Human Recombinant Hyposialylated Erythropoietin, Methods of Purification and Therapeutic Uses Thereof

Abstract

The present invention relates to the fields of Biotechnology and Medicine and describes a pharmaceutical composition of recombinant human erythropoietin which is characterized by having a microheterogeneity of fucosylated N-glycans formed by bi, tri and tetra-antennary structures with mono and bi-sialylated sialic acid residues that represent 40-60% of the total glycans, trisialylated ones that represent 40-43% of the total glycans and tetrasialylated ones that represent 10-13% of the total glycans. This glycosylation pattern confers properties to said composition that allow for its use in disorders of the nervous system. The method of obtaining the pharmaceutical composition described herein is also described.

Claims

1. A pharmaceutical composition characterized in that it comprises as active ingredient a recombinant human erythropoietin (rhEPO) whose isoelectric point profile is between 4.25 and 5.85.

2. The pharmaceutical composition of claim 1 characterized in that the microheterogeneity of fucosylated N-glycans is formed by bi, tri and tetra-antennary structures, which have mono and bi-sialylated sialic acid residues in a range of 40-60% of the total of glycans, tri-sialylated ones in a range from 40-43% of the total glycans and tetra-sialylated ones in a range from 10-13% of the total glycans and a pharmaceutically acceptable excipient.

3. The pharmaceutical composition of claim 1 characterized in that the O-glycosylation site in serine 126 has 3 sialoforms that have between 0 and 2 sialic acid residues, where the monosialylated structure is the most abundant and represents between 78-82% of the total glycans and asialylated structures represent between 6-10% of the total glycans.

4. The pharmaceutical composition of claim 1 characterized in that the N-glycosylation site of asparagine 83 comprises: Fucosylated biantennary structures with 1 and 2 sialic acid residues, where said structures represent 8-12% of the total glycans, Fcosylated triantennary structures with 1, 2 and 3 sialic acid residues, where said structures represent 17-21% of the total glycans, Fucosylated tetraantennary structures with between 1 and 4 sialic acid residues, where said structures represent 27-31% of the total glycans and Fucosylated tetraantennary structures with N-acetyl lactosamine type 1 and 2 that present between 1 to 4 sialic acid residues, where said structures represent 38-42% of the total glycans.

5. The pharmaceutical composition of claim 1 characterized in that the pharmaceutically acceptable excipient comprises bioadhesive polymers and protein stabilizers.

6. The pharmaceutical composition of claim 5 characterized in that the bioadhesive polymers are selected from the group comprising: hydroxymethylcellulose, hydroxypropylcellulose and methylcellulose.

7. The pharmaceutical composition of claim 5 characterized in that the protein stabilizers are selected from the group comprising: L-tryptophan, L-leucine, L-arginine hydrochloride and L-histidine hydrochloride.

8. A method for obtaining an rhEPO with an isoelectric point profile between 4.25 and 5.85 without additional chemical and genetic modifications, comprising i) scaling up cells from a working cell bank to a cell density of ≥1×106 cells/mL: ii) fermenting the cells obtained in step (i) in a solution in stirred tank; and iii) collecting supernatant from said solution of step ii) and purifying said supernatant using chromatography, with a monolithic column as an anion exchanger with a Q quaternary ammonium ligand.

9. The method of claim 8, wherein the fermentation process of rhEPO is performed in stirred tank at a temperature of 35° C. and pH range between 7.2-7.3, in a protein-free culture medium supplemented with glutamine to give a final concentration of between 8 and 12 mmol/L.

10. The method of claim 8, wherein a 20 mmol/L Tris 10 mmol/L HCl solution with pH 8 and conductivity 1.5 mS/cm is used and as equilibrium buffer and a 50 mmol/L sodium acetate solution with pH 4.41 as elution buffer in the purifying step.

11. A pharmaceutical composition whose active ingredient is the rhEPO obtained by the method of claim 8.

12. The use of the pharmaceutical composition of claim 1 for the treatment of dementia, stroke, Parkinson's disease, ataxia, craniocerebral trauma, glaucoma, autism, hypoxia of the newborn, multiple sclerosis, amyotrophic lateral sclerosis and neurological damage induced by trauma, intoxication or radiation.

13. A method for the treatment of a subject in need of such treatment comprising the administration of the pharmaceutical composition of claim 1 in an administration range from 0.1 mg to 4 mg in a volume of 1 mL one to three times per week for 6 to 12 months.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0046] FIG. 1. Isoform profile: A) Definition of cut of the isoforms; B) SN generated in the cultures under the conditions of temperature and pH evaluated.

[0047] FIG. 2. Proportion of hyposialylated isoforms in the cultures under the temperature and pH conditions evaluated.

[0048] FIG. 3. Proportion of hyposialylated isoforms in each culture under the conditions evaluated at pilot scale.

[0049] FIG. 4. Relative intensity of less acidic isoforms in the SN.

[0050] FIG. 5. Influence of mobile phase pH and conductivity on the static adsorption capacity in the Q quaternary ammonium ligand of rhEPO isoforms: (A) hyposialylated (basic), (B) acidic.

[0051] FIG. 6. Breakthrough curves A): Chromatographic matrix “Q SFF”, B): Monolithic column “CIM QA”.

[0052] FIG. 7. Influence of elution buffer pH on the performance of basic and acidic isoforms.

[0053] FIG. 8. Distribution of the hyposialylated isoforms of rhEPO at production scale.

[0054] FIG. 9. Study of the N-glycans of the hyposialylated rhEPO by zwitterionic hydrophilic interaction liquid chromatography coupled to mass spectrometry.

[0055] FIG. 10. Extracted ions electropherogram (EIE) of hyposialylated rhEPO 0126 glycopeptide glycoforms observed resulting from digestion with: a) trypsin and neuraminidase and b) trypsin.

[0056] FIG. 11. EIE of N83 glycopeptide observed glycoforms resulting from trypsin and neuraminidase digestion.

[0057] FIG. 12. EIE of hyposialylated rhEPO N83 glycopeptide observed glycoforms resulting from digestion with: a) trypsin and neuraminidase and b) trypsin.

[0058] FIG. 13: Cell viability profile of the astrocyte culture at 24 and 48 hours after cell damage with 8% dimethylsulfoxide (DMSO) and subsequent treatment with hyposialylated rhEPO.

[0059] FIG. 14. Kaplan-Meier chart, viability during the 7 days of observation.

[0060] FIG. 15. Neurological status of animals at 24 hours after infarction.

[0061] FIG. 16. Effect of the application of different rhEPO isoforms on the number of reticulocytes in the normocytemic mouse model.

[0062] The present invention is further elaborated with the following examples and figures. However, these examples are not meant to be construed as limiting the scope of the invention.

EXAMPLES

Example 1. The pH and Temperature Influence rhEPO Isoform Profile at Laboratory Scale

[0063] From the CHO cell line transfected with the human EPO gene, a seed cell bank was obtained which was adapted to grow in suspension in a protein-free medium. The complete adaptation to this culture medium was obtained after 37 generations representing 25 days of cultivation.

[0064] To evaluate the performance of the cell line in the presence of different pH and temperature conditions, an experimental design was carried out where both variables were combined. The experimental conditions are shown in Table 1. The pH of the culture was controlled with the addition of 0.5 mol/L sodium hydroxide.

TABLE-US-00001 TABLE 1 Experimental conditions to evaluate the influence of pH and temperature variables on the rhEPO isoform profile. Initial cell concentration (cell/mL) 0,3 × 10.sup.6 Cell viability (%) >90 Culture time (days) 7 Incubation temperature (° C.) 35 37 Work volume (mL) 300 pH 7,2 7,3 7,5

[0065] The isoform profile of rhEPO corresponding to the samples of SN generated in the cultures in the different conditions evaluated was determined by isoelectric focusing. A mixture of ampholins with a pH range from 2 to 5 and from 3 to 10 was used, an internal EPOCIM® reference material was used as control. The intensity percentage of each of the isoforms in the samples was analyzed by densitometry, using the Gene Tools program. Isoforms with pH values ranging from 2.80 to 4.25 were defined as acidic isoforms and the ones with pH values in the range from 4.25 to 6.55 as basic isoforms (FIG. 1A).

[0066] FIG. 1B shows that the isoform profile was strongly influenced by temperature. Regardless of pH, when the SN samples corresponding to each condition were compared at a temperature of 37° C. with the control, the 7 acid isoforms are observed. On the other hand, at 35° C. a loss of acidic isoforms was observed, which was more pronounced in the conditions with pH 7.2 and 7.3.

[0067] FIG. 2 shows that under the conditions of pH 7.2 and 7.3 and temperature 35° C., the total of the isoforms obtained (100%) were basic isoforms.

Example 2. The pH and Temperature Influence rhEPO Isoform Profile at Pilot Scale

[0068] The impact of the best culture conditions obtained at laboratory scale (Temperature 35° C., pH 7.2 and 7.3) in a more controlled and favorable environment for the cell culture was evaluated. From the seed cell bank described in Example 1, four fermentation runs were designed using a ST type bioreactor (Infors-AGCH 4103, Bottmingen) of 3.5 L effective volume. The operating conditions are shown in Table 2. Initial cell viability was greater than 90% in all the conditions evaluated.

TABLE-US-00002 TABLE 2 Operating conditions corresponding to each fermentation run. Condition Condition Condition Condition Parameters 1 2 3 4 Initial Xv (cell/mL) 0,5 × 106 1,76 × 106 1,08 × 106 2,31 × 106 Operation temperature 37 35 35 37 (° C.) pH 7,41 7,2 7,3 - Work volume (L) 1,5 3 3 3 Total cultivation 5 7 7 7 time (days) Air flow (mL/min) 15 15 15 15 Rotational impeller speed (rpm) 150 150 150 150 Dilution rate (vvd) — 0,3 0,3 0,3

[0069] Samples were taken at the end of each fermentation and the isoform profile of the SN samples generated in the cultures for the conditions evaluated in the bioreactor was determined by means of the isoelectric focusing technique. Once these profiles were obtained, Gene Tools program was used and starting from the image corresponding to the isoelectric focusing gel, the intensity of each band present in the gel was determined and the proportion of the hyposialylated isoforms for each condition was evaluated. As can be seen in FIG. 3 in culture conditions 2 and 3 (35° C., pH 7.2-7.3), the proportion of hyposialylated isoforms was higher than in conditions 1 and 4. These results corroborate the findings obtained at laboratory scale, that is, that by modifying the operating conditions to have a pH 7.2-7.3 and a temperature of 35° C., the isoform profile of rhEPO is modified, a greater intensity being achieved in those with pH values between 4.25 and 5, 85.

Example 3. By Increasing the Concentration of Glutamine in the Culture Medium, the Expression of rhEPO Basic Isoforms is Favored

[0070] In order to assess whether the increase in the concentration of glutamine in the culture medium had an effect on the increase in hyposialylated isoforms, the performance of the rhEPO-producing CHO cell line was evaluated. A seed cell bank which was exposed to different glutamine concentrations of the culture medium was used, and an adaptation to the culture medium was carried out for 15 days. The assessment started with an initial cell concentration of 0.5×106 cel/mL in a final volume of 300 mL of culture medium, using rotating bottles that were kept in 37° C. incubators with an agitation speed of 600 rpm.

[0071] The final glutamine concentrations in the culture medium were: 8, 12 and 16 mmol/L and a culture medium with 6 mmol/L of glutamine was used as control.

[0072] The relative intensity of hyposialylated isoforms in the SN of the cultures was evaluated by densitometry after seven days of treatment with the different concentrations of glutamine.

[0073] FIG. 4 shows that with each of the three variants evaluated a higher proportion of hyposialylated isoforms was obtained with respect to the control, thus corroborating that with an increase in the concentration of glutamine in the culture medium, the obtainment of hyposialylated isoforms in the SN is favored. The highest value of these isoforms was observed with 8 mmol/L of glutamine (87%).

Example 4. The pH and Conductivity Influence the Adsorption of the Acidic and Basic Isoforms in the Q Strong Quaternary Ammonium Anion Exchanger at a Laboratory Scale

[0074] With the objective of determining the pH and conductivity conditions that favor the greatest adsorption of acidic isoforms in the Q strong quaternary ammonium anion exchanger the following buffer solutions were evaluated: 20 mmol/L sodium phosphate (anhydrous dibasic and monobasic dihydrate) and 20 mmol/L Tris 10 mmol/L HCl. Buffer solutions were studied at different pH and conductivity values, pH between 6 and 8 and conductivity between 1.5 and 5 mS/cm.

[0075] The greatest adsorption of acidic isoforms to the exchanger was observed with conditions of pH 6 and conductivity 1.50 mS/cm. On the other hand, the adsorption of the hyposialylated (basic) isoforms is maximized at pH 8 and conductivity 1.50 mS/cm. The results are shown in FIG. 5.

Example 5. Q Strong Anionic Matrix that Uses the Monolithic Column Technology has Better Performance in Terms of Capacity to Separate rhEPO Isoforms as Compared to the One that Uses the Conventional Technology of Chromatographic Gels at a Laboratory Scale

[0076] The dynamic adsorption capacity (Q) of each technology under study was calculated with two breakthrough curves, using two of the linear flow rates recommended by the manufacturer of each technology 100 cm/h and 600 cm/h for the technology of chromatographic gels (Q SFF) and 156 cm/h and 624 cm/h for the monolithic column (CIM QA). The equilibrium solution used for experiments with the conventional technology and for the monolithic column technology was Tris-HCL buffer with pH 8 and conductivity 1.5 mS/cm, according to the results obtained in Example 4.

[0077] The rhEPO sample was applied to the columns and at the exit from them samples were taken at different times. The concentration of proteins (C) in each sample was determined by spectrophotometry. With the known values of C in each of the samples, the fraction of non-adsorbed protein C/C.sub.0 was calculated. The load, which is the mass of protein applied to the column per unit volume of gel given in mg of rhEPO/mL of Q matrix was calculated for each time.

[0078] In FIG. 6 the values of C/C0 vs. dynamic adsorption capacity are represented graphically in breakthrough curves that allow to know the Q of the gel per fraction of non-adsorbed protein C/C.sub.0=0.1 at the specified flow rates. FIG. 6A shows the breakthrough curve obtained with the conventional technology of chromatographic gels where the lowest flow rate was the one with the highest dynamic capacity. FIG. 6B shows that the monolithic column has very similar dynamic adsorption capacity with the two linear flow rates tested, so it can be operated at higher flow rates, without experiencing variations in the dynamic adsorption capacity of the matrix.

[0079] Table 3 shows the Q values obtained with the different measurements performed.

TABLE-US-00003 TABLE 3 Values of Q obtained with each of the flow rates studied in Q SFF and CIM QA columns. Air flow Column (cm/h) Q (rhEPO/mL of gel) QSFF 100 7,82 600 5,58 CIM QA 156 8,25 624 7,79

[0080] When comparing the results of the Q studies carried out in strong anion exchangers in packed bed and the monolithic column technology, it was observed that both technologies have similar Q at the lowest linear velocity studied. However, at the highest linear velocity the monolithic column has a dynamic adsorption capacity 1.40 times greater than that of the traditional chromatographic matrix evaluated. Therefore, it can be concluded that the monolithic column allows for the increase of the work flow of the process without affecting the capacity of processing the rhEPO mass to be purified.

Example 6. By Lowering the pH, the Elution of the Hyposialylated rhEPO Isoforms in the Monolithic Column at Laboratory Scale is Favored

[0081] Experimental tests were performed using Q SFF and CIM QA columns. The equilibrium solution used for both technologies was the Tris-HCl buffer at pH 8 and conductivity of 1.5 mS/cm, according to Example 4. The working linear velocity of the Q SFF column was 600 cm/h and that of the monolithic column was 624 cm/h. The product used for the experiments was the one obtained from the “Sephadex” G-25 molecular exclusion chromatographic column, after the change of buffer to that of equilibrium solution mentioned above was carried out. For the elution of the basic isoforms several runs were performed with the monolithic anion-exchange column using the 50 mmol/L Tween 20 sodium acetate buffer at 0.01% with pH values: 4.41; 4.81; 5.06; 5.20.

[0082] The recoveries obtained in the elution of the hyposialylated (basic) isoforms and the acidic isoforms in each run are shown in FIG. 7. From the analysis of these results it was determined that as the pH of the elution buffer solution of basic isoforms increases their performance decreases, therefore the 50 mmol/L sodium acetate buffer at pH 4.41 was selected for the elution.

Example 7. The Process of Obtaining the Hyposialylated rhEPO at Production Scale is Consistent in Terms of Separation of Isoforms and Purity Degree

[0083] From the seed cell bank described in Example 1, a fermentation run was carried out the initial cell viability being greater than 90%. The fermentation stage was performed at a temperature of 35° C., in a protein-free culture medium which was supplemented to reach a final concentration of 8 mmol/L of glutamine, the pH of the medium was kept between 7.2 and 7.3.

[0084] Once four harvests were obtained, the purification stage was performed where for the critical step the anion exchanger with Q quaternary ligand was used using a monolithic column. A Tris-HCl solution at pH 8 and conductivity of 1.5 mS/cm was used as equilibrium buffer and the 50 mmol/L sodium acetate buffer pH 4.41 was used for the elution.

[0085] The isoform profile of the four batches of active raw material obtained after the purification process was determined by means of the isoelectric focusing technique. A mixture of ampholins with a pH range from 2 to 5 and from 3 to 10 was used, and an internal EPOCIM® reference material was used as control. FIG. 8 shows the consistency in the separation of the isoforms where an isoform profile configured by six major isoforms is observed of which only two are shared with the control.

[0086] In addition, the sialic acid content of the purified isoforms was determined according to the procedure for the determination of this molecule described in Europe 8.0 Pharmacopoeia (2014) and the purity by reverse phase HPLC. Table 4 shows the sialic acid content and purity results obtained.

TABLE-US-00004 TABLE 4 Sialic acid content and purity results of the hyposialylated rhEPO. Sialic Acid (sialic acid mol/mol of RP-HPLC Product protein) (%) Batch 1 5,1 96,24 Batch 2 5,5 98,03 Batch 3 6,5 98,38 Batch 4 6,9 98,29

[0087] The content of sialic acid was less than 10 mols of sialic acid/protein molecule and the purity was greater than 95% in the four batches from which it can be concluded that the process to obtain hyposyalylated rhEPO guarantees the obtainment of isoforms in a pH range between 4.25 and 5.85, which is different from that observed in NeuroEPO.

Example 8. The Tertiary Structure Associated with Glycosylation of the Hyposialylated rhEPO Shows a Characteristic Microheterogeneity

Glycan Analysis

[0088] To study its N-glycans profile hyposialylated rhEPO underwent a process of denaturation and enzymatic digestion with Peptide N-glycosidase F (PNGase F). Once the N-glycans were released, they were purified by solid phase extraction using HyperSep Hypercab SPE cartridges (Mancera-Arteu, M. et al. (2016) Anal. Chim. Acta 940: 92-103.

[0089] Subsequently, they were derivatized according to the procedure described by Giménez et al. in 2015. (Gimenez, E et al. (2015) Anal. Chim. Acta, 866: 59-68).

[0090] A zwitterionic-hydrophilic interaction capillary liquid chromatography coupled to mass spectrometry was performed following the methodology described by Mancera-Arteu, M. et al. (2016) Anal. Chim. Acta, 940: 92-103.

[0091] FIG. 9 shows the mass spectra of three of the glycans detected in the hyposialylated rhEPO. As can be seen, glycan 3Ant2SiA1Fuc is the most abundant, while glycan 4Ant4SiA1Fuc is found in decreased quantities.

[0092] Table 5 shows the percentage of relative area of glycans according to structure.

TABLE-US-00005 TABLE 5 Percentage of relative area of glycans according to structure. Glycans Area (%)** Biantennary structures 10,3 Triantennary structures 17,2 Tetraantennary structures 72,5

[0093] Table 6 shows the glycans detected with the corresponding relative areas, monoisotopic experimental molecular mass (Mexp) and mass error. As can be seen there is a significant number of glycans with less sialylated structures.

TABLE-US-00006 TABLE 6 N-glycans detected in the hyposialylated rhEPO by zwitterionic-hydrophilic interaction capillary liquid chromatography coupled to mass spectrometry. Glycans Area Area (%)* Area (%) ** Mexp[Da] Error 2 Ant 2Ant1SiA1Fuc 608.586 2.91 10.3 2154.8011 9.39 2Ant2SiA1Fuc 2.321.812 11.31 2445.8983 8.34 3 Ant 3Ant1SiA1Fuc 448.587 2.18 17.2 2519.9309 6.25 3Ant2SiA1Fuc 1.872.328 8.98 2811.0307 7.25 3Ant3SiA1Fuc 2.562.955 12.46 3102.1243 6.58 4Ant 4Ant1SiA1Fuc 589.168 2.94 30.1 28885.0669 8.01 4Ant2SiA1Fuc 2.122.871 10.03 3176.1571 6.09 4Ant3SiA1Fuc 4.243.325 20.74 3467.2550 5.60 4Ant4SiA1Fuc 1.575.249 7.80 3758.2975 5.11 4Ant 1LacAc 4Ant1LacNAc1SiA1Fuc 205.545 1.01 35.0 3250.1917 5.45 4Ant1LacNAc2SiA1Fuc 1.506.731 3.67 3541.2889 5.75 4Ant1LacNAc3SiA1Fuc 807.621 7.01 3832.3847 3.44 4Ant1LacNAc4SiA1Fuc 8.539.613 3.00 4123.4849 4.12 4Ant 2LacAc 4Ant2LacNAc2SiA1Fuc 185.318 0.88 3.32 3906.4126 1.11 4Ant2LacNAc3SiA1Fuc 433.265 2.06 4197.4971 5.26 4Ant2LacNAc4SiA1Fuc 322.987 1.59 4488.5519 6.33 4Ant 3LacAc 4Ant3LacNAc1SiA1Fuc 68.423 0.32 4.081 3980.5185 24.3 4Ant3LacNAc2SiA1Fuc 1.092.708 0.3 4271.4698 16.70 4Ant3LacNAc3SiA1Fuc 83.360 0.38 4562.5930 13.0 4Ant3LacNAc4SiA1Fuc 95.315 0.43 4853.6515 16.6 *It represents the relative area with respect to the total of glycans detected. ** it represents the relative area grouped by antennas with respect to the total of glycans detected.

[0094] As can be seen in Table 6, glycans with more sialylated structures (four sialic acid molecules) are found in a low ratio. Instead, structures with a sialic acid molecule that have not been found in other rhEPOs such as 3Ant1SiA1Fuc, 4Ant1SiA1Fuc, 4Ant1LacNAc1SiA1Fuc and 4Ant3LacNAc1SiA1Fuc are detected. The glycan 4Ant3Sia1Fuc is abundant in the hyposialylated rhEPO. Also, it can be seen that the percentage of the relative area by antennas with respect to the total glycans detected is different from other rhEPO and that the structures with one or two sialic acid residues represent more than 50% of the relative area by antennas.

Glycopeptide Analysis

[0095] To detect all the glucoforms present in the O.sub.126 and N.sub.83 glycopeptides, the rhEPO and the rhEPO-CRS (pharmacopoeia reference product) were subjected to digestion with trypsin and neuraminidase (rhEPO HS-TN). All sialoforms in each glucoform detected were studied from the trypsin digestion analysis (rhEPO HS-T). The samples were analyzed by mass spectrometry according to the procedure described by Giménez, E. et al. (2011) Rapid Commun Mass Spectrom. 25: 2307-2316.

[0096] FIGS. 10A and B show the EIE of O.sub.126 glycopeptide glycoforms detected in the rhEPO HS-TN and HS-T digestates, respectively. In the first one, a single peak corresponding to that of the O.sub.126/0SiA glycoform is observed, since the neuraminidase produces a complete desialylation of the glycopeptide, so that all the sialoforms become a single glycoform. In contrast, when digested with trypsin only, three sialoforms with 0, 1 and 2 sialic acid molecules are observed.

[0097] Table 7 shows the O.sub.126 sialoforms detected with the corresponding relative areas, monoisotopic experimental molecular mass (Mexp) and mass error.

TABLE-US-00007 TABLE 7 Glycoforms detected in O.sub.126 glycopeptide of hyposialylated and CRS rhEPO digested with trypsin. rhEPO-CRS-T Glycopeptide Rt 0126 [min] Rel Area [%] Mexp[Da] Error[ppm] 0SiA 12,1 1,61 1829,9169 15,0 1SiA 14,6 60,4 2121,0166 15,0 2SiA 18,4 38,0 2412,1175 15,0 hyposialylated-T EPO 0SiA 14,0 8,88 1829,9173 15,2 1SiA 17,2 80,5 2121,0173 15,3 2SiA 22,1 10,7 2412,1215 17,1

[0098] The results obtained show that the most abundant sialoform is the one that contains one sialic acid molecule. Both the percentage of relative area of asialylated isoforms and of those with only one sialic acid are higher as compared to the ones described for other rhEPO, finding in the proportion of both sialoforms differences with the CRS rhEPO. In the case of the N83 glycopeptide, when analyzing the rhEPO HS-TN digest, six peaks were detected, corresponding to six different glycoforms without sialic acid. The EIEs obtained are shown in FIG. 11 and the detected glycoforms in Table 8. The results obtained show that in the N.sub.83 site there is a higher percentage of complex desialylated tetraantennary structures.

TABLE-US-00008 TABLE 8 Detected glicoforms of EPO HS-TN N.sub.83 glycopeptide Glycopeptide.sub.83 R.sub.t[min] Area.sub.rel[%] Mexp[Da] Error[ppm] 2Ant0SiA1Fuc 15,4 17,12 4126,9263 13,0 3Ant0SiA1Fuc 15,7 27,33 4492,0674 13,9 4Ant0SiA1Fuc 15,9 24,30 4857,1996 12,9 4Ant1LacNAc0SiA1 16,2 12,50 5222,3335 12,3 4Ant2LacNAc0SiA1 16,4 6,75 5587,3257 13,6 4Ant3LacNAc0SiA1 15,9 12,0 5952,6219 14,8

[0099] FIG. 12A shows that there is no separation between the different tetraantennary structures in the rhEPO HS-TN digestate. On the contrary, when the sialylated glycoforms of the rhEPO HS-T digestate are analyzed, a clear separation between them is observed, those with three sialic acid residues having greater intensity (FIG. 121B). These results agree with the findings from the study of glycans.

[0100] Table 9 shows all sialoformns detected and the corresponding relative areas, monoisotopic experimental molecular masses (Mexp) and mass error.

TABLE-US-00009 TABLE 9 Glycoforms of N83 glycopeptide detected in trypsin and trypsin/neuroaminidase digestates. Glycopeptide N.sub.83 Ref area (%) Area (%)** Mexp[Da]Error[ppm] 2 Ant 2Ant1SiA1Fuc 2,54 10,47 4417,9833 3,43 2Ant2SiA1Fuc 7,93 4709,1311 14,3 3Ant 3Ant1SiA1Fuc 9,17 19,56 4783,1079 1,58 3Ant2SiA1Fuc 7,49 5074,2569 12,1 3Ant3SiA1Fuc 2,90 5365,3564 12,2 4Ant 4Ant1SiA1Fuc 4,82 29,13 5148,2690 7,08 4Ant2SiA1Fuc 9,58 5439,4017 13,6 4Ant3SiA1Fuc 9,06 5730,5066 14,5 4Ant4SiA1Fuc 5,67 6021.4347 14,0 4Ant 1LacAc 4Ant1LacNAc1SiA1Fuc 4,40 23,79 5513,3202 8,08 4Ant1LacNAc2SiA1Fuc 7,67 5804,5468 14,9 4Ant1LacNAc3SiA1Fuc 5,96 6095,5354 3,31 4Ant1LacNAc4SiA1Fuc 5,76 6386,8708 34,4 4Ant 4Ant2LacNAc2SiA1Fuc 6,42 17,05 6169,8030 34,1 2LacAc 4Ant2LacNAc3SiA1Fuc 4.81 6460,6122 11,7 4Ant2LacNAc4SiA1Fuc 5,82 6751,6217 23,9

[0101] Results corroborate that the hyposialylated rhEPO is characterized by having less sialylated structures (for example, 2Ant1SiA1Fuc or 3Ant1SiA1Fuc) when the protein is digested only with trypsin. When the percentage of relative areas of the mono and bisialylated structures was verified, results show that they represent approximately 60%. The tetraantennary structures in the N83 glycopeptide, are found in the highest percentage as compared to other glycoforms, those with two sialic acid residues being the ones with the greatest proportion.

Example 9. Hyposialylated rhEPO has a Restorative Effect on Astrocytes Against the Cytotoxicity of DMSO

[0102] PG4 astrocyte cell line was cultivated in Dulbecco's Modified Eagle's Medium (DMEM) culture medium high in glucose supplemented with 3.7 g/L NaHCO.sub.3 and 10% fetal bovine serum (FBS). The cells were incubated for 24 hours at a temperature of 37° C. in an atmosphere of 5% CO.sub.2/95% air. After the specified time the SN was extracted and the cells were subjected to damage caused with 8% DMSO and incubated again for 24 h. Subsequently, the SN was removed and 2% DMEM culture medium with different concentrations of hyposialylated rhEPO (1.25; 2.5; 5 and 10 ng/mL) was added. At 24 h and 48 h Alamar Blue reagent was added, then the cells were incubated for 6 h and a reading was made at 540/630 nm. All incubations were performed at a temperature of 37° C. in an atmosphere of 5% CO.sub.2/95% air.

[0103] It can be seen in FIG. 13 how at the different concentrations 1.25; 2.5; 5 and 10 ng/mL of hyposialylated rhEPO the cells were able to regain their viability, the best condition being 1.25 ng/mL of hyposialylated rhEPO, which demonstrates the restorative capacity of astrocytes.

Example 10. Hyposialylated rhEPO has Better Neuroprotective Activity than NeuroEPO

[0104] Gerbils from Mongolia were used to develop the model of permanent unilateral ischemia following the methodology described by Kahn K. in 1972, (Kahn K. (1972) Minneap 22: 510-515). Subsequently, the animals were randomized into five experimental groups:

[0105] Group 1: Vehicle 10 μL of the vehicle

[0106] Group 2: Treated with 0.142 mg/kg of hyposialylated rhEPO

[0107] Group 3: Treated with 0.0142 mg/kg of hyposialylated rhEPO

[0108] Group 4: Treated with 0.142 mg/kg of NeuroEPO

[0109] Group 5: Treated with 0.0142 mg/kg of NeuroEPO

[0110] Each treatment was administered by IN route, three times a day for four days. The animals were evaluated during the first 4 days of treatment, as well as during recovery the next 3 days.

[0111] The results show a significantly greater survival rate of the animals treated with the hyposialylated rhEPO in the doses of 0.142 and 0.0142 mg/kg when compared with the vehicle, while the NeuroEPO only decreased mortality significantly in the dose 0.142 mg/kg but not in the 0.0142 mg/kg (FIG. 14).

[0112] The neurological evaluation showed the neuroprotective effect of hyposialylated rhEPO (NeuroEPO plus), which significantly decreased the values of the neurological scale in the animals treated with it in the doses used, as compared with the placebo group and with the groups treated with NeuroEPO, in which the same result obtained with hyposialylated rhEPO could only be achieved with the 0.142 mg/kg dose. (FIG. 15). (Duncan statistical test, equal letters p>0.05; different letters p>0.05).

Example 11 Hyposialylated rhEPO does not Increase the Reticulocyte Count in the Normocytemic Mouse Model

[0113] B6D2F1 female mice were randomized into 6 experimental groups of 6 animals each and received a single dose of treatment as follows:

[0114] Group 1: 0.003 mg/mL of rhEPO working reference material in a 200 μL volume by subcutaneous route (SC).

[0115] Group 2: 0.006 mg/mL of hyposialylated rhEPO in a 200 μL volume by SC route.

[0116] Group 3: 0.5 mg/mL of hyposialylated rhEPO in a 200 μL volume by IN route.

[0117] Group 4: 1 mg/mL of hyposialylated rhEPO in a 200 μL volume by IN route.

[0118] Group 5: 2 mg/mL of hyposialylated rhEPO in a 200 μL volume by IN route.

[0119] Group 6: Control, 200 μL excipient by SC route.

[0120] The results of the reticulocyte count are shown in FIG. 16. As can be seen there were no significant differences between the animals in the control group and those in the groups treated with hyposialylated rhEPO both by SC and IN. The reticulocyte count of the animals treated with the rhEPO work reference material was significantly higher than the one of the animals in the control group and in the group treated with hyposialylated rhEPO, with both routes used. (Duncan test, equal letters p>0.05; different lettersp<0.05).

[0121] The results demonstrated that hyposialylated EPO does not increase the reticulocyte count even at the highest dose established for this trial, which shows that it has no hematopoietic effect.