The Helicobacter pylori ureC gene codes for a phosphoglucosamine mutase.

The function of UreC, the product of a 1,335-bp-long open reading frame upstream from the urease structural genes (ureAB) of Helicobacter pylori, was investigated. We present data showing that the ureC gene product is a phosphoglucosamine mutase. D. Mengin-Lecreulx and J. van Heijenoort (J. Biol. Chem. 271:32-39, 1996) observed that UreC is similar (43% identity) to the GlmM protein of Escherichia coli. Those authors showed that GlmM is a phosphoglucosamine mutase catalyzing interconversion of glucosamine-6-phosphate into glucosamine-1-phosphate, which is subsequently transformed into UDP-N-acetylglucosamine. The latter product is one of the main cytoplasmic precursors of cell wall peptidoglycan and outer membrane lipopolysaccharides. The present paper reports that, like its E. coli homolog glmM, the H. pylori ureC gene is essential for cell growth. It was known that growth of a lethal conditional glmM mutant of E. coli at a nonpermissive temperature can be restored in the presence of the ureC gene. We showed that complete complementation of the glmM mutant can be obtained with a plasmid overproducing UreC. The peptidoglycan content and the specific phosphoglucosamine mutase activity of such a complemented strain were measured; these results demonstrated that the ureC gene product functions as a phosphoglucosamine mutase. Homologs of the UreC and GlmM proteins were identified in Haemophilus influenzae, Mycobacterium leprae, Clostridium perfringens, Synechocystis sp. strain PCC6803, and Methanococcus jannaschii. Significant conservation of the amino acid sequence of these proteins in such diverse organisms suggests a very ancient common ancestor for the genes and defines a consensus motif for the phosphoglucosamine mutase active site. We propose renaming the H. pylori ureC gene the glmM gene.     

Reaction mechanism of phosphoglucosamine mutase from Escherichia coli.

The phosphoglucosamine mutase (GlmM) from Escherichia coli, specifically required for the interconversion of glucosamine-6-phosphate and glucosamine-1-phosphate (an essential step in the pathway for cell-wall peptidoglycan and lipopolysaccharide biosyntheses) was purified to homogeneity and its kinetic properties were investigated. The enzyme was active in a phosphorylated form and catalysed its reaction according to a classical ping-pong bi-bi mechanism. The dephosphorylated and phosphorylated forms of GlmM could be separated by HPLC and coupled MS showed that only one phosphate was covalently linked to the active site of the enzyme. The site of phosphorylation was clearly identified as Ser102 in the 445-amino acid polypeptide. GlmM was also capable of catalysing the interconversion of glucose-1-phosphate and glucose-6-phosphate isomers, although at a much lower (1400-fold) rate. Interestingly, the mutational change of the Ser100 to a threonine residue resulted in a 20-fold increase of the nonspecific phosphoglucomutase activity of GlmM, suggesting that the presence of either a serine or a threonine at this position in the consensus sequence of hexosephosphate mutases could be one of the factors that determines the specificity of these enzymes for either sugar-phosphate or amino sugar-phosphate substrates.     

Identification of the Pseudomonas aeruginosa glmM gene, encoding phosphoglucosamine mutase

A search for a potential algC homologue within the Pseudomonas aeruginosa PAO1 genome database has revealed an open reading frame (ORF) of unknown function, ORF540 in contig 54 (July 1999 Pseudomonas genome release), that theoretically coded for a 445-amino-acid-residue polypeptide (I. M. Tavares, J. H. Leit?o, A. M. Fialho, and I. Sá-Correia, Res. Microbiol. 150:105-116, 1999). The product of this gene is here identified as the phosphoglucosamine mutase (GlmM) which catalyzes the conversion of glucosamine-6-phosphate to glucosamine-1-phosphate, an essential step in the formation of the cell wall precursor UDP-N-acetylglucosamine. The P. aeruginosa gene has been cloned into expression vectors and shown to restore normal peptidoglycan biosynthesis and cell growth of a glmM Escherichia coli mutant strain. The GlmM enzyme from P. aeruginosa has been overproduced to high levels and purified to homogeneity in a six-histidine-tagged form. Beside its phosphoglucosamine mutase activity, the P. aeruginosa enzyme is shown to exhibit phosphomannomutase and phosphoglucomutase activities, which represent about 20 and 2% of its GlmM activity, respectively.     

Identification of M. tuberculosis Rv3441c and M. smegmatis MSMEG_1556 and Essentiality of M. smegmatis MSMEG_1556

The normal growth of mycobacteria attributes to the integrity of cell wall core which consists of peptidoglycan (PG), arabinogalactan (AG) and mycolic acids. N-acetyl glucosamine (GlcNAc) is an essential component in both PG and AG of mycobacterial cell wall. The biosynthetic pathway for UDP-N-acetylglucosamine (UDP-GlcNAc), as a sugar donor of GlcNAc, is different in prokaryotes and eukaryotes. The conversion of glucosamine-6-phosphate to glucosamine-1-phosphate, which is catalyzed by phosphoglucosamine mutase (GlmM), is unique to prokaryotes. Bioinformatic analysis showed that Msm MSMEG_1556 and Mtb Rv3441c are homologous to Ec GlmM. In this study, soluble Msm MSMEG_1556 protein and Mtb Rv3441c protein were expressed in E. coli BL21(DE3) and their phosphoglucosamine mutase activity were detected. In order to further investigate the essentiality of MSMEG_1556 for the growth of M. smegmatis, we generated a conditional MSMEG_1556 knockout mutant, which harbored thermo-sensitive rescue plasmid carrying Mtb Rv3441c. As the rescue plasmid was unable to complement MSMEG_1556 deficiency at 42°C, MSMEG_1556 knockout mutant did not grow. The dramatic morphological changes of MSMEG_1556 knockout mutant after temperature shift from 30°C to 42°C have been observed by scanning electron microscope. These results demonstrated that MSMEG_1556 is essential for growth of M. smegmatis. This study provided evidence that GlmM enzyme could be as a potential target for developing anti-tuberculosis drugs.     

Identification of the Streptococcus gordonii glmM gene encoding phosphoglucosamine mutase and its role in bacterial cell morphology, biofilm formation, and sensitivity to antibiotics

Phosphoglucosamine mutase (EC 5.4.2.10) catalyzes the interconversion of glucosamine-6-phosphate into glucosamine-1-phosphate, an essential step in the biosynthetic pathway leading to the formation of peptidoglycan precursor uridine 5'-diphospho-N-acetylglucosamine. The gene (glmM) of Escherichia coli encoding the enzyme has been identified previously. We have now identified a glmM homolog in Streptococcus gordonii, an early colonizer on the human tooth and an important cause of infective endocarditis, and have confirmed that the gene encodes phosphoglucosamine mutase by assaying the enzymatic activity of the recombinant GlmM protein. Insertional glmM mutant of S. gordonii did not produce GlmM, and had a growth rate that was approximately half that of the wild type. Morphological analyses clearly indicated that the glmM mutation causes marked elongation of the streptococcal chains, enlargement of bacterial cells, and increased roughness of the bacterial cell surface. Furthermore, the glmM mutation reduces biofilm formation and increases sensitivity to penicillins relative to wild type. All of these phenotypic changes were also observed in a glmM deletion mutant, and were restored by the complementation with plasmid-borne glmM. These results suggest that, in S. gordonii, mutations in glmM appear to influence bacterial cell growth and morphology, biofilm formation, and sensitivity to penicillins.     

Copurification of glucosamine-1-phosphate acetyltransferase and N-acetylglucosamine-1-phosphate uridyltransferase activities of Escherichia coli: characterization of the glmU gene product as a bifunctional enzyme catalyzing two subsequent steps in the pathway for UDP-N-acetylglucosamine synthesis.

The glmU gene product of Escherichia coli was recently identified as the N-acetylglucosamine-1-phosphate uridyltransferase activity which catalyzes the formation of UDP-N-acetylglucosamine, an essential precursor for cell wall peptidoglycan and lipopolysaccharide biosyntheses (D. Mengin-Lecreulx and J. van Heijenoort, J. Bacteriol. 175:6150-6157, 1993). Evidence that the purified GlmU protein is in fact a bifunctional enzyme which also catalyzes acetylation of glucosamine-1-phosphate, the preceding step in the same pathway, is now provided. Kinetic parameters of both reactions were investigated, indicating in particular that the acetyltransferase activity of the enzyme is fivefold higher than its uridyltransferase activity. In contrast to the uridyltransferase activity, which is quite stable and insensitive to thiol reagents, the acetyltransferase activity was rapidly lost when the enzyme was stored in the absence of reducing thiols or acetyl coenzyme A or was treated with thiol-alkylating agents, suggesting the presence of at least one essential cysteine residue in or near the active site. The acetyltransferase activity is greatly inhibited by its reaction product N-acetylglucosamine-1-phosphate and, interestingly, also by UDP-N-acetylmuramic acid, which is one of the first precursors specific for the peptidoglycan pathway. The detection in crude cell extracts of a phosphoglucosamine mutase activity finally confirms that the route from glucosamine-6-phosphate to UDP-N-acetylglucosamine occurs via glucosamine-1-phosphate in bacteria.     

Substitution of aspartate and glutamate for active center histidines in the Escherichia coli phosphoenolpyruvate:sugar phosphotransferase system maintain phosphotransfer potential

The active center histidines of the Escherichia coli phosphoenolpyruvate:sugar phosphotransferase system proteins; histidine-containing protein, enzyme I, and enzyme IIA(Glc) were substituted with a series of amino acids (serine, threonine, tyrosine, cysteine, aspartate, and glutamate) with the potential to undergo phosphorylation. The mutants [H189E]enzyme I, [H15D]HPr, and [H90E]enzyme IIA(Glc) retained ability for phosphorylation as indicated by [(32)P]phosphoenolpyruvate labeling. As the active center histidines of both enzyme I and enzyme IIA(Glc) undergo phosphorylation of the N(epsilon2) atom, while HPr is phosphorylated at the N(delta1) atom, a pattern of successful substitution of glutamates for N(epsilon2) phosphorylations and aspartates for N(delta1) phosphorylations emerges. Furthermore, phosphotransfer between acyl residues: P-aspartyl to glutamyl and P-glutamyl to aspartyl was demonstrated with these mutant proteins and enzymes.     

Functional cloning and mutational analysis of the human cDNA for phosphoacetylglucosamine mutase: identification of the amino acid residues essential for the catalysis

In Saccharomyces cerevisiae, phosphoacetylglucosamine mutase is encoded by an essential gene called AGM1. The human AGM1 cDNA (HsAGM1) and the Candida albicans AGM1 gene (CaAGM1) were functionally cloned and characterized by using an S. cerevisiae strain in which the endogenous phosphoacetylglucosamine mutase was depleted. When expressed in Escherichia coli as fusion proteins with glutathione S-transferase, both HsAgm1 and CaAgm1 proteins displayed phosphoacetylglucosamine mutase activities, demonstrating that they indeed specify phosphoacetylglucosamine mutase. Sequence comparison of HsAgm1p with several hexose-phosphate mutases yielded three domains that are highly conserved among phosphoacetylglucosamine mutases and phosphoglucomutases of divergent organisms. Mutations of the conserved amino acids found in these domains, which were designated region I, II, and III, respectively, demonstrated that alanine substitutions for Ser(64) and His(65) in region I, and for Asp(276), Asp(278), and Arg(281) in region II of HsAgm1p severely diminished the enzyme activity and the ability to rescue the S. cerevisiae agm1Delta null mutant. Conservative mutations of His(65) and Asp(276) restored detectable activities, whereas those of Ser(64), Asp(278), and Arg(281) did not. These results indicate that Ser(64), Asp(278), and Arg(281) of HsAgm1p are residues essential for the catalysis. Because Ser(64) corresponds to the phosphorylating serine in the E. coli phosphoglucosamine mutase, it is likely that the activation of HsAgm1p also requires phosphorylation on Ser(64). Furthermore, alanine substitution for Arg(496) in region III significantly increased the K(m) value for N-acetylglucosamine-6-phosphate, demonstrating that Arg(496) serves as a binding site for N-acetylglucosamine-6-phosphate.     

Contribution of phosphoglucosamine mutase to the resistance of Streptococcus gordonii DL1 to polymorphonuclear leukocyte killing

Phosphoglucosamine mutase (GlmM; EC 5.4.2.10) catalyzes the interconversion of glucosamine-6-phosphate to glucosamine-1-phosphate, an essential step in the biosynthetic pathway leading to the formation of the peptidoglycan precursor uridine 5'-diphospho- N -acetylglucosamine. We have recently identified the gene ( glmM ) encoding the enzyme of Streptococcus gordonii , an early colonizer on the human tooth and an important cause of infective endocarditis, and indicated that the glmM mutation in S. gordonii appears to influence bacterial cell growth, morphology, and sensitivity to penicillins. In the present study, we assessed whether the glmM mutation also affects escape from polymorphonuclear leukocyte (PMN)-dependent killing. Although no differences in attachment to human PMNs were observed between the glmM mutant and the wild-type S. gordonii , the glmM mutation resulted in increased sensitivity to PMN-dependent killing. Compared with the wild type, the glmM mutant induced increased superoxide anion production and lysozyme release by PMNs. Moreover, the glmM mutant is more sensitive to lysozyme, indicating that the GlmM may be required for synthesis of firm peptidoglycans for resistance to bacterial cell lysis. These findings suggest that the GlmM contributes to the resistance of S. gordonii to PMN-dependent killing. Enzymes such as GlmM could be novel drug targets for this organism.     

Phosphoglycerates and Protein Phosphorylation: Identification of a Protein Substrate as Glucose-1,6-Bisphosphate Synthetase

We have previously reported the occurrence of two endogenous protein phosphorylation systems in mammalian brain that are enhanced in the presence of 3-phosphoglycerate (3PG) and ATP. We present here a study of one of these systems, the phosphorylation of the 72-kDa protein (3PG-PP72). This system was separated into the substrate, 3PG-PP72, and a kinase by ammonium sulfate fractionation, hydroxyapatite chromatography, and hydrophobic interaction HPLC. The substrate protein was shown to be directly phosphorylated with [1-32P]1,3-bisphosphoglycerate [( 1-32P]1,3BPG) with an apparent Km of 1.1 nM. Nonradioactive 1,3BPG inhibited 32P incorporation in the presence of [gamma-32P]ATP and 3PG. Phosphopeptide mapping and phosphoamino acid analyses indicated that the site of phosphorylation of 3PG-PP72 observed in the presence of 3PG and ATP is a serine residue identical to that observed with [1-32P]1,3BPG. Moreover, [32P]phosphate incorporated into 3PG-PP72 in the presence of 3PG and ATP was removed by subsequent incubation with glucose-1-phosphate or glucose-6-phosphate. Finally, 3PG-PP72 showed chromatographic behaviors identical to those of glucose-1,6-bisphosphate (G1,6P2) synthetase. Based upon these observations, we conclude that 3PG-PP72 is G1,6P2 synthetase and that it is phosphorylated directly by 1,3BPG, which is formed from 3PG and ATP by 3PG kinase present in a crude 3PG-PP72 preparation.     

Effect of Phosphoglucosamine Mutase on Biofilm Formation and Antimicrobial Susceptibilities in M. smegmatis glmM Gene Knockdown Strain

UDP-N-acetylglucosamine (UDP-GlcNAc) is a direct glycosyl donor of linker unit (L-Rhamnose-D-GlcNAc) and an essential precursor of peptidoglycan in mycobacteria. Phosphoglucosamine mutase (GlmM) is involved in the formation of glucosamine-1-phosphate from glucosamine-6-phosphate, the second step in UDP-GlcNAc biosynthetic pathway. We have demonstrated that GlmM protein is essential for the growth of M. smegmatis. To facilitate the analysis of the GlmM protein function in mycobacteria, a tetracycline inducible M. smegmatis glmM gene knockdown strain was constructed by using an antisense RNA technology. After induction with 20 ng/ml tetracycline, the expression of GlmM protein in glmM gene knockdown strain was significantly decreased, resulting in a decline of cell growth. The morphological changes of glmM gene knockdown strain induced with 20 ng/ml tetracycline have been observed by scanning electron microscope and transmission electron microscope. Furthermore, insufficient GlmM protein reduced the biofilm formation and increased the sensitivity to isoniazid and ethambutol in M. smegmatis, indicating that GlmM protein had effect on the biofilm formation and the senstivity to some anti-tuberculosis drugs targeting the cell wall. These results provide a new insight on GlmM functions in mycobacteria, suggesting that GlmM could be a potential target for development of new anti-tuberculosis drug.     

Phosphoprotein with phosphoglycerate mutase activity from the archaeon Sulfolobus solfataricus

When soluble extracts of the extreme acidothermophilic archaeon Sulfolobus solfataricus were incubated with [gamma-(32)P]ATP, several proteins were radiolabeled. One of the more prominent of these, which migrated with a mass of approximately 46 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), was purified by column chromatography and SDS-PAGE and subjected to amino acid sequence analysis via both the Edman technique and mass spectroscopy. The best match to the partial sequence obtained was the potential polypeptide product of open reading frame sso0417, whose DNA-derived amino acid sequence displayed many features reminiscent of the 2,3-diphosphoglycerate-independent phosphoglycerate (PGA) mutases [iPGMs]. Open reading frame sso0417 was therefore cloned, and its protein product was expressed in Escherichia coli. Assays of its catalytic capabilities revealed that the protein was a moderately effective PGA mutase that also exhibited low levels of phosphohydrolase activity. PGA mutase activity was dependent upon the presence of divalent metal ions such as Co(2+) or Mn(2+). The recombinant protein underwent autophosphorylation when incubated with either [gamma-(32)P]ATP or [gamma-(32)P]GTP. The site of phosphorylation was identified as Ser(59), which corresponds to the catalytically essential serine residue in bacterial and eucaryal iPGMs. The phosphoenzyme intermediate behaved in a chemically and kinetically competent manner. Incubation of the (32)P-labeled phosphoenzyme with 3-PGA resulted in the disappearance of radioactive phosphate and the concomitant appearance of (32)P-labeled PGA at rates comparable to those measured in steady-state assays of PGA mutase activity.     

Functional homology of 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase, phosphoglycerate mutase, and 2,3-bisphosphoglycerate mutase

The bisphosphatase domain of the rat liver 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase has been shown to exhibit a structural similarity to yeast phosphoglycerate mutase and human red blood cell 2,3-bisphosphoglycerate mutase including very similar active site sequences with a histidyl residue being involved in phospho group transfer. The liver bifunctional enzyme was found to catalyze the hydrolysis of glycerate 1,3-bisphosphate to glycerate 3-phosphate and inorganic phosphate. The Km for glycerate 1,3-bisphosphate was 320 microM and the Vmax was 11.5 milliunits/mg. Incubation of the rat liver enzyme with [1-32P]glycerate 1,3-bisphosphate resulted in the formation of a phosphoenzyme intermediate, and the labeled amino acid was identified as 3-phosphohistidine. Tryptic and endoproteinase Lys-C peptide maps of the 32P-phosphoenzyme labeled either with [2-32P]fructose 2,6-bisphosphate or [1-32P]glycerate 1,3-bisphosphate revealed that 32P-radioactivity was found in the same peptide, proving that the same histidyl group accepts phosphate from both substrates. Fructose 2,6-bisphosphate inhibited competitively the formation of phosphoenzyme from [1-32P]glycerate 1,3-bisphosphate. Effectors of fructose-2,6-bisphosphatase also inhibited phosphoenzyme formation. Substrates and products of phosphoglycerate mutase and 2,3-bisphosphoglycerate mutase also modulated the activities of the bifunctional enzyme. These results demonstrate that, in addition to a structural homology, the bisphosphatase domain of the bifunctional enzyme has a functional similarity to phosphoglycerate mutase and 2,3-bisphosphoglycerate mutase and support the concept of an evolutionary relationship between the three enzyme activities.     

Phosphorylation of glucokinase from rat liver in vitro by protein kinase A with a concomitant decrease of its activity

Glucokinase, purified from rat liver, was phosphorylated to an extent of 1 mol [32P]-phosphate/mol of enzyme when incubated with [32P]ATP and protein kinase A from pig or rabbit muscle. The phosphate was bound to serine residues. K0.5 increased and Vmax decreased upon phosphorylation. The phosphate group was removed during incubation of the phosphorylated glucokinase with alkaline phosphatase. Enzymatically inactive glucokinase was not phosphorylated by the protein kinase.     

Bovine brain Na+, K+-stimulated ATP phosphohydrolase studied by a rapid-mixing technique. Detection of a transient [32P]phosphoenzyme formed in the presence of potassium ions.

1. Conditions for binding of [gamma-32P]ATP to bovine brain Na+,K+-stimulated ATPase were investigated by the indirect technique of measuring the initial rate of 32P-labelling of the active site of the enzyme. 2. At 100 muM [gamma-32P]ATP in the presence of 3 mM MgCl2, approximately the same very high rate of formation of [32P]phosphoenzyme was obtained irrespective of whether [gamma-32P]ATP was added to the enzyme simultaneously with, or 70 ms in advance of the addition of NaCl. A comparatively slow rate of phosphorylation was obtained at 5 muM[gamma-32P]ATP without preincubation. However, on preincubation of the enzyme with 5 muM[gamma-32P]ATP a rate of formation of [32P]phosphoenzyme almost as rapid as at 100 muM[gamma-32P]ATP was observed. 3. A transient [32P]phosphoenzyme was discovered. It appeared in the presence of K+, under conditions which allowed extensive binding of [gamma-32P]-ATP. The amount of [gamma-32P]ATP that could be bound to the enzyme seemed to equal the amount of [32P] phosphorylatable sites. 4. The formation of the transient [32P] phosphoenzyme was inhibited by ADP. The transient [32P] phosphoenzyme was concluded mainly to represent the K+-insensitive and ADP-sensitive E1-32P. 5. When KCl was present in the enzyme solution before the addition of NaCl only a comparatively slow rate of phosphorylation was observed. On preincubation of the enzyme with [gamma-32]ATP an increase in the rate of formation of [32P] phosphoenzyme was obtained, but there was no transient [32P]-phosphoenzyme. The transient [32P]phosphoenzyme was, however, detected when the enzyme solution contained NaCl in addition to KCl and the phosphorylation was started by the addition of [gamma-32P]ATP.     

Determination of the roles of active sites in F1-ATPase by controlled affinity labeling

The affinity reagents 3'-O-(5-fluoro-2,4-dinitrophenyl) [alpha-32P]ATP (FDNP-[alpha-32P]ATP) and 3'-O-(5-fluoro-2,4-dinitrophenyl) [8-14C]ATP (FDNP-[14C]ATP) were synthesized and used to characterize the structure and function of the three active sites in F1-ATPase. FDNP-[alpha-32P]ATP was found to bind covalently to F1 up to two DNP-[alpha-32P]ATP labels per F1 in the absence of Mg2+ without decreasing the ATPase activity. However, when MgCl2 was subsequently added to the reaction mixture, the enzyme could be further labeled with concomitant decrease in ATPase activity that is consistent with the complete inactivation of one enzyme molecule by an affinity label at the third ATP-binding site. Partial hydrolysis of the FDNP-[14C]ATP-labeled enzyme and sequencing of the isolated peptide indicated that the affinity label was attached to Lys-beta 301 at all three active sites. Samples of F1 with covalent affinity label on Lys-beta 301 were also used to reconstitute F1-deficient submitochondrial particles. The reconstituted particles were assayed for ATPase and oxidative phosphorylation activities. These results show that the catalytic hydrolysis of ATP either by F1 in solution or by F0F1 complex attached to inner mitochondrial membrane takes place essentially at only one active site, but is promoted by the binding of ATP at the other two active sites, and that ATP synthesis during oxidative phosphorylation takes place at all three active sites [corrected].     

Two mammalian glucosamine-6-phosphate deaminases: a structural and genetic study

Glucosamine-6-phosphate deaminase (EC 3.5.99.6) is an allosteric enzyme that catalyzes the reversible conversion of D-glucosamine-6-phosphate into D-fructose-6-phosphate and ammonium. Here we describe the existence of two mammalian glucosamine-6-phosphate deaminase enzymes. We present the crystallographic structure of one of them, the long human glucosamine-6-phosphate deaminase, at 1.75 A resolution. Crystals belong to the space group P2(1)2(1)2(1) and present a whole hexamer in the asymmetric unit. The active-site lid (residues 162-182) presented significant structural differences among monomers. Interestingly the region with the largest differences, when compared with the Escherichia coli homologue, was found to be close to the active site. These structural differences can be related to the kinetic and allosteric properties of both mammalian enzymes.     

The phosphorylation of subunits of complex I from bovine heart mitochondria

In bovine heart mitochondria and in submitochondrial particles, membrane-associated proteins with apparent molecular masses of 18 and 10 kDa become strongly radiolabeled by [(32)P]ATP in a cAMP-dependent manner. The 18-kDa phosphorylated protein is subunit ESSS from complex I and not as previously reported the 18 k subunit (with the N-terminal sequence AQDQ). The phosphorylated residue in subunit ESSS is serine 20. In the 10 kDa band, the complex I subunit MWFE was phosphorylated on serine 55. In the presence of protein kinase A and cAMP, the same subunits of purified complex I were phosphorylated by [(32)P]ATP at the same sites. Subunits ESSS and MWFE both contribute to the membrane arm of complex I. Each has a single hydrophobic region probably folded into a membrane spanning alpha-helix. It is likely that the phosphorylation site of subunit ESSS lies in the mitochondrial matrix and that the site in subunit MWFE is in the intermembrane space. Subunit ESSS has no known role, but subunit MWFE is required for assembly into complex I of seven hydrophobic subunits encoded in the mitochondrial genome. The possible effects of phosphorylation of these subunits on the activity and/or the assembly of complex I remain to be explored.