Sites In Human Brain Research Articles

Opioids and obstructive sleep apnea.

Freire C, Sennes LU, Polotsky VY. Opioids and obstructive sleep apnea. J Clin Sleep Med. 2022;18(2):647-652.

Autoradiographic mapping of synaptic vesicle glycoprotein 2A in non-human primate and human brain.

Synaptic vesicle glycoprotein 2A (SV2A) has been previously characterized as an imaging biomarker for assessment of synaptic density in positron emission tomography (PET) studies of patients with neurological conditions. To provide detailed maps of the brain localization of SV2A autoradiography studies were carried out using the SV2A radioligand [11 C]UCB-J and whole hemisphere sections of non-human primate (NHP) and human brain. Binding of [11 C]UCB-J was observed in all evaluated grey matter structures of the primate brain, with highest density in the caudate nucleus and cortex and lowest density in pons and globus pallidus. The density of [11 C]UCB-J binding sites in human brain showed a good correlation with that in NHP brain. Binding of [11 C]UCB-J in the white matter was very low relative to that in grey matter containing structures and was only inhibited to a minor extent by co-incubation with a saturating concentration of unlabelled UCB-J. The high-resolution images obtained in the present study may aid the interpretation of data acquired in human subjects examined using [11 C]UCB-J in PET studies. In addition, observation of low binding for [11 C]UCB-J in white matter (centrum semiovale) supports that this structure can be used as a reference region for quantitative analysis of [11 C]UCB-J PET data.

Transcriptome-wide Discovery of microRNA Binding Sites in Human Brain

The orchestration of brain function requires complex gene regulatory networks that are modulated, in part, by microRNAs (miRNAs). These noncoding RNAs associate with argonaute (Ago) proteins in order to direct posttranscriptional gene suppression via base pairing with target transcripts. In order to better understand how miRNAs contribute to human-specialized brain processes and neurological phenotypes, identifying their targets is of paramount importance. Here, we address the latter by profiling Ago2:RNA interactions using HITS-CLIP to generate a transcriptome-wide map of miRNA binding sites in human brain. We uncovered ∼ 7,000 stringent Ago2 binding sites that are highly enriched for conserved sequences corresponding to abundant brain miRNAs. This interactome points to functional miRNA:target pairs across >3,000 genes and represents a valuable resource for accelerating our understanding of miRNA functions in brain. We demonstrate the utility of this map for exploring clinically relevant miRNA binding sites that may facilitate the translation of genetic studies of complex neuropsychiatric diseases into therapeutics.

RNA editing of the Q/R site of GluA2 in different cultured cell lines that constitutively express different levels of RNA editing enzyme ADAR2

Adenosine deaminase acting on RNA 2 (ADAR2) catalyzes RNA editing at the glutamine/arginine (Q/R) site of GluA2, and an ADAR2 deficiency may play a role in the death of motor neurons in ALS patients. The expression level of ADAR2 mRNA is a determinant of the editing activity at the GluA2 Q/R site in human brain but not in cultured cells. Therefore, we investigated the extent of Q/R site-editing in the GluA2 mRNA and pre-mRNA as well as the ADAR2 mRNA and GluA2 mRNA and pre-mRNA levels in various cultured cell lines. The extent of the GluA2 mRNA editing was 100% except in SH-SY5Y cells, which have a much lower level of ADAR2 than the other cell lines examined. The ADAR2 activity at the GluA2 pre-mRNA Q/R site correlated with the ADAR2 mRNA level relative to the GluA2 pre-mRNA. SH-SY5Y cells expressed higher level of the GluA2 mRNA in the cytoplasm compared with other cell lines. These results suggest that the ADAR2 expression level reflects editing activity at the GluA2 Q/R site and that although the edited GluA2 pre-mRNA is readily spliced, the unedited GluA2 pre-mRNA is also spliced and transported to the cytoplasm when ADAR2 expression is low.

Parametric mapping of 5HT 1A receptor sites in human brain: Evaluations of the HYPOTIME and MLSRTM methods

Parametric mapping of 5HT 1A receptor sites in human brain: Evaluations of the HYPOTIME and MLSRTM methods

Parametric and Regional Maps of Free Serotonin 5HT1A Receptor Sites in Human Brain as Function of Age in Healthy Humans

Serotonin 5HT(1A)-binding sites can be detected in living human brain with the positron-emitting antagonist [(11)C]WAY-100635. Previous measurements of the availability of [(11)C]WAY-binding sites in normal aging are equivocal, in part because of the greatly variable binding of this ligand. To test the null hypothesis that the binding potential (pB) of 5HT(1A) sites remains constant with age; 19 healthy volunteers aged 23-73 years (8 women, 11 men) underwent positron emission tomography. To determine pBs, we applied a novel tissue reference method of analysis, Estimation of Reversible Ligand Binding and Receptor Density (ERLiBIiRD) (Gjedde, 2003; Rosa-Neto et al, 2004), which extrapolates measures of specific binding to an estimated steady-state. We compared these estimates in the two age groups with results obtained with the conventional Logan Plot and Simplified Reference Tissue Method (SRTM) applied to both regions of interest-based as parametric analyses. The regional distribution of specific binding of free sites [(11)C]WAY-100635 was similar to that reported in previous studies, with the highest pBs in limbic structures and the raphé nuclei. Although the results of the three methods differed, pBs in the elderly subjects consistently were lower than those of young subjects. Thus, the correlation between pB and age applied to regions-of-interest revealed significant decline of pB at the rate of 3 or 4% per decade, and a 10% decline of the global mean 5HT(1A) receptor-pB in elderly relative to young subjects. The results demonstrate that the number of available 5HT(1A)-binding sites declines with age.

Contributions of protein phosphatases PP1, PP2A, PP2B and PP5 to the regulation of tau phosphorylation

Abnormal hyperphosphorylation of tau is believed to lead to neurofibrillary degeneration in Alzheimer's disease (AD) and other tauopathies. Recent studies have shown that protein phosphatases (PPs) PP1, PP2A, PP2B and PP5 dephosphorylate tau in vitro, but the exact role of each of these phosphatases in the regulation of site-specific phosphorylation of tau in the human brain was unknown. Hence, we investigated the contributions of these PPs to the regulation of tau phosphorylation quantitatively. We found that these four phosphatases all dephosphorylated tau at Ser199, Ser202, Thr205, Thr212, Ser214, Ser235, Ser262, Ser396, Ser404 and Ser409, but with different efficiencies toward different sites. The K(m) values of tau dephosphorylation catalysed by PP1, PP2A and PP5 were 8-12 microm, similar to the intraneuronal tau concentration of human brain, whereas the K(m) of PP2B was fivefold higher. PP2A, PP1, PP5 and PP2B accounted for approximately 71%, approximately 11%, approximately 10% and approximately 7%, respectively, of the total tau phosphatase activity of human brain. The total phosphatase activity and the activities of PP2A and PP5 toward tau were significantly decreased, whereas that of PP2B was increased in AD brain. PP2A activity negatively correlated to the level of tau phosphorylation at the most phosphorylation sites in human brains. Our findings indicate that PP2A is the major tau phosphatase that regulates its phosphorylation at multiple sites in human brain. The abnormal hyperphosphorylation of tau is partially due to a downregulation of PP2A activity in AD brain.

Evidence that RNA editing modulates splice site selection in the 5-HT2C receptor gene.

Adenosine to inosine editing of mRNA from the human 5-HT2C receptor gene (HTR2C) occurs at five exonic positions (A-E) in a stable stem-loop that includes the normal 5' splice site of intron 5 and is flanked by two alternative splice sites. Using in vitro editing, we identified a novel editing site (F) located in the intronic part of the stem-loop and demonstrated editing at this site in human brain. We have shown that in cell culture, base substitutions to mimic editing at different combinations of the six sites profoundly affect relative splicing at the normal and the upstream alternative splice site, but splicing at the downstream alternative splice site was consistently rare. Editing combinations in different splice variants from human brain were determined and are consistent with the effects of editing on splicing observed in cell culture. As RNA editing usually occurs close to exon/intron boundaries, this is likely to be a general phenomenon and suggests an important novel role for RNA editing.

Mt1 Melatonin receptor in the primate adrenal gland: inhibition of adrenocorticotropin-stimulated cortisol production by melatonin.

The pineal hormone melatonin participates in circadian, seasonal, and reproductive physiology. The presence of melatonin binding sites in human brain and peripheral tissues is well documented. However, in the mammalian adrenal gland, low-affinity melatonin binding sites have been detected only in the rat by some but not all authors. Conflicting evidence for a regulatory role of melatonin on adrenal cortisol production, prompted us to investigate this possibility in a New World primate, the capuchin monkey. Expression of melatonin receptors in the adrenal cortex was demonstrated through pharmacological characterization and autoradiographic localization of 2-[125I]iodomelatonin binding sites (dissociation constant = 96.9 +/- 15 pM; maximal binding capacity = 3.8 +/- 0.4 fmol/mg protein). The mt1 identity of these receptors was established by cDNA sequencing. Melatonin treatment of dispersed cells and explants from adrenal gland did not affect basal cortisol production. However, cortisol production stimulated by 100 nM ACTH was significantly inhibited by low melatonin concentrations (0.1-100 nM); this inhibitory effect was reversed by the mt1/MT2 melatonin antagonist luzindole. Melatonin also inhibited dibutyril-cAMP-stimulated cortisol production, suggesting that melatonin acts through a cAMP-independent signaling pathway. The present data demonstrate that the primate adrenal gland cortex expresses functional mt1 melatonin receptors and shows that melatonin inhibits ACTH-stimulated cortisol production.

Allosteric modulation of [3H]-CGP39653 binding through the glycine site of the NMDA receptor: further studies in rat and human brain.

Binding of D,L-(E)-2-amino-4-[(3)H]-propyl-5-phosphono-3-pentenoic acid ([(3)H]-CGP39653), a selective antagonist at the glutamate site of the NMDA receptor, is modulated by glycine in rat brain tissue. We have further investigated this phenomenon in rodent and human brain by means of receptor binding and quantitative autoradiography techniques. In rat cerebral cortical membranes the glycine antagonist 3-[2-(Phenylaminocarbonyl)ethenyl]-4,6-dichloro-indole-2-carboxylic acid sodium salt (GV150526A) did not change basal [(3)H]-CGP39653 binding, but competitively reversed the high affinity component of [(3)H]-CGP39653 binding inhibition by glycine, with a pK(B) value of 8.38, in line with its affinity for the glycine site (pK(i)=8.49 vs. [(3)H]-glycine). Glycine (10 microM) significantly decreased [(3)H]-CGP39653 affinity for the NMDA receptor (with no change in the B(max)), whereas enhanced L-glutamate affinity (P<0.05, paired-samples Student's t-test). In rat brain sections the addition of GV150526A (30 microM) to the incubation medium increased [(3)H]-CGP39653 binding to 208% of control (average between areas), indicating the presence of endogenous glycine. The enhancement presented significant regional differences (P<0.05, two-way ANOVA), with striatum higher than cerebral cortex (282 and 187% of control, respectively; P<0.05, Fisher's LSD). On the contrary, there was not any significant variation in affinity values of [(3)H]-CGP39653, L-glutamate, glycine and GV150526A in striatal and cortical membranes. These results confirmed the existence of regionally distinct NMDA receptors subtypes with different glycine/glutamate allosteric modulation. Whole brain autoradiography revealed an uneven distribution of [(3)H]-CGP39653 binding sites in human brain. High levels of binding were determined in hippocampus and in cingulate, frontoparietal and insular cortex. Intermediate to low levels of binding were found in diencephalic nuclei and basal ganglia. [(3)H]-CGP39653 binding was increased to 216% of control (mean between areas) by 30 microM GV150526A. The enhancement, however, did not present significant regional differences. These results introduce GV150526A as a useful tool to identify NMDA receptor subtypes by means of receptor autoradiography; moreover, they demonstrate that the allosteric inhibition of [(3)H]-CGP39653 binding by glycine parallels an increase in receptor affinity to the endogenous ligand L-glutamate. Finally, this study provides the first detailed anatomical description of the regional distribution of [(3)H]-CGP39653 binding sites in human brain.

Distribution of angiotensin IV binding sites (AT 4 receptor) in the human forebrain, midbrain and pons as visualised by in vitro receptor autoradiography

Angiotensin IV and other AT 4 receptor agonists, improve memory retention and retrieval in the passive avoidance and swim maze learning paradigms. Angiotensin IV binding sites (also known as the AT 4 receptors) are widely distributed in guinea pig and monkey ( Macaca fascicularis) brains where high densities of the binding sites have been detected in the hippocampus, neocortex and motor nuclei. However, the distribution of the binding sites in the human brain is not known. We have recently localised the angiotensin IV binding sites (AT 4 receptors) in post-mortem human brain using iodinated Nle-angiotensin IV, a higher affinity and more stable analogue of angiotensin IV. This radioligand bound with relatively high affinity and specificity to angiotensin IV binding sites. In competition studies on consecutive sections through the prefrontal cortex and claustrum, angiotensin IV, Nle-angiotensin IV and LVV-hemorphin 7 competed for the binding of 125I[Nle]-angiotensin IV with nanomolar affinities. Angiotensin II and the AT 1 and AT 2 receptor antagonists were ineffective in competing for the binding at concentrations of up to 10 μM. We found high densities of 125I[Nle]-angiotensin IV binding sites throughout the cerebral cortex including the insular, entorhinal, prefrontal and cingulate cortices. Very high densities of the binding sites were observed in the claustrum, choroid plexus, hippocampus and pontine nucleus. Some thalamic nuclei displayed high densities of binding including the anteroprincipal, ventroanterior, anteromedial, medial dorsal and ventrolateral nuclei. The caudate nucleus, putamen, many amygdaloid nuclei and the red nucleus all displayed moderate densities of binding with a higher level detected in the substantia nigra pars compacta. In the hypothalamus, high densities binding sites were found in the ventromedial nucleus with lower levels in the dorsomedial and paraventricular nuclei. The distribution of 125I[Nle]-angiotensin IV binding sites in the human brain is similar to that found in other species and supports multiple roles for the binding sites in the central nervous system, including facilitation of memory retention and retrieval.

Affinity of (+/-)-pindolol, (-)-penbutolol, and (-)-tertatolol for pre- and postsynaptic serotonin 5-HT(1A) receptors in human and rat brain.

There is considerable interest in the use of drugs that selectively block presynaptic (somatodendritic) serotonin 5-HT(1A) receptors for the adjunctive treatment of major depressive disorder. The 5-HT(1A)/beta-adrenoceptor ligands (+/-)-pindolol, (-)-tertatolol, and (-)-penbutolol are currently under clinical investigation, and knowledge of their affinity at different populations of central 5-HT(1A) receptors is needed. Here we have determined the affinity of these drugs for presynaptic and postsynaptic 5-HT(1A) receptors in postmortem human and rat brain using receptor autoradiography and the selective 5-HT(1A) radioligand [(3)H]WAY-100635. The binding of [(3)H]WAY-100635 was specific and saturable and showed high affinity in the rat dorsal raphe nucleus and hippocampus (K(D) = 1.5-1.7 nM). In competition studies, the three compounds had nanomolar affinity and produced monophasic displacement of [(3)H]WAY-100635 binding in all regions of both species. (-)-Penbutolol and (-)-tertatolol had similar affinity for pre-and postsynaptic 5-HT(1A) receptors in both rat and human brain. However, in the human, but not the rat, the affinity of (+/-)-pindolol in dorsal raphe nucleus (K(i) = 8.9 +/- 1. 1 nM) was slightly but significantly higher than that in hippocampus (K(i) = 14.4 +/- 1.5 nM in CA1). In summary, our data show that (+/-)-pindolol, (-)-tertatolol, and (-)-penbutolol are all high-affinity ligands at native human and rat 5-HT(1A) receptors. (-)-Penbutolol and (-)-tertatolol do not discriminate between the pre- and postsynaptic 5-HT(1A) sites tested in either species, but (+/-)-pindolol showed a slightly higher affinity for the presynaptic site in human brain. Further work is needed to establish whether the latter difference is clinically relevant.

I 2-Imidazoline receptors and monoamine oxidase B enzyme sites in human brain: covariation with age

An association between monoamino oxidase B enzyme (MAO-B) sites and I 2-imidazoline receptors (I 2-IR) has been established in human brain on the basis of correlational studies of radiolabelled binding sites. Because both MAO-B and I 2-IR densities increase with aging, the age at death could be partially involved in the observed correlations. The evaluation of two independent but similar datasets demonstrated that the linear correlation between MAO-B and I 2-IR densities dropped from 0.70 ( P=0.02) and 0.41 ( P=0.14) in the crude analyses to the negligible values of 0.07 ( P=0.84) and 0.09 ( P=0.75) when the age at death was controlled for in the statistical analyses. The results lead to conclude, contrary to former interpretations, that there is not any statistical association linking both MAO-B catalytic unit sites and I 2-IR densities in the human brain.

[formula omitted]]Tiagabine binding to GABA uptake sites in human brain

The binding of [ 3 H ]tiagabine (( RS-1-(4,4-(3-methyl-2-thienyl)-3-butenyl)-3-piperidine carboxylic acid) to homogenates of frozen post-mortem human brain has been characterized. Inhibition experiments with γ-aminobutyric acid (GABA), GABA uptake inhibitors, ligands active at postsynaptic GABA receptors and receptors for other neurotransmitters, suggest that [ 3 H ]tiagabine binds with high affinity to GABA uptake sites. Inhibition and kinetic experiments suggests that 70%–80% of the binding is to a high affinity site. Saturation experiments showed that the binding was saturable. B max was 3.4 pmol/mg protein and K d 16 nM in frontal cortex. The dissociation constants ( K d) measured in kinetic and equilibrium experiments were in the same range (16–56 nM). The regional distribution was studied in nine brain regions and the binding was heterogenous, with the highest binding in frontal cortex and parietal cortex and the lowest binding in nucleus caudatus and putamen. This is, to our knowledge, the first study on [ 3 H ]tiagabine binding in human tissue. It is concluded that [ 3 H ]tiagabine binding can be used as a specific marker for the GABA transporter GAT-1 in homogenates of human brain.

Distribution of transferrin and ferritin binding in normal and multiple sclerotic human brains.

Delivery of iron to the brain traditionally has been considered the responsibility of transferrin. However, transferrin receptors in brain are located primarily within gray matter areas rather than in the iron rich white matter tracts. In this report we present the first demonstration of ferritin binding sites in human brain and provide evidence that these binding sites are primarily in white matter tracts. This distribution of ferritin binding is opposite of that seen for the distribution of the transferrin receptor in normal adult human brain. Ferritin binds to human brain tissue in a competitive and saturable manner with a dissociation constant of 0.35 nM and a binding site density of 116.7 fmol/mg protein. In brain tissue from multiple sclerotic (MS) patients the normal pattern of transferrin and ferritin binding distributions is disrupted. Ferritin binding is absent in the lesion itself and in the immediate periplaque region within the white matter but returns to normal as the distance from the lesion becomes greater. In direct contrast to ferritin binding, transferrin binding in the MS tissue is present in the white matter tracts, but only in the periplaque region. The periplaque region also contains transferrin receptor positive cells (as determined by immunocytochemistry) morphologically consistent with oligodendrocytes. Gray matter binding of transferrin in MS patients appears normal. These data provide the initial evidence of ferritin binding in human brain, address the enigma of the apparent absence of an iron delivery system to the iron-rich white matter, and suggest loss of ferritin binding is involved in or is a consequence of demyelination associated with MS.

Autoradiographic characterization, anatomical distribution, and developmental pattern of a new 5-HT site in human brain.

Autoradiographic characterization, anatomical distribution, and developmental pattern of a new 5-HT site in human brain.

Characterization of the 5-HT4 binding site in human brain.

In this study, drug inhibition and saturation experiments on the binding of the highly selective 5-HT4 antagonist [3H]GR 113808 were performed in human brain membranes so as to better characterize this binding site. Drug competition studies were carried out by incubating 0.2 nM [3H]GR 113808 in the presence of increasing concentrations of six different drugs, i.e. 5-HT, 5-CT, ondansetron, tropisetron, BIMU 1 and BIMU 8 (mixed 5-HT3 and 5-HT4 agonists). The binding displaced by 5-HT showed a drug inhibition constant (Ki) value of 197nM. The use of 5-CT or ondansetron also showed the existence of single-site models albeit with Ki values in the micromolar range (11,5 microM). Tropisetron, BIMU 1 and BIMU 8 displaced bound [3H]GR 113808 according to a two-site binding model, with the high affinity component in the nanomolar range and the low affinity site in the micro or milimolar range. Saturation experiments revealed high binding densities in basal ganglia (187 fmol/mg in putamen, and 149 fmol/mg in caudate nucleus), while lower densities were observed in cortical regions (49 fmol/mg in temporal cortex, 45 fmol/mg in parietal cortex and 71 fmol/mg in cingulate cortex). The apparent affinity (Kd) was similar in the brain regions studied, ranging from 0.13 to 0.34 nmol/l. Despite the enrichment of 5-HT receptors in human brain, their functional correlate in brain diseases remains to be clarified.

Autoradiographic characterization of [3H]inositol (1,4,5) trisphosphate and [3H]inositol (1,3,4,5) tetrakisphosphate binding sites in human brain.

Autoradiographic techniques were used to investigate the characteristics of tritiated inositol(1,4,5)trisphosphate ([3H]IP3) and inositol (1,3,4,5) tetrakisphosphate ([3H]IP4) binding to human brain. In brain sections [3H]IP3 exhibited a two-site binding with KD values of 87 nM and 9.3 microM respectively for the higher and lower affinity sites. [3H]IP4 also bound to two sites with KD values of 43 nM and 1.4 microM, respectively. With the conditions fixed in this study, [3H]IP3 and [3H]IP4 autoradiography in the cortex, caudate, hippocampus and cerebellum were performed. The most prominent [3H]IP3 binding among these regions was found in the cerebellum, particularly in the molecular layer. Within the hippocampus, the subiculum and the CA1 region showed much more prominent binding than the other subfields. [3H]IP4, binding was fairly homogeneous in the regions studied, with the exception of a slightly higher binding in the molecular layer of the cerebellum.

Differential psychopathology and patterns of cerebral glucose utilisation produced by ( S)- and ( R)-ketamine in healthy volunteers using positron emission tomography (PET)

Until recently, racemic ketamine ( S-ketamine/ R-ketamine=50:50) has been used to study NMDA receptor hypofunction in relation to pathophysiological models of schizophrenia. Ketamine given to normal humans in subanesthetic doses produces a model psychosis including both positive and negative symptoms of schizophrenia. More recently it has been shown that at subanesthetic doses the pure ( S)- and ( R)-ketamine enantiomeres interact differently with the NMDA and sigma receptor sites in human brain. It was found that ( S)-ketamine binds with a 3–4 time higher affinity to the PCP binding site of the NMDA receptor than ( R)-ketamine, and that at these concentrations ( R)-ketamine interacts also weakly with the sigma receptor sites, where ( S)-ketamine binds only negligibly. To further investigate the role of NMDA-receptor mediated neurotransmission in schizophrenic psychosis, the effects of pure ( S)- and ( R)-ketamine enantiomeres on brain energy metabolism in normal humans using positron emission tomography and [ 18F]fluorodeoxyglucose (FDG) are reported here. Psychotomimetic doses of ( S)-ketamine increased cerebral metabolic rates of glucose (CMRglu) markedly in the frontal cortex including the anterior cingulate, parietal and left sensorimotor cortex, and in the thalamus. The metabolic changes in the frontal and left temporal cortex correlated with ego-disintegration and hallucinatory phenomena. Equimolar doses of ( R)-ketamine tended to decrease CMRglu across brain regions and significantly suppressed CMRglu in the temporomedial cortex and left insula. ( R)-ketamine did not produce psychotic symptoms, but a state of relaxation. The ( S)-ketamine-induced metabolic hyperfrontality appears to parallel similar metabolic findings in acute psychotic schizophrenic patients and encourages further investigations of glutamatergic disturbances in schizophrenia.

Genomic Organization and the 5′ Flanking Region of the γ Subunit of the Human Amiloride-sensitive Epithelial Sodium Channel

The amiloride-sensitive epithelial sodium channel (ENaC) complex is made up of at least three different subunits alpha, beta, and gamma, which are developmentally regulated, selectively expressed, and variously up-regulated by steroid hormones. To understand mechanisms involved in regulation of the gamma subunit, we have determined the structure of the human gammaENaC gene. By 5' rapid amplification of cDNA ends, primer extension analysis, and nuclease protection assay, we identified transcription start sites in human brain, kidney, and lung. A human genomic library was screened and overlapping cosmid clones that span approximately 50 kilobases and contain the hgammaENaC gene were identified. The 5'-untranslated region is 141 bases long, and the translation start codon is contained within the second exon. The human gene spans at least 35 kilobases. The 5' end of the gene including portions of 5' flanking genomic DNA and the first intron are G + C rich and contain several CpG dinucleotides, consistent with a CpG island. The 5' flanking region contains no CCAAT or TATA-like elements but does contain two GC boxes as well as several putative transcription factor binding sites including AP-2, Sp1, CRE, PEA-3, and NF-IL6. This is the first description of the structural organization and the 5' flanking region of a member of the epithelial sodium channel complex.