This ring is made of brass and fits closely on to the- column. The centring ring has three small projections which centre the column in the apparatus but do not disturb the even flow of liquid. The adhesion of the agarose layer to the glass- can be improved by previously coating the column with a thin layer of agarose and * then allowing this layer to dry out in the air. The column can then be coated with a further layer of agarose, which adheres well. The agarose column can be used as it is or it can be chemically cross-linked with NaOH-epichlorohydrin. It is then possible to allow the column to dry for ease of storage or transportation.

For the coating of a 20 mm diameter column, the coating apparatus is mounted as described and then heated to 70 C. 100 ml of 1% agarose solution at 90 - 100 C is poured in. The magnetic stirrer controller is- adjusted so that the height difference in the two columns is I..5 cm^*. The electrophoresis column, which contains no cooling water is placed in the apparatus using the centring, ring.. The upper part of the column is centred in the apparatus and the column is^; firmly clamped. Enough 1% agarose solution is poured into the apparatus so that the agarose- circuit is complete. Cooling water, at 10 C, is allowed to run through the apparatus for 1 minute. The column is removed and the thin agarose layer adhering to it is allowed to dry on the surface of the column. The column is emptied of cooling water and replaced in the apparatus. Cooling water^* is allowed, to run- through the* column* for 3 minutes-. The' column is then removed, when it will have a. thin layer of agarose adhering, to it.- The- agarose is then cross-linked. It is first dipped in 2..5^" N sodium hydroxide- for 4 hours and then into a. solution of 8% epichlorohydrin in toluene for at least 8 hours-, and finally it is washed in slowly running water for 1 day.

The column can then be soaked in a suitable leading electro¬ lyte and used for displacement electrophoresis. In preparing tubular gels by this method it has been found that the rate of deposition of the agarose decreased markedly with the rate of pumping. At a rate of flow equivalent to a head of 1.5 cm, with a molten agarose temperature of 70 C and a cooling water temperature of 10 C, a layer of agarose just under 1 mm thick was built up in two minutes. It was - 2.4 - slightly conical, the layer being a little thicker at the top of the tube than at the bottom. However the coating was perfectly even and uniform in cross-section. At a lower speed of circulation, corresponding to a head of 0.5 cm the rate of build-up of the- coating was- much faster and after 2 minutes the coating was rather uneven although its-average cross- section was uniform. At higher speeds of circulation," corresponding to a head of 2.5 - 3_*0 cm the rate of coating was very low. In 2 minutes the layer was barely perceptible and after 6 minutes it was seen to be asymmetrical.

^

O

Claims

CLAIMS 1. A device for use in displacement electrophoresis comprisinq a gel on which substances are to be separated having the form of a tube supported on one side only by means of a supporting member.

2. A device according to Claim 1 in which the supporting member is tubular.

3. A device according to Claim 1 or 2 in which the gel is supported on the external surface of the tubular supporting member.

4-. A device according to Claim 1, 2 or 3 which the gel material is agarose or polyacrylamide.

5- A device according to any of Claims 1 to 4 in which the thickness of the gel is up to 3 mm, especially 1 - 2 mm-.

6. A method for preparing a device according to any of the preceding claims in which the gel is formed on the supporting member in a mould.

7- A method for producing a device according to any of Claims 1 to 5 in which a stream of gel-forming material is caused to flow on a surface of the supporting member and the gel is formed by cooling.

8. A method according to Claim 7 in which the gel-forming material flows in the axial direction along the outer surface of a tubular supporting member which is cooled internally.

9« A method of performing displacement electrophoresis on a gel characterised in that the gel is in the form of a tube and

OMPI the displacement occurs along the length of the tube.

10. A method according to Claim 9 in which the displacement occurs in the downward direction.

11. A method according to Claim 9 or^' 10 in which the gel is in contact with an inert water-immiscible insulating liquid when not in contact with electrolyte.

12. A method according to Claim 9_> 10 or 11, in which the displacement is monitored by means of mobility marker sub¬ stances in the gel.

13. A method according to Claim 12 in which the marker substances are fluorescent ampholytes.

14. A method according to any of Claims 9 to 13 applied to the separation of proteins, peptides, or amino acids.

15. Displacement electrophoresis apparatus comprising a tubular gel coated on the external surface of an inner tube arranged for the passage of a coolant therethrough and an outer jacket for the flow of displacing electrolyte in contact with the gel in the axial direction.

16. Apparatus according to Claim 15, arranged to be mounted with the tubular axes vertical and for flow of displacing electrolyte in the downward direction.

17• Fluorescent ampholytes, being mobility marker substances for use in displacement electrophoresis by means of a device, method, or apparatus according to any of the preceding claims.