Structural Study of the Complex of cblC Methylmalonic Aciduria and Homocystinuria-Related Protein MMACHC with Cyanocobalamin

Abstract

MMACHC is an essential protein for the body to metabolise vitamin B₁₂, and its deficiency will cause cblC-type methylmalonic aciduria and homocystinuria. MMACHC can interact with cyanocobalamin (a type of vitamin B₁₂) cofactor and plays an important role in targeting cyanocobalamin to the enzyme of interest. In this paper, the GST-tag fusion-tagged MMACHC protein was successfully expressed by Escherichia coli (E. coli) low-temperature induction, and the high-purity MMACHC protein was successfully purified by affinity chromatography and gel filtration. Further, the crystal structure of MMACHC and cyanocobalamin complex was obtained with a resolution of 1.93 Å using X-ray diffraction. By analysing the complex structure of MMACHC and cyanocobalamin, we revealed the reasons for the diversity of MMACHC substrates and explained the reasons for the differences in disease conditions caused by different MMACHC site mutations. The acquisition of the complex structure of MMACHC and cyanocobalamin will play a significant role in promoting research on the metabolic pathway of vitamin B₁₂.

1. Introduction

Methylmalonic aciduria (MMA) is a heterogeneous autosomal recessive metabolic disorder of cobalamin (vitamin B₁₂) [1], which encompasses different complement groups, namely cblA–cblG, cblJ, cblX, and mut [2,3,4,5]. Cobalamin C (cblC) deficiency is the most common disorder and causes methylmalonic aciduria combined with homocystinuria. Previous studies suggested that homocysteine toxicity, glutathione metabolism, oxidative stress, the effects of the pharmacological activation, impaired methylation/phosphorylation, mislocalisation of RNA-binding proteins, and endoplasmic reticulum stress may contribute to disease onset [6,7,8,9,10,11,12,13].

The gene responsible for the cblC deficiency, which is called methylmalonic aciduria type C and homocystinuria (MMACHC), has been identified on chromosome 1p34.1 and catalyses the decyanation and dealkylation of cobalamin [14].

Vitamin B₁₂ is a very large molecular organic compound, which can be called different cobalt amine according to the connection to a beta on cobalt different functional group (R group). In humans, there are two active cofactor forms of cobalamin, adenosylcobalamin (AdoCbl) for methylmalonyl CoA mutase and methylcobalamin (MeCbl) for methionine synthase (Figure 1). Although humans cannot synthesise cobalamin, we can ingest it from the diet or receive it from gut bacteria [2]. However, as optical instability, the most incoming cobalamin from the diet is an inactive form—namely, cyanocobalamin (CNCbl). MMACHC catalyses CNCbl reductive decyanation to yield cob(II)alamin and cyanide. Cob(II)alamin can be converted to MeCbl or AdoCbl in the further reaction. Previous studies reported that MMACHC catalyses alkylcobalamins, such as MeCbl and AdoCbl, through a dealkylation reaction. This reaction may be involved in the regulation of MeCbl and AdoCbl in vivo [15].

The full-length MMACHC contains 282 amino acids, and the mass weight is 31.7 kDa [14]. The primary sequence of MMACHC is well conserved among mammals, but no homologous proteins are known in prokaryotes. MMACHC includes an unconventional vitamin B₁₂-binding motif, and residues 122-HXXGX-126-154-GG-156 are important for vitamin B₁₂ binding. Residues 181–282 in MMACHC display apparent structural similarity to the C-terminal domain (residues 152–239) of TonB protein, a domain that directly interacts with bacterial outer membrane receptors for import of vitamin B₁₂ and iron-containing compounds [16]. Therefore, it was speculated that the C-terminal domain of MMACHC may also interact with the lysosomal transporter LMBD1, which helped vitamin B₁₂ enter the cytoplasm [17]. Using surface plasmon resonance and bacterial two-hybrid system, it was confirmed that MMACHC can interact with methylmalonic aciduria and homocystinuria type D protein (MMADHC) [18].

Previous studies identified that MMACHC mutations are the cause of cblC deficiency [14,19]. cblC methylmalonic aciduria and homocystinuria caused by MMACHC mutations are the most common congenital defects in vitamin B₁₂ metabolism [20]. That is because the metabolism of vitamin B₁₂ has a significant relationship with the metabolism of homocysteine and folic acid [21]. For example, methionine synthetase requires mecobalamin as a coenzyme to catalyse the conversion of homocysteine from N⁵-methyltetrahydrofolate to methionine. If MMACHC deficiency leads to mecobalamin deficiency, the synthesis of methionine is blocked, and excessive accumulation of homocysteine leads to homocystinuria. More critically, the regeneration of a tetrahydrofolic acid will be greatly affected, and tetrahydrofolic acid is a tool for methyl transport, which is required for the synthesis of both purine and pyrimidine. As a result, the nucleic acid synthesis disorder leads to abnormal cell division, which may eventually lead to megaloblastic anemia, also known as pernicious anemia [22].

There are two distinct cblC type defect phenotypes in terms of age of onset. Early-onset patients present with failure to thrive, megaloblastic anaemia, epilepsy, and neurodegenerative signs; late-onset patients present acute neurological symptoms after 4 years of age including spasticity, delirium, and psychosis. Pigmentary retinopathy with perimacular degeneration is part of the clinical disease manifestations [23,24]. Thus far, at least 60 different mutations in the MMACHC gene have been reported. Besides missense mutations, there are also many causing a putative premature truncation or deletion of the protein. The most common missense mutation is G147D [25,26], which is resulting in vitamin B₁₂ unresponsive disease and early-onset disease. R161Q is the other common mutation that binds CNCbl with lower affinity than wild-type’s and is associated with late-onset disease [25].

The mechanism of action of the decyanation and dealkylation processes of MMACHC is a very interesting question. It has been confirmed that the decyanation and dealkylation processes of MMACHC are carried out through two completely different chemical reactions [15,17]. In the presence of nicotinamide adenine dinucleotide phosphate (NADPH) and flavin adenine dinucleotide (FAD), CNCbl is converted to cyanide and CoB (II) Alamin through a reductant cyanidation process [17]. Alkylcobalamin requires nucleophilic substitution to produce CoB (I) Alamin in the presence of glutathione and then reoxidation to CoB (II) Alamin. At present, it has been confirmed that both MMACHC and C2–C6 alkylcobalamin have the ability to dealkylate [15]. Previous studies also reported that MMACHC was responsible for the early processing of both CNCbl (decyanation) and alkylcobalamins [27]. MMACHC does not belong to any known family of glutathione transferases, suggesting that MMACHC may represent a new class of glutathione transferases.

MMACHC is a 31.7 kDa protein [14], which may interact with LMBD1 and MMADMC and can decyanate and dealkylate in two completely different chemical reactions. Moreover, the substrate range is also quite wide and can help the structure transformation of cobalamin. This led to a great deal of interest in its protein structure. How does MMACHC contain so much functionality in such a small space?

2. Materials and Methods

2.1. Construction of MMACHC Expression Vector

The full-length MMACHC gene (residues 1–282, UniProt ID: Q9Y4U1) was cloned into plasmid pGEX-4T-1, which produces an N-terminally GST-tagged protein with a thrombin digestion site. Then, the MMACHC-pGEX-4T-1 expression plasmid was introduced into E. coli Rosetta(DE3)pLysS to express the recombinant MMACHC.

2.2. Overexpression and Purification of MMACHC

For overexpression of MMACHC, Rosetta(DE3)pLysS cells carrying the MMACHC expression vector were grown in LB medium with ampicillin at 37 °C. Isopropyl-1-thio-b-D-galactopyranoside (1 mM, final concentration) was added to the bacterial cultures (OD600 ≈ 0.8) to induce recombinant protein expression, and the cells were grown for 12 h with shaking at 16 °C. The bacterial pellet was resuspended in buffer A (140 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄, 5% glycerol, pH 7.3). Cells were lysed by passing through a microfluidiser (1250 bar; 1 bar = 100 kPa) three times; then, the lysate was centrifuged at 30,000× g for 30 min. The supernatant was loaded onto a 5 mL GSTrap column (GE Healthcare) that had been equilibrated in buffer A. After 20 column volumes of buffer A washing, GST-tagged MMACHC was eluted by using buffer B (20 mM Tris-HCl pH 8.0, 150 mM NaCl, 10 mM GSH, 5% glycerol). GST-tag was cut by using thrombin at 20 °C for 16 h and removed by GSTrap. MMACHC were loaded onto a Superdex 75 16/60 column that was equilibrated in buffer C (25 mM Tris-HCl pH 8.0, 150 mM NaCl, 5% glycerol). The resulting protein peak was concentrated to 15 mg/mL by using a 10 kDa MWCO Amicon Ultra and fresh TCEP was added to a final concentration of 2 mM. Aliquots were snap-frozen in liquid nitrogen and stored at −80 °C until they were used for crystallisation. Selenomethionine-labeled MMACHC (SeMet- MMACHC) protein was expressed in E. coli BL21 (DE3) cells using the M9 medium protocol. SeMet-MMACHC was expressed and purified using the same method as described for the wild-type MMACHC.

2.3. Crystallisation of MMACHC with CNCbl

Crystals of MMACHC with CNCbl were grown by hanging drop vapor diffusion at 18 °C. Equal volumes of 15 mg/mL protein were mixed with reservoir solution containing 25% PEG2000MME, 0.1 M Bis-Tris pH 6.5, 0.2 M Li₂SO₄.

2.4. Data Collection and Structure Determination

For the data collection, the crystals were soaked in a reservoir solution containing 20% glycerol and flash-cooled in liquid nitrogen. X-ray diffraction data were collected at beamline BL17U1 of Shanghai Synchrotron Radiation Facility (SSRF), People’s Republic of China [28]. The Se-Met data were processed by using the HKL2000 software suite [29], and the initial SAD phase was solved by using SHELXC/D/E and hkl2map. The model was built using ARP/wARP and COOT [30] and then refined with Phenix [31,32,33]. The apo data were processed by using the xia2 software suite [34] and the molecular replacement method was carried out by using PHASER [35]. After molecular replacement, maximum-likelihood-based refinement of the atomic position, and temperature factors were performed with Phenix [31,32,33], and the atomic model was built with the program COOT [30]. Crystallographic statistics for the final models are shown in

Table 1. Data collection and refinement statistics.

	MMACHC with CNCbl
Data collection
Space group	C 1 2 1
a, b, c (Å)	72.97 96.33 47.26
α, β, γ (°)	90.00 111.10 90.00
Wavelength (Å)	0.97915
Resolution (Å) a	55.6–1.93 (2.03–1.93)
CC1/2 (a)	99.8 (99.8)
Unique reflections a	22,121
Rmeas (%) a	4.8 (104)
Mean I/σ (I) a	11.8 (3.2)
Completeness (%) a	96.14 (96.81)
Multiplicity a	2.6 (2.6)
Refinement
Resolution (Å)	33.37–1.93 (2.00–1.93)
Rwork/Rfreeb	0.1602 (0.2020)
No.atoms
Protein	1963
Ligand	108
Water	208
Average B factors (Å2)
Protein	28.02
Ligand	27.23
Water	28.9
RMS deviations
Bond lengths (Å)	0.0163
Bond angles (°)	1.34
Ramachandran plot
Favoured (%)	98.74
Allowed (%)	1.26
Outliers (%)	0

^a The values in parentheses are for the outermost shell. ^b R_free is the R_work based on 5% of the data excluded from the refinement. $R_{\mathrm{meas}}=\sum _{hkl}\sqrt{n/\left(n-1\right)}\sum _{i=1}^n|{\mathrm{I}}_{\mathrm{i}}\left(hkl\right)-\langle \mathrm{I}\left(hkl\right)\rangle |/\sum _{hkl}\sum _i{\mathrm{I}}_{\mathrm{i}}\left(hkl\right)$, where $\langle \mathrm{I}\left(hkl\right)\rangle$is the mean intensity of a set of equivalent reflections. $R_{\mathrm{work}}=\sum _{hkl}\Vert {\mathrm{F}}_{\mathrm{obs}}|-|{\mathrm{F}}_{\mathrm{calc}}\Vert /\sum _{hkl}|{\mathrm{F}}_{\mathrm{obs}}|$ where F_obs and F_calc are observed and calculated structure factors, respectively.

. Figures were prepared with PYMOL (http://www.pymol.org, 16 February 2022) [36].

3. Results

3.1. MMACHC Purification and Gel-Filtration Chromatography

Using MMACHC-pGEX-4T-1 prokaryotic expression vector, MMACHC recombinant protein was successfully expressed in E. coli Rosetta(DE3)pLysS strain. After GST affinity chromatography purification, about 2 mg of MMACHC protein was obtained from 500 mL of culture medium. The MMACHC was then further purified by gel-filtration chromatography (Figure 2A). The apparent molecular weight of the protein was measured by gel-filtration chromatography and was about 45 kDa, which was slightly higher than theoretical molecular weight and SDS–PAGE results. The results of 15% SDS–PAGE showed that the protein purity reached more than 90%.

3.2. Overall Structure of MMACHC with CNCbl

The MMACHC and CNCbl complex crystal (PDB code: 7WUZ) belonged to space group C2, with unit-cell parameters a = 72.97 Å, b = 96.33 Å, c = 47.26 Å, β = 111.10°. Detailed crystallographic data statistics for MMACHC and CNCbl complex structure are shown in Table 1. There were 287 residues in the asymmetric unit, including five residues introduced from the plasmid constructs, of which 241 residues can be matched to the electron density map.. The C terminus of MMACHC (242–282) was not observed (no clear electron density). The refined structure contained 2315 non-H protein atoms, 208 water molecules, 1 CNCbl molecule, 1 sulphate ion, and 1 tartrate ion (Figure 3A,B). Overall, 98.74% of all residues were located in the favoured region of the Ramachandran diagram. The R_work and R_free factors are 16.02% and 20.20%, respectively.

To investigate the conservation of MMACHC in eukaryotes, we extracted amino acids of MMACHC derived from different eukaryotes for multiple sequence alignment, which showed that MMACHC is relatively conserved in different eukaryotes, with three highly conserved amino acids (Figure 3C).

3.3. CNCbl Occupies the Ligand-Binding Site of MMACHC

To better determine the interaction of MMACHC with CNCbl and the presence of the MMACHC ligand-binding region, electron density maps of the ligand-binding sites were extracted from the overall electron density maps. Remarkably, a strip-shaped electron density map of CNCbl was found in the ligand-binding sites (consistent with the previous MMACHC and AdoCbl complex) (Figure 4), which also demonstrates the interaction of MMACHC with CNCbl and the presence of the MMACHC ligand-binding region.

3.4. Structural Comparison of MMACHC Combined with CNCbl and AdoCbl

To better clarify the difference between human MMACHC with CNCbl and AdoCbl (PDB code: 3SOM) [37], the structure of them was superposed (rmsd: 0.313, TM score: 0.6443). Comparing the two structures revealed that adenine was replaced by cyano in the MMACHC-CNCbl structure, and this cyano formed a certain spatial blockage that induced a certain shift in MMACHC (Figure 5A). In order to verify the role of the cyano group, the structures of MMACHC with CNCbl and AdoCbl were further compared, and the results (Figure 5B,C) showed that the presence of the cyano group contributed to the instability of the α-helix and β-sheet of MMACHC (the part labelled by the rectangular box).

4. Conclusions and Discussion

Methylmalonic aciduria combined with homocystinuria is a group of haematological and neurological disorders caused by congenital defects in the metabolism of vitamin B₁₂. The gene responsible for the cblC type of defect has been named methylmalonic aciduria type C and homocystinuria (MMACHC) [14]. The cblC type C methylmalonic aciduria combined with homocystinuria caused by the MMACHC mutations are the most common congenital defect in vitamin B₁₂ metabolism and have the highest incidence.

Here, we well explained the substrate diversity of MMACHC and the different severity of symptoms caused by different mutation sites of MMACHC by resolving the structure of the MMACHC complex with CNCbl.

Firstly, there is a large space at the site where the cobalamin R group is located, large enough to hold a variety of different groups, such as methyl, hydroxyl, and adenosine, providing a structural basis for the substrate diversity of MMACHC.

Secondly, the severity of symptoms caused by different mutation sites in MMACHC varied. For example, patients with G147D mutation usually develop before one year of age, with symptoms such as dysplasia, seizures, lethargy and neurodegeneration, and even blood and eye abnormalities. This is because G147 is located in the region where MMACHC binds CNCbl, and the mutation disables MMACHC’s ability to bind CNCbl (Figure 6), so patients experience distress shortly after birth. In contrast, patients with R161Q mutation, which usually develops after the age of one and has symptoms such as dementia, behavioural problems, and myelopathy, have R161 located close to the location of the cyano and further away from the region that binds CNCbl, resulting in MMACHC-R161Q mutation having only one-fifth the ability to bind CNCbl of the wild type (Figure 6).

The structures shown by these two groups are quite consistent with the structures resolved in this study. The determination of the crystal structure of the MMACHC and cyanocobalamin complexes contributes to a better understanding of the metabolic pathways of vitamin B₁₂.