Identification of 19 protein S gene mutations in patients with phenotypic protein S deficiency and thrombosis. Protein S Study Group

Protein S is a protein C-dependent and independent inhibitor of the coagulation cascade. Deficiency of protein S is an established risk factor for venous thromboembolism. We have used a strategy of specific amplification of the coding regions and intron/exon boundaries of the active protein S gene (PROS1) and direct single-strand solid phase sequencing, to seek mutations in 35 individuals with phenotypic protein S deficiency. Nineteen point mutations (16 novel) in 19 probands (or relatives of probands) with venous thromboembolism are reported here. Fifteen of the 19 mutations were expected to be causal and included 10 missense mutations (Lys9Glu, Glu26Ala, Gly54Glu, Cys145Tyr, Cys200Ser, Ser283Pro, Gly340Asp, Cys408Ser, Ser460Pro, and Cys625Arg). Three of the 15 mutations resulted in premature stop codons (delete T 635 producing a stop codon at position 126, Lys368stop and Tyr595stop) and two were at intron/exon boundaries (+1 G to A in intron d and +3 A to C in intron j). Of the remaining four mutations, three were within intronic sequence and one was a silent mutation within the coding region and did not alter amino acid composition. In two of the 10 missense mutations, reduced plasma protein S activity compared with antigen level suggested the presence of variant (type II) protein S.     

His1069Gln and six novel Wilson disease mutations: analysis of relevance for early diagnosis and phenotype

In the present study we examined 33 German and 10 Cuban unrelated Wilson disease (WND) index patients and their relatives. The common His1069Gln mutation accounted for 42% of all WND chromosomes in the German series and the haplotype C was found to be highly predictive for this mutation. Six WND gene mutations have not been described previously and involved a splice site at intron 18 (3903 + del1G), a termination codon in the copper-binding region of exon 2 (Cys271X), and missense mutations in transmembrane region 2 (Gly710Ala), in transmembrane region 3 (Tyr741Cys), in the DKTGT motif (Thr1031Ile) and in the ATP loop region (Gly1176Arg). In 15 German WND index patients and three sibs both WND mutations could be determined and a genotype-phenotype correlation was attempted. Patients homozygous for the His1069Gln mutation showed almost the complete range of clinical presentations, and thus in our study this mutation is not associated with a late, neurological presentation.     

The insulin-like growth factor (IGF)binding protein 1 binding epitope on IGF-I probed by heteronuclear NMR spectroscopy and mutational analysis

NMR spectroscopy studies and biosensor interaction analysis of native and site-directed mutants of insulin-like growth factor I (IGF-I) was applied to identify the involvement of individual residues in IGF-I binding to IGF-binding protein 1 (IGFBP-1). Backbone NMR chemical shifts were found to be affected by IGFBP-1 binding in the following residues: Pro2, Glu3, Cys6, Gly7, Gly19, Pro28-Gly30, Gly32, Arg36, Arg37, Gln40-Gly42, Pro63, Lys65, Pro66, and Lys68-Ala70. Three IGF-I arginine side chains were identified by NMR to participate in IGFBP-1 binding. All IGF-I arginine residues were replaced by alanines, using site-directed mutagenesis, in four single substituted variants, IGF-I(R21A), IGF-I(R50A), IGF-I(R55A), and IGF-I(R56A), and one double replacement mutant, IGF-I(R36A/R37A). Biosensor interaction analysis binding studies demonstrate the involvement of Arg36-Arg37 and Arg50 in IGFBP-1 binding, while experiments with the IGF-I receptor implicate Arg21, Arg36-Arg37, and Arg56 as part of the receptor binding epitope. These overlapping binding surfaces explain why IGF-I receptor and IGFBP-1 binding to IGF-I is competitive. The C terminus of free, but not IGFBP-1-bound, IGF-I is found to exist in two distinct, NMR-detectable conformations at 30 degreesC. One possible explanation for this structural heterogeneity could be cis-trans isomerization of the Cys6-Cys48 disulfide bond.     

Evolution of precipitates in lead-implanted aluminum: A backscattering and channeling study

BACKGROUND: Thirty-six mutations that cause Gaucher disease, the most common glycolipid storage disorder, are known. Although both alleles of most patients with the disease contain one of these mutations, in a few patients one or both disease-producing alleles have remained unidentified. Identification of mutations in these patients is useful for genetic counseling. MATERIALS AND METHODS: The DNA from 23 Gaucher disease patients in whom at least one glucocerebrosidase allele did not contain any of the 36 previously described mutations has been examined by single strand conformation polymorphism (SSCP) analysis, followed by sequencing of regions in which abnormalities were detected. RESULTS: Eight previously undescribed mutations were detected. In exon 3, a deletion of a cytosine at cDNA nt 203 was found. In exon 6, three missense mutations were identified: a C-->A transversion at cDNA nt 644 (Ala176-->Asp), a C-->A transversion at cDNA nt 661 that resulted in a (Pro182-->Thr), and a G-->A transition at cDNA nt 721 (Gly202-->Arg). Two missense mutations were found in exon 7: a G-->A transition at cDNA nt 887 (Arg257-->Gln) and a C-->T at cDNA nt 970 (Arg285-->Cys). Two missense mutations were found in exon 9: a T-->G at cDNA nt 1249 (Trp378-->Gly) and a G-->A at cDNA nt 1255 (Asp380-->Asn). In addition to these disease-producing mutations, a silent C-->G transversion at cDNA nt 1431, occurring in a gene that already contained the 1226G mutation, was found in one family. CONCLUSIONS: The mutations described here and previously known can be classified as mild, severe, or lethal, on the basis of their effect on enzyme production and on clinical phenotype, and as polymorphic or sporadic, on the basis of the haplotype in which they are found. Rare mutations such as the new ones described here are sporadic in nature.     

Selection and characterization of beta-lactam-beta-lactamase inactivator-resistant mutants following PCR mutagenesis of the TEM-1 beta-lactamase gene

Mechanism-based inactivators of beta-lactamases are used to overcome the resistance of clinical pathogens to beta-lactam antibiotics. This strategy can itself be overcome by mutations of the beta-lactamase that compromise the effectiveness of their inactivation. We used PCR mutagenesis of the TEM-1 beta-lactamase gene and sequenced the genes of 20 mutants that grew in the presence of ampicillin-clavulanate. Eleven different mutant genes from these strains contained from 1 to 10 mutations. Each had a replacement of one of the four residues, Met69, Ser130, Arg244, and Asn276, whose substitutions by themselves had been shown to result in inhibitor resistance. None of the mutant enzymes with multiple amino acid substitutions generated in this study conferred higher levels of resistance to ampicillin alone or ampicillin with beta-lactamase inactivators (clavulanate, sulbactam, or tazobactam) than the levels of resistance conferred by the corresponding single-mutant enzymes. Of the four enzymes with just a single mutation (Ser130Gly, Arg244Cys, Arg244Ser, or Asn276Asp), the Asn276Asp beta-lactamase conferred a wild-type level of ampicillin resistance and the highest levels of resistance to ampicillin in the presence of inhibitors. Site-directed random mutagenesis of the Ser130 codon yielded no other mutant with replacement of Ser130 besides Ser130Gly that produced ampicillin-clavulanate resistance. Thus, despite PCR mutagenesis we found no new mutant TEM beta-lactamase that conferred a level of resistance to ampicillin plus inactivators greater than that produced by the single-mutation enzymes that have already been reported in clinical isolates. Although this is reassuring, one must caution that other combinations of multiple mutations might still produce unexpected resistance.     

Identification of novel mutations in the dihydropyrimidine dehydrogenase gene in a Japanese patient with 5-fluorouracil toxicity

5-Fluorouracil (5-FU) is used widely in the treatment of several common neoplasms. Dihydropyrimidine dehydrogenase (DPD) is the initial and rate-limiting enzyme in the catabolism of 5-FU. Several recent studies have described a pharmacogenetic disorder in which cancer patients with decreased DPD activity develop life-threatening toxicity following exposure to 5-FU. We reported recently the first Japanese case of decreased DPD activity accompanied by severe 5-FU toxicity. The present study describes the results of molecular analysis of this patient and her family, in which three novel mutations (Arg21Gln, Val335Leu, and Glu386Ter) of the gene coding for DPD were identified. We also revealed that Arg21Gln and Glu386Ter are on the same allele and that Val335Leu is on the other allele, on the basis of analysis of the family genome. Expression analysis in Escherichia coli showed that Val335Leu and Glu386Ter led to mutant DPD protein with significant loss of enzymatic activity and no activity, respectively. The Arg21Gln mutation, however, resulted in no decrease in enzymatic activity compared with the wild type. The present data represent the first molecular genetic analysis of DPD deficiency accompanied by severe 5-FU toxicity in a Japanese patient.     

Cysteine-scanning mutagenesis of helix IV and the adjoining loops in the lactose permease of Escherichia coli: Glu126 and Arg144 are essential. off

Cys-scanning mutagenesis has been applied to the remaining 45 residues in lactose permease that have not been mutagenized previously (from Gln100 to Arg144 which comprise helix IV and adjoining loops). Of the 45 single-Cys mutants, 26 accumulate lactose to > 75% of the steady state observed with Cys-less permease, and 14 mutants exhibit lower but significant levels of accumulation (35-65% of Cys-less permease). Permease with Phe140-->Cys or Lys131-->Cys exhibits low activity (15-20% of Cys-less permease), while mutants Gly115-->Cys, Glu126-->Cys and Arg144-->Cys are completely unable to accumulate the dissacharide. However, Cys-less permease with Ala or Pro in place of Gly115 is highly active, and replacement of Lys131 or Phe140 with Cys in wild-type permease has a less deleterious effect on activity. In contrast, mutant Glu126-->Cys or Arg144-->Cys is inactive with respect to both uphill and downhill transport in either Cys-less or wild-type permease. Furthermore, mutants Glu126-->Ala or Gln and Arg144-->Ala or Gln are also inactive in both backgrounds, and activity is not rescued by double neutral replacements or inversion of the charged residues at these positions. Finally, a mutant with Lys in place of Arg144 accumulates lactose to about 25% of the steady state of wild-type, but at a slow rate. Replacement of Glu126 with Asp, in contrast, has relatively little effect on activity. None of the effects can be attributed to decreased expression of the mutants, as judged by immunoblot analysis. Although the activity of most of the single-Cys mutants is unaffected by N-ethylmaleimide, Cys replacement at three positions (Ala127, Val132, or Phe138) renders the permease highly sensitive to alkylation. The results indicate that the cytoplasmic loop between helices IV and V, where insertional mutagenesis has little effect on activity [McKenna, E., et al. (1992) Proc. Natl. Acad. Sci. U.S.A. 89, 11954-11958], contains residues that play an important role in permease activity and that a carboxyl group at position 126 and a positive charge at position 144 are absolutely required.     

Analysis of single-strand conformation polymorphisms by capillary electrophoresis with laser induced fluorescence detection

Detection of point mutations in genomic DNA is important for diagnosis of inherited characteristics and genetic diseases. A point mutation in a specific region of DNA amplified by polymerase chain reaction (PCR) can be detected with single-strand conformation polymorphism (SSCP) analysis. Analysis of SSCP by laser-induced fluorescence capillary electrophoresis in entangled polymer solution (CE-LIF) has been developed in the present paper. K-ras genes including seven mutations were amplified with primer labeled with Texas Red at its 5' end. The labeled PCR products were dissociated to single strands by heating and separated with capillary gel electrophoresis and He-Ne laser-excited fluorescence detection. Our results suggest that all fragments having normal (Gly) and mutated (Ala, Arg, Cys, Ser, Val, Asp) sequences at codon 12 can be distinguished. Analysis of SSCPs with CE-LIF is well suited for clinical analysis of SSCPs because of its high sensitivity, resolution, reproducibility and speed.     

RDS/peripherin gene mutations are frequent causes of central retinal dystrophies

Patients from 76 independent families with various forms of mostly central retinal dystrophies were screened for mutations in the RDS/peripherin gene by means of SSCP analysis and direct DNA sequencing. Two nonsense mutations (Gln239ter, Tyr285ter), five missense mutations (Arg172Trp, Lys197Glu, Gly208Asp, Trp246Arg, Ser289Leu), and one single base insertion (Gly208insG), heterozygous in all cases, were detected. Only one of these mutations, Arg172Trp, has been reported previously. Cosegregation of the mutation with the disease phenotype could be established in selected families. Other missense mutations were excluded from a panel of 55-75 control subjects. The patients showed remarkable variation in phenotype and disease expression not only between cases with different mutations but also between affected members of the same family. This study indicates that RDS/peripherin mutations are a frequent cause of various types of central retinal dystrophies and that the RDS/peripherin gene exhibits a broad spectrum of allelic mutations. Comparative analysis of known mutations allowed us to hypothesise that the deleterious effect of RDS/peripherin gene mutations is the result of different molecular mechanisms.     

Factor V Arg306-->Thr (factor V Cambridge) and factor V Arg306-->Gly mutations in venous thrombotic disease

We investigated the prevalence of two reported mutations of the factor V gene (factor V Arg306-->Thr, or factor V Cambridge, and factor V Arg306-->Gly) in 104 relatively young patients with verified venous thrombosis and in 208 age-, sex- and race-matched controls, in order to establish whether the two mutations are associated with increased predisposition for venous thrombosis. PCR amplification followed by BstNI and MspI digestion was employed to determine the genotypes, and each mutation was confirmed by DNA sequencing. Among the controls, one individual was found to be heterozygous for the factor V Arg306-->Thr mutation and one heterozygous for the factor VArg306-->Gly mutation; none of the patients carried either mutation. Our findings do not support factor V Cambridge and factor V Arg306-->Gly as risk factors for venous thrombosis.     

Identification of two missense mutations in the GIP receptor gene: a functional study and association analysis with NIDDM: no evidence of association with Japanese NIDDM subjects

Gastric inhibitory polypeptide (GIP) potently stimulates insulin secretion from pancreatic islets in the presence of glucose as an incretin. Because the insulinotropic effect of GIP is reduced in NIDDM, it should be clarified whether defects in the GIP receptor gene contribute to the impaired insulin secretion in NIDDM. Using genomic DNA samples from Japanese NIDDM and non-NIDDM subjects, we have investigated the entire coding region of the GIP receptor gene by polymerase chain reaction-single strand conformational polymorphism (PCR-SSCP). We have identified two missense mutations, Gly198-->Cys (Gly198Cys) in exon 7 and Glu354-->Gln (Glu354Gln) in exon 12. Investigation of the function of GIP receptor with either of these mutations reveals a half-maximal stimulation value of GIP-induced cAMP response in Chinese hamster ovary cells expressing the GIP receptor with Gly198Cys of 6.3 +/- 1.2 x 10(-10) mol/l (n = 3), which was considerably higher than that of the normal GIP receptor, 9.4 +/- 3.8 x 10(-12) mol/l GIP (n = 3), whereas that of the GIP receptor with Glu354Gln was not significantly different from that of the normal GIP receptor. To assess the possible role of the GIP receptor gene in genetic susceptibility to NIDDM, we have examined the allelic frequencies of Gly198Cys and Glu354Gln in NIDDM and control subjects. Association studies show no relationship between NIDDM and either of the two mutations.     

Mutational analyses support a model for the HRV2 2A proteinase

The proteinase 2A of human rhinovirus 2 is a cysteine proteinase which contains a tightly bound Zn ion thought to be required for structural integrity. A three-dimensional model for human rhinovirus type 2 proteinase 2A (HRV2 2A) was established using sequence alignments with small trypsin-like Ser-proteinases and, for certain regions, elastase. The model was tested by expressing selected proteinase 2A mutants in bacteria and examining the effect on both intramolecular ("cis") and intermolecular ("trans") activities. The HRV2 proteinase 2A is proposed to have a two domain structure, with the catalytic site and substrate binding region on one face of the molecule and a Zn-binding motif on the opposite face. Residues Gly 123, Gly 124, Thr 121, and Cys 101 are proposed to be involved in the architecture of the substrate binding pocket and to provide the correct environment for the catalytic triad of His 18, Asp 35, and Cys 106. Residues Tyr 85 and Tyr 86 are thought to participate in substrate recognition. The presence of an extensive C-terminal helix, in which Asp 132, Arg 134, Phe 130, and Phe 136 play important roles, explains why mutations in this region are generally detrimental to proteinase activity. The proposed Zn-binding motif comprises Cys 52, Cys 54, Cys 112, and His 114. Exchange of these residues inactivates the enzyme. Furthermore, as measured by atom emission spectroscopy, Zn was absent from purified preparations of proteinase 2A in which His 114 had been replaced by Asn. The absence of disulphide bridges was confirmed by subjecting highly purified HRV2 proteinase 2A to one- and two-step alkylation procedures.     

Characterization of two naturally occurring mutations in the second epidermal growth factor-like domain of factor VII

We investigated the mechanisms responsible for severe factor VII (FVII) deficiency in homozygous Italian patients with either Gly97Cys or Gln100Arg mutations in the second epidermal growth factor domain of FVII. Transient expression of complementary DNA coding for the mutations in COS-1 cells showed impaired secretion of the mutant molecules. Using stably transfected Chinese hamster ovary (CHO) cells, we performed pulse-chase labeling studies, immunohistochemistry, and experiments with inhibitors of protein degradation, showing that FVII-Cys97 did not accumulate intracellularly but was degraded in a pre-Golgi, nonlysosomal compartment by a cysteine protease. In stably transfected CHO cells expressing FVII-Arg100, the level of intracellular FVII was not increased by several inhibitors of protein degradation, but FVII-Arg100 was retained in the endoplasmic reticulum for a longer period of time than wild-type FVII. FVII-Arg100 had a lower apparent molecular weight than did wild-type FVII under nondenaturing conditions, which is attributable to misfolding due to abnormal disulfide bond formation.     

Structure-function correlations of calcium binding and calcium channel activities based on 3-dimensional models of human annexins I, II, III, V and VII

The annexins are a family of calcium-dependent phospholipid-binding proteins which share a high degree of primary sequence similarity. Using a model of the crystal structure of annexin V as a template, 3-dimensional models of human annexins I, II, III and VII were constructed by homology modeling (J. Greer, J. Mol. Biol. 153, 1027-1042, 1981; J.M. Chen, G. Lee, R.B. Murphy, R.P. Carty, P.W. Brant-Rauf, E. Friedman and M.R. Pincus, J. Biomolec. Str. Dyn. 6, 859-87, 1989) for the 316 amino acid portions corresponding to the annexin V structure published by Huber et al. (J. Mol. Biol. 223, 683-704, 1992). These methods were used to study structure-function correlations for calcium ion binding and calcium channel activity. Published experimental data are specifically shown to be consistent with the annexin models. Possible intramolecular disulfide bridges were identified in annexin I (between Cys297 and Cys316) and in annexins II and VII (between Cys115 and Cys243). Each of the annexin models have 3 postulated calcium binding sites, usually via a Gly-Xxx-Gly-Thr loop with an acidic Glu or Asp residue 42 positions C-terminal to the first Gly. Despite a nonconserved binding site sequence, annexins I and II are able to coordinate calcium in domain 3 since the residue in the second loop position is directed toward the solvent away from the binding pocket. This finding also suggests a mechanism for a conformational change upon binding calcium. Highly conserved Arg and acidic sidechains stabilize the channel pore structure; annexin channels probably exist in a closed state normally. Arg271 may be involved in channel opening upon activation: basic residue 254 can stabilize Glu112, which allows Arg271 to interact with residue 95 instead of Glu112. Residue 267, found on the convex surface at the pore opening, may also be important in modifying channel activity.     

Beta-2 adrenoceptor genetic variation is associated with genetic predisposition to essential hypertension: The Bergen Blood Pressure Study

We tested the hypothesis that genetic variation in the beta-2 adrenoceptor gene is associated with a genetic predisposition to hypertension. Offspring of two hypertensive parents were compared with offspring of two normotensive parents. The subjects were participants of the Bergen Blood Pressure Study, where couples were recruited in 1963 to 1964 and re-examined in 1990. We studied offspring of those couples in which both partners were either hypertensive or normotensive in both examinations. Twenty-three hypertensive and 22 normotensive families met the inclusion criteria. DNA samples from the first born of hypertensive family-history offspring and normotensive family-history offspring were analyzed. We used multiplex sequencing and specifically examined the promoter and the N-terminal portion of the beta-2 adrenoceptor gene. We found four genetic variants: at position -47, a C-->T substitution in the 5' leader cistron causing an Arg-->Cys exchange, at -20, a T-->C substitution, at +46 an A-->G substitution leading to an Arg16-->Gly exchange, and at +79, a C-->G substitution leading to a Gln27-->Glu exchange. The frequency of the Arg16 allele was significantly higher in the hypertensive family-history offspring compared to normotensive family-history offspring (58% vs. 28% P < 0.011). We constructed haplotypes for the four intragenic variants and found significant linkage dysequilibrium. In particular, the 5' leader cistron mutant with the wild type alleles at the other loci was significantly more frequent in offspring of hypertensive parents, compared to offspring of normotensive parents. We also performed a relative risk analysis comparing the Gly/Gly, Arg/Gly, and Arg/Arg alleles, which implicated the Arg-containing allele. Finally, we analyzed the effect of genotype on blood pressure in the offspring. We found a significant step-wise effect for all four polymorphisms examined. Our data suggest that the Arg variant of the Arg-->Gly exchange is associated with parental hypertension and higher blood pressure values in this northern European population.     

Probing cellular protein targets of H2O2 with fluorescein-conjugated iodoacetamide and antibodies to fluorescein

Fatty acid ethyl ester synthase-III metabolizes both ethanol and carcinogens. Structure-function studies of the enzyme have not been performed in relation to site specific mutagenesis. In this study, three residues (Gly 32, Cys 39 and His 72) have been mutated to observe their role in enzyme activity. Gly to Gln, Cys to Trp and His to Ser mutations did not affect fatty acid ethyl ester synthase activity, but His to Ser mutant had less than 9% of control glutathione S-transferase activity. The apparent loss of transferase activity reflected a 28 fold weaker binding constant for glutathione. Thus, this study indicates that Gly and Cys may not be important for synthase or transferase activities however, histidine may play a role in glutathione binding, but it is not an essential catalytic residue of glutathione S-transferase or for fatty acid ethyl ester synthase activity.     

Six previously undescribed pyruvate kinase mutations causing enzyme deficiency

Erythrocyte pyruvate kinase deficiency is the most common cause of hereditary nonspherocytic hemolytic anemia. We present 6 previously undescribed mutations of the PKLR gene associated with enzyme deficiency located at cDNA nt 476 G-->T (159Gly-->Val), 884 C-->T (295Ala-->Val), 943 G-->A (315Glu-->Lys), 1022 G-->A (341Gly-->Asp), 1511 G-->T (504Arg-->Leu), and 1528 C-->T (510Arg-->Ter). Two of these mutations are near the substrate binding site: the 315Glu-->Lys (943A) mutation may be involved in Mg2+ binding and 159Gly-->Val (476T) mutation has a possible effect on ADP binding. Four of six mutations produce deduced changes in the shape of the molecule. Two of these mutations, 504Arg-->Leu (1511T) and 510Arg-->Ter (1528T), are located at the interface of domains A and C. One of them (510Arg-->Ter) is a deletion of the C-terminal residues affecting the integrity of the protein. The 504Arg-->Leu mutation eliminates a stabilizing interaction between domains A and C. Changes in amino acid 341(nt 1022) from Gly to Asp cause local perturbations. The mutation 295Ala-->Val (884T) might affect the way pyruvate kinase interacts with other molecules. We review previously described mutations and conclude that there is not yet sufficient data to allow us to draw conclusions regarding genotype/phenotype relationship.     

Familial distal renal tubular acidosis is associated with mutations in the red cell anion exchanger (Band 3, AE1) gene

All affected patients in four families with autosomal dominant familial renal tubular acidosis (dRTA) were heterozygous for mutations in their red cell HCO3-/Cl- exchanger, band 3 (AE1, SLC4A1) genes, and these mutations were not found in any of the nine normal family members studied. The mutation Arg589--> His was present in two families, while Arg589--> Cys and Ser613--> Phe changes were found in the other families. Linkage studies confirmed the co-segregation of the disease with a genetic marker close to AE1. The affected individuals with the Arg589 mutations had reduced red cell sulfate transport and altered glycosylation of the red cell band 3 N-glycan chain. The red cells of individuals with the Ser613--> Phe mutation had markedly increased red cell sulfate transport but almost normal red cell iodide transport. The erythroid and kidney isoforms of the mutant band 3 proteins were expressed in Xenopus oocytes and all showed significant chloride transport activity. We conclude that dominantly inherited dRTA is associated with mutations in band 3; but both the disease and its autosomal dominant inheritance are not related simply to the anion transport activity of the mutant proteins.     

Identification of a phosphate regulatory site and a low affinity binding site for glucose 6-phosphate in the N-terminal half of human brain hexokinase

Crystal structures of human hexokinase I reveal identical binding sites for phosphate and the 6-phosphoryl group of glucose 6-phosphate in proximity to Gly87, Ser88, Thr232, and Ser415, a binding site for the pyranose moiety of glucose 6-phosphate in proximity to Asp84, Asp413, and Ser449, and a single salt link involving Arg801 between the N- and C-terminal halves. Purified wild-type and mutant enzymes (Asp84 --> Ala, Gly87 --> Tyr, Ser88 --> Ala, Thr232 --> Ala, Asp413 --> Ala, Ser415 --> Ala, Ser449 --> Ala, and Arg801 --> Ala) were studied by kinetics and circular dichroism spectroscopy. All eight mutant hexokinases have kcat and Km values for substrates similar to those of wild-type hexokinase I. Inhibition of wild-type enzyme by 1,5-anhydroglucitol 6-phosphate is consistent with a high affinity binding site (Ki = 50 microM) and a second, low affinity binding site (Kii = 0.7 mM). The mutations of Asp84, Gly87, and Thr232 listed above eliminate inhibition because of the low affinity site, but none of the eight mutations influence Ki of the high affinity site. Relief of 1,5-anhydroglucitol 6-phosphate inhibition by phosphate for Asp84 --> Ala, Ser88 --> Ala, Ser415 --> Ala, Ser449 --> Ala and Arg801 --> Ala mutant enzymes is substantially less than that of wild-type hexokinase and completely absent in the Gly87 --> Tyr and Thr232 --> Ala mutants. The results support several conclusions. (i) The phosphate regulatory site is at the N-terminal domain as identified in crystal structures. (ii) The glucose 6-phosphate binding site at the N-terminal domain is a low affinity site and not the high affinity site associated with potent product inhibition. (iii) Arg801 participates in the regulatory mechanism of hexokinase I.