International Journal of Biological Macromolecules 64 (2014) 267–275
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Inhibition of methemoglobin formation in aqueous solutions under aerobic conditions by the addition of amino acids Yuping Wei a,b , Chunlong Li a,b , Liang Zhang a,b , Zhiguo Su a , Xia Xu a,∗ a b
State Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, PR China University of Chinese Academy of Sciences, Beijing 100049, China
a r t i c l e
i n f o
Article history: Received 22 July 2013 Received in revised form 6 December 2013 Accepted 8 December 2013 Available online 14 December 2013 Keywords: Amino acid Hemoglobin Methemoglobin Molecular simulation Stability
a b s t r a c t Hemoglobin (Hb) as an important iron-containing oxygen-transport protein is easily oxidized to the ferric met-form, methemoglobin (metHb), and loses the capacity of binding oxygen during storage. In this study, the experimental data indicate that the presence of Tyr and Glu significantly suppress the metHb formation in the Hb solutions in aqueous environment under aerobic conditions at the temperature of 25 and 37 ◦ C, respectively. At pO2 of 144 Torr the metHb percentage in the Hb solutions was the lowest with less than 10% at day 7 after incubation with Tyr at the ratio of 24 at pH 9.5 at 25 ◦ C. At 37 ◦ C, the metHb percentage did not reach 5% after 12 h of incubation with Glu at the ratio of 24 at pH 9. Molecular simulation analysis suggest that the presence of Tyr or Glu may contribute to the formation of the breakwater network, the stabilization of distal histidine, the changes in the size of heme pocket, and eventually result in the inhibition of metHb formation. This study provides insight into a new design for Hb-oxygen based carriers with strongly inhibition of metHb formation in aqueous environment under aerobic conditions, even at physiological temperature in vitro. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Hemoglobin (Hb), a tetrametric protein in red blood cells, is composed of four polypeptide chains (two ␣- and two -chains), each of which containing a hydrophobic pocket bound to a prosthetic heme group. This group consists of a Fe(II) ion and a porphyrin ring as the O2 -binding site to carry and deliver oxygen. In the prosthetic heme group, the iron atom locating at the center of the porphyrin ring forms four bonds with porphyrin nitrogen, one covalent bond with the histidine residue and one bond with exogenous ligands (Fig. 1) [1]. The formation of metHb causes Hb to lose the capacity of binding oxygen under aerobic conditions. At a steady state the distal His58 stabilizes the bound dioxygen through a hydrogen bond. Upon the nucleophilic attack of water molecules, the FeO2 center is open to the proton on the acidic pH side or hydroxide anion on the basic side. This results in an irreversible displacement of the bound
Abbreviations: BSA, bovine serum albumin; CD, circular dichroism; Hb, hemoglobin; metHb, methemoglobin; PAGE, polyacrylamide gel electrophoresis; ROS, reactive oxygen species; Tris–HCl, Tris[hydroxymethyl-aminomethane]–HCl; TEMED, N,N,N ,N -tetramethyl ethylenediamine. ∗ Corresponding author at: 1 Beiertiao, Zhongguancun Haidian District, Beijing 100190, PR China. Tel.: +86 10 82544939; fax: +86 82544939. E-mail address: [email protected] (X. Xu). 0141-8130/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijbiomac.2013.12.010
dioxygen in the form of superoxide ion (O2 − ) and a replacement of the sixth coordination position originally binding with oxygen by water or hydroxide [2] and eventually leads to the formation of metHb [2,3]. Therefore, the inhibition of metHb formation is essential to preservation of Hb before clinical applications. In vivo the percentage of metHb in red blood cells is less than 5% due to the presence of systematic reduction. However, in vitro due to the absence of the metHb reduction system which is removed during processes of Hb purification [4–7], the metHb formation and protein denaturation are the major problems in the Hb-containing systems and acellular-type Hb-based oxygen carriers [8]. These Hb-containing systems and Hb-based oxygen carriers, such as liposome-encapsulated Hb [9], Hb vesicles [10,11], and microparticles with Hb loading [12–15], are considered as potential candidates for red-blood-cell substitutes [16–19]. It is hence that the efficient inhibition of metHb formation in the Hb-containing systems and Hb-based oxygen carriers has far-reaching significance in clinical applications. Various approaches including cryopreservation and freezedrying have been investigated to stabilize Hb [20,21]. However, these methods are time-consuming and process-complicated. Hb preserved either alone or in oxygen carriers in aqueous environment is much more convenient for infusion in emergency situations. Therefore, the inhibition of metHb formation is the primary method for the stabilization of Hb during its preservation in aqueous environment. Previous reports have demonstrated that
268
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
Fig. 1. Schematic representation of Hb autoxidation process in aqueous environment. At steady state the distal histidine (58) stabilizes the bound dioxygen through a hydrogen bond. Electron transfer from N to N at His58 weakens hydrogen bond between His58 and FeO2 center. Nucleophilic attack of entering water molecule can result in the irreversible displacement of the bound dioxygen in the form of superoxide ion (O2 − ), and facilitate a full charge transfer from Fe (II) to O2 . This eventually causes the formation of metHb.
under anaerobic conditions the introduction of reductants or L-Tyr can suppress the metHb formation [6,22,23]. The thermal stability of Hb in aqueous environment can be improved in the presence of sarcosine and sorbitol [24]. MetHb formation can also be inhibited by the addition of glutamate [25] and antioxidant enzymes [26,27]. In addition, the metHb formation is strongly pH-dependent, and the nucleophilic attack to FeO2 during the metHb formation is more significantly enhanced under acidic conditions than alkaline [3,28]. Temperature should be another important factor to affect protein stability. However, the report about the effect of temperature on metHb formation is rare. Although it is contributive to the stabilization of Hb by pH adjustment and temperature control, the methods for suppressing the metHb formation in aqueous environment under aerobic conditions are still needed to be improved. The aim of this study was to suppress the metHb formation in aqueous environment under aerobic conditions. Due to the high stability, low cost and good safety, amino acids are the potential stabilizers of proteins. This study was the first report to examine the effects of five different amino acids on the inhibition of metHb formation at different pHs and different temperatures, and further the effect of the mole ratio of amino acid to Hb on the metHb formation. The possible mechanism behind the inhibiting effect of amino acids on the metHb formation was discussed based on the molecular simulation. 2. Experimental procedures 2.1. Materials Donated red blood cells were obtained from the Anzhen Hospital (Beijing, China). Tris[hydroxymethyl-aminomethane]–HCl (Tris–HCl), acetonitrile, methanol, glycine and NaCl were purchased from Sinopham Chemical Reagent (China). Acrylamide, N,N -methylene bisacrylamide, coomassie blue G-250, acetic acid/ethyl alcohol and ammonium persulfate were purchased from Xinglong (China). N,N,N ,N -tetramethyl ethylenediamine (TEMED) was purchased from Merck (UK). l-Trp, l-Cys, l-Lys, l-Tyr, sodium glutamate and bovine serum albumin (BSA, MW66000) were purchased from Sigma–Aldrich (USA). Anion exchangers and Q Sepharose Fast Flow (Q-SFF) resin used in this study were produced from GE Healthcare (USA). Dialysis membrane with cut-off molecular weight of 3500 Da was purchased from Union Carbide (USA). 2.2. Preparation and purification of Hb from red blood cells Hb was isolated from the donated blood using hemolysis. The red blood cells were suspended in 20 mM Tris–HCl buffer
(pH 8.0) at the ratio of 1:2 (v/v) at 4 ◦ C for 24 h. After hemolysis, the suspension of lysed cells was centrifuged at 2300 × g for 1 h at 4 ◦ C. The supernatant was purified using the ion exchange chromatography method with Q-SFF resin at room temperature [29,30]. Briefly, the supernatant was loaded with an injector and eluted with a linear gradient of buffer from 100% buffer A (20 mM Tris–HCl, pH 8.0) to 75% buffer B (20 mM Tris–HCl plus 0.2 M NaCl, pH 8.0) in 5-fold volume of column. The elution was monitored with a UV detector at 280 nm. The Hb solution after purification was concentrated to 150 g/l using a dialysis membrane with the cut-off molecular weight of 3500 Da. 2.3. Storage 150 g/l of Hb in 20 mM Tris–HCl buffer was diluted to 50 g/l using Tris–HCl buffer (pH 7.4). The Hb diluted solutions were incubated at pH 7, 8, 9, 9.5 and 10 (20 mM Boric acid sodium hydroxide buffer for pH 10) at pO2 of 144 Torr at 25 or 37 ◦ C. The stock solutions of Glu, Lys, Tyr, Cys and Trp were prepared at the concentration of 0.04 M at pH 7. For Tyr, the stock solution was prepared at around pH 10 due to the low solubility of Tyr below pH 10. Glu, Lys, Cys, Tyr or Trp was introduced to the Hb solutions at the mole ratio of 24, respectively, and the pH was adjusted to the required pH using the Tris–HCl buffer (20 mM). Then the Hb solutions were incubated at 25 ◦ C for up to 7 days or 37 ◦ C for up to 21 h at pO2 of 144 Torr at pH 8, 9 and 9.5, respectively. In addition, Tyr and Glu were introduced in the Hb solutions at the mole ratio of 8, 16 and 24 at pH 9 and 9.5, respectively, and incubated at 25 ◦ C for up to 2 days. The pH at the end point for all the tested conditions was measured using a pH meter (Sartorius, Germany). 2.4. Measurement of metHb content The Hb solutions were diluted from 50 g/l to 0.5 g/l before assay. The pH of the diluted Hb solutions measured using a pH meter was in the range of 6–8 for all the tested conditions. The absorption of the diluted Hb solutions (0.5 g/l) at 560, 576 and 630 nm was measured using a UV-Vis spectrometer (Shanghai Spectrum, China). The content of metHb was determined using the following equations described by Benesch et al., which was valid between pH 6.2 and pH 8.8 [31]. A560 = ε560 CdeoxyHb + ε560 CoxyHb + ε560 CmetHb
(1)
A576 = ε576 CdeoxyHb + ε576 CoxyHb + ε576 CmetHb
(2)
A630 = ε630 CdeoxyHb + ε630 CoxyHb + ε630 CmetHb
(3)
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
269
Fig. 2. The purification of human Hb. (A) A typical elution pattern for the purified human Hb. Column (200 mm in length, 26 mm in internal diameter) was packed with 60 ml Q-sepharose fast flow resin. RBC lysate injected into the column via Superloop (5 ml) was eluted using a linear gradient of buffer from 100% buffer A to 75% buffer B and followed by 100% buffer B. The highest peak represents Hb, other peaks represent impurities. (B) Native-PAGE of the purified Hb solution. Native stacking gel with 3% acrylamide (pH 6.8) and native separating gel with 12.5% acrylamide (pH 8.8) were used. BSA was used as the molecular marker. Gels were stained in coomassie blue G-250 and destained in acetic acid/ethyl alcohol aqueous solution.
where A560 , A576 and A630 are the absorption at 560, 576 and 630 nm, ε560 , ε576 and ε630 are the extinction coefficients at 560, 576 and 630 nm chosen according to the pH of solutions as described by Benesch et al. [31]. ε560 , ε576 and ε630 are 0.749, 0.850 and 0.208 at pH 8.8, 0.405, 0.406 and 0.401 at pH 7.0, 0.373, 0.358 and 0.412 at pH 6.2, respectively. CdeoxyHb , CoxyHb and CmetHb are the mole concentrations of deoxy, oxy and met hemoglobin, respectively. 2.5. Circular dichroism (CD) spectra The secondary structure of Hb was determined using a CD spectrometer (Jasco-J810, Japan) at 25 ◦ C. The spectra were scanned at 200 nm/min with a response time of 1 s and recorded in the range of 190–260 nm and in the Soret region. The results were expressed as ellipticity in mdeg. The proportion of ␣ helix in Hb was calculated using the Jasco-J810 analysis software. 2.6. Fluorescence measurements The structural changes of Hb after incubation with Tyr at pH 9.5 at 25 ◦ C or Glu at pH 9 at 37 ◦ C were investigated using a Hitachi-4000 spectrophotometer (Japan). The emission spectra were recorded between 300 nm and 500 nm at the excitation wavelength of 280 nm. 2.7. Electrophoresis Polyacrylamide gel electrophoresis (PAGE) was performed in a single-sided vertical system equipped with an electrophoresis power supply (Tanon, China). Native stacking gel (3% acrylamide, pH 6.8) and native separating gel (12.5% acrylamide, pH 8.8) were prepared using TEMED with ammonium persulfate to catalyze the polymerization of acrylamide. The upper and lower tank buffers (pH 8.8) contained glycine as the mobile anion. The samples with equal volume of loading buffer were added into the wells. BSA was used as the molecular marker. Gels were stained in coomassie blue G-250 and destained in acetic acid/ethyl alcohol aqueous solution. 2.8. Molecular simulation Molecular dynamics (MD) simulation was performed using the Gromacs 4.5.4 and the Gromos96 G43a1 force field31 in combination with Simple Point Charge (Extended) (SPCE) water model [32]. The structural files of Hb, 1GZX.pdb and 1R1X.pdb with the high˚ were obtained from the Protein Data Bank. The resolution of 1.25 A, file of amino acid (.pdb) was generated by the Visualizer Studio 3.1.
All bound waters were removed from the structural files of Hb. Both the temperature and pressure coupling were applied to the 5 ns molecular simulation. Molecular docking was carried out using the AutoDock 4.0 to assess the conformation of amino acids in the heme pocket with the lowest free binding energy. The parameters including Lamarckian Genetic Algorithm (LGA) with the population size of 150 individuals, maximum of 2.5 million energy evaluations and maximum of 27,000 generations were employed during molecular docking, and other parameters during the simulation were used as default [33]. 2.9. Statistical analysis Experiments were carried out at least in triplicate for each condition. Data represent mean values with error bar denoting the standard deviation of the means. One-way analysis of variance (ANOVA) was performed to assess the statistical significance. A probability of p < 0.05 was considered to be significant difference. 3. Results 3.1. Hb preparation and purification Human Hb was isolated by hemolysis and purified by an ion exchange chromatograph. As shown in Fig. 2A, the first peak represented impurities and the second indicated the majority of Hb. The elution was started after the first peak with a linear gradient of buffer from 100% buffer A to 75% buffer B and followed by a step gradient of 100% buffer B to remove most of the impurity proteins. The majority of Hb was eluted when a large peak appeared in the chromatogram. To detect the purified Hb, the native-PAGE was performed. A single band corresponding to the molecular weight of 64.5 kDa in electrophoresis was observed (Fig. 2B), indicating that the purified human Hb was obtained. 3.2. The effect of pH on Hb stability pH has significant influence on the formation of metHb. Here, the percentage of metHb in the Hb solutions at various pH conditions at 25 ◦ C and 37 ◦ C was determined. As shown in Fig. 3A, at 25 ◦ C, a V-shape of the percentage was formed with approximately 12.70 ± 3.56% Hb converted to metHb at pH 10 at day 1 (p < 0.001), while only 3.08 ± 0.54% at pH 9. With the incubation time increasing, there was still the lowest metHb level at pH 9 than that at other pHs, indicating that the Hb solutions at pH 9 was more stable than that at other pHs. It has been reported that the metHb level at or
270
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
Fig. 3. The percentage of metHb in the Hb solutions under different pH conditions. (A) The metHb content in 50 g/l of Hb solution after 2 days of incubation at pH 7, 8, 9, 9.5 and 10 at 25 ◦ C. (B) The metHb content in 50 g/l of Hb solution after 12 h of incubation at pH 7, 8, 9, 9.5 and 10 at 37 ◦ C (Asterisk ‘*’ indicates the significant difference relative to the lowest metHb percentage in the same group, p < 0.01).
below 10% does not significantly alter the ability of hemoglobin to deliver oxygen to tissues [34]. Therefore, in this study the 10% level of metHb was used as a reference standard. At 37 ◦ C for 6 h, the percentage of metHb was below 10% at pH 7, 8 and 9, and much more than 10% at pH 10 (Fig. 3B). After 12 h, the minimum percentage of metHb was 5.48 ± 0.43%% at pH 8, whereas about 9–10% at pH 7 and 9. However, there was no significant difference in the metHb level between pH 8 and pH 7 and 9 (Fig. 3B). These results revealed that there was relatively lower metHb content at pH 9 at 25 ◦ C, and pH 8 at 37 ◦ C than that at other pHs. 3.3. The metHb formation in the presence of amino acids The Glu, Lys, Trp and Cys were introduced into 50 g/l of Hb solutions at the mole ratio of 24, respectively, and kept at pH 8 and pH 9 at 25 ◦ C and 37 ◦ C. Due to the low solubility of Tyr below pH 10, the pH of Hb solution after the introduction of Tyr was at pH 9.5, therefore the effect of Tyr on the metHb formation was determined at pH 9.5. At pH 8 and 9 at 25 ◦ C, the introduction of Glu, Lys, Trp and Cys did not inhibit the metHb formation. Interestingly, it was noticed that the metHb formation was effectively inhibited by the presence of Tyr at pH 9.5 at 25 ◦ C (Fig. 4A, p1d < 0.001, p2d = 0.001). At 37 ◦ C, the metHb formation was significantly suppressed by the presence of Glu at pH 9, while not by Tyr, Cys, Trp and Lys under other conditions (Fig. 4B). After 12 h of incubation in the presence of Glu, the percentage of metHb in the Hb solutions was 2-fold less than that of the blank at pH 9 (p = 0.036). Since Cys, Trp and Lys had no suppressing effects on the metHb formation at either 25 ◦ C or 37 ◦ C, therefore, Cys, Trp and Lys was not used for further experiments. To test the suppressing effect of Tyr and Glu with storage time, the percentage of metHb in the Hb solutions in the presence of Tyr
Fig. 4. The percentage of metHb in the Hb solutions after incubation with Glu, Lys, Trp, Cys and Tyr under different conditions. (A) The percentage of metHb in the presence of Glu, Lys, Trp and Cys at pH 8 and 9 and Tyr at pH 9.5 at 25 ◦ C. (B) The percentage of metHb in the presence of Glu, Lys, Trp, Cys and Tyr at pH 8 and 9 and Tyr at pH 9.5 at 37 ◦ C. The Hb solution in the absence of amino acids was used as the blank (Asterisk ‘*’ indicates the significant difference in the inhibition of metHb formation relative to the metHb percentage of the blank in the same group, p < 0.05).
for 7 days at 25 ◦ C or Glu for 21 h at 37 ◦ C was determined, respectively. At 25 ◦ C, compared to the percentage of metHb in the Hb solutions without Tyr, the metHb formation during 7 days of incubation was dramatically suppressed by the presence of Tyr (Fig. 5A, p < 0.001). At day 7, the percentage of metHb in the presence of Tyr at pH 9.5 reached 9.61 ± 1.36%, whereas the percentage in the absence of Tyr at pH 9.5 was extremely high, approximately 16% at day 5 (the metHb contents of the controls at pH 9 and 9.5 after day 5 were very high and data was not shown in the histogram). At 37 ◦ C, the metHb formation was significantly inhibited by the presence of Glu at pH 9 (Fig. 5B, p15 h = 0.001, p18 h < 0.001, p21 h < 0.001). After 21 h of incubation with Glu, the metHb formation was 8.59 ± 1.27% at pH 9, just half the percentage of the blank. 3.4. The effect of the amino acid concentration on the metHb formation The concentration of amino acid in the Hb solutions is an important factor affecting the metHb formation. Here we examined the effect of mole ratio of Tyr to Hb on the metHb formation. It was interesting to notice that the concentration of Tyr in the Hb solution did have a significant effect on the suppression of metHb formation. With the increase in the Tyr concentration from the ratio of 8–24 at pH 9.5 at 25 ◦ C, the inhibition became more efficient (Fig. 6). Although there was no statistically significant difference between different ratios at day 1, the percentage of metHb in the Hb solutions was 2.47 ± 0.81% at the ratio of 24, approximately 5 times less than that at the ratio of 8 at day 2 (p = 0.002). Glu had the similar effects on the suppression of metHb formation with Tyr, however, after
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
271
Fig. 5. The percentage of metHb after long-term incubation with Glu and Tyr under different conditions. (A) The percentage of metHb at pH 9.5 at 25 ◦ C in the presence of Tyr for 7 days. (B) The percentage of metHb at pH 8 and 9 at 37 ◦ C in the presence of Glu for 21 h. The Hb solution in the absence of amino acid was used as the blank (Asterisk ‘*’ indicates the significant difference relative to the metHb percentage of the blank in the same group, p < 0.01).
2 days incubation with Glu, the percentage of metHb was much higher than 10% (p = 0.003). 3.5. CD spectra CD spectroscopy was used to reveal the structural changes of proteins at the secondary structure level. CD spectra of the initial Hb and Hb after incubation with Glu or Tyr was recorded in the UV
range of 190–260 nm and in the Soret region. The spectra of the Hb after 3 days of incubation with Tyr at pH 9.5 at 25 ◦ C were practically identical with that of the initial Hb. The spectra showed very high similarity in the maximum at 208 and 222 nm (Fig. 7A). Furthermore, the proportion of ␣-helix for the Hb after the incubation with Tyr for 3 days at 25 ◦ C or with Glu for 18 h at 37 ◦ C was very similar with the initial Hb, at around 63% (Fig. 7B). Although the proportion of ␣-helix for the Hb after the incubation with Tyr for 7 days at 25 ◦ C decreased a little to 57%, the decrease was not statistically significant (p = 0.052). These results suggest that the Hb after the incubation still maintained its secondary structure. In the Soret region, it was interesting to notice that the CD data (ellipticity) was increased by around 25% at about 420 nm (corresponding to the oxygenated Hb) [35] after 3 days of incubation with Tyr at pH 9.5 at 25 ◦ C, similar with the initial position after 7 days of incubation (Fig. 7C). Differently, there was a considerable increase at about 420 nm after 18 h of incubation with Glu at pH 9 at 37 ◦ C in comparison with that after 12 h of incubation. 3.6. Fluorescence spectrum
Fig. 6. The percentage of metHb in the Hb solutions after incubation with different concentrations of Tyr and Glu. Hb solutions with Tyr and Glu were incubated at pH 9.5 at 25 ◦ C and at pH 9 at 37 ◦ C, respectively. The ratios of Tyr or Glu to Hb at 8, 16 and 24 were used (Asterisk ‘*’ indicates the significant difference relative to the ratio of 8 in the same group, p < 0.01).
The intrinsic fluorescence of proteins can provide considerable information about the structure and dynamics since it changes with conformational transitions and denaturation [36]. As shown in Fig. 8A, the fluorescence intensity of Hb solutions was significantly increased at the wavelength of around 320 nm once Tyr was introduced. The increase in the fluorescence intensity of Hb solutions around 320 nm was correlated with the increase in the concentration of Tyr (Supplementary data). The result implies that
272
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
Fig. 9. Native-PAGE for Hb after incubation with Tyr and Glu. Hb after 3 and 7 days of incubation with Tyr at the ratio of 24 at pH 9.5 at 25 ◦ C and after 12 and 18 h of incubation with Glu at the ratio of 24 at pH 9 at 37 ◦ C was used for electrophoresis, respectively. BSA was used as the molecular marker. Gels were stained in coomassie blue G-250 and destained in acetic acid/ethyl alcohol aqueous solution.
3.7. Electrophoresis
Fig. 7. The CD analysis for the secondary structure of Hb. (A) The CD spectra for the secondary structure of Hb. (B) The percentage of ␣-helix in the secondary structure of Hb. The CD data were analyzed using the Jasco-J810 software (p = 0.052). (C) The CD spectra of Hb in the Soret region. Hb after 3 and 7 days of incubation with Tyr at the ratio of 24 at pH 9.5 at 25 ◦ C and 12 and 18 h of incubation with Glu at the ratio of 24 at pH 9 at 37 ◦ C was scanned. The fresh Hb solution (Hb initial) in the absence of amino acids was used as the control.
the increased fluorescence intensity when Tyr is added is probably due to the fluorescence from Tyr itself. With the incubation time increasing at pH 9.5 at 25 ◦ C, the fluorescence intensity of Hb solutions was just slightly increased after 7d incubation compared with that of Hb solutions at the very beginning of incubation. In addition, there was no significant change in the fluorescence intensity of Hb in the presence of Glu at pH 9 at 37 ◦ C (Fig. 8B).
To test the integrity of Hb after incubation with amino acids, the native-PAGE was performed. Electrophoresis analysis revealed that only a single band corresponding to the molecular weight of 64.5 kDa was observed after 3 and 7 days of incubation with Tyr at pH 9.5 at 25 ◦ C and after 12 and 18 h of incubation with Glu at pH 9 at 37 ◦ C (Fig. 9). This result indicated that Hb did not aggregate or degrade after the incubation in the presence of Tyr or Glu under above conditions. 3.8. Molecular simulations Ratios of 1, 8, 16 and 24 were chosen to test the effect of ratios of amino acid to Hb on the adsorption of amino acid to Hb using the MD simulation. After 5 ns, the system reached equilibrium with little change in root mean square deviation (RSMD). At the low concentration of Tyr, there was no adsorption between Tyr and Hb (Fig. 10A). With the increase in the concentration of Tyr to the ratio of 16, Tyr started to adsorb to the region closed to the heme. At the ratio of 24, two Tyr molecules (the molecules in yellow color
Fig. 8. Fluorescence spectra of Hb at pO2 of 144 Torr in aqueous environment. (A) The fluorescence intensity of Hb after the incubation with Tyr at pH 9.5 at 25 ◦ C. (B) The fluorescence intensity of Hb after the incubation with Glu pH 9 at 37 ◦ C.
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
273
Fig. 10. The interactions of Tyr and Glu with Hb based on the molecular simulation. (A) The snapshot of Tyr or Glu to Hb at 5 ns at different Tyr/Hb ratios at 25 ◦ C simulated using the Gromacs. Yellow: Tyr bound to heme region; gray: Tyr not bound to heme region. (B) The combining conformation of Hb with Tyr, Glu, Trp, Lys and Cys simulated using the AutoDock. (C) The heme pocket size (distance between the backbone of His58 and His87) at 5 ns in the presence of Tyr at 25 ◦ C and Glu at 37 ◦ C simulated using Gromacs. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
in Fig. 10A) were interacted with two heme sites. Similar patterns were observed for Glu (Fig. 10A). The conformation of Tyr, Glu, Trp, Lys and Cys in the heme pocket at their lowest free binding energy was calculated using the AutoDock. The results were shown in Fig. 10B. In the presence of Tyr, hydrogen bonds were formed between carboxy terminal of Tyr and His58, carboxy terminal of Tyr and Lys61, and between amino terminal of Tyr and porphyrin ring. The aromatic side chain of Tyr was interacted with Phe46 and His45. In the presence of Glu and Cys, hydrogen bonds were generated between carboxy terminal of Glu and His58, carboxy terminal of Glu and Lys61, and between
amino terminal of Glu and porphyrin ring, but Glu was not interacted with Phe46 and His45. In contrast, there was no hydrogen bond formed between carboxy terminal of Trp/Lys and His 58, His 45, Lys 61 or Phe46. Fig. 10C shows the heme pocket size according to the MD simulation when the Hb solutions were incubated with Tyr at 25 ◦ C and Glu at 37 ◦ C, respectively. There was little change in the RMSD after 5 ns. It was interesting to notice that the distance between the backbone of His58 and His87 varied with temperatures and amino acids in the system. In the presence of Tyr, the heme pocket size was 14.09 A˚ at 25 ◦ C. When the temperature was elevated to 37 ◦ C,
274
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
˚ However, in the presence of the pocket size increased to 14.77 A. ˚ 0.78 A˚ larger than that in Glu, the pocket size increased to 14.85 A, the presence of Tyr at 25 ◦ C. When the temperature was elevated ˚ to 37 ◦ C, the pocket size decreased to 14.36 A.
4. Discussion The high stability of Hb is essential for the Hb containing systems when used as blood substitutes. This study was designed to investigate the effects of amino acids as the potential stablizers on preventing Hb from the metHb formation under different conditions, including pH, temperature and the ratio of amino acids to Hb. The results provide a possible clue for improving the stability of Hb and minimizing the metHb formation in aqueous environment under aerobic conditions in vitro. Previous study indicates that the metHb formation is strongly depended on the pH [3,28]. The metHb formation in aqueous environment at 25 ◦ C was in a V-shape with the lowest percentage at pH 9 (Fig. 3A), which was consistent with the previous study on bovine heart myoglobin [37]. However, the pH at which there was the lowest percentage of metHb shifted from pH 9 to pH 8 at 37 ◦ C (Fig. 3B). This is probably caused by the change in the energy barrier which works as an electron transfer from the Fe (II) to O2 leading to the displacement of O2 − from Hb [3]. In this study, Cys, Trp and Tyr were selected as the potential stabilizers to inhibit the metHb formation for their capacities of reduction [38]. Lys as a positive charge amino acid may be able to have interactions with the negative charge of Hb at the physiological pH and therefore was chosen as another potential stabilizer. Similarly, Glu, a negative charge amino acid, was also chosen in this study since its potential interactions with hemes through the positive charge in their surrounding area. The experimental results demonstrated that Glu at 37 ◦ C and Tyr at 25 ◦ C can suppress the metHb formation (Figs. 4 and 5). The mechanism behind the inhibiting effect induced by Glu and Tyr is probably caused by the network formed between carboxy terminal of Tyr/Glu and His58, His45, Lys61 or Phe46, which could inhibit the formation of an activated state between distal histidine and O2 , and the electron transfer from Fe (II) to the bound O2 , and further lead to the suppressing of metHb formation. In contrast, Trp and Lys are not able to prohibit the metHb formation which may be due to no hydrogen bonds generated between the amino acid and His58 and Lys61 (Fig. 10B). In the presence of Cys, although the hydrogen bond was generated between Cys and distal histidine, no inhibiting effect of Cys on the metHb formation was observed in aqueous environment under aerobic conditions (Fig. 4). This is consistent with the previous study demonstrating that the introduction of Cys under aerobic conditions leads to the generation of active oxygens during the autoxidation of Cys and significantly promotes the conversion of Hb to metHb [5]. This adverse effect on the conversion of Hb to metHb may blanket the capacity of inhibiting the formation of an activated state between distal histidine and O2 , and the electron transfer from Fe (II) to the bound O2 . This could be the reason why the presence of Cys did not result in the suppression of metHb under aerobic conditions, conversely accelerate the metHb formation (Fig. 4). Hb stability is related to protein matrices which act as a breakwater of the FeO2 center against aqueous solvent [39]. The metHb formation can be efficiently inhibited in aqueous environment under aerobic condition when FeO2 is embedded in protein matrices, which eventually leads to the slowing down of the autoxidation rate of Hb [40]. The results from the molecular simulations indicated that the adsorption of Tyr or Glu to the region closed to the hemes occurred at the mole ratio of 24 (Fig. 10A). According to
the results from the CD spectra in the Soret region, the conformation surrounding the heme region may be changed due to the introduction of Tyr and Glu (Fig. 7C). The interactions between the heme group and Tyr or Glu in aqueous environment may become an extra barrier against the nucleophilic displacement and the following transfer of an electron from Fe (II) to the bound O2 through a hydrogen bond of the globin moiety [3]. In addition, the hydrogen bond generated between amino acid (Tyr or Glu) and distal histidine could inhibit the formation of an activated state between distal histidine and dioxygen and the electron transfer from Fe (II) to the bound O2 [40]. These actions may play critical roles in the stabilization of Hb. Glu and Tyr have been shown to considerably suppress the metHb formation in rats in vivo and in isolated rat blood [25] and under anaerobic conditions [6,22,23]. However, this study was the first to point out that the inhibiting effect of amino acid on metHb formation was dependent on temperature. The presence of Glu and Tyr was capable to inhibit the metHb formation at 37 ◦ C and 25 ◦ C, respectively (Figs. 4 and 5). The simulation results suggested that the presence of Tyr led to an increase in the heme pocket size when the temperature was elevated from 25 ◦ C to 37 ◦ C. Differently, the introduction of Glu resulted in a decrease in the heme pocket size when the temperature was elevated from 25 ◦ C to 37 ◦ C (Fig. 10C). The increase in the size of the heme pocket allows the entering of water molecules to attack FeO2 center on both acidic and basic side [3]. The size of the heme pocket for the initial crystal structure was 14.47 A˚ (Supplementary data), which ˚ but was larger than that in the presence of Tyr at 25 ◦ C (14.09 A) ˚ In contrast, an opposite trend smaller than that at 37 ◦ C (14.77 A). was observed in the presence of Glu. Therefore, the variation of the heme pocket may be an elucidation for the different suppressing effects on the metHb formation provided by Tyr and Glu at the different temperatures. The various capacities of Tyr and Glu on the inhibition of metHb formation at different temperatures imply that the stabilization of Hb may be controlled by the microenvironment around the heme group, including the formation of breakwater network, the status of distal histidine and the size of heme pocket. The CD spectroscopy is used to determine the conformation of polypeptide chains of proteins. When a polypeptide chain of a protein has ␣-helical conformation, there are two negative peaks at 208 nm and 222 nm in the CD spectra [41,42]. The proportion of ␣-helix for the initial Hb and the Hb after incubation with Tyr for 7 days at 25 ◦ C or with Glu for 18 h at 37 ◦ C did not show significant changes (Fig. 7), suggesting that Hb still maintained its secondary structure. Furthermore, the native-PAGE indicated that Hb did not aggregate or degrade after the incubation at 25 ◦ C and 37 ◦ C at pO2 of 144 Torr in aqueous environment (Fig. 9), suggesting that the denaturation of Hb in aqueous environment under aerobic conditions was mainly caused by the metHb formation rather than by the degradation or aggregation. In the Soret region, the changes at about 420 nm after the incubation with Tyr at 25 ◦ C or with Glu at 37 ◦ C indicates that Tyr or Glu may change the conformation surrounding of the hemes through the potential interactions with the surrounding residues of the hemes (Figs. 7C and 10B), and further change the microenvironment around the hemes [43]. The structural changes of proteins can also be reflected by the changes of protein fluorescence intensity. When Tyr was introduced into Hb solutions, the fluorescence intensity was significantly increased compared with the fresh Hb solutions, implying that the presence of Tyr in the Hb solutions could be a reason for the dramatic changes in the fluorescence intensity. The fluorescence intensity with the emission at around 320 nm was not markedly increased with the incubation time when Tyr was present, suggesting there was no significantly change in the structure of Hb (Fig. 8A). The fluorescence spectra of Hb after 18 h of incubation with Glu at pH 9 at
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
37 ◦ C was also not significantly changed compared with the fresh Hb solutions (Fig. 8B). 5. Conclusions In this study, Tyr and Glu can effectively inhibit the formation of metHb without leading to changes in the structure of Hb in Hb-containing systems in aqueous environment under aerobic conditions at different temperatures. The suppressing effect induced by the presence of Tyr and Glu may be related to the change in the microenvironment around the heme group, including the formation of the breakwater network, the change in the size of heme pocket and the stabilization of distal histidine. Inhibition of the metHb formation achieved by the introduction of Tyr and Glu may be much easier and less expensive than other methods, such as engineered protein. These results may provide a new insight into the design for Hb-oxygen based carriers with high stability of Hb in aqueous environment under aerobic conditions, and even at physiological temperature in vitro. Acknowledgements This research was supported by the Chinese Academy of Sciences, and the National Natural Science Foundation of China (21176238). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. ijbiomac.2013.12.010. References [1] [2] [3] [4]
D. Bossi, B. Giardina, Mol. Aspects Med. 17 (1996) 117–128. C.C. Winterbourn, Environ. Health Persp. 64 (1985) 321–330. K. Shikama, Prog. Biophys. Mol. Biol. 91 (2006) 83–162. L.P. Stratton, A.S. Rudolph, W.K. Knoll Jr., S. Bayne, M.C. Farmer, Hemoglobin 12 (1988) 353–368. [5] S. Takeoka, H. Sakai, T. Kose, Y. Mano, Y. Seino, H. Nishide, E. Tsuchida, Bioconjug. Chem. 8 (1997) 539–544. [6] T. Atoji, M. Aihara, H. Sakai, E. Tsuchida, S. Takeoka, Bioconjug. Chem. 17 (2006) 1241–1245. [7] T. Kanias, J.P. Acker, Cryobiology 58 (2009) 232–239.
275
[8] G.M. Tush, R.J. Kuhn, Ann. Pharmacother. 30 (1996) 1251–1254. [9] A.S. Rudolph, A. Sulpizio, P. Hieble, V. MacDonald, M. Chavez, G. Feuerstein, J. Appl. Physiol. 82 (1997) 1826–1835. [10] H. Sakai, H. Horinouchi, K. Tomiyama, E. Ikeda, S. Takeoka, K. Kobayashi, E. Tsuchida, Am. J. Pathol. 159 (2001) 1079–1088. [11] H. Sakai, P. Cabrales, A.G. Tsai, E. Tsuchida, M. Intaglietta, Am. J. Physiol. Heart C 288 (2005) H2897–H2903. [12] K. Chen, T.J. Merkel, A. Pandya, M.E. Napier, J.C. Luft, W. Daniel, S. Sheiko, J.M. DeSimone, Biomacromolecules 13 (2012) 2748–2759. [13] Y. Xiong, A. Steffen, K. Andreas, S. Muller, N. Sternberg, R. Georgieva, H. Baumler, Biomacromolecules 13 (2012) 3292–3300. [14] R.M. Winslow, Nat. Med. 3 (1997) 474. [15] E. Tsuchida, Artificial Red Cells: Materials, Performances, and Clinical Study as Blood Substitutes, Wiley, New York, 1995, pp. 288. [16] T.M. Chang, Artif. Cells Blood Sub. 25 (1997) 1–24. [17] L. Djordjevich, I.F. Miller, Exp. Hematol. 8 (1980) 584–592. [18] F. D’Agnillo, A.I. Alayash, Blood 98 (2001) 3315–3323. [19] J.G. Riess, Chem. Rev. 101 (2001) 2797–2920. [20] B.A. Kerwin, M.C. Heller, S.H. Levin, T.W. Randolph, J. Pharm. Sci. 87 (1998) 1062–1068. [21] M.C. Heller, J.F. Carpenter, T.W. Randolph, J. Pharm. Sci. 85 (1996) 1358–1362. [22] S. Takeoka, T. Ohgushi, H. Sakai, T. Kose, H. Nishide, E. Tsuchida, Artif. Cells Blood Sub. 25 (1997) 31–41. [23] H. Sakai, K.I. Tomiyama, K. Sou, S. Takeoka, E. Tsuchida, Bioconjug. Chem. 11 (2000) 425–432. [24] R. Di Domenico, R. Lavecchia, Biotechnol. Lett. 22 (2000) 335–339. [25] A.M. Genkin, M.S. Volkov, Bull. Exp. Biol. Med. 49 (1960) 486–487. [26] Y. Teramura, H. Kanazawa, H. Sakai, S. Takeoka, E. Tsuchida, Bioconjug. Chem. 14 (2003) 1171–1176. [27] V. Nadithe, Y.H. Bae, Int. J. Biol. Macromol. 47 (2010) 603–613. [28] W.D. Brown, L.B. Mebine, J. Biol. Chem. 244 (1969) 6696–6701. [29] C.T. Andrade, L.A. Barros, M.C. Lima, E.G. Azero, Int. J. Biol. Macromol. 34 (2004) 233–240. [30] G. Sun, A.F. Palmer, J. Chromatogr. B 867 (2008) 1–7. [31] R.E. Benesch, R. Benesch, S. Yung, Anal. Biochem. 55 (1973) 245–248. [32] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark, H.J. Berendsen, J. Comput. Chem. 26 (2005) 1701–1718. [33] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, J. Comput. Chem. 30 (2009) 2785–2791. [34] R. Linberg, C.D. Conover, K.L. Shum, R.G. Shorr, Cells Blood Sub. 26 (1998) 133–148. [35] M. Nagai, Y. Sugita, Y. Yoneyama, J. Biol. Chem. 244 (1969) 1651–1653. [36] T. Chen, S. Zhu, H. Cao, Y. Shang, M. Wang, G. Jiang, Y. Shi, T. Lu, Spectrochim. Acta A 78 (2011) 1295–1301. [37] Y. Sugawara, A. Matsuoka, A. Kaino, K. Shikama, Biophys. J. 69 (1995) 583–592. [38] S. Takeoka, Y. Teramura, T. Atoji, E. Tsuchida, Bioconjug. Chem. 13 (2002) 1302–1308. [39] M. Tsuruga, A. Matsuoka, A. Hachimori, Y. Sugawara, K. Shikama, J. Biol. Chem. 273 (1998) 8607–8615. [40] K. Shikama, Chem. Rev. 98 (1998) 1357–1374. [41] T. Chatterjee, A. Pal, S. Dey, B.K. Chatterjee, P. Chakrabarti, PLoS ONE 7 (2012) e37468. [42] N. Sreerama, R.W. Woody, Method. Enzymol. 383 (2004) 318–351. [43] Q. Chen, I. Lalezari, R.L. Nagel, R.E. Hirsch, Biophys. J. 88 (2005) 2057–2067.
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Inhibition of methemoglobin formation in aqueous solutions under aerobic conditions by the addition of amino acids Yuping Wei a,b , Chunlong Li a,b , Liang Zhang a,b , Zhiguo Su a , Xia Xu a,∗ a b
State Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, PR China University of Chinese Academy of Sciences, Beijing 100049, China
a r t i c l e
i n f o
Article history: Received 22 July 2013 Received in revised form 6 December 2013 Accepted 8 December 2013 Available online 14 December 2013 Keywords: Amino acid Hemoglobin Methemoglobin Molecular simulation Stability
a b s t r a c t Hemoglobin (Hb) as an important iron-containing oxygen-transport protein is easily oxidized to the ferric met-form, methemoglobin (metHb), and loses the capacity of binding oxygen during storage. In this study, the experimental data indicate that the presence of Tyr and Glu significantly suppress the metHb formation in the Hb solutions in aqueous environment under aerobic conditions at the temperature of 25 and 37 ◦ C, respectively. At pO2 of 144 Torr the metHb percentage in the Hb solutions was the lowest with less than 10% at day 7 after incubation with Tyr at the ratio of 24 at pH 9.5 at 25 ◦ C. At 37 ◦ C, the metHb percentage did not reach 5% after 12 h of incubation with Glu at the ratio of 24 at pH 9. Molecular simulation analysis suggest that the presence of Tyr or Glu may contribute to the formation of the breakwater network, the stabilization of distal histidine, the changes in the size of heme pocket, and eventually result in the inhibition of metHb formation. This study provides insight into a new design for Hb-oxygen based carriers with strongly inhibition of metHb formation in aqueous environment under aerobic conditions, even at physiological temperature in vitro. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Hemoglobin (Hb), a tetrametric protein in red blood cells, is composed of four polypeptide chains (two ␣- and two -chains), each of which containing a hydrophobic pocket bound to a prosthetic heme group. This group consists of a Fe(II) ion and a porphyrin ring as the O2 -binding site to carry and deliver oxygen. In the prosthetic heme group, the iron atom locating at the center of the porphyrin ring forms four bonds with porphyrin nitrogen, one covalent bond with the histidine residue and one bond with exogenous ligands (Fig. 1) [1]. The formation of metHb causes Hb to lose the capacity of binding oxygen under aerobic conditions. At a steady state the distal His58 stabilizes the bound dioxygen through a hydrogen bond. Upon the nucleophilic attack of water molecules, the FeO2 center is open to the proton on the acidic pH side or hydroxide anion on the basic side. This results in an irreversible displacement of the bound
Abbreviations: BSA, bovine serum albumin; CD, circular dichroism; Hb, hemoglobin; metHb, methemoglobin; PAGE, polyacrylamide gel electrophoresis; ROS, reactive oxygen species; Tris–HCl, Tris[hydroxymethyl-aminomethane]–HCl; TEMED, N,N,N ,N -tetramethyl ethylenediamine. ∗ Corresponding author at: 1 Beiertiao, Zhongguancun Haidian District, Beijing 100190, PR China. Tel.: +86 10 82544939; fax: +86 82544939. E-mail address: [email protected] (X. Xu). 0141-8130/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijbiomac.2013.12.010
dioxygen in the form of superoxide ion (O2 − ) and a replacement of the sixth coordination position originally binding with oxygen by water or hydroxide [2] and eventually leads to the formation of metHb [2,3]. Therefore, the inhibition of metHb formation is essential to preservation of Hb before clinical applications. In vivo the percentage of metHb in red blood cells is less than 5% due to the presence of systematic reduction. However, in vitro due to the absence of the metHb reduction system which is removed during processes of Hb purification [4–7], the metHb formation and protein denaturation are the major problems in the Hb-containing systems and acellular-type Hb-based oxygen carriers [8]. These Hb-containing systems and Hb-based oxygen carriers, such as liposome-encapsulated Hb [9], Hb vesicles [10,11], and microparticles with Hb loading [12–15], are considered as potential candidates for red-blood-cell substitutes [16–19]. It is hence that the efficient inhibition of metHb formation in the Hb-containing systems and Hb-based oxygen carriers has far-reaching significance in clinical applications. Various approaches including cryopreservation and freezedrying have been investigated to stabilize Hb [20,21]. However, these methods are time-consuming and process-complicated. Hb preserved either alone or in oxygen carriers in aqueous environment is much more convenient for infusion in emergency situations. Therefore, the inhibition of metHb formation is the primary method for the stabilization of Hb during its preservation in aqueous environment. Previous reports have demonstrated that
268
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
Fig. 1. Schematic representation of Hb autoxidation process in aqueous environment. At steady state the distal histidine (58) stabilizes the bound dioxygen through a hydrogen bond. Electron transfer from N to N at His58 weakens hydrogen bond between His58 and FeO2 center. Nucleophilic attack of entering water molecule can result in the irreversible displacement of the bound dioxygen in the form of superoxide ion (O2 − ), and facilitate a full charge transfer from Fe (II) to O2 . This eventually causes the formation of metHb.
under anaerobic conditions the introduction of reductants or L-Tyr can suppress the metHb formation [6,22,23]. The thermal stability of Hb in aqueous environment can be improved in the presence of sarcosine and sorbitol [24]. MetHb formation can also be inhibited by the addition of glutamate [25] and antioxidant enzymes [26,27]. In addition, the metHb formation is strongly pH-dependent, and the nucleophilic attack to FeO2 during the metHb formation is more significantly enhanced under acidic conditions than alkaline [3,28]. Temperature should be another important factor to affect protein stability. However, the report about the effect of temperature on metHb formation is rare. Although it is contributive to the stabilization of Hb by pH adjustment and temperature control, the methods for suppressing the metHb formation in aqueous environment under aerobic conditions are still needed to be improved. The aim of this study was to suppress the metHb formation in aqueous environment under aerobic conditions. Due to the high stability, low cost and good safety, amino acids are the potential stabilizers of proteins. This study was the first report to examine the effects of five different amino acids on the inhibition of metHb formation at different pHs and different temperatures, and further the effect of the mole ratio of amino acid to Hb on the metHb formation. The possible mechanism behind the inhibiting effect of amino acids on the metHb formation was discussed based on the molecular simulation. 2. Experimental procedures 2.1. Materials Donated red blood cells were obtained from the Anzhen Hospital (Beijing, China). Tris[hydroxymethyl-aminomethane]–HCl (Tris–HCl), acetonitrile, methanol, glycine and NaCl were purchased from Sinopham Chemical Reagent (China). Acrylamide, N,N -methylene bisacrylamide, coomassie blue G-250, acetic acid/ethyl alcohol and ammonium persulfate were purchased from Xinglong (China). N,N,N ,N -tetramethyl ethylenediamine (TEMED) was purchased from Merck (UK). l-Trp, l-Cys, l-Lys, l-Tyr, sodium glutamate and bovine serum albumin (BSA, MW66000) were purchased from Sigma–Aldrich (USA). Anion exchangers and Q Sepharose Fast Flow (Q-SFF) resin used in this study were produced from GE Healthcare (USA). Dialysis membrane with cut-off molecular weight of 3500 Da was purchased from Union Carbide (USA). 2.2. Preparation and purification of Hb from red blood cells Hb was isolated from the donated blood using hemolysis. The red blood cells were suspended in 20 mM Tris–HCl buffer
(pH 8.0) at the ratio of 1:2 (v/v) at 4 ◦ C for 24 h. After hemolysis, the suspension of lysed cells was centrifuged at 2300 × g for 1 h at 4 ◦ C. The supernatant was purified using the ion exchange chromatography method with Q-SFF resin at room temperature [29,30]. Briefly, the supernatant was loaded with an injector and eluted with a linear gradient of buffer from 100% buffer A (20 mM Tris–HCl, pH 8.0) to 75% buffer B (20 mM Tris–HCl plus 0.2 M NaCl, pH 8.0) in 5-fold volume of column. The elution was monitored with a UV detector at 280 nm. The Hb solution after purification was concentrated to 150 g/l using a dialysis membrane with the cut-off molecular weight of 3500 Da. 2.3. Storage 150 g/l of Hb in 20 mM Tris–HCl buffer was diluted to 50 g/l using Tris–HCl buffer (pH 7.4). The Hb diluted solutions were incubated at pH 7, 8, 9, 9.5 and 10 (20 mM Boric acid sodium hydroxide buffer for pH 10) at pO2 of 144 Torr at 25 or 37 ◦ C. The stock solutions of Glu, Lys, Tyr, Cys and Trp were prepared at the concentration of 0.04 M at pH 7. For Tyr, the stock solution was prepared at around pH 10 due to the low solubility of Tyr below pH 10. Glu, Lys, Cys, Tyr or Trp was introduced to the Hb solutions at the mole ratio of 24, respectively, and the pH was adjusted to the required pH using the Tris–HCl buffer (20 mM). Then the Hb solutions were incubated at 25 ◦ C for up to 7 days or 37 ◦ C for up to 21 h at pO2 of 144 Torr at pH 8, 9 and 9.5, respectively. In addition, Tyr and Glu were introduced in the Hb solutions at the mole ratio of 8, 16 and 24 at pH 9 and 9.5, respectively, and incubated at 25 ◦ C for up to 2 days. The pH at the end point for all the tested conditions was measured using a pH meter (Sartorius, Germany). 2.4. Measurement of metHb content The Hb solutions were diluted from 50 g/l to 0.5 g/l before assay. The pH of the diluted Hb solutions measured using a pH meter was in the range of 6–8 for all the tested conditions. The absorption of the diluted Hb solutions (0.5 g/l) at 560, 576 and 630 nm was measured using a UV-Vis spectrometer (Shanghai Spectrum, China). The content of metHb was determined using the following equations described by Benesch et al., which was valid between pH 6.2 and pH 8.8 [31]. A560 = ε560 CdeoxyHb + ε560 CoxyHb + ε560 CmetHb
(1)
A576 = ε576 CdeoxyHb + ε576 CoxyHb + ε576 CmetHb
(2)
A630 = ε630 CdeoxyHb + ε630 CoxyHb + ε630 CmetHb
(3)
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
269
Fig. 2. The purification of human Hb. (A) A typical elution pattern for the purified human Hb. Column (200 mm in length, 26 mm in internal diameter) was packed with 60 ml Q-sepharose fast flow resin. RBC lysate injected into the column via Superloop (5 ml) was eluted using a linear gradient of buffer from 100% buffer A to 75% buffer B and followed by 100% buffer B. The highest peak represents Hb, other peaks represent impurities. (B) Native-PAGE of the purified Hb solution. Native stacking gel with 3% acrylamide (pH 6.8) and native separating gel with 12.5% acrylamide (pH 8.8) were used. BSA was used as the molecular marker. Gels were stained in coomassie blue G-250 and destained in acetic acid/ethyl alcohol aqueous solution.
where A560 , A576 and A630 are the absorption at 560, 576 and 630 nm, ε560 , ε576 and ε630 are the extinction coefficients at 560, 576 and 630 nm chosen according to the pH of solutions as described by Benesch et al. [31]. ε560 , ε576 and ε630 are 0.749, 0.850 and 0.208 at pH 8.8, 0.405, 0.406 and 0.401 at pH 7.0, 0.373, 0.358 and 0.412 at pH 6.2, respectively. CdeoxyHb , CoxyHb and CmetHb are the mole concentrations of deoxy, oxy and met hemoglobin, respectively. 2.5. Circular dichroism (CD) spectra The secondary structure of Hb was determined using a CD spectrometer (Jasco-J810, Japan) at 25 ◦ C. The spectra were scanned at 200 nm/min with a response time of 1 s and recorded in the range of 190–260 nm and in the Soret region. The results were expressed as ellipticity in mdeg. The proportion of ␣ helix in Hb was calculated using the Jasco-J810 analysis software. 2.6. Fluorescence measurements The structural changes of Hb after incubation with Tyr at pH 9.5 at 25 ◦ C or Glu at pH 9 at 37 ◦ C were investigated using a Hitachi-4000 spectrophotometer (Japan). The emission spectra were recorded between 300 nm and 500 nm at the excitation wavelength of 280 nm. 2.7. Electrophoresis Polyacrylamide gel electrophoresis (PAGE) was performed in a single-sided vertical system equipped with an electrophoresis power supply (Tanon, China). Native stacking gel (3% acrylamide, pH 6.8) and native separating gel (12.5% acrylamide, pH 8.8) were prepared using TEMED with ammonium persulfate to catalyze the polymerization of acrylamide. The upper and lower tank buffers (pH 8.8) contained glycine as the mobile anion. The samples with equal volume of loading buffer were added into the wells. BSA was used as the molecular marker. Gels were stained in coomassie blue G-250 and destained in acetic acid/ethyl alcohol aqueous solution. 2.8. Molecular simulation Molecular dynamics (MD) simulation was performed using the Gromacs 4.5.4 and the Gromos96 G43a1 force field31 in combination with Simple Point Charge (Extended) (SPCE) water model [32]. The structural files of Hb, 1GZX.pdb and 1R1X.pdb with the high˚ were obtained from the Protein Data Bank. The resolution of 1.25 A, file of amino acid (.pdb) was generated by the Visualizer Studio 3.1.
All bound waters were removed from the structural files of Hb. Both the temperature and pressure coupling were applied to the 5 ns molecular simulation. Molecular docking was carried out using the AutoDock 4.0 to assess the conformation of amino acids in the heme pocket with the lowest free binding energy. The parameters including Lamarckian Genetic Algorithm (LGA) with the population size of 150 individuals, maximum of 2.5 million energy evaluations and maximum of 27,000 generations were employed during molecular docking, and other parameters during the simulation were used as default [33]. 2.9. Statistical analysis Experiments were carried out at least in triplicate for each condition. Data represent mean values with error bar denoting the standard deviation of the means. One-way analysis of variance (ANOVA) was performed to assess the statistical significance. A probability of p < 0.05 was considered to be significant difference. 3. Results 3.1. Hb preparation and purification Human Hb was isolated by hemolysis and purified by an ion exchange chromatograph. As shown in Fig. 2A, the first peak represented impurities and the second indicated the majority of Hb. The elution was started after the first peak with a linear gradient of buffer from 100% buffer A to 75% buffer B and followed by a step gradient of 100% buffer B to remove most of the impurity proteins. The majority of Hb was eluted when a large peak appeared in the chromatogram. To detect the purified Hb, the native-PAGE was performed. A single band corresponding to the molecular weight of 64.5 kDa in electrophoresis was observed (Fig. 2B), indicating that the purified human Hb was obtained. 3.2. The effect of pH on Hb stability pH has significant influence on the formation of metHb. Here, the percentage of metHb in the Hb solutions at various pH conditions at 25 ◦ C and 37 ◦ C was determined. As shown in Fig. 3A, at 25 ◦ C, a V-shape of the percentage was formed with approximately 12.70 ± 3.56% Hb converted to metHb at pH 10 at day 1 (p < 0.001), while only 3.08 ± 0.54% at pH 9. With the incubation time increasing, there was still the lowest metHb level at pH 9 than that at other pHs, indicating that the Hb solutions at pH 9 was more stable than that at other pHs. It has been reported that the metHb level at or
270
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
Fig. 3. The percentage of metHb in the Hb solutions under different pH conditions. (A) The metHb content in 50 g/l of Hb solution after 2 days of incubation at pH 7, 8, 9, 9.5 and 10 at 25 ◦ C. (B) The metHb content in 50 g/l of Hb solution after 12 h of incubation at pH 7, 8, 9, 9.5 and 10 at 37 ◦ C (Asterisk ‘*’ indicates the significant difference relative to the lowest metHb percentage in the same group, p < 0.01).
below 10% does not significantly alter the ability of hemoglobin to deliver oxygen to tissues [34]. Therefore, in this study the 10% level of metHb was used as a reference standard. At 37 ◦ C for 6 h, the percentage of metHb was below 10% at pH 7, 8 and 9, and much more than 10% at pH 10 (Fig. 3B). After 12 h, the minimum percentage of metHb was 5.48 ± 0.43%% at pH 8, whereas about 9–10% at pH 7 and 9. However, there was no significant difference in the metHb level between pH 8 and pH 7 and 9 (Fig. 3B). These results revealed that there was relatively lower metHb content at pH 9 at 25 ◦ C, and pH 8 at 37 ◦ C than that at other pHs. 3.3. The metHb formation in the presence of amino acids The Glu, Lys, Trp and Cys were introduced into 50 g/l of Hb solutions at the mole ratio of 24, respectively, and kept at pH 8 and pH 9 at 25 ◦ C and 37 ◦ C. Due to the low solubility of Tyr below pH 10, the pH of Hb solution after the introduction of Tyr was at pH 9.5, therefore the effect of Tyr on the metHb formation was determined at pH 9.5. At pH 8 and 9 at 25 ◦ C, the introduction of Glu, Lys, Trp and Cys did not inhibit the metHb formation. Interestingly, it was noticed that the metHb formation was effectively inhibited by the presence of Tyr at pH 9.5 at 25 ◦ C (Fig. 4A, p1d < 0.001, p2d = 0.001). At 37 ◦ C, the metHb formation was significantly suppressed by the presence of Glu at pH 9, while not by Tyr, Cys, Trp and Lys under other conditions (Fig. 4B). After 12 h of incubation in the presence of Glu, the percentage of metHb in the Hb solutions was 2-fold less than that of the blank at pH 9 (p = 0.036). Since Cys, Trp and Lys had no suppressing effects on the metHb formation at either 25 ◦ C or 37 ◦ C, therefore, Cys, Trp and Lys was not used for further experiments. To test the suppressing effect of Tyr and Glu with storage time, the percentage of metHb in the Hb solutions in the presence of Tyr
Fig. 4. The percentage of metHb in the Hb solutions after incubation with Glu, Lys, Trp, Cys and Tyr under different conditions. (A) The percentage of metHb in the presence of Glu, Lys, Trp and Cys at pH 8 and 9 and Tyr at pH 9.5 at 25 ◦ C. (B) The percentage of metHb in the presence of Glu, Lys, Trp, Cys and Tyr at pH 8 and 9 and Tyr at pH 9.5 at 37 ◦ C. The Hb solution in the absence of amino acids was used as the blank (Asterisk ‘*’ indicates the significant difference in the inhibition of metHb formation relative to the metHb percentage of the blank in the same group, p < 0.05).
for 7 days at 25 ◦ C or Glu for 21 h at 37 ◦ C was determined, respectively. At 25 ◦ C, compared to the percentage of metHb in the Hb solutions without Tyr, the metHb formation during 7 days of incubation was dramatically suppressed by the presence of Tyr (Fig. 5A, p < 0.001). At day 7, the percentage of metHb in the presence of Tyr at pH 9.5 reached 9.61 ± 1.36%, whereas the percentage in the absence of Tyr at pH 9.5 was extremely high, approximately 16% at day 5 (the metHb contents of the controls at pH 9 and 9.5 after day 5 were very high and data was not shown in the histogram). At 37 ◦ C, the metHb formation was significantly inhibited by the presence of Glu at pH 9 (Fig. 5B, p15 h = 0.001, p18 h < 0.001, p21 h < 0.001). After 21 h of incubation with Glu, the metHb formation was 8.59 ± 1.27% at pH 9, just half the percentage of the blank. 3.4. The effect of the amino acid concentration on the metHb formation The concentration of amino acid in the Hb solutions is an important factor affecting the metHb formation. Here we examined the effect of mole ratio of Tyr to Hb on the metHb formation. It was interesting to notice that the concentration of Tyr in the Hb solution did have a significant effect on the suppression of metHb formation. With the increase in the Tyr concentration from the ratio of 8–24 at pH 9.5 at 25 ◦ C, the inhibition became more efficient (Fig. 6). Although there was no statistically significant difference between different ratios at day 1, the percentage of metHb in the Hb solutions was 2.47 ± 0.81% at the ratio of 24, approximately 5 times less than that at the ratio of 8 at day 2 (p = 0.002). Glu had the similar effects on the suppression of metHb formation with Tyr, however, after
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
271
Fig. 5. The percentage of metHb after long-term incubation with Glu and Tyr under different conditions. (A) The percentage of metHb at pH 9.5 at 25 ◦ C in the presence of Tyr for 7 days. (B) The percentage of metHb at pH 8 and 9 at 37 ◦ C in the presence of Glu for 21 h. The Hb solution in the absence of amino acid was used as the blank (Asterisk ‘*’ indicates the significant difference relative to the metHb percentage of the blank in the same group, p < 0.01).
2 days incubation with Glu, the percentage of metHb was much higher than 10% (p = 0.003). 3.5. CD spectra CD spectroscopy was used to reveal the structural changes of proteins at the secondary structure level. CD spectra of the initial Hb and Hb after incubation with Glu or Tyr was recorded in the UV
range of 190–260 nm and in the Soret region. The spectra of the Hb after 3 days of incubation with Tyr at pH 9.5 at 25 ◦ C were practically identical with that of the initial Hb. The spectra showed very high similarity in the maximum at 208 and 222 nm (Fig. 7A). Furthermore, the proportion of ␣-helix for the Hb after the incubation with Tyr for 3 days at 25 ◦ C or with Glu for 18 h at 37 ◦ C was very similar with the initial Hb, at around 63% (Fig. 7B). Although the proportion of ␣-helix for the Hb after the incubation with Tyr for 7 days at 25 ◦ C decreased a little to 57%, the decrease was not statistically significant (p = 0.052). These results suggest that the Hb after the incubation still maintained its secondary structure. In the Soret region, it was interesting to notice that the CD data (ellipticity) was increased by around 25% at about 420 nm (corresponding to the oxygenated Hb) [35] after 3 days of incubation with Tyr at pH 9.5 at 25 ◦ C, similar with the initial position after 7 days of incubation (Fig. 7C). Differently, there was a considerable increase at about 420 nm after 18 h of incubation with Glu at pH 9 at 37 ◦ C in comparison with that after 12 h of incubation. 3.6. Fluorescence spectrum
Fig. 6. The percentage of metHb in the Hb solutions after incubation with different concentrations of Tyr and Glu. Hb solutions with Tyr and Glu were incubated at pH 9.5 at 25 ◦ C and at pH 9 at 37 ◦ C, respectively. The ratios of Tyr or Glu to Hb at 8, 16 and 24 were used (Asterisk ‘*’ indicates the significant difference relative to the ratio of 8 in the same group, p < 0.01).
The intrinsic fluorescence of proteins can provide considerable information about the structure and dynamics since it changes with conformational transitions and denaturation [36]. As shown in Fig. 8A, the fluorescence intensity of Hb solutions was significantly increased at the wavelength of around 320 nm once Tyr was introduced. The increase in the fluorescence intensity of Hb solutions around 320 nm was correlated with the increase in the concentration of Tyr (Supplementary data). The result implies that
272
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
Fig. 9. Native-PAGE for Hb after incubation with Tyr and Glu. Hb after 3 and 7 days of incubation with Tyr at the ratio of 24 at pH 9.5 at 25 ◦ C and after 12 and 18 h of incubation with Glu at the ratio of 24 at pH 9 at 37 ◦ C was used for electrophoresis, respectively. BSA was used as the molecular marker. Gels were stained in coomassie blue G-250 and destained in acetic acid/ethyl alcohol aqueous solution.
3.7. Electrophoresis
Fig. 7. The CD analysis for the secondary structure of Hb. (A) The CD spectra for the secondary structure of Hb. (B) The percentage of ␣-helix in the secondary structure of Hb. The CD data were analyzed using the Jasco-J810 software (p = 0.052). (C) The CD spectra of Hb in the Soret region. Hb after 3 and 7 days of incubation with Tyr at the ratio of 24 at pH 9.5 at 25 ◦ C and 12 and 18 h of incubation with Glu at the ratio of 24 at pH 9 at 37 ◦ C was scanned. The fresh Hb solution (Hb initial) in the absence of amino acids was used as the control.
the increased fluorescence intensity when Tyr is added is probably due to the fluorescence from Tyr itself. With the incubation time increasing at pH 9.5 at 25 ◦ C, the fluorescence intensity of Hb solutions was just slightly increased after 7d incubation compared with that of Hb solutions at the very beginning of incubation. In addition, there was no significant change in the fluorescence intensity of Hb in the presence of Glu at pH 9 at 37 ◦ C (Fig. 8B).
To test the integrity of Hb after incubation with amino acids, the native-PAGE was performed. Electrophoresis analysis revealed that only a single band corresponding to the molecular weight of 64.5 kDa was observed after 3 and 7 days of incubation with Tyr at pH 9.5 at 25 ◦ C and after 12 and 18 h of incubation with Glu at pH 9 at 37 ◦ C (Fig. 9). This result indicated that Hb did not aggregate or degrade after the incubation in the presence of Tyr or Glu under above conditions. 3.8. Molecular simulations Ratios of 1, 8, 16 and 24 were chosen to test the effect of ratios of amino acid to Hb on the adsorption of amino acid to Hb using the MD simulation. After 5 ns, the system reached equilibrium with little change in root mean square deviation (RSMD). At the low concentration of Tyr, there was no adsorption between Tyr and Hb (Fig. 10A). With the increase in the concentration of Tyr to the ratio of 16, Tyr started to adsorb to the region closed to the heme. At the ratio of 24, two Tyr molecules (the molecules in yellow color
Fig. 8. Fluorescence spectra of Hb at pO2 of 144 Torr in aqueous environment. (A) The fluorescence intensity of Hb after the incubation with Tyr at pH 9.5 at 25 ◦ C. (B) The fluorescence intensity of Hb after the incubation with Glu pH 9 at 37 ◦ C.
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
273
Fig. 10. The interactions of Tyr and Glu with Hb based on the molecular simulation. (A) The snapshot of Tyr or Glu to Hb at 5 ns at different Tyr/Hb ratios at 25 ◦ C simulated using the Gromacs. Yellow: Tyr bound to heme region; gray: Tyr not bound to heme region. (B) The combining conformation of Hb with Tyr, Glu, Trp, Lys and Cys simulated using the AutoDock. (C) The heme pocket size (distance between the backbone of His58 and His87) at 5 ns in the presence of Tyr at 25 ◦ C and Glu at 37 ◦ C simulated using Gromacs. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
in Fig. 10A) were interacted with two heme sites. Similar patterns were observed for Glu (Fig. 10A). The conformation of Tyr, Glu, Trp, Lys and Cys in the heme pocket at their lowest free binding energy was calculated using the AutoDock. The results were shown in Fig. 10B. In the presence of Tyr, hydrogen bonds were formed between carboxy terminal of Tyr and His58, carboxy terminal of Tyr and Lys61, and between amino terminal of Tyr and porphyrin ring. The aromatic side chain of Tyr was interacted with Phe46 and His45. In the presence of Glu and Cys, hydrogen bonds were generated between carboxy terminal of Glu and His58, carboxy terminal of Glu and Lys61, and between
amino terminal of Glu and porphyrin ring, but Glu was not interacted with Phe46 and His45. In contrast, there was no hydrogen bond formed between carboxy terminal of Trp/Lys and His 58, His 45, Lys 61 or Phe46. Fig. 10C shows the heme pocket size according to the MD simulation when the Hb solutions were incubated with Tyr at 25 ◦ C and Glu at 37 ◦ C, respectively. There was little change in the RMSD after 5 ns. It was interesting to notice that the distance between the backbone of His58 and His87 varied with temperatures and amino acids in the system. In the presence of Tyr, the heme pocket size was 14.09 A˚ at 25 ◦ C. When the temperature was elevated to 37 ◦ C,
274
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
˚ However, in the presence of the pocket size increased to 14.77 A. ˚ 0.78 A˚ larger than that in Glu, the pocket size increased to 14.85 A, the presence of Tyr at 25 ◦ C. When the temperature was elevated ˚ to 37 ◦ C, the pocket size decreased to 14.36 A.
4. Discussion The high stability of Hb is essential for the Hb containing systems when used as blood substitutes. This study was designed to investigate the effects of amino acids as the potential stablizers on preventing Hb from the metHb formation under different conditions, including pH, temperature and the ratio of amino acids to Hb. The results provide a possible clue for improving the stability of Hb and minimizing the metHb formation in aqueous environment under aerobic conditions in vitro. Previous study indicates that the metHb formation is strongly depended on the pH [3,28]. The metHb formation in aqueous environment at 25 ◦ C was in a V-shape with the lowest percentage at pH 9 (Fig. 3A), which was consistent with the previous study on bovine heart myoglobin [37]. However, the pH at which there was the lowest percentage of metHb shifted from pH 9 to pH 8 at 37 ◦ C (Fig. 3B). This is probably caused by the change in the energy barrier which works as an electron transfer from the Fe (II) to O2 leading to the displacement of O2 − from Hb [3]. In this study, Cys, Trp and Tyr were selected as the potential stabilizers to inhibit the metHb formation for their capacities of reduction [38]. Lys as a positive charge amino acid may be able to have interactions with the negative charge of Hb at the physiological pH and therefore was chosen as another potential stabilizer. Similarly, Glu, a negative charge amino acid, was also chosen in this study since its potential interactions with hemes through the positive charge in their surrounding area. The experimental results demonstrated that Glu at 37 ◦ C and Tyr at 25 ◦ C can suppress the metHb formation (Figs. 4 and 5). The mechanism behind the inhibiting effect induced by Glu and Tyr is probably caused by the network formed between carboxy terminal of Tyr/Glu and His58, His45, Lys61 or Phe46, which could inhibit the formation of an activated state between distal histidine and O2 , and the electron transfer from Fe (II) to the bound O2 , and further lead to the suppressing of metHb formation. In contrast, Trp and Lys are not able to prohibit the metHb formation which may be due to no hydrogen bonds generated between the amino acid and His58 and Lys61 (Fig. 10B). In the presence of Cys, although the hydrogen bond was generated between Cys and distal histidine, no inhibiting effect of Cys on the metHb formation was observed in aqueous environment under aerobic conditions (Fig. 4). This is consistent with the previous study demonstrating that the introduction of Cys under aerobic conditions leads to the generation of active oxygens during the autoxidation of Cys and significantly promotes the conversion of Hb to metHb [5]. This adverse effect on the conversion of Hb to metHb may blanket the capacity of inhibiting the formation of an activated state between distal histidine and O2 , and the electron transfer from Fe (II) to the bound O2 . This could be the reason why the presence of Cys did not result in the suppression of metHb under aerobic conditions, conversely accelerate the metHb formation (Fig. 4). Hb stability is related to protein matrices which act as a breakwater of the FeO2 center against aqueous solvent [39]. The metHb formation can be efficiently inhibited in aqueous environment under aerobic condition when FeO2 is embedded in protein matrices, which eventually leads to the slowing down of the autoxidation rate of Hb [40]. The results from the molecular simulations indicated that the adsorption of Tyr or Glu to the region closed to the hemes occurred at the mole ratio of 24 (Fig. 10A). According to
the results from the CD spectra in the Soret region, the conformation surrounding the heme region may be changed due to the introduction of Tyr and Glu (Fig. 7C). The interactions between the heme group and Tyr or Glu in aqueous environment may become an extra barrier against the nucleophilic displacement and the following transfer of an electron from Fe (II) to the bound O2 through a hydrogen bond of the globin moiety [3]. In addition, the hydrogen bond generated between amino acid (Tyr or Glu) and distal histidine could inhibit the formation of an activated state between distal histidine and dioxygen and the electron transfer from Fe (II) to the bound O2 [40]. These actions may play critical roles in the stabilization of Hb. Glu and Tyr have been shown to considerably suppress the metHb formation in rats in vivo and in isolated rat blood [25] and under anaerobic conditions [6,22,23]. However, this study was the first to point out that the inhibiting effect of amino acid on metHb formation was dependent on temperature. The presence of Glu and Tyr was capable to inhibit the metHb formation at 37 ◦ C and 25 ◦ C, respectively (Figs. 4 and 5). The simulation results suggested that the presence of Tyr led to an increase in the heme pocket size when the temperature was elevated from 25 ◦ C to 37 ◦ C. Differently, the introduction of Glu resulted in a decrease in the heme pocket size when the temperature was elevated from 25 ◦ C to 37 ◦ C (Fig. 10C). The increase in the size of the heme pocket allows the entering of water molecules to attack FeO2 center on both acidic and basic side [3]. The size of the heme pocket for the initial crystal structure was 14.47 A˚ (Supplementary data), which ˚ but was larger than that in the presence of Tyr at 25 ◦ C (14.09 A) ˚ In contrast, an opposite trend smaller than that at 37 ◦ C (14.77 A). was observed in the presence of Glu. Therefore, the variation of the heme pocket may be an elucidation for the different suppressing effects on the metHb formation provided by Tyr and Glu at the different temperatures. The various capacities of Tyr and Glu on the inhibition of metHb formation at different temperatures imply that the stabilization of Hb may be controlled by the microenvironment around the heme group, including the formation of breakwater network, the status of distal histidine and the size of heme pocket. The CD spectroscopy is used to determine the conformation of polypeptide chains of proteins. When a polypeptide chain of a protein has ␣-helical conformation, there are two negative peaks at 208 nm and 222 nm in the CD spectra [41,42]. The proportion of ␣-helix for the initial Hb and the Hb after incubation with Tyr for 7 days at 25 ◦ C or with Glu for 18 h at 37 ◦ C did not show significant changes (Fig. 7), suggesting that Hb still maintained its secondary structure. Furthermore, the native-PAGE indicated that Hb did not aggregate or degrade after the incubation at 25 ◦ C and 37 ◦ C at pO2 of 144 Torr in aqueous environment (Fig. 9), suggesting that the denaturation of Hb in aqueous environment under aerobic conditions was mainly caused by the metHb formation rather than by the degradation or aggregation. In the Soret region, the changes at about 420 nm after the incubation with Tyr at 25 ◦ C or with Glu at 37 ◦ C indicates that Tyr or Glu may change the conformation surrounding of the hemes through the potential interactions with the surrounding residues of the hemes (Figs. 7C and 10B), and further change the microenvironment around the hemes [43]. The structural changes of proteins can also be reflected by the changes of protein fluorescence intensity. When Tyr was introduced into Hb solutions, the fluorescence intensity was significantly increased compared with the fresh Hb solutions, implying that the presence of Tyr in the Hb solutions could be a reason for the dramatic changes in the fluorescence intensity. The fluorescence intensity with the emission at around 320 nm was not markedly increased with the incubation time when Tyr was present, suggesting there was no significantly change in the structure of Hb (Fig. 8A). The fluorescence spectra of Hb after 18 h of incubation with Glu at pH 9 at
Y. Wei et al. / International Journal of Biological Macromolecules 64 (2014) 267–275
37 ◦ C was also not significantly changed compared with the fresh Hb solutions (Fig. 8B). 5. Conclusions In this study, Tyr and Glu can effectively inhibit the formation of metHb without leading to changes in the structure of Hb in Hb-containing systems in aqueous environment under aerobic conditions at different temperatures. The suppressing effect induced by the presence of Tyr and Glu may be related to the change in the microenvironment around the heme group, including the formation of the breakwater network, the change in the size of heme pocket and the stabilization of distal histidine. Inhibition of the metHb formation achieved by the introduction of Tyr and Glu may be much easier and less expensive than other methods, such as engineered protein. These results may provide a new insight into the design for Hb-oxygen based carriers with high stability of Hb in aqueous environment under aerobic conditions, and even at physiological temperature in vitro. Acknowledgements This research was supported by the Chinese Academy of Sciences, and the National Natural Science Foundation of China (21176238). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. ijbiomac.2013.12.010. References [1] [2] [3] [4]
D. Bossi, B. Giardina, Mol. Aspects Med. 17 (1996) 117–128. C.C. Winterbourn, Environ. Health Persp. 64 (1985) 321–330. K. Shikama, Prog. Biophys. Mol. Biol. 91 (2006) 83–162. L.P. Stratton, A.S. Rudolph, W.K. Knoll Jr., S. Bayne, M.C. Farmer, Hemoglobin 12 (1988) 353–368. [5] S. Takeoka, H. Sakai, T. Kose, Y. Mano, Y. Seino, H. Nishide, E. Tsuchida, Bioconjug. Chem. 8 (1997) 539–544. [6] T. Atoji, M. Aihara, H. Sakai, E. Tsuchida, S. Takeoka, Bioconjug. Chem. 17 (2006) 1241–1245. [7] T. Kanias, J.P. Acker, Cryobiology 58 (2009) 232–239.
275
[8] G.M. Tush, R.J. Kuhn, Ann. Pharmacother. 30 (1996) 1251–1254. [9] A.S. Rudolph, A. Sulpizio, P. Hieble, V. MacDonald, M. Chavez, G. Feuerstein, J. Appl. Physiol. 82 (1997) 1826–1835. [10] H. Sakai, H. Horinouchi, K. Tomiyama, E. Ikeda, S. Takeoka, K. Kobayashi, E. Tsuchida, Am. J. Pathol. 159 (2001) 1079–1088. [11] H. Sakai, P. Cabrales, A.G. Tsai, E. Tsuchida, M. Intaglietta, Am. J. Physiol. Heart C 288 (2005) H2897–H2903. [12] K. Chen, T.J. Merkel, A. Pandya, M.E. Napier, J.C. Luft, W. Daniel, S. Sheiko, J.M. DeSimone, Biomacromolecules 13 (2012) 2748–2759. [13] Y. Xiong, A. Steffen, K. Andreas, S. Muller, N. Sternberg, R. Georgieva, H. Baumler, Biomacromolecules 13 (2012) 3292–3300. [14] R.M. Winslow, Nat. Med. 3 (1997) 474. [15] E. Tsuchida, Artificial Red Cells: Materials, Performances, and Clinical Study as Blood Substitutes, Wiley, New York, 1995, pp. 288. [16] T.M. Chang, Artif. Cells Blood Sub. 25 (1997) 1–24. [17] L. Djordjevich, I.F. Miller, Exp. Hematol. 8 (1980) 584–592. [18] F. D’Agnillo, A.I. Alayash, Blood 98 (2001) 3315–3323. [19] J.G. Riess, Chem. Rev. 101 (2001) 2797–2920. [20] B.A. Kerwin, M.C. Heller, S.H. Levin, T.W. Randolph, J. Pharm. Sci. 87 (1998) 1062–1068. [21] M.C. Heller, J.F. Carpenter, T.W. Randolph, J. Pharm. Sci. 85 (1996) 1358–1362. [22] S. Takeoka, T. Ohgushi, H. Sakai, T. Kose, H. Nishide, E. Tsuchida, Artif. Cells Blood Sub. 25 (1997) 31–41. [23] H. Sakai, K.I. Tomiyama, K. Sou, S. Takeoka, E. Tsuchida, Bioconjug. Chem. 11 (2000) 425–432. [24] R. Di Domenico, R. Lavecchia, Biotechnol. Lett. 22 (2000) 335–339. [25] A.M. Genkin, M.S. Volkov, Bull. Exp. Biol. Med. 49 (1960) 486–487. [26] Y. Teramura, H. Kanazawa, H. Sakai, S. Takeoka, E. Tsuchida, Bioconjug. Chem. 14 (2003) 1171–1176. [27] V. Nadithe, Y.H. Bae, Int. J. Biol. Macromol. 47 (2010) 603–613. [28] W.D. Brown, L.B. Mebine, J. Biol. Chem. 244 (1969) 6696–6701. [29] C.T. Andrade, L.A. Barros, M.C. Lima, E.G. Azero, Int. J. Biol. Macromol. 34 (2004) 233–240. [30] G. Sun, A.F. Palmer, J. Chromatogr. B 867 (2008) 1–7. [31] R.E. Benesch, R. Benesch, S. Yung, Anal. Biochem. 55 (1973) 245–248. [32] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark, H.J. Berendsen, J. Comput. Chem. 26 (2005) 1701–1718. [33] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, J. Comput. Chem. 30 (2009) 2785–2791. [34] R. Linberg, C.D. Conover, K.L. Shum, R.G. Shorr, Cells Blood Sub. 26 (1998) 133–148. [35] M. Nagai, Y. Sugita, Y. Yoneyama, J. Biol. Chem. 244 (1969) 1651–1653. [36] T. Chen, S. Zhu, H. Cao, Y. Shang, M. Wang, G. Jiang, Y. Shi, T. Lu, Spectrochim. Acta A 78 (2011) 1295–1301. [37] Y. Sugawara, A. Matsuoka, A. Kaino, K. Shikama, Biophys. J. 69 (1995) 583–592. [38] S. Takeoka, Y. Teramura, T. Atoji, E. Tsuchida, Bioconjug. Chem. 13 (2002) 1302–1308. [39] M. Tsuruga, A. Matsuoka, A. Hachimori, Y. Sugawara, K. Shikama, J. Biol. Chem. 273 (1998) 8607–8615. [40] K. Shikama, Chem. Rev. 98 (1998) 1357–1374. [41] T. Chatterjee, A. Pal, S. Dey, B.K. Chatterjee, P. Chakrabarti, PLoS ONE 7 (2012) e37468. [42] N. Sreerama, R.W. Woody, Method. Enzymol. 383 (2004) 318–351. [43] Q. Chen, I. Lalezari, R.L. Nagel, R.E. Hirsch, Biophys. J. 88 (2005) 2057–2067.
