See discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/235881916

Interaction between DAHP synthase and chorismate mutase endows new regulation on DAHP synthase activity in

ARTICLE in APPLIED MICROBIOLOGY AND BIOTECHNOLOGY · MARCH 2013 Impact Factor: 3.34 · DOI: 10.1007/s00253-013-4806-0 · Source: PubMed

CITATIONS READS 3 36

6 AUTHORS, INCLUDING:

Di Liu Shuang-Jiang Liu Chinese Academy of Sciences Chinese Academy of Sciences

69 PUBLICATIONS 2,177 CITATIONS 144 PUBLICATIONS 2,027 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Di Liu Retrieved on: 23 September 2015 Appl Microbiol Biotechnol DOI 10.1007/s00253-013-4806-0

BIOTECHNOLOGICALLY RELEVANT AND PROTEINS

Interaction between DAHP synthase and chorismate mutase endows new regulation on DAHP synthase activity in Corynebacterium glutamicum

Pan-Pan Li & De-Feng Li & Di Liu & Yi-Ming Liu & Chang Liu & Shuang-Jiang Liu

Received: 3 January 2013 /Revised: 19 February 2013 /Accepted: 20 February 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract Previous research on Corynebacterium gluta mutase of Mycobacterium tuberculosis predicted the inter- micum revealed that 3-deoxy-D-arabino-heptulosonate 7- action surfaces of the putative DSCg-CMCg complex. The phosphate synthase (DSCg, formerly DS2098) interacts with residues and structural domains that contributed chorismate mutase (CMCg, formerly CM0819). In this study, to the interaction surfaces were experimentally identified to 212 219 we investigated the interaction by means of structure-guided be the SPAGARYE sequence of DSCg and the 60 64 97 101 mutation and enzymatic assays. Our results show that the SGGTR loop and C-terminus ( RGKLG )ofCMCg. interaction imparted a new mechanism for regulation of

DAHP activity: In the absence of CMCg,DSCg activity Keywords 3-deoxy-D-arabino-heptulosonate 7-phosphate was not regulated by prephenate, whereas in the presence (DAHP) synthase . Chorismate mutase (CM) . Regulation . of CMCg, prephenate markedly inhibited DSCg activity. Shikimate pathway . Corynebacterium glutamicum Prephenate competed with the phosphoenolpyr- uvate, and the inhibition constant (Ki) was determined to be 0.945 mM. Modeling based on the structure of the Introduction complex formed between DAHP synthase and chorismate 3-Deoxy-D-arabino-heptulosonate 7-phosphate (DAHP) : : : synthases (DSs; EC 2.5.1.54) catalyze the condensation of P.-P. Li Y.-M. Liu C. Liu S.-J. Liu phosphoenolpyruvate (PEP) and erythrose 4-phosphate to State Key Laboratory of Microbial Resources, Institute of Microbiology, Chinese Academy of Sciences, form DAHP in bacteria, fungi, and plants (Herrmann 1995; Beijing 100101, People’s Republic of China Pittard 1996; Romero et al. 1995). Because DSs catalyze the first step of the shikimate pathway (Herrmann 1995), they D.-F. Li are tightly regulated by a variety of mechanisms. Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, People’s Republic of China Escherichia coli have three DSs that are sensitive to feed- back regulation by (Phe), (Trp), or D. Liu (Tyr), respectively (Kikuchi et al. 1997; Romero et Network Information Center, Institute of Microbiology, al. 1995). Corynebacterium glutamicum has two DSs, Chinese Academy of Sciences, Beijing 100101, People’s Republic of China NCgl0950 DS and NCgl2098; the former is sensitive to feedback inhibition by Tyr and, to a much lesser extent, by : * Y.-M. Liu S.-J. Liu ( ) Phe and Trp, whereas the latter (hereafter designated DSCg), Environmental Microbiology and Biotechnology Research Center, is slightly sensitive to feedback inhibition by Trp but is Institute of Microbiology, Chinese Academy of Sciences, Beichen West Road No. 1, Chaoyang District, insensitive to Tyr and Phe (Liao et al. 2001; Liu et al. Beijing 100101, People’s Republic of China 2008). Bacillus subtilis possesses only one DS, which is e-mail: [email protected] insensitive to Phe, Tyr, and Trp, but is regulated by the Appl Microbiol Biotechnol shikimate pathway intermediates prephenate and chorismate were obtained from the (PDB: 2B7O; (Wu et al. 2005). Recently, the DS from Mycobacterium PDB: 2W1A). The protein structure and prephenate docking tuberculosis (DSMt) has been shown to be subject to unique were analyzed by means of the PyMOL program (Delano : DSMt is not sensitive to any single 2002) and Ligand Explorer (http://www.pdb.org/pdb/home/ aromatic amino acid, but a combination of Trp and Phe home.do). significantly inhibits its activity (Webby et al. 2010). More recently, dynamic cross talk between remote binding sites of DNA extraction and manipulation catalytic products has been identified as the molecular basis for the allosteric regulation (Jiao et al. 2012). The total genomic DNA of C. glutamicum was isolated

DSMt plays a key role in the activation and feedback according to the procedure of Tauch et al. (Tauch et al. regulation of the branch-point chorismate mutase 1995). DNA restriction enzyme digestion, plasmid isolation, from M. tuberculosis (CMMt) (Schneider et al. 2008). and agarose gel electrophoresis were carried out as described Interestingly, DSMt forms a complex with CMMt, and complex by Samsbrook et al. (1989). E. coli strains were transformed formation leads to substantial enhancement and synergistic by electroporation according to the method of Tauch et al. regulation of CM activity by Phe and Tyr in M. tuberculosis (2002). (Sasso et al. 2009). In contrast, we previously found that in

C. glutamicum,DSCg activity is significantly enhanced by PCR amplification of DNA fragments and construction chorismate mutase (CMCg), whereas no enhancement of of plasmids CMCg activity by DSCg is observed (Li et al. 2009), even though DSCg and CMCg show 77 and 57 % sequence identities PCR reactions were performed with Pfu DNA polymerase to DSMt and CMMt,respectively. (TransGen, Beijing, China). PCR products were purified with In this study, we investigated the interaction between DSCg an agarose gel DNA fragment recovery kit (Tiangen, Beijing, and CMCg. Structure modeling predicted that DSCg and CMCg China). Plasmids pET21-DSCg, pET28-CMCgM1, and should form a complex similar to the DSMt-CMMt complex pET28-CMCgM2 (Table 1) for expression of wild type and (Sasso et al. 2009). In the presence of CMCg, we found that the mutant DSCg and CMCg in E. coli cells were constructed. DSCg activity was regulated by prephenate, whereas in the Primers used for amplification of the entire target genes or absence of CMCg, no such regulation was observed. The DNA fragments are listed in Table 1. amino acid residues that contributed to the interaction inter- faces of the putative DSCg-CMCg complex were identified by QuikChange site-directed mutagenesis means of site-directed mutagenesis and deletion. The plasmids for expression of site-directed mutant proteins

DSCgM1, DSCgM2, CMCgM3, and CMCgM4 were Materials and methods constructed as described by Wang et al. (2008). PCR re- actions were performed with Pfu DNA polymerase. Bacterial strains, plasmids, and cultivation Reaction mixtures without primers were run in parallel and used as controls. DpnI was added to PCR products to digest The bacterial strains and plasmids used in this study are listed the template plasmids; after 3 h at 37 °C, the mixture was in Table 1. E. coli strains were grown aerobically on a rotary transformed into E. coli XL1-Blue by electroporation for shaker (150 rpm) at 37 °C in Luria-Bertani (LB) broth or on replication of target plasmids. Primers used for site-directed LB agar [1.5 % (wt/vol)] plates. C. glutamicum strains were mutagenesis are listed in Table 1. routinely grown at 30 °C in LB broth on a rotary shaker

(150 rpm). Kanamycin and ampicillin were used at concen- Heterologous expression of wild type and mutant DSCg −1 trations of 50 and 100 μg×ml ,respectively. and CMCg in E. coli cells, preparation of cellular lysates, and protein purification Sequence and structure analyses

Plasmids pET28-CMCg, pET28-CMCgM1, pET28-CMCgM2, The sequences of the M. tuberculosis and C. glutamicum pET28-CMCgM3, pET28-CMCgM4, pET21-DSCg, pET21- DSs and CMs were retrieved from GenBank (accession nos. DSCgM1, and pET21-DSCgM2 were transformed into E. coli 000962 and 003450). Protein sequence alignments and anal- BL21 (DE3) by electroporation for heterologous expression yses were carried out using the BLAST tool on the NCBI of the wild-type and mutated DS and CM genes. Recombinant website (http://www.ncbi.nlm.nih.gov)andtheClustalW protein synthesis in E. coli BL21 (DE3) cells was initiated by multiple sequence alignment program (Thompson et al. the addition of 0.2 mM isopropyl β-D-1-thiogalactopyrano-

1997). The structures of DSMt and the DSMt-CMMt complex side, and the mixtures were cultured for 3 h at 30 °C. plMcoilBiotechnol Microbiol Appl

Table 1 Bacterial strains, plasmids, and primers used in this study cc Relevant characteristic/sequence Source/reference/note

Strain E. coli XL1-Blue supE44 hsdR17 recA1 endA1 gyrA46 thi relA1 lac-F′[proAB+lacIq lacZΔM15 Tn10(tetT)] Stratagene (Cat. no. 200249) BL21 (DE3) hsdS gal (λcIts857 ind-l Sam7 nin-5 lacUV5-T7 gene 1) Novagen (Cat. no. 69450-3) C. glutamicum RES167 Restriction-deficient mutant of ATCC13032, Δ (cglIM-cglIR-cglIIR) From University of Bielefeld Plasmid pET28a Expression vector with N-terminal hexahistidine affinity tag Novagen

pET28-CMCg pET28a derivative for expression of CMCg (Li et al. 2009) pET28-CMCg M1 Expression vector of CMCg M1: deletion of residues 98–100 This study pET28-CMCg M2 Expression vector of CMCg M2: deletion of residues 97–101 This study 60 64 60 64 pET28-CMCg M3 Expression vector of CMCg M3: residues SGGTR are mutated to AAAAA This study 46 57 pET28-CMCg M4 Expression vector of CMCg mutant: R A, R A This study pET21a Expression vector with no tag Novagen

pET21-DSCg pET28a derivative for expression of DSCg This study 212 219 212 219 pET21-DSCg M1 Expression vector of DSCg M1: residues SPAGARYE are mutated to be AAAAAAAAA This study 387 394 387 394 pET21-DSCg M2 Expression vector of DSCg M2: residues HFDKVIDE are mutated to be AAAAAAAA This study Primer

CMCg Fr GACCATATGACTAATGCAGGTGAC This study CMCg M1Rr CGAATTCTTATCCGCGTCCCATGCGCAGG This study CMCg M2Rr CGAATTCTTATCCCATGCGCAGC This study CMCg M3Fr CGCATGAGCGCGGCCGCAGCAGCTCTCGTGCACACCCGAGAAG This study CMCg M3Rr GTGCACGAGAGCTGCTGCGGCCGCGCTCATGCGTGTTTTTCCGATG This study CMCg M4Fr1-1 GGTGAAACGCGCCACGAAGATTTCCCAAACCATC This study CMCg M4Rr2-1 GAAATCTTCGTGGCGCGTTTCACCGCATCGAGG This study CMCg M4Fr1-2 CATCGGAAAAACAGCCATGAGCTCGGGCGGAAC This study CMCg M4Rr2-2 CGAGCTCATGGCTGTTTTTCCGATGGTTTGGG This study DSCg Fr CTCTAGAAAAGGAGGACATATGAATAGGGGTGTGAGTTGG This study DSCg Rr GTGGATCCTGTGGAGCGGAGTTATCTTGA This study DSCg M1Fr GAACCGCGAGTTCGTTGCGAACGCCGCAGCTGCTGCAGCCGCCGCGGCTCTTGCTCGTGAGATCGAC This study DSCgM1Rr CGAGCAAGAGCCGCGGCGGCTGCAGCAGCTGCGGCGTTCGCAACG This study DSCgM2Fr CATCCAATGGCTACAAGACCCGTGCCGCAGCTGCTGCAGCCGCCGCGGTCCAGGGCTTCTTCGAGGTC This study DSCgM2Rr GCCCTGGACCGCGGCGGCTGCAGCAGCTGCGGCACGGGTCTTG This study Appl Microbiol Biotechnol

Recombinant DS proteins in E. coli BL21 (DE3) cells were DS activity synthesized similarly but at 16 °C. Cells were harvested by centrifugation at 10,000×g, washed twice with 0.2 % KCl DS activity was determined according to the procedure de- solution, resuspended in 50 mM Tris–HCl buffer (pH7.5), scribed by Siehl (1997). The assay mixture contained 50 mM and disrupted by sonication in an ice-water bath. The cellular Tris–HCl buffer (pH7.5), 5 mM PEP, 2 mM erythrose-4- lysate was centrifuged at 20,000×g for 30 min at 4 °C, and the phosphate, and enzyme in a volume of 75 μl. The mixture was supernatant was used for protein purification. incubated at 30 °C for 10 min. The reaction was initiated by the

Recombinant wild type and mutant CMCg proteins were addition of enzyme and terminated by the addition of 400 μlof purified with a His-Bind protein purification kit (Novagen, 10 % (wt/vol) trichloroacetic acid. Reaction mixtures without Madison, WI) according to the manufacturer’s instructions. enzyme were run in parallel and used as controls. One unit of

Wild-type DSCg and mutant DSCgM1 were purified by ammo- activity was defined as the amount of enzyme that catalyzed the nium sulfate fractionation and chromatography on a HiTrap Q synthesis of 1 μmol of DAHP per minute at 30 °C. For calcula- HP column (GE Healthcare, Little Chalfont, UK). The wild- tions, we used 4.5×104M−1×cm−1 as the extinction coefficient type DSCg was recovered from ammonium sulfate solution at for DAHP at 549 nm (ε549nm) (Schoner and Herrmann 1976). 20–35 % saturation. DSCgM1 was recovered from ammonium Inhibition of DS activity by prephenate was determined by sulfate solution at 30–50 % saturation. Wild-type DSCg and inclusion of prephenate in the assay mixtures. Parallel exper- mutant DSCgM1 crude extracts were individually loaded onto iments without enzyme in the assay mixture were run as a HiTrap Q HP column. Both DSCg and DSCgM1 were eluted controls. Inhibition constants (Ki)werecalculatedbymeans at NaCl concentrations ranging from 500–700 mM. Proteins of Dixon plots (Dixon 1953). were stored at −20 °C following concentration. Measurements of protein concentrations Chorismate mutase activity Protein concentrations were measured according to the CM activity was assayed by means of the procedure of method of Bradford (1976), with bovine serum albumin as Davidson and Hudson (1987).Theassaymixturecontained the standard. 50 mM Tris–HCl buffer (pH7.5, containing 1 mM dithiothreitol) and 1 mM chorismicacidinatotalvolumeof 0.2 ml. After the mixture was incubated at 37 °C for 5 min, Results enzyme was added to initiate the reaction, and 10 min later, the reaction was stopped by the addition of 0.4 ml of 1 M HCl. One Regulation of DSCg activity by prephenate in the presence unit of CM activity was defined as the amount of enzyme that of CMCg and identification of a putative prephenate binding catalyzed the conversion of 1 μmol of chorismate to prephenate site on CMCg per minute under the conditions used. For calculations, we used 3 −1 −1 17.5×10 M ×cm as the extinction coefficient at 320 nm In this study, we observed that in the presence of CMCg, (ε320nm) for phenylpyruvate (Davidson and Hudson 1987). 2 mM prephenate inhibited DSCg activity by 45.6 %. The Ki

Fig. 1 Prediction of chorismate/prephenate binding sites in the puta- Prephenate was modeled in the PEP of DSCg in the tive DSCg-CMCg complex. (a) Key amino acid residues involved in the complex. (I) amino acid residues involved in the binding of PEP; (II) binding of prephenate to the complex. (b) Amino acid residues in- putative amino acid residues involved in the binding of prephenate. volved in the binding of PEP and prephenate to the complex. Hydrogen bonds are indicated with broken lines Appl Microbiol Biotechnol

Table 2 Specific activities of DS and CM and mutants. Values are Cg Cg Reexamination of the DSMt-CMMt structure and exploration averages from three parallel determinations and standard errors are of the interaction surfaces between DS and CM provided Cg Cg

Enzyme or mutant Specific activity (units/mg) The structures of DSMt and CMMt have previously been solved (PDB: 2B7O; PDB: 2QBV) (Kim et al. 2006; Webby DAHP synthase activity et al. 2005). DSMt contains a core (β/α)8 triosephosphate DSCg 85.1±1.9 barrel domain and CMMt consists of three α- DSCgM1 56.4±3.0 helices. Recently, the structure of the DSMt-CMMt complex DSCgM2 No activity (PDB: 2W1A) also became available (Sasso et al. 2009). In Chorismate mutase activity this study, we reexamined these structures in an attempt to CMCg 124.0±6.9 show the interaction interface between DSMt and CMMt. Sasso CMCgM1 114.2±7.8 et al. (2009)reportedthattheDSMt-CMMt complex is com- CM M2 112.7±12.4 Cg posed of a central core of four DSMt subunits sandwiched CM M3 97.1±5.8 Cg between two CMMt dimers. Our examination of the DSMt- CM M4 No activity Cg CMMt complex structure showed that two regions, designated I and II, play a key role in the interaction between DSMt and CMMt (Fig. 2a). Region I consists of two α-helices (α7, 383HFDRIVDEVQGFFEVHRAL401; α8, 445NTQQSLE 462 64 68 value for prephenate was determined to be 0.945 mM. The LAFLVAEMLRD )ofDSMt and the SGGTR loop and 100 104 inhibition by prephenate was competitive against PEP, but C-terminus ( RGRLG )ofCMMt. Region II contains the 208 215 not against erythrose 4-phosphate. SPAGARYE loop between α2a and α2b of DSMt and 71 79 89 97 Using the crystal structure of the DSMt-CMMt complex H2 ( HSREMKVIE ) and H3 ( KDLAILLLR )ofCMMt (Sasso et al. 2009), a binding site for analogs of chorismate (Fig. 2b). and prephenate was observed. We further identified the corre- DSCg and CMCg showed 77 and 57 % sequence identity sponding amino acid residues in CMCg (Figs. 1a, b), and to DSMt and CMMt, respectively. Therefore, we used the 46 mutated two of the key amino acid residues, Arg and structure of DSMt-CMMt as a reference to identify the amino 57 Arg . The resulting mutant, CMCgM4, showed no CM activ- acid residues that mediate the interaction between DSCg and 46 57 ity (Table 2), suggesting that Arg and Arg were essential CMCg. The sequences of DSCg and CMCg were aligned with for CMCg activity. To determine whether the inactive CMCgM4 those of DSMt and CMMt, respectively (Fig. 2b). Residues in 60 64 97 101 interacted with DSCg, we investigated the effect of prephenate the SGGTR and RGKLG regions of CMCg and the 212 219 387 394 on DS activity in the presence of CMCgM4, and found that the SPAGARYE and HFDKVIDE regions of DSCg regulatory effect was retained but was somewhat reduced corresponded to residues in the 64SGGTR68 (orange) and 100 104 (Table 3).PrephenatereducedDSactivityby26.6%inthe RGRLG (magenta) regions of CMMt and to residues in 208 215 383 390 presence of mutant CMCgM4, whereas DS activity was re- the SPAGARYE (blue) and HFDRIVDE (red) duced by 45.6 % in the presence of the wild-type CMCg. regions of DSMt (Fig. 2a), respectively. We deduced that

Table 3 Effects of prephenate Enzyme assays Remained DS activity (%) Specific DS activity (units/mg) (2 mM) on DSCg and DSCgM1 activities when CMCg or CMCg mutants were present in the re- DSCg+CMCg 100.0±5.8 95.3±5.5 action mixtures. Values are av- DSCg+CMCg+prephenate 54.4±6.3 51.8±6.0 erages from three parallel DS +CM M1 100±2.3 93.0±2.1 determinations and standard er- Cg Cg rors are provided DSCg+CMCgM1+prephenate 86.8±3.3 80.7±3.1 DSCg+CMCgM2 100±15.1 87.3±13.2

DSCg+CMCgM2+prephenate 84.8±16.2 74.0±14.1

DSCg+CMCgM3 100±7.7 81.7±6.2

DSCg+CMCgM3+prephenate 70.8±8.3 57.8±6.8

DSCg+CMCgM4 100±5.6 70.3±3.9

DSCg+CMCgM4+prephenate 73.4±11.0 51.6±7.7

DSCgM1+CMCg 100±6.0 58.2±3.5

DSCgM1+CMCg+prephenate 89.6±15.8 52.1±9.2 Appl Microbiol Biotechnol

Fig. 2 (a) Structure modeling of DSMt-CMMt interface regions: (cyan)CMMt monomer, (green)DSMt monomer, (H1, H2, and H3) α- helices of CMMt. Mutated sites (orange, magenta, red, and blue). (b) Identification of corresponding regions of the interaction interface of the DSCg-CMCg complex by sequence alignment: (H) α- helix, (L) loop; (*)CMCgM1 or DSCgM1, (△)CMCgM2 or DSCgM2, (●)CMCgM3, and (▲)CMCgM4

these amino acid residues were involved in the interaction between DSCg and CMCg, and therefore, we subjected them to mutagenesis analysis.

Construction of DSCg and CMCg mutants and determination of their catalytic activities

We constructed two DSCg mutants [DSCgM1 212 219 387 394 ( SPAGARYE )andDSCgM2 ( HFDKVIDE )] 98 100 and three CMCg mutants [CMCgM1 ( GRL ), CMCgM2 97 101 60 64 ( RGRLG ), and CMCgM3 ( SGGTR )] by replacing or deleting the amino acid residues indicated in parentheses. Upon induction with isopropyl ß-D-1-thiogalactopyrano- Fig. 3 Models of the structures of DSCg and the DSCg-CMCg complex: side, recombinant E. coli BL21 (DE3) cells promptly syn- (yellow)DSCg,(green)DSCg in the DSCg-CMCg complex, and (magenta, blue, and cyan) three loops above the DS catalytic center discussed in thesized the target proteins. All target proteins were Cg the text. These models were based on the structures of DSMt (PDB: obtained as soluble proteins, except DSCgM2, which was 2B7O) and the DSMt-CMMt complex (PDB: 2W1A) Appl Microbiol Biotechnol

obtained as insoluble inclusion bodies and showed no ac- compete with PEP for the binding site in the DSCg-CMCg tivity. The proteins were purified from recombinant E. coli complex, we modeled the putative DSCg-CMCg complex BL21 (DE3) cells, and their catalytic activities were deter- and DSCg structures using structural information for the mined. Enzymatic activity assays demonstrated that the DSMt-CMMt complex and DSMt alone (Sasso et al. 2009; activities of all the CMCg mutants were nearly identical to Webby et al. 2005). The model suggested that prephenate the original activities, indicating that the mutated amino acid binds at the binding site of PEP in the catalytic center of residues were not essential for activity. DSCg. Further analysis revealed that three loops, 138PRSSDLDGNGLPN150 (magenta), 375NTFTASNG 386 418 446 Effect of site-directed mutagenesis on prephenate regulation YKTR (blue), and G(AA)27P (cyan), were located of DS activity near the catalytic center of DSCg (Fig. 3). Comparison of the structural models for DSCg and the DSCg-CMCg complex 418 446 To explore the DSCg and CMCg interaction and the regula- revealed that the G(AA)27P loop in DSCg was flexible, tion of DSCg by prephenate, we determined DS activities as previously reported for DSMt by Webby et al. (2005), with wild-type DSCg and mutants of CMCg, as well with whereas in the complex, the loop was rigid owing to its mutant DSCg and wild-type CMCg (Table 3). The regulatory interaction with CMCg. We hypothesized that the flexible 418 446 effect of prephenate on DSCg activity in the presence of G(AA)27P loop prevented the entrance of prephenate CMCgM1 or CMCgM2 was significantly reduced compared into the catalytic center of DSCg. This hypothesis would with the effect in the presence of CMCg:DSCg activity was explain why prephenate inhibited DSCg activity by compet- reduced by 13.2 and 15.2 % in the presence of CMCgM1 or ing for PEP in the presence of CMCg but not in its absence. CMCgM2, respectively. The regulatory effect of prephenate Our hypothesis is supported by a report that conformational in the presence of CMCgM3 remained strong (prephenate changes in amino acid loops near the catalytic center of the reduced DSCg activity by 29.2 %). Regulation of DS activity DS in Thermotoga maritima regulate the enzyme’s activity by prephenate was retained in the DSCgM1-CMCg complex, (Shumilin et al. 2004). The new knowledge on the regula- but to a lesser extent than in the DSCg-CMCg complex tion of DS activity from this study might be useful for the (Table 3). Both CMCgM1 and CMCgM2 have a truncated construction of more productive strains of aromatic amino C-terminus, whereas CMCgM3 carries a deletion far from acids with C. glutamicum. the C-terminus. In consideration of the results described in this and previous sections, we concluded that the C-terminus Acknowledgments This work was supported by a grant from the Ministry of Science and Technology of China (973-project, grant no. of CMCg was important for the CMCg and DSCg interaction. 2012CB721104).

Discussion

C. glutamicum is a commercial important amino acid pro- References ducer and plays a key role in the production of glutamate, lysine, and various vitamins (Ikeda and Nakagawa 2003). In Bradford MM (1976) A rapid and sensitive method for the quantization this study, we examined the interaction between CMCg and of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 DSCg with bioinformatic tools and experimental confirma- Davidson BE, Hudson GS (1987) Chorismate mutase-prephenate de- tion. Taken together, these results suggest that disruption of hydrogenase from Escherichia coli. Methods Enzymol 142:440– the amino acid residues of either DSCg or CMCg at the 450 interaction interface of the putative DSCg-CMCg complex DeLano WL (2002) Unraveling hot spots in binding interfaces: prog- affected the interaction between the two proteins and thus ress and challenges. Curr Opin Struct Biol 12:14–20 Dixon M (1953) The determination of constants. altered the regulatory effect of prephenate on the DS activity Biochem J 55:170–171 of the DSCg-CMCg complex. We concluded that prephenate Herrmann KM (1995) The shikimate pathway: early steps in the – regulation of DSCg activity interactions relied on the inter- of aromatic compounds. Plant Cell 7:907 919 Ikeda M, Nakagawa S (2003) The Corynebacterium glutamicum ge- action surfaces between DSCg and CMCg molecules. nome: features and impacts on biotechnological processes. Appl However, additional mechanism for prephenate regulation Microbiol Biotechnol 62:99–109 of DSCg activity might exist. As indicated by our results, Jiao W, Hutton RD, Cross PJ, Jameson GB, Parker EJ (2012) Dynamic prephenate competed for PEP in the presence of CMCg but cross-talk among remote binding sites: the molecular basis for unusual synergistic allostery. J Mol Biol 415:716–726 not in DSCg alone. Considering that PEP is one of the sub- Kikuchi Y, Tsujimoto K, Kurahashi O (1997) Mutational analysis of strates for DAHP synthase, it would be interesting to ex- the feedback sites of phenylalanine-sensitive 3-deoxy-D-arabino- plore further that CMCg also had binding site for prephenate heptulosonate-7-phosphate synthase of Escherichia coli. Appl and/or PEP. To elucidate why prephenate was able to Environ Microbiol 63:761–762 Appl Microbiol Biotechnol

Kim SK, Reddy SK, Nelson BC, Vasquez GB, Davis A, Howard AJ, Siehl DL (1997) Inhibitors of EPSP synthase, glutamine synthetase, Patterson S, Gilliland GL, Ladner JE, Reddy PT (2006) Biochem- and synthesis. In: Row RM, Burton JD, Kuhr RJ (eds) ical and structural characterization of the secreted chorismate Herbicide Activity: Toxicology, Biochemistry and Molecular Bi- mutase (Rv1885c) from Mycobacterium tuberculosis H37Rv: an ology. IOS Press, Amsterdam, pp 37–67 *AroQ enzyme not regulated by the aromatic amino acids. J Shumilin IA, Bauerle R, Wu J, Woodard RW, Kretsinger RH (2004) Bacteriol 188:8638–8648 Crystal structure of the reaction complex of 3-deoxy-D-arabino- Li PP, Liu YJ, Liu SJ (2009) Genetic and biochemical identification of heptulosonate-7-phosphate synthase from Thermotoga maritima the chorismate mutase from Corynebacterium glutamicum. Mi- refines the catalytic mechanism and indicates a new mechanism of crobiology 155:3382–3391 allosteric regulation. J Mol Biol 341:455–466 Liao HF, Lin LL, Chien HR, Hsu WH (2001) Serine 187 is a crucial Tauch A, Kassing F, Kalinowski J, Pühler A (1995) The Corynebac- residue for allosteric regulation of Corynebacterium glutamicum terium xerosis composite transposon Tn5432 consists of two 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase. FEMS identical insertion sequences, designated IS1249, flanking the Microbiol Lett 194:59–64 erythromycin resistance gene ermCX. Plasmid 34:119–131 Liu YJ, Li PP, Zhao KX, Wang BJ, Jiang CY, Drake HL, Liu SJ (2008) Tauch A, Kirchner O, Loffler B, Gotker S, Pühler A, Kalinowski J Corynebacterium glutamicum contains 3-deoxy-D-arabino- (2002) Efficient electrotransformation of Corynebacterium heptulosonate 7-phosphate synthases that display novel biochem- diphtheriae with a mini-replicon derived from the Corynebacte- ical features. Appl Environ Microbiol 74:5497–5503 rium glutamicum plasmid pGA1. Curr Microbiol 45:362–367 Pittard AJ (1996) Biosynthesis of aromatic amino acids in Escherichia Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG coli and Salmonella. Cellular and Molecular Biology. American (1997) The CLUSTAL_X windows interface: flexible strategies Society of Microbiology, Washington, pp 458–484 for multiple sequence alignment aided by quality analysis tools. Romero RM, Roberts MF, Phillipson JD (1995) Anthranilate synthase Nucleic Acids Res 25:4876–4882 in microorganisms and plants. Phytochemistry 39:263–276 Wang RH, Chen RC, Liu RZ (2008) A modified method of Samsbrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory QuikChange site-directed mutagenesis. J Xiamen Univ 47:z2 manual, 2nd edn. Cold Spring Harbor Laboratory Press, New York Webby CJ, Baker HM, Lott JS, Baker EN, Parker EJ (2005) The Sasso S, Okvist M, Roderer K, Gamper M, Codoni G, Krengel U, Kast structure of 3-deoxy-D-arabino-heptulosonate 7-phosphate P (2009) Structure and function of a complex between chorismate synthase from Mycobacterium tuberculosis reveals a common mutase and DAHP synthase: efficiency boost for the junior part- catalytic scaffold and ancestry for type I and type II enzymes. J ner. EMBO J 28:2128–2142 Mol Biol 354:927–939 Schneider CZ, Parish T, Basso LA, Santos DS (2008) The two chorismate Webby CJ, Jiao W, Hutton RD, Blackmore NJ, Baker HM, Baker EN, mutases from both Mycobacterium tuberculosis and Mycobacterium Jameson GB, Parker EJ (2010) Synergistic allostery, a sophisti- smegmatis: biochemical analysis and limited regulation of promoter cated regulatory network for the control of aromatic amino acid activity by aromatic amino acids. J Bacteriol 190:122–134 biosynthesis in Mycobacterium tuberculosis. J Biol Chem Schoner R, Herrmann KM (1976) 3-Deoxy-D-arabino-heptulosonate 7- 285:30567–30576 phosphate synthase. Purification, properties, and kinetics of the Wu J, Sheflyan GY, Woodard RW (2005) Bacillus subtilis 3-deoxy-D- tyrosine-sensitive isoenzyme from Escherichia coli. J Biol Chem arabino-heptulosonate 7-phosphate synthase revisited: resolution 251:5440–5447 of two long-standing enigmas. Biochem J 390:583–590