Issue published March 1, 1997

T o identify receptor-associated proteins that may contribute to the specificity of insulin and IGF-I signaling responses, a mouse embryo library was screened using the yeast two-hybrid system. Multiple receptor-interactive clones encoding the SH2 domain of the adapter protein Grb10 were isolated. Subsequent cloning of the full-length Grb10 sequence from a mouse fat cDNA library defined a previously unknown Grb10 variant, that appears to be the predominant isoform in mouse tissues. Receptor-deficient R- cells (fibroblasts from mice with homologous disruption of the IGF-I receptor gene) and transfected R- cells expressing either insulin receptors (R-IR cells) or IGF-I receptors (R+ cells) were used to investigate the specificity of Grb10 interaction with the two related receptors. Hormone-activated insulin receptors in R-IR cells coprecipitated with three species, all recognized as Grb10 isoforms by specific Grb10 antibody. Under the same conditions, Grb10 was essentially undetectable in IGF-I receptor immunoprecipitates from stimulated R+ cells. Grb10 association with insulin receptors was maximal at 10 nM insulin stimulation and sustained from 5-10 min after hormone stimulation in R-IR cells. In conclusion, Grb10 interacts preferentially with insulin vs. IGF-I receptors in intact cells and, thus, may have a role in mediating insulin receptor-specific cellular responses.

T he 5' end of the mRNA-encoding sterol regulatory element binding protein-1 (SREBP-1) exists in two forms, designated 1a and 1c. The divergence results from the use of two transcription start sites that produce two separate 5' exons, each of which is spliced to a common exon 2. Here we show that the ratio of SREBP-1c to 1a transcripts varies markedly among organs of the adult mouse. At one extreme is the liver, in which the 1c transcript predominates by a 9:1 ratio. High 1c:1a ratios are also found in mouse adrenal gland and adipose tissue and in human liver and adrenal gland. At the other extreme is the spleen, which shows a reversed 1c:1a ratio (1:10). In five different lines of cultured cells, including the HepG2 line derived from human hepatocytes, the 1a transcript predominated (1c:1a ratio < 1:2). In mouse 3T3-L1 preadipocytes, the 1a transcript was present, but the 1c transcript was not detectable. When these cells were differentiated into adipocytes by hormone treatment in culture, the amount of 1a transcript rose markedly (8.2-fold), and the 1c transcript remained virtually undetectable. We conclude that the SREBP-1a and 1c transcripts are controlled independently by regulatory regions that respond differentially to organ-specific and metabolic factors.

W e have produced transgenic mice whose livers express a dominant positive NH2-terminal fragment of sterol regulatory element binding protein-1c (SREBP-1c). Unlike full-length SREBP-1c, the NH2-terminal fragment enters the nucleus without a requirement for proteolytic release from cell membranes, and hence it is immune to downregulation by sterols. We compared SREBP-1c transgenic mice with a line of transgenic mice that produces an equal amount of the NH2-terminal fragment of SREBP-1a. SREBP-1a and -1c are alternate transcripts from a single gene that differ in the first exon, which encodes part of an acidic activation domain. The 1a protein contains a long activation domain with 12 negatively charged amino acids, whereas the 1c protein contains a short activation domain with only 6 such amino acids. As previously reported, livers of the SREBP-1a transgenic mice were massively enlarged, owing to accumulation of triglycerides and cholesterol. SREBP-1c transgenic livers were only slightly enlarged with only a moderate increase in triglycerides, but not cholesterol. The mRNAs for the LDL receptor and several cholesterol biosynthetic enzymes were elevated in SREBP-la transgenic mice, but not in 1c transgenic mice. The mRNAs for fatty acid synthase and acetyl CoA carboxylase were elevated 9- and 16-fold in la animals, but only 2- and 4-fold in 1c animals. Experiments with transfected cells confirmed that SREBP-1c is a much weaker activator of transcription than SREBP-1a when both are expressed at levels approximating those found in nontransfected cells. SREBP-1c became a strong activator only when expressed at supraphysiologic levels. We conclude that SREBP-1a is the most active form of SREBP-1 and that SREBP-1c may be produced when cells require a lower rate of transcription of genes regulating cholesterol and fatty acid metabolism.

W ild-type (Agt+/+) and homozygous angiotensinogen deletion mutant (Agt-/-) littermates were placed on normal (NS) or low Na diet (LS) for 2 weeks. Plasma aldosterone levels (P(aldo)) were comparable during NS, and similarly elevated during LS in Agt+/+ and Agt-/-. Moreover, in both, the elevation in P(aldo) was accompanied by marked increase in adrenal zona glomerulosa cells and adrenal P450aldo mRNA. Agt-/- mice were distinguished from Agt+/+ mice by their higher plasma K level, by approximately 1.5 and approximately 3.8 mEq/liter during NS and LS, respectively. Within the Agt-/- group, P(aldo) was directly proportional to plasma K. The importance of K for the hyperaldosteronism during dietary Na restriction was verified by the observation that superimposition of K restriction led to hypotension in Agt+/+ and uniform death in Agt-/- mice along with a reduction in P(aldo) by 75 and 90%, respectively. Thus, suppression of potassium, but not angiotensin, led to a marked attenuation of hyperaldosteronism during dietary Na restriction. Therefore, (a) a powerful angiotensin-independent mechanism exists for the hyperaldosteronism during LS; (b) high K is a central component of this mechanism; (c) contrary to current belief, the tonic effect of high K on aldosterone synthesis and release does not require an intact renin-angiotensin system; and (d) normally, intermediary feedback signals for hyperaldosteronism, i.e., both hypotension and high K, are effectively masked by aldosterone actions.

N on-insulin-dependent diabetes mellitus (NIDDM) is considered a polygenic disorder in which insulin resistance and insulin secretory defect are the major etiologic factors. Homozygous mice with insulin receptor substrate-1 (IRS-1) gene knockout showed normal glucose tolerance associated with insulin resistance and compensatory hyperinsulinemia. Heterozygous mice with beta cell glucokinase (GK) gene knockout showed impaired glucose tolerance due to decreased insulin secretion to glucose. To elucidate the interplay between insulin resistance and insulin secretory defect for the development of NIDDM, we generated double knockout mice with disruption of IRS-1 and beta cell GK genes by crossing the mice with each of the single gene knockout. The double knockout mice developed overt diabetes. Blood glucose levels 120 min after intraperitoneal glucose load (1.5 mg/g body wt) were 108 +/- 24 (wild type), 95 +/- 26 (IRS-1 knockout), 159 +/- 68 (GK knockout), and 210 +/- 38 (double knockout) mg/dl (mean +/- SD) (double versus wild type, IRS-1, or GK; P < 0.01). The double knockout mice showed fasting hyperinsulinemia and selective hyperplasia of the beta cells as the IRS-1 knockout mice (fasting insulin levels: 0.38 +/- 0.30 [double knockout], 0.35 +/- 0.27 [IRS-1 knockout] versus 0.25 +/- 0.12 [wild type] ng/ml) (proportion of areas of insulin-positive cells to the pancreas: 1.18 +/- 0.68%; P < 0.01 [double knockout], 1.20 +/- 0.93%; P < 0.05 [IRS-1 knockout] versus 0.54 +/- 0.26% [wild type]), but impaired insulin secretion to glucose (the ratio of increment of insulin to that of glucose during the first 30 min after load: 31 [double knockout] versus 163 [wild type] or 183 [IRS-1 knockout] ng insulin/mg glucose x 10(3)). In conclusion, the genetic abnormalities, each of which is nondiabetogenic by itself, cause overt diabetes if they coexist. This report provides the first genetic reconstitution of NIDDM as a polygenic disorder in mice.

P revious studies have indicated that nitric oxide (NO) released from Kupffer cells modulates biological viability of cocultured hepatoma cells. This study was designed to evaluate the mechanisms by which Kupffer cells synthesize and release NO in reponse to cocultured hepatoma cells. Kupffer cells isolated from male Wistar rats were cocultured with rat hepatoma cell line, AH70 cells. The sum of nitrite and nitrate levels increased in the culture medium of Kupffer cells with AH70 cells as compared with those of Kupffer cells or AH70 cells alone. Increased expressions of iNOS and iNOS mRNA in Kupffer cells cocultured with AH70 cells were detected by an immunofluorescence staining and a fluorescence in situ hybridization study, respectively. A fluorescence in situ DNA-protein binding assay revealed that NF-kappaB activation occurs in Kupffer cells and activated NF-kappaB moved into the nuclei preceding to an increased production of NO. Oxidative stress indicated by dichlorofluorescein fluorescence was observed in Kupffer cells cocultured with AH70 cells. An increased calcium mobilization indicated as increased fluo-3-associated fluorescence was also induced in Kupffer cells after coculture with AH70 cells. Monoclonal antibodies directed against rat CD18 and ICAM-1, as well as TMB-8, a calcium inhibitor, prevented the calcium mobilization, active oxygen production, and NF-kappaB activation in addition to the increased production of NO. Pyrrolidine dithiocarbamate, an inhibitor of oxidative NF-kappaB activation, diphenylene iodonium, an NADPH oxidase inhibitor, and quinacrine, a phospholipase A2 inhibitor, significantly attenuated the increase in dichlorofluorescein fluorescence, NF-kappaB activation, and NO production. Therefore, this study suggests that CD18/ICAM-1-dependent cell-to-cell interaction with hepatoma cells causes calcium mobilization and oxidative activation of NF-kappaB, which may lead to the increased production of NO in Kupffer cells.

C GP 51901 is a non-anaphylactogenic mouse/human chimeric anti-human IgE antibody that binds to free IgE and surface IgE of IgE-expressing B cells but not to IgE bound to high affinity IgE receptors (Fc epsilonR1) on mast cells and basophils or low affinity IgE receptors (Fc epsilonR2) on other cells. A phase 1 double-blind, placebo-controlled, single dose study with doses of 3, 10, 30, and 100 mg of CGP 51901 was conducted in 33 pollen-sensitive subjects who had raised levels of serum IgE and received either intravenous CGP 51901 or placebo. The administration of CGP 51901 was well tolerated and resulted in a decrease of serum free IgE levels in a dose-dependent manner, with suppression after 100 mg of CGP 51901 reaching > 96%. Time of recovery to 50% of baseline IgE correlated with the dose of administered antibody and ranged from a mean of 1.3 d for the 3 mg to 39 d for the 100 mg dose. Total IgE, comprised of free and complexed IgE, increased as stored and newly synthesized IgE bound to CGP 51901. Complexed IgE was eliminated at a rate comparable with the terminal half-life of free CGP 51901 (11-13 d at all doses). Only one subject showed a weak antibody response against CGP 51901. We conclude that the use of anti-human IgE antibody is safe and effective in reducing serum IgE levels in atopic individuals and provides a potential therapeutic approach to the treatment of atopic diseases.

O xidative modification of low density lipoprotein has been suggested as patho-physiologically relevant process in atherogenesis and the lipid peroxidizing enzyme 15-lipoxygenase may be involved. For experimental evidence on the in vivo action of this enzyme in the time course of plaque formation we analyzed the lipid extracts of lesional areas representing various stages of human atherogenesis for the occurrence of specific 15-lipoxygenase products. In advanced human lesions the degree of oxygenation of the lesion lipids measured as hydroxy linoleic acid/linoleic acid ratio varied between 0.2 and 3.2%. Here an unspecific pattern of oxygenated lipids that did not differ from the pattern formed during copper-catalyzed LDL oxidation was detected. In both cases an enantiomer ratio (S/R-ratio) of 13-hydroxy-9Z,11E-octadecadienoic acid (13-HODE) of approximately 1:1 was found. In young human lesions which were obtained from the collection of the pathological determinants of atherosclerosis in youth (PDAY) program the hydroxy linoleic acid/linoleic acid ratio was much smaller (variation between 0.05 and 0.6%), and a significant share of specific 15-lipoxygenase products was detected (S/R-ratio of 13-hydroxy linoleic acid of 54 +/- 3.1/46 +/- 3.1 [mean +/- SD]). These data suggest that the 15-lipoxygenase is enzymatically active on endogenous substrates in young human lesions and thus, may be of patho-physiological importance for early atherogenesis. In advanced human plaques the 15-lipoxygenase may be functionally silent and specific lipoxygenase products formed in earlier stages may be decomposed or superimposed by large amounts of nonenzymatic lipid peroxidation products.

S ecretory leukocyte protease inhibitor (SLPI) is a serine protease inhibitor found in fluids lining mucosal surfaces. In addition to its primary function as an antiprotease, SLPI may also influence cellular functions associated with enzyme synthesis and retroviral infection. In this study, SLPI was examined for its effect on signaling events involved in the production of matrix metalloproteinases (MMPs) by monocytes. Addition of SLPI before stimulation with concanavalin A or LPS resulted in a significant inhibition of monocyte prostaglandin H synthase-2 (PGHS-2), a pivotal enzyme in the PGE2-cAMP dependent pathway of monocyte MMP synthesis. Suppression of PGHS-2 was detected with 0.1 microg/ml of SLPI with a substantial inhibition at 1 and 10 micro/ml. Attenuation of PGHS-2 by SLPI was accompanied by decreased production of PGE2 resulting in the suppression of interstitial collagenase (MMP-1) and gelatinase B (MMP-9) that was reversed by PGE2 or Bt2cAMP. The inhibitory effect of SLPI was largely independent of its antiprotease activity because SLPI muteins, with significantly lower antiprotease activity, also suppressed the induction of PGHS-2 and MMPs. The inhibitory effects of SLPI did not involve the modulation of monokine production since TNF-alpha and IL-10 were unaffected. These findings demonstrate that SLPI also functions as a potent antiinflammatory agent by interfering with the signal transduction pathway leading to monocyte MMP production.

W e attempted to elicit active anaphylaxis to ovalbumin, or passive IgE- or IgG1-dependent anaphylaxis, in mice lacking either the Fc epsilonRI alpha chain or the FcR gamma chain common to Fc epsilonRI and Fc gammaRI/III, or in mice lacking mast cells (KitW/ KitW-v mice), and compared the responses to those in the corresponding wild-type mice. We found that the FcR gamma chain is required for the death, as well as for most of the pathophysiological changes, associated with active anaphylaxis or IgE- or IgG1-dependent passive anaphylaxis. Moreover, some of the physiological changes associated with either active, or IgG1-dependent passive, anaphylactic responses were significantly greater in Fc epsilonRI alpha chain -/- mice than in the corresponding normal mice. Finally, while both KitW/KitW-v and congenic +/+ mice exhibited fatal active anaphylaxis, mast cell-deficient mice exhibited weaker physiological responses than the corresponding wild-type mice in both active and IgG1-dependent passive systemic anaphylaxis. Our findings strongly suggest that while IgE antibodies and Fc epsilonRI may influence the intensity and/or kinetics of some of the pathophysiological changes associated with active anaphylaxis in the mouse, the mortality associated with this response can be mediated largely by IgG1 antibodies and Fc gammaRIII.

I n mouse mast cells, both Fc epsilonRI and Fc gammaRIII are alpha beta gamma2 tetrameric complexes in which different alpha chains confer IgE or IgG ligand recognition while the signaling FcR beta and gamma chains are identical. We used primarily noninvasive techniques (changes in body temperature, dye extravasation) to assess systemic anaphylactic responses in nonanesthetized wild-type, Fc epsilonRI alpha chain -/- and FcR gamma chain -/- mice. We confirm that systemic anaphylaxis in mice can be mediated largely through IgG1 and Fc gammaRIII and we provide direct evidence that these responses reflect activation of Fc gammaRIII rather than Fc gammaRI. Furthermore, we show that Fc gammaRIII-dependent responses are more intense in normal than in congenic mast cell-deficient KitW/KitW-v mice, indicating that Fc gammaRIII responses have mast cell-dependent and -independent components. Finally, we demonstrate that the upregulation of cell surface expression of Fc gammaRIII seen in Fc epsilonRI alpha chain -/- mice corresponds to an increased association of Fc gammaRIII alpha chains with FcR beta and gamma chains and is associated with enhanced Fc gammaRIII-dependent mast cell degranulation and systemic anaphylactic responses. Therefore, the phenotype of the Fc epsilonRI alpha chain -/- mice suggests that expression of Fc epsilonRI and Fc gammaRIII is limited by availability of the FcR beta and gamma chains and that, in normal mice, changes in the expression of one receptor (Fc epsilonRI) may influence the expression of functional responses dependent on the other (Fc gammaRIII).

M onocyte chemotactic protein-4 (MCP-4) is a newly identified C-C chemokine with potent eosinophil chemoattractant properties. We describe studies of its biological activity in vitro to induce chemotaxis of peripheral blood eosinophils and to induce histamine release from IL-3-primed peripheral blood basophils. MCP-4 and eotaxin caused a similar rise in eosinophil intracytoplasmic Ca2+ and complete cross-desensitization. MCP-4 also abolished the eosinophil Ca2+ response to MCP-3 and partially desensitized the response to macrophage inflammatory protein-1alpha. MCP-4 activated cell migration via either CCR2b or CCR3 in mouse lymphoma cells transfected with these chemokine receptors. MCP-4 inhibited binding of 125I-eotaxin to eosinophils and CCR3-transfected cells and inhibited 125I-MCP-1 binding to CCR2b-transfectants. MCP-4 mRNA was found in cells collected in bronchoalveolar lavage of asthmatic and nonasthmatic subjects and was prominently expressed in human lung and heart. MCP-4 mRNA was expressed in several human bronchial epithelial cell lines after cytokine stimulation. Pretreatment of BEAS-2B epithelial cells with the glucocorticoid budesonide inhibited MCP-4 mRNA expression. These features make MCP-4 a candidate for playing a role in eosinophil recruitment during allergic respiratory diseases.

R esting lymphocyte survival is dependent upon the expression of Bcl-2, yet the factors responsible for maintaining lymphocyte Bcl-2 protein expression in vivo are largely unknown. Natural killer (NK) cells are bone marrow-derived lymphocytes that constitutively express the beta and common gamma(c) subunits of the IL-2 receptor (R) as a heterodimer with intermediate affinity for IL-2. IL-15 also binds to IL-2Rbeta gamma(c) and is much more abundant in normal tissues than IL-2. Mice that lack the IL-2 gene have NK cells, whereas mice and humans that lack IL-2R gamma(c) do not have NK cells. Further, treatment of mice with an antibody directed against IL-2Rbeta results in a loss of the NK cell compartment. These data suggest that a cytokine other than IL-2, which binds to IL-2Rbeta gamma(c), is important for NK cell development and survival in vivo. In the current report, we show that the recently described IL-15R(alpha) subunit cooperates with IL-2Rbeta gamma(c) to transduce an intracellular signal at picomolar concentrations of IL-15. We demonstrate that resting human NK cells express IL-15R(alpha) mRNA and further, that picomolar amounts of IL-15 can sustain NK cell survival for up to 8 d in the absence of serum. NK cell survival was not sustained by other monocyte-derived factors (i.e., TNF-alpha, IL-1beta, IL-10, IL-12) nor by cytokines known to use gamma(c) for signaling (i.e., IL-4, IL-7, IL-9, IL- 13). One mechanism by which IL-15 promotes NK cell survival may involve the maintenance of Bcl-2 protein expression. Considering these functional properties of IL-15 and the fact that it is produced by bone marrow stromal cells and activated monocytes, we propose that IL-15 may function as an NK cell survival factor in vivo.

W e examined the effect of ventilation strategy on lung inflammatory mediators in the presence and absence of a preexisting inflammatory stimulus. 55 Sprague-Dawley rats were randomized to either intravenous saline or lipopolysaccharide (LPS). After 50 min of spontaneous respiration, the lungs were excised and randomized to 2 h of ventilation with one of four strategies: (a) control (C), tidal volume (Vt) = 7 cc/kg, positive end expiratory pressure (PEEP) = 3 cm H2O; (b) moderate volume, high PEEP (MVHP), Vt = 15 cc/kg; PEEP = 10 cm H2O; (c) moderate volume, zero PEEP (MVZP), Vt = 15 cc/kg, PEEP = 0; or (d) high volume, zero PEEP (HVZP), Vt = 40 cc/kg, PEEP = 0. Ventilation with zero PEEP (MVZP, HVZP) resulted in significant reductions in lung compliance. Lung lavage levels of TNFalpha, IL-1beta, IL-6, IL-10, MIP-2, and IFNgamma were measured by ELISA. Zero PEEP in combination with high volume ventilation (HVZP) had a synergistic effect on cytokine levels (e.g., 56-fold increase of TNFalpha versus controls). Identical end inspiratory lung distention with PEEP (MVHP) resulted in only a three-fold increase in TNFalpha, whereas MVZP produced a six-fold increase in lavage TNFalpha. Northern blot analysis revealed a similar pattern (C, MVHP < MVZP < HVZP) for induction of c-fos mRNA. These data support the concept that mechanical ventilation can have a significant influence on the inflammatory/anti-inflammatory milieu of the lung, and thus may play a role in initiating or propagating a local, and possibly systemic inflammatory response.

F amilial combined hyperlipidemia (FCH) is a common genetic lipid disorder in Western societies. In a recent report (Dallinga-Thie, G.M., X.D. Bu, M. van Linde-Sibenius Trip, J.I. Rotter, A.J. Lusis, and T.W.A. de Bruin. J. Lipid Res., 1996, 36:136-147) we have studied three restriction enzyme polymorphisms: XmnI, and MspI sites 5' of the apo AI gene and SstI site in the 3' untranslated region of exon 4 of the apo CIII gene in 18 FCH pedigrees, including 18 probands, 178 hyperlipidemic relatives, 210 normolipidemic relatives, and 176 spouses. DNA variations in the apo AI-CIII-AIV gene cluster had a modifying effect on plasma triglycerides, LDL cholesterol, and apolipoprotein CIII levels. In this study, combinations of haplotypes were analyzed to further characterize their interactions and effect on the expression of severe hyperlipidemia in FCH subjects. A specific combination of haplotypes with one chromosome carrying the X1M1S2 (1-1-2) haplotype and the other the X2M2S1 haplotype (2-2-1) was significantly more frequent in hyperlipidemic relatives (6%) than in normolipidemic relatives (3%) and spouses (0.5%). Associated with this combination of haplotypes were significantly elevated plasma cholesterol (P < 0.0001), triglycerides (P < 0.0001), and apo CIII (P < 0.001) levels when compared to the wild type combination of haplotypes 1-1-1/1-1-1. The only spouse with this specific combination of haplotypes showed a severe hyperlipidemic phenotype, similar to FCH. Furthermore, nonparametric sibpair linkage analysis revealed significant linkage between these markers in the gene cluster and the FCH phenotype (MspI P = 0.0088, SstI P = 0.044, and XMS haplotype P = 0.037). The present findings confirm that the apo AI-CIII-IV gene cluster contributes to the FCH phenotype, but this contribution is genetically complex. An epistatic interaction between different haplotypes of the gene cluster was demonstrated. The S2 allele on one haplotype was synergistic to the X2M2 allele on the other haplotype in its hyperlipidemic effect. Therefore, two different susceptibility loci exist in the gene cluster, demonstrating the paradigm of complex genetic contribution to FCH.

S tressful experience during early brain development has been shown to produce profound alterations in several mechanisms of adaptation, while several signs of behavioral and neuroendocrine impairment resulting from neonatal exposure to stress resemble symptoms of dysregulation associated with major depression. This study demonstrates that when applied concomitantly with the stressful challenge, the steroid GABA(A) receptor agonist 3,21-dihydropregnan-20-one (tetrahydrodeoxycorticosterone, THDOC) can attenuate the behavioral and neuroendocrine consequences of repeated maternal separation during early life, e.g., increased anxiety, an exaggerated adrenocortical secretory response to stress, impaired responsiveness to glucocorticoid feedback, and altered transcription of the genes encoding corticotropin-releasing hormone (CRH) in the hypothalamus and glucocorticoid receptors in the hippocampus. These data indicate that neuroactive steroid derivatives with GABA-agonistic properties may exert persisting stress-protective effects in the developing brain, and may form the basis for therapeutic agents which have the potential to prevent mental disorders resulting from adverse experience during neonatal life.

T he annual urinary screening of Japanese children above 3 yr of age has identified a progressive proximal renal tubular disorder characterized by low molecular weight proteinuria, hypercalciuria, and nephrocalcinosis. The disorder, which has a familial predisposition and occurs predominantly in males, has similarities to three X-linked proximal renal tubular disorders that are due to mutations in the renal chloride channel gene, CLCN5. We have investigated four unrelated Japanese kindreds with this tubulopathy and have identified four different CLCN5 mutations (two nonsense, one missense, and one frameshift). These are predicted to lead to a loss of chloride channel function, and heterologous expression of the missense CLCN5 mutation in Xenopus oocytes demonstrated a 70% reduction in channel activity when compared with the wild-type. In addition, single-stranded conformation polymorphism (SSCP) analysis was found to be a sensitive and specific mutational screening method that detected > 75% of CLCN5 mutations. Thus, the results of our study expand the spectrum of clinical phenotypes associated with CLCN5 mutations to include this proximal renal tubular disorder of Japanese children. In addition, the mutational screening of CLCN5 by SSCP will help to supplement the clinical evaluation of the annual urinary screening program for this disorder.

S timulation of human neutrophils with chemoattractants FMLP or platelet activating factor (PAF) results in different but overlapping functional responses. We questioned whether these differences might reflect patterns of intracellular signal transduction. Stimulation with either PAF or FMLP resulted in equivalent phosphorylation and activation of the mitogen-activated protein kinase (MAPk) homologue 38-kD murine MAP kinase homologous to HOG-1 (p38) MAPk. Neither FMLP nor PAF activated c-jun NH2-terminal MAPk (JNKs). Under identical conditions, FMLP but not PAF, resulted in significant p42/44 (ERK) MAPk activation. Both FMLP and PAF activated MAP kinase kinase-3 (MKK3), a known activator of p38 MAPk. Both MAP ERK kinase kinase-1 (MEKK1) and Raf are activated strongly by FMLP, but minimally by PAF. Pertussis toxin blocked FMLP-induced activation of the p42/44 (ERK) MAPk cascade, but not that of p38 MAPk. A specific p38 MAPk inhibitor (SK&F 86002) blocked superoxide anion production in response to FMLP and reduced adhesion and chemotaxis in response to PAF or FMLP. These results demonstrate distinct patterns of intracellular signaling for two chemoattractants and suggest that selective activation of intracellular signaling cascades may underlie different patterns of functional responses.

T he temporal synthesis of the P21 protein of Borrelia burgdorferi and the development of the humoral response to this antigen was assessed in infected mice. p21 is a member of the ospE-F gene family and its protein, P21, has been shown to be expressed by B. burgdorferi within infected mice but not by spirochetes cultured in vitro. P21 was not detected on B. burgdorferi in unfed or engorged Ixodes dammini (also known as I. scapularis) ticks, further supporting the postulate that P21 synthesis is specific for the mammalian host. In B. burgdorferi-infected mice, ospE mRNA and OspE antibodies were observed at 7 d, whereas p21 mRNA and P21-specific antibodies were detected at 21-28 d, suggesting that p21 is expressed later than ospE. Moreover, ospA mRNA was not discernible until day 14, indicating that ospA, like p21, is not expressed in the early stages of tick-transmitted murine Lyme borreliosis. Because p21 is expressed during infection in mice, we assessed the human humoral response to P21. 28% (34 of 122) of the patients with either early- or late-stage Lyme disease, and 33% (11 of 33) of the individuals with Lyme arthritis had P21 antibodies, suggesting that a P21 response may serve, at least partially, as a marker of infection. Active immunization with recombinant P21 did not protect C3H mice from tick-borne B. burgdorferi infection, and passive transfer of P21 antiserum to infected mice did not alter the course of disease. These data suggest that the antigenic structure of B. burgdorferi changes during the early stages of murine infection.

D ystrophic mineralization remains the leading cause of stenotic or regurgitant failure in native human and porcine bioprosthetic heart valves. We hypothesized that cellular expression of noncollagenous matrix proteins (osteopontin, osteocalcin, and osteonectin) that regulate skeletal mineralization may orchestrate valvular calcification. Porcine bioprosthetic heart valves and native human heart valves obtained during replacement surgery were analyzed for cells, matrix proteins that regulate mineralization, and vessels. Cell accumulation and calcification were correlated for both valve types (rho = 0.75, P = 0.01, native; rho = 0.42, P = 0.08, bioprosthetic). Osteopontin expression correlated with cell accumulation (rho = 0.58, P = 0.04) and calcification (rho = 0.52, P = 0.06) for bioprosthetic valves. Osteocalcin expression correlated with calcification (rho = 0.77, P = 0.04) and cell accumulation (rho = 0.69, P = 0.07) in native valves. Comparisons of calcified versus noncalcified native and bioprosthetic valves for averaged total matrix protein mRNA signal score revealed increased noncollagenous proteins mRNA levels in calcified valves (P = 0.07, group I vs. group II; P = 0.02, group III vs. group IV). When stratified according to positive versus negative mRNA signal status, both calcified bioprosthetic valves (P = 0.03) and calcified native valves (P = 0.01) were significantly more positive for noncollagenous proteins mRNA than their noncalcified counterparts. Local cell-associated expression of proteins regulating mineralization suggests a highly coordinated mechanism of bioprosthetic and native valve calcification analogous to physiologic bone mineralization. Modulation of cellular infiltration or cellular expression of matrix proteins that regulate mineralization, may offer an effective therapeutic approach to the prevention of valve failure secondary to calcification.

R ecent studies have revealed that familial hypertrophic cardiomyopathy (FHC) is caused by missence mutations in myosin heavy chain or other sarcomeric proteins. To investigate the functional impact of FHC mutations in myosin heavy chain, mutants of Dictyostelium discoideum myosin II equivalent to human FHC mutations were generated by site-directed mutagenesis, and their motor function was characterized at the molecular level. These mutants, i.e., R397Q, F506C, G575R, A699R, K703Q, and K703W are respectively equivalent to R403Q, F513C, G584R, G716R, R719Q, and R719W FHC mutants. We measured the force generated by these myosin mutants as well as the sliding velocity and the actin-activated ATPase activity. These measurements showed that the A699R, K703Q, and K703W myosins exhibited unexpectedly weak affinity with actin and the lowest level of force, though their ATPase activity remained rather high. F506C mutant which has been reported to have benign prognosis exhibited the least impairment of the motile and enzymatic activities. The motor functions of R397Q and G575R myosins were classified as intermediate. These results suggest that the force level of mutant myosin molecule may be one of the key factors for pathogenesis which affect the prognosis of human FHC.

T he accelerated formation of advanced glycation end products (AGEs) and the overexpression of transforming growth factor beta (TGF-beta) have both been implicated in the pathogenesis of diabetic microvascular and macrovascular complications. Previous studies in our laboratory have demonstrated that the vascular changes in diabetes include hypertrophy of the mesenteric vasculature. To examine the role of AGEs in this process, streptozotocin-induced diabetic rats and control animals were randomized to receive aminoguanidine, an inhibitor of AGE formation, or no treatment. Animals were studied at 7 d, 3 wk, and 8 mo after induction of diabetes. When compared with control animals, diabetes was associated with an increase in mesenteric vascular weight and an increase in media wall/lumen area. By Northern analysis, TGF-beta1 gene expression was increased 100-150% (P < 0.01) and alpha1 (IV) collagen gene expression was similarly elevated to 30-110% compared to controls (P < 0.05). AGEs and extracellular matrix were present in abundance in diabetic but not in control vessels. Treatment of diabetic rats with aminoguanidine resulted in significant amelioration of the described pathological changes including overexpression of TGF-beta1 and alpha1 (IV) collagen. These data implicate the formation of AGEs in TGF-beta overexpression and tissue changes which accompany the diabetic state.

T he epidermal growth factor receptor (HER1) has been implicated in regenerative growth and proliferative diseases of the human bladder epithelium (urothelium), however a cognate HER1 ligand that can act as a growth factor for normal human urothelial cells (HUC) has not been identified. Here we show that heparin-binding EGF-like growth factor (HB-EGF), an activating HER1 ligand, is an autocrine regulator of HUC growth. This conclusion is based on demonstration of HB-EGF synthesis and secretion by primary culture HUC, identification of HER1 as an activatable HB-EGF receptor on HUC surfaces, stimulation of HUC clonal growth by HB-EGF, inhibition of HB-EGF-stimulated growth by heparin and of log-phase growth by CRM 197, a specific inhibitor of HB-EGF/HER1 interaction, and identification of human urothelium as a site of HB-EGF precursor (proHB-EGF) synthesis in vivo. ProHB-EGF expression was also detected in the vascular and detrusor smooth muscle of the human bladder. These data suggest a physiologic role for HB-EGF in the regulation of urothelial proliferation and regeneration subsequent to mucosal injury. Expression of proHB-EGF is also a feature of differentiated vascular and detrusor smooth muscle in the bladder. Because proHB-EGF is known to be the high affinity diphtheria toxin (DT) receptor in human cells, synthesis of the HB-EGF precursor by human urothelium also suggests the possibility of using the DT-binding sites of proHB-EGF as an in vivo target for the intraluminal treatment of urothelial diseases.

P -selectin is expressed on activated endothelium and platelets where it can bind monocytes, neutrophils, stimulated T cells, and platelets. Because recruitment of these cells is critical for atherosclerotic lesion development, we examined whether P-selectin might play a role in atherosclerosis. We intercrossed P-selectin-deficient mice with mice lacking the low density lipoprotein receptor (LDLR) because these mice readily develop atherosclerotic lesions on diets rich in saturated fat and cholesterol. The atherogenic diet stimulated leukocyte rolling in the mesenteric venules of LDLR-deficient mice, and the increase in adhesiveness of the vessels was P-selectin-dependent. Most likely due to the reduced leukocyte interaction with the vessel wall, P-selectin-deficient mice on diet for 8-20 wk formed significantly smaller fatty streaks in the cusp region of the aortae than did P-selectin-positive mice. This difference was more prominent in males. At 37 wk on diet, the lesions in the LDLR-deficient animals progressed to the fibrous plaque stage and were distributed throughout the entire aorta; their size or distribution was no longer dependent on P-selectin. Our results show that P-selectin-mediated adhesion is an important factor in the development of early atherosclerotic lesions, and that adhesion molecules such as P-selectin are involved in the complex process of atherosclerosis.

T o determine the extent to which the effect of a physiologic increment in epinephrine (EPI) on glucose production (GP) arises indirectly from its action on peripheral tissues (muscle and adipose tissue), epinephrine was infused intraportally (EPI po) or peripherally (EPI pe) into 18-h-fasted conscious dogs maintained on a pancreatic clamp. Arterial EPI levels in EPI po and EPI pe groups rose from 97 +/- 29 to 107 +/- 37 and 42 +/- 12 to 1,064 +/- 144 pg/ml, respectively. Hepatic sinusoidal EPI levels in EPI po and EPI pe were indistinguishable (561 +/- 84 and 568 +/- 75 pg/ml, respectively). During peripheral epinephrine infusion, GP increased from 2.2 +/- 0.1 to 5.1 +/- 0.2 mg/kg x min (10 min). In the presence of the same rise in sinusoidal EPI, but with no rise in arterial EPI (during portal EPI infusion), GP increased from 2.1 +/- 0.1 to 3.8 +/- 0.6 mg/kg x min. Peripheral EPI infusion increased the maximal gluconeogenic rate from 0.7 +/- 0.4 to 1.8 +/- 0.5 mg/ kg x min. Portal EPI infusion did not change the maximal gluconeogenic rate. The estimated initial increase in glycogenolysis was approximately 1.7 and 2.3 mg/kg x min in the EPI pe and EPI po groups, respectively. Gluconeogenesis was responsible for 60% of the overall increase in glucose production stimulated by the increase in plasma epinephrine (EPI pe). Elevation of sinusoidal EPI per se had no direct gluconeogenic effect on the liver, thus its effect on glucose production was solely attributable to an increase in glycogenolysis. Lastly, the gluconeogenic effects of EPI markedly decreased (60-80%) its overall glycogenolytic action on the liver.

T he induction of prostaglandin G/H synthase (PGHS; prostaglandin endoperoxide synthase, cyclooxygenase) by proinflammatory cytokines accounts, at least in part, for the altered eicosanoid biosynthesis in inflammatory diseases. In secondary cultures of normal human bronchial epithelial cells (NHBECs), interferon-gamma (IFN-gamma, 10 ng/ml for 24 h) increased the amount of prostaglandin E2 (PGE2) released in response to stimulation with exogenous arachidonic acid (5 microM). The enhanced production of PGE2 reflected the upregulation of PGHS-2 as indicated by enhanced expression of PGHS-2 RNA and increased recovery of PGHS-2 protein in NHBECs. IFN-gamma did not alter the production of PGE2 in A549 cells (a human lung adenocarcinoma cell line) or 6-keto-PGF1alpha in human umbilical vein endothelial cells (HUVECs), although prostaglandin release and/or the expression of PGHS-2 RNA in these cell lines was upregulated by other proinflammatory cytokines. Induction of PGHS-2 RNA in IFN-gamma-treated NHBECs, which peaked at 24 h, suggested the presence of an intermediary substance regulating the expression of PGHS-2. When the binding between the epidermal growth factor (EGF) receptor and its ligands was disrupted by a neutralizing antibody (LA-1), IFN-gamma failed to upregulate the release of PGE2 and the expression of PGHS-2 RNA in NHBECs. Furthermore, IFN-gamma induced the expression of RNAs for a number of ligands at the EGF receptor TGF-alpha; heparin-binding EGF-like growth factor (HB-EGF); and amphiregulin in NHBECs, and when administered exogenously, these ligands increased PGE2 release from NHBECs. Heparin at the concentration that neutralized the function of amphiregulin, or antibodies against TGFalpha or HB-EGF also reduced the release of PGE2 from IFN-gamma-stimulated NHBECs. These data are consistent with the presence of an autocrine growth factor/EGF receptor loop regulating PGHS-2 expression and PGE2 synthesis in bronchial epithelial cells.

T he mechanism of cooperation between IL-5 and eotaxin for the selective accumulation of eosinophils at sites of allergic inflammation is unknown. In this investigation we have used IL-5 deficient mice to define the relationship between this cytokine and eotaxin in the regulation of blood eosinophilia and eosinophil homing and tissue accumulation. Both IL-5 and eotaxin could independently induce a rapid and pronounced blood eosinophilia in wild type mice when administered systemically. In contrast, only eotaxin induced a pronounced blood eosinophilia in IL-5 deficient mice. The eosinophilic response induced by intravenous eotaxin in wild type mice did not correlate with a significant reduction in the level of bone marrow eosinophils, whereas intravenous IL-5 resulted in depletion of this store. These results suggest the existence of two mechanisms by which eosinophils can be rapidly mobilized in response to intravenous eosinophil chemoattractants; first, mobilization of an IL-5 dependent bone marrow pool, and second, an eotaxin-induced sequestration of eosinophils from tissues into the blood. Subcutaneous injection of eotaxin induced a local tissue eosinophilia in wild type mice but not in IL-5 deficient mice. Furthermore, tissue eosinophilia in wild type mice, but not in IL-5 deficient mice, was enhanced by adoptive transfer of eosinophils or the administration of intravenous IL-5. However, pretreatment of IL-5 deficient mice with intraperitoneal IL-5 for 72 h restored eosinophil homing and tissue accumulation in response to subcutaneous eotaxin. We propose that eotaxin secreted from inflamed tissue may play an important role in initiating both blood and tissue eosinophilia in the early phases of allergic inflammation. Furthermore, IL-5 is not only essential for mobilizing eosinophils from the bone marrow during allergic inflammation, but also plays a critical role in regulating eosinophil homing and migration into tissues in response to eotaxin and possibly other specific chemotactic stimuli.

S tudies determined the effects of chronic changes in sodium diet on the expression, regulation, and function of different angiotensin II (ANG II) receptor subtypes in renal resistance vessels. Rats were fed low- or high-sodium diets for 3 wk before study. Receptor function was assessed in vivo by measuring transient renal blood flow responses to bolus injections of ANG II (2 ng) into the renal artery. ANG II produced less pronounced renal vasoconstriction in rats fed a low- compared with high-sodium diet (16% vs. 56% decrease in renal blood flow, P < 0.001). After acute blockade of ANG II formation by iv enalaprilat injection in sodium-restricted animals, ANG II produced a 40% decrease in renal blood flow, a level between untreated dietary groups and less than high salt diet. Intrarenal administration of angiotensin II receptor type 1 (AT1) receptor antagonists losartan or EXP-3174 simultaneously with ANG II caused dose-dependent inhibition of ANG II responses. Based on maximum vasoconstriction normalized to 100% ANG II effect in each group, AT1 receptor antagonists produced the same degree of blockade in all groups, with an apparent maximum of 80-90%. In contrast, similar doses of the angiotensin II receptor type 2 (AT2) receptor ligand CGP-42112 had only a weak inhibitory effect. In vitro equilibrium-saturation 125I-ANG II binding studies on freshly isolated afferent arterioles indicated that ANG II receptor density was lower in the low- vs. high-sodium animals (157 vs. 298 fmol/mg, P < 0.04); affinity was similar (0.65 nM). Losartan and EXP-3174 displaced up to 80-90% of the ANG II binding; fractional displacement was similar in both diet groups. In contrast, the AT2 receptor analogues PD-123319 and CGP-42112 at concentrations < 10(-6) M had no effect on ANG II binding. RT-PCR assays revealed the expression of both angiotensin II receptor type 1A (AT(1A)) and angiotensin II receptor type 1B (AT(1B)) subtypes in freshly isolated afferent arterioles, while there was very little AT2 receptor expression. Total AT1 receptor mRNA expression was suppressed by low sodium intake to 66% of control levels, whereas it was increased to 132% of control by high-sodium diet, as indicated by ribonuclease protection assay. Receptor regulation was associated with parallel changes in AT(1A) and AT(1B) expression; the AT(1A)/AT(1B) ratio was stable at 3.7. We conclude that AT1 receptors are the predominant ANG II receptor type in renal resistance vessels of 7-wk-old rats. Chronic changes in sodium intake caused parallel regulation of expression and amount of receptor protein of the two AT1 receptor genes that modulate receptor function and altered reactivity of renal vessels to ANG II.

P ulmonary immune responses are suited to determine mechanisms of lymphocyte elimination, as lung inflammation must be regulated tightly to preserve gas exchange. The self-terminating response of primed C57BL/6 mice to intratracheal challenge with the T cell-dependent Ag sheep erythrocytes (SRBC) was used to test the importance of lung lymphocyte apoptosis in pulmonary immunoregulation. Apoptosis of alveolar and interstitial lymphocytes was demonstrated morphologically, by three independent methods to detect DNA fragmentation, and by surface expression of phosphatidylserine. Apoptotic lymphocytes were exclusively CD4-, CD8-, B220-, but many were CD3+ and Thy 1+. Inhibiting apoptosis by in vivo cyclosporine treatment prolonged lung lymphocyte accumulation following SRBC challenge. Experiments using mice homozygous for the lpr or gld mutations showed that pulmonary lymphocyte apoptosis depended on expression of Fas (CD95) and its ligand (Fas-L). Pulmonary inflammation increased on repeated intratracheal SRBC challenge of lpr/lpr mice, in contrast to the waning response in normal mice. These results confirm that in situ lymphocyte apoptosis contributes to termination of immune responses in nonlymphoid organs, probably because of activation-induced cell death, and may be important in inducing tolerance to repeated antigen exposure.

R eactivation of herpes simplex virus type 2 (HSV-2) occurs intermittently as perceived clinically and by viral culture. We performed a series of studies to evaluate the frequency and pattern of HSV-2 reactivation using both viral isolation and HSV PCR assay. Daily samples of genital secretions were obtained from 27 HSV-2 seropositive women; a subset of subjects obtained samples while receiving oral acyclovir 400 mg PO twice a day. HSV DNA was detected in genital swab specimens on 28% of 1,410 d compared with 8.1% of days by viral isolation. 11 of 20 women had HSV DNA detected on > 20% of days, 4 on > 50%, and 2 on > 75% of days; in contrast, none of the women shed on > 21% of days by viral isolation. The daily administration of oral acyclovir promptly reduced the frequency of HSV DNA detection by a median of 80%. Within 3-4 d of discontinuing daily acyclovir, HSV DNA again appeared in the genital area. HSV-2 shedding in the genital mucosa occurs much more frequently than previously appreciated. This frequent reactivation likely plays a role in the epidemic spread of genital herpes worldwide.

R ecombinant adenoviruses (Ads) efficiently transfer foreign genes into hepatocytes in vivo, but the duration of transgene expression is limited by the host immune response which precludes gene expression upon readministration of the virus. To test if this immune response can be abrogated by oral tolerization, we instilled protein extracts of a recombinant adenovirus type-5 via gastroduodenostomy tubes into bilirubin-UDP-glucuronosyltransferase-1 (BUGT1)-deficient jaundiced Gunn rats. Control rats received BSA. Subsequent intravenous injection 5 x 10(9) pfu of a recombinant adenovirus-expressing human BUGT1 (Ad-hBUGT1) resulted in hepatic expression of human BUGT1 (hBUGT1) with reduction of serum bilirubin levels by 70%. After 2 mo serum bilirubin increased gradually. In orally tolerized rats, but not in controls, a second dose of the virus on day 98 markedly reduced serum bilirubin again. In the tolerized rats, the development of antiadenoviral neutralizing antibodies and cytotoxic lymphocytes were markedly inhibited, and transplantation of their splenocytes into naive Gunn rats adoptively transferred the tolerance, indicating a role for regulatory cells. Lymphocytes from the tolerized rats hyperexpressed TGFbeta1, IL2, and IL4 upon exposure to viral antigens, whereas IFNgamma expression became undetectable. Thus, oral tolerization with adenoviral antigens permits long-term gene expression by repeated injections of recombinant adenoviruses.

T o explore potential roles for lactogenic hormones in human fetal development, we examined the distribution and ontogenesis of expression of prolactin receptors (PRLRs) in human fetal tissues at 7.5-14 wk of gestation and in tissues of the embryonic and fetal rat on days e12.5-e20.5. Histochemical analysis of PRLR immunoreactivity in the human fetus and fetal rat revealed novel and unexpected patterns of receptor expression. Most remarkable was the appearance in early fetal development of intense PRLR immunoreactivity in tissues derived from embryonic mesoderm, including the periadrenal and perinephric mesenchyme, the pulmonary and duodenal mesenchyme, the cardiac and skeletal myocytes, and the mesenchymal precartilage and maturing chondrocytes of the endochondral craniofacial and long bones, vertebrae and ribs. Striking changes in the cellular distribution and magnitude of expression of PRLRs were noted in many tissues during development. In the fetal adrenal the initial mesenchymal PRLR expression is succeeded by the emergence of PRLR immunoreactivity in deeper fetal cortical cell layers. In the fetal kidney and lung, the invagination of cortical mesenchyme is accompanied by progressive PRLR immunoreactivity in bronchial and renal tubular epithelial cells. In the pancreas, the PRLR is expressed primarily in acinar cells and ducts in early gestation; in late gestation and in the postnatal period, the PRLR is expressed predominantly in pancreatic islets, co-localizing with insulin and glucagon. Finally in fetal hepatocytes, PRLR immunoreactivity increases significantly between embryonic days e52 and e96 in the human fetus and between days e16.5 and e18.5 in the fetal rat. In addition to playing important roles in reproduction, lactation, and immune function, the lactogenic hormones likely play roles in tissue differentiation and organ development early in gestation.

4 0 recipients of first cadaver kidney transplants were given perioperative donor vertebral bone marrow infusions (DBMC), compared with 100 controls who did not receive donor bone marrow. The immunosuppressive regimen included OKT3, Tacrolimus, and steroid maintenance therapy, and, in some patients, newly introduced mycophenolate mofetil. This report describes the 24-mo actuarial follow-up and several immunological monitoring studies including sequential measurements of donor bone marrow lineage subset chimerism by the recently reported PCR-flow assay. This is a sensitive in situ PCR detection system for donor versus recipient histocompatibility genes as well as cell surface CD epitope markers using flow cytometry. The results indicate (a) the stabilization of the donor CD3+ and CD34+ cells in recipient peripheral blood at levels below 1% between 6 mo and 1 yr postoperatively, with a 10-fold higher level of donor cell chimerism of these lineages in recipient iliac crest marrow; (b) significantly lower levels of chimerism in peripheral blood up to 6 mo postoperatively in patients who had early acute (reversible) rejection episodes compared with those who did not; (c) a higher degree of chimerism seen in patients who were class II MHC HLA DR identical with their donors; (d) the identification of a high proportion of the donor bone marrow derived CD3 dimly staining subset of T cells (to which regulatory functions have been ascribed) in recipient peripheral blood and especially in recipient bone marrow; and (e) an unexpectedly increased susceptibility to clinically significant infections (primarily viral), and even death in the DBMC-infused group, compared with controls, but no graft losses because of rejection in the DBMC-infused group. Mixed lymphocyte culture assays showed a trend toward a greater number of nonspecifically low reactors in the DBMC group, as well as a greater number of nonspecifically high reactors in the controls (P = 0.058). The autologous mixed lymphocyte reaction also indicated a trend towards nonspecific immune activation in the DBMC group. Finally, anti-cytomegaloviral IgG antibody reactivity was significantly inhibited in the DBMC group 4-6 mo postoperatively (P = < 0.05). In the controls, there were no donor cell lineages detected by PCR-flow in the peripheral blood. These rather unexpected findings, indicating a more depressed cellular and humoral immune capacity in the DBMC cadaver kidney transplant recipients in this relatively early follow-up period, are discussed relevant to chimerism, MHC restriction, and suppressor activity brought about by specialized DBMC subsets, which still need to be defined.

F ive lipoxygenase (5-LO) is the first committed enzyme in the metabolic pathway leading to the synthesis of the leukotrienes. We examined genomic DNA isolated from 25 normal subjects and 31 patients with asthma (6 of whom had aspirin-sensitive asthma) for mutations in the known transcription factor binding regions and the protein encoding region of the 5-LO gene. A family of mutations in the G + C-rich transcription factor binding region was identified consisting of the deletion of one, deletion of two, or addition of one zinc finger (Sp1/Egr-1) binding sites in the region 176 to 147 bp upstream from the ATG translation start site where there are normally 5 Sp1 binding motifs in tandem. Reporter gene activity directed by any of the mutant forms of the transcription factor binding region was significantly (P < 0.05) less effective than the activity driven by the wild type transcription factor binding region. Electrophoretic mobility shift assays (EMSAs) demonstrated the capacity of wild type and mutant transcription factor binding regions to bind nuclear extracts from human umbilical vein endothelial cells (HUVECs). These data are consistent with a family of mutations in the 5-LO gene that can modify reporter gene transcription possibly through differences in Sp1 and Egr-1 transactivation.