WO2002070111A2

WO2002070111A2 - Method for the manufacture of gradient gels and method for analysing biomolecules on ultra-long ipg gels

Info

Publication number: WO2002070111A2
Application number: PCT/DK2002/000149
Authority: WO
Inventors: Peter Mose Larsen; Stephen J. Fey
Original assignee: Syddansk Universitet
Priority date: 2001-03-07
Filing date: 2002-03-07
Publication date: 2002-09-12
Also published as: AU2002237206A1; WO2002071048B1; WO2002070111A3; WO2002070111A8; WO2002071048A8; WO2002071048A1

Abstract

The invention provides a method for the manufacture of gradient gels, in particular IPG gels, by use of a very dense gel precursor solution, i.e. casting solutions having a density of at least 1.15 kg/L, as well as the polyacrylamide gels as such. The invention also provides a electrophoresis chamber in which a drum carrying a pH-gradient strip is arranged. Furthermore, the invention provides a method of separating a protein mixture in which a moderate voltage gradient is applied.

Description

METHOD FOR THE MANUFACTURE OF GRADIENT GELS AND METHOD FOR ANALYSING BIOMOLECULES ON ULTRA-LONG IPG GELS

FIELD OF THE INVENTION

The present invention relates to a method for analysing virtually all proteins (including basic and very acidic proteins) on ultra-long IPG gels, and to a method for providing ultra- long IPG gels. This method provides very reproducible results, qualitatively as well as quantitatively. The extremely high resolution obtained obviates the need to run the second dimension commonly used in two-dimensional gel electrophoresis. The invention also provides a suitable electrophoresis chamber for handling of ultra-long gels or gel strips.

BACKGROUND OF THE INVENTION

Since 1975, (O'Farrell PH. High resolution two-dimensional electrophoresis of proteins. J Biol Chem. 1975 May 25;250(10):4007-21) complex mixtures of proteins have been separated by means of two-dimensional gel electrophoresis in which the physical separation of the proteins in the first dimension gel is based on a separation after the proteins' isoelectric point (proteins are chains of amino acids and as such can have many positive and negative charges and the pH at which they have an equal number of eachtype of charge of them is called their isoelectric point (pi)). This is referred to as an isoelectric focussing (IEF) of the proteins. However, a single IEF gel cannot resolve all of the proteins present in a single cell type since it is expected that there are in excess of 20,000different proteins in a typical cell. Therefore many researchers who want to study as many of the expressed proteins as possible (proteomics) have used a second "dimension" - a second gel wherein the proteins are separated at right angles to the first IEF gel. This is called two-dimensional gel electrophoresis (2DGE).

In the original technology, proteins were separated according to their pi-value by means of carrier ampholytes, i.e. zwitterionic chemicals which create the pH-gradient in which the proteins move and start to focus. When they reach the pH-value that equals their pl-value, the proteins reach a metastable state andthey are said to be focussed.

First, the gel mixture to be used for the first dimension gel is mixed. This mixture is composed of an acrylamide and a bisacrylamide solution, a detergent (e.g. Nonidet p40, Triton, CHAPS), urea, thiourea, water and carrier ampholytes. Then TEMED and ammonium persulphate are added where after the solution is left to polymerise in thin glass tubes (typically from 1 to 3 mm in diameter and 12 to 24 cm long. After the polymerisation is completed (usually overnight), the gels are ready for use. This means that the pH-gradient is created when voltage is applied to the gel. Therefore, voltage is applied to the gels (e.g. 250 - 1200 V for a period of time (which will vary depending on the ampholytes and other chemicals in the sample)) and when the right electrical parameters (the correct voltage and current) have been obtained, the gels are "prefocussed" and are said to be ready for use. Then, the sample, which is to be analysed, can be applied to the gel. Subsequently, the electrophoresis is started again and, this time, the proteins are separated as they pass down through the newly created pH-gradient until they reach their pl-value. This is how proteins in the pH range from approx. 3.5 to 7 are analysed.

However, this isoelectric focussing does not function for the basic proteins when using the conventional carrier ampholytes technology. There are two reasons for this phenomenon; firstly, carrier ampholytes are broken down in a basic environment when urea is present and this explains why a metastable pH-gradient is never established. Consequently, the basic proteins cannot focus at their pl-value. Secondly, and perhaps even more important, urea is broken down during focussing in basic environments, creating cyanate ions. These cyanate ions react with the side chains of the basic amino acids in the proteins to be separated. By slowly carbamylating these side chains, the isoelectric point for the protein is continually changing during the electrophoresis and this makes it difficult for the protein to focus. Thirdly, there is a greater flow of water due to electroendosmosis at basic pH- values and this also makes it more difficult to focus proteins.

Thus, an alternative technology called non-equilibrium pH-gradient electrophoresis

(NEPHGE) was published in 1977, (O'Farrell PZ, Goodman HM, O'Farrell PH. High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell. 1977 Dec; 12(4) : 1133-41), In this technique, the proteins are applied to polymerised gels and, subsequently, the electrophoresis started. Then the proteins started to migrate in the gel at the same time as the ampholytes start to form their pH-gradient. The consequence is that a metastable equilibrium is never established in these gels and that explains why the proteins will pass out of the gel if the electrophoresis is continued for too long a period of time. Thus, the NEPHGE-gel is not particularly stable, as the method is very sensitive to e.g. the amount of protein on the gel, the salt content in the sample, the temperature in the room etc. Consequently, few laboratories have analysed the basic proteins, as the NEPHGE-gels were not sufficiently reproducible to be used in many labs.

For many years, these and other serious problems were the driving force for attempts to find a replacement for the carrier ampholytes making it possible to establish a stable and reproducible pH-gradient. The breakthrough came with the Immobiline technology (IPG gels) (1) Bjellqvist B, Ek K, Righetti PG, Gianazza E, Gorg A, Westermeier R, Postel W. Isoelectric focusing in immobilized pH-gradients: principle, methodology and some applications. J Biochem Biophys Methods. 1982 Sep;6(4):317-39. 2) Westermeier R, Postel W, Weser J, Gorg A. High-resolution two-dimensional electrophoresis with isoelectric focusing in immobilized pH-gradients. J Biochem Biophys Methods. 1983 Dec;8(4):321-30. 3) Gorg A, Postel W, Gunther S. The current state of two-dimensional electrophoresis with immobilized pH-gradients. Electrophoresis. 1988 Sep;9(9):531-46. In this technology, 5 different low molecular weight functional groups have been substituted onto the acrylamide molecule thus obtaining acrylamide derivatives with different pKa-values (pH where half of the side chains are ionised and half are not). By casting a gradient gel by mixing two different acrylamide solutions containing different mixtures of the various substituted acrylamide species it is possible to produce a polyacrylamide gel where the pH-

10 gradient is covalently bound to the acrylamide matrix of the gel. In practice this is done by using different density solutions obtained by adding glycerol to the acrylamide solutions. By starting the casting process with solution A and then adding an increasing amount of solution B until at the end only solution B is used, then a gel is obtained (after the polymerisation is completed) which is characterised by the pKa-value of solution A at the

15 bottom and by the pKa-value of solution B on the top. Since the pH-gradient is covalently fixed to the matrix, the salt content of the sample and the amount of protein in the gel do not influence the pH-gradient and this makes the separation using these IPG gels very reproducible (Strahler JR., Hanash SM, Somerlot L, Bjellqvist B, Gorg A. Effect of salt on the performance of immobilized pH-gradient isoelectric focusing gels. Electrophoresis.

20 1988 Feb;9(2):74-80).

This technology has worked quite well, but there now remain at least two serious problems with this technology.

The first serious problem with the existing technology is that the number of different proteins that are present in the cell is expected to lie somewhere in the range from 20,000

25 to 50,000. This problem was not really appreciated when the IPG's were first introduced because it was only possible to detect 2-3,000 protein species (and even worse, it was very difficult to identify most of these). However, the number of proteins that can be revealed by this technology has been steadily increasing (due to small accumulative methodological improvements) so that now it is actually possible to detect up to 6,000

30 proteins in standard gels (covering the pH range from 4 to 7 and 6 to 11 (corresponding to the traditional IEF and NEPHGE gels)).l) Fey SJ, Nawrocki A, Larsen MR, Gorg A, Roepstorff P, Skews GN, Williams R, Mose Larsen P. Proteome analysis of Saccharomyces cerevisiae: a methodological outline. Electrophoresis. 1997 Aug; 18(8): 1361-72. 2) Nawrocki A, Larsen MR,Podtelejnikov AV, Jensen ON, Mann M, Roepstorff P, Gorg A, Fey

35 SJ, Mose Larsen P. Correlation of acidic and basic carrier ampholyte and immobilized pH- gradient two-dimensional gel electrophoresis patterns based on mass spectrometric protein identification. Electrophoresis. 1998 May; 19(6): 1024-35, and 6) Fey SJ and MoseLarsen P. 2D or not 2D. Two-dimensional gel electrophoresis. Curr Opin Chem Biol. 2001 Feb;5(l):26-33.). However, with the introduction of sensitive mass spectrophotometers, it has now been shown that the majority of the protein spots detected actually contain at least two proteins, and sometimes three, four or more - confirming that the number of protein species present is significantly higher than the number of spots detected.

One of the advantages of the IPG technology is that only a very few chemicals are needed to make the pH-gradient in the gels, and therefore it is straightforward to cast gels which cover a narrower pH range. These gels could for example cover only one or even only a half or a quarter pH unit and several manufacturers have recently put onto the market gels which cover narrow pH ranges (e.g. 3.5 - 4.5, 4.0 - 5.0, 4.5 - 5.5, 5.0 - 6.0, 5.5 - 6.7). Using the best gel recipes currently available these gels are capable of resolving a total of over 15,000 protein spots from a single cell or tissue lysate (op. sit Fey SJ and Mose Larsen P. 2D or not 2D. Two-dimensional gel electrophoresis. Curr Opin Chem Biol. 2001 Feb;5(l):26-33). Mass spectrometry suggests at current limits of detection that there are multiple protein species in only about 5% of these spots).

Thus, the progress in this area has turned out to be a major problem - because it is not good enough to quantitate the protein spots because it is impossible to know the relative contributions of each of the proteins in the spot in the standard gels which cover several pH units. Quantitation can be carried out for the narrow range gels because of their increased resolving power, but it is necessary to run five or more gels and thus have a correspondingly increased amount of sample. For many situations, for example involving human biopsy samples, it simply may not be possible to collect this amount of tissue. Even when it is possible, the larger the sample taken is, the more likely that it represents a mixture of cell types (e.g. muscle cells, blood vessels and cells, nerve cells, connective tissue etc.) and this mixture will mask the important specific changes in the tissue which have been caused by disease or treatment.

The second serious problem relates to proteins with pi-values outside the pH range from 4 to 7.

Once again the analysis of basic proteins has proved to be difficult - this time actually more difficult in the IPG system than in the NEPHGE system. The source of the problem lies in the fact that the basic acrylamide derivatives polymerise very rapidly and this is one of the reasons why it has been difficult to obtain uniform and reproducible pH-gradients. Today, it is possible to buy IPG gels in the basic range: e.g. pH 6-9, 6-11 (both from Amersham Bioscience) and 7-10 (from BioRad) and, naturally, it is possible to synthesise gels covering any desired pH range. There have also been problems with the focussing process (for example, the most basic proteins have run out of the gel before other proteins are focussed, while other proteins give rise to streaks on the gel (i.e. never focus to sharp spots on the gel).

In addition, no narrow range basic IPG gel strips are commercially available and only one gradient (covering the pH range from 10 to 12) has been published together with a gel image to demonstrate that it works (Gorg A. IPG-Dalt of very alkaline proteins. Methods Mol Biol. 1999;112: 197-209).

IPG gels covering narrow pH ranges which extend more acidic than pH 4 or more basic than pH 9 suffer from electroendosmosis (the pumping action of immobilised charges on free flowing water molecules). This leads to the defocusing of the protein spots as the proteins get carried with the water.

Thus, these have been, and still are, the biggest problems in connection with protein separation in the proteome analysis. Several attempts have been made to solve this problem and the presently most favoured method is to increase the voltage applied to the gel (to 8,000V in the IPGphor or even 10,000V in the Protean IEF Cell) in order to obtain a steeper electrical gradient which theoretically should provide a sharper focusing effect. This has helped somewhat in the acidic range and partly worked in the neutral region of the basic pH-gradients. However, there is no improvement whatsoever in the range above pH 8.5. Furthermore, this very strong voltage applied to the gels has caused a far stronger electroendosmosis, causing water to migrate through the gel instead of the proteins and this has had a negative effect on the focussing (the proteins are in practice swimming against a current). Another negative effect has been that water is accumulates at the anode which clearly implies that the proteins do not focus in this area, as they will float on top of the gel in the accumulated water. This effect can be partially counteracted by including some isopropanol (typically 5% v/v) in the gel (Gorg A. IPG-Dalt of very alkaline proteins. Methods Mol Biol. 1999; 112: 197-209).

Other types of reducing substances have been added, e.g. dithiothreitol (DTT) 2- mercaptoethanol, and phosphines, but with negative result. Another attempt to solve the problem with the basic proteins has been to make an ultra wide gradient (e.g. covering the pH range from 4 to 12) so that the electroendosmotic flow is opposed by a flow from the opposite end. This works well but suffers from the problem that the spot density becomes so high that most spots represent multiple protein species, giving serious problems for quantitation [Gorg A, Obermaier C, Boguth G, Weiss W. Recent developments in two- dimensional gel electrophoresis with immobilized pH-gradients: wide pH-gradients up to pH 12, longer separation distances and simplified procedures. Electrophoresis. 1999;20(4- 5) :712-7. BREIF DESCRIPTION OF THE INVENTION

The present invention provides a method for the manufacture of gradient gels, said method comprising the steps of:

(i) arranging a casting mould having two parallel, vertical inner surfaces which define walls 5 of an enclosure;

(ii) providing at least two gel precursor solutions including a first gel precursor solution and a second gel precursor solution, said first gel precursor solution having a first density and said second gel precursor solution comprising a density increasing agent and having a second density,

10 (iii) casting the gradient gel by providing the at least two gel precursor solutions to the casting mould to a height of h meters so as to establish the gradient gel, the ratio between the second density and the first density being at least 1.15: 1, said at least two gel precursor solutions being provided to the casting mould so that the volume corresponding to the lowermost 10% of the gel precursor solutions predominantly, such as by at least

15 80%, consist of the second gel precursor solution, and so that the volume corresponding to the uppermost 10% of the gel precursor solutions predominantly, such as by at least 80%, consist of the first gel precursor solution.

The present invention also provides long gradient gels prepared according to the above method.

20 These gels will also be well suited for the separation of other biomolecules (e.g. DNA, RNA, oligonucleotides, peptides, analogues of them all and various types of sugar or sugar containing molecules). These gels also can be used in combination with a wide variety of biomolecule dissolving solution.

Furthermore the resolution of these gels can be substantially increased by making the gel 25 thinner than normal e.g. 1 mm, 0.5 mm or 0.1 mm or even less.

The present invention further provides an electrophoresis chamber for pH-gradient strips, said chamber comprising a housing defining a chamber enclosure, a drum arranged within the chamber enclosure, a lid or door as a part of the housing for accessing the chamber enclosure, means for rotating the drum around its axis of rotation, at least one pH-

30 gradient strip(s) having a length to width ratio of at least 10: 1, said drum having arranged the at least one pH-gradient strip(s) on its surface of rotation, said strip(s) being arranged so that the longitudinal direction thereof is substantially perpendicular to, or at an angle of at the most 20° to, the axis of rotation of the drum, each end of the strip(s) being held by one or more clamps, said clamps having conductive parts for electrically connecting each 5 end of the strip(s) to an external high voltage power supply so as to apply a voltage over the opposite ends of the strip(s). Finally, the present invention also provides a method of separating a protein mixture by means of a IPG gel, said method comprising the steps of

(a) providing the IPG gel having a length of λ meters; (b) providing the protein mixture to be separated;

(c) applying the protein mixture to the gel;

(d) positioning electrodes at or near each of the ends of the gel;

(e) applying a voltage over the electrodes, said voltage representing the following profile:

(i) an initial voltage of 0-100 V (ii) a first voltage increase gradient of in average 100-800 V/h for 1-10 hours up to an intermediate voltage of in the range of 300-2000 V,

(iii) a second voltage increase gradient of in average 500-2,000 V/h for (3-15 m^"1 x λ) hours from the intermediate voltage up to a final voltage of (3,000-40,000 nrf¹ x λ) V, (iv) the final voltage for (15-300 m^"1 x λ) hours.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 illustrates a simple apparatus for casting linear gradient gels.

Figure 2 illustrates the electrical profile of an IPG gel during the separation of proteins. The voltage profile (V) is programmed into the power-pack and the current usage (mAmp) is not regulated - but is measured to follow the progress of the electrophoresis with respect to time (hours).

Figure 3 illustrates the drum of an electrophoresis chamber according to the invention.

Figure 4 illustrates one possible configuration of spacers designed so that the high density of the solution needed to cast the long gels cannot displace the spacer and make the gel casting chamber leak. Normally, the vertical spacers usually abut onto the horizontal spacer and there is no top spacer.

DETAILED DESCRIPTION OF THE INVENTION

Manufacture of gradient gels, such as immobilised pH-gradient gels ("IPG gels")

When used within the following aspect (manufacture of gradient gels), the term "gel" is intended to cover any gel or gel-type separation matrices. True gels such as polyacrylamide gels can be considered as one large molecule due a suitable degree of cross-linking, whereas gel-type "gels" can be considered an assembly of high molecular weight molecules, such as polysaccharides, e.g. agar, etc. The "gels" can be for example of the type corresponding to either of the dimensions described in the introduction for 2D electrophoresis - i.e. capable of separating biomolecules according to either their isoelectric point, such as in IPG gels, or according to their molecular weight. In the following the example taken is the former, namely separation according to their isoelectric point (specifically IPG gels), whereas the latter possibility is equally valid as will be understood by the person skilled in the art.

This being said, a particularly relevant embodiment of this aspect of the invention relates to the manufacture of IPG gels.

If one assumes that there are 20,000 different protein species in a particular cell extract, then a 1 meter long gel would have 200 proteins per cm or 20 per mm if the proteins were evenly distributed.

Assuming that such long gels could be made, the subsequent protein detection could be carried out by autoradiography which has a resolution of about 0.01 mm. This could then be followed by infusing trypsin or some other specific cleavage enzyme used in mass spectrometry plus a matrix solution (to aid the peptide desorption e.g. cyano-4- hydroxycinnamic acid) into the gel and then performing mass spectrometry to identify the proteins present. One disadvantage of this technique would be that the proteolytic digestion of the protein bands prior to mass spectrometry will undoubtedly cause diffusion and thus a broadening of the band.

However, currently, mass spectrometers are capable of resolving mixtures of up to 10 proteins at once (and this capability is increasing in parallel with their sensitivity). The diameter of the laser beam footprint on the target used to excite peptides for mass spectrometry can be less than 0.1 mm.

Thus, although it is not yet possible to resolve all of the proteins cleanly from highly complex mixtures, the technologies are close to reaching this goal. Improvements, for example in being able to cast a gel that is 5 meter long covering the relevant pH range (e.g. 2-12), would dramatically change the situation. Additional alternatives would be to run five 1 meter long gels, or to run gels that cover sections of the pH-gradient, e.g., as single (partially overlapping) pH units.

The key advantages of one aspect of the invention, namely the method for the manufacture of a gradient gel (such as an "IPG gel"), is that it now becomes possible to greatly increase the length and thus the resolution of the gradient gels to the point where it is not necessary to carry out the second dimension of the electrophoresis. This will have several significant effects: 1. Separation of the proteins in one dimension is very compatible with detection, quantitation and identification of the proteins in a highly automated format.

2. Separation of the proteins in long one dimension gels will increase the amount of protein that can be applied to the gel and thus facilitate detection of low abundance

5 proteins and post-translational modifications.

3. Elimination of the second dimension greatly simplifies the image analysis (one of the most time-consuming steps in proteome analysis) which currently has to be carried out on an essentially manual basis through lack of suitable computer programs. Although many programs exist, each image requires at least 8 hrs of manual editing to

10 correct spot boundaries, to add spots not detected, and to correct mismatches.

4. The use of only one dimension will increase the options available for using different additives in the gel. At present they have to be compatible with both gel dimensions or be changed by washing in between, and this is a difficult step because the proteins should remain soluble at all times.

15 This being said, the method according to this aspect of the invention comprises the steps of:

(i) arranging a casting mould having two parallel, vertical inner surfaces which define walls of an enclosure; (ii) providing at least two gel precursor solutions including a first gel precursor solution and

20 a second gel precursor solution, said first gel precursor solution having a first density and said second gel precursor solution comprising a density increasing agent and having a second density,

(iii) casting the gradient gel by providing the at least two gel precursor solutions to the casting mould to a height of h meters so as to establish the gradient gel, the ratio between

25 the second density and the first density being at least 1.15: 1, said at least two gel precursor solutions being provided to the casting mould so that the volume corresponding to the lowermost 10% of the gel precursor solutions predominantly, such as by at least 80%, consist of the second gel precursor solution, and so that the volume corresponding to the uppermost 10% of the gel precursor solutions predominantly, such as by at least

30 80%, consist of the first gel precursor solution.

In an initial step, a casting mould having two substantially parallel, vertical inner surfaces, which define walls of an enclosure, is arranged. The gel is typically cast between fairly thick glass plates that define the inner surfaces. Other materials may in principle also be used, e.g. steel, aluminium, titanium, etc., but glass will of course allow for visual 35 inspection of the gel during manufacture. Such glass plates (or other plates) are arranged in a vertical manner so that they define walls of an enclosure. The enclosure is further defined by "spacers" which keeps a predetermined distance between the plates. The spacers are typically made of one piece or they may be constructed in such a way that they interlock so that they cannot be pushed out of place by the dense solutions. See Figure 4 for a possible spacer design. The (glass) plates are typically held together with the spacers by means of strong clamps so that there is no liquid leakage from the enclosure.

The thickness of conventional glass plates is 3-4 mm. When casting long gels, e.g. gels of at least 0.5 m, the glass plates will normally need to be even thicker, such as at least 7 mm, because the weight of the gel precursor solution is sufficient to bend the glass due to the height of the enclosure. If no precautions are taken, a gel that is thicker at the bottom and in the middle than at the top and at the edges will be formed and this will lead to irreproducible electrophoresis on the gel strips in that the proteins will run to various distances depending upon how thick the gel is.

The enclosure defined by the (glass) plates and the spacers will determine the shape of the gel. Typical dimensions are, height between 0.5 m and 5 m, preferably between 1.0 m to 2.0 m, width between 10 cm and 100 cm, preferably between 20 cm and 40 cm, thickness (between plates) about between 0.01 and 3 mm, preferably between 0.05 and 1 mm.

The height of the gel casting mould is normally about, but of course at least, the desirable height of the long gradient gel.

In a preferred embodiment, one of the parallel inner surfaces is adapted to form covalent bonds with the gel under the polymerisation conditions. Thus, the gel is cast on surface to which the acrylamide will bind (for example "Gel bond®" from FMC Products). This allows for thorough washing prior to use (see below). The purpose of this particular surface (e.g. gel bond®) is to reduce the osmotic distortion of the gel during the process of washing away the density increasing agent(s) and to simplify the problems of handling (including the drying) of the large gels produces by the instant method. In the present invention, either one of the glass plates, or, preferably, a separate film mounted on one or the glass plates may allow for easier handling of the gel. It should be understood that the separate film should be detachable from the glass plate after casting of the gel. A separate film is advantageous in that it also allows for cutting of the gel into thin strips.

In another initial step of the method, at least two gel precursor solutions including a first gel precursor solution and a second gel precursor solution are provided. Typically only two gel precursor solutions are utilised, but one skilled in the art will be familiar with the devices capable of delivering solutions from a large number (e.g. 6-8) different precursor solutions, see, e.g., Altland K, Altland A. Pouring wide-range immobilized pH-gradient gels with a window of extremely flattened slope. Electrophoresis. 1990 Apr; ll(4):337-42; Altland K. IPGMAKER: a program for IBM-compatible personal computers to create and test recipes for immobilized pH-gradients. Electrophoresis. 1990 Feb; 11(2): 140-7; Altland K, Altland A. Pouring reproducible gradients in gels under computer control: new devices for simultaneous delivery of two independent gradients, for more flexible slope and pH range of immobilized pH-gradients. Clin Chem. 1984 Dec;30(12 Pt 1):2098-103.

By definition, the first gel precursor solution has a first density and the second gel precursor solution has a second density. It should be understood, in particular in view of the following, that the second density is larger that the first density.

The second gel precursor solution comprises a density increasing agent in order to provide a relatively higher density of the second gel precursor solution compared with the first gel precursor solution. This being said, the density of the second precursor solution is preferably at least 1.15 kg/L, such as at least 1.20 kg/L, e.g. at least 1.25 kg/L, or at least 1.30 kg/L.

The first gel precursor solution typically has a density of at the most 1.05 kg/L, such as at the most 1.03 kg/L, e.g. at the most 1.02 kg/L, or at the most 1.01 kg/L.

In one embodiment, wherein the first gel precursor solution comprises acrylamide, bisacrylamide, and a catalyst and has a density of at the most 1.03 kg/L, and the second gel precursor solution comprises acrylamide, bisacrylamide, caesium chloride, and TEMED and has a density of at least 1.15 kg/L.

Suitable examples of density increasing agents are any salt or compound that is highly soluble and can be used to produce a dense solution according to the requirements stipulated herein. Some examples of density increasing agents are those selected from sucrose, glycerol, Dextran T500®, Ficol®, Percoll® (all three from Amersham Biosciences) , urea, guanidinium chloride, soluble salts of lead (e.g. lead sulphate or nitrate), mercury (e.g mercurous nitrate), caesium chloride (CsCI), caesium sulphate (Cs₂S0₄), caesium formate, osmium tetroxide, and sodium bromide (NaBr), in particular dextran T500, Percoll, caesium chloride (CsCI), caesium sulphate (Cs₂S0₄), caesium formate, and sodium bromide (NaBr). Preferred density increasing agents are caesium chloride (CsCI), caesium sulphate (Cs₂S0₄), caesium formate, and sodium bromide (NaBr).

The density of a solution of CsCI can be conveniently calculated using the formula, in which if p is the desired density of the solution, then the number of grams of CsCI that have to be added per millilitre solution equals (4ρ -4)/(4 - p). The use of density increasing agents like CsCI at the high concentrations proposed here would normally interfere with or prevent electrophoresis taking place and therefore it is necessary to wash them out of the gel before electrophoresis.

Each of the gel precursor solutions comprises a gel-forming agent, a cross-linking agent, and a catalyst preferably in the same concentration in first and second precursor solution. Gradients in these gel-forming agents are well known in the art and are an additional option in these ultra-long gels.

The gel precursor solutions also comprises pH-functional polymerisable compounds, e.g. low molecular weight functional groups substituted onto the acrylamide molecule, thus obtaining acrylamide derivatives with different pKa-values.

In the most typical embodiment, the gel-forming agent is acrylamide and the cross-linking agent is bisacrylamide. The catalyst may be any applicable in view of the gel-forming agent and cross-linking agent selected, and may in the case of acrylamide and bisacrylamide be TEMED.

The gel precursor solutions may be selected in such a manner that the difference between the pH of the first gel precursor solution and the pH of the second gel precursor solution is about 1 pH unit, such as about 3 pH units or even about 6 or about 10 pH units. Examples are the gradient ranges covered by the first gel precursor solution and the second gel precursor solution are 3-12, 3-10, 4-7, 6-9, 6-11, 3.5-4.5, 4.0-5.0, 4.5-5.5, 5.0-6.0, 5.5- 6.7, 3-6, 5-8, and 7-10.

An example of two gel precursor solutions is given in table 1 below. This recipe is intended for a 1 meter long gel, which covers the pH range from pH 4.0 to 7.0.

Table 1. Recipies for the light and heavy solutions for casting a long gradient gel

a) Acrylamide solution is 28.8% acrylamide : 1.2% bisacrylamide (weight/volume) b) TEMED is N,N, ,N'-Tetra-methyl-ethylenediamine. The solutions are prepared and then degassed before the ammonium persulphate (APS) and TEMED are added. This is done to remove any effects of the atmospheric oxygen on the polymerisation process. Ammonium persulphate and TEMED are added immediately before casting the gels. (See also Fey et al., Electrophoresis, 1997, vol. 18, 1361 - 1372 for further details.)

In a subsequent step, the gradient gel is cast by providing the at least two gel precursor solutions to the casting mould in such a manner that the volume corresponding to the lowermost 10% of the gel precursor solutions predominantly, such as by at least 80%, e.g. by at least 90%, consist of the second gel precursor solution (the most dense solution), and so that the volume corresponding to the uppermost 10% of the gel precursor solutions predominantly, such as by at least 80%, e.g. by at least 90%, consist of the first gel precursor solution (the least dense solution).

Many procedures for the casting of one-dimensional gels have been published and practised and thus these will be familiar to one skilled in the art. Similarly, many methods exist for the casting of a gradient, whether it be a pH-gradient or a gradient in the percentage of the acrylamide support matrix (which will affect the resolution of proteins according to their molecular weight). An illustrative example of a set-up for providing the gel precursor solutions to the mould is illustrated in Figure 1. Figure 1 illustrates a gel casting chamber (G) connected to a 2 chambered device into which two acrylamide solutions can be filled. Initially taps T_x and T₂ are closed. Heavy and light solutions of acrylamide plus gradient monomers of a particular concentration and mixture are filled into chambers H and L respectively. A gradient in density and gradient monomer can be cast as described below. Such a device for casting gradient gels is also described in "Gel electrophoresis of Proteins" 3^rd edition Edited by B.D. Hames Oxford University Press, chapter 5 "Conventional isoelectric focusing in gel slabs, in capillaries and immobilised pH- gradients," by Righetti PG, Bossi A and Gelfi C. ISBN 0-19-963640-0 and Gianazza E, Dossi G, Celentano F, Righetti PG. Isoelectric focusing in immobilized pH gradients: generation and optimization of wide pH intervals with two-chamber mixers. J Biochem Biophys Methods. 1983 Sep;8(2): 109-33, and many other configurations exist.

The gel precursors solutions are provided to a height of h meters so as to establish the gradient gel of a similar length in the direction of the pH-gradient. In order to achieve the objects of the invention, the ratio between the second density and the first density is at least 1.15: 1, such as at least 1.20: 1, e.g. at least 1.20: 1, in particular 1.30: 1. As a preferred additional requirement, the density of the second precursor solution is preferably at least 1.15 kg/L, such as at least 1.20 kg/L, e.g. at least 1.25 kg/L, or at least 1.30 kg/L.

The polymerisation of the gel-forming agent and the cross-linking agent will in principle start at a slow rate as soon as a catalyst is added, i.e. even when preparing the gel precursor solutions. Thus, the catalyst should not be added until shortly before the solutions are provided to the mould. Alternatively, a photo-catalyst or thermo-catalyst may be used. One example of a frequently used catalyst is TEMED.

Another feature to retard the polymerisation process is to cool the gel solutions down to 4°C or lower (i.e. on melting ice) and at this temperature the reaction proceeds very slowly.

The rate of polymerisation should be slow; i.e. ideally the gel should appear polymerised only after 1-2 hrs. This is important because polymerisation is an exothermic and an autocatalytic process. Thus once polymerisation starts in one region; it proceeds to "grow" out from this initiation site rather like a crystal. This can cause local heating and this in turn can disturb the gradient. Actual polymerisation continues for several hours and it is advisable for the sake of reproducibility to leave the gels to polymerise in a uniform temperature environment overnight.

A strong neutral buffer can also be included into the solutions so as to control the pH of the gel so that gels polymerised with basic acrylamides do not polymerise too fast (which would otherwise cause local heating and would disturb the pH-gradient). The washing step (see below) will then cause that this buffer is effectively removed from the gel prior to its use. This buffer can be for example Tris-HCI, phosphate or any similar buffer in the pH range from pH 4 to 7.

In a final step, the gradient gel is washed so as to reduce the concentration of the density increasing agents by at least 90%, or under even better circumstances by more than 95%, such as to virtually eliminate any density increasing agent. Also, the content of any buffers may be reduced to the same extent.

Thus, after the gel is fully polymerised (typically at least 12 hrs after casting), it is removed from the glass plates, and washed extensively to remove the unwanted density increasing agents and any buffers. Advantageously, the gel is attached to a polymer film "gel bond" so as to ease handling. This washing step can also be used to introduce desired other compounds into the gel, for example the detergent SDS or urea (to improve protein solubility) or (low concentrations, e.g. 2% of) glycerol (for storage purposes). If the gel has been covalently attached to a glass plate then, once the washing is completed the gel is removed from the glass plate.

As an example, the gels may be washed three times for 30 minutes each in distilled water on a horizontal shaker (with forwards and backwards motion for uniformity) or inside a rotating cylinder. The gel can then be washed twice in a desired buffer solution (for example 2% glycerol if the gel has to be dried at this point) to introduce the desired components.

In the case where, for some reason, "gel bond" or the like cannot be used, the gel may be washed through a successive series of buffers in which the salt concentration is stepwise reduced (e.g. 1 M, 0.5 M, 0.25 M, 0.1 M, 0.02 M, 0 M NaCI). Another option is to use a more rigid acrylamide derivative to in part substitute for the acrylamide in the gel (e.g. Duracryl® supplied by Millipore).

After washing, the gel may be dried. This normally takes about 24 hrs depending upon the ventilation, room temperature, percentage of glycerol introduced, etc. Care should be taken to keep the gel completely free of dust as this will otherwise cause problems when proteins are detected for example with silver staining, fluorescence or by mass spectrometry. (Most dust contains human keratin in amounts that will mask out the detection of proteins in the gel by mass spectrometry).

Subsequent to drying of the gel, the gel may be covered with a sheet of protective plastic (as is well known to one skilled in the art) and cut into strips, e.g. 3 mm wide strips, so that the full range of the pH-gradient is represented on the resulting strip. Industrial cutting devices (e.g. rotating wheel cutters) are normally needed to cut these long gels. This being said, the width of the gel (strip) can be even larger so that preparative gels can be run with many times the load of a conventional gel. For example if the normal gel is 3 mm wide and one can load 600 μg of protein, then if the gel were 20 cm wide (which is a typical width of the gel when it is first cast before it is cut into analytical strips), it would be possible to load 180 mg of protein.

When dried, the gel (strips) can be stored for long periods of time and need to be rehydrated before use as will be known for the skilled person. As an example, a 100 x 0.3 x 0.05 cm dried gel strip can be rehydrated with 1.25 ml of lysis buffer and left for at least 6 hrs in a constant temperature environment and protected from dust before it is suitable for electrophoresis.

The rehydrated gel strips may be utilised as conventionally described, or may be utilised in connection with the constant temperature electrophoresis chamber as described below.

The above-mentioned aspect of the present invention also relates to a gradient gel, such as an immobilised pH-gradient gel ("IPG" gel), obtainable by the method described above. Such a gel may cover the pH range of 4-9, in particular the pH range of 2-12. Electroendosmosis effects can be avoided if gradients cover wide pH ranges.

Alternatively, the gel covers a pH ranges of about 3 pH units, about 1 pH unit, or about 0.1 pH unit.

It should be understood that the invention provided herewith renders it possible to design long IPG gels covering quite a broad range of pH-values, e.g. pH 4-9 or even pH 2-12. Such gels will be very useful for the analysis of complex protein mixtures in which proteins, oligonucleotides or other biomolecules with quite different pl-values are present. Also, the invention is applicable for the manufacture of long IPG gels covering a narrow range of pH-values, e.g. around 1 pH unit, around 0.5 pH unit or around 0.25 pH unit. Such gels are very useful for refined separation of a mixture of proteins, oligonucleotides or other biomolecules having almost identical pi-values.

The gradient gel may as such have a length, h, in the longitudinal direction (the direction of the gradient - corresponding to the vertical height of the mould) of at least 0.5 m, such as at least 1.0 m, in particular at least 2.0 m.

Thus, the length of the gel can be increased to at least 1 meter, obviating the need to run the gel in the second dimension. A further increase in effective length is possible by running multiple gels each covering narrow (overlapping) pH ranges. When utilising longer gels, handling may be facilitated by coiling the gel around the drum, e.g. as described below for the electrophoresis chamber. As described above, the gel may be cut in narrow strips. Thus, in one embodiment, the gradient gel will have a length to width ratio of at least 10: 1, such as at least 25: 1, e.g. at least 50: 1, or even at least 100: 1. Such strips are particularly useful in connection with the electrophoresis chamber described below.

Although the above has been described for the preparation of fairly long gels, it is believed that the method as such will also be advantageous for the preparation of relatively shorter gels.

The gels can be used as described below for the electrophoresis chamber. If the gels are run on a flat bed device instead of the drum, then the samples can be applied using any of the standard devices, e.g. direct application of the sample to the gel surface, paper wicks, cups or indentations in the gel itself.

Electrophoresis chamber

In another aspect, the present invention relates to an electrophoresis chamber that is particularly useful for use in connection with the long gels described above. The chamber will reduce the operational variation and thus increase the value of the data obtained in that the (gel) strips are processed in a closed environment in order to protect the operator from the high DC voltages used, and in order to protect the strips while they are processed from atmospheric C0₂, 0₂ and dust.

The gel is curved around the drum to reduce the size of the instrument facilitating control and regulation of the temperature and humidity under which the strips are run.

Temperature control may be effected by recirculation of a coolant through the centre of the drum. Humidity within the chamber can be regulated by a water trough located within the chamber, e.g. under the drum.

Thus, the present invention provides an electrophoresis chamber for pH-gradient strips, said chamber comprising a housing defining a chamber enclosure, a drum arranged within the chamber enclosure, a lid or door as a part of the housing for accessing the chamber enclosure, means for rotating the drum around its axis of rotation, at least one pH- gradient strip(s) having a length to width ratio of at least 10: 1, such as at least 25: 1, e.g. at least 50: 1, or even at least 100: 1, said drum having arranged the at least one pH- gradient strip(s) on its surface of rotation, said strip(s) being arranged so that the longitudinal direction thereof, i.e. the direction of the gradient, is substantially perpendicular to, or at an angle of at the most 20° to, the axis of rotation of the drum, each end of the strip(s) being held by one or more clamps, said clamps having conductive parts for electrically connecting each end of the strip(s) to an external high voltage power supply so as to apply a voltage over the opposite ends of the strip(s).

The width of the drum can be chosen so that many analytical strips can be run in parallel, or a very wide preparative gel can be run.

An embodiment of the drum of a chamber is illustrated in Figure 3. In this embodiment, three pH-gradient strips are arranged on the surface of rotation of the drum, and the strips are arranged substantially parallel to each other. In this embodiment, i.e. embodiment where two or more strips are arranged perpendicular to the axis of rotation of the drum, the clamps for holding the ends of the strips may be in the form of a clamp assembly. More specifically, the electrophoresis chamber illustrated in Figure 3 consists of a drum through which a cooling liquid can flow and which is coated with an electrically nonconducting surface. For gels of approx. 1 m, the diameter of the drum is about 40 cm and the height depends only upon how many gels need to be run. A cylinder height (width of surface of revolution) of about 30 cm is suitable to run many gels (more than 40). The two clamps (electrodes) marked (+) and (-) also act to hold the strips in place and are thus mounted on springs (or other tension device). If desired it is possible to insert an electrode wick between the electrode and the gel (this can first be attached to the electrode while it is in the open position (i.e. away from the drum). In the embodiment illustrated, 3 gels are held tightly against the drum by one pair of electrodes. In another arrangement the strips could be held down by individual electrodes so that different types of strip (requiring different electrical profiles), or different types of sample (containing widely different amounts of salt or protein concentration) could be run in parallel optionally using different gradients for each strip.

The sample can be either absorbed into the strip during rehydration of the gel or it can be applied with a sample cup applicator afterwards. A suitable sample applicator is a small device that curves to fit the drum and has either wells or a porous material to hold the sample. Once the strips are mounted on the drum, the sample can be applied by moving the sample applicator against the drum in such a way that the wells or porous material (e.g. filter paper) come in contact with the strips. The sample can then be pipetted into the cups or onto the porous material. Alternatively the sample can have been applied to the applicator previously to being placed in contact with the strips. The materials that the applicator is made of must be totally inert to the sample - i.e. they should not react and the proteins in the sample must not bind to the applicator. The applicator is normally electrically insulated. The housing of the chamber is not shown.

The above-mentioned embodiment is useful when the length of the strips is slightly shorter than the length of the circumference of the drum. Even longer strips, i.e. strips having a length of more than the length of the circumference, e.g. about 2, 3 or even 5 or more times the length of the circumference of the drum, may also be processed in a chamber by spiralling the strip around the drum.

Thus, alternatively (not illustrated), one or more pH-gradient strip(s) is/are spirally arranged around the surface of rotation of the drum at an angle of in the range of 2° to 15° to the axis of rotation of the drum. If more than one strip is arranged, the strips are arranged substantially parallel to each other. In this case the conductive parts of the clamps should be designed so that they only came into contact with the ends of the strips and that there was no possibility for short-circuit from one part of the strip to another. In this instance two separate clamps are normally necessary.

Means for electrical connection to an external high voltage power supply will be apparent for the person skilled in the art.

Although, the drum in principle may carry any type of matrix for the fractionation of proteins and other biomolecules, e.g. gradient strips of any kind, it is particularly relevant to have arranged thereon a pH-gradient gel strip (IPG gel strip) as defined herein.

The chamber should also have means for regulating the humidity in the chamber enclosure so as to reduce or eliminate evaporation from the (gel) strip(s). This may simply be achieved by having a trough filled with water in the bottom of the chamber. Optionally, the evaporation of water from this trough may be electrically assisted (e.g. heated). Regulation of the humidity is particularly relevant where electrophoresis is conducted over a long period of time, such as 10 hours or more.

C0₂ in the atmosphere may cause the pH to drift lowering the quality of the result of the electrophoresis, in particular for pH regions above pH 7.0.

Thus, in order to establish an atmosphere, which will not have detrimental effect on the electrophoresis process, it is desirable to include within the chamber an open container comprising a strong base, such as an aqueous base, in order to neutralise any atmosphric C0₂ present or formed in the process. An aqueous solution of sodium hydroxide will be suitable. Alternatively, the sodium hydroxide can be added to the water trough. As a further alternative, or as a supplement, nitrogen can be introduced into the electrophoresis chamber, e.g. at a rate of approx. 2-5 L/hr.

Also, it is often desirable to establish and maintain an inert atmosphere, e.g. an argon or N₂ atmosphere, most often a N₂ atmosphere, to prevent any unwanted oxidation of the proteins during the long electrophoresis interval when the proteins are exposed to high electrical fields. Finally it is very desirable, considering that the individual proteins will usually be present in amounts less that a nanogram or a picrogram or a femtogram, that the atmosphere inside the apparatus is clean of any dust, bacteria or (larger) organic molecules. This can also be achieved by continuous flushing with nitrogen so that there is always a slight over- pressure in the apparatus.

Furthermore, the chamber may also comprise means for regulating the temperature in the chamber enclosure, e.g. temperature regulation may effected by circulation of a termostated liquid through the centre of the drum (see Figure 3), or by circulating a thermostated liquid within the chamber enlosure or through channels in the housing.

This being said, the temperature should preferably be regulated to between 5°C and 75°C, preferably between 15 and 60°C for samples where nucleic acids or their analogues are analysed, or between 10 and 40°C for samples where proteins are analysed in SDS- containing solutions, or between 10 and 30°C for samples where proteins are analysed in urea containing solutions. Advantageously, the temperature is kept at a fairly constant level during the electrophoresis process.

Loading of the gel strip(s) can be achieved by clamping the rehydrated gels to the drum using one of two electrodes and then rotating the drum to wind on the gels. When the gels are in place, the other electrode is used to hold the gels to the drum's surface. Alternative methods could envisage a multiple gel-strip holder which is clipped onto the drum or where a series of individual strips are clamped each by individual electrodes.

The sample (e.g. protein mixture) is typically applied to the strip at or near the more pH neutral end or region thereof. The reason for this precaution is that proteins may become degraded at high pH or attacked (carbamylated) by urea (when urea is used). Thus, this is particularly essential for basic pH ranges, i.e. pH above 7.0, and for pH ranges below 5.0 (e.g. pH 4.0 - 5.0). For other pH regions, i.e. pH 5.0-7.0, this precaution is normally not necessary.

If the samples have not already been loaded into the gels during rehydration, one way of applying a sample (protein mixture) to the strips is to rotate the drum so that the neutral region of the gels that are loaded is directly underneath the sample applicator positioned above the centre of the drum, see Figure 3. The sample applicator may be a separate facility of the chamber. The sample applicator is lowered onto the gels,. Thus, the electrophoresis chamber preferably also comprises means for locking the drum so as to avoid undesired rotation upon application of the sample. The electrophoresis can be started and after a period of time (which will depend upon the sample and the type of analysis being carried out) the sample applicator can be removed again. The sample (protein mixture) is normally applied to the pH neutral part of the pH-gradient (gel) strip, or at the end corresponding to the most neutral pH. As an example, a sample (e.g. 400 μl) is applied to a 6 cm piece of filter paper (of equal width to the gel, e.g. 3 mm) and this is then laid on top of the gel (close to the neutral region of the gel) (see Figure 3). Thus for the pH ranges that are described here as needing sample application by cup loading, the quantitative results are decisively better as the standard deviations are far smaller than what is usual for acidic or basic proteins if the samples are applied in the manner described rather than rehydrating the gel with the sample.

The procedures for gel loading, sample loading, electrophoresis, protein detection, processing (enzyme digestion or protein elution), and transfer for mass spectrometry can be fully automated in the electrophoresis chamber.

Gel electrophoresis

A standard 18 cm long IPG gel covering the pH range from 4 to 7 from Amersham Biosciences using the traditional buffer system (urea, thiourea, DTT, and ampholytes) usually requires about 50,000 Vhrs (Volt x hours) to separate the proteins. A gel that is 100 cm long requires approximately 200,000 Vhrs. Since conventional power-packs usually deliver only 3,500 V (Amersham Biosciences) the standard gels run for times in excess of 14 hrs. The long gels would require about 2.3 days. Higher voltages could be applied but even so these gels would need to run for about 20 hrs. However, when gels are run at high voltages, the risk exists that local heating can break down the urea. The cyanate ions that are produced would then attack the proteins and produce an artifactual series of charge modification variants.

The present invention provides a solution to this problem by devising a useful voltage profile for running IPG gels, in particular gels covering the basic pH range.

Thus, the present invention provides a method of separating a protein mixture by means of a IPG gel, said method comprising the steps of

(a) providing the IPG gel having a length of λ meters;

(b) providing the protein mixture to be separated; (c) applying the protein mixture to the gel;

(d) positioning electrodes at or near each of the ends of the gel;

(i) an initial voltage of 0-100 V

(ii) a first voltage increase gradient of in average 100-500 V/h for 1-10 hours up to an intermediate voltage of in the range of 300-1500 V, (iii) a second voltage increase gradient of in average 500-10,000 V/h for (3-15 m^"1 x λ) hours from the intermediate voltage up to a final voltage of (5,000-25,000 rrf¹ x λ) V, (iv) the final voltage for (5-50 m^"1 x λ) hours.

One initial step, step (a), in the method is to provide a suitable IPG gel. As a matter of definition, the gel has a length of λ meters in the longitudinal direction, i.e. in the direction corresponding to the direction of the gradient. The gel is preferably a gel as defined above, i.e. a gel prepared by the novel method of manufacture. The gel preferably has a length of at least 0.5 m, such as at least 1.0 m, e.g. at least 2.0 m.

The gel is preferably placed on a constant temperature surface. In a preferred embodiment, the electrophoresis chamber described above is utilised. Alternatively, the gel may be placed on a constant temperature platform designed similar to the Multiphore (but made longer than 1 meter) or in a constant temperature non-electrically conducting liquid bath.

Another initial step, step (b), in the method is to provide a suitable protein mixture. The person skilled in the art will know how to prepare such a protein mixture for gel electrophoresis, see, e.g., Harder A, Wildgruber R, Nawrocki A, Fey SJ, Larsen PM, Gorg A. Comparison of yeast cell protein solubilization procedures for two-dimensional electrophoresis. Electrophoresis. 1999 Apr-May;20(4-5):826-9. and 2) op cit: Nawrocki A, Larsen MR, Podtelejnikov AV, Jensen ON, Mann M, Roepstorff P, Gorg A, Fey SJ, Larsen PM. Correlation of acidic and basic carrier ampholyte and immobilized pH gradient two- dimensional gel electrophoresis patterns based on mass spectrometric protein identification. Electrophoresis. 1998 May; 19(6): 1024-35. 3) op cit: Fey SJ, Nawrocki A, Larsen MR, Gorg A, Roepstorff P, Skews GN, Williams R, Mose Larsen P. Proteome analysis of Saccharomyces cerevisiae: a methodological outline. Electrophoresis. 1997 Aug; 18(8): 1361-72 and these are hereby incorporated by reference.

The protein mixture is subsequently, in step (c), applied to the gel. The person skilled in the art will know several method for performing this step. However, the protein mixture is typically applied to the gel at or near the more pH neutral end or region thereof. The reason for this precaution is that proteins may become degraded at high pH or attacked (carbamylated) by urea (when urea is used). Thus, this is particularly essential for basic pH ranges, i.e. pH above 7.0, and for pH ranges below 5.0 (e.g. pH 4.0 - 5.0). For other pH regions, i.e. pH 5.0-7.0, this precaution is normally not necessary.

In one final step before the actual electrophoresis is started, the electrode wicks and electrodes (e.g. claims as described for the electrophoresis chamber above) are positioned at the ends of the gel. Application of a voltage over the electrodes is conducted in such a manner that the voltage increase is gentle so as not to activate the urea to attack the proteins. Therefore a slow increase is applied at first in order to remove a substantial part of the salt(s), e.g. at least 70%, preferably 95% of the salts, from the gel. If salt is present in the sample, then desalting will occur during the beginning of the process. This can be visualised by measuring the current flow during the electrophoresis. See for example the peaks at the beginning of the electrophoresis in Figure 2.

The length of the gel will have a certain impact on the voltage necessary and the time required for "desalting" and for the actual electrophoresis. The voltage profile according to the invention can be described as follows in relation to the length (λ) of the gel:

(i) an initial voltage of 0-100 V

(ii) a first voltage increase gradient of in average 100-800 V/h for 1-10 hours up to an intermediate voltage of in the range of 300-2000 V,

(iii) a second voltage increase gradient of in average 500-2,000 V/h for (3-15 m^"1 x λ) hours from the intermediate voltage up to a final voltage of (3,000-40,000 m^"1 x λ) V, (iv) the final voltage for (15-300 m^"1 x λ) hours.

Particularly promising voltage profiles are:

(i) an initial voltage of 0-50 V

(ii) a first voltage increase gradient of in average 300-500 V/h for 1-10 hours up to an intermediate voltage of in the range of 500-1500 V,

(iii) a second voltage increase gradient of in average 1,000-1,500 V/h for (1-10 m^"1 x λ) hours from the intermediate voltage up to a final voltage of (5,000-300,000 m^"1 x λ)

V,

(iv) the final voltage for (15-300 rrf¹ x λ) hours.

The total Vhrs applied in steps (i)-(iv) is very dependant upon the type of pH range that the gel covers, the purity of the water and the reagents used, the means used to prepare the sample and the total amount of salt present. The total kVhrs will typically be in the range of plus or minus 100% from the kVhrs values given for a selection of pH gradients in Table 3, in particular plus or minus 50% from these values and preferably plus or minus 25% from these values. The variation in total kVHrs is also reflected in the broad ranges given above.

By the described profile, it has been the aim to ensure that the current does not become too high, typically above 40 μA (μamp) per 1 mm² cross-sectional area of the gel. To make the gradient more "gentle" as described above, it is possible to extend the length of the steps (ii) and (iii) so that the gels experience more time at lower voltages. During this period, low molecular weight molecules migrate (salts, buffers, ampholytes and small peptides) through the gel.

The total kVhrs (V times hours) necessary may be determined by simple optimisation procedures, e.g. by running gels for different total Vhrs and then selecting the pattern that gives the best protein pattern. This has to be done because, as mentioned above, the quality of the water, reagents and samples varies from lab to lab.

The present method is particularly useful for running basic gels (by definition gels covering pH ranges that extend above pH 8, e.g. covering the pH ranges 6-9, 6-10, 6-11, etc.). Some modifications or precautions are preferably applied when processing basic gels. Such modifications may also be applied for non-basic gels. Advantageous, the basic gels normally require less volt-hours (Vhrs) than gels covering the acidic or narrow pH ranges.

The running procedure is identical to the one described above except for the following:

In order to establish an atmosphere, which will not have detrimental effect on the electrophoresis process, it is desirable to include within the chamber in which the process is conducted an open container comprising a strong base, such as an aqueous base, in order to neutralise any C0₂ present or formed in the process. An aqueous solution of sodium hydroxide (e.g. 5M sodium hydroxide) will be suitable. Alternatively, or as a supplement, nitrogen can be introduced into the electrophoresis chamber, e.g. at a rate of approx. 2-5 L/hr.

The temperature of the electrophoresis should preferably not be allowed to exceed 25°C (to avoid urea breakdown) or to fall below 15°C (otherwise the urea might start to precipitate). Preferably the temperature should be maintained at around 20°C throughout the run.

When ramping up the voltage, the current should preferably not exceed 60 μA per mm², in particular not exceed 40 μA per mm², such as not above 30 μm per mm², as this can give rise to cyanate ions.

If salt is present in the sample, then desalting will occur during the beginning of the process (as described above). Furthermore, the final voltage used is far lower than the final voltage conventionally used for the 18 cm gels. The voltage applied in connection with the method does not cause a breakdown of large amounts of urea into cyanate ions in the basic environment (see above). Therefore, the avoidance of carbamylation is of particular importance. These features are significant for obtaining a far better focussing and a substantial improvement of the reproducibility of the quantitative data. Finally, when working with samples of high specific activity (large number of CPM and little protein), a very small amount of a non-radioactive sample is added (about 50 μg/gel of a cell lysate (this is an amount far larger than the radioactive sample which is usually added). There are two reasons for this addition.

Firstly, if a small amount of cyanate ions should be generated during the focussing time, then these ions can react with the protein present. If there is only radioactive proteins present then the cyanate ions will be in excess and so substantially affect the image obtained. If however a small amount of non-radioactive protein has been added, this will be in excess to the cyanate ions present and so their affect will be 'absorbed' primarily by these non-radioactive proteins. Thus most of the radioactive proteins will not be modified and so the protein pattern obtained will not be substantially affected. The second reason is similar - high affinity, non-specific binding sites exist in the gel (especially at basic pH's which tend to depolymerise the gel) and these can bind to the protein.

The following represents particular aspects of the present method:

The method, wherein the concentration of 0₂ in the air around the gel is reduced by displacement with an inert gas like nitrogen.

The method, wherein the thickness of the gel is between 0.01 and 3 mm, preferably between 0.05 and 1 mm.

The method, wherein the thickness of the gel is reduced to facilitate infusion of compounds into the gel for further characterising the proteins present.

The means, where a dye or a fluorescent dye is infused into the gel in order to be able to quantitate the amount of proteins present.

The method, where a proteolytic enzyme is diffused into the gel to digest the proteins (such enzymes being trypsin, lys-c and similar)

The method, where a laser or UV sensitive dye is infused which aids the release of the protein or peptides from the gel for mass spectrometry.

The method, whereby the amount of protein can be measured in the gel by passing the gel past a narrow slit or window, in particular where the device is a scintillation counter which can measure the amount of radioactivity present in at least part of the gel. In particular the method where the device is a light source and a camera, optionally a fluorescent camera which can capture images of the gel strip so that the amount of fluorescence or staining can be used to estimate the amount of protein present in at least part of the gel. The methods, gels and chamber of the invention offers several advantages:

1. They renders it possible to recover the separated proteins from the gels for further processing - e.g. for mass spectrometry, antibody production.

2. Detection of proteins is greatly simplified because the gel can be scanned through either a scintillation counter or a fluorescence camera to detect and quantitate the protein present.

3. Image analysis is considerably simpler for ID gels than 2D gels and this thus eliminates the major bottleneck in this technology. The image analysis can be coupled directly with the mass spectrometry so that the proteins are quantitated and identified in one operation without the need for manual intervention.

4. The quantitative results are decisively better as the standard deviations are far smaller than what is usual for acidic or basic proteins.

5. Proteins can be simply recovered from the gels for further processing - e.g. for mass spectrometry, antibody production.

6. Enzyme digestion can be carried out in the gel prior to mass spectrometry without the need to cut the band or spot out of the gel.

EXAMPLES

Example 1 - Preparation of a long gel

The set-up illustrated in Figures 1 and 4 was used in the following for the preparation of gels.

The glass plates used were 110 x 20 x 0.7 cm, cleaned meticulously and siliconised to reduce surface tension. A sheet of plastic (gel bond), that had been chemically activated so that it would bind covalently to acrylamide, was placed onto one of the glass plates. This was held in place by a thin film of silicone grease and tight contact was achieved by using a rolling pin to press the plastic onto the glass until it was possible to see Newton's rings at the glass-plastic interface over most of the area of the plates. The spacers and remaining the glass plate were then assembled into the apparatus shown schematically in Figure 4 (casting mould). The dimensions of the gel casting chamber were 110 x 20 x 0.05 cm (internal distance between the glass plate and the plastic).

This gel casting mould was then connected to the gradient mixer as shown in Figure 1. The diameter of each of the chambers of the mixer was 1.5 cm and their height was 15 cm. The gradient mixer was positioned above the top of the casting chamber and the silicone tubing connecting the two was lead through a peristaltic pump (close to the bottom of the casting chamber.

The dense and less dense acrylamide solutions are prepared and kept at 4°C (on melting ice).

The solutions are prepared and then degassed before the ammonium persulphate (APS) and TEMED are added. This is done to remove any effects of the oxygen, dissolved in the solutions, on the polymerisation process.

APS and TEMED should be added immediately before casting the gels to start the polymerisation process.

Initially taps Ti and T₂ are closed. Sixty millilitres of the first precursor solution of acrylamide plus IPG monomers of a particular concentration and mixture are filled into chamber (L (for Light solution)) and stirred with a magnetic stirrer. Forty six millilitres of the second precursor solution, containing a different mixture and concentration of IPG monomers plus an agent to increase the density of the solution are filled into chamber (H (for Heavy solution). An example of the recipe is given in the table below.

The following precursor solutions can be used to produce a gel that covers the pH range from pH 4.0 to 7.0 appear from the following table.

a) Acrylamide solution is 28.8% acrylamide : 1.2% bisacrylamide (weight/volume) b) TEMED is N,N,N,N'-Tetra-methyl-ethyIenediamine.

Tap T₂ is opened and the light acrylamide solution is allowed to flow to the entrance of G. Tap T₂ is shut. The difference in the volumes of amount of solution in L and H is adjusted so that the remaining liquid in the gradient maker now weights the same so that there is no flow between L and H when tap Ti is opened. Tap T₂ is opened and the peristaltic pump switched on. The acrylamide solution is now allowed to flow slowly into G. When G is full there is a gradient of both density and IPG monomers in G. The gel is then left to polymerise overnight at room temperature.

An exponential gradient can be obtained by closing L to the atmosphere so that the volume of liquid in L remains constant. In this case, twice as much of the second precursor solution is needed)

Please note that if the gradient is cast from the top of the casting chamber, then the heavy solution should be filled into chamber L and the light solution into chamber H but otherwise the procedure is essentially the same. There are some situations where this is preferable for example if the heavy solution contains an unstable component.

The next morning, the gel is removed from the casting chamber and placed into a stainless steel washing tray (130 x 50 x 5 cm) and washed 5 times with 1 L of distilled water each time for 30 minutes each time. Finally the gel is washed twice in 2% glycerol. The gel is then clipped to the back of a laminar flow hood (which recirculates clean air) or some other suitable place and left overnight to dry at room temperature.

The next morning the basic end of the gel is marked (in this case where the top spacer was) with a permanent marker pen and the gel is covered with a sheet of transparent celluloid plastic. The gel is then cut into 3mm wide strips using an industrial wheel cutter. A 20 cm wide gel produces about 50 usable strips that are 3mm wide (the outer 2 cms are discarded due to potential edge effects). These strips are stored in plastic bags at -20°C until needed. Example 2 - Separation of a protein mixture

In this examples, the a protein mixture is separated by one-dimensional gel electrophoresis utilising the long gel prepared in Example 1, i.e. a 1 meter long gel covering a pH-gradient range of 4.0-7.0, and the electrophoresis chamber described herein.

A protein sample (protein mixtures) is dissolved in lysis buffer so that 400 μl contains 2 x 10⁶ cpm [³⁵S]-methionine and 20 μg proteins.

One meter long gels are reswollen in 1.25 ml lysis buffer (7 M Serdolyte® deionised urea, 2 M serdolyte® deionised thiourea, 2% g/v Chaps, 1% g/v DTT, 2% v/v Pharmalyte® 3-10 (the actual amount of Pharmalyte® needed varies from batch to batch)) for a minimum of 10 hrs at room temperature in a closed reswelling cassette (e.g. covered with a glass plate and sealed with parafilm). The cassette should be made of a protein non-absorbent material (e.g. PTFE).

The electrophoresis chamber is utilised according to the following protocol:

The side chambers in the chamber are filled with 0.1 M NaOH, 0.1% SDS to humidify the atmosphere and remove the C0₂ from the atmosphere. (This is not necessary for gels that only cover the acidic region (up to pH 7)).

The water bath is switched on at least 1 hr before applying the samples in order to regulate the temperature of the running apparatus to a predetermined temperature, e.g. about 20°C.

The supply of nitrogen gas to the running apparatus is turned on. A low flow rate (e.g. 2- 5 L/h) is desirable so that the gels are not dried out. This can be bubbled through the solution in the bottom of the chamber to humidify the gas if needed.

The power-pack is programmed to give a suitable gradient, in this case a gradient suitable for a 1 meter long gel covering the pH range of 4.0-7.0 (see Table 2 and Figure 2). Switch off the current check at the start, if the power pack has such a safety feature.

Table 2: Electrophoretic parameters for running a lm long IPG gel covering the pH range from 4.0 to 7.0

a) The run is finished after phase 3, but the additional phase 4 is added so that the electrophoresis is not terminated before the operator is ready.

The milliampere and watt limits are normally kept constant unless more than 10 gels are to be run, in which case they are increased proportionately.

A corresponding voltage profile for a narrow range gel (pH range from 5.5 - 6.5) would be:

The shape of the electrical profiles for other gradient ranges is similar but the following values have advantageously been used for the maximum voltage and total volt-hours (Vhrs):

Table 3: Sample load method and running requirements for a selection of different IPG gels covering different pH ranges.

As will be apparent for the person skilled in the art, the exact running conditions need to be optimised for each type of gradient, buffer system used, and possibly also sample type (they may contain different amounts of salt).

Samples can be applied to the gel during the passive rehydration phase if the pH range depending upon the gel type as indicated in the table. If the sample cannot be introduced into the gels by passive reswelling it has to be applied using the small silicone "cups" or sample strips made of a porous material (e.g. filter paper). The sample has to be applied at the most neutral region of the IPG strip (e.g. at the pH 5 end of a gradient that covers the pH range from 3 to 5) or at the neutral region if that is covered by the gel in order to avoid protein precipitation or modification due to the extreme pH.

The electrophoresis chamber is closed. The computerised monitoring of the electrophoresis is started and the electrophoresis is then initiated.

When the electrophoresis is completed, the power supply is switched off, and the long gels are removed and treated appropriately. One application could be to dry the gel, expose to a phosphor-imager plate (the gel may need to be cut because the maximum size of the phosphor-imager plate is 35 times 43 cms) to quantitate protein expression. Subsequently the gel could be soaked in a trypsin solution to digest the proteins in preparation for mass spectrometry.

Claims

1. A method for the manufacture of a gradient gel, said method comprising the steps of: (i) arranging a casting mould having two parallel, vertical inner surfaces which define walls of an enclosure; (ii) providing at least two gel precursor solutions including a first gel precursor solution and a second gel precursor solution, said first gel precursor solution having a first density and said second gel precursor solution comprising a density increasing agent and having a second density, (iii) casting the gradient gel by providing the at least two gel precursor solutions to the casting mould to a height of h meters so as to establish the gradient gel, the ratio between the second density and the first density being at least (1.15): 1, said at least two gel precursor solutions being provided to the casting mould so that the volume corresponding to the lowermost 10% of the gel precursor solutions predominantly, such as by at least 80%, consist of the second gel precursor solution, and so that the volume corresponding to the uppermost 10% of the gel precursor solutions predominantly, such as by at least 80%, consist of the first gel precursor solution.

2. The method according to claim 1, wherein the density of the second precursor solution is at least 1.15 kg/L.

3. The method according to any of the preceding claims, in which the density increasing agent is selected from dextran T500, Percoll, caesium chloride (CsCI), caesium sulphate (Cs₂S0 ), caesium formate, and sodium bromide (NaBr).

4. The method according to any of the preceding claims, wherein the density of the first precursor solution is at the most 1.05 kg/L.

5. The method according to any of the preceding claims, further comprising the step of (iv) washing the gradient gel so as to reduce the concentration of the density increasing agents by at least 90%.

6. The method according to any of the preceding claims, wherein each of the gel precursor solutions comprises a gel-forming agent, a cross-linking agent and a catalyst.

7. The method according to claim 4, wherein the gel-forming agent is acrylamide, the cross-linking agent is bisacrylamide and the catalyst is TEMED.

8. The method according to any of the preceding claims, wherein the second gel precursor solution comprises acrylamide, bisacrylamide, and caesium chloride, and has a density of at least 1.15 Kg/L.

9. The method according to any of the preceding claims, wherein one of the parallel inner 5 surfaces is adapted to form covalent bonds with the gel under the polymerisation conditions.

10. The method according to any of the preceding claims, wherein the height h is at least 0.5 m.

11. A gradient gel obtainable by the method defined in any of the claims 1-10.

10 12. The gradient gel according to claim 11 which at least covers the pH range of 4-10.

13. The gradient gel according to claim 11 which covers a pH ranges of about 3, or a pH range of about 1, or a pH range of about 0.1 pH units

14. The gradient gel according to any of the claims 11-13, said gel having a length in the longitudinal direction (i.e. in the direction of the gradient) of at least 0.5 m.

15 15. The gradient gel according to any of the claims 11-14, which has a length to width ratio of at least 10: 1.

16. An electrophoresis chamber for pH-gradient strips, said chamber comprising a housing defining a chamber enclosure, a drum arranged within the chamber enclosure, a lid or door as a part of the housing for accessing the chamber enclosure, means for rotating

20 the drum around its axis of rotation, at least one pH-gradient strip(s) having a length to width ratio of at least 10: 1, said drum having arranged the at least one pH-gradient strip(s) on its surface of rotation, said strip(s) being arranged so that the longitudinal direction thereof is substantially perpendicular to, or at an angle of at the most 20° to, the axis of rotation of the drum, each end of the strip(s) being held by one or more 5 clamps, said clamps having conductive parts for electrically connecting each end of the strip(s) to an external high voltage power supply so as to apply a voltage over the opposite ends of the strip(s).

17. The chamber according to claim 16 which has two or more pH-gradient strips arranged on the surface of rotation of the drum, said strips being arranged substantially parallel 0 to each other.

18. The chamber according to claim 16 which has a pH-gradient strip spirally arranged around the surface of rotation of the drum at an angle of in the range of 2° to 15° to the axis of rotation of the drum.

19. The chamber according to any of the claims 16-18, wherein the pH-gradient strip(s) 5 is/are as defined in any of the claims 11-15.

20. The chamber according to any of the claims 16-19, wherein the length to width ratio is at least 50: 1.

21. The chamber according to any of the claims 16-20, which further comprises means for regulating the temperature in the chamber enclosure.

10 22. The chamber according to any of the claims 16-21, which further comprises means for regulating the humidity in the chamber enclosure.

23. A method of separating a protein mixture by means of a IPG gel, said method comprising the steps of

15 (a) providing the IPG gel having a length of λ meters;

(b) providing the protein mixture to be separated;

(c) applying the protein mixture to the gel;

(d) positioning electrodes at or near each of the ends of the gel;

(e) applying a voltage over the electrodes, said voltage representing the following 20 profile:

(i) an initial voltage of 0-100 V

(iii) a second voltage increase gradient of in average 500-2,000 V/h for 25 (3-15 m^"1 x λ) hours from the intermediate voltage up to a final voltage of

(3,000-40,000 m^"1 x λ) V,

(iv) the final voltage for (15-300 m^"1 x λ) hours.

24. The method according to claim 23, wherein the voltage represents the following profile: 0 (i) an initial voltage of 0-50 V

(iii) a second voltage increase gradient of in average 1,000-1,500 V/h for (2-10 m^"1 x λ) hours from the intermediate voltage up to a final voltage of (5,000- 30,000 m^"1 x λ) V,

(iv) the final voltage for (15-300 m^"1 x λ) hours.

25. The method according to any of the claims 23-24, wherein the strip covers at least a part of the pH interval 6-12.

5 26. The method according to any of the claims 23-25, wherein the protein mixture, if the gel covers the pH range of 2-5 or 7-12, is applied at or near the more pH neutral end or region of the gel.

27. The method according to any of the claims 23-26, wherein the concentration of the C0₂ in the air around the gel is reduced either by using a strong alkali or by displacement

10 with an inert gas like nitrogen.

28. The method according to any of the claims 23-27, wherein the concentration of 0₂ in the air around the gel is reduced by displacement with an inert gas like nitrogen.

29. The method according to any of the claims 23-28, wherein the thickness of the gel is between 0.01 and 3 mm, preferably between 0.05 and 1 mm.

15 30. The method according to any of the claims 23-29, wherein the thickness of the gel is reduced to facilitate infusion of compounds into the gel for further characterising the proteins present.

31. The means according to claims 23-30, where a dye or a fluorescent dye is infused into the gel in order to be able to quantitate the amount of proteins present.

20 32. The method according to claim 23-31, where a proteolytic enzyme is diffused into the gel to digest the proteins (such enzymes being trypsin, lys-c and similar)

33. The method according to claim 23-32, where a laser or UV sensitive dye is infused which aids the release of the protein or peptides from the gel for mass spectrometry.

34. The method according to any of the claims 23-33, wherein the IPG gel is as defined in 25 any of the claims 11-15.

35. The method according to any of the claims 23-34, wherein the gel has a length in the longitudinal direction (i.e. in the direction of the gradient) of at least 0.5 m.

36. The method according to any of the claims 23-35, which has a length to width ratio of at least 10: 1.

37. The method according to any of the claims 23-36, whereby the amount of protein can be measured in the gel by passing the gel past a narrow slit or window

38. The method according to claim 37, where the device is a scintillation counter which can measure the amount of radioactivity present in at least part of the gel.

39. The method according to claim 38, where the device is a light source and a camera, optionally a fluorescent camera which can capture images of the gel strip so that the amount of fluorescence or staining can be used to estimate the amount of protein present in at least part of the gel.