[Biomarkers for early diagnosis of Alzheimer's disease: current update and future directions]. / Diagnostic biologique de la maladie d'Alzheimer: avancées, limites et perspectives.

INTRODUCTION: The increased prevalence of the sporadic form of Alzheimer's disease (AD) has become a significant health issue in the elderly population. The need for early diagnosis is imperative because this, along with the development of novel therapeutic treatments, would permit the rapid and perhaps more efficient treatment of these debilitating disorders early on. BACKGROUND: Over the last decade, the potential use of certain biomarkers in the cerebrospinal fluid (CSF), and more recently, in the plasma has been investigated. Among the candidates studied includes the neurotoxic amyloid beta peptide and the Tau protein. However, although these two proteins have been clearly shown to be directly related to the pathophysiology of this disorder, it has proven difficult to establish a clear relationship between plasma or CSF levels of Abeta and Tau and the incidence and severity of AD in patients. This is due in part to differences in methodologies related to the detection sensitivity, as well as the variations in the biological data and consequent interpretation of the biochemical and biological data. Peripheral cells, in particular platelets and skin fibroblasts, could be an alternative solution as peripheral biological markers for the early diagnosis of AD. These cells are easily accessible from patients. Furthermore, they would provide a means not only to validate potential therapeutic strategies, but also to study the mechanisms involved in the development of AD, including APP processing. PERSPECTIVES: A combined strategy using both a fundamental mechanistic and an analytical approach of patient peripheral cells will allow the identification of new biological markers for AD, and hence permit immediate therapeutic strategies to be implemented.

Sequential expression patterns of apoptosis- and cell cycle-related proteins in neuronal response to severe or mild transient hypoxia.

Severe hypoxia was shown to induce apoptotic death in developing brain neurons, whereas mild hypoxia was demonstrated to stimulate neurogenesis. Since the apoptotic process may share common pathways with mitosis, expression profiles of proteins involved in apoptosis or the cell cycle were analyzed by immunohistochemistry and/or western blotting, in relation with cell outcome of cultured neurons from fetal rat forebrain subjected to either lethal (6 h) or non-lethal (3 h) hypoxia (95% N(2)/5% CO(2)). Hypoxia for 6 h led to apoptosis that was inhibited by the cell cycle blocker olomoucine. Transient overexpression of proliferating cell nuclear antigen was followed by increasing expression of p53, p21, Bax and caspases, whereas Bcl-2 and heat shock proteins were progressively repressed. Conversely, a 3-h hypoxic insult initiated neuronal mitosis, with increased thymidine incorporation. In these conditions, levels of proliferating cell nuclear antigen, Rb, Bcl-2 and heat shock proteins were persistently elevated, while expression of p53, p21, Bax and caspases gradually decreased. These data confirm that hypoxia promotes cell cycle activation, whatever the stress intensity. This process is then aborted following apoptosis-inducing hypoxia, whereas sublethal insult would trigger neurogenesis, at least in developing brain neurons in vitro, by stimulating timed expression of neurogenic and survival-associated proteins.

Allele varepsilon4 of apolipoprotein E gene is less frequent in Down syndrome patient of the Sicilian population and has no influence on the grade of mental retardation.

We evaluated the allele (and genotype frequencies in 60 Down syndrome (DS), 25 mothers and 57 controls from Sicily and its relation with mental retardation. DS patients and sex ratio (M:F) was 22.1+/-10.5 and 1.14, respectively. Allele varepsilon4 and varepsilon3 frequencies were respectively lower (P=0.015) and higher (P=0.005) in DS patients compared to controls. Genotype varepsilon3/varepsilon4 and varepsilon3/varepsilon3 were less (P=0.03) and more frequent (P=0.001) in DS patients, with respective odd ratios of 0.31 (CI at 95%: 0.18-0.49) and of 4.4 (CI at 95%: 3.4-5.7). No difference of allele (distribution was found in function of the grades of mental retardation according to DMS-IV. Our results show that the implication of Apo-E4 in the pathogenesis of Alzheimer disease cannot be extrapolated in that of dementia of DS.

ApoE protects cortical neurones against neurotoxicity induced by the non-fibrillar C-terminal domain of the amyloid-beta peptide.

Although the genetic link between the epsilon 4 allele of apolipoprotein E (apoE) and Alzheimer's disease (AD) is well established, the apoE isoform-specific activity underlying this correlation remains unclear. We have recently characterized the interaction of the soluble the amyloid-beta peptide (A beta) with model membrane and demonstrated that non-fibrillar A beta peptide, including N-terminal truncated forms of A beta, induced apoptotic cell death in primary rat cortical neurones in vitro. To further investigate the potential interaction between apoE and A beta in the pathogenesis of AD, we have determined the effect of apoE isoforms on the neurotoxicity of non-fibrillar A beta peptides. We demonstrate here that the apoE2 and E3 isoforms protect cortical neurones against apoptotic cell death induced by a non-fibrillar form of the A beta(1-40), A beta(12-42), A beta(29-40) and A beta(29-42) peptides, whereas apoE4 had no effect. This effect involves the formation of stable complexes between apoE and the C-terminal domain (e.g. amino acids 29-40) of A beta(1-40). Interestingly, apoE had no effect on the toxicity induced by aggregated A beta peptides, suggesting a lack of interaction between apoE and amyloid fibrils. Our results provide evidence that interaction with the C-terminal domain of A beta, apoE2 and E3, but not apoE4, inhibits the interactions of the non-fibrillar A beta peptide with the plasma membrane of neurones, A beta peptide aggregation and subsequent neurotoxicity.

A nonfibrillar form of the fusogenic prion protein fragment [118-135] induces apoptotic cell death in rat cortical neurons.

Neuronal loss is a salient feature of prion diseases. However, its cause and mechanism, particularly its relationship with the accumulation and precipitation of the pathogenic, protease-resistant isoform PrP(Sc) of the cellular prion protein PrP(C), are still an enigma. Several studies suggest that neuronal loss could occur through a process of programmed cell death, which is consistent with the lack of inflammation in these conditions. By analogy with the pathological events occurring during the development of Alzheimer's disease, controversies still exist regarding the relationship between amyloidogenesis, prion aggregation, and neuronal loss. We recently demonstrated that a prion protein fragment (118-135) displayed membrane-destabilizing properties and was able to induce, in a nonfibrillar form, the fusion of unilamellar liposomes. To unravel the mechanism of prion protein neurotoxicity, we characterize the effects of the human Pr[118-135] peptide on rat cortical neurons. We demonstrate that low concentrations of the Pr[118-135] peptide, in a nonfibrillar form, induce a time- and dose- dependent apoptotic cell death, including caspase activation, DNA condensation, and fragmentation. This toxicity might involve oxidative stress, because antioxidant molecules, such as probucol and propyl gallate, protect neurons against prion peptide toxicity. By contrast, a nonfusogenic variant Pr[118-135, 0 degrees ] peptide, which displays the same amino acid composition but several amino acid permutations, is not toxic to cortical neurons, which emphasizes the critical role of the fusogenic properties of the prion peptide in its neurotoxicity. Taken together, our results suggest that the interaction between the Pr[118-135] peptide and the plasma membrane of neurons might represent an early event in a cascade leading to neurodegeneration.

Molecular basis of Alzheimer's disease.

Despite an exponential production of data, Alzheimer's disease (AD) remains an enigma. Unresolved questions persist in the face of the heterogeneity of this neuropathology. Recent progress in understanding mechanisms for AD results from the study of amyloid precursor protein (APP) metabolism and the involvement of senile plaque-associated proteins. In addition to the amyloid cascade hypothesis, alternative schemes emerge, in which the amyloid peptide is not the primary effector of the disease. Perturbations of vesicular trafficking, the cytoskeletal network, and membrane cholesterol distribution could be central events. Furthermore, since the physiological role of APP, presenilins, and apolipoprotein E in the central nervous system are not completely understood, their involvement in AD etiology remains speculative. New actors have to be found to try to explain sporadic cases and non-elucidated familial cases.

The nonfibrillar amyloid beta-peptide induces apoptotic neuronal cell death: involvement of its C-terminal fusogenic domain.

The toxicity of the nonaggregated amyloid beta-peptide (1-40) [A beta(1-40)] on the viability of rat cortical neurons in primary culture was investigated. We demonstrated that low concentrations of A beta peptide, in a nonfibrillar form, induced a time- and dose-dependent apoptotic cell death, including DNA condensation and fragmentation. We compared the neurotoxicity of the A beta(1-40) peptide with those of several A beta-peptide domains, comprising the membrane-destabilizing C-terminal domain of A beta peptide (e.g., amino acids 29-40 and 29-42). These peptides reproduced the effects of the (1-40) peptide, whereas mutant nonfusogenic A beta peptides and the central region of the A beta peptide (e.g., amino acids 13-28) had no effect on cell viability. We further demonstrated that the neurotoxicity of the nonaggregated A beta peptide paralleled a rapid and stable interaction between the A beta peptide and the plasma membrane of neurons, preceding apoptosis and DNA fragmentation. By contrast, the peptide in a fibrillar form induced a rapid and dramatic neuronal death mainly through a necrotic pathway, under our conditions. Taken together, our results suggest that A beta induces neuronal cell death by either apoptosis and necrosis and that an interaction between the nonfibrillar C-terminal domain of the A beta peptide and the plasma membrane of cortical neurons might represent an early event in a cascade leading to neurodegeneration.

Laminin 1 attenuates beta-amyloid peptide Abeta(1-40) neurotoxicity of cultured fetal rat cortical neurons.

A growing amount of evidence indicates the involvement of extracellular matrix components, especially laminins, in the development of Alzheimer's disease, although their role remains unclear. In this study, we clearly demonstrate that laminin 1 inhibits beta-amyloid peptide (Abeta)-induced neuronal cell death by preventing the fibril formation and interaction of the Abeta peptide with cell membranes. The presence of laminin at a laminin/Abeta peptide molar ratio of 1:800 significantly inhibits the Abeta-induced apoptotic events, together with inhibition of amyloid fibril formation. The inhibitory effects of laminin 1 were time- and dose-dependent, whereas laminin 2 had less effect on Abeta neurotoxicity. A preincubation of laminin and Abeta was not required to observe the protective effect of laminin, suggesting a direct interaction between laminin 1 and Abeta. Moreover, laminin had no effect on the toxicity of the fibrillar Abeta peptide, suggesting an interaction of laminin with nonfibrillar species of the Abeta peptide, sequestering the peptide in a soluble form. These data extend our understanding of laminin-dependent binding of Abeta and highlight the possible modulation role of laminin regarding Abeta aggregation and neurotoxicity in vivo.

Molecular determinants of the interaction between the C-terminal domain of Alzheimer's beta-amyloid peptide and apolipoprotein E alpha-helices.

In a previous work, we predicted and demonstrated that the 29-42-residue fragment of beta-amyloid peptide (Abeta peptide) has in vitro capacities close to those of the tilted fragment of viral fusion proteins. We further demonstrated that apolipoprotein E2 and E3 but not apolipoprotein E4 can decrease the fusogenic activity of Abeta(29-42) via a direct interaction. Therefore, we suggested that this fragment is implicated in the neurotoxicity of Abeta and in the protective effects of apolipoprotein E in Alzheimer's disease. Because structurally related apolipoproteins do not interact with the Abeta C-terminal domain but inhibit viral fusion, we suggested that interactions existing between fusogenic peptides and apolipoproteins are selective and responsible for the inhibition of fusion. In this study, we simulated interactions of all amphipathic helices of apolipoproteins E and A-I with Abeta and simian immunodeficiency virus (SIV) fusogenic fragments by molecular modeling. We further calculated cross-interactions that do not inhibit fusion in vitro. The results suggest that interactions of hydrophobic residues are the major event to inhibit the fusogenic capacities of Abeta(29-42) and SIV peptides. Selectivity of those interactions is due to the steric complementarity between bulky hydrophobic residues in the fusogenic fragments and hydrophobic residues in the apolipoprotein C-terminal amphipathic helices.

Beta-amyloid peptide interacts specifically with the carboxy-terminal domain of human apolipoprotein E: relevance to Alzheimer's disease.

Growing evidence indicates the involvement of apolipoprotein E (apoE) in the development of late-onset and sporadic forms of Alzheimer's disease, although its exact role remains unclear. We previously demonstrated that beta-amyloid peptide (Abeta) displays membrane-destabilizing properties and that only apoE2 and E3 isoforms inhibit these properties. In this study, we clearly demonstrate that the carboxy-terminal lipid-binding domain of apoE (e.g., residues 200-299) is responsible for the Abeta-binding activity of apoE and that this interaction involves pairs of apoE amphipathic alpha-helices. We further demonstrate that Abeta is able to inhibit the association of the C-terminal domain of apoE with lipids due to the formation of Abeta/apoE complexes resistant to sodium dodecyl sulfate-polyacrylamide gel electrophoresis. On the contrary, the amino-terminal receptor-binding domain of apoE (e.g., residues 129-169) is not able to form stable complexes with Abeta. These data extend our understanding of human apoE-dependent binding of Abeta by involving the C-terminal domain of apoE in the efficient formation of apoE/Abeta complex.

Peptides in membranes: tipping the balance of membrane stability.

This review describes a class of peptides that associate with lipids in membranes and are commonly known as 'oblique-orientated peptides'. Owing to an asymmetric distribution of hydrophobic residues along the axis of the alpha-helix, such peptides can destabilize membranes or lipid cores, thereby facilitating such cellular processes as vesicular fusion or protein transport across subcellular compartments, as well as remodelling of lipid cores.

Specific modulation of the fusogenic properties of the Alzheimer beta-amyloid peptide by apolipoprotein E isoforms.

C-terminal fragments of the Alzheimer amyloid peptide (amino acids 29-40 and 29-42) have physico-chemical properties related to those of the fusion peptides of viral proteins and they are able to induce the fusion of liposomes in vitro. We proposed that these properties could mediate a direct interaction of the amyloid peptide with cell membranes and account for part of the cytotoxicity of the amyloid peptide. In view of the epidemiologic and biochemical linkages between the pathology of Alzheimer's disease and apolipoprotein E (apoE) polymorphism, we examined the potential interaction between the three common apoE isoforms and the C-terminal fragments of the amyloid peptide. We show that, at low concentration, only apoE2 and apoE3 are potent inhibitors of the amyloid peptide fusogenic and aggregational properties, whereas the apoE4 isoform has no effect. We further show that the protective effect of apoE is mediated by the formation of stable apoE/amyloid peptide complexes, as determined by tryptophan emission fluorescence measurements and by gel electrophoresis. The interaction specificity between apoE2 and apoE3 and the amyloid fragments is demonstrated here, since other apolipoproteins (e.g. apolipoprotein A-I and A-II), with similar amphipathic structures, do not interact with the amyloid C-terminal fragments. Finally, we show that, reciprocally, the amyloid peptide can interact directly with the apoE2 and apoE3 isoforms to decrease or perturb their normal association with lipids. These data suggest that the 29-40 and 29-42 domains of the amyloid peptide could be critical for the amyloid-apoE interaction, and that apoE2 and apoE3 isoforms, but not apoE4, could play a protective role against the formation of amyloid aggregates and/or against their interaction with cellular membranes.

The 118-135 peptide of the human prion protein forms amyloid fibrils and induces liposome fusion.

The prion protein (PrPC) is a glycoprotein of unknown function normally found at the surface of neurons and of glial cells. It is involved in diseases such as bovine spongiform encephalopathy, and Creutzfeldt-Jakob disease in the human, where PrPC is converted into an altered form (termed PrPSc). PrPSc is highly resistant towards proteolytic degradation and accumulates in the central nervous system of affected individuals. By analogy with the pathological events occuring during the development of Alzheimer's disease, controverses still exist regarding the relationship between amyloidogenesis, prion aggregation and neuronal loss. To unravel the mechanism of PrP neurotoxicity and understand the interaction of PrP with cellular membranes, a series of natural and variant peptides spanning residues 118 to 135 of PrP was synthesized. The potential of these peptides to induce fusion of unilamellar lipid vesicles was investigated. According to computer modeling calculations, the 120 to 133 domain of PrP is predicted to be a tilted lipid-associating peptide, and to insert in a oblique way into a lipid bilayer through its N-terminal end. In addition to amyloidogenic properties exhibited in vitro by these peptides, peptide-induced vesicle fusion was demonstrated by several techniques, including lipid- and core-mixing assays. Elongation of the 120 to 133 peptide towards the N- and C-terminal ends of the PrP sequence showed that the 118 to 135 PrP peptide has maximal fusogenic properties, while the variant peptides had no effect. Due to their high hydrophobicity, all peptides tested were able to interact with liposomes to induce leakage of encapsulated calcein. We demonstrate also that the propensity of the peptides to fold as an alpha-helix increases their fusogenic activity, thus accounting for the maximal fusogenic activity of the most stable helix at residues 118 to 135. These data suggest that, by analogy with the C-terminal domain of the beta-amyloid peptide, the fusogenic properties exhibited by the prion peptides might contribute to the neurotoxicity of these peptides by destabilizing cellular membranes.

Fusogenic properties of the C-terminal domain of the Alzheimer beta-amyloid peptide.

A series of natural peptides and mutants, derived from the Alzheimer beta-amyloid peptide, was synthesized, and the potential of these peptides to induce fusion of unilamellar lipid vesicles was investigated. These peptide domains were identified by computer modeling and correspond to respectively the C-terminal (e.g. residues 29-40 and 29-42) and a central domain (13-28) of the beta-amyloid peptide. The C-terminal peptides are predicted to insert in an oblique way into a lipid membrane through their N-terminal end, while the mutants are either parallel or perpendicular to the lipid bilayer. Peptide-induced vesicle fusion was demonstrated by several techniques, including lipid-mixing and core-mixing assays using pyrene-labeled vesicles. The effect of peptide elongation toward the N-terminal end of the entire beta-amyloid peptide was also investigated. Peptides corresponding to residues 22-42 and 12-42 were tested using the same techniques. Both the 29-40 and 29-42 beta-amyloid peptides were able to induce fusion of unilamellar lipid vesicles and calcein leakage, and the amyloid 29-42 peptide was the most potent fusogenic peptide. Neither the two mutants or the 13-28 beta-amyloid peptide had any fusogenic activity. Circular dichroism measurements showed an increase of the alpha-helical content of the two C-terminal peptides at increasing concentrations of trifluoroethanol, which was accompanied by an increase of the fusogenic potential of the peptides. Our data suggest that the alpha-helical content and the angle of insertion of the peptide into a lipid bilayer are critical for the fusogenic activity of the C-terminal domain of the amyloid peptide. The differences observed between the fusogenic capacity of the amyloid 29-40 and 29-42 peptides might result from differences in the degree of penetration of the peptides into the membrane and the resulting membrane destabilization. The longer peptides, residues 22-42 and 12-42, had decreased, but significant, fusogenic properties associated with perturbation of the membrane permeability. These data suggest that the fusogenic properties of the C-terminal domain of the beta-amyloid peptide might contribute to the cytotoxicity of the peptide by destabilizing the cell membrane.

Single-step purification of two functional human apolipoprotein E variants hyperexpressed in Escherichia coli.

We have cloned, from total human liver RNA, the cDNA encoding apolipoprotein E3 (apoE3). Site-directed mutagenesis was used to obtain the cDNA encoding the apoE4 isoform, a major variant of this apolipoprotein in man. These two cDNAs were subcloned into the procaryotic expression vector pAHRS. A polyhistidine tag was added at the NH2-termini of the recombinant proteins (apoE3 and apoE4) to enable rapid purification. The resulting plasmids (pAHRS-apoE3 and pAHRS-apoE4) were introduced into the Escherichia coli strain BL21(DE3). Recombinant strains were grown at 37 degrees C in a Luria and Bertani medium and the addition of isopropyl beta-thiogalactoside resulted in the expression of large amounts of apoE protein (40.5 kDa), representing at least 15% of cellular proteins. The recombinant apoE isoforms were purified, under denaturating conditions, in one step by affinity chromatography on a Ni-chelated agarose column, yielding to about 20 mg of 96% pure protein per liter of culture. Compared to plasma apoE3 purified from human very low density lipoproteins, the two renatured recombinant apoE isoforms have the same secondary structure content, as revealed by circular dichroism measurement. Moreover, the recombinant apoE3 isoform shares similar properties for the association with lipids, compared to the human protein, indicating that the addition of the amino-terminal polyhistidine peptide does not influence the structure and the lipid binding properties of this recombinant apoE isoform. No differences in the secondary structure of recombinant apoE4 were detected, whereas this isoform presents specific reactivity with lipids. This simple and rapid procedure for the expression and the purification of functional recombinant apoE should therefore enable structural and physiological studies requiring large amounts of these apolipoproteins.

Apolipoprotein (apo) E genotype and apoE concentration determine binding of normal very low density lipoproteins to HepG2 cell surface receptors.

The clinical relevance of apoE concentration in lipoprotein fractions should be evaluated. We investigated the impact of the common apolipoprotein (apo) E polymorphism in conjunction with very low density lipoprotein (VLDL) apoE concentration on the receptor binding properties of VLDL preparations from 17 normolipidemic subjects of the HepG2 cell surface receptors. All six apoE genotypes were studied. When apoE genotype alone was considered, two subgroups could be distinguished: VLDL without apoE isoform E2 (VLDL-3/3, VLDL-3/4, and VLDL-4/4) showed significantly higher affinity than VLDL with apoE2 (VLDL-4/2, VLDL-3/2, and VLDL-2/2). Once we adjusted for VLDL apoE content, we observed that VLDL affinity to HepG2 cell surface receptors decreased, according to apoE genotype, in the following order: VLDL-4/4 (100%) > VLDL-3/4 (93%) > VLDL-3/3 (82%) > VLDL-4/2 (53%) > VLDL-3/2 (36%) > VLDL-2/2 (30%). Moreover, we found that VLDL apoE concentration could modify isoform-specific binding. An analysis in 47 subjects showed that the concentration of total VLDL protein and the VLDL apoE concentration varied considerably. The variation of VLDL apoE was independent of apoE genotype and corresponding serum apoE levels. We conclude that, in addition to apoE genotype, apoE content of VLDL is an important determinant of the receptor binding properties of VLDL.

[Alzheimer disease: hypotheses implicating apolipoproteins E]. / Maladie d'Alzheimer: hypothèses impliquant les apolipoprotéines E.

The most promising discovery in the study of Alzheimer's disease (AD) markers is undoubtedly the implication of apolipoprotein E (apoE). The gene of this apoliprotein is located on chromosome 19 and is characterized by three common alleles epsilon 2, epsilon 3 and epsilon 4 giving raise to 6 genotypes and 6 protein phenotypes. ApoE is well known for its role in cholesterol transport. Different studies have been performed, giving major arguments in favor of an important role of apoE in the pathophysiology of AD. These include the discovery of the relationship between the epsilon 4 allele and AD, the ability of this protein to form complexes with beta amyloid protein (A beta) in seniles plates, its presence in neurofibrillary tangles and in vessels of AD patients. Another important finding is the differential interaction between the different isoforms of apoE and tau protein. Some of the hypotheses implicate the role of different apoE isoforms on the growth and extension of neurones, perhaps by a receptor mediated pathway. It has been suggested that apoE acts as a pathological chaperone protein, or alternatively that it protects neurons by regulation of the cell membrane and modifying calcium homeostasis. It is clear that apoE genotype determination alone cannot be used for diagnosis of AD. The presence of epsilon 4 allele is only one of several risk factors for the development of the disease. Other factors may also be implicated and are the subject of ongoing research.

Apolipoprotein E: an important gene and protein to follow in laboratory medicine.

The human apolipoprotein (apo) E gene is polymorphic, with three common alleles (epsilon 2, epsilon 3, epsilon 4) coding for three isoforms (E2, E3, E4). The isoforms differ from each other by a single amino acid substitution, and also differ in their binding affinity for the four apo E receptors. Apo E polymorphism is an important determinant of risk for the development of cardiovascular and Alzheimer diseases, the prevalence of the epsilon 4 allele being increased in both kinds of patients compared with control subjects. Furthermore, the prevalence of the epsilon 4 allele differs among populations (range 5-40%, respectively, for Taiwanese and Papua New Guineans). Genotyping or phenotyping needs to be introduced in clinical laboratories. The choice of the method should be based on the types of patients who are examined. The apo E genotype is also a determinant of apo E plasma concentration. Standardization of apo E measurement is an important prerequisite before investigating the clinical interest of plasma apo E concentration. Determination of apo E genotype/phenotype and later the plasma concentration are expected to yield useful clinical laboratory information.

Expression and role of the human liver UDP-glucuronosyltransferase UGT1*6 analyzed by specific antibodies raised against a hybrid protein produced in Escherichia coli.

Characterization of human UDP-glucuronyltransferases (UGTs) has been limited by the unavailability of probes selective for each of several highly related isoforms. To better understand the role of this superfamily in the metabolism of drugs and xenobiotics, we describe a molecular/immunological strategy for discriminating the implication of each human isoenzyme in this process. Specific polyclonal antibodies were generated against the divergent amino-terminal domain of the UGT isoform UGT1*6 which is involved in the detoxification of nucleophilic compounds related to phenols and naphthols in human liver. The novel approach consists of the expression of a N-terminal UGT polypeptide fused to Staphylococcus aureus protein A in Escherichia coli and a single step purification of the fusion protein by immunoaffinity chromatography. Immunoblot and immunoinhibition analysis showed that the antibodies raised against the fusion protein selectively recognized both the denaturated and the native forms of UGT1*6, when expressed in V79 cell lines, but not three other recombinant UGT isoenzymes. In human liver microsomes, specific immunoinhibition analysis demonstrated that glucuronidation by UGT1*6 represented 20 to 50% of the total 1-naphthol UGT activity with a good correlation with the amount of protein selectively quantified on immunoblot. The specific expression of UGT1*6 was found to be significantly reduced in tumoral tissues but enhanced in cholestatic livers, when compared with healthy hepatic tissues. Interestingly, in human kidney microsomes, antibodies revealed a high level of UGT1*6 expression on immunoblot and inhibited 1-naphthol glucuronidation up to 55%, indicating that this isoform is also expressed in kidney and extensively contributes to phenol glucuronidation in this tissue.

Expression of the human UDP-glucuronosyltransferase UGT1*6 in Escherichia coli. Influence of bacterial signal peptides on the production and localization of the recombinant protein.

The membrane-bound human liver UDP-glucuronosyltransferase UGT1*6 was expressed in Escherichia coli. Exchange of the natural signal peptide by the bacterial signal peptides of pclB or OmpT proteins considerably increased the level of expression and, as the natural signal peptide, targeted the protein to the membranes. The extent of maturation of SpelB-UGT1*6 precursor was about 30%. No processing of sOmpT-UGT1*6 occurred but the processing rate of this precursor could be significantly increased by mutagenesis of the first two amino acid residues of the mature sequence. These expression vectors allowed us to produce high levels of recombinant mature UGT1*6 required for further structural studies.