From: International Conference on Isoelectric Focusing and Isotachophoresis, Annals New York Acad. Sci. 209, 515-529 (1973)
DR. RODBARD: This afternoon we have a very fortunate, perhaps unprecedented, opportunity to discuss several aspects of electrophoresis theory and principles. In opening this conference two days ago, Dr. Catsimpoolas referred to one general classification of electrophoretic methods: (I) moving boundary, as exemplified by application of the Tiselius apparatus; (2) zone electrophoresis. including paper, starch, and polyacrylamide gel electrophoresis; (3) disc electrophoresis, also known as zone electrophoresis in multiphasic buffer systems or mutiphasic zone electrophoresis; (4) isotachophoresis, which is a term that first appeared in an issue of Science Tools1 and was apparently the consensus definition of the authors oE papers in that LKB publication; and finally (5) isoelectric focusing. which has also been referred to as pH gradient electrophoresis and "isoelectric line spectra."2,3
I hope that the panel will discuss the interrelationships among these various methods, indicate where they rest on a common theoretical physical basis, clarify the distinctions between them, and attempt to avoid redundancy of nomendature. In short, I hope we will be able to define which concepts are identical, which are closely related, and which might be different. Finally, we might attempt to develop a unified or general approach to electrophoresis.
Dr. Catsimpoolas has charged me with raising the question as to what is the proper terminology for these various methods and to consider the possibility of renaming the Confrence if that were felt to be desirable. I'd like to begin with consideration of isoelectric focusing and propose,an alternative dassification scheme, namely, the distinction between equilibrium, steady-state, and transient aspects of electrophoresis. Items 1&endash; 4 (TABLE l) are all generally regarded as kinetic or transport processes, whereas the formulation generally given for isoelectric focusing is that of an "equilibrium" process somewhat analogous to density gradient centrifugation.
I would like to ask Professor Rilbe if he feels that isoelectric focusing also has a transient-state component, and whether we can regard this as a kinetic process, or a transport process. In other words, how does isoelectric-focusing relate to the other forms (1&endash; 4 of TABLE l) of electrophoresis? And, can we develop a theoretical basis for isoelectric focusing in common with the other methods of electrophoresis?
DR. H. RILBE: I don't remember if I was the first to create the term "isoelectric focusing." Was I? Alexander Kolin used the same word before me.
DR. RODBARD: Actually, Kolin did speak of the "focusing" of proteins into thin zones in one of his articles.2 He also referred to these as isoelectric line spectra.3 At that time he was talking about mobility spectra and isoelectric spectra. Certainly, he was one of the pioneers in the development of pH gradient electrophoresis.
DR. RILBE: I remember that, and I remember also his term "isoelectric spectra," which I myself mentioned in the first lecture.4
I also remember that there was a Swiss worker, Schumacher, working in the inorganic field, who developed a method for what he called "focusing ion exchange":5 it wasn't simply ion exchange, because he used an electric current; His process was similar to isoelectric focusing, the difference being that his complex formers were not protons, but compounds similar to EDTA and other complex formers for inorganic ions.
The phenomenon that he described and the analytical procedures that he developed&emdash;mainly with paper as a medium&emdash;were based on the same principle: that is, the use of electric current, forcing inorganic ions to be isoelectric, to condense, to focus into zones, and then remain stationary.
In our work on isoelectric focusing of proteins, we have restricted ourselves to protons as the only complex partners in making this isoelectric condensation.
In regard to the second question: Is there a transient state? Of course there is. Everybody knows that. But, I personally am not interested in it because the theory for the steady state is so simple, whereas if this theory were available for the transient state, it would be extremely complicated. I don't think it would be useful for anything
DR RODBARD: Actllally, the one place where I would propose that a kinetic treatment might be useful, is in trying to predict how long it takes for the gradient to be established and then, again, how long it takes for the protein to reach its final steady-state position.
I'd like to ask Dr. Ornstein and Dr. Jovin to comment on whether they feel that isoelectric focusing is an equilibrium process in a thermodynamic sense, or whether it is a transport process? Furthermore, in view of the role of electroendosmosis, which has been much
1. Moving boundary electrophoresis
2. Zone electrophoresis
3. Disc electrophoesis or multiphasic zone electrophoresis
4. Isotachophoresis (ITP)
5. Isoelectric focusing (IF) or pH gradient electrophoresis
discussed in the last few days,6 can we regard isoelectric focusing as an equilibrium, or as a steady-state system at any time?
DR. L ORNSTEIN: I think that everything depends upon what time scale you want to use. For example, one can look at the process in terms of how long it takes for a species to move in an ampholyte gradient to a point of constant pH, independent of what else happens thereafter. For example, let's consider a rather poor isoelectric focusing setup where, in fact&emdash;for reasons we will disregard for the moment&emdash;the whole pattern is drifting. Yet, you can have a situation where, if you feed the test sample in from both ends, it focuses to the same level, and that level drifts.
If the sample focuses to a constant pH and then that pH zone drifts, then I think to a first approximation, even in the presence of the perturbation phenomena, you can consider this as a nontransport phenomenon. A transport phenomenon is superimposed on this, but if the drift is slow, it can be neglected to a first approximation. If the drift is large, it can't be neglected. For most purposes, it is reasonable to consider focusing as an "equilibrium-type system." Consider an isoelectric focusing setup where Ampholine was focused into a pH 3 &endash; 10 gradient. Large quantities of glycine are then added to the two reservoirs. Below pH 6, glycine is partially a cation, and above 6 it is an anion. The upper pK is approximately 9.7 and the lower near 2.3. The result is that glycine will move in from both reservoirs toward the isoelectric point, which is approximately 6. As long as you have an infinite source of glycine at both ends, glycine will accumulate in the middle and broaden the isoelectric zone in the region of 6, thus, in a sense. producing a plateau near pH 6.
Well, glycine is an ampholyte with two pKs&emdash;low and high. Water is an ampholyte with two pKs&emdash;low and high&emdash;only further apart. Given sufficient time, water moves exactly the same as you would expect glycine to. So, in effect, the endosmotic flow that I was talking about the other day6 is this flow of water in from both ends toward the middle. Now, it depends entirely upon the voltage gradient and the time scale of your experiment, whether that perturbing transport is to be considered serious or not.
I think that for a general understanding, it's best to think of isoelectric focusing as an equilibrium method. Some of the other methods under discussion (TABLE l) are steady-state processes, and all of the others are transport processes.
DR. T. M. JOVIN: From a strictly thermodynamic point of view, isoelectric focusing cannot be regarded as an equilibrium method because there is necessarily a net transport of at least two constituents within a system to maintain the electrical field. In this sense, it is really indifferent whether you use glycinate and glycinium ions or hydroxyl ions and protons as constituents. The method is impossible in the absence of transport and, therefore, I think is best regarded from that standpoint as a steady-state method.
On the other hand, if one considers merely the macromolecular species, it may then assume equilibrium or steady-state distributions under the influence of a constant field and otherwise constant conditions. Nonetheless, the system as a whole certainly is not at a thermodynamic equilibrium.
DR. RILBE: Regarding the mass transport aspect of isoelectric focusing: True, there must be mass transport to make transport of electricity possible. If you look at the focusing experiment without such complications as electroendosmosis, we find that there is no visible, or at most a very slow, mass transport. The isoelectric spectrum remains there day after day, largely unchanged. What can the explanation be? What mass particle transport occurred? My answer to this is that the hydrogen ions transport the entire electrical current.
Consider an isoelectric column with anode and cathode and consider a focused zone of a carrier ampholyte. In one region (e.g., the upper part of the column near the cathode) the cations predominate, while anions predominate in the lower region near the anode. Accordingly, the cation moves upward, but after migrating a short distance upward it loses its proton and moves down again in the form of an anion. The cations move up, they lose the proton, and then move back again in the form of anions. The sum of all these migrations is that protons, and only protons, are transported through. Therefore, one can say that protons are the carrier of the entire current, and we have mass transport of protons (otherwise electric transport is impossible). But, the net result is zero transport of the carrier ampholytes and the (macromolecular) amphoteric species.
DR. JOV1N: A human being eating the same amount of material that he dissipates in a day also may show no alterations over long periods of time, but he is not in a thermodynamic equilibrium: this is a dissipative system, a steady-state system, and clearly not one of equilibrium. In this sense, this is not an unreasonable analogy: there is an inflow of energy to the system which is compensated by the dissipation of heat and the system remains at a steady-state "equilibrium" if you like but this is not a thermodynamic equilibrium.
DR. RODBARD: How does this compare with sedimentation equilibrium or density gradient centrifugation, for instance, from this point of view?
DR. JOVIN: The gravitational field is a conservative field, whereas an electrical field is not. There is no such thing as mass transport equilibrium.
DR. RILBE: I think these two phenomena, isoelectric condensation and isopycnic condensation in an ultracentrifuge, are completely analogous. In the centrifuge cell you have a driving force; you have a condensing driving force (that is, the gravitational force) on the one side and the diffuuional force on the other. The net result is a condensing force. If one replaces the gravitational force by the electrical force, one obtains isoelectric condensation instead. There is no other difference as far as I can see, so if one process is an equilibrium process, the other must be also.
DR. JOVIN: I still would wish to maintain that the basis of the isoelectric focusing method is the creation of a pH gradient. Without the pH gradient, we do not have the method. The fact that we have a pH gradient means that there is a gradient in the concentration of the hydrogen ion through the system. That means the chemical potential of the hydrogen ion is not the same throughout the system. Then I would find it impossible to describe this as an equilibrium method if we do not wish to ignore the participation of the hydrogen ion. But, since the hydrogen ion is the basis of the technique, I feel that the process must be treated, accordingly, as one of non-equilibrium.
DR. S. HJERTEN: I think it is most important that we examine the similarities between different methods. We should not only consider electrophoresis, we should also consider other methods as well, as Dr. Rilbe was considering here (centrifugation, isopycnic centrifugation, and isoelectric focusing). I think we should continue along this line
If we can find similarities between different methods (electrophoretic, chromatographic, or centrifugation methods), we should also try to find similarities in the rationale and principles for these methods. I think it is more important to relate these methods to electrophoresis than to discuss their names. We should extend the analogies between electrophoresis and isopycnic centrifugation to chromatography as well.
Assume we have a chromatographic column filled with beads or with a gel of some kind (e.g., dextran, polyacrylamide, agarose, etc.). Let us assume that we apply a zone of protein on top of this column and elute with such a high concentration of ammonium sulfate that the proteins will precipitate at the top of the column. Then, suppose we continue elution with a gradient in the concentration of ammonium sulfate, so that after some time the concentration of the ammonium sulfate is so low that the protein will dissolve and migrate down through the column. However, the protein will migrate faster than the ammonium sulfate, so that after some time it will begin to precipitate again because the ammonium sulfate concentration is sufficiently high (if the ammonium sulfate gradient remains steep enough). This process will be repeated over and over again. I believe that this chromatographic method is analogous with the isopycnic centrifugation method or isoelectric focusing.
I want to stress the similarities that you can find with many other techniques. Dr. J. C. Giddings has shown that you can find very similar equations in chromatography, electrophoresis, centrifugation, and other methods as well.7,8 This analogy has also been pointed out by Dr. C. J. 0. R. Morris in the Conference in Guilford, England, December, 1971. We should take this into serious consideration when we discuss terminology and attempt to develop a unified theory not only for electrophoresis but for all separation methods.
DR. RODBARD: We shall now close our consideration of isoelectric focusing, at least temporarily, and I would like to turn to items 1&endash;4 of TABLE 1 to examine the interrelations among these, and even to ask whether these are not the same thing, or different aspects of the same thing. Dr. Everaerts, how did you distinguish between these various forms of electrophoresis?
DR. F. M. EVERAERTS: You can compare boundary electrophoresis with chromatography and with frontal analysis. At the beginning of the fractionation, one has all components together in a zone at the end of the column. After migration has occurred, ideally, onc obtains a zone containing only a single component. Generally, the slower migrating zones will contain progressively more components.
In zone electrophoresis, you have an effect somewhat analogous to elution (from a chromatography column), and this eluting effect is accomplished in this case by ions (or more exactly, by the voltage gradient). You can compare it with the familiar elution chromatography. Thus, you never will reach a steady state in conventional zone electrophoresis (in the absence of multiphasic buffer systems), or in the moving boundary case.
With regard to isotachophoresis (or steady-state stacking or displacement electrophoresis), there is the important phenomenon of obtaining a true "steady state." The separation in isotachophoresis can be compared with moving boundary electrophoresis, where there is just a brief period (depending on several factors, such as the sample load introduced and the concentration of electrolytes) when one obtains steady-state transport. In disc electrophoresis you have a combination of both isotachophoresis and zone electrophoresis. In the upper gel you have stacking or steady-state stacking, whereas in the lower gel you have a form of zone electrophoresis, when one considers the sample macromolecule being fractionated. You can also find an "equilibrium," where you find steady-state stacking in the upper zone and a steady-state moving boundary (the "front") in the lower gel.
DR. RODBARD: Dr. Ornstein, I wonder if you'd comment on this matter. Specifically, would you accept the idea that the first stage, the so called stacking stage of your procedure, is analogous or identical with isotachophoresis? And, do we have a steady-state moving boundary and/or isotachophoresis in the lower gel?
DR. ORNSTEIN: Under ideal conditions, the process in our so-called spacer gel (which is one of the two upper gels), is identical to isotachophoresis. Also, under ideal conditions, in the lower (separation) gel (we never have ideal conditions similar to those in moving boundary electrophoresis), we observe zone electrophoresis, with the difference that at the starting point in zone electrophoresis all our fractions ideally are already preseparated&endash;&endash;ranged in contiguous "stacks" of "disks."
However, these operational distinctions tend to blur the unity of all these methods. First, consider moving boundaly electrophoresis: If your tube is long enough and you run it long enough, as the moving boundaries and the zones finally move out from one another, you have a transition from moving boundary to zone electrophoresis. And, if you look at what is happening at the leading and trailing boundaries of each of these zones, you see phenomena that are closely related to steady-state stacking.
Dr. Jovin pointed out that for steady-state stacking, at a particular boundary, the two species on either side have to disappear across the boundary. That makes for a clear understanding of the method, but consider the following experiment. Suppose we put bromophenol blue in the upper reservoir and it feeds continuously, in low concentrations, into the region ahead of the proteins and just behind the chloride (or leading ion) in a typical disc electrophoresis run. Thus, we have "stacking" of bromophenol blue, but because of the continuous source in the upper reservoir, it doesn't disappear as you cross the boundary behind the stack in which bromophenol blue is concentrated. If you were to increase the concentration of bromophenol blue in the upper reservoir, you would increase the concentrations behind that boundary between the stack of bromophenol blue and the next slower component, and in the limit you no longer have a stacking step. This experiment illustrates that one can have stacking, depending upon whether there is or is not a source of additional ions [bromophenol blue, the sample rnacromolecule(s), or an additional buffer constituent]. Thus, depending entirely upon the actual operating_conditions, each of these methods (TABLE 1, 1-4) (leaving aside the isoelectric focusing) transform one into the other. Even isoelectric focusing represents a hybrid of some of these during the transient state when the pH gradient is being formed. Accordingly, it is useless to make great distinctions among these methods, except when you choose models to describe the special properties of each system.
I want to make one general comment on terminology and classification. I think we should try to avoid arguing about whose favorite term for a particular phenomenon is best or the first. For example, I think it hardly matters what my definition of disc electrophoresis was, if usage has changed its meaning, with some exceptions. I don't think it's useful to try to change the habits of a large number of people in the use of words. I think it is useful to examine misunderstandings that lead to ambiguity, where different people are using the same term to designate different things, and I think we should try to dear those up. But, I don't think we should discuss which of a pair of otherwise equal terms is the preferable one.
Although many people now use the term "disc electrophoresis" in a way differently than I intended (and therefore we probably by now should accept the public's view of what is disc electrophoresis), the definition in my paper was an electrophoretic system in which there are discontinuities (of pH, ionic strength, buffer composition, and gel concentration) intentionally built into the system.9 This term comes primarily from the first part of discontinuity. Therefore, in a sense it was meant to be equivalent to what Dr. Jovin means by multiphasic zone electrophoresis.
DR. JOVIN: I'd like to return to the experiment discussed by Dr. Ornstein. First, if you increase the bromophenol blue in the upper buffer, it will not change the concentration of the bromophenol blue in the stack, but merely increase the length of the stack once it has completely formed. Second, I think the explanation for the statement Dr. Ornstein made regarding the hypothetical bromophenol blue experiment really goes back to his comment about the whole process depending upon whose clock you use, that is, the temporal extent of the experiment. If you waited long enough in the example he has cited, all of the bromophenol blue would achieve its proper position in the stack in the steady-state situation. This is the same argument that arises in consideration of the use of Ampholine spacers in isotachophoresis The completion of stacking depends upon how long you wait. As I tried to indicate before, if you do not fulfill the correct steady-state moving boundary conditions, this could take a great deal longer than anybody envisions. For example, if one requires a system to completely invert the order of the electrolytes, so that what you originally set up physically as a leading electrolyte ultimately becomes a trailing ion, the process might take so long that we could never really achieve the objectives of the experiment.
DR. ORNSTEIN: I agree with everything that Dr. Jovin said, with one exception. In the case of adding the anion (bromophenol blue) in the upper reservoir, the concentration of bromophenol blue in the stack depends upon the proportion of the current carried by that anion in the most "trailing solution." If the concentration of that anion in the reservoir is increased, and if it has a higher mobility than he main trailing ion, then the proportion of current that it carries increases, ultimately leading to the breakdown of steady-state stacking. However, use of low concentrations of additives to the upper rcservoir permits one to do what I call "SDS (sodium dodecyl sulfate) disc eiectrophoresis." By adding controlled low concentrations of SDS to the upper reservoir, one obtains some complexing of SDS with some of the proteins, and the proteins are continuously exposed to a (nearly) constant concentration of SDS that depends on the concentration of SDS in the upper reservoir.
This technique can also be used for introducing sulfhydryl groups into the gel, for example, using thioglycolic acid. If your protein is one that requires a sulfhydryl atmosphere to prevent denaturation, then you can put thioglycolic acid in the upper reservoir (in an anionic run) at a suitable concentration to produce the desired continuous bathing concentration throughout the gel during the entire run. The maximal allowable concentration compatible with steady-state stacking will depend upon the proportion of the current carried by the trailing anion proper and the additive in the upper reservoir.
DR. JOVIN: I agree that you can raise the concentration of the additive to a level higher than the equilibrium concentration it would achieve if it were stacked under the steady-state stacking conditions. But, I still maintain that if one does eventually achieve the steady-state situation, the concentration of that constituent will be determined by the nature of the leading ion in the phase, as given by the so-called Kohlrausch regulating function in the stack.
DR. ORNSTEIN: I agree that the steady-state concentration of bromophenol blue (SDS, thioglycolate, etc.) is obtained in the stack; however, I'm considering the concentration (e.g., of bromophenol blue, SDS, thioglycolate) behind the stack.
DR. JOVIN: Behind the stack you can have any concentration. This is a transient state that is not defined by any available equations.
DR. ORNSTElN: It is not transient if you have an infinite source of that ion (i.e., an infinite reservoir with a fixed concentration of additive), although it doesn't have to be very enormous in practice.
ED. NOTE: Dr. Jovin uses the term transient, in contradistinction to "steady-state" because the concentrations of ionic species are continuously chaanging. Dr. Ornstein is using the term "transient" with respect to duration of the effect.
DR. RODBARD: We all agree that there is stacking of buffer ions in the upper or stacking gel, and hopefully (but not always) there is also stacking of the proteins and of the bromophenol blue (or other tracking dye) in the conventional electrophoretic run of this type. However, it is not generally appreciated that there is also a stack (i.e., a moving boundary in the lower gel) during the resolution or separation phase of the run; the bromophenol blue usually remains in this stack, and other rapidly migrating, highly charged, species (e.g., proteins, peptides, nucleotides, etc.) often remain in the staclc, unless you superimpose enough molecular sieving to retard them below the mobility of the "front" or pi-lambda boundary.l0
DR. ORNSTEIN: If the molecular sieving effect isn't sufficient then you can increase the Tris concentration in the lower gel, thereby raising the pH and increasing the relative mobility of the trailing ion (e.g., glycine), and permit it to overtake those rather fast components. Then they will unstack, but by changing the concentration of the constituent ions in the lower.gel, you can arrange it so that all species except the very small fast ions unstack.
DR. RODBARD: Similarly, one can arrange it so that the species of interest remains in the stack; then this becomes identical with steady-state stacking or with isotachophoresis without spacers.
DR. ORNSTEIN: Absolutely.
DR. RODBARD: I think many people are of the opinion, or have somehow gained the impression, that ampholyte spacer ions are necessary for isotachophoresis or steady-state stacking, or that it is desirable to have ampholytes present. I'd like to ask the panel: Are ampholytes (i.e., amphoteric electrolytes) necessary as spacers to have the types of separations that we've seen under the name of isotachophoresis?
DR. EVERAERTS: The mobility differences between the various Ampholine spacers are so small that it seems incredible to me that you can separate them in such a short run. I believe that you have a pH gradient between the leading and terminating electrolytes. Initially, the ampholytes are between the leading and terminating electrolytes, and the proteins are mixed together with the ampholytes. In the electric field you may obtain a very good separation of those proteins, But, my opinion is that the proteins are not separated into discrete zones separated by the ampholytes, but they are moving together with the ampholytes. I would be very interested if someone has sliced gels to see if one ever obtains proteins in a zone free of ampholytes, or whether one always obtains a mixture of proteins and ampholytes. I believe that you will find proteins and ampholytes always intermixed, so that you have a system very comparable to isoelectric focusing.
DR. JOVIN: I concur completely with that. If one expects all of the Ampholine components (that presumably encompass a gradation of electrophoretic mobilities&emdash;although this has not been experimentally established) to assume a steady-state position in the stack, one must be severely disappointed. The conditions will rarely be met for this to be true. Further, if the Ampholine components themselves do not stack, then it is impossible for the proteins to be "stacked" in between the spacers.
Regarding the original question: There is absolutely no necessity to use an amphoteric substance as the so-called spacer in isotachophoresis. The spacer merely has to be a substance that assumes a relative or a constituent mobility that falls within the range of the proteins under the conditions of the experiment. It can be a simple buffer ion such as Tris+, or any other ion or polyion, or it may be something similar to an Ampholine component. But, there is no a priori reason for the latter to be the case.
DR. RODBARD: One could use compounds of one particular class but varying in the number of charged groups. In one extreme case, the charged groups could all be of the same sign, so that the compound isn't an ampholyte and has no pI.
DR. JOVIN: I can envision steady-state stacking as being a useful preparative technique, by setting up a system where there might be 10 zones of "spacers," with the expectation that the protein(s) would fall somewhere in these 10 zones. The distribution of proteins within each zone would depend upon the distribution of mobilities. If the protein of interest happens to be in a zone with relatively few other proteins, then a very high initial purification might be possible. As an analytical technique for proteins in heterogeneous mixtures, I think it is hopeless to ever imagine that one would get useful information in the presence of multiple "spacers."
DR. ORNSTEIN: It is also unlikely that the distribution of components presently available in Ampholine is ideally suited for spacing of proteins in steady-state isotachophoresis. However, if the spacers are properly designed, or if one selects from mixtures the ones that satisfy the appropriate moving boundary equation, then one should be able to get very satisfactory spacing of very many (but not all) proteins. There will be pairs of proteins that, in conjunction with a particular spacer ion or in conjunction with the adjacent protein band, will not satisfy the boundary conditions and will therefore stay mixed.
One can use whole serum with a supplement of hemoglobin for spacers: This can be used in large quantities, such as 5-10 ml of serum, applied to a 2.5 cm diameter tube. There are sufficient differences in optical density for the hemoglobin, transferrin, and hemoglobin-haptoglobin complexes, and some of the other proteins, so that you can visualize the stacks. This experiment has shown that a fair number of proteins form very thin stacks. If one uses a large amount of protein, the stack has a "square" shape; that is, an albumin band may be 5 mm long, but with very sharp leading and trailing boundaries. Similarly, the hemoglobin and the transferrin stacks may be 1 or 2 mm long, again with very sharp edges.
This indicates that at least those visible components have stacked with respect to the protein zones that are immediately adjacent. This encourages one to believe that a reasonable number of proteins will migrate in steady-state stacking in such systems. Certainly, several workers have evidence for this. Accordingly, I'm somewhat more encouraged than Dr. Jovin about the likelihood of obtaining properly designed spacers serving that purpose.
DR. RODBARD: The experiment just described by Dr. Ornstein provides presumptive evidence of stacking (of hemoglobin, transferrin, etc.). However, the threshold of detection by the naked eye is such that the 'sharpness' of boundaries may be very misleading unless spectrophotometric, pH, and/or conductance measurements are used.
DR. EVERAERTS: Dr. Ornstein, if you have two proteins with a difference of mobility, let's say for simplicity 20 x 10-5 cm2 sec-1 V-1, and you bring a spacer in between, then you reduce the difference in mobilty between adjacent zones. Would it not be preferable to design and make detectors to resolve the small narrow zones, rather than to introduce spacers in between the proteins, thereby making the separation more difficult?
DR. ORNSTEIN: For analytical purposes, I agree that having detectors is one of the reasonable approaches. However, for preparative purposes, spacers are clearly advantageous..
DR. EVERAERTS: I agree.
DR. ORNSTEIN: Dr. Everaerts mentioned a 20 x 10-5 difference in mobility&emdash;an enormous difference. Clearly, if one has such an enormous difference, then there is a wide range of species available for use as spacers; therefore this becomes a very easy and thus uninteresting example. The interesting and challenging case arises when one is dealing with small differences in mobility.
DR. EVERAERTS: Yes, but that is what I intended. If I stipulate the difference of 20 and you introduce not one but, let's say 10 spacers, this may result in a mixed zone of spacers together with the two proteins.
DR. ORNSTEIN: If the two proteins show a large difference in mobility, and you introduce a spacer, unless the spacer has a mobility very close to one of the proteins, you will still have extremely sharp boundaries between each of them and the spacer. The beauty of steady-state stacking (or isotachophoresis) is that the conditions of the boundaries counteract diffusion spreading. Thus, the boundaries are extremely sharp for molecules with low diffusion coeflicients, such as proteins. The boundaries are of the order of a few microns thick when the voltage gradients are similar to those in a typical run: For a 5 mm diameter column with 0.01 ionic strength, one is usually dealing with a boundary only 10 µm thick. This degree of sharpness is such that Dr. Everaerts' conductivity detection system will not have sufficient resolution to record the true sharpness of those boundaries. It is very difficult to fabricate a conductivity detector that would have the necessary resolving power.
DR. JOVIN: I agree. As I and others have pointed out, there is a transmissivity, that is, a polarity of transmission of effects in moving boundary electrophoresis, that determines the constitution, concentrations of constituents, pH, ionic strength, and all the other properties of the trailing phase. We can introduce an arbitrarily complicated mixture of proteins above the gel and permit it to stack. If every component fulfills the steady-state moving boundary conditions and assumes a position in this stack, the final or trailing phase that forms in back of everything will be the same as it would have been in the absence of proteins. That is, there is no alteration of this transmissivity effect from the leading to the trailing phase due to the introduction of the protein(s).
DR. RODBARD: To summarize, the spacers in isotachophoresis need not be ampholytes. In the case of Kendall's classic experiments, the spacers were strong ions. In some of Dr. Everaerts' experiments, some of the fatty acids could be regarded as spacers between other fatty acids in the series. The same applies to the use of nucleotides. The spacers need not be ampholytes or amphoteric. Furthermore, there need not be any spacers at all to have a succession of moving boundaries between different components, each moving at the same speed. Suppose that the components of interest are adjacent. If you then increase the mass of any one of these, the width of the zone increases because the concentration is regulated, and one can obtain either analytical or preparative resolution with steady-state stacking. Steady-state stacking, or at least the concept of stacking and unstacking, applies not only in the upper gel, which is commonly called the stacking gel, but also in the lower gel of the so-called disc electrophoretic system. If I have understood the panel correctly, I believe we can conclude that moving boundary electrophoresis, zone electrophoresis, disc, multiphasic zone electrophoresis, and isotachophoresis are all the same thing occurring in either single or multiple buffer phases. I wonder if we can reach some kind of consensus of this panel regarding this. Would anyone take exception to that?
DR. ORNSTEIN: Certainly, they all can be classified under transport electrophoresis. In that sense, they are all the same thing. You need the same basic equations. The full set of appropriate equations, not setting aside any particular subsets that are relatively unimportant for particular versions, can be applied to the whole gamut of electrophoretic methods under discussion. Certainly, there is a unity for the full set of equations that justifies these methods being looked at collectively. But, of course, some of them (e.g., moving boundary and zone electrophoresis) can be distinguished in useful operational ways, and both of these can be distinguished in useful operational ways from steady-state stacking or isotachophoresis.
DR. RODBARD: Dr. Ornstein, you note that the basic equations are the same. I think we should reiterate that the equations that we're dealing with are essentially simple fundamental principles, for instance the conservation of mass, electroneutrality, the acid base equilibria (or multiple simultaneous chemical equilibria). and the laws of diffusion (Fick's first and second laws) which, in turn, derive from the law of conservation of mass. These are the only (or at least the major) physical principles involved here. I believe that even the electroendosmosis effects can be regarded in this same setting.
DR. ORNSTEIN: I would add to this the immensely important practical effects of the interaction of the macromolecules with the gel matrix, which can either be regarded as viscous effects, if they're looked at in a most general way, as momentum transfer effects, or as sieving effects.
DR. RODBARD: I would be happy to discuss the effects of the gel or other matrix, or the unified theory of gel electrophoresis and gel filtration.11 However, due to lack of time, I would like to ask the panel to restrict attention, for the moment, to the primarily "electrophoretic"effects, ignoring the molecular sieving effects.
I conclude we are all in agreement that we're dealing with one essentially unified phenomenon: the physical principles operating here are the same in all of these cases (TABLE 1, 1&endash; 4), and actually what we're changing are the initial boundary conditions, that is initial setup rather than the mathematical laws underlying the process.
I wonder whether it would not be possible to achieve any consensus on how we should designate these forms of electrophoresis. The use of the term "ionic migration method".has been mentioned.l2 Also, the concepts and terminology of stacking, unstacking, and steady-state stacking were introduced by Ornstein by the time of the meeting of the New York Academy of Sciences, 1964.9 The concepts of stacking limits, selective stacking, selective unstacking, and restacking were introduced by Jovin.l0 Finally, the term "isotachophoresis" apparently was coined by a group of authors publishing in Science Tools in 1969l
I wonder if the panel members would like to discuss the choice of terminology to see whether there are any preferences among the various options available.
DR. EVERAERTS: May I add one additional name: displacement electrophoresis. first used by Dr. H. P. Martin. In 1942, he carried out some analytical separations of amino acids in collaboration with Dr. Longsworth, and they coined this term.
DR. ORNSTEIN: My feeling is that, ultimately, usage determines and justifies which term is appropriate. So long as there is no misunderstanding and everyone understands that steady-state stacking and isotachophoresis and the earlier term "ionic migration" are synonymous then there is no problem. The trouble with Kendall's term (ionic migration) was that it could easily apply to all of electrophoresis and was too ambiguous and general for the special aspects emphasized here. I think that one desires a term that is both descriptive and easily remembered. Steady-state stacking has fewer syllables and it has alliteration. The term "isotachophoresis," at least, sounds as if it is one of the electrophoretic methods. I think both of those are useful. My guess is that, because isotachophoresis is derived from the Greek and has certain appeal in this regard, it will prevail. I see nothing wrong with that, so long as their is no misunderstanding.
DR. JOVIN: I think we should make it clear that, if we use the term isotachophoresis, it need not refer to the necessary addition of the spacer ions, whatever form they take. If we use a term such as this to indicate the process of steady-state s~acking, then it need not involve any further considerations of the presence or absence of spacers.
DR. ORNSTEIN: Haglund's definition of the term was as follows:
The name Isotachophoresis has been given to an elctrophoretic principle which in the past has been known by several names: 'ionic migration,' 'ionic mobility analysis,' 'Cons dectrophoresis,' 'displacement electrophoresis,' 'electromigration'and 'moving boundary analyis.'l
He did not indude steady-state stacking, but it also indudes steady-state stacking.9 It may turn out that people will tend to only use this term when there are spacers. just as they now use the term disc electrophoresis for a much more restricted set of phenomena that Dr. Davis and I intended with our definition of disc electrophoresis.9
DR. HAGLUND: I really ought to have induded reference to steady-state stacking. I think I corrected this to some extent (but not completely) on the next pagel when I mentioned that isotachophoresis was the same as steady-state stacking.
I wish to explain the origin of the term isotachophoresis. At that time, 5 years ago, we were not working with disc electrophoresis, but instead, we were dealing with a similar principle in thin layer electrophoresis. We were interested in the potential use of spacers. We were in contact with Russian workersl3 who have done a great amount of work in this technique in Leningrad. We encountered a long list of previously suggested names. A group of us were collaborating, induding workers from LKB, Eindhoven, Stockholm, and the protein laboratory at Copenhagen. Initially there were five or six groups, although later on several other groups were added. We felt that we had to unify some nomendature, at least within the group, to facilitate understanding and communication. We decided to coin a new term. Using the Greek, we combined "iso" for same, and "tacho" for speed, together with "phoresis" for migration. These concepts were synthesized by the terminology "isotachophoreds." This term has not been restricted to Science Tools publication of LKB; it has been utilized in the general scientific literature.
DR. A. CHRAMBACH: It is entirely possible to designate the use of electrophoretic steady-state stacking in the presencec of spacers by the term "isotachophoresis," in contradistinction to the use of multiphasic buffer systems in the absence of spacer ions, which could be designated by the term steady-state stacking. Historically, most of the users of this method learned about it from Ornstein,9 many years before any other term was to be found in the literature. Accordingly, one could retain both terms but arbitrarily use them to distinguish between application of the same principles in the presence or the absence of spacers.
DR. HJERTEN: Dr. Ornstein said that as long as there is no misunderstanding, we can use the well-established term. This is quite correct.
DR. ORNSTEIN: Even the more recent one&emdash;either one&emdash;as long as there is no misunderstanding.
DR. HJERTEN: I don't agree with this. I wish to reemphasize the similarities, and these similarities should be carried over into the terminology for several methods. This should facilitate the use of a method in the best way to obtain optimal resolution. It is important to examine the similarities between the numerous fractionation methods available. Let us again examine the similarities between chromatography and electrophoresis. Consider displacement chromatography: Its characteristic feature is that the concentration of a substance in a zone will reflect the nature of the substance, whereas the width of the zone is a measure of the amount of substance present. This applies to displacement chromatography and is closely analogous to the process of isotachophoresis. Therefore, we should try to find similar words for these two methods, as well as for other methods that are based upon the same principles.
DR. ORNSTEIN: I would have to severely disagree about thc two examples that you choose. That is, the striking thing about isotachophoresis is the sharp boundaries between the individual species. This doesn't exist in displacement chromatography at all, so that's a very poor comparison.
DR. HJERTEN: Theoretically, bands are very, very much narrower in a column through exclusion from the column packing material. Theoretically, there are no disdnctions between displacement chromatography and electrophoresis.
DR. JOVIN: The main distinction between displacement chromatography and electrophoresis is that in one case gravity is the sole restoring force, and in the other case there is an electrical field that we can control at will.
DR. ORNSTEIN: In electrophoresis, the electrical field provides the restoring force. That's the important thing. There is a restoring force in electrophoresis, but not in chromatography.
DR. HJERTEN: It is the restoring force that will give rise to these two phenomena that I mentioned and that will give rise to a single mathematical formula by which you can describe both.
DR. JOVIN: It seems that there is another term we haven't even considered yet, that is "steady-state electrophoresis." With such a term, we should observe all of these methods in which restoring forces are present, such as isotachophoresis and so forth.
DR. RILBE: Concerning steady-state electrophoresis, I'm afraid that there is a risk, a possibility of confusion. When Tiselius worked with a process similar to what we now call isoelectric focusing, in 1941, he utilized the term stationary electroIysis This term may be confused with steady-state electrophoresis.
I think that the term isotachophoresis is a good invention. Nevertheless, I am disappointed, because I agree with Dr. Hjerten that the analogy with displacement chromatography is so complete, so it is a pity that "displacement electrophoresis," as used by Dr. Everaerts, was not recommended instead.
As an alternative, one could introduce the term corresponding to "isotachophoresis" into the field of displacement chromatography, for example "isotachochromatophoresis." Again, it is unfortunate to omit the term "electro," but this would result in a longer term yet, "isotachochomatoelectrophoresis."
DR. HJERTEN: Let us extend these analogies to centrifugation. If you perform ultracentrifugation you will also obtain what we could call "displacement centrifugation." We have observed this many times, and have worked with this for several years. Thus, we have similar effects in ultracentrifugation, in chromatography, and electrophoresis. Hopefully, we could develop a common nomenclature and terminology.
DR. ORNSTElN: My objection (to the analogies with chromatography and centrifugation) is that one of the important differences is the role of the counterion in the electrophoretic methods, especially when one is dealing with proteins in a buffer consisting of weak acids and bases. The reason why Kendall's methodl2 lay unused for proteins from the 1920s until early 1960 and 1961, when Davis and I first obtained successful results with the method, was that there was no explicit treatment or appreciation of the role of the counterion&emdash;mainly, the importance of using a weak counterion as the base for anionic runs, or a weak acid for cationic runs, to control the mobilities of the ions in the stack and the mobilities of the trailing ions. Because the counterion plays such a dramatic role (in steady-state stacking), and there is no counter part in either of the other two cases (centrifugation and chromatography), I think that squeezing the chromatographic methods and the centrifugation methods into a straightjacket with steady-state stacking (isotachophoresis) is going to divert attention away from a very important consideration.
DR. RODBARD: A number of new names have been proposed. Dr. Chrambach proposed that the term isotachophoresis be reserved for steady-state stacking in the presence of spacers to distinguish it from the stacking in the absence of spacers. This corresponds to what has already occurred in practice. Of course, there was nothing about Ornstein's original definition of steady-state stacking that would exclude the use of spacers. Also, Dr. Haglund points out, in discussing Ornstein-Davis' disc electrophoresis, that: "When the potential is applied, the starting zone, while migrating in the buffer gradient, undergoes a considerable ooncentration and sharpening. This is nothing but isotachophoresis, and will, when allowed to develop, results in a considerable separation."l
I think we can turn this argument around and conclude that isotachophoresis is nothing but stacking.
DR. H. HAGLUND: I agree.
DR. RODBARD: Other statements to this effect also appear in Reference 1. We also have the term steady-state electrophoresis and a number of others: steady-state stacking (SSS); isotachophoresis (ITP); multiphasic zone electrophoresis (MZE); ionic migration method; steady-state electrophoresis; displacement electrophoresis; and isotachochromatophoresis. I'd like-to see whether there is any strong feeling about Dr. Chrambach's suggestion to reserve use of the term isotachophoresis for stacking in the presence of spacer ions.
DR. HAGLUND: Unfortunately, it is difficult to know what are spacers and what are not. Certainly, the principle itself cannot distinguish whether an ion is a spacer ion or whether it is a sample ion. The same ions (e.g., partially hydrolyzed protein) can be used as spacers, or they can be sample ions, depending on the application. Thus, I'm afraid that the proposed distinction may break down. I think we should try to approach a single terminology for all types of electrophoresis where all ions move at the same speed.
DR. RODBARD: Of course, operationally, we know whether or not we have deliberately introduced spacer ions (e.g., ampholytes). If we are to regard the terms SSS and ITP as synonymous, then certainly from the viewpoint of historical precedent, I think the term steady-state stacking would definitely have priority: It was the term that was described first and applied first to the same phenomenon.
DR. ORNSTEIN: Well, as the one who was responsible for originating and defining it, I think one should pay attention to what happens in natural languages. Among those reading the scientific literature, many persons have a concept of isotachophoresis that is at least close to being correct. Perhaps there are more people who "understand" the term isotachophoresis then there are persons who "understand" the term steady-state stacking. Even though the term SSS may have some priority in terms of history, it isn't useful to press the point.
DR. JOVIN: Except in one sense. The public needs a convenient bibliographic reference as a source of information. Now, the disc electrophoretic method is no problem, as the papers by Drs. Ornstein9 and B. J. Davis14 are among the most cited papers in the scientific literature today. Unfortunately, I don't believe there are equivalent situations for the other methods. What is needed is a complete, current, comprehensive review to make the subject more easily understandable to the large number of people using these techniques.
DR. RODBARD: I'd like to ask the panel members where we go from here? What are the frontiers in the theory of electrophoresis? What are the assumptions that we can now break down to attempt to obtain a more general description of these processes? For instance, most, if not all, of the theories have ignored the role of the hydroxyl ion and the proton, and we ignore the diffusion spreading of boundaries in the stack that can. under some circumstances (particularly, low field strength), become appreciable. Instead, these boundaries are being treated as though they were infinitely sharp.
DR. ORNSTElN: In the literature, the behavior of the boundaries has been discussed in great detail by Longsworth and McInness and briefly by myself.9 Those interested in that problem can find enough in the literature to understand it.
You say that the contribution of the hydroxyl and hydrogen ion have been ignored: In some cases they've been implicitly ignored because of no mention, and in other cases, say within a pH range of A to B, people will ignore them because their contribution is insignificant. That isn't ignoring it. That's being sufficiently explicit so that no one should be misled into thinking that you can always ignore the contribution of hydrogen and hydroxyl ions.
DR. RODBARD: Would other panelists like to discuss where we go from here with electrophoresis theory? There are certain obvious directions, for instance extending Jovin's theory from consideration of divalent weak electrolytes to polyvalent weak electrolytes and extending Jovin's and Ornstein's theory of binary systems, containing essentially a single anion and a single cation, to multicomponent systems where there might be 3, 4, or more ionic species (e.g., in the presence of SDS or thioglycolate) or to amphoteric compounds in general.
DR. RILBE: I would note that the classical moving boundary theory doesn't require application of the moving boundary equations to the hydrogen ions. After applying the moving boundary equations to all other ionic species, the hydrogen and hydroxyl ion concentrations are given by the electrical neutrality condition. If you know the concentration of all weak electrolytes in a zone, you also know the pH and [H+] and [OH&endash;]. Thus, we have not negelected the role of the hydrogen ion concentration.
DR. JOVIN: I'd like to emphasize this and I think, as Dr. Rilbe pointed out in 1948, it's difficult to treat the hydrogen ion separately, because it is not the sole ionic constituent of any separate substance. Thus, it is impossible to attempt to consider the hydrogen ion out of context with the other weak electrolytes.
DR. RODBARD: We must now close the discussion for lack of time. I hope that this discussion has served to reduce the rather bewildering array of at least six different terms (although we have also generated new ones), where a number of people felt they were dealing with different phenomena. The discussants have noted the interrelationships among the various methods: in particular, method (1&endash; 4 of TABLE l) are seen as different aspects of the same phenomena. Although we may make useful operational distinctions between some of them, the fundamental underlying principles are the same, and these methods represent a spectrum or continuum such that each can merge with the others. Methods 3 4 of TABLE 1 are identical and should have been classified as one. Finally, isoelectric focusing shares several features with the other methods, especially in the transient phase prior to establishment of the pH gradient. The final pH gradient may be regarded as an "equilibrium" or as a form of steady-state electrophoresis, but not as a true thermodynamic equilibrium. We can conclude that we have perhaps come somewhat closer toward the goal of unification of these various forms of electrophoresis and, possibly, other separation methods as well.
REFERENCES
1. HAGLUND, H. 1970. Sci. Tools 17: 1.
2. KOLIN, A. 1958. In Methods of Biochemical Analysis. D. Glick, Ed. Vol. 6: 259. Interscience. New York, N. Y.
3. KOLIN, A. 1955. Proc. Nat. Acad. Sci. USA 41: 101.
4. RILBE, H. This monograph.
5. SCHUMACHER, E. 1957. Helv. Chim. Acta 40: 2322.
6. ORNSTEIN, L. This monograph, discussion following paper by Chrambach et al.
7. GIDDINGS, J. C. & K. DAHLGREN. 1971. Separation Sci. 6: 345.
8. GIDDINGS, J.C. 1969. Separation Sci. 4: 181.
9. ORNSTEIN, L. 1964. Ann. N.Y. Acad. Sci. 121: 321.
10. JOVIN, T. M. This monograph; 1973. Biochemistry 12: 871.
11. RODBARD, D. & A. CHRAMBACH. 1970. Proc. Nat. Acad. Sci. USA 65: 970.
12. KENDALL, J. 1928. Science 67: 163.
13. KONSTANTINOV, B. P. & O. V. OSHURROVA. 1966. Soviet Phys.Tech. Phys. (English Transl.) 11: 693.
14. DAVIS, B. J. 1964. Ann. N.Y. Acad. Sci. 121: 404.