Kumar et al. 10.1073/pnas.0705054104. |
Fig. 8. Absorption spectra of purified DosS and DosT. Electronic absorption spectra of purified DosS (pink) and DosT (blue) in 20 mM Tris buffer (pH 7.5) possessed a, b (shown in Inset with 500- to 600-nm region enlarged) and Soret bands, indicating the presence of heme group bound to the protein. Numbers indicate wavelength in nm.
Fig. 9. Pyridine hemochromogen assay of DosS and DosT. Pyridine hemochromogen assay was performed with DosS and DosT as described (1). DosS and DosT spectra are characteristic of heme type b.
1. Berry EA, Trumpower BL (1987) Anal Biochem 161:1-15.
Fig. 10. Removal of DTH from samples before air exposure. DosS (3 mM in 20 mM Tris buffer, pH 7.5) was treated with 30 mM DTH followed by passing samples twice through size exclusion columns [D-Salt Polyacrylamide Desalting Columns (6K; Pierce)] to remove DTH. Note the absence of the DTH peak in the size-exclusion sample. Subsequently, samples were exposed to air for 30 s.
Fig. 11. Reaction of NO and CO with oxy DosT. As purified DosT (3 mM in Tris, pH 7.5) was exposed to NO or CO. CO was able to "replace" bound oxygen, whereas NO did not react with oxy heme of DosT.
SI Text
Materials and Methods
Cloning, Expression and Purification of DosS and DosT
. The complete ORFs of dosS and dosT were PCR-amplified from Mtb H37Rv genomic DNA and cloned into pET15b (Novagen Darmstadt, Germany). To generate His-SUMO-tagged gene products, dosS and dosT PCR products were cloned in-frame with the 11-kDa small ubiquitin related modifier (SUMO) tag in pET15b. SUMO is a small eukaryotic polypeptide that has been shown to increases solubility of bacterial proteins (1). His-tagged or His-SUMO-tagged fusion proteins were overproduced in E. coli Rosetta (DE3) cells in Luria Broth (Difco, Franklin Lakes, NJ) supplemented with 20 mM hemin. Purified DosS and DosT were obtained in both the soluble and insoluble fractions. Soluble proteins were extracted by using Profinity IMAC Ni-charged resin (Bio-Rad, Hercules, CA) as recommended by the manufacturer. His-tag and His-SUMO-tagged proteins gave identical absorption, EPR spectroscopy, and autokinase results. Fresh DosS and DosT were prepared every 2 weeks.
DosS and DosT Sample Preparation of Spectroscopic Analysis.
To determine the oxidation status of the heme iron, air-exposed protein was divided into two aliquots and treated with 300 mM KCN. The met form of heme reacts with KCN to produce a met-CN complex characterized by single peak at 540 nm. When necessary, samples were treated with 300 mM Fe(CN)6-3. Heme content and identity (type A, B, C) were determined as described (2).
To test whether CO and NO can act as ligands for DosS or DosT, samples were deoxygenated for 15 min with a constant flow of argon gas and transferred to the anaerobic glove box. Samples were then treated with DTH, and the DTH was subsequently removed by size-exclusion chromatography. For NO exposure, protein samples were treated with a 200-fold molar excess proline NONOate (Cayman, Ann Arbor, MI). For CO exposure, a constant stream of CO (100%) was flushed through the protein samples.
Expression Analysis of the Dormancy Regulon in Response to CO.
Mtbcells were grown to OD600 of 0.8 in 7H9 medium and treated for 3 h with 1% CO or 50 mM CO obtained either from CO dissolved in PBS buffer or the carbon monoxide-releasing molecule tricarbonylchlororuthenium(II) dimer (CORM-2). The release of CO from CORM-2 and the amount of CO dissolved in PBS was assessed spectrophotometrically by measuring the conversion of deoxymyoglobin (deoxy-Mb) to carbonmonoxy myoglobin (MbCO) as described (3). Mtb cells were harvested and RNA was isolated as described (4). First-strand synthesis was performed by using the iScript Select cDNA Synthesis Kit (Bio-Rad) with the primers 16SR: ATGTCGCAAGGTTAACCCGCGTGCC; dosRR: AGCAAC CGCGACACGTAGTTC; hspXR: GTTGGCTTCCCTTCCGAAACC; and fdxAR: CCG GTTTGCACGCACCACAATC. Expression of dosR, hspX, or fdxA were analyzed with real-time PCR using iQ SYBR Green Supermix (Bio-Rad) and a BioRad iCycler 5 with an iQ Multicolor Real-Time PCR Detection System (Bio-Rad). The primer sets used were 16SF: GAAGAA TGAGCCTGCGAGTC and 16SR: GGTCCAGAACACGCCACTAT; dosRF: CGGTCG CTGGTGGACAATC and dosRR: TTTCGGCTAGGAACATTCG; hspXF: CGCACCGAGCAGAAGGAC and hspXR: CCGCCACCGACACAGTAA; fdxAF: CCTATGTGATCGGTAGTGA and fdxAR: GGGTTGATGTAGAGCATT. Data analysis was performed with the iQ Multicolor Real-Time PCR Detection System Optical Software System (Bio-Rad), version iQ5. PCR efficiencies were normalized to obtain accurate expression levels.
Notes
Note 1
. Heme Fe protoporphyrin IX gives rise to two p®p* electronic transitions at ~400 nm (soret or B band) and 500-600 nm (Qv, Qo, or b and a bands) whose wavelengths are indicative of the oxidation, spin, and coordination states of the heme iron. Usually a soret band between 350 and 400 nm is indicative of penta-coordinated heme iron, whereas a sharp band at 400-412 nm is indicative of hexa-coordinated high-spin heme (5). However, bands between 412 and 450 nm are usually indicative of a hexa-coordinated low-spin heme. A weak band (e.g., DosS) observed in the 600- to 650-nm region (5) is characteristic of a high-spin heme protein.
Note 2
. Most heme sensor proteins contain PAS domains (Per/Arnt/Sim) that are frequently reported to function as input modules for sensing oxygen, redox potential, and light (6). However, DosS (Rv3132c) and DosT (Rv2027c) lack a PAS domain, rather motif annotation have established that both proteins are GAF-domain (cGMP-regulated phosphodiesterases, adenylyl cyclases, and FhlA) proteins. GAF-domain proteins are frequently involved in cyclic nucleotide binding and sensing of small molecules (7).
Note 3
. DosS and DosT were deoxygenated by passing argon gas through the purified protein samples for 15 min and then transferred to an anaerobic glove box. To reduce DosS, or to deoxygenate oxy-DosT, samples were treated with a 10-fold molar excess of DTH, which was then removed by passing samples twice through a size-exclusion column inside the anaerobic glove box. DTH removal was confirmed by the loss of a characteristic DTH peak at 317 nm (SI Fig. 11). In addition, the redox state of the heme iron was monitored by UV-visible spectroscopy and showed changes consistent with removal of DTH (e.g., reduced-state spectrum revert to the oxidized-state spectrum, or deoxy state reverts to the oxy-state spectrum). When relevant, the absorption spectra of the proteins were recorded before and after exposure of air (bubbling for 30-120 s). Protein samples were then characterized by EPR spectroscopy and treatment with KCN or Fe(CN)63-.
Note 4.
Because the purpose of this experiment was to assess the redox state of the heme iron of DosS or DosT, we only showed the g = 6 region. Unfortunately, the g = 2 region cannot be examined because of the large quantity of Fe(CN)6-3 present in the DosT sample. Fe(CN)63- is a low spin iron species, making it difficult to distinguish between the spin state of heme iron and the ferricyanide iron.
Note 5.
We did not observe any change in absorption spectra of met DosS treated with a 200-fold molar excess of NO, suggesting little or no reaction. At molar excesses, NO can reduce and then nitrosylate heme in a process referred to as reductive nitrosylation (8). Although the precise mechanism of this reaction remains under investigation, a characteristic product is the ferrous-NO complex. Failure to detect this species by EPR further suggests that under these experimental conditions, significant ferric-NO reactions were not occurring.
Note 6.
NO can oxidize oxy heme to generate met heme and nitrate. However, our data suggest that 600 mM NO was not able to oxidize 3 mM oxy DosT.
Note 7.
As an additional verification, met DosS was deoxygenated, transferred to the glove box, and reduced with DTH, which was subsequently removed by size-exclusion chromatography. The sample was divided into two aliquots; one aliquot was analyzed for autokinase activity inside the glove box, whereas the other sample was exposed to atmospheric air for 2 min before performing the autokinase activity. The results were essentially identical to the experiment described in Results and thus further validate that ferrous DosS showed increased autokinase activity compared with met DosS.
Note 8.
Similarly, as in Note 7, DosT was treated with DTH inside the glove box, followed by size-exclusion chromatography to remove the DTH. The sample was divided into two aliquots, and one aliquot was removed from the glove box and exposed to atmospheric air. The two samples were analyzed for autokinase acitivity inside the glove box (deoxy) and on the bench (oxy), respectively. The results were essentially identical to the experiment described in Results demonstrating that bound O2 inhibits DosT autokinase activity.
1. Esposito D, Chatterjee DK (2006) Curr Opin Biotechnol 17:353-358.
2. Berry EA, Trumpower BL (1987) Anal Biochem 161:1-15.
3. Motterlini R, Clark JE, Foresti R, Sarathchandra P, Mann BE, Green CJ (2002) Circ Res 90:E17-E24.
4. Butcher PD, Mangan JA, Monahan IM (1998) Methods Mol Biol 101:285-306.
5. Smulevich G, Neri F, Marzocchi MP, Welinder KG (1996) Biochemistry 35:10576-10585.
6. Gu YZ, Hogenesch JB, Bradfield CA (2000) Annu Rev Pharmacol Toxicol 40:519-561.
7. Martinez SE, Beavo JA, Hol WG (2002) Mol Interv 2:317-323.
8. Ford PC, Fernandez BO, Lim MD (2005) Chem Rev 105:2439-2455.