The CBS subdomain of inosine 5'-monophosphate dehydrogenase regulates purine nucleotide turnover

Abstract

Inosine 5'-monophosphate dehydrogenase (IMPDH) catalyses the rate-limiting step in guanine nucleotide biosynthesis. IMPDH has an evolutionary conserved CBS subdomain of unknown function. The subdomain can be deleted without impairing the in vitro IMPDH catalytic activity and is the site for mutations associated with human retinitis pigmentosa. A guanine-prototrophic Escherichia coli strain, MP101, was constructed with the subdomain sequence deleted from the chromosomal gene for IMPDH. The ATP content was substantially elevated in MP101 whereas the GTP content was slighty reduced. The activities of IMPDH, adenylosuccinate synthetase and GMP reductase were two to threefold lower in MP101 crude extracts compared with the BW25113 wild-type strain. Guanine induced a threefold reduction in the MP101 ATP pool and a fourfold increase in the GTP pool within 10 min of addition to growing cells; this response does not result from the reduced IMPDH activity or starvation for guanylates. In vivo kinetic analysis using 14-C tracers and 33-P pulse-chasing revealed mutation-associated changes in purine nucleotide fluxes and turnover rates. We conclude that the CBS subdomain of IMPDH may coordinate the activities of the enzymes of purine nucleotide metabolism and is essential for maintaining the normal ATP and GTP pool sizes in E. coli.

Fig. 1. The de novo and salvage pathways for purine nucleotide biosynthesis

A, adenine; Hx, hypoxanthine; G, guanine; X, xanthine; AR, adenosine; HxR, inosine; GR, guanosine; AMPs, adenylosuccinate. purA, adenylosuccinate synthetase (AMPsS); purB, adenylosuccinate lyase; guaB, IMP dehydrogenase (IMPDH); guaA, GMP synthetase (GMPS); deoD, purine nucleoside phosphorylase; hpt, guanine/hypoxanthine phosphoribosyltransferase; gpt, guanine phosphoribosyltransferase; gsk, guanosine/inosine kinase; apt, adenine phosphoribosyltransferase; add, adenosine deaminase; guaC, GMP reductase (GMPR); purF, amidophosphoribosyl transferase; amn, AMP nucleosidase. Shown in red and green arrows are reactions that may have a lower or a higher flux, respectively, in the conditions of a high GTP pool in the MP101 mutant. Dashed lines reflect feedback regulation.

Fig. 2. The effect of guaB^ΔCBS mutation on nucleotide pools

A, The structure of Streptococcus pyogenes IMPDH tetramer (PDB code 1ZFJ); the catalytic domains are shown in yellow, blue, orange and green; the deleted CBS domain is shown in each monomer in purple with respect to the sequence homology with E. coli GuaB. The positioning of the CBS pair within the IMPDH sequence is shown as a schematic bar representation (a full ClustalW alignment of E. coli BW25113 guaB⁺, E. coli MP101 guaB^ΔCBS and S. pyogenes IMPDH proteins is presented in Fig. S1). B, Nucleotide pools in strains grown on minimal media. The average of 5 independent measurements is given. Error bars are standard deviations. C, Purine-induced nucleotide pool responses in BW25113 guaB⁺ and MP101 guaB^ΔCBS. 30 μg/ml of the indicated purine base was added at 0 min. Vertical axis, nucleotide pool (mM); horizontal axis, time (min).

Fig. 3. Effect of IMPDH^ΔCBS overexpression on MP101 guaB^ΔCBS purine nucleotide pools

A. Constitutive overexpression of IMPDH^ΔCBS in MP101 guaB^ΔCBS. MP101 cells containing the pGUA22 plasmid were grown to a mid-exponential phase in MOPS media supplemented with 100 μg/ml carbenicillin and harvested by centrifugation. BugBuster reagent (Novagen) containing 30 μg/ml PMSF was used for protein extraction (1 ml per 100 OU₆₀₀). The extract was cleared by centrifugation and IMPDH activity per mg of protein was measured using the ¹⁴C-IMP-based assay as described in the Experimental Procedures section. The asterisk indicates the number of independent measurements. B. ATP and GTP pools of minimal-media grown MP101 cells containing the pGUA22 plasmid for IMPDH^ΔCBS overexpression versus the pCR2.1-TOPO-E control vector. C. Guanine-induced purine nucleotide pool response in minimal media-grown MP101 guaB^ΔCBS cells overexpressing IMPDH^ΔCBS. Vertical axis, nucleotide pool (mM); horizontal axis, time from addition of 30 μg/ml guanine (min).

Fig. 4. The effect of adenylate and guanylate nucleotides on IMPDH activity in E. coli BW25113 guaB⁺ extract

The concentrations of nucleotide additives were chosen to be 2 – 5 times greater than physiological with the exception of GMP which was used at a concentration at least 10 times greater than physiological (Neidhardt et al., 1987). The presence in the reaction of 4 mM Mg(CH₃COO)₂ is indicated with a plus sign. The experiment was conducted in 4 replicates. Error bars are standard deviations.

Fig. 5. Purine nucleotide pools in BW25113 guaB⁺ and MP101 guaB^ΔCBS during steady-state growth on minimal media supplemented with indicated purine bases

The cells were cultured in the presence of a purine base and ³³P_i for at least 2 hrs before nucleotide pool measurement. A. Nucleotide pools during purine-supplemented steady state growth. B. Guanine-induced purine nucleotide pool responses in MP101 guaB^ΔCBS cells grown in the presence of adenine, hypoxanthine or xanthine. 30 μg/ml guanine was added at 0 min to MP101 cells growing on MOPS minimal media supplemented with 10 μg/ml of the indicated compounds.

Fig. 6. The kinetics of ¹⁴C-purine labeling

Approximate initial rates were obtained from linear fits to the first three data points, as shown A, Kinetics of guanine salvage. [8-¹⁴C]-guanine (57 mCi/mmol) was added to the cell culture media to a final concentration of 32 μM. B, Salvage of hypoxanthine and IMP partitioning. 1.36 μM [¹⁴C(U)]-hypoxanthine (291 mCi/mmol) was added to the media simultaneously with 159 μM (24 μg/ml) unlabeled guanine. Blue symbols and lines, BW25113 guaB⁺; orange symbols and lines, MP101 guaB^ΔCBS. “●”, solid lines, adenylate nucleotides (ATP + ADP + AMP); “▲”, dashed lines, guanylate nucleotides (GTP + GDP + GMP); “□”, the sum of adenylate nucleotides and guanylate nucleotides; “■”, IMP; “♦”, adenylate/guanylate ratio.

Fig. 7. Turnover rates of ATP and GTP phosphates

A, no purine supplementation; B, 30 μg/ml guanine added simultaneously with excess unlabeled phosphate (see Materials and Methods). With the exception of the curve for MP101 GTP in the presence of guanine, the curves are single-exponential fits with offset.