Surfactant-like particles mediate tissue-specific functions in epithelial cells.

Lipid-rich, unilamellar membranes appear to be relatively common structures lining the apical or 'exposed' surface of epithelial cells. They have now been described in the intestinal tract from the esophagus to the rectum and have been isolated from tissues, such as the stomach, the small bowel, the colon, and the bladder. The presence of a lining layer in the lungs has been known for some time, and its functions, structure, and metabolism have been extensively studied, as can be gleaned from the multitude of reports presented at this symposium. The 'other' surfactants, however, have attracted far less attention and have been investigated in detail in only a few reports. This paucity of information, when compared to the pulmonary system, is most likely due to the fact that a generalized function (sufficiency state) or disease (deficiency state) has not yet been recognized for either the intestinal or urinary forms of surfactant. It seems reasonable to assume that the role of the SLP will vary, at least in part, with the organ or tissue with which it is associated, although the widespread nature of the membrane assumes that some functions (e.g. protective) will be shared. Thus, pulmonary surfactant's primary function in the lung may be to reduce surface tension and prevent lung collapse; but it also plays a significant part in the lung's defenses against bacterial and/or chemical invasion. It is hoped that future studies will shed some light on the function of the various SLPs and lead to a better appreciation for their role in both maintaining a healthy environment and contributing to the proper functioning of their host tissues.

Regulation of surfactant-like particle secretion by Caco-2 cells.

Surfactant-like particle (SLP) is a phosphatidylcholine (PC)-rich membrane produced in the small intestine, and its secretion is increased by fat feeding. In Caco-2 cells known to produce SLP, preincubation with [(3)H]palmitate labelled the SLP and was used as a marker for newly secreted membrane. SLP-associated PC and protein (d=1.07-1.08 g/ml in a linear non-equilibrium NaBr gradient) were secreted in parallel with triacylglycerols (TG) and at a rate about twice the control rate in response to feeding cells with an oleate/egg PC mixture. Cholesterol and apolipoprotein A-I identified only a small peak corresponding to high-density lipoprotein (HDL), but the largest peak corresponded with SLP (d=1.07-1.08). Palmitate incorporation into PC showed a similar small peak migrating at the density of HDL, but most labelled PC secreted from the cells was due to SLP. PC secretion, alkaline phosphatase activity, and newly synthesized immunoprecipitated SLP proteins from conditioned serum-free media migrated together at a density of >/=1.21 g/ml in a lipoprotein NaBr step gradient, and represented SLP. Glycerol incorporated into TG migrated at a peak density of 1.12 g/ml, consistent with HDL secretion from cells incubated in serum-free media. These data confirm that the secreted PC in SLP is distinct from lipoprotein particles. Incorporation of [(3)H]palmitate into the PC fraction of either whole cell homogenate or isolated brush border membranes was not affected by oleate/egg PC feeding. Both Pluronic L-81, an inhibitor of chylomicron secretion, and BMS-197636-02, a microsomal triglyceride transfer protein inhibitor, blocked the secretion of both TG and PC. Elevation of intracellular cAMP levels that stimulate surfactant secretion from type II pneumocytes caused a 50% reduction in SLP and TG secretion from Caco-2 cells. These results confirm the SLP response to fat feeding found in vivo, further supporting a role for SLP in TG secretion from the enterocyte, and show that the regulation of SLP secretion differs from that of pulmonary surfactant.

Role of intestinal surfactant-like particles as a potential reservoir of uropathogenic Escherichia coli.

The binding of uropathogenic Escherichia coli is mediated at the tips of pili by the PapG adhesin, which recognizes the Galalpha(1-4)Gal disaccharide on the uroepithelial surface. These receptors have been identified unequivocally in the human and murine urinary tracts but not in intestinal epithelium, yet uropathogenic E. coli strains are commonly found in normal colonic microflora. The gastrointestinal tract from duodenum to rectum elaborates a phospholipid-rich membrane particle with surfactant-like properties. In these studies, we report that purified murine particles contain a receptor recognized by the class I PapG adhesin because: (1) PapD-PapG complexes and class I pili bound to surfactant-like particles in a solid-phase assay, whereas binding was not detected in microvillous membranes derived from the same tissues, (2) purified PapD-PapG complex bound to a glycolipid receptor detectable in lipid extracts from the particles, and (3) soluble Galalpha(1-4)Gal inhibited the adhesin by 72% from binding to surfactant-like particles. The Galalpha(1-4)Gal receptor present in the intestinal surfactant-like particle which overlies the intestinal mucosa could provide one means to establish an intestinal habitat for uropathogenic E. coli.

Intestinal fatty acid binding protein may favor differential apical fatty acid binding in the intestine.

The intestinal mucosa metabolizes fatty acids differently when presented to the lumenal or basolateral membrane. Expression of both liver and intestinal fatty acid binding proteins (L- and I-FABPs) uniquely in the enterocyte offers a possible explanation of this phenomenon. An organ explant system was used to analyze the relative binding of fatty acids to each protein. More fatty acid was bound to L-FABP than to I-FABPs (28% vs. 6% of cytosolic radioactivity), no matter on which side the fatty acid was added. However, a 2-3-fold increase in fatty acid binding to the intestinal paralog was noted after apical addition of palmitic or oleic acid in mucosa from chow fed rats. When oleic acid was added apically, a 1.4-fold increase in binding to I-FABP was observed in mucosa derived from chronically fat fed rats, consistent with the previously observed 50% increase in the content of that protein. Immunocytochemical localization of both FABPs in vivo demonstrated an apical cytoplasmic localization in the fasting state, and redistribution to the entire cytoplasm after fat feeding. These data are consistent with the hypothesis that I-FABP may contribute to the metabolic compartmentalization of apically presented fatty acids in the intestine.

Binding of pili from uropathogenic Escherichia coli to membranes secreted by human colonocytes and enterocytes.

PapG adhesins mediate the binding of uropathogenic Escherichia coli. Although receptors for these adhesins have not been demonstrated in intestinal epithelia, the colonic microflora includes strains of uropathogenic E. coli. We now report that surfactant-like particles secreted by the human intestine contain receptors for PapG adhesins and may provide an intestinal habitat for uropathogenic bacteria.

Vitamin A up-regulates expression of bone-type alkaline phosphatase in rat small intestinal crypt cell line and fetal rat small intestine.

Vitamin A is a potent inducer for liver/bone/kidney alkaline phosphatase (L/B/K ALP) in a variety of tissues. However, the evidence for induction of L/B/K ALP by vitamin A in small intestine is limited. In this study, we investigated the influence of vitamin A on L/B/K ALP expression in rat small intestinal crypt IEC-6 cells and fetal rat small intestine. Treatment of IEC-6 cells with all-trans retinoic acid (RA) increased the levels of activity, protein and mRNA of L/B/K ALP, whereas enterocyte-specific proteins, including intestinal ALP, sucrase-isomaltase and glucose transporter-2, were not induced. The reverse transcription-polymerase chain reaction technique revealed that this L/B/K ALP transcript had the bone-type but not the liver-type leader exon. IEC-6 cells constitutively expressed mRNAs of all subtypes of retinoic acid receptor (RAR) and retinoid X receptor (RXR) at varied concentrations. Among these receptor mRNAs, RARbeta mRNA quickly responded to RA treatment, and the level was doubled within 4 h. Gel mobility shift assay showed that RA induced an RXRE-binding activity in IEC-6 cells. The L/B/K ALP transcript, expressed in fetal rat small intestine, also contained the bone-type leader exon. Intragastric administration of 10 mg retinyl acetate to pregnant rats from gestational d 7 to 15 increased the levels of this transcript and enzyme in 15-d fetal rat small intestine. Our results suggest that vitamin A may be an important regulator for L/B/K ALP expression in fetal rat small intestine as well as in IEC-6 cells.

Caco-2 cells express a combination of colonocyte and enterocyte phenotypes.

Caco-2 cells are derived from a human colonic adenocarcinoma, but differentiate into small intestinal-like cells after confluence. While this enterocytic differentiation has been well studied, the presumed parallel loss of colonocyte function has not been as thoroughly examined. To follow the phenotype for both tissues, Western blots were performed using antisera recognizing liver/bone/kidney alkaline phosphatase and surfactant-like particle proteins found in normal human colon, along with antisera against the small bowel representatives of the same proteins. Antisera against proteins enriched in either enterocytes (alpha1-antitrypsin) or colonocytes (surfactant protein A) were also evaluated. Alkaline phosphatase activity increased from 3 to 18 days post-confluence. Activity at 3 days post-confluence derived substantially from both isomers. Thereafter, the colonic (liver/bone/kidney) isomer declined to low levels as the content of the enterocytic isomer rose. A similar pattern was found with colonic (decreasing expression) and enterocytic (increasing expression) surfactant-like particle proteins. In particular, the content of larger enterocytic particle proteins (97 and 116 kDa) increased with time in culture. Expression of alpha1-antitrypsin increased early and remained high, whereas surfactant protein A generally declined after the third day post-confluency. In summary, undifferentiated Caco-2 cells express very low levels of proteins characteristic of either colonocytes or enterocytes. Immediately after confluence, they expressed proteins characteristic of both cell types. Thereafter, the content of colonocyte-specific proteins decreased, whereas those specific for the enterocyte increased. The timing and degree of this phenotypic switch have implications for the interpretation of experiments using Caco-2 cells as a model of small intestinal function.

Effects of maternal dexamethasone therapy on fetal lung development in the rhesus monkey.

A large body of evidence demonstrates that antenatal glucocorticoids can accelerate fetal lung maturation. The purpose of this study was to delineate the optimal dose of dexamethasone and to determine whether a single- or multiple-injection regimen of the same total dexamethasone dosage was more effective in accelerating pulmonary development. Pregnant rhesus monkeys were injected with varying doses of dexamethasone or vehicle as either a single-bolus injection or as four separate injections, each spaced 12 hours apart. All maternal treatments were begun exactly 72 hours prior to delivery, and all fetuses were delivered by cesarean section at 135 +/- 1 days gestation. When a single injection of dexamethasone was used, fetal liver weight increased in a dose-related fashion. Fetal and maternal cortisol and fetal blood glucose concentrations were also influenced by increasing dexamethasone dosages. Fetal pulmonary phospholipids, however, were unchanged at all steroid doses examined. Multiple injections of dexamethasone generally produced more pronounced effects, even though the total dexamethasone dose remained the same. Thus, liver weight and fetal and maternal cortisol, glucose, and insulin levels were significantly influenced by the multiple administration of dexamethasone. In addition, total lung phosphatidylcholine, surfactant phosphatidylcholine, the surfactant-phosphatidylcholine-to-total-phosphatidylcholine ratio, and the surfactant-disaturated-phosphatidylcholine-to-total-lung-disaturated- phosphatidylcholine ratio were elevated after the multiple-injection regimen. A total dose of 0.5 mg dexamethasone/kg maternal body weight given in four separate injections appeared to produce the most beneficial results. These data suggest that low-dose, antenatal glucocorticoid treatment can effectively accelerate the biochemical maturation of the fetal lung.

Appearance of surfactant-like particles in apical medium of Caco-2 cells may occur via tight junctions.

Intestinal alkaline phosphatase (AP) is secreted by Caco-2 cells bound to surfactant-like particles (SLP), which can be localized by electron microscopy to the basolateral space and the intestinal lumen, especially over tight junctions. To investigate the hypothesis that SLP are secreted basolaterally and enter the lumen through the tight junction, Caco-2 cells were incubated with agents known to increase permeability at tight junctions. Cytochalasin D and phorbol 12-myristate 13-acetate increased Caco-2 cell monolayer permeability and the appearance of particles in apical medium two- to threefold, as monitored by mannitol movement and AP activity, respectively. Blocking the apical secretory pathway by nocodazole or colchicine had no effect on either parameter. Estimation of SLP content demonstrated an increase in apical media particles similar to that determined by AP activity. Quantitative image analysis established that apical SLP content increased 4-10 times, whereas total cell particle content remained unchanged. These data indicate that SLP may be secreted initially into the basolateral space and then transported to the intestinal lumen through the tight junctions.

Two rat intestinal alkaline phosphatase isoforms with different carboxyl-terminal peptides are both membrane-bound by a glycan phosphatidylinositol linkage.

Two cDNAs encode rat intestinal alkaline phosphatases having completely different carboxyl-terminal peptides; one is hydrophobic and fulfills the consensus requirements for glycan phosphatidylinositol linkage, and the other is neither hydrophobic nor hydrophilic, but contains a small amino acid domain (-NSASS-) just distal to a region of 17 threonine residues. Constructs were created using 80% of the amino-terminal portion of one alkaline phosphatase and the carboxyl-terminal portions of each of the isoforms. Both of the carboxyl-terminal peptides supported glycan phosphatidylinositol linkage as demonstrated by the following criteria: 1) plasma membrane targeting in transfected COS-1 cells, 2) release of transfected alkaline phosphatase by phosphatidylinositol-specific phospholipase C, 3) appearance of the trypanosome variable glycoprotein cross-reacting determinant after phospholipase C treatment, 4) ethanolamine incorporation into newly synthesized enzyme, 5) loss of phospholipase C release after mutation of the omega and omega + 2 positions in the putative linkage site, -NSA-, and 6) evidence of surface membrane localization by immunofluorescence using antibody against rat intestinal alkaline phosphatase. These data demonstrate that a predicted hydrophobic carboxyl-terminal sequence is not essential for glycan phosphatidylinositol linkage. Moreover, because both isomers are membrane-bound, the origin of soluble enzyme in the serum is likely to arise from the action of serum phosphatidylinositol-specific phospholipase C.

The two mRNAs encoding rat intestinal alkaline phosphatase represent two distinct nucleotide sequences.

Rat duodenal mucosa contains two mRNAs (2.7 and 3.0 kb) encoding intestinal alkaline phosphatase (IAP), but the mechanism for their production has not been clear. By means of the polymerase chain reaction (PCR), we isolated a fragment that identifies a second rat IAP (rIAP-II), differing from the rIAP-I sequence in the coding and 3' untranslated regions and encoding a COOH-terminal sequence predicted to be hydrophilic. By means of probes unique to each sequence, rIAP-I identified the 2.7-kb mRNA, and rIAP-II the 3.0-kb mRNA. The different structures and differential regulation of the mRNAs after fat feeding demonstrate the presence of two rat IAP transcripts.

Glutamine synthetase activity of developing astrocytes is inhibited in vitro by very low concentrations of lead.

This study has dealt with the inhibition by lead of glutamine synthetase (GS) activity in homogenates of mixed glial primary cultures, 95% enriched in differentiating astrocytes. A 70% inhibition was observed with a lead concentration of only 2.5 microM. Prevention of the inhibition by addition of EDTA or dithiothreitol is compatible with the conclusion that the effect is mediated by binding of lead ion to sulfhydryl moieties of the enzyme. Among several other cations tested, only mercury, which has a similarly high binding affinity for sulfhydryl moieties, inhibited the enzyme. The inhibitory effect of lead was relatively specific, since no inhibition of another astrocytic marker enzyme, lactate dehydrogenase, of the oligodendroglial marker enzyme, 2',3'-cyclic nucleotide 3'-phosphohydrolase, or of the plasma membrane marker, Na,Ka-ATPase, was observed with concentrations of lead that produced a 70% decrease of GS. Because of the critical role of GS in regulation of extracellular glutamate, the findings raise the possibility that glutamate-induced neuronal injury is involved in the genesis of the cognitive defects associated with chronic low-level lead exposure in young children.

Brain damage induced by prenatal exposure to dexamethasone in fetal rhesus macaques. I. Hippocampus.

Neurotoxic effects of prenatal administration of dexamethasone were examined in the fetal rhesus monkey brain at 135 and 162 days of gestation (term is 165 days). In an experimental design mimicking human clinical trials, dexamethasone was given intramuscularly to pregnant monkeys on day 132 (single injection with doses of 0.5, 5, or 10 mg/kg maternal body weight) or on days 132 and 133 (multiple injections at 12-h intervals with 0.125 x 4, 1.25 x 4, or 2.5 mg/kg x 4). The fetuses were delivered by caesarean section on day 135 or day 162 and hippocampal slices were prepared for evaluation. Light and electron microscopic observation revealed decreased numbers of pyramidal neurons in the hippocampal CA regions and of granular neurons in the dentate gyrus associated with degeneration of neuronal perikarya and dendrites. Axodendritic synaptic terminals of the mossy fibers in the CA3 hippocampal region showed pronounced degeneration. Degeneration was dose-dependent and multiple injections induced more severe damage than single injections of the same total dose. Even the lowest dose (0.5 mg/kg, which is similar to the dose used in human clinical trials) produced these changes. Degenerative changes induced by dexamethasone treatment (5 mg/kg) on days 132 and 133 were also clearly evident in fetuses studied at 162 days. Therefore, caution is recommended in the use of prenatal corticosteroids in premature deliveries.

Hypoxic injury to developing glial cells: protective effect of high glucose.

Hypoxic injury to differentiating glial cells is a critical event in the development of periventricular leukomalacia, the major hypoxic-ischemic lesion of the premature infant. This study has addressed the effects of hypoxia on differentiating glial cells, primarily astrocytes. Primary cultures of dissociated newborn rat brain, which are composed predominantly of differentiating astroglia, were used. Efflux of lactate dehydrogenase, an enzyme enriched in astroglia, was used to quantitate cellular injury. Three major findings are reported. First, differentiating astrocytes were resistant to hypoxic injury for many hours, although by 24 h of hypoxia severe cellular injury (lactate dehydrogenase efflux of 86% of total and morphologic changes) was obvious. Second, increase of glucose in the culture medium from the approximately physiological concentration of 5.6 to 15 mM had a marked protective effect versus hypoxia, i.e. lactate dehydrogenase efflux was totally prevented during 24 h of hypoxia in 15 mM glucose. Third, the protective effect of high glucose appeared to be related to increased utilization by glycolysis, because there was a direct correlation between the resistance to hypoxic cellular injury and the amount of lactate generated and of glucose consumed by the cells. Thus, the cells with the lowest lactate dehydrogenase efflux (and highest glucose supplementations) had medium lactate concentrations as high as 32-36 mM. These concentrations of lactate are approximately double the reported threshold concentration of lactate considered to produce cellular necrosis in in vivo models of hypoxic injury, primarily in mature animals. The data raise the possibility that hypoxic injury to differentiating glia can be prevented or ameliorated by increase in glucose availability.

Effect of lactate on glucose incorporation into fetal lung phospholipids.

Fetal rabbit lung explants were incubated with 3.0 mM glucose and varying levels of lactate. An increase in lactate concentrations resulted in a decrease in glucose incorporation into total disaturated phosphatidylcholine. Glucose utilization for surfactant phosphatidylcholine synthesis was also reduced by approximately 35% in the presence of 5.0 mM lactate. The decreased incorporation of glucose occurred in the fatty acid portion of both total tissue disaturated phosphatidylcholine and surfactant phosphatidylcholine. The effect of lactate on glucose incorporation into pulmonary phospholipids was not affected by the presence of pyruvate in concentrations up to 500 microM. Pyruvate alone produced only a slight decrease in glucose utilization for lung phospholipid production. These data indicate that glucose and lactate are competitive substrates for late gestation surfactant phospholipid fatty acid synthesis, and that lactate is potentially a very important substrate for fetal lung development.

Evidence for lactate utilization for fetal lung glycogen synthesis.

Fetal rabbit lungs from 23 day gestation animals were used to investigate the potential role of lactate as a substrate for fetal lung glycogen synthesis. Fetal lactate dehydrogenase activity was approximately twice that found in the adult lung, while the activity of phosphoenolpyruvate carboxykinase was elevated fourfold over the adult value. Pyruvate carboxylase activities were similar in both fetal and adult lungs. Studies employing fetal lung explants in organ culture indicated that the presence of both glucose and lactate may be necessary for glycogen accumulation in the developing fetal lung. These data support the hypothesis that lactate is an important precursor for fetal lung glycogen.

Palmitate and glucose oxidation by fetal type II pneumocytes.

Fetal type II pneumocytes in organotypic culture can oxidize both palmitate and glucose, with glucose being converted to CO2 at a rate substantially greater than that of palmitate. Glucose can be oxidized by both the pentose shunt pathway and the tricarboxylic acid cycle. Palmitate oxidation to CO2 is increased by carnitine and reduced by glucose and unsaturated fatty acids. These data suggest that glucose may be an important oxidative substrate during late fetal life and that fatty acids may play a relatively minor role in type II cell oxidative metabolism.

Fetal lung development in male and female nonhuman primates.

Indices of lung maturation were assessed in 58 rhesus fetuses at five gestational ages during the last trimester of nonhuman primate pregnancy to determine whether fetal sex influences lung maturation. In addition to analysis of whole lung phospholipids, glycogen, protein, DNA, and pressure-volume curves surfactant fraction phosphatidylcholine (PC) was quantitated following isolation by sucrose gradient centrifugation and a combination of predictors were assessed by all possible subsets regression to attain a composite "maturity index." For the total population, there was a uniform progression in physical growth characteristics, lung destensibility and stability and phospholipids with advancing gestation. The quantitative change in surfactant fraction PC concentration for both sexes was considerably greater than that observed for whole lung PC between 135 days gestation and term. Further, the increase in surfactant PC occurred in association with improving lung destensibility and deflation stability prior to maximum changes in the whole lung PC or disaturated PC concentration. There were no statistically discernible differences in biochemical or physiological assessment between sexes at any gestational age. These data in nonhuman primates suggest that documented differences in survival from the respiratory distress syndrome between males and females do not result from a discordance in lung maturation as a function of time throughout the last trimester of gestation.

Changes in surfactant phospholipids in fetal rat lungs from normal and diabetic pregnancies.

The purposes of this study were to adapt and evaluate further a pulmonary surfactant isolation method applicable to unperfused fetal rat lung, to quantitate key phospholipids phosphatidylcholine (disaturated phosphatidylcholine, and phosphatidylglycerol) of the isolated material during the last 3 days of gestation, and to determine if abnormalities in surfactant phospholipids were present in fetuses of diabetic pregnancies. A simplified scheme of sucrose gradient centrifugation proved useful for small scale preparations of material enriched in the phospholipids most characteristic of pulmonary surfactant. It was shown that fetal blood phospholipids did not contaminate the surfactant fraction and therefore would not produce artifacts in assessment of lung maturational changes. Analyses of subcellular fractions isolated at 19.5, 20.5, and 21.5 days revealed that the percentages of disaturated phosphatidylcholine relative to total phospholipids were 23-44% in the surfactant preparations and 14-21% in the residual (nonsurfactant) fractions, while the disaturated phosphatidylcholine/phosphatidylcholine ratios were 0.62 +/- 0.06 and 0.41 +/- 0.03, respectively. Summation of the amounts of individual phospholipids in the two fractions yielded data that were nearly identical to the concentrations of these compounds in whole fetal lung samples analyzed independently, implying that losses during the surfactant isolation technique were negligible. The concentrations of phosphatidylcholine, disaturated phosphatidylcholine, phosphatidylglycerol, and total phospholipids increase markedly (more than 10-fold) and progressively in surfactant fractions prepared from normal fetal rat lung at 19.5, 20.5, and 21.5 days of gestation. In contrast, the residual fractions showed no changes from 19.5 to 20.5 days and then relatively modest increases from 20.5 to 21.5 days, except for phosphatidylglycerol, which increased markedly.(ABSTRACT TRUNCATED AT 250 WORDS)