F Levenez

CHU de Lyon - Groupement Hospitalier Edouard Herriot, Lyons, Rhône-Alpes, France

Are you F Levenez?

Claim your profile

Publications (22)80.23 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: The trefoil factor TFF3 is a peptide predominantly produced by mucus-secreting cells in the small and large intestines. It has been implicated in intestinal protection and repair. The mechanisms that govern TFF3 secretion are poorly understood. The aim of this study was, therefore, to evaluate the influence of neurotransmitters, hormonal peptides and mediators of inflammation on the release of TFF3. For this purpose, an isolated vascularly perfused rat colon preparation was used. After a bolus administration of 1 ml isotonic saline into the lumen, TFF3 secretion was induced by a 30-min intra-arterial infusion of the compounds to be tested. TFF3 was evaluated in the luminal effluent using a newly developed radioimmunoassay. TFF3 was barely detected in crude luminal samples. In contrast, dithiothreitol (DTT) treatment of the effluent revealed TFF3 immunoreactivity, which amounted to about 0.3 pmol min(-1) cm(-1) in the basal state. Gel chromatography of DTT-treated luminal samples revealed a single peak that co-eluted with the monomeric form of TFF3. TFF3 was not detected in the portal effluent. Bethanechol (10(-6)-10(-4) M), vasoactive intestinal peptide (VIP, 10(-8)-10(-7) M) or bombesin (10(-8)-10(-7) M) induced a dose-dependent release of TFF3. In contrast, substance P evoked a modest release of TFF3, whereas calcitonin gene-related peptide (CGRP), somatostatin, neurotensin or peptide YY (PYY) did not modify TFF3 secretion. The degranulator compound bromolasalocid, 16,16-dimethyl PGE2 (dmPGE2) or interleukin-1-beta (IL-1-beta) also evoked a marked release of TFF3. In conclusion, TFF3 in the colonic effluent is present in a complex. This association presumably involves a disulfide bond. Additionally, the present results suggest a role for enteric nervous system and resident immune cells in mediation of colonic TFF3 secretion.
    Regulatory Peptides 10/2001; 101(1-3):35-41. · 2.06 Impact Factor
  • Gastroenterology 01/2001; 120(5). · 12.82 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The intestinal peptide guanylin regulates the electrolyte/water transport in the intestinal epithelium. The aim of the present study was to investigate the mechanisms that modulate its secretion in the isolated vascularly perfused rat colon by using a specific guanylin RIA. Intraarterial infusion of bethanechol (10(-4) M) or bombesin (10(-7) M) elicited a significant 6-fold increase in the release of guanylin immunoreactivity (G-IR) in the lumen. Bombesin-stimulated G-IR secretion was strongly reduced by tetrodotoxin, whereas atropine had no effect. VIP (10(-7) M) induced a moderate release of G-IR, whereas substance P, calcitonin gene-related peptide, peptide YY, somatostatin, and neurotensin were without effect. Dimethyl-PGE2 (1.4 x 10(-5) M) or interleukin-1beta (2.5 x 10(-10) M) induced a 3-fold increase in G-IR in the lumen, whereas the degranulator compound bromolasalocid did not stimulate guanylin secretion. Forskolin (10(-5) M) or sodium nitroprusside (10(-4)-10(-3) M) induced a significant release of G-IR. In contrast, PMA (10(-7) M) or ionophore A23187 (10(-6) M) did not modify basal secretion of G-IR. Upon stimulation of guanylin release with bombesin or bethanechol, an increase in G-IR in the portal effluent was also detected. The release of G-IR in the portal effluent was 40-fold lower than that of G-IR into the luminal perfusate. Additionally, analysis with gel chromatography revealed that the immunoreactive material released in the lumen or in the portal effluent coeluted with the 15-amino acid peptide originally isolated from rat intestine. In conclusion, the present data suggest that the enteric nervous system and immune cells may modulate guanylin release from the rat colon. The release of guanylin in the lumen and portal effluent suggests that this peptide may exert both luminal/paracrine and hormonal effects.
    Endocrinology 08/2000; 141(7):2594-9. · 4.72 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Truncated glucagon-like peptide (GLP)-1 is a potent incretin. Its synthesis and secretion are modulated by food, but the influence of individual nutrients remains to be established. The hypothesis that protein hydrolysates (peptones) can directly regulate both GLP-1 secretion and proglucagon (PG) gene transcription was tested in this study, ex vivo in the isolated vascularly perfused rat intestine and in vitro in the murine enteroendocrine cell line STC-1. Peptones were albumin egg hydrolysate (AEH) and meat hydrolysate (MH). We demonstrate in these two models that peptones dose-dependently stimulate GLP-1 release, whereas isocaloric quantities of bovine serum albumin or of an amino acid mixture had no stimulatory effect. A strong and rapid increase of PG RNA level was observed in STC-1 cells treated with peptones (14-fold and 7-fold increase after 4 h of incubation with 3% wt/vol MH and AEH, respectively). Peptones also increased the PG RNA level in the colonic PG-expressing cell line GLUTag. In contrast, peptones did not modify the PG RNA level in two pancreatic glucagon-producing cell lines, namely, the RINm5F and INR1G9 cells. The peptone effect in STC-1 cells was completely abolished by blocking transcription before MH treatment. The stability of proglugacon transcripts was not modified by MH treatment, but nascent transcripts were more abundant in STC-1 cells preincubated with MH. Finally, MH treatment strongly stimulated (15-fold stimulation) the transcriptional activity of two PG gene promoter fragments (-1100 and -350 base pair) linked to the CAT reporter gene transiently transfected in STC-1 cells. Overall, peptones evoke an as yet undescribed release of GLP-1 when brought into contact with native intestinal L-cells or with STC-1 enteroendocrine cells. The increased transcription of the glucagon gene in the latter system suggests an important role of protein hydrolysates in the control of not only the secretion but also the synthesis of the incretin hormone.
    Diabetes 08/1998; 47(7):1038-45. · 7.90 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We investigated the role of low-doses of bombesin in the regulation of exocrine secretion in the pancreas of the conscious pig. In ten growing castrated male Large White pigs, bombesin was infused intravenously for 1 h at doses of 0 to 500 pmol/kg/h under a stimulation of secretin (36 pmol/kg/h). In six pigs, bombesin (50 pmol/kg/h) was administered alone for 2 h and its effect on pancreatic secretion was compared to that of an infusion of secretin. The pancreatic juice and the blood were collected at 15-min intervals for use in assays of protein in the juice and gastrin in the plasma. When bombesin was infused alone or in combination with secretin, the volume secreted was not altered. The protein output was not altered by secretin, but was increased by the infusion of bombesin, in a dose-dependent manner, reaching a plateau at 250 pmol/kg/h. The plasma gastrin levels were increased by bombesin, starting with the 50 pmol/kg/h dose. This effect was maximal at a dose of 100 pmol/kg/h. The levels remained below those measured after a standard meal, demonstrating that the effect of bombesin on the studied parameters is of physiological significance.
    Regulatory Peptides 05/1998; 74(1):41-5. · 2.06 Impact Factor
  • Gastroenterology 01/1998; 114. · 12.82 Impact Factor
  • Gastroenterology 01/1998; 114. · 12.82 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Studies performed on pig indicated that its pancreas was insensitive to the gastrointestinal hormone cholecystokinin (CCK) and suggested that its secretions were rather under the control of the neurotransmitter acetylcholine. This study was performed to determine reasons for this insensitivity by comparing secretory responses to different secretagogues and establishing the dominant CCK receptor type. Pancreatic acini prepared from weaned piglets were evaluated for their sensitivity to carbamylcholine (Cch), caerulein, JMV-180, and secretin. RNA were extracted for CCK-A and CCK-B receptor expression using specific cRNA probes. Results indicate that pig pancreatic acini are sensitive to Cch and relatively insensitive to caerulein with no response to JMV-180, a CCKA agonist, or secretin; MK-329, a CCK-A receptor antagonist, significantly inhibited caerulein-induced enzyme secretion from 10(-8) M. The pig pancreas expresses few CCK-A mRNA receptors but a majority of CCK-B. These data demonstrate that the pig pancreas expresses a majority of CCK-B receptors. In conclusion, the pig pancreas possesses a large majority of CCK-B receptors responsible for their low sensitivity to CCK.
    The American journal of physiology 09/1996; 271(2 Pt 1):E397-402. · 3.28 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The effect of cholecystokinin (CCK) on the pancreas was investigated in the pig in two experiments. Fifteen pigs were fed a diet containing 17 or 48% protein with or without MK329 (4.5 mg per meal). MK329 enhanced the postprandial peak of plasma CCK during the first 30 min, but pancreas adaptation to high protein was not affected. Sixteen pigs were divided into two groups: 12 pigs were infused with CCK-8 + secretin for 1 h and four pigs received a standard meal. In both groups, pancreatic secretion tests were performed under infusion of the vehicle alone or with MK329. After CCK + secretin, MK329 (65-500 micrograms/kg/h) did not alter CCK plasma levels and reduced the early pancreatic protein response by about 30%. Enzyme outputs in pancreatic juice were modestly affected by MK329. After the meal, MK329 (500 micrograms/kg/h) doubled the postprandial peak of plasma CCK and lowered the pancreatic protein output by 35-40% for the first 30 min. We suggest that (a) pancreatic adaptation to high dietary protein is not mediated via CCK-A receptors and (b) the stimulation of pancreatic protein secretion by a meal or by exogenous CCK-8 is mediated partly by CCK-A receptors.
    Pancreas 08/1995; 11(1):86-94. · 2.95 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Secretory proteins are segregated into two pathways out of the trans-Golgi network of regulated secretory cells. To identify proteins specifically secreted by pathways other than the one leading to zymogen granule exocytosis in the exocrine pancreas, conscious permanently cannulated pigs were perfused with atropine to inhibit the regulated fusion of granules. Atropine almost totally inhibited the protein secretion after 1 h of perfusion. The secretion of GP-2, a glycosyl phosphatidylinositol-anchored protein of the zymogen granule membrane, was partially inhibited but was never totally abolished by atropine perfusion. The pattern of proteins secreted under atropine was almost totally different. Soluble GP-2 was the major secretory product. Its specific activity increased 60 times over its normal level in all other conditions. This secretion clearly originated from nonregulated pathways. Results suggest that during the atropine block the apical plasmalemma could be the source of the released GP-2 and that the sustained nature of this release is compatible with a replenishment of the plasmalemma with GP-2 by the continuous exocytosis of vesicles from the nonregulated pathways.
    The American journal of physiology 12/1994; 267(5 Pt 1):G764-71. · 3.28 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this work was to study whether stimulation or destruction of sensory afferents can modulate pancreatic secretion. The neurotoxin capsaicin is specific for a subpopulation of small diameter primary afferent neurons. Small doses of capsaicin were administered to anesthetized rats as intraduodenal or intragastric bolus injections to stimulate digestive sensory fibers, and pancreatic secretory response was measured. In addition, several high-dose subcutaneous capsaicin injections were administered 10 days before the experiments began, in order to inactivate sensory fibers. Basal and 2-deoxy-D-glucose (2DG)-stimulated pancreatic secretion was then measured. Intraduodenal capsaicin (96-3,050 micrograms/kg) induced a progressive (peak response 40-60 min after the injection), dose-related and long-lasting (> 180 min) increase in pancreatic output of sodium, bicarbonate, and total protein. The maximal response was obtained with 964 micrograms/kg capsaicin; it amounted to about 15% of the maximal response to exogenous cholecystokinin octapeptide (CCK8). The response was not decreased by atropine, hexamethonium, vagotomy, a mixture of adrenoceptor antagonists (prazosin + idazoxan + propranolol), or by the CCKB receptor antagonist L365,260. In contrast, the CCKA receptor antagonist L364,718 reduced by 30-40% the sodium and bicarbonate response and reduced by 90% the protein response induced by capsaicin, but not the response induced by methacholine or 2DG. However, intraluminal capsaicin did not release CCK in a preparation of isolated perfused duodeno-jejunum. Intragastric capsaicin did not significantly change pancreatic secretion. Capsaicin pretreatment had no effect on basal and 2DG-stimulated secretion, but abolished the response to intraduodenal capsaicin. In conclusion, intraduodenal capsaicin can stimulate external pancreatic secretion in anesthetized rats through capsaicin-sensitive sensory neurons.(ABSTRACT TRUNCATED AT 250 WORDS)
    Pancreas 03/1994; 9(2):203-11. · 2.95 Impact Factor
  • S. Beucher, F Levenez, M Yvon, T Corring
    http://dx.doi.org/10.1051/rnd:19940611. 01/1994;
  • S BEUCHER, F LEVENEZ, M YVON, T CORRING
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigated the ability of gastric digestive products from casein to stimulate cholecystokinin release by intestinal cells using the isolated vascularly perfused rat duodenojejunum. Casein digests were prepared with an in vitro system simulating gastric digestion and emptying.The luminal infusion of the digesta emptied from the artificial stomach for the first 10 minutes produced a sharp rise of portal cholecystokinin-like immunoreactivity to 300% of basal, followed by a well-sustained plateau secretion until the end of the infusion. The residual casein fraction of this digest brought about a modest cholecystokinin secretion, while the peptide component was as strong a stimulant as total digest. The peptide responsible for this effect was the glycomacropeptide that is a glycosylated fragment (106–169) of κ-casein. Only the slightly glycosylated forms of the peptide originating from variant A of κ-casein were active. The carbohydrate-free peptide did not alter basal cholecystokinin. The highly glycosylated forms of the peptide and the slightly glycosylated peptide from κ-casein variant B induced only a transient and low rise of portal cholecystokinin. The removal of N-acetylneuraminic acid from the active peptide suppressed its effect, while the infusion of an N-acetylneuraminic acid solution induced only a very low response.It is concluded that the glycomacropeptide released from dietary casein during gastric digestion can stimulate cholecystokinin release by intestinal cells in the rat. A well-defined structure is required for the peptide activity. A part of the peptide chain and some glycosidic chains containing N-acetylneuraminic acid, especially those bound to the amino acid residue threonyl 31 of caseinomacropeptide variant A, would be involved in this structure.
    Journal of Nutritional Biochemistry - J NUTR BIOCHEM. 01/1994; 5(12):578-584.
  • [Show abstract] [Hide abstract]
    ABSTRACT: In conscious pigs, i.v. infusion of serial doses of neurotensin (NT, actual doses of 0.24-59.4 pmol.kg-1.min-1) on a background of secretin resulted in a linear increase of plasma NT-like immunoreactivity (NT-LI), as measured with an antiserum that requires the biologically active C-terminal part of the molecule for recognition. The NT infusions were accompanied by a dose-related increase of pancreatic volume and bicarbonate and protein output. The threshold plasma NT-LI concentrations for significant increases of pancreatic enzymes and of pancreatic fluid and bicarbonate were 46.6 +/- 7.0 and 161.3 +/- 10.8 pM, respectively. Food intake was followed by a sharp pancreatic response and a progressive increase of plasma NT-LI level to a peak of about 27.7 +/- 3.0 pM from the basal level 11.8 +/- 1.6 pM. The carbohydrate fraction of the meal was predominantly responsible for the NT release observed after meal intake. High-performance liquid chromatography analysis of plasma samples collected during NT infusion or after meal ingestion revealed one immunoreactive peak coeluting with intact NT. Pancreatic polypeptide was released on infusion of high, supraphysiological doses of NT, while plasma somatostatin remained at low basal values. It is concluded that in the pig, intact NT is released after a standard meal predominantly through the carbohydrate fraction of the diet, NT is a stimulant of exocrine pancreas secretion, and the modest pancreatic response to physiologic increments of plasma NT-LI on peptide infusion is not attributable to indirect inhibition of the exocrine pancreas by NT-released pancreatic polypeptide or somatostatin.
    Pancreas 06/1990; 5(3):306-13. · 2.95 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Food intake enhances the release of intestinal cholecystokinin (CCK) in the pig but the contribution of individual nutrients to the CCK response has not yet been established in this species. Six hogs (mean weight 50 kg) were fitted with a duodenal fistula for instillation of nutrients and with portal (PV) and carotid (CA) catheters for blood sampling. After a 24-h fast, the animals received 1,000 ml of isotonic solution containing 440 kcal of carbohydrate (starch hydrolysate), or of protein (casein hydrolysate) or fat (Intralipid) or a control saline solution by 60-min intraduodenal perfusion after a 60-min control period during which the animals received saline. Portal and peripheral blood samples were collected at 15-min intervals for CCK radioimmunoassay. Intraduodenal perfusion of fat provoked a sharp increase in CCK-Like immunoreactivity (CCK-LI) in PV (peak 76.6 +/- 12.2 pM from basal 10.8 +/- 1.2 pM) and in peripheral blood (peak 46.7 +/- 8.4 pM from basal 9.1 +/- 1.0 pM). The protein hydrolysate induced a transient increase in plasma CCK-LI during the first 30 min of intestinal perfusion (PV: peak 40.1 +/- 5.0 pM from basal 11.9 +/- 1.4 pM; CA: 31.8 +/- 4.0 pM from basal 8.5 +/- 0.8 pM). The transient effect of proteins on CCK release might reflect the consequence of somatostatin release from intestinal stores. Starch hydrolysate promptly raised plasma CCK-LI level to a plateau value (PV: 52.5 +/- 13.1 pM from basal 11.9 +/- 1.4 pM; CA: 35.4 +/- 8.0 from basal 8.5 +/- 0.8 pM).(ABSTRACT TRUNCATED AT 250 WORDS)
    Reproduction Nutrition Development 02/1990; 30(2):267-75. · 1.90 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Fourteen castrated male Large White pigs, weighing 42.5 +/- 1.0 kg, were fitted with biliary and duodenal fistulae for biliary secretion studies. Furthermore, catheters were placed in a carotid artery for blood sampling and in a jugular vein for peptide infusion. Bile was automatically restituted to the animals and continuously sampled for analysis on experimental days. Following an 8 day recovery period, infusion studies were performed after an overnight fast. After a 30 min basal period, sustained biliary flow and bile acid output were obtained and maintained throughout the assay with secretin (36 pmol/kg/h) and CCK-8 (600 pmol/kg/h) infusion. Then, 200, 400, 600, 800 or 1200 pmol/kg/h of porcine pancreatic polypeptide (PP) were infused for 60 min. Secretin plus CCK infusion was continued for 1 h after PP infusion was stopped. Each dose of PP was given on a separate day. Biliary flow was not affected by PP except for the dose of 400 pmol/kg/h. On the contrary, bile acid concentration and output decreased with the lowest dose of PP (200 pmol/kg/h). As soon as the first dose of PP was infused, bile acid concentration and output fell to about 60% of values obtained with secretin plus CCK. Plasma levels of PP were below or similar to postprandial values for 200, 400 and 600 pmol/kg/h and they were significantly larger with 800 and 1200 pmol/kg/h. Bile acid concentration and output did not return to values obtained with secretin plus CCK infusion after cessation of PP infusion. In conclusion, porcine PP given in physiological doses to the pig decreases bile acid output whereas biliary flow remains unaffected.
    Regulatory Peptides 02/1990; 27(1):139-47. · 2.06 Impact Factor
  • http://dx.doi.org/10.1051/rnd:19900213. 01/1990;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In conscious pigs, intravenous infusion of serial doses of cholecystokinin octapeptide (CCK8; 2.9-232.3 pmol.kg-1.min-1) upon a background of secretin resulted in a linear increase of plasma CCK-like immunoreactivity (CCK-LI) concentration and evoked a dose-related increase of pancreatic volume and bicarbonate and protein outputs. The threshold plasma CCK-LI concentration for significant pancreatic response was 103.8 +/- 10.2 pM using a CCK8 dose of 8.8 pmol.kg-1.min-1. The maximum pancreatic response was observed for a plasma CCK-LI level of 498.0 +/- 15.3 pM using 77.2 pmol CCK8.kg-1.min-1. In anesthetized pigs, the threshold plasma CCK-LI concentration for pancreatic response was 1500 pM (actual CCK8 dose of 60.3 pmol.kg-1.min-1). The physiological relevance of this finding was assessed by comparing the food-induced increase of pancreatic secretion with that of plasma CCK-LI. Food ingestion was followed by a sharp pancreatic response and by a progressive increase of plasma CCK-LI to a peak increment of about 15 pM. Gel chromatography of portal and peripheral plasma from fed animals revealed three major peaks in the volumes of CCK33/39 and CCK8, and in a volume intermediate between CCK33/39 and CCK8. An additional minor component eluted ahead of CCK33/39. CCK8, which is one of the CCK components released after food intake, appears to be a fairly weak pancreatic stimulant in pigs.
    Canadian Journal of Physiology and Pharmacology 12/1989; 67(11):1391-7. · 1.56 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Fourteen castrated male Large White pigs, weighing 42.5 +/- 1.0 kg, were fitted with pancreatic and duodenal fistulae for pancreatic secretion studies. Moreover, catheters were placed in a carotid artery for blood sampling and in a jugular vein for peptide infusion. Pancreatic juice was automatically restituted to the animals and continuously sampled for analysis on experimental days. Following an 8-day recovery period, perfusion studies were performed after an overnight fast. After a 30-min basal period, sustained pancreatic flow and protein output were obtained and maintained throughout the assay with secretin (36 pmol/kg/h) and CCK-8 (600 pmol/kg/h) infusion. Then, 200, 400, 600, 800 or 1200 pmol/kg/h of porcine pancreatic polypeptide (PP) were infused for 60 min. Secretin + CCK infusion was continued for 1 h after PP infusion was stopped. Each dose of PP was given on a separate day. Neither pancreatic flow nor bicarbonate output were affected whatever the dose of infused PP. On the contrary, protein concentration and output decreased with the lowest dose of PP (200 pmol/kg/h) and the diminution was more pronounced with the other doses. With 600 pmol/kg/h as well as with 800 and 1200 pmol/kg/h of PP, pancreatic protein output fell to about 20% of values obtained with secretin + CCK. Plasma levels of PP were below or similar to postprandial values for 200, 400 and 600 pmol/kg/h and they were significantly larger with 800 and 1200 pmol/kg/h. Protein concentration and output returned to values obtained with secretin + CCK infusion after cessation of PP infusion. In conclusion, porcine PP given in physiological doses to the pig decreases pancreatic protein output whereas pancreatic flow remains unaffected.
    Regulatory Peptides 02/1989; 24(1):55-65. · 2.06 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The peripheral plasma concentrations of immunoreactive motilin, pancreatic polypeptide (PP), somatostatin and gastrin were measured in 7 pigs fasted to 24 h and subsequently fed a standard meal. Plasma motilin peaked during the last part of phase II activity of the migrating myoelectric complex (MMC) sequence (25.2 +/- 2.3 pM), the lowest value being recorded during phase I (10.6 +/- 1.5 pM) after a 24 h fast. Plasma motilin remained at a low level during the digestive pattern of duodenal activity, no fluctuation occurring when the first postprandial MMC recurred. At variance analysis, gastrin and PP were not released phasically with MMC in the fasting state, while at autocovariance both peptides tended to fluctuate during the MMC sequence with positive and negative peaks at regular intervals along MMC cycles. No variation of plasma somatostatin was observed in the fasting animals. These findings argue against a major role of circulating PP, gastrin and somatostatin-like components in the control of fasted and post absorptive duodenal motility in pigs while the role of motilin remains equivocal.
    Regulatory Peptides 11/1988; 23(1):27-35. · 2.06 Impact Factor

Publication Stats

183 Citations
80.23 Total Impact Points

Institutions

  • 2000
    • CHU de Lyon - Groupement Hospitalier Edouard Herriot
      Lyons, Rhône-Alpes, France
  • 1998
    • French Institute of Health and Medical Research
      Lutetia Parisorum, Île-de-France, France
  • 1989–1998
    • French National Institute for Agricultural Research
      Lutetia Parisorum, Île-de-France, France