Characterization and structure of glyceraldehyde-3-phosphate dehydrogenase type 1 from Escherichia coli

The crystal structure of full-length glyceraldehyde-3-phosphate dehydrogenase type 1 (GAPDH1) from Escherichia coli was determined at 1.88 Å resolution. Analysis of the NAD⁺-bound form showed some differences between the structures of E. coli GAPDH1 and human GAPDH. As E. coli GAPDH1 shares 100% identity with GAPDH from Shigella sonnei, its structure may help in finding a drug for the treatment of shigellosis.

Abstract

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is a key enzyme in the glycolytic pathway that catalyzes the conversion of d-glyceraldehyde 3-phosphate to 1,3-diphosphoglycerate. Here, the full-length GAPDH type 1 from Escherichia coli (EcGAPDH1) was cloned and overexpressed, and the protein was purified. Biochemical analyses found that the optimum reaction temperature and pH of EcGAPDH1 were 55°C and 10.0, respectively. The protein has a certain amount of thermostability. Crystals of EcGAPDH1 were obtained using the sitting-drop vapor-diffusion technique and X-ray diffraction data were collected to 1.88 Å resolution. Characterization of the crystals showed that they belonged to space group P4₁2₁2, with unit-cell parameters a = b = 89.651, c = 341.007 Å, α = β = γ = 90°. The structure of EcGAPDH1 contains four subunits, each of which includes an N-terminal NAD⁺-binding domain and a C-terminal catalytic domain. Analysis of the NAD⁺-bound form showed some differences between the structures of EcGAPDH1 and human GAPDH. As EcGAPDH1 shares 100% identity with GAPDH from Shigella sonnei, its structure may help in finding a drug for the treatment of shigellosis.

1. Introduction  

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is an essential enzyme in the process of glycolysis, in which it catalyses the oxidative phosphorylation of d-glyceraldehyde 3-phosphate (G3P) to 1,3-diphosphoglycerate (BPG) in the presence of inorganic phosphate and nicotinamide adenine dinucleotide (NAD⁺). In addition, GAPDH contributes to other biological processes, including membrane fusion (Zhang et al., 2007 ▸), the export of nuclear RNA, DNA replication and repair (Kosova et al., 2017 ▸), cell senescence (Nicholls et al., 2012 ▸) and apoptosis (Fukuhara et al., 2001 ▸). Recent research has shown that GAPDH type 1 from Escherichia coli (EcGAPDH1) is also involved in DNA repair, protein folding and protein synthesis (Ferreira et al., 2013 ▸).

EcGAPDH1 shares its sequence with GAPDH from the pathogenic Shigella sonnei, the causative agent of the diarrheal disease shigellosis, and shows only low identity to human GAPDH. In the light of this, we can use EcGAPDH1 as a model to study the structure of GAPDH from Shigella sonnei. Moreover, since the sequence identity of EcGAPDH1 to human GADPH is lower than that to S. sonnei GAPDH, this may allow us to find a way to stop glycolysis in S. sonnei without affecting human cells.

2. Materials and methods  

2.1. Cloning, expression and purification  

The target fragment of EcGAPDH1 was amplified from E. coli using PCR. The amplified DNA fragment and the pET-28a vector (Novagen, USA) were digested by the NheI/XhoI restriction endonucleases and ligated by T4 DNA ligase (New England Biolabs, USA). The confirmed recombinant vectors were transformed into E. coli strain BL21 (DE3). The bacteria were grown at 37°C in LB medium with 50 mg l⁻¹ kanamycin and 34 mg l⁻¹ chloramphenicol. EcGAPDH overexpression was induced by adding 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) when the OD₆₀₀ reached 0.8. After induction for 10 h, the bacteria were harvested by centrifugation at 5000g for 5 min. The pellet was suspended in lysis buffer consisting of 300 mM NaCl, 50 mM sodium phosphate, 10 mM imidazole pH 8.0 and was homogenized using a high-pressure homogenizer (JN-3000 Plus, JNBIO, China) at a pressure of 12 kg cm⁻². Cell debris was removed by centrifugation at 15 000 rev min⁻¹ for 40 min. The supernatant was loaded onto a HisTrap HP column (GE Healthcare) and the targeted protein was eluted with elution buffer consisting of 300 mM NaCl, 50 mM sodium phosphate, 50–500 mM imidazole pH 8.0. The protein was further purified by size-exclusion chromatography on a HiPrep 26/60 Sephacryl S-200 HR column (GE Healthcare) with 400 mM NaCl, 40 mM Tris–HCl pH 8.0 as the mobile phase (Zhang et al., 2014 ▸). The collected protein was concentrated and buffer-exchanged. The final concentration of EcGAPDH was 60 mg ml⁻¹ in buffer consisting of 4 mM Tris–HCl pH 8.0, 5 mM NaCl. The purity of EcGAPDH1 was analyzed by 12% SDS–PAGE and the concentration was determined using the Bradford assay. 20 µl EcGAPDH1 protein was dispensed into a PCR tube, flash-frozen in liquid nitrogen and stored at −80°C. The His tag was not removed during the enzymatic assay or the crystallization steps. Macromolecule-production information is summarized in Table 1 ▸.

Table 1. Macromolecule-production information.

Source organism	E. coli BL21 (DE3)
Forward primer	CACGCTAGCATGAGTAAAGTTGGTATTAACGG
Reverse primer	CACCTCGAGTCAGAGTTTAGCGAATTTTTCGA
Cloning vector	pET-28a
Expression vector	pET-28a
Expression host	E. coli BL21 (DE3)
Complete amino-acid sequence of the construct produced	HHHHHHSSGLVPRGSHMASMSKVGINGFGRIGRLVLRRLLEVKSNIDVVAINDLTSPKILAYLLKHDSNYGPFPWSVDFTEDSLIVDGKSIAVYAEKEAKNIPWKAKGAEIIVECTGFYTSAEKSQAHLDAGAKKVLISAPAGEMKTIVYNVNDDTLDGNDTIVSVASCTTNCLAPMAKALHDSFGIEVGTMTTIHAYTGTQSLVDGPRGKDLRASRAAAENIIPHTTGAAKAIGLVIPELSGKLKGHAQRVPVKTGSVTELVSILGKKVTAEEVNNALKQATTNNESFGYTDEEIVSSDIIGSHFGSVFDATQTEITAVGDLQLVKTVAWYDNEYGFVTQLIRTLEKFAKL

2.2. Enzymatic assay of recombinant EcGAPDH  

The standard oxidative phosphorylation reaction of EcGAPDH1 was followed using a DU800 UV–visible spectrophotometer (Beckman, USA) to monitor the change in NADH concentration at 340 nm. The standard reaction mixture (200 µl) consisted of 40 mM triethanolamine, 50 mM K₂HPO₄, 2 mM d-glyceraldehyde 3-phosphate, 2 mM NAD⁺, 0.2 mM EDTA pH 8.9 (Cook et al., 2009 ▸). The reaction was initiated by adding 0.022 µg purified EcGAPDH1 per 100 µl. All samples were tested three times and denatured EcGAPDH1 was used as a blank control.

The optimum temperature of EcGAPDH1 was determined using different temperatures ranging from 30 to 60°C in the standard reaction solution. The optimum pH of EcGAPDH1 was determined at the optimum temperature in 40 mM tri­ethanolamine, 50 mM K₂HPO₄, 0.2 mM EDTA in the pH range 6–12.

The influences of temperature, pH, organic solvents and detergents on EcGAPDH1 were analyzed by measuring the residual enzyme activity at the optimum temperature and pH. Thermostability was measured after incubating the enzyme for 2.5 h at temperatures of 30, 35, 40, 45, 50, 55, 60 and 65°C. The effect of pH on the enzyme activity was measured after incubating the enzyme for 2 h in phosphate-buffered saline (PBS) in the pH range 6–12 at room temperature. The tolerance of EcGAPDH1 to organic solvents and detergents was determined after incubation at room temperature for 4 h. The organic solvents included 15% acetone, acetonitrile, ethanol and 2-propanol and the detergents included 1% Tween 20, Triton X-100 and sodium dodecyl sulfate (SDS).

2.3. Crystallization and data collection  

The initial crystallization conditions for EcGAPDH1 were screened using the commercial crystallization kits Crystal Screen HT, Index HT, Natrix HT, PEG/Ion HT and SaltRx HT from Hampton Research and Wizard I–IV from Emerald Bio. A total of 672 conditions were screened in 96-well plates using the sitting-drop vapor-diffusion method by mixing 0.2 µl 30 mg ml⁻¹ protein solution with 0.2 µl reservoir solution and equilibrating against 40 µl reservoir solution. Crystals formed in several conditions. We optimized the conditions by modifying the protein concentration, PEG 400 concentration and pH over a wide range. We thus generated fusiform-like crystals within a week in a reservoir solution consisting of 0.1 M calcium acetate, 0.1 M sodium acetate pH 6.5, 25%(w/v) PEG 400. These crystals were soaked for about 30 s in a cryoprotectant solution that consisted of 70% reservoir solution and 30%(w/v) PEG 400 before being flash-cooled in a nitrogen stream. Crystallization information is summarized in Table 2 ▸.

Table 2. Crystallization.

Method	Vapor diffusion, sitting drop
Plate type	96-well plates
Temperature (K)	293
Protein concentration (mg ml−1)	30
Buffer composition of protein solution	4 mM NaCl, 5 mM Tris–HCl pH 8.0
Composition of reservoir solution	100 mM sodium acetate pH 4.6, 30%(w/v) PEG 400, 200 mM calcium acetate
Volume and ratio of drop	2.4 µl, 1:1
Volume of reservoir (µl)	100

X-ray diffraction data were collected at a wavelength of 0.97915 Å at 100 K using a MAR DTB detector on beamline BL17U1 at Shanghai Synchrotron Radiation Facility (SSRF), People’s Republic of China. The oscillation angle was 1°. The diffraction data were indexed, integrated and scaled with the HKL-2000 package (Otwinowski & Minor, 1997 ▸). Data-collection and processing statistics are summarized in Table 3 ▸.

Table 3. Data collection and processing.

Diffraction source	Beamline BL17U1, SSRF
Wavelength (Å)	0.97915
Temperature (K)	100
Detector	ADSC Quantum 315 CCD
Crystal-to-detector distance (mm)	250
Rotation range per image (°)	1
Total rotation range (°)	360
Exposure time per image (s)	0.2
Space group	P41212
a, b, c (Å)	89.65, 89.65, 341.01
α, β, γ (°)	90, 90, 90
Mosaicity (°)	0.625
Resolution range (Å)	50–1.88 (1.91–1.88)
Total No. of reflections	1016362
No. of unique reflections	114194
Completeness (%)	99.9 (100)
Multiplicity	8.9 (9.8)
〈I/σ(I)〉	13.5 (2.0)
R r.i.m. † (%)	21.8 (89.7)

^†Estimated R _r.i.m. = R _merge[N/(N − 1)]^1/2, where N is the data multiplicity.

2.4. Structure solution and refinement  

The structure of EcGAPDH1 was determined by molecular replacement in the CCP4 suite (Winn et al., 2011 ▸) using the structure of GAPDH from methicillin-resistant Staphylococcus aureus MRSA252 (MrsaGADPH; PDB entry 3lvf; Mukherjee et al., 2010 ▸) as a search model. Model building was carried out by Coot (Emsley et al., 2010 ▸) and refinement used Phenix (Liebschner et al., 2019 ▸) and REFMAC5 (Murshudov et al., 2011 ▸). The result of refinement was assessed by the values of R _work and R _free. The stereochemical quality of the model was evaluated by PROCHECK (Laskowski et al., 1993 ▸). All graphics were visualized using PyMOL (Schrödinger). Data-collection and processing statistics are summarized in Table 3 ▸.

3. Results and discussion  

We confirmed the identity of EcGAPDH1 by comparison with S. sonnei and human GAPDH (HuGAPDH) using the Basic Local Alignment Search Tool from the National Center for Biotechnology Information. The sequence of EcGAPDH1 shares 100% identity with that of GAPDH from S. sonnei, but shares only 45% identity with that of HuGAPDH (Supplementary Fig. S1).

EcGAPDH1 was purified by Ni–NTA affinity chromatography and gel-filtration chromatography. The results showed that EcGAPDH1 had greater than 99% purity as assessed by Quantity One (Bio-Rad; Fig. 1 ▸). The results from gel-filtration chromatography showed that EcGAPDH1 mainly exists as tetramers and that a small part exists in the form of polymers in solution buffer.

Figure 1.

Gel-filtration profiles of EcGAPDH1. (a) Top: gel-filtration profile of EcGAPDH on a HiPrep 26/60 Sephacryl S-100 HR column. EcGAPDH eluted at 108 ml. Bottom: 12% SDS–PAGE of wild-type protein eluted from gel filtration. (b) Gel-filtration standards (Bio-Rad, California, USA) on a HiPrep 26/60 Sephacryl S-100 HR column. Bovine thyroglobulin, with a molecular weight (MW) of 670 kDa, eluted at 92.3 ml. Bovine γ-globulin (bovine), with a MW of 158 kDa, eluted at 104.6 ml. Chicken ovalbumin, with a MW of 44 kDa, eluted at 132 ml. Horse myoglobin, with a MW of 17 kDa, eluted at 171.3 ml. Vitamin B₁₂, with a MW of 1.35 kDa, eluted at 286 ml.

The optimum reaction temperature was 55°C and the enzymatic activity drastically decreased to 50% at 60°C (Fig. 2 ▸ a). The results also showed pH 10.0 to be the optimum reaction pH (Fig. 2 ▸ b). Analysis of the thermostability of EcGAPDH1 showed that the enzyme retains 100% activity after incubation for 2.5 h at temperatures of 30, 35, 40, 45 and 50°C. EcGAPDH1 has higher thermostability than GAPDH from pea seed (Gani et al., 2016 ▸). This characteristic may help E. coli to survive under extreme conditions. The enzyme activity sharply decreased to 20% at 55°C and was abolished at 60°C (Fig. 2 ▸ c), suggesting that EcGAPDH1 has a certain amount of temperature tolerance but is easy to inactivate. EcGAPDH1 also showed at least 100% enzyme activity after incubation for 2 h at room temperature at pH 6–11 (Fig. 2 ▸ d). EcGAPDH1 activity decreased by 40–60% in 1% Tween 20 or 15% acetone, ethanol or 2-propanol, and almost all activity was lost in 15% acetonitrile. Furthermore, 1% SDS had little effect on EcGAPDH1 activity, while 1% Triton X-100 increased its activity (Fig. 2 ▸ e), suggesting varying effects of organic solvents and detergents on the activity of EcGAPDH.

Figure 2.

Characterization of EcGAPDH1. (a) The optimum temperature of EcGAPDH. (b) The optimum pH of EcGAPDH. (c) Effect of different temperatures on the activity of EcGAPDH. (d) Effect of pH on the activity of EcGAPDH. (e) Effect of organic solvents and detergents on the activity of EcGAPDH. The enzyme activities in (c), (d) and (e) was measured at 328 K and pH 10 under standard conditions. The enzyme activity without changes in temperature, pH and organic solvent was defined as 100%.

Fusiform-like single crystals (Supplementary Fig. S2a) were obtained by combining 1.2 µl 30 mg ml⁻¹ protein solution with 1.2 µl reservoir solution and equilibrating against 100 µl reservoir solution. We obtained a single crystal that diffracted X-rays and collected 360° of data as shown in Table 1 ▸. The results showed that the EcGAPDH1 crystal belonged to space group P4₁2₁2, with unit-cell parameters a = b = 89.651, c = 341.007, α = β = γ = 90°, and diffracted to 1.88 Å resolution (Supplementary Fig. S2b). As the molecular weight of EcGAPDH1 is almost 36 kDa, the Matthews coefficients calculated for four molecules in the asymmetric unit of the crystal was 2.4 ų Da⁻¹ and the solvent content was 48.78%. Crystallographic statistics for data collection and structure refinement are summarized in Tables 3 ▸ and 4 ▸, respectively Each asymmetric unit contains four subunits named O, P, Q and R (Fig. 3 ▸ a). The refined structure has three noncrystallographic axes named P, Q and R (Ferreira et al., 2013 ▸). Using the O subunit as a reference to superpose the backbone atoms of the P, Q and R subunits, the root-mean-square deviation (r.m.s.d.) values are 0.168 Å (287 atoms), 0.133 Å (289 atoms) and 0.158 Å (280 atoms), respectively. The results indicate that the conformations of the four different subunits are very similar (Fig. 3 ▸ a).

Table 4. Structure refinement.

Resolution range (Å)	37.81–1.88 (1.95–1.88)
Completeness (%)	99.9
No. of reflections, working set	108650
No. of reflections, test set	5753
Final R cryst	0.180
Final R free	0.246
No. of NAD molecules	4
No. of water molecules	544
R.m.s. deviations
Bonds (Å)	0.008
Angles (°)	0.863
Average B factor (Å2)	23.78
Ramachandran plot
Most favored (%)	96
Allowed (%)	0.075
Disallowed (%)	0

Figure 3.

Overall structure of EcGAPDH1. (a) Overall structure of EcGAPDH showing a cartoon representation of the four subunits in the asymmetric unit: P (cyan), O (magenta), Q (green) and R (orange). The four NAD⁺ molecules are shown as sticks. (b) The amino acids that interact with NAD⁺ and the active amino acid in the P subunit. (c) Cartoon representation of the monomer structure of EcGAPDH containing an NAD⁺-binding domain and a catalytic domain.

Each of the subunits can be divided into two domains: the N-terminal NAD⁺-binding domain and the C-terminal catalytic domain. The NAD⁺-binding domain is typically a Rossman fold containing eight β-strands, namely β1 (Lys3–Asn7), β2 (Asp28–Asn33), β3 (Val58–Phe60), β4 (Ser64–Val67), β5 (Lys70–Tyr75), β6 (Ile92–Glu95), β7 (Lys116–Ile119) and β8 (Ile144–Ser146). The strands are connected by either helices or short loops. β3 and β5 are antiparallel to the other six parallel β-strands. There are four α-helices in this domain: α1 (Gly10–Val23), α2 (Ser37–His47), α3 (Ser102–Ser106) and α4 (Gln107–Ala112). The catalytic domain contains eight mixed β-sheets, β9 (Ile168–Ala178), β10 (Ile205–His207), β11 (Leu226–Val231), β12 (Ser239–Leu247), β13 (Phe270–Thr273), β14 (Ser289–Asp292), β15 (Glu297–Val301) and β16 (Leu304–Tyr313), and three long α-helices, α5 (Ser149–Gly167), α6 (Thr252–Thr264) and α7 (Gln317–Lys332). The catalytically active residues Cys150 and His177 are situated in α5 and β9, respectively (Fig. 3 ▸ c).

There are four subunits in the structures of EcGAPDH1, HuGAPDH (PDB entry 1u8f; Jenkins & Tanner, 2006 ▸) and MrsaGAPDH (PDB entry 3lvf; Mukherjee et al., 2010 ▸). The r.m.s.d.s on C^α atoms are 1.006 (1152 atoms) and 0.711 (1197 atoms) after superimposition of the C^α atoms of HuGAPDH and MrsaGAPDH onto those of EcGAPDH. A structure-based sequence alignment shows that HuGAPDH and MrsaGAPDH share 45% and 58% identity with EcGAPDH, respectively (Fig. 4 ▸).

Figure 4.

Structure-based alignment of the EcGAPDH1, HuGAPDH and MrsaGAPDH sequences. Secondary-structure elements for EcGAPDH are shown above the alignment. Strictly conserved residues are shown in a white font on a red background. Amino acids that interact with NAD⁺ are represented in cyan boxes where the amino acids are conserved and in a green box where they differ. The active amino acid is shown in a yellow box.

3D structural analysis of EcGAPDH1 compared with the structures of HuGAPDH and MrsaGAPDH shows that the main difference is the loop conformation, especially the S-loop (Supplementary Fig. S3). The S-loop is a long winding region formed by amino acids Ala178–Ile205 along the R dyad groove.

Further analysis of the hydrogen bonds between the NAD⁺ cofactor and amino acids in the structures of EcGAPDH1, HuGAPDH and MrsaGAPDH (Fig. 5 ▸) shows that several amino acids, Ile12, Arg11, Asp34 and Ser120 using EcGAPDH1 numbering, are highly conserved in all three structures, with one amino acid that differs. In EcGAPDH1, the N atom of the adenine of NAD⁺ forms a hydrogen bond to Lys78. This amino acid was changed to an arginine and a proline at position 78 in HuGAPDH and MrsaGAPDH, respectively. The N atom of the nicotinamide of NAD⁺ and the side chain of Asn316 form hydrogen bonds in the structures of HuGAPDH and MrsaGAPDH. However, this hydrogen bond disappears in the EcGAPDH1 structure. A new hydrogen bond between the O atom of the main chain of Asn315 and the N atom of the adenine of NAD⁺ is formed. This difference might allow us to find a way to stop glycolysis in S. sonnei without affecting human cells.

Figure 5.

NAD⁺-binding sites of (a) EcGAPDH1, (b) HuGAPDH and (c) MrsaGAPDH. The hydrogen bonds formed between residues and NAD⁺ are shown as yellow dotted lines. The one differing amino acid is highlighted with a circle.

In summary, we purified and characterized EcGAPDH1, crystallized the protein and analyzed its structure. The structure of EcGAPDH1 may facilitate the understanding of the catalytic process and function of GAPDH in S. sonnei.

Supplementary Material

PDB reference: glyceraldehyde-3-phosphate dehydrogenase type 1, 7c5f

Funding Statement

This work was funded by National Key Research and Development Program of China grant 2016YFA0500600. Science and Technology Research Program of Shanghai grant 19DZ2282100.