During metabolism of sphingomyelin (SM) and glycosphingolipids, the metabolites ceramide, sphingosine, and sphingosine-1-phosphate (S1P) are generated. These compounds are important messengers that regulate numerous cellular functions, including cell growth, differentiation, and apoptosis (1–3). Ceramide and sphingosine are proapoptotic, acting via several intracellular downstream targets. Sphingosine kinase phosphorylates sphingosine to S1P, which acts on G protein–coupled plasma membrane receptors. S1P promotes cell growth and angiogenesis and regulates platelet function and lymphocyte migration. The term ceramide/S1P sphingolipid rheostat emphasizes the crucial importance of this ratio in regulation of multiple cellular functions (4).
Abnormal sphingolipid signaling may contribute to the pathology in cystic fibrosis (CF). The uptake and degradation of S1P is retarded in cultured murine airway epithelial cells, suggesting that prolonged S1P signaling may be an important feature of a proinflammatory vicious circle (5). In normal nasal or tracheal epithelial cells, Pseudomonas aeruginosa infection results in an activation of acid sphingomyelinase (acid-SMase). This enzyme may be involved in the formation of ceramide-enriched membrane domains (lipid rafts) that are required for internalisation of bacteria and reorganization of receptors, as well as for intracellular signaling molecules. Ligand-receptor activity located to these lipid rafts regulates induction of cell death and the controlled release of cytokines from cells, and may thus be important during P aeruginosa infections in patients with CF (6). Stimulation of a human CF airway epithelial cell line, IB3-1, with P aeruginosa resulted in decreased formation of ceramide due to loss of acid-SMase activity compared with controls (7). Ito et al (8) and Ramu et al (9) reported a direct inhibitory effect on the CF transmembrane conductance regulator (CFTR) function by ceramide in the basolateral membrane of airway epithelial cells. Cholesterol/SM interacting mechanisms may influence CFTR-mediated chloride secretion in airway (10) and colon (11) epithelial cells.
In the gut SM is digested by the brush border enzymes alkaline sphingomyelinase (alk-SMase) and neutral ceramidase (CDase) to sphingoid bases and fatty acids, which are well absorbed (12). Absorbed sphingosine is converted to S1P, most of which is further degraded to palmitic acid (13). The formation of ceramide, sphingosine, and S1P, and the metabolism of S1P to hexadecenal via S1P-lyase or hydrolysis via S1P-phosphatases, is therefore of great interest with regard to the intestinal pathology in CF. It is interesting to note that alk-SMase, by its phospholipase C activity, was shown to hydrolyse and inactivate platelet activating factor (14). This proinflammatory lipid messenger is increased in asthma, inflammatory bowel disease, and neonatal necrotizing enteritis (15), and may induce lung pathology mediated via acid-SMase–induced ceramide formation (16).
Alk-SMase and CDase are expressed in the gut as the differentiated epithelium develops at late fetal stage both in rats and in humans. Thus, significant levels of alk-SMase and neutral CDase were found in meconium of both preterm (from gestational week 23) and term human infants (17). Adding SM to a milk replacement formula was shown to influence gut differentiation and maturation in newborn rats (18).
The intestinal pathophysiology of the F508del CF mouse model, which exhibits the CFTR processing defect similarly to its human counterpart, shows defects of macromolecular secretion (mucus plugging) and of ion transport (19,20). The Cftrtm1Eur strain expresses levels of mRNA for F508del CFTR that appear comparable to the levels of wild-type CFTR in control animals (20,21). This results in a low but detectable (<20% of normal) Cl− secretory response to an increase in cellular cyclic adenosine monophosphate, and intestinal disease characterised by goblet cell hyperplasia and lipid resorption deficiency (22). In humans expression of the mutated F508del CFTR also is observed (23). Residual CFTR activity is observed in rectal biopsies of a minority of patients with CF, including homozygous F508del mutants (24). Distinct but not life-threatening gastrointestinal pathology is an important feature of this phenotype (25).
An altered formation of sphingolipid metabolites from dietary and/or endogenous sources could affect intestinal maturation and contribute to development of intestinal pathology in CF. We therefore asked whether enzymes in the gastrointestinal tract involved in the generation of ceramide and sphingosine, and thus indirectly in the generation of S1P, are normally expressed in F508del homozygous and heterozygous mice. We measured alk- and acid-SMase, and neutral- and acid-CDase, in the small intestine and colon of wild-type and F508del homozygous and heterozygous mice (Cftrtm1Eur). We also analyzed other tissues of interest in CF pathology.
Our main findings suggest that alk-SMase and neutral CDase are normally expressed in the small intestine and colon in this model, and the activity of acid-SMase is lower in the ileum of F508del homozygous mice and the neutral CDase higher in the spleen. We also report differences in body weight and size of the intestine and colon.
MATERIALS AND METHODS
C57BL/6J mice (F12 backcross) heterozygous for the F508del mutation, strain Cftrtm1Eur(20,26), were obtained from the Erasmus Medical Centre. Embryo derivation into a specific pathogen-free condition was performed upon arrival at the Karolinska Institutet. The mice were bred and housed in isolator cages in a specific pathogen-free facility at the Karolinska Institutet. All of the mice were fed with standard diet (R36, Lactamin, Stockholm, Sweden) and water ad libitum. The mice were bred as heterozygotes and offspring were genotyped by polymerase chain reaction analysis followed by digestion of the polymerase chain reaction product with SspI. The following primer pairs were used:
All of the experiments were conducted in accordance with institutional guidelines for animal care and use.
Preparation of Tissues for Assays
Organs were harvested at Karolinska Institutet and shipped on dry ice to Lund University. Frozen organs were thawed on ice and the stomach and caecum were removed. The small intestine was rinsed with saline and the content collected. The length and weight of small intestines were then measured before dividing the intestine into 4 equal segments of about 8 to 10 cm. Each piece was cut open and the mucosa was scraped with a plastic scraper and put in a glass homogenizer. The mucosa from each segment was homogenized in 0.9 mL buffer (50 mmol/L Tris-HCl, 2 mmol/L ethylenediaminetetraacetic acid, 1 mmol/L phenylmethylsulphonyl fluoride, 1 mmol/L benzamidine, 0.5 mmol/L dithiothreitol, and 2.5 mg/mL bile salt mixture; all chemicals from Sigma, St Louis, MO). The length and weight of the colon were measured and the mucosa was scraped and homogenized in 0.5 mL buffer. A piece of liver, 1 kidney, the whole spleen, a piece of lung, and an aliquot of faeces from each mouse were homogenized in 1-mL buffer. All of the homogenates were sonicated for 10 seconds twice and centrifuged at 15,000g for 10 minutes. The supernatants were saved and used for assays. Protein determination of each supernatant was performed with BioRad DC (detergent compatible) Protein assay kit (BioRad, Stockholm, Sweden).
A substrate solution for assay of neutral CDase was prepared by sonicating 14C-octanoyl-D-erythro-sphingosine (American Radiolabeled Chemicals, St Louis, MO) (27) in 50 mmol/L Tris-HCl pH 7.0 and 10 mmol/L NaTC for 2 minutes on ice. For acid ceramidase, 50-mmol/L Tris-Maleate pH 5.0 and 10-mmol/L Na TC buffer were used. For each assay, the enzyme was added to the substrate solution as 10% of the volume and incubated in a water bath at 37°C for 60 minutes. The reaction was stopped by adding 6 volumes of methanol:chloroform:heptane (20,25,28) and 2 volumes of 0.05 mol/L K2CO3, K2B2O2 pH 10 (27). After mixing, the tubes were centrifuged at 9000g for 10 seconds on a MSE Micro Centaur centrifuge for Eppendorf tubes. An aliquot of the upper phase was taken to liquid scintillation counting in a Packard TriCarb 2100 liquid scintillation counter. The total disintegrations per minute in the upper phase were calculated. The total disintegrations per minute and percentage of released fatty acids was calculated and partitioning of released radioactives-free fatty acid considered as described earlier (27).
Alk-SMase activity was determined by a method described previously (12). Briefly, samples (5 μL) were mixed with 95 μL of 50 mmol/L Tris/HCl buffer (pH 9.0) containing 0.15 mol/L NaCl, 2 mmol/L ethylenediaminetetraacetic acid, 6 mmol/L taurocholate (assay buffer), and 0.80 μmol/L [14C] SM (∼8000 disintegrations per minute) and incubated at 37°C for 30 or 60 minutes. For acid-SMase, Tris-Maleate buffer (pH 5.0) containing 0.15 mol/L NaCl and 0.12% Triton X-100 was used as assay buffer. The reaction was terminated by adding 0.4 mL of chloroform/methanol (2:1, vol/vol) followed by centrifugation at 9000g for 3 seconds. An aliquot (100 μL) of upper phase containing the cleaved phosphocholine was analysed for radioactivity by liquid-scintillation counting.
Results are presented as mean values and standard error of the mean (SEM). The data were analysed by nonpaired 2-tailed Jonckheere-Terpstra test and P < 0.05 was considered statistically significant. Wilcoxon-Mann-Whitney test was used as post hoc. F508del mice were compared with mice with no mutation. Graphs and statistics were done with Prism 4 for Windows software (GraphPad, La Jolla, CA).
Body and Organ Size
We examined several parameters on 7 mice homozygous for the F508del mutation (here denoted F508del), 6 heterozygotes, and 6 mice with no mutation (here denoted as wild-type), all of which were 15 to 20 weeks of age. There were 5 males and 14 females randomly distributed in the experimental groups.
The body weights of the F508del animals were significantly reduced. These animals weighed 20% less than wild-type mice and 10% less than heterozygous animals (Fig. 1A). In F508del animals, the length of small intestine was increased by about 30% (Fig. 1B) and the weight without content was increased by about 60% (Fig. 1C). Moreover, the weight (grams) per centimeter of colon was 30% higher in F508del mice (Fig. 1D) compared with heterozygotes and wild-type mice. All of the significant values in Figure 1A–D were confirmed by Wilcoxon-Mann-Whitney post hoc test.
Small Intestine and Colon
The small intestine was divided into 4 equal parts; protein concentrations and enzyme activities of alk-SMase and acid-SMase, and neutral- and acid- CDase were measured in mucosal homogenates. There was higher activity (nanomoles per milligrams protein) of alk-SMase and neutral CDase in the 2 middle parts, but no significant differences between the genotypes (Fig. 2A and B). In distal parts, the activity of acid-SMase was almost 30% lower in F508del mice than in controls (P < 0.05) (Fig. 2C), but this significance was only confirmed in the third part by Mann-Whitney post hoc test. There was no significant difference of acid-CDase activity (Fig. 2D). In intestinal contents, there were no significant differences in the enzyme levels (nano- or picomoles per milliliter) between the genotypes (Fig. 3A–D). Similarly, the activity of alk-SMase and neutral CDase in colonic mucosa of F508del mice was not significantly different compared with controls (Fig. 4A and B), and we found no detectable activity of acid-SMase.
In the lungs the mice had a lower activity of neutral CDase, alk-SMase, and acid-CDase in comparison to the small intestine with no significant differences between the 3 genotypes (Fig. 5A, B, and D). The activity of acid-SMase was in the same range as in the intestine but 10% lower in F508del compared with controls (not significant). The data exclude 1 mouse in the F508del group because of an extreme value (39.5 pmol · mg−1 · hour−1). There was no difference in activities of acid-CDase between the genotypes (Fig. 5C and D).
The data for neutral CDase and acid-SMase activity in liver, kidney, and spleen are shown in Table 1. For acid-SMase activity, there were no statistically significant differences, although higher values were observed in the F508del mice in all 3 tissues; however, the activity of neutral CDase in spleen was 40% higher in the F508del mice compared with controls (P = 0.015). A Mann-Whitney post hoc test did not confirm this significance (P = 0.071). There were no differences in weight between spleens in these genotypes. Neutral CDase activity in homozygotes was unchanged in kidneys and elevated by 10% in livers. In general, the activity of neutral CDase was higher in kidneys and livers than in lungs but lower than in the intestines.
This study examined alk-SMase, neutral CDase, acid-SMase, and acid-CDase in F508del animals with a focus on the intestine but some interesting results in other organs also are discussed. A major finding was that alk-SMase and neutral CDase were normally expressed in the small intestine in this model (Figs. 2 and 3), despite that the mice showed intestinal abnormalities consistent with the CF phenotype (25).
Another interesting observation is that the activity of acid-SMase was evenly distributed throughout the small intestine in the F508del mice, whereas it was increased in distal parts in normal and heterozygous mice (Fig. 2C). In wild-type mice, expression of CFTR protein exhibits a proximal–distal gradient, being highest in the duodenum and lowest in the ileum (28). In view of the inhibitory effect of ceramide on CFTR (8), the lower acid-SMase activity in the ileum may decrease ceramide formation and thereby enhance the residual CFTR function. Another possibility is that the decreased acid-SMase activity in the distal small intestine is related to the exposure of the ileum to bacteria. Further studies investigating these hypotheses are warranted.
Alk-SMase and neutral CDase are both intestinal brush border enzymes, which are in part released by bile salts or by tryptic cleavage (29). Both enzymes have been purified and characterized from luminal bile salt eluates in rat and from human duodenal contents. The pure enzymes were found to be stable during incubation with trypsin and chymotrypsin (27). Thus, the levels in gut lumen reflect the release from the mucosa independent of any difference in pancreatic protease activity between wild-type and CF mice. Our finding that the proportions of enzyme activities found in mucosal tissue and in the gut lumen were similar in F508del, heterozygous, and wild-type animals indicates that the enzymes are processed and transferred to the brush borders in a normal way in CF animals. Accordingly, the vesicular traffic from the site of synthesis to the insertion in the apical membrane, which has been suggested to be a defect in CF (30), must be normal for these enzymes. Based upon this, it is likely that alk-SMase and neutral CDase may participate in the digestion of dietary SM and endogenous sphingolipids in a typical way. It was earlier shown in patients with CF that whereas alkaline phosphatase is lower in small intestinal biopsies, lactase, sucrase, and maltase remained at normal levels (31). Because lactase-phlorizine hydrolase has been implicated in the first step of glycolipid hydrolysis (32), this provides further evidence that sphingolipid digestion may be normal in CF. We did not study the neonatal development of the enzymes in the CF animal model used but postulate that the enzymes participate in milk sphingolipid digestion during suckling in these animals as in normal rats and humans (17).
Both length and weight of the small intestine and weight per centimeter length of the colon in F508del mice were significantly larger than in wild-type mice. Although noticed by others (33), this phenomenon has not initiated detailed morphological or mechanistic studies. It may reflect adaptive changes to increased mechanical stress, defective secretion, or low-grade chronic inflammation, but also a more direct involvement of the CFTR gene in intestinal growth and differentiation. The increased size of the intestine in the F508del mice may to some degree affect the total activity of the SMases and CDases. It is interesting to note that intestinal adaptation to postoperative ileus recently was found to be linked to increased levels of S1P and ceramide phosphate (34). Linking this observation to the finding that S1P is degraded at a slower rate than is typical in CF (5), it may be speculated that phosphorylated sphingolipid metabolites may be involved also in the gut hypertrophy of the F508del CF mouse model observed in this study. In the colon, we found an increase of neutral CDase activity (not significant, P = 0.36) (Fig. 4B) and a slight decrease of alk-SMase activity (not significant, P = 0.69) in F508del mice, which may lead to reduced levels of ceramide. This is in line with the established link between CF and increased frequency of gastrointestinal malignancies such as colonic adenocarcinoma and that sphingolipid signaling is altered in this disorder (35,36).
We have found that alk-SMase is expressed in lungs and in bronchial mucosa at a much lower level than in the small intestine (Duan et al, unpublished data). Others have demonstrated the presence of neutral CDase in the lungs (37). In this study, we confirmed the presence of all 4 enzymes in the lungs, with no significant difference between the wild-type and F508del mice in this respect. The F508del mouse model developed spontaneous lung disease, but our results do not suggest any direct involvement of the 4 enzymes in the inflammatory process (Fig. 5A–D) (38). The physiological role of alk-SMase in the lung is presently unknown. The levels are much lower than in the gut, but are clearly measurable and the enzyme is an ectoenzyme (12). The possibility that SM in the outer leaflet of the plasma membrane, as well as platelet activating factor released by inflammatory cells, may be substrates for the enzyme in the lung therefore remains an interesting possibility. If so, alk-SMase may have complementary functions to acid-SMase (3) under both normal conditions and in patients with CF with airway infections. This possibility needs to be explored further. In this study, the level of acid-SMase activity in lungs was similar to that in the gut.
The activity of acid-SMase in kidneys and liver was not higher in the CFTR mutated mice than in the wild-type (Table 1). In the spleen, however, the activity of neutral CDase was significantly higher in F508del than in heterozygotes and wild-type mice, but a Wilcoxon-Mann-Whitney post hoc test did not confirm this significance (P = 0.071). Higher activity of neutral CDase generates more substrate to the different types of sphingosine kinases expressed in the spleen (39). Because the S1P generated is required by lymphocytes to exit lymphoid organs in response to inflammation (40), one may speculate that the increased splenic neutral CDase level in CF mice may influence lymphocyte traffic.
In summary, this article shows for the first time that the intestine and lungs in F508del CF mice express normal levels of alk-SMase and neutral ceramidase, whereas acid SMase is lower in the ileum. Further studies should examine whether the lower level of acid-SMase in the ileum may be related to gut inflammation and/or the inhibitory effect of ceramide on CFTR. We also found intestinal hypertrophy of F508del animals, which motivates future morphological studies. The higher neutral CDase activity in the spleens of F508del animals may reflect the need of precursors to S1P formation during egress of B and T cells (40).
The skillful assistance of Vivi Adermark is gratefully acknowledged. The statistical analysis was done by biostatistician Fredrik Nilsson at Competence Centre for Clinical Research (Region Skåne, Lund).
1. Chalfant CE, Spiegel S. Sphingosine 1-phosphate and ceramide 1-phosphate: expanding roles in cell signaling. J Cell Sci 2005; 118:4605–4612.
2. Cuvillier O. Sphingosine in apoptosis signaling. Biochim Biophys Acta 2002; 1585:153–162.
3. Gulbins E, Li PL. Physiological and pathophysiological aspects of ceramide. Am J Physiol Regul Integr Comp Physiol 2006; 290:R11–R26.
4. Cuvillier O, Pirianov G, Kleuser B, et al
. Suppression of ceramide-mediated programmed cell death by sphingosine-1-phosphate. Nature 1996; 381:800–803.
5. Boujaoude LC, Bradshaw-Wilder C, Mao C, et al
. Cystic fibrosis
transmembrane regulator regulates uptake of sphingoid base phosphates and lysophosphatidic acid: modulation of cellular activity of sphingosine 1-phosphate. J Biol Chem 2001; 276:35258–35264.
6. Grassme H, Jendrossek V, Riehle A, et al
. Host defense against Pseudomonas aeruginosa
requires ceramide-rich membrane rafts. Nat Med 2003; 9:322–330.
7. Zeidan Y, Ya H. Loss of acid sphingomyelinase
induction by Pseudomonas aeruginosa
in cystic fibrosis
. 30th European Cystic Fibrosis
Conference, Belek, Turkey, June 12–13, 2007.
8. Ito Y, Sato S, Ohashi T, et al
. Reduction of airway anion secretion via CFTR in sphingomyelin
pathway. Biochem Biophys Res Commun 2004; 324:901–908.
9. Ramu Y, Xu Y, Lu Z. Inhibition of CFTR Cl- channel function caused by enzymatic hydrolysis of sphingomyelin
. Proc Natl Acad Sci USA 2007; 104:6448–6453.
10. White NM, Corey DA, Kelley TJ. Mechanistic similarities between cultured cell models of cystic fibrosis
and Niemann-pick type C. Am J Respir Cell Mol Biol 2004; 31:538–543.
11. Lam RS, Shaw AR, Duszyk M. Membrane cholesterol content modulates activation of BK channels in colonic epithelia. Biochim Biophys Acta 2004; 1667:241–248.
12. Duan RD, Nilsson A. Sphingolipid hydrolyzing enzymes in the gastrointestinal tract. Methods Enzymol 2000; 311:276–286.
13. Nilsson A. The presence of sphingomyelin
- and ceramide-cleaving enzymes in the small intestinal tract. Biochim Biophys Acta 1969; 176:339–347.
14. Wu J, Nilsson A, Jonsson BA, et al
. Intestinal alkaline sphingomyelinase
hydrolyses and inactivates platelet-activating factor by a phospholipase C activity. Biochem J 2006; 394:299–308.
15. Amer MD, Hedlund E, Rochester J, et al
. Platelet-activating factor concentration in the stool of human newborns: effects of enteral feeding and neonatal necrotizing enterocolitis. Biol Neonate 2004; 85:159–166.
16. Goggel R, Winoto-Morbach S, Vielhaber G, et al
. PAF-mediated pulmonary edema: a new role for acid sphingomyelinase
and ceramide. Nat Med 2004; 10:155–160.
17. Duan RD, Cheng Y, Jonsson BA, et al
. Human meconium contains significant amounts of alkaline sphingomyelinase
, neutral ceramidase
, and sphingolipid metabolites. Pediatr Res 2007; 61:61–66.
18. Motouri M, Matsuyama H, Yamamura J, et al
. Milk sphingomyelin
accelerates enzymatic and morphological maturation of the intestine in artificially reared rats. J Pediatr Gastroenterol Nutr 2003; 36:241–247.
19. Grubb BR. Ion transport across the normal and CF neonatal murine intestine. Am J Physiol 1999; 277:G167–G174.
20. Van Doorninck JH, French PJ, Verbeek E, et al
. A mouse model for the cystic fibrosis
delta F508 mutation. EMBO J 1995; 14:4403–4411.
21. French PJ, van Doorninck JH, Peters RH, et al
. A delta F508 mutation in mouse cystic fibrosis transmembrane conductance regulator
results in a temperature-sensitive processing defect in vivo. J Clin Invest 1996; 98:1304–1312.
22. Bijvelds MJ, Bronsveld I, Havinga R, et al
. Fat absorption in cystic fibrosis
mice is impeded by defective lipolysis and post-lipolytic events. Am J Physiol Gastrointest Liver Physiol 2005; 288:G646–G653.
23. Strong TV, Boehm K, Collins FS. Localization of cystic fibrosis transmembrane conductance regulator
mRNA in the human gastrointestinal tract by in situ hybridization. J Clin Invest 1994; 93:347–354.
24. Bronsveld I, Mekus F, Bijman J, et al
. Residual chloride secretion in intestinal tissue of deltaF508 homozygous twins and siblings with cystic fibrosis
. The European CF Twin and Sibling Study Consortium. Gastroenterology 2000; 119:32–40.
25. Grubb BR, Gabriel SE. Intestinal physiology and pathology in gene-targeted mouse models of cystic fibrosis
. Am J Physiol 1997; 273(2 Pt 1):G258–G266.
26. Scholte BJ, Davidson DJ, Wilke M, et al
. Animal models of cystic fibrosis
. J Cyst Fibros 2004; 3(Suppl 2):183–190.
27. Olsson M, Duan RD, Ohlsson L, et al
. Rat intestinal ceramidase
: purification, properties, and physiological relevance. Am J Physiol Gastrointest Liver Physiol 2004; 287:G929–G937.
28. Ameen N, Alexis J, Salas P. Cellular localization of the cystic fibrosis transmembrane conductance regulator
in mouse intestinal tract. Histochem Cell Biol 2000; 114:69–75.
29. Wu J, Liu F, Nilsson A, et al
. Pancreatic trypsin cleaves intestinal alkaline sphingomyelinase
from mucosa and enhances the sphingomyelinase
activity. Am J Physiol Gastrointest Liver Physiol 2004; 287:G967–G973.
30. Ameen NA, van Donselaar E, Posthuma G, et al
. Subcellular distribution of CFTR in rat intestine supports a physiologic role for CFTR regulation by vesicle traffic. Histochem Cell Biol 2000; 114:219–228.
31. Van Biervliet S, Eggermont E, Marien P, et al
. Combined impact of mucosal damage and of cystic fibrosis
on the small intestinal brush border enzyme activities. Acta Clin Belg 2003; 58:220–224.
32. Kobayashi T, Suzuki K. The glycosylceramidase in the murine intestine. Purification and substrate specificity. J Biol Chem 1981; 256:7768–7773.
33. Mailleau C, Paul A, Colin M, et al
. Glycoconjugate metabolism in a cystic fibrosis
knockout mouse model. Mol Genet Metab 2001; 72:122–131.
34. Dragusin M, Wehner S, Kelly S, et al
. Effects of sphingosine-1-phosphate and ceramide-1-phosphate on rat intestinal smooth muscle cells: implications for postoperative ileus. FASEB J 2006; 20:1930–1932.
35. Neglia JP, FitzSimmons SC, Maisonneuve P, et al
. The risk of cancer among patients with cystic fibrosis
. Cystic Fibrosis
and Cancer Study Group. N Engl J Med 1995; 332:494–499.
36. Duan RD. Anticancer compounds and sphingolipid metabolism in the colon. In Vivo 2005; 19:293–300.
37. Longo CA, Tyler D, Mallampalli RK. Sphingomyelin
metabolism is developmentally regulated in rat lung. Am J Respir Cell Mol Biol 1997; 16:605–612.
38. Legssyer R, Huaux F, Lebacq J, et al
. Azithromycin reduces spontaneous and induced inflammation in DeltaF508 cystic fibrosis
mice. Respir Res 2006; 7:134.
39. Fukuda Y, Kihara A, Igarashi Y. Distribution of sphingosine kinase activity in mouse tissues: contribution of SPHK1. Biochem Biophys Res Commun 2003; 309:155–160.
40. Schwab SR, Cyster JG. Finding a way out: lymphocyte egress from lymphoid organs. Nat Immunol 2007; 8(12):1295–1301.