Abstract
To produce recombinant hemoglobin in Escherichia coli, sufficient intracellular heme must be present, or the protein folds improperly and is degraded. In this study, coexpression of human hemoglobin genes and Plesiomonas shigelloides heme transport genes enhanced recombinant hemoglobin production in E. coli BL21(DE3) grown in medium containing heme.
Numerous attempts have been made to find an alternative to human blood for patients who require blood transfusions (for recent reviews, see references 1 and 6). One approach is to harvest recombinant hemoglobin expressed in Escherichia coli. When human α- and β-globin chains are expressed in E. coli, hemoglobin can be recovered from soluble fractions of cells (5). However, the chains are degraded unless sufficient heme is present to allow assembly into tetrameric hemoglobin (13-15). The problem has been addressed previously by adding large quantities of heme to the culture medium of E. coli strains that are more permeable to heme than other strains. Although E. coli can synthesize heme, nonpathogenic strains lack high-affinity heme transport systems.
Plesiomonas shigelloides is an intestinal pathogen that uses heme iron (2). Its heme transport system consists of HugA, the outer membrane heme receptor; TonB and ExbBD, which allow HugA to move heme into the periplasm; HugB, which moves heme across the periplasm; and HugC and HugD, which move heme into the cytoplasm (4). Three additional proteins, HugWXZ, may be required for the utilization of heme as an iron source but not for heme transport (4, 8).
A recent study demonstrated that the expression of a heme receptor gene in E. coli increases the heme content and activity of plasmid-encoded catalase (11). In our study, we assessed hemoglobin production in E. coli BL21(DE3) coexpressing human hemoglobin and P. shigelloides heme transport genes.
The P. shigelloides heme utilization system supports the growth of BL21(DE3) in heme-containing medium.
Heme utilization assays were performed to demonstrate that BL21(DE3) coexpressing human hemoglobin genes (on pHB0.0) and P. shigelloides heme utilization genes (on pHUG17.1) can transport heme and use it as an iron source. BL21(DE3)/pHB0.0 with pHUG17.1 or the parent vector (pWSK29) were grown for 20 h in M9 medium (10) containing 7.6 μM heme (Sigma Chemicals) and 100 μg/ml of the iron chelator ethylenediamine-di-(o-hydroxyphenyl acetic acid) (EDDA) (Complete Green Company) (9). BL21(DE3)/pHB0.0 with the heme utilization system reached an absorbance of 1.03 at a wavelength of 600 nm, compared to 0.09 for BL21(DE3)/pHB0.0 with the parent vector. Both strains had absorbances of approximately 1.0 in M9 medium with 20 μM FeSO4 and below 0.10 in M9 medium with EDDA and no heme.
BL21(DE3) containing hemoglobin genes and P. shigelloides heme transport genes produces high levels of hemoglobin.
To assess hemoglobin production, BL21(DE3)/pHB0.0 with or without the heme transport system was grown in L broth with 6.25 μg/ml EDDA to allow the expression of heme transport genes. After 8 h, 40 μg/ml isopropyl-β-d-thiogalactopyranoside (IPTG) was added to induce the expression of hemoglobin genes which are controlled by a tac promoter, and 15.2 μM heme was added to increase intracellular heme. After 16 h of additional growth, cultures were flushed with carbon monoxide (CO) for 20 min, adjusted to an absorbance of 0.50 at 700 nm, and scanned on a Spectronic Genesys spectrophotometer at 350- to 700-nm wavelengths.
Both cultures showed a peak at 419 nm, consistent with CO hemoglobin (7) (Fig. 1A), but the peak for BL21(DE3) coexpressing the hemoglobin genes (on pHB0.0) and heme transport genes (on pHUG21) was markedly higher than that for BL21(DE3) expressing the hemoglobin genes alone. No peak was detected in the scan of E. coli containing the heme transport genes alone [BL21(DE3)/pDSV1, pHUG21], indicating that the heme transport system did not cause the accumulation of heme-containing proteins which could contribute to the spectrum.
FIG. 1.
Hemoglobin production in BL21(DE3) transformed with various plasmids. (A) Absorbance scans of live cultures of BL21(DE3) with and without hemoglobin or P. shigelloides heme transport genes. Solid line, E. coli containing hemoglobin and heme transport genes [(BL21(DE3)/pHB0.0, pHUG21]; dotted line, E. coli containing hemoglobin genes alone [BL21(DE3)/pHB0.0, pWSK29]; dashed line, E. coli containing heme transport genes alone [BL21(DE3)/pDSV1, pHUG21]. Values on the x axis are wavelengths in nanometers. (B) First derivative spectrum obtained from absorbance scans. Derivative values are shown beside each arrow. Solid line, BL21(DE3) containing heme transport and hemoglobin genes; dotted line, BL21(DE3) containing hemoglobin genes alone. (C) Immunoblot of soluble (S) and insoluble (I) fractions from 0.004 OD unit of cells at a wavelength of 700 nm. Lanes 1 and 4, BL21(DE3) containing heme transport genes alone; lanes 2 and 5, BL21(DE3) containing hemoglobin genes alone; lanes 3 and 6, BL21(DE3) containing hemoglobin and heme transport genes; lane 7, 0.45 μg of hemoglobin (Sigma Chemicals).
The first derivative profiles of the absorbance scans were obtained (Fig. 1B). Graves et al. (3) found that first derivative values of live cultures correlate with quantities of purified hemoglobin isolated from cells. When three independent assays were performed, BL21(DE3) with hemoglobin and heme transport genes or with hemoglobin genes alone had average derivative values of 0.134 ± 0.002 and 0.023 ± 0.002, respectively.
To determine if derivative values correlated with levels of soluble hemoglobin, cells were lysed and centrifuged to separate soluble fractions found in supernatants from insoluble fractions found in pellets (13). Immunoblots using goat anti-human hemoglobin antibody (BiosPacific Fortron Bio Science, Inc.) and horseradish peroxidase-labeled rabbit anti-goat antibody (Bio-Rad) were performed on protein samples that had been electrophoresed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and densitometry of autoradiograms was performed using the UVP AutoChemi system. BL21(DE3) with hemoglobin and heme transport genes produced substantially higher levels of soluble hemoglobin than with hemoglobin genes alone (Fig. 1C, lanes 3 and 2, respectively). Moreover, most of the hemoglobin produced by the latter strain was in the insoluble fraction (compare lane 2 of Fig. 1C to lane 5). No signal was generated by either fraction of BL21(DE3) with heme transport genes alone (lanes 1 and 4), indicating that antibody did not cross-react with P. shigelloides heme transport or E. coli proteins.
Analysis of soluble fractions from three experiments indicated that BL21(DE3) with hemoglobin and heme transport genes produced 5.3 times more hemoglobin (standard deviation, ±0.50) than BL21(DE3) with hemoglobin genes alone. These data match the 5.8-fold increase (standard deviation, ±0.58) in average derivative spectrum values from assays of BL21(DE3) with hemoglobin and heme transport genes versus with hemoglobin genes alone.
Hemoglobin production in BL21(DE3) containing hemoglobin and heme transport genes on the same plasmid.
pHB0.0hug contains both hemoglobin and heme transport genes (Fig. 2A; Table 1). Hemoglobin production was assessed in BL21(DE3) transformed with the one-plasmid system (pHB0.0hug, a high-copy-number plasmid) or cotransformed with the two-plasmid system (pHB0.0, a high-copy-number plasmid, and pHUG21, a low-copy-number plasmid). Unlike BL21(DE3) with the two-plasmid system, the strain with the one-plasmid system produced high derivative spectrum values when grown in L broth without EDDA, presumably because of the higher copy number of the heme transport genes on pHB0.0hug. Thus, the strain with the one-plasmid system was grown in L broth and that with the two-plasmid system in L broth with EDDA.
FIG. 2.
Hemoglobin production in BL21(DE3) containing the one-plasmid or two-plasmid system. (A) Genetic and partial restriction enzyme map of genes on pHB0.0hug. Genes are indicated with horizontal arrows, and promoters are indicated with vertical bent arrows. Restriction enzyme sites are as follows: B, BamHI; H, HindIII; N, NheI; S, SalI; and W, BsiWI. (B) Immunoblot of soluble (S) and insoluble (I) fractions. Lanes 1 and 3, BL21(DE3) containing the two-plasmid system (pHB0.0, pHUG21); lanes 2 and 4, BL21(DE3) containing the one-plasmid system (pHB0.0hug); lane 5, 0.45 μg of hemoglobin. (C) Densitometry analysis of immunoblots of soluble fractions. Data are from three independent sets of samples analyzed on the same immunoblot; soluble fractions from 0.004 OD unit of cells at a wavelength of 700 nm were loaded. Open circles, hemoglobin control; filled squares, BL21(DE3) containing the two-plasmid system; filled triangles, BL21(DE3) containing the one-plasmid system.
TABLE 1.
Bacterial strains and plasmids used in this study
| Strain or plasmid | Description | Source or reference |
|---|---|---|
| Strain | ||
| E. coli BL21(DE3) | F−ompT gal dcm lon hsdSB (rB− mB−; E. coli B strain), with DE3, a λ prophage carrying the T7 RNA pol gene | |
| Plasmids | ||
| pHUG17.1 | Apr; the 14-kb HindIII fragment from pHPS6 (4) was cloned into pWSK29; contains hugAWXZ, tonB, exbBD, and hugBCD from P. shigelloides | This study |
| pHUG21 | Apr; pHUG17.1 in which the BsiWI fragment containing hugWXZ was removed; contains hugA, tonB, exbBD, and hugBCD from P. shigelloides | This study |
| pHB0.0 | Tcr; contains human α- and β-chain genes controlled by a tac promoter | Somatogen |
| pDSV1 | Tcr; pHB0.0 in which the HindIII-BamHI fragment containing human α- and β-chain genes was removed | This study |
| pHB0.0hug | Tcr; pHB0.0 containing the HindIII fragment from pHUG21; contains human α- and β-chain genes, as well as hugA, tonB, exbBD, and hugBCD from P. shigelloides | This study |
| pWSK29 | Apr; low-copy-number cloning vector | 12 |
The results of spectrophotometric hemoglobin assays indicated that average derivative spectrum values of both the single- and double-plasmid strains were 0.134, with standard deviations below 0.008. However, when protein fractions were analyzed, BL21(DE3) with the one-plasmid system (Fig. 2B, lane 2) produced more soluble hemoglobin than with the two-plasmid system (lane 1). Spectrophotometric scans were performed on soluble fractions isolated from both strains. Fractions from BL21(DE3) containing the one-plasmid system had an average derivative value 1.36-fold higher (standard deviation, ±0.02) than that from BL21(DE3) with the two-plasmid system, suggesting that BL21(DE3) with the one-plasmid system did indeed produce more soluble holo-hemoglobin.
To estimate the amount of soluble hemoglobin, samples from 0.004 optical density (OD) unit of cells obtained in three independent experiments were analyzed on an immunoblot containing known amounts of hemoglobin. Data points for BL21(DE3) with the two-plasmid system (Fig. 2C) clustered around 0.5 μg, whereas those for BL21(DE3) with the one-plasmid system ranged from 0.6 μg to 0.75 μg. The average amount of soluble hemoglobin produced per 0.004 OD unit was 0.48 μg for BL21(DE3) with the two-plasmid system and 0.65 μg for BL21(DE3) with the one-plasmid system (Table 2). To estimate the amount of hemoglobin produced per ml of culture, the average absorbance of the culture over 0.004 OD unit was multiplied by the average quantity of hemoglobin per 0.004 OD unit. BL21(DE3) with the one-plasmid system produced 244 μg of hemoglobin per ml, whereas BL21(DE3) with the two-plasmid system produced 142 μg per ml (Table 2). BL21(DE3) with the hemoglobin genes alone produced approximately 24 μg of hemoglobin per ml. Thus, BL21(DE3) with the one-plasmid system produced 10 times more soluble hemoglobin than BL21(DE3) with the hemoglobin genes alone. Additional work is needed to determine whether this approach of coexpressing hemoglobin and heme transport genes in E. coli is an economically feasible method of producing recombinant hemoglobin.
TABLE 2.
Analysis of hemoglobin production in BL21(DE3) containing various plasmids
| Addition to BL21(DE3) | Plasmid(s) | Absorbance of culture at 24 h at a wavelength of 700 nm | Amt (μg) of soluble hemoglobin per 0.004 OD unit | Amt (μg) of soluble hemoglobin per ml of culture |
|---|---|---|---|---|
| Two-plasmid system | pHB0.0, pHUG21 | 1.18 ± 0.06 | 0.48 ± 0.02 | 142 ± 13.0 |
| One-plasmid system | pHB0.0hug | 1.50 ± 0.08 | 0.65 ± 0.05 | 244 ± 7.4 |
| Hemoglobin genes alone | pHB0.0, pWSK29 | 1.06 ± 0.01 | 0.09 ± 0.01 | 24 ± 2.5 |
Acknowledgments
This work was supported by grant development funds from the University of Texas of the Permian Basin, National Institutes of Health grant 1 R15 HL079992-01 (D. P. Henderson), NIH grants GM 35649 and HL 47020, and Welch Foundation grant C-612 (J. S. Olson).
We thank Shelley Payne and Elizabeth Wyckoff for their advice concerning the project.
Footnotes
Published ahead of print on 1 August 2008.
REFERENCES
- 1.Chang, T. M. S. 2006. Blood substitutes based on nanobiotechnology. Trends Biochem. 24:372-377. [DOI] [PubMed] [Google Scholar]
- 2.Daskaleros, P. A., J. A. Stoebner, and S. M. Payne. 1991. Iron uptake in Plesiomonas shigelloides: cloning of the genes for heme-iron uptake system. Infect. Immun. 59:2706-2711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Graves, P. E, D. P. Henderson, M. J. Horstman, B. J. Solomon, and J. S. Olson. 2008. Enhancing stability and expression of recombinant human hemoglobin in E. coli: progress in the development of a recombinant HBOC source. Biochim. Biophys. Acta. doi: 10.1016/j.bbapap.2008.04.012. [DOI] [PubMed]
- 4.Henderson, D. P., E. E. Wyckoff, C. E. Rashidi, H. Verlei, and A. L. Oldham. 2001. Characterization of the Plesiomonas shigelloides genes encoding the heme iron utilization system. J. Bacteriol. 183:2715-2723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hoffman, S. J., D. L. Looker, J. M. Roehrich, P. E. Cozart, S. L. Durfee, J. L. Tedesco, and G. L. Stetler. 1990. Expression of fully functional tetrameric human hemoglobin in Escherichia coli. Proc. Natl. Acad. Sci. USA 87:8521-8525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Inayat, M. S., A. C. Bernard, V. S. Gallicchio, B. A. Garvy, H. L. Elford, and O. R. Oakley. 2006. Oxygen carriers: a selected review. Transfus. Apher. Sci. 34:25-32. [DOI] [PubMed] [Google Scholar]
- 7.Looker, D., A. J. Mathews, J. O. Neway, and G. L. Stetler. 1994. Expression of recombinant human hemoglobin in Escherichia coli. Methods Enzymol. 231:364-374. [DOI] [PubMed] [Google Scholar]
- 8.Oldham, A. E., T. A. Wood, and D. P. Henderson. 2008. Plesiomonas shigelloides hugZ encodes an iron-regulated heme binding protein required for heme iron utilization. Can. J. Microbiol. 54:97-102. [DOI] [PubMed] [Google Scholar]
- 9.Rogers, H. J. 1973. Iron-binding catechols and virulence in Escherichia coli. Infect. Immun. 7:438-444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sambrook, J., and D. W. Russell. 2001. Molecular cloning. a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
- 11.Varnado, C. L., and D. C. Goodwin. 2004. System for the expression of recombinant hemoproteins in Escherichia coli. Protein Express. Purif. 35:76-83. [DOI] [PubMed] [Google Scholar]
- 12.Wang, R. F., and S. R. Kushner. 1991. Construction of versatile low-copy-number vectors for cloning, sequencing, and gene expression in Escherichia coli. Gene 100:195-199. [PubMed] [Google Scholar]
- 13.Weickert, M. J., and S. R. Curry. 1997. Turnover of recombinant human hemoglobin in Escherichia coli occurs rapidly for insoluble and slowly for soluble globin. Arch. Biochem. Biophys. 348:337-346. [DOI] [PubMed] [Google Scholar]
- 14.Weickert, M. J., and I. Apostol. 1998. High-fidelity translation of recombinant human hemoglobin in Escherichia coli. Appl. Environ. Microbiol. 64:1589-1593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Weickert, M. J., M. Pagratis, C. B. Glascock, and R. Blackmore. 1999. A mutation that improves soluble recombinant hemoglobin accumulation in Escherichia coli in heme excess. Appl. Environ. Microbiol. 65:640-647. [DOI] [PMC free article] [PubMed] [Google Scholar]