Effects of structurally different lipid emulsions on human neutrophil migration

Clinical Nutrition (2000) 19(5): 327–331 & 2000 Harcourt Publishers Ltd doi:10.1054/clnu.2000.0113, available online at http://www.idealibrary.com on

ORIGINAL ARTICLE

E¡ects of structurally di¡erent lipid emulsions on human neutrophil migration G. J. A.WANTEN*, D. ROOS{ A. H. J. NABER{ Departments of *Gastroenterology and {Hepatology, University Hospital, Nijmegen, {CLB, Sanquin Blood Supply Foundation and Laboratory of Experimental Clinical Immunology, Academic Medical Centre, University of Amsterdam, Amsterdam,The Netherlands (Correspondence to: GJAW, Department of Gastroenterology, University Hospital Nijmegen, Geert Grooteplein Zuid 8, 6525 GA Nijmegen,The Netherlands)

AbstractöAim: To test the hypothesis that structurally di¡erent lipid emulsions have distinct immunomodulatory properties, we analysed neutrophil migration in the presence of various lipid emulsions. Method: Neutrophils of 8 volunteers were pre-incubated in medium or physiological 2.5 mM emulsions containing long-chain (LCT), medium-chain (MCT), mixed LCT/MCT, a-tocopherol-enriched LCT/MCT (LCT/MCT-E) or structured triglycerides (SL). Thereafter, the cells were put on top of 3 mm-pore-sized cell culture ¢lters and incubated for one hour in the presence or absence of a chemo-attractant. Neutrophil migration was measured as the percentage of cells that had passed the ¢lter in the presence (chemotaxis) or absence (random migration) of a chemotactic factor. Results: Compared to lipid-free incubation (19+1%) random neutrophil migration signi¢cantly decreased with LCT/MCT (11+2%), LCT/MCT-E (12+2) and MCT (5+2%), while LCT (18+3%) and SL (20+1%) had no e¡ect. N-formyl-methionyl-leucyl-phenylalanine- (fMLP, 1078 M) or zymosan-activated-serum-induced (ZAS, 10%) ¢lter passage under lipid-free conditions amounted to 61+14% and 70+13%, respectively. These values decreased with LCT/MCT to 11+9% and 15+7%; with LCT/MCT-E to 18+10% and 28+12%; with SL to 39+18% and 57+14%, and with MCT to 5+2% and 10+6%, (all P50.01), while LCT had no e¡ect. Compared to LCT/MCT, the a-tocopherol-enriched formulation signi¢cantly increased ZAS- and fMLP-induced chemotaxis. fMLP-induced chemotaxis decreased in direct proportion to LCT/MCT triglyceride concentration. Conclusions: Human neutrophil migration is distinctively inhibited by structurally di¡erent lipid emulsions, depending on triglyceride chain-length and concentration as well as a-tocopherol content. & 2000 Harcourt Publishers Ltd.

Lipofundin, consists of a mixture of 50% medium-chain triglycerides (MCT, with saturated fatty acids of 6 to 12 carbons) and 50% LCT. LCT/MCT and MCT emulsions decreased chemotaxis in an in vitro study (13), whereas in vivo no effects of LCT/MCT were found (14). Data on the influence of a novel Lipofundin formulation, enriched with a-tocopherol (LCT/ MCT-E), on neutrophil migration are lacking. The same applies for recently developed synthetic structured lipids (SL), with medium-chain and long-chain fatty acids randomly distributed within one triglyceride molecule. It has been suggested that SL emulsions may be clinically advantageous to LCT and LCT/MCT emulsions (15–16). We recently reported that LCT/MCT and MCT emulsions, contrary to LCT and SL, have distinct effects on in vitro neutrophil oxygen radical formation, adhesion and cellular signalling pathways (17–20). In the present study we investigate the effect of various structurally different emulsions, in a physiological concentration (i.e. concentrations reached in the circulation when administering TPN), on random neutrophil migration and chemotaxis.

Key words: lipid; emulsion; neutrophil; migration; human; tocopherol

Introduction Total parenteral nutrition (TPN) is widely used in clinical practice to maintain or improve the nutritional status of patients. Lipid emulsions establish an integral part of TPN regimens, both as an energy source and as essential fatty-acid supply. The increased risk for infectious complications associated with the use of TPN has been related to immune-suppressive effects of the lipid components, especially with respect to phagocyte functions (1–3). Neutrophil migration is an early event in the activation of the non-specific defence system and represents an important aspect of granulocyte function. Intralipid, the most widely used lipid formulation, is a long-chain triglyceride (LCT) emulsion that has shown detrimental effects on neutrophil migration in some (4) but not all (5) in vitro studies, a controversy remaining after several in vivo studies (6–12). Another emulsion, 327

328

LIPID EMULSIONS AND NEUTROPHIL MIGRATION

Materials and Methods

Cell isolation

Eight volunteers (4 male, 4 female, mean age 30+8 years, range 27–45), were included in the study. After an overnight fast, blood samples were drawn and processed as described below. Serum triglyceride content (mean 0.94+0.29 mM, range 0.55–1.56) was determined colorimetrically on a Hitachi 747 analyser.

Granulocytes were purified from blood anticoagulated with lithium heparin (21). The blood, diluted 1:1 with PBS with 0.4% (w/v) trisodium citrate (pH 7.4), was placed on Percoll (r 1.076 g/ml) and centrifuged (7006g, 18 minutes, 258C). The pellet was suspended in 50 ml of ice-cold lysis solution for 10 minutes. After centrifugation (5 min, 4006g, 48C), the remaining erythrocytes were lysed in fresh lysis solution for another 5 minutes. The cells were then washed and resuspended in incubation medium to a final concentration of 26106/ml and were kept at room temperature. Cytospin preparations were 497% pure and 499% viable as determined by May-Gru¨nwald/Giemsa and trypan blue staining.

Reagents Reagents were from Sigma Chemicals (St Louis, MO, USA) unless stated otherwise. N-formyl-methionylleucyl-phenylalanine (fMLP) was stored as a 1-mM stock in dimethyl sulfoxide at 7208C. Hank’s balanced salt solution (HBSS) was from Life Technologies (Paisley, Scotland). Phosphate-buffered saline (PBS) contained Na+ 163.9 mM, Cl7 140.3 mM, HPO2ÿ 4 10.9 mM and H2POÿ 4 1.8 mM (pH 7.4). Isotonic lysis solution contained 155 mM NH4Cl, 10 mM KHCO3 and 0.1 mM EDTA (pH 7.4). Incubation medium contained HBSS supplemented with 0.5% (w/v) human serum albumin (HSA; from Behring, Westwood, MA, USA). Percoll (r 1.129 g/ml at 208C) was from Pharmacia Biotech AB, Uppsala, Sweden. LCT emulsion (Intralipid 20%) was from Pharmacia & Upjohn AB, Stockholm, Sweden. LCT/MCT and a-tocopherol-enriched (170+ 40 mg) LCT/MCT emulsions (Lipofundin 20%) were from B. Braun Melsungen AG, Melsungen, Germany. SL emulsion (Fat Emulsion 73403) was from Pharmacia & Upjohn Parenterals, Stockholm, Sweden. A pure 20% MCT emulsion was provided by Pharmacia & Upjohn (Clayton, NC, USA). For lipid emulsion characteristics see Table 1. Blood samples were collected in 10 ml Monoject tubes (Sherwood Medical, Ballymoney, N. Ireland) with 143 USP units of lithium heparin.

Chemotactic factors fMLP (1078 M final concentration) or zymosan activated serum (ZAS, 10% v/v) were used as chemotactic factors. ZAS was prepared as follows. Zymosan particles were washed twice and resuspended in medium to a final concentration of 10 mg/ml. After incubation (30 min, 378C) of 900 ml of human serum and 100 ml of zymosan particles and centrifugation (10 min, 7006g), a 10% solution (v/v) of the supernatant was used as chemotactic factor. Chemotactic assay Neutrophil chemotaxis was evaluated as described previously, using MilliCell-PC culture plate inserts equipped with Isopore polycarbonate membranes with a diameter of 12 mm, a membrane-thickness of 10 mm and a pore size of 3.0 mm (Sigma Chemicals, St Louis,

Table 1 Composition and characteristics of lipid emulsions according to amanufacturer or bliterature (Steger et al. (28)). NA=none added NIA=no information available LCT Fractionated soy bean oil (g/l) Medium-chain triacylglycerols (g/l) Fatty acid (% w/w of total) Caproic acid (C6:0) Caprylic acid (C8:0) Capric acid (C10:0) Lauric acid (C12:0) Palmitic acid (C16:0) Stearic acid (C18:0) Oleic acid (C18:1) Linoleic acid (C18:2) Linolenic acid (C18:3) Arachidonic acid (C20:4) Structured triacylglycerols (g/l) Mean mol. trigl. weight Fraction. egg phospholipids (g/l) Glycerol (g/l) Tocopherol (mg/l): a b d g pH

LCT/MCT(-E)

200 0 — — — — 9 5 25 55 8 1 0 865 12 22.5 8.8a NIA 43.1a 84.4a 8.0

100 100

16.9b 4.3b 119.8b 44.3b

0.5 28.5 20 1 6.5 2 11 26 4 0.5 0 634 12 25 a 50 (200) NIA NIA NIA 8.0

21.8 NIA 53.9 27.9

SL

MCT

0 0

0 200

0.1 23.4 10.4 0.2 7.5 3.2 16.2 33.2 4.2 — 200 683 12 22.5 NIA NIA NIA NIA 6.5–8.5

2% 50–63% 36–47% +2% — — — — — — 0 505 12 20 NAa NAa NAa NAa 8.0

CLINICAL NUTRITION

MO, USA) (22). Six hundred ml of chemotactic factor in PBS was added to the bottom chamber of the culture plate. After pre-incubation of cells (16106/ml) with 2.5 mM lipid emulsions or medium (10 min, 378C), 400 ml of neutrophil-lipid suspension was loaded above the filter of the cell culture inserts and incubated for 1 hour at 378C. After rinsing of the filter bottom surface to include adherent cells, the total number of cells that had passed the membrane was counted in a hemocytometer. Results are expressed as the percentage of cells that had passed the membrane after incubation, in the presence or absence of chemo-attractant. Light microscopy revealed no significant cell adherence to the membrane bottom surface after rinsing. Statistical analysis Wilcoxon’s signed ranks test (two-tailed) was used for analysis of results. The significance level was set at 0.05. Values are given as mean+standard deviation.

329

Fig. 1 MLP-induced chemotaxis of purified human neutrophils in the presence of various LCT/MCT concentrations. Results, presented as the percentage membrane passage (3-mm pore-size, 10-mm thickness polycarbonate filter) compared to lipid-free conditions after 1 h of incubation at 378C, are representative of four experiments performed in duplicate.

triacylglycerol concentration, reaching a plateau at 2 mM (Figure 1). Results Neutrophil migration under influence of various lipid emulsions and in the presence (chemotaxis) or absence (random migration) of ZAS or fMLP was measured as the percentage of cells that had passed a culture insert membrane with known pore-size and membrane thickness. These results are shown in Table 2. Compared to membrane passage under lipid-free conditions neutrophil chemotaxis significantly decreased with LCT/ MCT, LCT/MCT-E, SL and MCT. All MCT-containing emulsions decreased chemotaxis (with both stimulants) compared to LCT and SL (all P50.02). Compared to lipid-free incubation random neutrophil migration significantly decreased with LCT/MCT, LCT/ MCT-E and MCT, while LCT and SL had no effect. Compared to LCT/MCT, the a-tocopherol-enriched formulation LCT/MCT-E slightly but significantly increased both ZAS- and fMLP-induced chemotaxis (p-values 0.004 and 0.008, respectively), but not random migration. For LCT/MCT the detrimental effect on chemotaxis was shown to be directly proportional to Table 2 Migration of isolated human neutrophils (as % membrane passage of the total numer of cells) after incubation (1 h at 378C) in 2.5 mM lipid emulsions in the presence [chemotaxis, with fMLP or zymosan-activated serum (ZAS)] or absence (random migration) of chemotactic factors. Results in mean+SD of 8 independent experiments. Significant changes compared to lipid-free incubation are indicated: aP50.001, bP50.02 Random migration lipid-free medium LCT LCT/MCT LCT/MCT-E SL MCT

19+1 18+3 11+2a 12+2a 20+1 5+2a

ZAS

FMLP

70+13 67+11 15+7a 28+12a 57+14b 10+6a

61+14 53+10 11+9a 18+10a 39+18a 5+2a

Discussion Our results show, in agreement with studies regarding other aspects of phagocyte function, that structurally different lipid emulsions distinctly influence human neutrophil function in vitro (17–20). One previous study demonstrated both an increased expression of adhesion molecules as well as an increase in functional adhesion in the presence of MCT-, but not LCT- or SL emulsions (19). Increased adhesion and cell ‘stickiness’ due to neutrophil activation may well explain the detrimental effect of these emulsions on cell migration in the present study. This is further supported by the observation that incubation in MCT-emulsions, contrary to LCT, results in a significantly decreased fluorescence anisotropy (23). The latter most probably reflects an increase in membrane fluidity. It has been shown that such modulation of the physicochemical nature of the cell membrane may activate neutrophils (24, 25). Importantly, in this study no evidence for a detrimental effect of the LCT emulsion Intralipid on neutrophil migration was found, supporting the results of previous studies (5,7,9–12). The question remains what emulsion constituent is responsible for the observed differences. Because the emulsifier of both the pure LCT and MCT emulsions (from the same manufacturer) is the same, this factor can be excluded as an explanation. Triglyceride structure is another aspect in which the emulsions under investigation differ. Indeed, the (triglyceride dosedependent) impairment in chemotaxis of cells incubated with medium-chain-fatty-acid (MCFA) containing emulsions, compared to the LCT emulsion indicates

330

LIPID EMULSIONS AND NEUTROPHIL MIGRATION

that triglyceride chain-length is important in the interaction of lipid emulsions and the phagocyte system. Bellinati-Pires et al. reported similar findings in an in vitro study with 10% LCT/MCT and MCT emulsions, employing a similar technique (Boyden chamber) to analyze neutrophil migration (13). The rather mild inhibiting effect of the structured lipid emulsion compared to the LCT/MCT and MCT emulsions in our study might suggest that the MCFA effect is mitigated in the presence of long-chain fatty acids within the same triglyceride molecule. Another discriminator is the content of lipid-soluble anti-oxidant vitamin E (tocopherol) isoforms. Tocopherol occurs as the a-, b-, g- or d-isomer, depending on the number and/or position of methyl groups on the chromanol ring. a-Tocopherol, the form with the highest biological activity, is found in the highest concentration in human plasma and tissues, whereas gtocopherol, with a relative biological activity of 0.15, is present in at least equally high concentrations in the diet (26, 27). Although the a-tocopherol-enriched LCT/ MCT-E formulation showed a (slight) increase in neutrophil chemotaxis compared to LCT/MCT, there was no effect on random migration. Taken together with the fact that the a-tocopherol-content of LCT/MCT is much higher than that of the LCT emulsion (see Table 1) it therefore seems improbable that the effects of the MCT-containing emulsions can be attributed for a significant part to their vitamin E content. In conclusion human neutrophil migration is distinctly influenced by structurally different lipid emulsions, depending on their triglyceride chain-length and concentration as well as a-tocopherol content. These data indicate that triglycerides have distinct immunomodulatory properties, depending on their composition.

Acknowledgements We thank Mr. E. Carter, from Pharmacia & Upjohn Nutrition, for providing us with structured Fat Emulsion 73403 and pure MCT emulsion. This study was supported by grant no. 904-62-189 from the Netherlands Organisation for Scientific Research (NWO).

References 1. Freeman J, Goldmann DA, Smith NE, Sidebottom DG, Epstein MF, Platt R. Association of intravenous fat emulsions and coagulase-negative staphylococcal bacteremia in neonatal intensive care units. N Engl J Med 1990; 323: 301–308 2. Snydman D R, Murray S A, Kornfeld S J, Majka J A, Ellis C A. Total parenteral nutrition-related infections. Prospective epidemiological study using semiquantitative methods. Am J Med 1982; 73: 695–699 3. The Veterans Affairs Total Parenteral Nutrition Cooperative Study Group. Perioperative total parenteral nutrition in surgical patients. N Eng J Med 1991; 325: 525–532 4. Fisher G W, Hunter K W, Wilson S R, Mease A D. Diminished bacterial defence with Intralipid. Lancet 1980; 2: 819–820

5. English D, Roloff J S, Lukens JN, Parker P, Greene H L, Ghishan F K. Intravenous lipid emulsions and human neutrophil function. J Pediatr 1981; 99: 913–916 6. Nordestrom J, Jarstrand C, Wiernik A. Decreased chemotactic and random migration of leukocytes during Intralipid infusion. Am J Clin Nutr 1979; 32: 2416–2422 7. Palmblad J, Brostrom O, Lahnborg G, Voen A M, Venizelos N. Neutrophil functions during total parenteral nutrition and Intralipid infusion. Am J Clin Nutr 1982; 35: 1430–1436 8. Wiernick A, Jarstrand C, Julander I. The effect of intralipid on mononuclear and polymorphonuclear phagocytes. Am J Clin Nutr 1983; 37: 256–261 9. Ota D M, Jessup J M, Barcok C K et al. Immune function during intravenous administration of soybean oil emulsion. JPEN 1985; 9: 23–27 10. Escudier E F, Escudier B J, Henry-Amar M C, et al. Effects of infused Intralipid on neutrophil chemotaxis during total parenteral nutrition. JPEN 1986; 10: 596–598 11. Rasmussen A, Hessov I, Segel E. The effect of Intralipid on polymorphonuclear leukocytes. Clin Nutr 1988; 7: 37–41 12. Herson V C, Block C, Eisenfeld L, et al. Effects of intravenous fat infusions on neonatal neutrophil and platelet function. JPEN 1989; 13: 620–622 13. Bellinati-Pires, Waitzberg D L, Salgado M M. Carneiro-Sampaio M M S. Effect of medium- and long-chain triglycerides on human neutrophil migration. Braz J Med Biol Res 1992; 25: 369–373 14. Monico R, Dominioni L, Interdonato L et al. Effects of i.v. administration of an MCT-containing fat emulsion on neutrophil function and chemistry. Beitr Infusionsther Klin Ernahr 1988; 20: 36–43 15. Kruimel J W, Naber A H J, van der Vliet J A, Carneheim C, Katan M B, Jansen J B M J. Parenteral administration of structured triglycerides improves nitrogen balance in postoperative patients. JPEN 1997; 21: S34 16. Sandstrom R, Hyltander A, Korner U, Lundholm K. Structured triglycerides were well tolerated and induced increased whole body fat oxidation compared with long-chain triglycerides in postoperative patients. JPEN 1995; 19: 381–386 17. Kruimel J W, Curfs J H A J, Naber A H J, Wenker M A M, Jansen J B M J, Hoogkamp-Korstanje J A A. Physiological concentrations of lipid emulsions and leukocyte function. Clin Nutr 1994; 13 (suppl 1): 2 18. Wanten G J A, Naber A H J, Kruimel J W, Tool A T J, Roos D, Jansen J B M J. Influence of structurally different lipid emulsions on human neutrophil oxygen radical production. Eur J Clin Invest 1999; 29: 357–363 19. Wanten G J A, Geijtenbeek T B, Raymakers R, van der Meer J W M, Roos D, Jansen J B M J, Naber T H J. LCT/MCT- and MCT emulsions increase human neutrophil b2 integrin expression, degranulation and adhesion to ICAM-1, contrary to LCT- and structured lipid emulsions. JPEN 1999; 23: S8 20. Wanten G J A, van Emst-de Vries S E, Jansen J B M J, Naber T H J, Willems P H G M. LCT/ MCT- and MCT-, but not LCT- or structured lipid emulsions, induce oxygen radical formation and potentiate the zymosan-induced increase in cytosolic Ca2þ in human neutrophils in a protein kinase C-dependent manner. JPEN 1999; 23: S9 21. Kuijpers T W, Tool A T J, van der Schoot C E, Ginsel L A, Onderwater J J M, Roos D, Verhoeven A J. Membrane surface antigen expression on neutrophils: a reappraisal of the use of surface markers for neutrophil activation. Blood 1991; 4: 1105–1111 22. Van Lent P L E M, Van den Hoek A E M, Van den Bersselaar L A M, Spanjaards M F R, Van Rooijen N, Dijkstra C D, Van de Putte L B A, Van den Berg W B. In vivo role of phagocytic synovial lining cells in onset of experimental arthritis. Am J Pathol 1993; 143: 1226–1237 23. Wanten G, Naber A, Roos D. Human neutrophil membrane fluidity after incubation in structurally different lipid emulsions. Clin Nutr 1999; 18: S9 24. Miyahara M, Watanabe S, Okimasu E, Utsumi K. Chargedependent regulation of NADPH oxidase activity in guinea pig polymorphonuclear leukocytes. Biochem Biophys Acta 1987; 929: 253–262

CLINICAL NUTRITION

25. Shpungin S, Dotan I, Abo A, Pick E. Activation of the superoxideforming NADPH oxidase in a cell-free system by sodium dodecylsulfate. Absolute lipid dependence of the solubilized enzyme. J Biol Chem 1987; 262: 5592–5595 26. Parker R S. Dietary and biochemical aspects of vitamin E. Adv Food Nutr Res 1988; 33: 157–232 Submission date: 14 September 1999 Accepted: 4 February 2000

331

27. VERIS: 1990 Vitamin E Abstracts, prepared by Horwitt M K. LaGrange, I L, Henkel, p.p. vii–xi, 1991 28. Steger P J, Mu¨hlebach S F. Lipid peroxidation of IV lipid emulsions in TPN bags: The influence of tocopherols. Nutr 1989; 14: 179–185