Improved production of human hemoglobin in yeast by engineering hemoglobin degradation

With the increasing demand for blood transfusions, the production of human hemoglobin (Hb) from sustainable sources is increasingly studied. Microbial production is an attractive option, as it may provide a cheap, safe, and reliable source of this protein. To increase the production of human hemoglobin by the yeast Saccharomyces cerevisiae, the degradation of Hb was reduced through several approaches. The deletion of the genes HMX1 (encoding heme oxygenase), VPS10 (encoding receptor for vacuolar proteases), PEP4 (encoding vacuolar proteinase A), ROX1 (encoding heme-dependent repressor of hypoxic genes) and the overexpression of the HEM3 (encoding porphobilinogen deaminase) and the AHSP (encoding human alpha-hemoglobin-stabilizing protein) genes — these changes reduced heme and Hb degradation and improved heme and Hb production. The reduced hemoglobin degradation was validated by a bilirubin biosensor. During glucose fermentation, the engineered strains produced 18% of intracellular Hb relative to the total yeast protein, which is the highest production of human hemoglobin reported in yeast. This increased hemoglobin production was accompanied with an increased oxygen consumption rate and an increased glycerol yield, which (we speculate) is the yeast's response to rebalance its NADH levels under conditions of oxygen limitation and increased protein-production.     

Amino acid sequence of yeast hemoglobin. A two-domain structure.

The complete amino acid sequence of a hemoglobin from yeast (Candida norvegensis) has been determined by peptide and cDNA sequence analyses. The protein is composed of 387 amino acid residues and its amino terminus was blocked by an acetyl group. A computer search showed that the sequence of 155 N-terminal residues has 39% homology with that of Vitreoscilla hemoglobin. On the other hand, the sequence of 230 C-terminal residues showed a small, but notable, degree of similarity with that of a methemoglobin reductase found in human erythrocyte, i.e. NADH-cytochrome b5 oxido-reductase. We therefore conclude that yeast hemoglobin consists of two distinct domains; one is a heme-containing oxygen binding domain of the N-terminal region and the other is an FAD-containing reductase domain found in the C-terminal region.     

Oxygen binding and other physical properties of human hemoglobin made in yeast.

Wagenbach et al. (1991, BioTechnology, 9, 57-61) have recently developed a system for producing soluble recombinant tetrameric hemoglobin in yeast: hemoglobin begins to appear 4-5 h after induction with galactose, alpha- and beta-globin chains fold in vivo and endogeneously produced heme is incorporated into hemoglobin tetramers. We have further characterized the oxygen-binding properties, as well as the tetramer stability, of recombinant human Hb A made in yeast. After purification by ion-exchange chromatography, a single band at the same position as normal human Hb A was obtained using cellulose acetate electrophoresis. Although the oxy and deoxy forms of purified recombinant Hb A made in yeast were spectrophotometrically identical to native human Hb A, the oxygen-binding curve was shifted slightly left of that for native human Hb A. Further purification of recombinant hemoglobin by FPLC revealed two fractions: one (fraction B) with low cooperativity and high oxygen affinity, and the other (fraction A) with almost identical cooperativity and oxygen affinity compared with native human Hb A. The Bohr effect of fraction A was also identical to native human Hb A. Hemoglobin in fraction B with lowered cooperativity precipitated approximately 1.5 times faster than normal human Hb A during mechanical agitation, while hemoglobin in fraction A with normal cooperativity precipitated with kinetics identical to native human Hb A. These results suggest that some of the recombinant molecules made in yeast fold improperly, and that these molecules may exhibit decreased cooperativity for oxygen binding and decreased stability.(ABSTRACT TRUNCATED AT 250 WORDS)     

Sensing the host environment: recognition of hemoglobin by the pathogenic yeast Candida albicans

Adhesion to host cells and tissues is important for several steps in the pathogenesis of disseminated Candida albicans infections. Although such adhesion is evident in vivo and for C. albicans grown in vitro in complex medium, some adhesive activities are absent when cultures are grown in defined media. However, addition of hemoglobin to defined media restores binding and adhesion to several host proteins. This activity of hemoglobin is independent of iron acquisition and is mediated by a cell surface hemoglobin receptor. In addition to regulating expression of adhesion receptors, hemoglobin rapidly induces expression of several genes. One of these, a heme oxygenase, allows the pathogen to utilize exogenous heme or hemoglobin to acquire iron and to produce the cytoprotective molecules alpha-biliverdin and carbon monoxide. The specific recognition of and responses to hemoglobin demonstrate a unique adaptation of C. albicans to be both a commensal and an opportunistic pathogen in humans.     

Oxidation-reduction reactions of hemoglobin A, hemoglobin M Iwate, and hemoglobin M Hyde Park

The kinetics and equilibrium of the redox reactions of hemoglobin A, hemoglobin M Iwate, and hemoglobin M Hyde Park using the iron (II) and iron (III) complexes of trans-1,2-diaminocyclohexane-N,N,N',N'-tetraacetate (CDTA4-) as the reducing and oxidizing agents have been studied. With respect to the equilibrium it was found that hemoglobin M Iwate (where the beta chains were reduced) was more readily reduced than hemoglobin M Hyde Park (where the alpha chains are reduced). This difference was shown to be a result of a difference in the rate constant for reduction but not oxidation. The observed rate contants for the reduction of all three hemoglobins were shown to decrease with increasing pH. This was attributed to a decrease in the [T]/[R] ratio. The observed rate contants for the oxidation reaction were shown to increase with increasing pH. Accompanying this increase was a change in the kinetic profile for hemoglobin A from pseudo first order to one in which the rate increased as the extent of reaction increased. Inositol hexaphosphate had no effect on the rate of oxidation of deoxyhemoglobin A. This was a result of binding of FeCDTA2- or HCDTA3- to the protein. However, in the presence of inositol hexaphosphate, the reduction of methemoglobin A exhibited biphasic kinetics. This result was interpreted in terms of the production of a small amount of a conformation which was more readily reduced.     

The inhibition of hemoglobin C crystallization by hemoglobin F

We have reported that circulating CC erythrocytes containing HbO2 C crystals exhibit little or no Hb F suggesting that Hb F may inhibit the crystallization of Hb C. We report now that Hb F inhibits in vitro crystallization of HbO2 and HbCO C when compared to the effect of Hb A in a wide range of mixture proportions. For example, while HbCO C solutions form tetragonal C crystals within 25 min, no crystals form within 2 h with 30% Hb F, whereas 550 crystals/mm3 form with 30% Hb A. Furthermore, an increase in the percent of Hb A is correlated with a greater number of orthorhombic crystal formation rather than the tetragonal morphology observed with 100% Hb C. We also report that Hb A2 (containing delta chains that exhibit 10 sequence differences with beta chains) and Hb Lepore Boston-Washington (a fusion mutant of delta and beta chains that contains only six of these differences) both inhibit Hb C crystallization. By comparing the sequences of the three inhibitory hemoglobins, we conclude that position Gln-87 in the gamma chains is, at least partially, the cause of the inhibitory effect of Hb F on the crystallization of Hb C.     

Alternative diaspirins for modification of hemoglobin and sickle hemoglobin

Studies of modification of hemoglobin and of sickle hemoglobin by alternative aspirins have been extended to a series of new bis esters with a variety of substituted bridging diacids and to a group of mono esters with polar acyl groups. Rates of hydrolysis of these alternative aspirins have also been examined, and they reveal that a careful balance between stability and reactivity is essential for optimal activity. Four-carbon bridging groups have been found to be particularly effective, two of these raising the minimum gelling concentration of sickle hemoglobin by as much as 100%.     

Mouse hemoglobin during gestation. Absence of fetal hemoglobin

A "fetal hemoglobin' has been reported to exist during mouse gestation, Investigations using CMC chromatography, starch gel electrophoresis or isoelectric focusing have shown a hemoglobin band from fetal tissues, and blood was obtained which was different from the adult hemoglobin and designated a "fetal hemoglobin'. In the current study isoelectric focusing was used to study the hemoglobins existing in the tissues and blood during fetal and neonatal development and the results suggest there is no "fetal hemoglobin' present during gestation. It appears that the hemoglobin designated as "fetal' in our laboratory was a methemoglobin formed by an incomplete reaction of KCN with the hemoglobin. The additional hemoglobin bands which were obtained from fetal liver or neonatal spleen tissues appeared to be a modified adult hemoglobin.     

Specifically carboxymethylated hemoglobin as an analogue of carbamino hemoglobin. Solution and X-ray studies of carboxymethylated hemoglobin and X-ray studies of carbamino hemoglobin

Hemoglobin can be specifically carboxymethylated at its NH2-terminal amino groups (i.e. HbNHCH2COO-) to form the derivatives alpha 2Cm beta 2, alpha 2 beta 2Cm, and alpha 2Cm beta 2Cm, where Cm represents carboxymethyl. Previous studies (DiDonato, A., Fantl, W. J., Acharya, A. S., and Manning, J. M. (1983) J. Biol. Chem. 258, 11890-11895) suggested that these derivatives could be used as stable analogues of the corresponding carbamino (Hb-NHCOO-) forms of hemoglobin, adducts that are generated reversibly in vivo when CO2 combines with alpha-amino groups. In this paper we present x-ray diffraction studies of both carbamino hemoglobin and carboxymethylated hemoglobin that verify this proposal and we use the carboxymethylated derivatives to study the functional consequences of placing a covalently bound carboxyl group at the NH2 terminus of each hemoglobin subunit. Our studies also provide additional information concerning the oxygen-linked binding of anions and protons to Val-1 alpha. Difference electron density analysis of deoxy alpha 2Cm beta 2Cm versus the unmodified deoxyhemoglobin tetramer (deoxy alpha 2 beta 2) shows that the covalently bound carboxyl moieties replace inorganic anions that are normally bound to the free NH2-terminal amino groups in crystals of native deoxyhemoglobin grown from solutions of concentrated (2.3 M) ammonium sulfate. In the case of the beta-subunits, the carboxymethyl group replaces an inorganic anion normally bound between the alpha-amino group of Val-1 beta, the epsilon-amino group of Lys-82 beta, and backbone NH groups at the NH2-terminal end of the F'-helix. In the case of the alpha-subunits, the carboxymethyl group replaces an anion that is normally bound between the alpha-amino group of Val-1 alpha and the beta-OH group of Ser-131 alpha. A corresponding difference electron map of carbamino deoxyhemoglobin in low-salt (50 mM KCl) crystals shows that CO2 bound in the form of carbamate occupies the same two anion binding sites. The alkaline Bohr effect of alpha 2Cm beta 2 is only marginally lower (approximately 7%) than that of alpha 2 beta 2. Previous studies (Kilmartin, J. V., 1977) have shown that about 30% of the alkaline Bohr effect is the result of an oxygen-linked change in the pK alpha of Val-1 alpha, and O'Donnell et al., 1979, found that this portion of the Bohr effect is the result of the oxygen-linked binding of chloride to Val-1 alpha.(ABSTRACT TRUNCATED AT 400 WORDS)