RU2145873C1 - Method of commercial-scale purification of hemoglobin - Google Patents

Abstract

FIELD: hematology. SUBSTANCE: crude 0.1 to 20% hemoglobin solution is passed continuously through anionite or cationite chromatographic column until nearly complete saturation of column with hemoglobins and components with higher affinity, and further passage of solution proceeds to achieve supercharge of column, after which hemoglobins are replaced. Second chromatographic step is anticipated. EFFECT: extended choice of methods for commercial-scale hemoglobin purification. 10 cl, 16 dwg, 9 tbl, 18 ex

Description

 This invention relates to the purification of hemoglobin by chromatography and, more particularly, relates to a chromatographic process for the purification of hemoglobin on a commercial scale, which results in a hemoglobin solution containing at least 99% of the solute. This invention also relates to new hemoglobin preparations having unexpected beneficial properties for use as a blood substitute.

 The development of hemoglobin, the main blood substitute, continues to attract commercial attention, and recent developments have shown that mammalian blood cell hemoglobin, after appropriate modification, such as, for example, intramolecular cross-linking and, in some cases, polymerization, is a promising blood substitute . However, as this potential develops, requirements for hemoglobin purity are constantly increasing. At one time, it was assumed that hemoglobin only needed to be free from stroma, a condition that is achieved by flushing and soft leasing of red blood cells (red blood cells), followed by filtration of the lysate. Subsequently, it was found that the presence of trace amounts of impurities, such as phospholipids, leads to more specific reactions, for example, narrowing of blood vessels. Even after the drug underwent various stages of filtration, it still contained unacceptably high trace amounts of potentially dangerous impurities, such as red blood cell enzymes, modified and various forms of hemoglobin, phospholipids and surface antigens.

 One of the most important problems in the field of blood substitutes, especially in the field of hemoglobin-based oxygen carriers, is the development of effective processes for the rational and economical purification of hemoglobin on a commercial scale. This drug must be competitive with other potential substances used to increase plasma, blood, and oxygen-carrying fluids, which are offered as blood substitutes.

While the degree of separation and analysis time are important characteristics in analytical isolation, the following parameters are decisive parameters in preparative chromatography for use on a commercial or semi-commercial scale:
- the amount of substance allocated per unit of time, with a certain level of purity (throughput), and
- economic characteristics of the process, such as column size, medium, buffer solutions, equipment, recycling and reuse of components, reagents, etc.

 Hemoglobin purification methods that would meet the criteria of a cost-effective production method are not available in the literature.

 A hemoglobin-based blood substitute requires either a single hemoglobin form or, if more than one form is present, careful monitoring of the composition of known hemoglobin forms. Thus, a successful purification method is necessary in order to separate one form of hemoglobin from another, as well as to separate the desired form of hemoglobin from contaminating components of red blood cells, such as red blood cell enzymes, proteins, phospholipids and antigens.

 Chromatographic methods were used to purify hemoglobin solutions. US Pat. No. 4,925,474 to Hsia et al. Describes an affinity chromatography technique for purifying hemoglobin using columns in which a ligand having a predominant chemical affinity for the DPG portion of hemoglobin binds to the stationary phase of the column.

 The ion-exchange chromatography technique was also used to purify hemoglobin. The basic principles of ion exchange chromatography are well known. A mixture of various substances in solution is applied to a suitably prepared ion-exchange column. Each of the substances in the mixture has a different affinity for the chemical reactive groups of the column. By changing the conditions on the column, for example the pH of the solution, individual substances can bind or elute from the column selectively, so as to individually separate one of the substances from the mixture. The use of this technique for the purification of proteins, such as hemoglobin, is economically disadvantageous, except when used on a small scale and analytical work. For the purification of hemoglobin on an industrial scale for use, for example, as an oxygen carrier in resuscitation solutions (blood substitute), this technique for standard use is irrational. The amount of hemoglobin that needs to be absorbed and then eluted from the chromatographic column is so large that the required column sizes become impossibly large and expensive.

 Christensen et al., J. Biochem. Phys. 17 (1988), 143-154, described the chromatographic purification of human hemoglobin. The methodology used is a standard method of ion exchange chromatography and does not provide opportunities for economical production at the industrial level.

 Winslow and Chapman, “Preparing Hemoglobin Solutions on an Experimental Scale,” Methods in Enzymology (Pilot-scale Preparation of Hemoglobin Solutions), vol. 231, p3 (1984) described the preparation of strom-free hemoglobin (SFH), highly purified Ao hemoglobin and cross-linked hemoglobin using outdated human blood as a starting material. SFH was obtained by filtration. Hemoglobin Ao was obtained from SFH on a large scale chromatography using a strong ion exchange medium and a high ionic strength gradient.

 Both Christensen et al. And Winslow and Chapman described the above work at the Letterman Army Institute for Research (LAIR), now known as the Walter Reed Army Institute of Research.

US Pat. No. 5,084,588 to Rausch and Feola (Biopure) describes the use of the standard anion and cation exchange chromatography method for the separation and purification of hemoglobin. In the case of anion exchange chromatography, this patent describes three standard approaches:
a) binding of Hb at elevated pH and elution with a gradient with a decreasing pH or a step gradient with a lower pH value,
b) binding of Hb at high pH, low ionic strength and elution with a salt gradient,
c) a load with pH values at which hemoglobin does not bind to the anion exchanger and does not linger on the column, while impurities (more acidic impurities) linger on the column.

 Methods a) and b) are described in detail, but they are not suitable for production on a large scale due to limitations associated with low capacity, which does not allow to achieve sufficient separation of hemoglobin products. This load is usually 20-30 mg / ml and forces the use of unacceptably large and expensive columns for the purification of hemoglobin on a commercial scale. For example, for a single dose of 50 mg of a hemoglobin-based final oxygen carrier (HBOC), a column of 1.5-2.5 liters would be required.

 Although approach c) seems more rational, it gives in practice unmodified hemoglobin of an adult normal human AO, which, in terms of its chromatographic properties and the content of some basic impurities, such as HbAlc, does not seem satisfactory for practical use.

 Standard cation exchange chromatography approaches for purifying mammalian hemoglobin have similar limitations.

 US Pat. No. 5,084,588 also describes the preparation of bovine hemoglobin solutions in which more than 99.9% of the protein is present in the form of bovine hemoglobin, but data on the determination of protein components are not provided in this patent.

 The object of the present invention is a new chromatographic method for the purification of hemoglobin.

 Another object of the present invention is a chromatographic method that yields, on a commercial scale and in an economically accessible form, hemoglobin in an aqueous solution with a purity of more than 90% from a contaminated aqueous solution that contains at least 60% hemoglobin.

 Another and more specific object of the present invention is to obtain an aqueous solution of hemoglobin or cross-linked hemoglobin in commercial quantities, in which hemoglobin or cross-linked hemoglobin is at least 99% of the solute, and their use in the indicated degree of purity as a blood substitute with previously shown in vivo properties not described.

 The present invention is a method by which an aqueous solution of samples of pre-separated hemoglobin (Hb) with an Hb content of 60 to 99% of the solute is subjected to two-stage ion exchange chromatography. One of the steps is anion exchange chromatography and the other is cation exchange chromatography. These two types of ion exchange chromatography can be performed in any order. A crude solution of pre-separated samples of hemoglobin, for example, HbAo, is the feedstock for the first stage of chromatography, for example anion exchange chromatography, under conditions in which all components of the mixed solute are initially adsorbed on the solid phase of the chromatographic column. Relatively high pH values are required for anion exchange chromatography. The conditions are also selected in such a way that a very effective load on the column is achieved due to the use of the initial solution with low ionic strength, which leads to a situation in which almost all accessible sections of the solid chromatographic phase are filled with substances from the initial solution. The eluate collected at this stage does not contain a protein product. Substitution of HbAo and impurities with low affinity for the column under the selected conditions is then achieved by loading additional volumes of an aqueous initial solution, as described above. Under the conditions reached, referred to here as overload conditions, impurities with a higher affinity for the column are replaced by hemoglobin, and impurities with a lower affinity elute from the column. If the first stage of chromatographic separation is anion exchange chromatography, impurities with a higher acidity than the separated hemoglobin samples are adsorbed on the solid phase of the chromatographic column and are separated from the Hb samples, which appear in the eluate among other more basic impurities. In the case where cation exchange chromatography is chosen as the first stage, the more basic impurities are separated from the more acidic impurities and Hb samples that are in the eluate, due to the higher affinity of the first in the column.

 Then, at the second chromatographic stage of this method of the invention, a second ion-exchange process of partially purified Hb samples contained in the eluate of the first column is carried out under conditions of different pH values. When the eluate from the first column contains hemoglobin samples and more basic components, the second chromatography step is cation exchange chromatography, and vice versa. Again, this contaminated solution is supplied to the second column until a saturating and higher load is reached in order to create overload conditions in the column. Under overload conditions, samples of pre-separated hemoglobin are eluted from the column by substitution with impurities with high affinity in the selected conditions, and then collected in the eluate of the second column with a high degree of purity. The method of the invention can in most cases be called "two-stage self-replacing chromatography."

 As a result, Hb samples elute from the column in an exceptionally pure form, usually at more than 90% purity and with a commercially acceptable yield. Impurities that are more acidic than the separated Hb samples remain bound on one resin of the column, and more basic impurities on the other. Columns can be easily regenerated using traditional cleaning and regenerating methods.

 The most important factor for the successful implementation of this method, especially on a large, industrial scale, are selective, highly loaded columns with respect to impurities, allowing elution of Hb samples.

 When the method of the invention is used to purify human hemoglobin Ao, a new type of human hemoglobin is produced having a composition and properties not previously described in the literature. This hemoglobin preparation is, accordingly, another aspect of the present invention, independent of the purification method described herein. This product has high-value properties and characteristics that were previously undetected and not described in connection with any preparations of human hemoglobin AO and are especially useful when used as a blood substitute base. Among these properties, the complete absence of viral particles, especially the lipid particles of encapsulated viruses, the complete absence of vasoactivity when administered in vivo, the absence of toxicity to endothelial cells upon in vitro incubation, which is evidenced by the reduced release of lactate dehydrogenase in comparison with similar administration of other drugs, are important hemoglobin. In addition, with regard to the standard required hemoglobin properties for use as a blood substitute, such as the absence of endotoxins and other pyrogens, the product of the present invention also has improved properties compared to other hemoglobin preparations.

Accompanying drawings
FIG. 1 is a graphical illustration of an initial ion exchange portion of a preferred method of the present invention.

 FIG. 2 is a similar illustration of a second, cationic moiety of a preferred method of the present invention.

 Figure 3 is an analytical anion exchange chromatogram obtained according to the results of examples 2 and 3 below.

 Figure 4 is an analytical anion exchange chromatogram obtained according to the results of example 4 below.

 FIG. 5 is a graphical representation of the results of Example 8 below.

 FIG. 6 is a copy of a paper strip in the isoelectric focusing analysis of the product of Example 13 below.

 Fig. 7 is a copy of the results of a staining and blotting experiment described in Example 13.

 8 graphically represents the results of Example 14 below.

 FIG. 9 and 9a graphically represent the results of Example 15 below.

 Figure 10 graphically represents the results of example 16 below.

 Replacement chromatography or substitution elution, as it is commonly called, is a well-known method in the field of protein separation. This is one of three methods by which samples of substances are isolated from a chromatographic column. When this method is used to separate the corresponding proteins from a solution of their mixture, proteins with a higher affinity will replace proteins with a lower affinity on the matrix, so that the protein hierarchy in the substitution order in affinity is established from above within the column from entrance to exit. When proteins replace other proteins, a distinct separation line between protein samples is rare. Therefore, the use of this technique to obtain a highly purified solution of a single protein, such as hemoglobin, is unacceptable. If non-protein substances are selected for separation, then the chromatographic matrix will be associated with substances having a very high affinity, and the carrier will be difficult to regenerate for reuse.

 The main industrial use of the proposed method is to obtain purified adult hemoglobin Ao (HbAo) and thus it will be mainly described with the inclusion of literature references. This method, however, is not limited to purification of HbAo, but it is also applicable to other previously separated hemoglobin samples, such as selectively cross-linked hemoglobin derivatives, certain types of bovine hemoglobin, other types of hemoglobin, such as fetal hemoglobin, sickle cell hemoglobin, etc., genetically engineered hemoglobin, etc.

 The method of this invention is extremely cost-effective when used for the purification of HbAo on an industrial scale for the production of blood substitutes. It provides a particularly high-performance column capacity. The method therefore allows the use of relatively small columns to purify relatively large volumes of contaminated hemoglobin solution. It can be carried out periodically or continuously to clean a given batch of hemoglobin solution obtained from red blood cells (red blood cells) according to the standard procedure, solutions usually containing about 80% hemoglobin, but which usually contain a wide range of impurities, both protein-like and non-protein. Moreover, the contaminated hemoglobin solution itself is used as a means of substitution. This excludes the provision and use of various displacers with their inconvenience in service, complexity and high cost.

 A typical general method for preparing a purified HbAo solution, including the methodology of the present invention, begins with normal adult red blood cells (erythrocytes). They are combined and filtered to remove white blood cells. Cells are washed to remove serum proteins and lysed to extract hemoglobin and other cellular components. The lysate is filtered to remove red blood cell membranes (red blood cells), and then the resulting hemoglobin solution is diafiltered and concentrated. The contaminated hemoglobin extract is treated with carbon monoxide to form the CO-Hb complex and heated for pasteurization, thereby inactivating the impurities of the viruses in the solution. This mixture is then centrifuged and filtered to remove further residues and cell debris. After that, the solution is prepared for the implementation of the self-replacement method of the present invention.

 In a preferred embodiment of the invention, the first step of the chromatographic process is anion exchange and the second is cation exchange. The first anion exchange process is carried out at high pH, for example, pH 7-10, and preferably at pH 8.5 - 9.0 and under conditions of low ionic strength, i.e. low conductivity. Accordingly, the conductivity is less than 3 mS (mS - milliSiemens) and preferably less than 1 mS. Such conditions determine the need for a salt-free buffer. Targeted HbAo samples under these conditions initially bind to the column medium along with other substances. When the supply of the column with contaminated HbAo solution continues, those substances that exhibit a higher affinity for the matrix than NbAo, most often more acidic impurities, gradually replace HbAo and more basic impurities from the column. In the end, column overload is achieved, so that all HbAo and more basic impurities are replaced, leaving only more acidic impurities bound. In most cases, this load is in excess of the column load recommended by the manufacturer. Thus, at this stage, chromatographic separation of more acidic impurities from HbAo is achieved.

 The contaminated hemoglobin solution is supplied with a low linear flow rate, not higher than 10 cm / min and preferably 1 cm / min. This speed allows the establishment of kinetic equilibrium on the column. The concentration of the incoming solution is not a decisive factor, but a concentration in the range of 0.1-20%, preferably 2-7%, is suitable.

 The achievement of overload can be controlled by analysis of the solution flowing out of the column - when it does not contain at all or contains only traces of acidic impurities, then the maximum practical load of the column is achieved.

 FIG. 1 of the accompanying drawings illustrates, using diagrams, part of the anion exchange process of the present invention. An ion exchange column having the chemical groups —N + R3 (typically an anion exchange resin) shown in step A of this method in FIG. 1, is supplied with a solution containing a mixture of substances represented by asterisks for more acidic impurities, ellipses are HbAo and rectangles are more basic impurities. At the loading stage, stage B, basically all active chemical groups of N-R3 bind electrostatically to one of the substances contained in the mixture.

 An additional amount of the same solution is applied so that the column load reaches the overload conditions as shown in FIG. 1, stage C. Then, more and more basic impurities (rectangles) and some HbAo (ellipses) are replaced from there by more acidic components (stars), which exhibit a higher affinity for the column under these conditions. Continue to supply an additional amount of this solution until the column is overloaded, reaching stage D shown in FIG. 1, in which acidic substances replace substantially all of the HbAo from the column due to their higher affinity under the selected conditions. In this case, the liquid flowing out of the column is substantially free of acidic impurities. Thus, a sufficiently pure HbAo solution is obtained, but it still contains a certain amount of basic substances.

 The second stage of the preferred method of the present invention uses a cation exchange column, and an eluate is applied to it from the first column under similar overload conditions. Target HbAo and other undesirable, more basic impurities are bound in excess amounts. A partially purified hemoglobin solution from the first column continues to saturate the second column. Again, a relatively low solution transmission rate (10 cm / min or less preferably 1 cm / min) is used, allowing for the establishment of kinetic equilibrium between the solid and liquid phases in the eluate from the first column. A pH of 5-9 is maintained, preferably a pH of 6.5-8.5, and most preferably a pH of 7-8, before the eluate is applied to the cation exchange column. Use a solution with low ionic strength, i.e. specific conductivity less than 3 mS (mS - milliSiemens), preferably less than 1 mS. A specific conductivity of 2.5 mS is only useful if the flow rate is very low and the dilution of the solution used is very high. Overload conditions are used again. Protein-like impurities with high affinity, i.e. more basic than HbAo, replace HbAo on the column, and thus, HbAo elutes and returns in pure form.

 FIG. 2 of the accompanying drawings shows with diagrams, like FIG. 1, a part of the process carried out on a cation exchange resin containing SO3 groups. After adjusting the pH to the appropriate pH, the eluate is applied to the column and, initially, in stage A, the column is saturated with both HbAo (ellipses) and more basic impurities (rectangles) by electrostatic binding to the chemical groups of the column. Although these groups are depicted as sulfo groups, other possibilities, such as carboxyl groups, can equally be chosen. The application of the eluent solution to the column is continued until the overload conditions shown in FIG. 1, stage B, in which the more basic impurities, due to their greater affinity for the column under these conditions, replace HbAo on the column. While continuing to apply the solution until the column is overloaded, as shown in FIG. 2, step C, almost 100% of pure HbAo is replaced in this way and is in solution as an eluant from the column. Accordingly, the higher the column load, the higher the percentage of purified HbAo obtained from the column. The eluate is analyzed in order to determine the presence of basic impurities (rectangles) in it. When HbAo is no longer detected on the column during substitution, the maximum load is determined at this point.

 As used herein, the term “more basic impurities” refers to those impurities that elute from the column at a higher pH than is required for elution of HbAo. In contrast, the term "more acidic impurities" refers to those impurities that elute from the column at lower pH than is required for elution of HbAo. In a preferred method according to this invention, in which anion exchange chromatography is first performed, more basic impurities elute before HbAo, and more acidic impurities elute after HbAo. Depending on the use of the ion exchange medium for chromatography, various elution curves may be considered with slight deviations in the elution order.

 For both the first and second columns, the flow rate of the crude hemoglobin solution should be optimized for the selective concentration of this solution. A linear flow rate of 1 cm / min or less is optimal for hemoglobin solutions at a concentration of 2-7%. A flow rate above 2 cm / min can be successfully used to separate more dilute solutions, for example, less than 1% Hb. Ha separation by ion exchange chromatography at low load is usually not affected by the linear flow rate, however, for the method of the present invention, the kinetic equilibrium in the chromatography process is highly dependent on the flow rate. The concentration of the applied solution is not critical at any stage.

 Using the replacement chromatography method of the present invention, using first anion-exchange chromatography to remove more acidic components and then cation-exchange chromatography to remove more basic impurities, it is possible to achieve a high yield of exceptionally pure HbAo, essentially completely free of noticeable amounts of other proteins, such as serum and membrane proteins , on an industrial scale. The purity according to analytical anion exchange chromatography is 99% and usually more than 99.5%. The efficiency of the process is at least an order of magnitude higher than with the conventional chromatographic approach. In addition to removing protein-like impurities, the method of the present invention also allows extremely complete removal of modified and other undesirable forms of hemoglobin, giving highly pure HbAo (99.5%). In addition, as described in more detail below, virtually complete removal of encapsulated viral particles, both active and inactive, can be achieved.

 The advantages of the scale of the method of the present invention compared with a method using standard adsorption-elution standard ion exchange chromatography are shown in Table. 1. The numbers in the "self-substitution" column are obtained from the working example of the method of the present invention described below. The numbers in the adsorption-elution column are obtained from the aforementioned article by Winslow et al.

 From the analysis of these figures, it is clear that the self-chromatography by the method of the present invention can be carried out faster with sufficiently small columns. It can be carried out at low pressure and ambient temperatures. The effective capacity can be achieved 10-20 times higher than the capacity reported for standard chromatographic methods, making it possible to use columns with a relatively small volume. This leads to a decrease in water demand, reduced waste and the need for buffer solutions, all this significantly improves the industrial economy of this method.

 The purified hemoglobin solution, which is an eluate of a cation exchange column, can be diafiltered into a selected buffer and adjusted to the required concentration of HbAo.

 In the method of the present invention, the choice of ion-exchange gel material as the stationary phase is not critical. The choice can be made successfully within the modern level of science. Any of the usual commercially available gels of this type can be used here, provided that they can be successfully derivatized for use in the ion exchange method under conditions of selective affinity for the substances to be separated. General classes of carriers for ion exchange agents include macroporous, macrocellular and non-porous divenylbenzene - cross-linked polystyrenes or polymethacrylates, polyalkyleneamines, celluloses, agaroses, silica, ceramics, glass, alumina, metal particles, etc. Commercially available carriers include those sold under the trademarks POROS, Fractogel, Huper D, Dowex, Amberlite, Duolite, Bio-Rex, Chelex, Sephadex, Sepharose, Toyopearl, Macro-prep, Bio-protocol, Accell, Monotypes, etc. Functional groups tied to such carriers for ion exchange goals, include sulfonate, sulfopropyl, carboxymethyl, carboxylate, phospho, quaternary amine, quaternary aminoethyl, dimethylaminoethyl, polyethyleneamine, trimethylaminoethyl, diethylaminoethyl, aspartamide, etc.

 The method of the present invention can also work to obtain a hemoglobin solution of at least 99% purity and unexpectedly free of viral particles, active or inactive, especially lipid encapsulated viruses, such as HIV, herpes, simplex virus (HSV), bovine diarrhea virus, which in many cases, is a model for human hepatitis C virus.

 These are some of the most serious and dangerous viral contaminants of the blood. This is a very important additional achievement that was obtained using the present invention, and this achievement could not be predicted before this method was tested and the results were analyzed. Although the preferred process of this invention includes a pasteurization step for inactivating viruses prior to purification, as described above, it is particularly valuable to achieve the ability to remove viral particles as part of the purification process. The method makes it possible to remove from the blood substitute materials various active viral particles that can be preserved during pasteurization, and any residual unactivated viral particles. All this can be achieved without the use of any additional specific stage of removal of viral particles and without the addition of any special reagents or something like that to remove viruses. Exceptionally complete removal of the lipid encapsulated viruses can be achieved. A noticeable decrease in the number of other viruses was also achieved.

 As far as the applicant knows, the hemoglobin preparations of the present invention have a higher degree of purity than previously described preparations. Perhaps due to their unusually high degree of purity or, possibly, due to other, not yet fully explained factors, the products of this invention exhibit unexpected and highly useful properties not previously described for hemoglobin preparations, at least for preparations produced on a large scale.

 More specifically, the products of the present invention, HbAo, are free of all red blood cell membrane (RBC) proteins, as shown by isoelectric focusing and immunoblot analysis. Traces of human serum albumin (HSA) of less than 4 μg and usually 2-4 μg of HSA / 100 mg Hb are present in the preparations as detected by immunoabsorption methods. The only other detectable proteins that are sometimes found in the products of the invention are fragments of the hemoglobin alpha and beta globin chains, and they do not exceed 1 μg per 100 mg of hemoglobin. With this extremely high level of purity, the product of the present invention is unusual and exhibits unexpected properties described below. One of them is a decrease or absence of vasoconstrictor activity. Another property is the absence of toxicity to cultured endothelial cells.

 The highly purified hemoglobins previously described, both cross-linked and non-cross-linked, gave a pressor response upon infusion [1,2]. Since the products were apparently quite refined, existing theories believed that the proper properties of hemoglobin itself, which is responsible for this, must have appeared here. In particular, it was suggested that hemoglobin binds with nitric oxide, which causes this effect. Nitric oxide, also called an endothelium-forming relaxing factor, is a powerful vasodilator synthesized by endothelial cells using nitric oxide synthase as an intermediary. Since the hemoglobin molecule will bind nitric oxide with a high degree of affinity, it has been postulated and it is generally accepted that the intravenous administration of a hemoglobin solution, especially non-crosslinked hemoglobin solutions, will remove nitric oxide and cause a hypertensive effect with narrowing of blood vessels.

 The discovery that the products of the present invention do not exhibit vasopressor activity in models of isovolemic metabolism and hemorrhagic shock was also a complete surprise, as it contradicts modern theories and opinions regarding the properties inherent to hemoglobin. The present invention suggests that this is not the proper property of hemoglobin, which caused its vasoconstrictor activity (as previously reported), but at least partially related to impurities that are present in very small quantities and which have not been successfully removed before, despite to reports of a very high level of hemoglobin purity as previously described.

 The benefits of the lack of vasopressor activity of the products of the present invention are retained in the cross-linked and polymerized hemoglobins obtained from the products of the present invention, for example, using the method described in US application N 08/231945, April 21, 1994.

 In addition, the product of the present invention exhibits a higher degree of reduction in toxicity to endothelial cells when administered in vivo than the previously described hemoglobins. Studies with previously described solutions of oxyhemoglobin, including chromatographically pure hemoglobins, showed the presence of cell damage, as indicated by a marked increase in lactate dehydrogenase in the endothelial cells of the culture medium compared to its level in the medium of the control endothelial cell culture. Lactate dehydrogenase is a marker of cell damage. It was again postulated and recognized that this toxicity is an inherent property of oxyhemoglobin, since the tested solutions contained highly purified oxyhemoglobin [3].

 It was unexpectedly discovered that the hemoglobin according to the present invention is not a mediator in the release of lactate dehydrogenase. In contrast to the hemoglobin of the present invention, when incubating endothelial cells with hemoglobin solutions not containing stroma and purified by other chromatographic methods, a significant increase in the release of lactate dehydrogenase was noted.

 In general, it is important and obvious that any blood substitute and any component or raw material for a blood substitute must be sufficiently purified so that it is free from harmful impurities. Cell stroma, endotoxins and other pyrogens, non-hemoglobin proteins, phospholipids, etc., were previously recognized as harmful impurities. It was reported that former blood substitutes are “mostly free” of one or more of these impurities. However, the term “mostly free” is a relative subjective term, the interpretation and limitations of which depend on the method for determining the relevant parameters, and it has been used to describe the prior art. As a result of improved cleaning and analysis techniques, messages began to appear about blood substitutes and hemoglobin preparations of an increasingly high purity level. For example, previous scientific literature contains reports of hemoglobin solutions in which more than 99.9% of the protein is in one form or another of hemoglobin (see the aforementioned US Pat. No. 5,084,588, Rausch et al.). In accordance with the present invention, it is postulated that even such purity is not sufficient for successful blood substitutes, because the nature of the remaining impurities can cause vasoconstrictive effects and manifestations of cellular toxicity.

 It can be shown that stroma-free hemoglobin, which is obtained from the Walter Reed Army Research Institute (WRAIR), commonly recognized as the industry standard for purity, contains a large number (over 10) of various contaminating proteins, many of which are present in amounts less than 0, 01% of the total protein. Although, at first glance, this is a very small amount, however, hemoglobin as a blood substitute is administered to the subject in relatively large quantities, and therefore their danger is determined by the total amount of the injected mixture, and not by their concentration in the initial solution. One unit of hemoglobin-based blood substitute solution will contain 25-50 hemoglobin, so that 0.01% of impurities will be 2.5-5 mg. Among the wide variety of proteins present, there may be one or two that are extremely toxic at this level. The present invention provides hemoglobin preparations in which the content of non-hemoglobin proteins is lower than 0.5%, preferably lower than 0.1%, and it is likely, but not necessary, that as a result they exhibit a low vasoconstrictor effect and little activity with respect to endothelial cells. The invention also provides a method by which hemoglobin solutions can be obtained at an industrial level, as described above. Another important advantage of the products of the present invention is their ability to reduce procoagulant activity. Apparently, the content of agents that potentiate blood coagulation is very significantly reduced in these products compared to other available stromal preparations of hemoglobin that were previously described.

 The following non-limiting examples are described below to illustrate the invention.

EXAMPLE 1. Obtaining a crude hemoglobin lysate
Ninety-six units (approximately 30 L) of packaged red blood cells (RBCs) of the same type are combined and diluted with 24 L of 0.9% saline. This mixture is then filtered through 20 micron and 8 micron polypropylene filters to remove residual white blood cells.

After the white blood cells are filtered off, the RBCs are washed with 6 volumes of 0.9% saline to a constant volume by diafiltration using 3.5 m ² , 0.3 micron sepracor polyethersulfone hollow fiber membranes. The wash buffer is then replaced with 50 mmol Tris lysis buffer and RBCs are gradually dissolved in 6 volumes of this buffer.

The crude lysate is collected and concentrated from about 2.5% total hemoglobin (THb) to about 9.0% THb using Millipore (PT) membranes separating substances with a molecular weight of 30H. A container with a concentrated substance is saturated with CO to convert hemoglobin to the COHb form. The lysate is pasteurized at 62 + -2 ^° C. for 10 hours in a jacketed container. The pasteurized lysate is cooled and then centrifuged. Further deep filtration through 0.8 micron Millipore filters yields a pasteurized lysate for diafiltration on another 30K PTTK Millinor membrane. The material is diafiltered with a 5 mm Tris buffer at pH 8, until its conductivity is less than 0.3 mS (mS - milliSiemens), and pH - 8.9 + -0.2. Then it is diluted to 4.5-5% THb, and from that moment it is ready for subsequent chromatographic purification.

EXAMPLE 2. Self-replacement chromatography on anion exchange resin
A 1x10 cm column filled with Per Septive Rros HQ-50 anion exchange resin was washed with 4 volumes of 1N NaCl and equilibrated with 5 mm Tris buffer at pH 8.8. After equilibration, 50 ml of 3% (3g / 100ml) of the crude hemoglobin lysate (β 85% HbAo) in 5 mmol of Tris, prepared as described in Example 1, pH 8.8 at a rate of 0.78 ml / min (1 cm) is loaded onto the column. / min). This leads to the final overload of the column - 200 mg of solute per 1 ml of resin when compared with a standard capacity of 50 mg / ml (declared by the manufacturer). The eluate is collected. The column is then washed with 2 volumes of 5 mm Tris buffer, pH 8.8, and this eluate is combined with the eluate collected during application. The column was then washed with 1N NaCl to wash out the residual proteins, and this wash eluate was discarded. The column was regenerated using 3 volumes of 1M NaCl / HCl and 4 volumes of 1M NaOH. The hemoglobin eluate from the anion exchange resin was analyzed on an analytical anion exchange column using a pH gradient, as described in more detail below in Example 11. As shown, it contains 95-96% HbAo (FIG. 3), with a return of 90% HbAo applied to the column .

EXAMPLE 3. Self-replacement chromatography on anion exchange resin
A column (1 x 10 cm) filled with Merck Fractogel TMAE was washed with 4 volumes of 1 N NaCI and equilibrated with a 5 mm Tris buffer, pH 8.8. After equilibration, 50 ml of 3% (3g / 100ml) of the crude hemoglobin lysate (85% HbAo) in 5 mmol of Tris buffer prepared as described in Example 1, pH 8.8, at a rate of 0.78 ml / min ( 1 cm / min). This similarly leads to the final overload of the column. The eluate is collected. The column is then washed with 2 volumes of 5 mm Tris buffer, pH 8.8, and this eluate is combined with the eluate collected during application. The column is then washed with 1N NaCl to wash out the residual proteins and the wash eluate is discarded. The column is regenerated using 3 volumes of 0.5N HCl and 4 volumes of 1M NaOH. The hemoglobin eluate from the anion exchange resin is analyzed on an analytical anion exchange column using a pH gradient as described in more detail below in Example 11. It determines about 95-96% HbAo, as shown in FIG. 3, and 73% of the HbAo deposited on the column is returned. .

EXAMPLE 4. Self-replacement chromatography on a cation exchange resin
A 1x10 cm column filled with Perseptive Poros HS-50 is washed with 4 volumes of 1N NaCl and equilibrated with 5 mm Tris buffer, pH 7.5. After equilibration, 120 ml of 2% (2 g / 100 ml) of hemoglobin β95% HbAo purified on anion exchange resin in 5 ml of Tris buffer pH 7.5, prepared as described in Examples 2 and 3, is applied at a rate of 0.78 ml / min (1 cm / min). This finally leads to a significant overload of the column by about 5 times compared with the recommendation of the manufacturer. This eluate is collected. The column is then washed with 2 volumes of 5 mmol Tris buffer, pH 7.5, and this eluate is combined with the eluate collected upon application. The column is then washed with 1N NaCl to wash out the residual proteins and this solution is discarded. The column is regenerated using 3 volumes of 0.5 N HCl and 4 volumes of 1 M NaOH. The hemoglobin eluate from the cation exchange resin was analyzed on an analytical anion exchange column using a pH gradient, and the content was greater than 99% HbAo, as shown in FIG. 4, with a return of about 90% of the HbAo deposited on the column.

 EXAMPLE 5

 The effect of conductivity on self-replacement chromatography on anion exchange resin was studied (first stage).

 Effective replacement chromatography to a large extent depends on the ionic strength of the buffer and / or sample, which is determined by the conductivity. The experiment, the results of which are shown below in table. 2 was carried out on the basis of the crude lysate of Example 1 for the Poros HQ-50 anion exchange resin from PerSeptive Biosystems Inc. The column eluate was analyzed by analytical anion exchange chromatography. All fractions obtained in experiment 1 were free of detectable acidic impurities, while even the first fractions in experiment 2 were contaminated with acidic impurities.

EXAMPLE 6. Yield as a function of protein load
By overloading the column, a higher recovery of purified HbAo is achieved. The experiments described below in table. 3 show a very high overload of a 10 ml column with a Poros HS-50 cation exchange resin, which guarantees the adsorption of impurities at all binding sites, allowing for higher recovery of purified HbAo. Overload in the first experiment was 13.5 times more than the manufacturer's recommendations (308 mg protein / ml). The overload in the second experiment was about 5 times higher compared with the manufacturer's recommendation. High overload leads to a significant improvement in HbAo recovery from the column.

EXAMPLE 7. Comparison between small and large columns
The size exclusion chromatography process of the present invention can be used on large and small columns. Using sequentially the Rros HQ-50 anion exchange resin and the HS-50 cation exchange resin and columns of 15 and 10 liters, respectively, a comparison was made with a 10 ml column. The solution entering the Poros HQ-50 column was prepared as described in Example 1. The eluate from the HQ-50 column was directed to the HS-50 column. Tab. 4 shows the conditions and results of using a Poros HQ-50 column. Tab. 5 shows the results of using a Poros HS-50 column. The results show that process steps can be scaled to sizes of at least 10-15 liter columns for commercial use.

EXAMPLE 8. Analysis of the crude hemoglobin lysate and purified hemoglobin by isoelectric focusing (IEF)
An isoelectric focusing (IEF) analysis is performed using a basic agarose gel in which a stable pH gradient is achieved using commercially available ampholytes. Such ampholytes are used, 90% of which had a pH in the range of 5-8 and 10% in the pH range of 3.5-10. The agarose in the gel was at a final concentration of 1.25% and ampholyte was present at a final concentration of about 2%. In order to determine purity, the gel was heavily overloaded (1 mg protein per line). Electrophoresis was performed at low amperage to ensure the movement of the protein in a uniform straight strip. After reaching a steady state, the voltage and current were increased for a short period to obtain a sharper band. After electrophoresis, the protein on an agaroea gel was fixed using trichloroacetic acid and a solution of sulfosalicylic acid. After the fixation step, the gel was stained with Comasie-Blue to develop protein bands. To study the gel, it is dried in air and then a laser densitometer is used to obtain traces of electrofocused lines corresponding to the crude hemoglobin lysate and purified hemoglobin. In FIG. 5 shows a comparison between the IEF of the crude hemoglobin lysate obtained as described in Example 1 and the purified HbAo obtained according to the invention. The hemoglobin solution was purified using a Poros HQ-50 column, as described in Example 2, and a Poros HS-50 column, as described in Example 4, in series.

EXAMPLE 9. Immunoassay of a crude hemoglobin lysate and purified hemoglobin
An immunostaining technique was used to determine the amount of various contaminants in the crude hemoglobin lysate from Example 1 and the purified hemoglobin obtained according to the method of the invention. After electrofocusing the proteins on an agarose gel, the proteins were transferred onto nitrocellulose paper using capillary absorption. The remaining binding sites on paper were blocked by incubation of the stain in a solution of hydrolyzed fish gelatin. After incubation, the spot was washed with Tris-buffered saline (TBS) and then incubated with the primary antibody. (The primary antibody is the antibody against which the sample was tested, i.e., anti-human serum albumin (anti-HSA), which should be used to determine HSA in hemoglobin solutions). After this incubation, the stain was washed with TBS and then incubated with a second antibody associated with a labeling enzyme (e.g., horseradish peroxidase), which is an anti-primary antibody. After incubation with the secondary antibody, the spot was washed again with TBS and then the substrate for the marking enzyme was added. A colored precipitate will appear where the secondary antibody is bound to the primary antibody, which is associated with its antigen, thus indicating the amount and position of the antigen with respect to the primary antigen on the IEF gel.

 The results are shown in table. 6 at the end of the description, in which “+++++++” indicates a strong signal, “+” indicates a barely detectable signal, and “-” indicates that the signal is not detected.

 a) Using HSA standards on immunoblots, the amount of HSA present in purified HbAo was calculated - less than 5 picograms / mg Hb (less than 50 ppm particles per million).

 b) Traces of contaminants with pI β5,2 that mutually react with the anti-RBC antibody were determined. Other researchers also report similar proteins in their purified Hb solutions - see Christensen et al., Cit. lit.

 c) Anti-HSA antibodies present in anti-human plasma are determined by the presence of HSA in purified HbAo.

EXAMPLE 10. Human serum albumin - immunoassay (HSA-ID)
To determine the amount of HSA (human serum albumin) present in hemoglobin solutions, a chromatographic method is used. This method uses an antibody (anti-HSA) covalently linked to a column matrix. When a sample is applied to a column, everything except antibody antigen (HSA) passes through the application / washing step. When using an elution buffer, bound antigen (HSA) was eluted from the column and the peak area was used to count the antigen present in the sample.

 The ImmunoDetection cartridge (ID) used for the determination was obtained from PerSeptive Biosystems. The flow rate used for determination was 2 ml / min. A cartridge-shaped column (D 1.6mm x L 26 mm) was equilibrated with buffer (10 mm phosphate buffer + 300 mmol NaCl, pH 7.2) and then 50 μl of the sample was applied. A loading buffer was used to wash the column for 0.25 minutes, then an elution buffer (for 3.5 minutes) was used to elute the bound HSA from the column. The eluate was monitored on a Beckman spectrophotometer at 220 nm of 414 nm. The column was then balanced again with loading buffer for the next application. The amount of HSA present in purified hemoglobin, calculated using this method, was not more than 0.0005% by weight, based on hemoglobin.

EXAMPLE 11. Analytical anion exchange chromatography
The chromatographic methods used to analyze the purity of hemoglobin in the products in examples 2 and 4 were as follows. The definition was based on the method described by Muisman and Dozy (4). Samples were applied to the anion exchange column at high pH (pH 8.5) so that all proteins bind to the column, then a pH gradient of 8.5 to 6.5 was used to sequentially elute the proteins from the column. Using this chromatographic analysis, very close proteins (for example, different types of hemoglobin) can be separated. For this determination, a 4.6 mm x 100 mm analytical anion exchange column from PerSeptive Biosystems was used. The buffers used were: A) 25 mmol. Tris + 25 mmol. Bis-tris, pH 8.5; and B) 25 mmolar tris + 25 mmol. Bis-tris, pH 6.5. The determination was carried out at a flow rate of 5 ml / min. After the column was equilibrated with buffer B, 10 μl of the sample (conc. 10 mg THb / ml) was applied to the column and gradient elution was started (100% buffer A to 100% buffer B through 25 column volumes). The eluate was monitored in UV light at 280 nm to identify protein peaks. FIG. 3 is a chromatographic curve of the anion exchange purification of hemoglobin (product of Example 2), and FIG. 4 - purified HbAo (product of example 4) when analyzed by this method.

 This method is a standard assay for checking the purity of hemoglobin solutions. It has been found that the HbAo purified according to the invention is free of all contaminants observed in strom-free hemoglobin. Sometimes a small shoulder is observed at the peak of HbAo, which is attributed to the oxidized form of HbAo (Met HbAo). The purity in accordance with this invention, determined by this analytical method, was usually more than 99.5%.

EXAMPLE 12. Removal of viruses
The method of the invention was tested for its ability to remove viruses from crude hemoglobin.

To the samples of the crude lysate obtained as described in Example 1, and to the samples of a partially purified hemoglobin solution leaving the anion column, i.e., to the products of Examples 2 and 3, the test quantities of the following viruses were added in active form:
Poliovirus type 1
Human Immunodeficiency Virus (HIV)
Bovine diarrhea virus (which is a model of human hepatitis virus)
Herpes virus type I (HSV-1).

 Aliquots of the eluates from the respective columns were added to the test cell systems. In cases of polyvirus, herpes virus, and bovine diarrhea virus, the active virus reacts with the test cells, giving hemolysis zones. Well-known standard tests were carried out. In the case of the HIV virus, aliquots were added to the tissue culture medium. The amount of virus present was determined by the change in cell line growth. This change is referred to as TCID50 (the dose at which a tissue culture is infected by 50%). The results are shown below in table. 7.

 These results show a slight decrease in the amount of poliovirus when using both the anion-exchange self-substitution and cation-exchange self-substitution steps of the present invention, but a significant decrease in the virus in the case of human immunodeficiency virus (HIV), herpes virus (HSV) and bovine diarrhea virus after anion exchange chromatography and to a lesser extent - after cation exchange chromatography. This was a completely unpredictable result, which gave significant advantages when using the present invention.

EXAMPLE 13. Purity analysis by isoelectric focusing method and Western Blotting method
Isoelectric focusing (lEF) experiments were performed using the Saravis and Zamcheck method [5].

 IEF gels were obtained using agarose (1% final concentration), and a pH gradient was created using ampholytes (Pharmacia Biotech Inc., Piscataway, NY) so that 90% of the ampholytes had a pH in the range of 5-8 and 10% -in the range of 3.5-10. Typically, 10 μl of 100 mg / ml Hb solutions was applied to an IEF gel. After focusing the proteins, the gel was fixed with 10% trichloroacetic acid (TCA), dried, stained with Coomassie-Blue A3U. Typically, the total protein load per line was 1 mg.

 An isoelectric focusing (IEF) analysis was performed to compare the purity of SFH (Strom-free hemoglobin) and four (4) different NbAo samples purified according to the invention. 50 μg of total Hb or 1 ml of total Hb was applied to each pass. The results are shown in FIG. 6. Line 20 represents stroma-free hemoglobin at a full Hb load of 50 μg, lines 22, 24, 26 and 28 represent samples of the hemoglobin purified according to the invention at the same load. Line 30 represents SFH at 1 mg total load and lines 32, 34, 36 and 38 represent samples of Hb purified according to the invention at the same load. At a load of 50 μg of the total Hb sample, purified HbAo and SFH of the same purity were obtained. However, with a load of 1 mg of total hemoglobin, numerous red blood cell proteins, plasma proteins, or membrane proteins found in the SFH line are absent in the HbAo line. A very weak P1 protein band of about 5.2 was found for the purified HbAo line, possibly due to fragments of the globin chain. The bands above the HbAo band in the 50 μg line of the sample are due to the more positively charged Methb, which is formed during electrofocusing.

 After isoelectric focusing (IEF), the Western Blotting method was used, which included antibody staining (immunostaining), which facilitated a thorough study of the purity of the product of this invention. Thus, proteins from the IEF gel were transferred onto nitrocellulose paper using polymer absorption. The remaining binding sites on nitrocellulose paper were blocked by incubating the stain with hydrolyzed fish gelatin (Hipure Liquid Gelatin, Norland Products Inc., New Brunswick, N.Y.). The spot was then incubated with rabbit antibody in relation to human plasma. (Dako Corporation, Santa Barbara, CA, Lot # 010) (antibodies were diluted 1: 500 in Tris-buffered saline. After incubation overnight, the spot was washed with Tris-buffered saline (TBS) containing 1% fish gelatin, then incubated with a complex of goat anti-rabbit antibody with horseradish IgG peroxidase (HRP) (diluted 1: 1000 in TBS). After incubation for 1 hour, the spot was washed again with TBS containing 1% fish gelatin. The antigen-antibody-HPR complex was developed, using 4-chloro-1-naphthol (30 mg / 20 ml methanol) and hydrogen peroxide (50 μl 100 ml TBS). Zone purple precipitate indicates the presence of antigen-specific plasma proteins.

 Duplicate samples of stroma-free hemoglobin, Hb of the invention, lEF-Hb of the invention, purified hemoglobin from the Walter Reed Army Institute of Research (WRAIR) (purified as described in the aforementioned article by Hinslow et al.) And purified Hb from Sigma Corporation (which is described in their current catalog of diagnostic reagents as "Hemoglobin AO, reference H 0267") focused on IEF gel. Half of the gel was stained with Coomassie-Blue, the other half was transferred onto nitrocellulose paper. The spot was then stained for the manifestation of complexes of human plasma - an antibody of human plasma.

 The results are presented in Fig.7. In this drawing, line 40 is stroma-free IEF hemoglobin stained with Coomassie-Blue; lines 42, 44, 46 relate respectively to the HbAo line of the invention, the Hb of the WRAIR Institute and the Hb of Sigma Corporation. Line 48 is Western Blot, a human plasma antibody test for stroma-free hemoglobin, and lines 50, 52, and 54 relate to HbAo according to the invention, Hb of the WRAIR Institute and Hb Sigma Corporation, respectively.

 Significant amounts of plasma proteins were determined in SFH. In contrast, only very minor traces of human serum (HSA) were identified in the product line of the invention. No other plasma proteins have been identified in the product line of the invention. No detectable HSA level was found in HbAo WRAIR, but many other plasma proteins with bands lower and higher than HbAo were detected. The HbAo Sigma standard contains many contaminants originating from plasma, including a significant concentration of HSA.

EXAMPLE 14. The study of acute hypovolemia in conscious rats
The product of the present invention was tested for vasopressor activity.

 Sprague-Dawby rats, randomly divided into groups of 6 (n = 6), were allowed to acclimatize for 4 days, during which cannulation was performed using the Tabata and Chang technique (6). Appropriately cannulated animals received intravenous heparin, 150 IU / kg. Each arterial cannula was connected to a Stratham pressure transducer to determine blood pressure and heart rate. These parameters were constantly recorded on the Grass polygraph recorder. After registering a stable baseline, a lethal hemorrhagic shock was caused by the Wiggers method, which involves removing 67% of the total blood volume through another arterial catheter at 0.2 ml / min in 2 stages. Immediately after the shock, each rat received the test material at a rate of 0.5 ml / min, i.e. homologous blood, HbAo according to the invention (HbAo) or cross-linked HbAo according to the invention, which react with o-raffinose, a cross-linking reagent, as described in the aforementioned application for US patent 08 / 231.945 dated April 21, 1994 (volume x 3). Mean arterial pressure (MAP) and heart rate were recorded within 30 minutes after the end of the infusion. Each rat was then returned to a separate cage and body weight, hematocrit and survival were monitored for 14 days.

 The results are presented graphically in FIG. 8, which shows the graphical dependence of MAP on time.

 The data show that the product of the invention is a highly purified form of Ao hemoglobin, and has the same efficacy as complete blood in restoring blood pressure after hemorrhagic shock in conscious rats. There were no differences in the restoration of blood pressure and heart rate in the test groups between total blood and HbAo.

Example 15. Hemodynamic effects of isovolemic exchange blood transfusion in rats
The HbAo of the present invention was tested to evaluate the hemodynamic effect of isovolemic exchange transfusion of blood in conscious rats. The external iliac vein and rat artery were cannulated through the femoral region, see Keipert and Chang (8). After 24 hours of recovery, conscious animals were subjected to a continuous 50% isovolemic exchange blood transfusion. During an exchange blood transfusion, the volume of blood removed through the iliac artery was replaced with the test solution through the iliac vein at a rate of 0.5 ml / min, using an infusion pump operating at a constant speed. Blood pressure and pulse rate were constantly measured: 1) 30 minutes before the exchange, 2) during the exchange, 3) 2-3 hours after the exchange. Blood pressure and heart rate values were calculated based on the blood pressure curve.

 Human rat serum albumin, HSA, was transfused into one group of rats (number 4). For another group of rats (number 6), the stroma-free hemoglobin (SFH) was the test solution. For the third group of rats (number 8), the test solution was HbAo, a product of this invention.

 The results are presented in graphical form in FIG. 9 and 9a.

 For all test solutions, the blood pressure value increased by 10-20 mm Hg during the exchange transfusion procedure. With a control solution of albumin, the blood pressure dropped 15 mm Hg below the baseline during rest after the exchange and remained at a low level throughout the period after the exchange, approximately 2.5 hours (Fig. 9). These observations are consistent with previously published observations by Keipert and Chang (8).

 With stroma-free hemoglobin, the blood pressure value fell down to the main level at the end of the exchange period, but rose during the period after the exchange a maximum of 15 mm Hg above the main level within 30-40 minutes after the exchange (Fig. 9a).

 In contrast, for the HbAo of the present invention, the blood pressure after exchange returned to the baseline and increased in the post-exchange period by no more than 5 mm Hg, thereby keeping the MAP (average blood pressure) within the normal range.

 The results of this study show that the HbAo of the present invention lacks strong vaso-activity associated with exchange transfusion, compared with hemoglobin not containing stroma.

 Example 16

In vitro tissue culture assay for measuring oxidative damage to endothelial cells pretreated with hemoglobin preparations
The property of HbAo according to the present invention has been studied to cause oxidative damage to endothelial cells in culture. The release of lactate dehydrogenase activity in the medium is a marker of cell damage and cytotoxicity.

Pig thoracic aortic endothelial cells were cultured prior to fusion, subcultured in a ratio of 1: 3 in a repeating form. Within 48 hours, 7 cell repeats were used for the experiment. Cell damage was determined by measuring the amount of lactate dehydrogenase released from the cells into the medium. Before hemoglobin solutions were added, the medium was removed and the cells were washed twice with physiological phosphate buffered saline. Cells were incubated with hemoglobins (250 micromoles of heme) in a complete EGM medium from 6 to 24 hours at 37 ^o C (n = 4-6). Only cells incubated in complete EGM medium were used as a negative control. At the end of the incubation period, 50 microliters (μl) of medium were taken from each well and lactate dehydrogenase activity was determined by the Sigma determination method. Lactate dehydrogenase activity in each cell was compared with the activity that appears in the cells after adding 2% Triton X-100 (complete release of ZDH), and calculated as a percentage of the total release.

 After 6 hours of incubation, a significant increase in the release of lactate dehydrogenase was observed in the presence of hemoglobin not containing SFH stroma (74.4 ± 4.2%, p less than 0.005). A similar but smaller increase was observed with purified hemoglobin from the Walter Reed Army Institute of Research. No increase was observed with the purified HbAo according to this invention (39.8 ± 2.3%) when compared with the control (37.5 ± 2.2%). In the presence of stroma-free hemoglobin, the release of lactate dehydrogenase increases further after 24 hours of incubation (99.6 + 3.5%), while no changes were observed with the HbAo of this invention (41.6 ± 4.6%) in compared with control (34.1 ± 2.6%).

 Incubation of cultured endothelial cells with HbAo obtained by the method described in this invention did not cause lactate dehydrogenase activity of the cell damage marker. Therefore, cytotoxicity towards cultured cells is not a property of hemoglobin solutions, and HbAo purification removes the cytotoxic components present in stroma-free hemoglobin.

EXAMPLE 17. Removal of procoagulant activity
The product of the present invention has been tested to contain agents that potentiate blood coagulation. The method of Biessels et al. (9) was used.

 The guinea pigs were anesthetized with HyponormTM and a sylastic cannula was inserted into the carotid artery. After cannulation, 30% of the blood volume was taken (calculated as 2% of body weight) and replaced with an equal volume of the test solution. A 2% human albumin solution was infused at 2.5 ml / hr to maintain access to the cannula. For the beginning of the experiment, the beginning of albumin infusion was taken. Fibrinopeptide A (FPA), which is known to be formed from fibrinogen during blood coagulation, was measured before and after the administration of factor Xa (8–9 μg / kg) in a dose that alone does not cause coagulation. Coagulation samples were collected in an anticoagulant mixture of heparin (1000 u / ml), Trasylol (1000 u / ml) and sodium citrate (3.8%), separated, and the plasma was kept on ice.

 Plasma samples were processed as described for the determination of FPA by the method of Leesma, 1984 (10) and the resulting FPA was measured by the method of RIA, previously described in Gerrits (ll). Plasma proteins were precipitated with an equal volume of 40% PEG 6000 in phosphate buffered saline. The supernatant was heated in a boiling bath, centrifuged and stored for analysis. The results are shown below in table. 8.

 These data show that the product of the present invention essentially has no coagulant activity when compared with stroma-free hemoglobin. It is believed that such procoagulant activity is mainly associated with contamination of test solutions with lipids or phospholipids.

EXAMPLE 18. Determination of the amount of impurities of human whey proteins by ELISA
The amount of whey protein impurities in the purified hemoglobin preparation of the present invention was measured and compared with their amount in other available hemoglobin preparations by ELISA sandwich method.

 In the present example, an ELISA sandwich method was used according to the general methods and working methods of Butler, J.E. (12).

4 microtiter plates 1 mm in size of Dynatech 96 well were coated as solid phase with goat serum absorbed by rabbit and anti-human serum complex antibodies (Dako Corporation, Santa Barbara, CA, Lot # 72) diluted 1:40 with Tris-buffered saline. After an incubation of 2 hours at 37 ^° C., the plates were washed with Tris-buffered saline (TBS) and the remaining free portions of the plate were blocked by incubation with hydrolyzed fish gelatin (Hipure Liquid Gelatin, Norland Products Inc., Hew Brunswick, NY). After incubation for 1 hour, the plates were washed with TBS and test samples were added in two serial dilutions. After an incubation of 1 hour, the plates were washed again with TBS and incubated with rabbit serum solid coated with goat anti-human serum antibody, IgG fraction (Accurate Chemical and Scientific Corporation, Westbury HY, Lot # H036) containing 1% fish gelatin diluted 1: 1000 with TBS (Tris buffered saline). After incubation for 1 hour, the plate was washed with TBS and then incubated with purified rabbit anti-goat antibody complex, IgG fraction (horseradish peroxidase conjugated HRP) (BioRad Lot # 75312) diluted 1: 250 with TBS. After incubation for 2 hours, the plate was washed with TBS (Tris-buffered saline). The antigen-antibody-HRP complex was developed using o-phenylenediamine dichloride (OPD — 26 mg / 15 ml citric acid, pH 5.0). A pale yellow color indicated the presence of antigens, in particular plasma proteins. Absorption values were measured on a Molecular Devices Thermomax Microplate at 490 nm.

 Duplicate normal human serum samples (Dimension Laboratories Inc., Missisauga, Ontario, Lot # 30317), purified HbAo from Sigma Corporation (H-0267), purified HbAo from Walter Reed Army Institute of Research (WRAIR) (Lot # 92092) were analyzed. ) and purified HbAo from the present study. A summary of the results is given in table. 9. Obviously, the product of the invention contained no more than 0.0004% plasma proteins, while other tested samples contained at least an order of magnitude higher than these proteins.

 In all tests performed, the only whey protein that contained the product of the present invention was human serum albumin (HSA). The whey proteins found in other samples appear to be heterogeneous mixtures including HSA and other serum proteins.

List of references
1. Hess et al., "Systemic & Pulmonary Hypertension After Resuscitation with Cell-Free Hemoglobin", 1993.

 2. Chang et al., “Effect of Single Replacement ...”, BIOMAT, Art Cells and Immob. Biotech., 20 (2-4), pp 503-510 (1992).

 3. Simori et al., "Reaction on Human Endothelial Cells to Bovine Hemoglobin Solutions and Tumour Necrosis Factor."

 4. Huisman et al., "Studies on the Heterogeneity of Hemoglobins. IX - The Use of Tris-Hcl Buffers in the Anion-Exchange Chromatography of Hemoglobins", J. Chromatog., 19, 160-169, (1965).

 5. Saravis and Zamcheck.

 6. Tabata and Chang, Artificial Organs, 6, 213-214 (1982).

 7. Wiggler, C. J., Physiology of Shock, New York, The Common Fund, pp 138-139 (1950).

 8. Keipert and Chang, "Effecs of Partial ...", Vox. Sang 53, 7-14 (1987).

 9. Biessels et al. Annual Report of the Karl Landsteiner Fo undation, 1991.

 10. Leeksma et al. Blood 67: 650-654, 1986.

 11. Gerrits et al., Thromb. Res 5: 197, 1974.

 12. Butler, J.E., ELISA and other Solid Phase Immunoassays, edited by D. M. Kemeny and S. J. Callacombe, John Wiley and Sons Ltd, 1988, pp 1-180.

Claims (10)

 1. A method for purifying hemoglobin on a commercial scale from a crude solution containing protein-contaminating substances by ion exchange chromatography, characterized in that the crude solution is used in a concentration of 0.1-20%, the solution in the first stage is passed through a column filled with either an anion exchanger for cleaning a crude solution in which pollutants have a more acidic reaction than hemoglobin, or with a cation exchanger to clean a crude solution in which pollutants have a more basic reaction than oglobin, while the continuous passage of the crude solution in the first stage is carried out first until the column is completely saturated with hemoglobin and components with high affinity for the corresponding ion exchanger, and then until the column is overloaded with subsequent hemoglobin replacement, in the second stage, the eluate obtained at the first stage, through an anion or cation exchanger not selected in the first stage, while chromatography is also carried out first until the column is saturated with hemoglobin and substances with higher affinity for the column, and then to overload and the replacement of hemoglobin on it.  2. The method according to claim 1, characterized in that the first stage is anion exchange chromatography, and the second stage is cation exchange chromatography.  3. The method according to claim 2, characterized in that normal adult hemoglobin is used as the pre-selected hemoglobin.  4. The method according to claim 3, characterized in that the first stage - the anion-exchange process - is carried out at pH 5 - 9, using a passable solution with a low specific conductivity of less than 3 milliseconds (mS).  5. The method according to claim 4, characterized in that the first stage - the anion-exchange process - is carried out at a pH of 6.5 - 8.5, using a transmitted solution with a specific conductivity of less than 1 mS.  6. The method according to claim 4, characterized in that the passed solution is applied to the column at a speed of not higher than 10 cm per minute.  7. The method according to claim 4, characterized in that the second stage, the cation exchange process, is carried out at a pH of 6.5 to 8.5, using a passable solution with a specific conductivity of less than 3 mS.  8. The method according to claim 4, characterized in that the second stage, the cation exchange process, is carried out at a pH of 7-7.5, using a transmitted solution with a specific conductivity of less than 1 mS.  9. The method according to claim 7, characterized in that the passed solution is applied in the second stage to a cation exchange column with a speed of not higher than 10 cm per minute.  10. The method according to claim 7, characterized in that the concentration of the passed solution in the second stage of the cation exchange column is 2 to 6%.

Images