J Am Soc Nephrol 15: 2366–2372, 2004 as a Uremic Toxin Interferes with Neutrophil Chemotaxis

LUCIANO OTTONELLO,* PAOLA GNERRE,* MARIA BERTOLOTTO,* MARINA MANCINI,* PATRIZIA DAPINO,* RODOLFO RUSSO,† GIACOMO GARIBOTTO,† TOMMASO BARRECA,* and FRANCO DALLEGRI* *Division of Internal Medicine and †Division of Nephrology, Department of Internal Medicine and Medical Specialties, University of Genoa Medical School, Genoa, Italy

Abstract. Leptin is a pleiotropic molecule involved in energy Finally, the serum inhibitory activity can be effectively prevented homeostasis, hematopoiesis, inflammation, and immunity. Hypo- by immune depletion of leptin. The results also show, however, leptinemia characterizing starvation has been strictly related to that leptin by itself is endowed with chemotactic activity toward increased susceptibility to infection secondary to malnutrition. neutrophils. The two activities—inhibition of the cell response to Nevertheless, ESRD is characterized by high susceptibility to chemokines and stimulation of neutrophil migration—could be bacterial infection despite hyperleptinemia. Defects in neutrophils detected at similar concentrations. On the contrary, neutrophils 2ϩ play a crucial role in the infectious morbidity, and several uremic exposed to leptin did not display detectable [Ca ]i mobilization, ␤ toxins that are capable of depressing neutrophil functions have oxidant production, or 2- upregulation. The results dem- been identified. Only a few and contrasting reports about leptin onstrate that leptin is a pure chemoattractant devoid of secreta- and neutrophils are available. This study provides evidence that gogue properties that are capable of inhibiting neutrophil chemo- leptin inhibits neutrophil migration in response to classical che- taxis to classical neutrophilic chemoattractants. Taking into moattractants. Moreover, serum from patients with ESRD inhibits account the crucial role of neutrophils in host defense, the leptin- migration of normal neutrophils in response to N-formyl-methio- mediated ability of ERSD serum to inhibit neutrophil chemotaxis nyl-leucyl-phenylalanine with a strict correlation between serum appears as a potential mechanism that contributes to the establish- leptin levels and serum ability to suppress neutrophil locomotion. ment of infections in ERSD.

The obese (ob) (1) product, named leptin from the Greek by human monocytes (14) as well as the production of TNF term leptos, meaning thin, is a 16-kD nonglycosylated and IL-6 (15). Recent findings show that leptin positively hormone involved in the control of food intake (2). It is modulates mononuclear cell survival by interfering with the predominantly synthesized by adipocytes (3) to limit the intake apoptotic process (16). More striking, hypoleptinemia charac- of food, promote the breakdown of fat, and increase energy terizing starvation is strictly related to increased susceptibility expenditure (4,5). Indeed, spontaneous mutations in leptin or to infection secondary to malnutrition (12,17). Thus, leptin can its result in marked (6,7). Evidence is accu- be considered part of the recently categorized family of mol- mulating that leptin also plays a role in innate and acquired ecules produced by adipose tissue called adipokines, which are immunity (8,9). In fact, leptin and its receptor share structural capable of linking metabolism and immune homeostasis similarities with members of the long-chain helical cytokine (17,18). Adipokines include cytokines such as IL-1, IL-6, family, which includes IL-6, IL-11, and IL-12 (10,11). Con- IFN-␥, and TNF-␣ and chemokines such as IL-8, monocyte sistent with this view, leptin regulates T lymphocyte responses, chemotactic -1 (MCP-1), and macrophage inflamma- and, in particular, it polarizes T helper (Th) cells toward a Th1 tory protein-1␣ (MIP-1␣) (17). phenotype by enhancing proliferation and IL-2 production of Nevertheless, there are some remarkable exceptions to this naive T cells (12). Furthermore, leptin increases the secretion paradigm: In particular, ESRD is characterized by high sus- of TNF-␣, IL-6, and IL-12 by endotoxin-stimulated murine ceptibility to bacterial infection (19) despite high levels of peritoneal macrophages (13). In addition, leptin induces the leptinemia (20,21). In agreement with this observation, another expression and secretion of IL-1 (IL-1Ra) hyperleptinemic condition, obesity (22), is also associated with an increased incidence of infections (23). As far as ESRD is concerned, it is generally assumed that the defects in phago- Received April 20, 2004. Accepted June 28, 2004. cytic polymorphonuclear neutrophilic leukocytes (neutrophils) Correspondence to Dr. Luciano Ottonello, Dipartimento di Medicina Interna e Specialità Mediche, Viale Benedetto XV n. 6, I-16132 Genova, Italy. Phone: plays a crucial role in the infectious morbidity (24), and, ϩ39-010-3538686; Fax: ϩ39-010-3538686; E-mail: [email protected] indeed, several uremic toxins (e.g., molecules that are capable 1046-6673/1510-2366 of depressing neutrophil functions) have been identified Journal of the American Society of Nephrology (25,26). Despite active investigations regarding other immune Copyright © 2004 by the American Society of Nephrology cells, only a few reports about leptin and neutrophils are DOI: 10.1097/01.ASN.0000139321.98029.40 available. Furthermore, these works originated contrasting data J Am Soc Nephrol 15: 2366–2372, 2004 Neutrophil Chemotaxis Inhibition by Leptin 2367

about the capacity of leptin of modulating neutrophil activities. g, 45 min) and immediately tested in the chemotactic assays and in In particular, two papers from the group of Caldefie-Chezet leptin assays. (27,28) show that leptin is capable of triggering the oxidative and locomotory capacities of neutrophils without affecting Neutrophil Preparation phagocytosis. Conversely, Zarkesh-Esfahani et al. (29) did not Heparinized venous blood (10 U/ml heparin) was obtained from observe any direct effect of leptin on neutrophil activation. The healthy male volunteers after informed consent. Neutrophils were aim of the present work was to study the actual capability of isolated by dextran sedimentation and subsequent centrifugation on leptin to modulate neutrophil functional activities and, in case a Ficoll-Hypaque density gradient, as described previously (30). of positive results, to investigate a possible role of this hor- Contaminating erythrocytes were removed by hypotonic lysis (30). Then, neutrophils were washed three times with HBSS and resus- mone in the pathogenesis of neutrophil dysfunctions charac- pended in incubation medium at appropriate concentrations. Final terizing ESRD. cell suspension was Ͼ97% pure and Ͼ98% viable, as determined by usual assays (30). Materials and Methods Culture Medium and Reagents Neutrophil Locomotion Hanks’ balanced salt solution (HBSS; EuroCLone, Wetherby West, Neutrophil locomotion was studied using the leading front method, Yorkshire, UK) mixed with Dulbecco’s PBS (EuroClone; HBBS:PBS as described previously (30). Tests were conducted in duplicate using ␮ ϭ 3:1) containing 1 mg/ml BSA (Sigma Chemical Co., St. Louis, blind well chambers (NeuroProbe, Cabin John, MA) with a 3- m pore MO) was used as incubation medium throughout the study. Ficoll- size cellulose ester filter (Millipore, Milan, Italy) separating the cells ϫ 5 Hypaque (Lympholyte-I), Giemsa stain, and heparin were purchased (4 10 ) from the chemoattractant. After incubation at 37°C for 45 from Cedarlane Laboratories Ltd. (Hornby, Ontario, Canada), Merck min, the filters were removed, fixed in ethanol, stained with Harris (Darmstadt, Germany), and Roche (Milan, Italy), respectively. Fluo- hematoxylin, dehydrated, cleared with xylene, and mounted in Eukitt rescein diacetate, HEPES, N-formyl-methionyl-leucyl-phenylalanine (Kindler, GmbH, Freiburg, Germany). Duplicate chambers were run ␮ (FMLP), human recombinant C5a, and human recombinant leptin in each case, and the distance ( m) traveled by the leading front of ϫ were purchased from Sigma. FITC-conjugated anti-CD11b mAb 44 cells was measured at 400 magnifications; five randomly chosen (IgG1) and recombinant human IL-8 were purchased from Biosource fields were read for each filters. International (Camarillo, CA). Mouse anti-human leptin 44802 mAb 2ϩ was from R&D System Europe (Abingdon, UK). Fura-2 AM and Intracellular [Ca ]i Determination 2',7'-dichlorofluorescin-diacetate (DCFH-DA) were from Molecular Neutrophils (2.5 ϫ 106) were loaded with 2 ␮M fura-2 AM in Probes (Eugene, OR). Endotoxin contamination of the reagents used HBSS-HEPES 10 mM (pH 7.4; 30 min, 37°C, final volume 0.5 ml). was tested by manufacturers or directly by QLC-1000 Assay (Cam- Then, cell suspension was diluted 10-fold with HBSS-HEPES, incu- brex Bio Science Walkersville, Inc., Walkersville, MD). bated for 30 min at 37C°, washed twice, and resuspended in HBSS- HEPES. Fluorescence changes before and after addition of leptin or FMLP were monitored with Perkin-Elmer LS3 spectrofluorometer at Patients an excitation wavelength of 338 nm and an emission wavelength of The study population consisted of 18 patients with ESRD (7 men 510 nm (30). and 11 women; mean age, 62.3 Ϯ 17.8, x Ϯ 1 SD) under hemodialytic ϭ ϭ (n 13) or peritoneal dialytic (n 5) treatment. All were outpatients, Superoxide Anion Release Assay and the diagnoses were as follows: chronic glomerulonephritis (n ϭ The release of superoxide anion was studied by using a modifica- 6), hypertensive nephrosclerosis (n ϭ 9), and vasculitis (n ϭ 3). No tion of the method of Babior et al. (31) as described previously (30). patient had either history or clinical evidence of hepatic or gastroin- Briefly, neutrophils (5 ϫ 105) were incubated (20 min, 37°C, final testinal disease, infection, congestive heart failure, , or other volume 0.5 ml) with 80 ␮M ferricytochrome c, in the absence or endocrinopathies. Patients with active inflammation (C-reactive pro- presence of 300 U/ml superoxide dismutase (SOD). The reactions tein Ͼ10 mg/L) were excluded from the study. Eight patients dis- were then stopped by adding 2 ml of ice-cold 1 mM N-ethyl-male- played clinical signs of cardiovascular disease; six of them were imide, and the superoxide production was determined in the superna- previous smokers. Peritoneal dialysis patients had been peritonitis- tants from the OD550 of samples without SOD minus OD550 of free for at least 3 mo before the study. All patients received vitamin Ϫ samples with SOD using a extinction coefficient of 2.1 ϫ 10 4 M/cm. supplements that contained B vitamins and folic acid. None of the patients was receiving immune suppressive treatments, , or androgenic steroids. Reverse osmosis was used to purify water for Flow Cytometric Assessment of Neutrophil hemodialysis treatment. Serum was obtained from patients and from Oxidative Metabolism eight normal control subjects after informed consent. Leptin serum Flow cytometric analysis of neutrophil oxidative metabolism was concentrations were determined by a RIA method using reagents carried out according to Bass et al. (32), as described previously (33). supplied as a by DRG Instruments GmbH (Marburg, Germany). Briefly, neutrophils were preincubated (15 min, 37°C) with The lowest amount of leptin detectable in serum was 0.5 ng/ml. The DCFH-DA (5 ␮M). During the incubation time, DCFH-DA perme- within- and between-assay mean coefficients of variation were 3.9 and ated the cells, wherein it was cleaved by intracellular esterases to give 4.3%, respectively. Three selected sera from patients were immunode- nonfluorescence DCFH trapped within the cells. After washing in pleted with anti-leptin mAb (10 ␮g/ml). The antibody concentration PBS, the cells were incubated for 15 min at room temperature in the was chosen to achieve Ͼ90% neutralization of cytokine activity, presence or absence of leptin (200 ng/ml). Then, the cells were based on neutralization assays performed by the manufacturer. After incubated for additional 30 min at 37°C in the absence and presence incubation (overnight, 4°C), the sera were ultracentrifuged (10,000 ϫ of 100 nM FMLP. During this period, intracellular hydrogen peroxide 2368 Journal of the American Society of Nephrology J Am Soc Nephrol 15: 2366–2372, 2004 oxidized DCFH to give a green fluorescence DCFH. At the end of the incubation, the reaction was stopped by keeping the samples on ice until flow cytometric analysis was carried out using an EPICS XL flow cytometer (Coulter).

Flow Cytofluorimetric Analysis of CD11b Surface Expression The flow cytometric analysis of CD11b expression by neutrophils that were or were not exposed to 100 nM FMLP and/or 200 ng/ml leptin were performed as previously reported (33). FITC-conjugated anti-CD11b mAb 44 and appropriate isotype-matched mAb of irrel- evant specificity were used for the analysis. All of the flow cytometry experiments were carried with the use of an EPICS XL flow cytom- eter (Coulter).

Statistical Analyses Figure 2. Inhibition of C5a- and IL-8–induced neutrophil migration by human recombinant leptin. Neutrophils were incubated in the Data were expressed as mean Ϯ SD. Differences between two upper compartment of the chemotaxis chamber in the absence or groups were analyzed by Mann-Whitney U test. Differences among presence of various doses of leptin, and their locomotory response to three or more groups were analyzed by Friedman ANOVA test with 1 nM C5a or 1 nM IL-8 in the lower compartment was tested after 45 Dunn post test. Correlations were calculated by Spearman test. Sta- min of incubation. Results are expressed as mean Ϯ 1 SD of four tistical analyses were performed using GraphPad InStat version 4.01 experiments with neutrophils from different donors. 0 versus 50, for Windows (GraphPad Software, San Diego, CA). Differences were P Ͻ 0.05. accepted as significant at P Ͻ 0.05.

Results concentration was Ͻ0.0005 ng/ml. Thus, the observed phe- Leptin Inhibits Neutrophil Chemotaxis nomenon is unlikely related to endotoxin contamination. Nev- Using 10 nM FMLP as chemoattractant, the distance trav- ertheless, control experiments were performed in the presence eled by normal human neutrophils into the filters in 45 min was of 1 ␮g/ml polymyxin B, which inactivates endotoxin. When Ϯ ␮ Ϯ ϭ 125.6 14.4 m(x 1 SD, n 12). As shown in Figure 1, preincubated in medium alone, neutrophil migration to 10 nM the chemotactic response of neutrophils was inhibited by leptin FMLP in the absence or presence of 50 ng/ml leptin was 125.0 placed in the upper compartment of the migration chambers, in Ϯ 15.5 and 81.7 Ϯ 14.4 ␮m(xϮ 1 SD, n ϭ 3), respectively. a dose-dependent manner. Similarly, leptin also inhibited the When preincubated with 1 ␮g/ml polymyxin B, neutrophil chemotactic response of neutrophils to other chemoattractants migration to 10 nM FMLP in the absence or presence of 50 such as IL-8 and C5a (Figure 2). In these conditions, endotoxin ng/ml leptin was 122.0 Ϯ 16.5 and 79.4 Ϯ 10.0 ␮m(xϮ 1 SD, n ϭ 3), respectively.

Inhibition of Neutrophil Chemotaxis by Sera from ERDS Patients Is Related to Serum Levels of Leptin It is known that serum levels of leptin are increased in patients with chronic renal failure (20,21) and that sera from renal failure patients inhibits neutrophil locomotion (34,35). Therefore, the inhibition of neutrophil migration by sera from ERDS patients might be related to serum levels of leptin. In fact, serum from 18 ERDS patients inhibited the neutrophil chemotactic response to FMLP: Neutrophil migration in the presence of 25% ESRD serum in the upper compartment of migration chambers: 82.0 Ϯ 30.0 ␮m/45 min, x Ϯ 1 SD, n ϭ 18 (median, 80.8 ␮m/45 min; 25%, 58.4 ␮m/45 min; 75%, 111.0 ␮m/45 min); neutrophil migration in the presence of 25% control serum in the upper compartment of migration Figure 1. Dose-dependent inhibition of N-formyl-methionyl-leucyl- chambers: 119.3 Ϯ 15.4 ␮m/45 min, x Ϯ 1 SD, n ϭ 8 (median, phenylalanine (FMLP)-induced neutrophil migration by human re- 119.3 ␮m/45 min; 25%, 110.8 ␮m/45 min; 75%, 129.3 ␮m/45 combinant leptin. Neutrophils were incubated in the upper compart- min). Neutrophil migration to FMLP in the presence of ERDS ment of the chemotaxis chamber in the absence or presence of various doses of leptin, and their locomotory response to 10 nM FMLP in the serum versus neutrophil migration to FMLP in the presence of ϭ lower compartment was tested after 45 min of incubation. Results are normal serum: P 0.005. The concentrations of leptin in sera expressed as mean Ϯ 1 SD of three experiments with neutrophils from from ESRD patients were from 1.7 to 137.1 ng/ml (mean, different donors. 0 versus 1, P Ͻ 0.01; 0 versus 10, P Ͻ 0.01; 0 versus 70.10 Ϯ 50.41 ng/ml, x Ϯ 1 SD, n ϭ 18; median, 79.1 ng/mg; 50, P Ͻ 0.001. 25%, 10.2 ng/ml; 75%, 115.5 ng/ml). As depicted in Figure 3, J Am Soc Nephrol 15: 2366–2372, 2004 Neutrophil Chemotaxis Inhibition by Leptin 2369

in response to FMLP, whereas the undepleted serum samples were effective. This finding strongly supports the idea that the inhibitory effects exerted by serum from renal failure patients on neutrophil locomotory responses are actually mediated by leptin.

Leptin as a “Pure” Chemoattractant for Neutrophils When normal neutrophils are incubated in the upper com- partment of migration chambers, without the addition of che- moattractant in the lower compartment, they migrate sponta- neously. Under our experimental conditions, neutrophil spontaneous migration was 59.6 Ϯ 10.2 ␮m (mean Ϯ 1 SD, n ϭ 12). As shown in Figure 5, neutrophils that were exposed to Figure 3. Relationship between the concentration of leptin in 18 leptin that was added to the lower compartment of migration ESRD sera and neutrophil migration to FMLP in the presence of the chambers displayed a bell-shaped dose-response curve charac- same samples. The means of two determinations of leptin concentra- teristic of chemoattractants, reflecting the well-known inhibi- tions in sera were plotted against the levels of migration to 10 nM tion of neutrophil orientation and directed migration in re- FMLP of normal neutrophils in the presence of 25% sera expressed as sponse to high-dose chemoattractant stimulation (36). It is means of two different experiments. noteworthy that the stimulatory activity of leptin was detected at concentrations comparable to those found to inhibit the chemotactic response to FMLP. Control experiments that were neutrophil migration to FMLP in the presence of ESRD sera performed in presence of polymyxin B ruled out possible and leptin concentrations in the same samples is inversely interferences from contaminant endotoxin (data not shown). A correlated (Spearman r ϭϪ0.7110, P ϭ 0.0009). series of experiments were then performed to test the capability of leptin to trigger neutrophil activation and/or to interfere with Leptin-Depleted Sera from Chronic Renal Failure 2ϩ FMLP-induced functional responses. No [Ca ]i mobilization Patients Do not Affect the Chemotactic Response of was observed in Fura-2–loaded neutrophils that were exposed Normal Neutrophils to leptin, and leptin-treated neutrophils maintained their capac- Three inhibitory ESRD sera were selected to be immunode- 2ϩ ity to mount a rapid increase of [Ca ]i in response to FMLP pleted of leptin. After this procedure, the cytokine in each of 2ϩ (data not shown). Consistent with data in [Ca ]i assay, leptin the three sera was undetectable. Then, neutrophil migration to did not stimulate the respiratory burst or impair the FMLP- FMLP in the presence of undepleted and leptin-depleted sera triggered oxidative response by neutrophils determined as was tested in parallel assays. As shown in Figure 4, each of the spectrophotometric analysis of superoxide anion production three leptin-depleted sera did not inhibit neutrophil locomotion

Figure 4. Effect of three leptin-depleted ESRD sera on the neutrophil Figure 5. Dose-dependent induction of neutrophil migration by var- response to FMLP. Locomotory activity of neutrophils incubated in ious doses of leptin. Neutrophils were incubated in the upper com- the upper compartment of the chemotaxis chamber with undepleted partment of the chemotaxis chamber in medium, and their locomotory () or leptin-depleted (□) serum from three ESRD patients (8, 9, and response in the absence or presence of various doses of leptin in the 10) and exposed in the lower compartment for 45 min to 10 nM lower compartment was tested after 45 min of incubation. Results are FMLP. Results are expressed as mean Ϯ 1 SD of four experiments expressed as mean Ϯ 1 SD of five experiments with neutrophils from with neutrophils from different donors. Migration in presence of different donors. 0 versus 1, P Ͻ 0.05; 0 versus 10, P Ͻ 0.05; 0 versus undepleted versus leptin-depleted serum, P ϭ 0.0286. 50, P Ͻ 0.001. 2370 Journal of the American Society of Nephrology J Am Soc Nephrol 15: 2366–2372, 2004 and as flow cytometric analysis of intracellular oxidation of cells toward a further stimulation with the same chemotaxin 2'-7'-dichloroflourescein in DCFH-DA–loaded neutrophils but also prevent the cell locomotory response toward other (data not shown). Finally, in accordance with previous report, chemotaxins (43,44), our results suggest that by means of its leptin did not influence the expression of CD11b on neutro- chemokine-like activity, leptin is capable of inhibiting neutrophil phils and did not affect the CD11b upregulation induced by chemotaxis in response to other neutrophilic chemoattractants. FMLP (data not shown). Stimulation of neutrophils by classical chemoattractants, such as FMLP, IL-8, and C5a, results in a rapid and transient 2ϩ Discussion rise in [Ca ]I, and, indeed, this metabolic response has been Bacterial infections are a major cause of morbidity and considered for a long time a hallmark of chemoattractant- mortality among patients with ESRD (37). The high incidence triggered neutrophil locomotion (45). Nevertheless, several of bacterial infections must be considered a multifactorial pieces of evidence do not support this vision anymore. First, event related to diverse pathogenetic factors, such as advanced protrusive surface activity, gelsolin-actin complexes, and net age, comorbidities, and exposure to infectious risk factors actin assembly can occur in the absence of Ca2ϩ transient (46). during dialysis (19,38,39). A critical role in the pathogenesis of Furthermore, extracellular and intracellular Ca2ϩ chelators do increased susceptibility to infections is exerted by functional not block neutrophil migration in response to chemotactic abnormalities of neutrophils (24,40). These dysfunctions are factors (47,48), and chemotaxis, unlike superoxide anion pro- 2ϩ considered secondary to diverse causes, such as accumulation duction, does not depend on [Ca ]i enhancement in human of calcium within the cells, iron overload, interactions with neutrophils (49). More striking, a fast-growing family of li- biocompatible dialyzers and dialysis solutions, and several gands, such as Substance P, TGF-␤1, fibrinopeptide B, and Fas uremic molecules including low molecular weight ligand, induce neutrophil chemotaxis without increasing 2ϩ (LMWP) (25,26,40). LMWP are a recently categorized class of [Ca ]i levels and without activating oxidative metabolism or proteins that have a MW spectrum ranging from 1,000 to granule exocytosis (30,50,51). Consequently, chemoattractants 50,000 Da and are thought to be crucial for the pathogenesis of can be classified in two functional groups: classical chemoat- uremic syndrome, including malnutrition and increased sus- tractants (fMLP, C5a, and IL-8), which also evoke secretory ceptibility to infections (21,41). As far as neutrophil function is responses such as superoxide anion release or lysosomal de- regarded, six so-called granulocyte-inhibiting proteins (GIP) granulation, and pure chemoattractants, which are devoid of have been purified from ultrafiltrate of dialyzers and peritoneal secretagogue properties. Our results show that neutrophils that 2ϩ effluent (26). GIP exert their inhibitory activity toward differ- were exposed to leptin did not display detectable [Ca ]i ent functional activities of neutrophils, including oxidative mobilization or oxidant production. Consistent with our find- metabolism, chemotaxis, degranulation, and phagocytosis, ings, Zarkesh-Esfahani and et al. (29) recently showed that with consequent impairment of neutrophil-dependent antibac- neutrophils do not express the long form Ob-Rb, the receptor terial defense (21,26). Also, leptin has been categorized as a isoform mainly involved in the regulation of multiple intracel- LMWP, and, indeed, a role for this protein in some uremic lular signaling cascades, including the classic janus-activating manifestations, such as anorexia and weight loss, has been kinase signal transducer and activator of transcription (JAK- suggested (21,41). Here we provide evidence that leptin is STAT) pathway (5,52), which in turn is critical for phospho- 2ϩ capable of inhibiting in a dose-dependent manner neutrophil lipase C–dependent [Ca ]i rise induced by chemokine stim- migration in response to classical chemoattractants (FMLP, ulation (53). On the contrary, neutrophils express the short C5a, and IL-8). Moreover, serum from patients with ESRD form of the leptin receptor Ob-Ra capable of transducing inhibits migration of normal neutrophils in response to FMLP activating signal to the mitogen-activate protein kinase with a strict correlation between serum leptin levels and serum (MAPK) pathway without JAK-STAT activation (54). Accord- ability to suppress neutrophil locomotion. Finally, the serum ingly, TGF-␤1, a pure chemoattractant that is capable of stim- inhibitory activity can be effectively prevented by previous ulating neutrophil migration in a Ca2ϩ-independent manner, immune depletion of leptin. Taking into account the crucial required MAPK activation (55). In other words, it is suggestive role of neutrophils in host immune competence (42), the leptin- that classical chemoattractants, such as FMLP, induce neutro- mediated ability of serum from ESRD patients to inhibit neu- phil activation via both JAK-STAT and MAPK pathways, trophil chemotaxis appears as a potential mechanism contrib- whereas pure chemoattractants, including leptin, require for uting to the establishment of infections in ESRD. In other triggering cell locomotion only MAPK activation, which in the words, leptin must be considered a bona fide GIP detectable in case of leptin is mediated by Ob-Ra constitutively expressed by serum of ERSD patients. Nevertheless, further retrospective neutrophils. and/or perspective studies are necessary to establish a link It has been reported and herein confirmed that neutrophil between hyperleptinemia and increased susceptibility to infec- stimulation by leptin does not induce the upregulation of tions in ESRD patients. Our results confirm that leptin is also CD11b expression (29). Once again these data are in agree- endowed with chemotactic activity toward neutrophils. It is of ment with the well known incapacity of pure chemoattractants, note that the two activities—inhibition of the cell response to such as Substance P, TGF-␤1, to affect the expression of ␤2 FMLP or other chemokines and stimulation of neutrophil mi- (56). On the contrary, an aforementioned report re- gration—could be detected at similar concentrations. Taking garding CD11b expression and our data are in disagreement into account that certain chemotaxins not only desensitize the with data of Caldefie-Chezet et al. (27,28), which show that J Am Soc Nephrol 15: 2366–2372, 2004 Neutrophil Chemotaxis Inhibition by Leptin 2371 leptin is capable of activating an oxidative burst of neutrophils. reverses starvation-induced immunosuppression. Nature 394: We do not have a clear explanation for this discrepancy, but at 897–901, 1998 least one more consideration can be made. Our data suggest 13. Loffreda S, Yang SQ, Lin HZ, Karp CL, Brengman ML, Wang that hyperleptinemia induces deactivation of circulating neu- DJ, Klein AS, Bulkley GB, Bao C, Noble PW, Lane MD, Diehl trophils, which became incapable of responding to a subse- AM: Leptin regulates proinflammatory immune responses. FASEB J 12: 57–65, 1998 quent chemotactic stimulus. On the contrary, leptin-mediated 14. Gabay C, Dreyer MG, Pellegrinelli N, Chicheportiche R, Meier activation of circulating neutrophils should result in clinical CA: Leptin directly induces the secretion of interleukin 1 recep- pictures resembling neutrophilic vasculitis. In fact, hyperlep- tor antagonist in human monocytes. Clin Endocrinol Metab 86: tinemic clinical syndromes (e.g., chronic renal failure), as well 783–791, 2001 as hyperleptinemic animal models, do not show signs of neu- 15. Santos-Alvarez J, Goberna R, Sanchez-Margalet V: Human lep- trophilic hyperactivation. tin stimulates proliferation and activation of human circulating In conclusion, the results suggest that leptin behaves like a monocytes. Cell Immunol 194: 6–11, 1999 chemokine, capable of stimulating neutrophil locomotion and 16. Sanchez-Margalet V, Martin-Romero C, Santos-Alvarez J, Gob- desensitizing the cells to stimulation by another chemoattrac- erna R, Najib S, Gonzalez-Yanes C: Role of leptin as an immu- tant, and that leptin is responsible for chemotactic desensitiza- nomodulator of blood mononuclear cells: Mechanism of action. Clin Exp Immunol 133: 11–19, 2003 tion of neutrophils by sera from patients with chronic renal 17. Matarese G, La Cava A: The intricate interface between immune failure, taken as a disease model of hyperleptinemia. system and metabolism. Trends Immunol 25: 193–200, 2004 18. Rajala MW, Scherer PE: The adipocyte. At the crossroads of energy homeostasis, inflammation, and atherosclerosis. Endocri- References nology 144: 3765–3773, 2003 1. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman 19. Sarnak MJ, Jaber BL: Mortality caused by sepsis in patients with JM: Positional cloning of the mouse obese gene and its human end-stage renal disease compared with the general population. homologue. Nature 372: 425–432, 1994 Kidney Int 58: 1758–1764, 2000 2. Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabi- 20. Widjaja A, Kielstein JT, Horn R, von zur Muhlen A, Kliem V, nowitz D, Lallone RL, Burley SK, Friedman JM: Weight-reduc- Brabant G: Free serum leptin but not bound leptin concentrations ing effects of the plasma protein encoded by the obese gene. are elevated in patients with end stage renal disease. Nephrol Science 269: 543–546, 1995 Dial Transplant 15: 846–850, 2000 3. Maffei M, Fei H, Lee GH, Dani C, Leroy P, Zhang Y, Proenca 21. Clark WR, Gao D: Low-molecular weight proteins in end-stage R, Negrel R, Ailhaud G, Friedman JM: Increased expression in renal disease: Potential toxicity and dialytic removal mecha- adipocytes of ob RNA in mice with lesions of the nisms. J Am Soc Nephrol 13: S41–S47, 2002 and with mutations at the db . Proc Natl Acad SciUSA92: 22. Mantzoros CS: The role of leptin in human obesity and disease: 6957–6970, 1995 A review of current evidence. Ann Intern Med 130: 671–680, 4. Friedman JM, Halaas JL: Leptin and the regulation of body 1999 weight in mammals. Nature 395: 763–770, 1998 23. Lamas O, Marti A, Martinez JA: Obesity and immunocompe- 5. Ahima RS, Flier JS: Leptin. Annu Rev Physiol 62: 413–437, tence. Eur J Clin Nutr 56: S42–S45, 2002 2000 24. Vanholder R, Ringoir S: Infectious morbidity and defects of 6. Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, phagocytic function in end-stage renal disease: A review. JAm Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst JA, Soc Nephrol 3: 1541–1454, 1993 Cheetham CH, Earley AR, Barnett AH, Prins JB, O’Rahilly S: 25. Vanholder R, De Smet R, Waterloosn MA, Van Landschoot N, Congenital leptin deficiency is associated with severe early-onset Vogele P, Hoste E, Ringoir S: Mechanisms of uremic inhibition obesity in humans. Nature 387: 903–908, 1997 of phagocyte reactive species production: Characterization of the 7. Clement KC, Vaisse C, Lahlou N, Chabrol S, Pelloux V, Cassuto role of p-cresol. Kidney Int 47: 510–517, 1995 D, Gourmelen M, Dina C: A mutation in the human leptin 26. Haag-Weber M, Cohen G, Horl WH: Clinical significance of receptor gene cause obesity and pituitary dysfunction. Nature granulocyte-inhibiting proteins. Nephrol Dial Transplant 15: 392: 398–401, 1998 S15–S16, 2000 8. Fantuzzi G, Faggioni R: Leptin in the regulation of immunity, 27. Caldefie-Chezet F, Poulin A, Tridon A, Sion B, Vasson MP: inflammation, and hematopoiesis. J Leukoc Biol 68: 437–446, Leptin: A potential regulator of polymorphonuclear neutrophil 2000 bactericidal action? J Leukoc Biol 69: 414–418, 2001 9. Matarese G, La Cava A, Sanna V, Lord GM, Lechler RI, Fontana 28. Caldefie-Chezet F, Poulin A, Vasson MP: Leptin regulates func- S, Zappacosta S: Balancing susceptibility to infection and auto- tional capacities of polymorphonuclear neutrophils. Free Radic immunity: A role for leptin? Trends Immunol 23: 182–187, 2002 Res 37: 809–814, 2003 10. Madej T, Boguski MS, Bryant SH: Threading analysis suggests 29. Zarkesh-Esfahani H, Pockley AG, Wu Z, Hellewell PG, Weet- that the obese gene product may be a helical cytokine. FEBS Lett man AP, Ross RJM: Leptin indirectly activates human neutro- 373: 13–18, 1995 phils via induction of TNF-␣. J Immunol 172: 1809–1814, 2004 11. Baumann H, Morella KK, White DW, Dembski M, Bailon PS, 30. Ottonello L, Tortolina G, Amelotti M, Dallegri F: Soluble Fas Kim H, Lai CF, Tartaglia LA: The full-length leptin receptor has ligand is chemotactic for human neutrophilic polymorphonuclear signaling capabilities of interleukin 6-type cytokine receptors. leukocytes. J Immunol 162: 3601–3606, 1999 Proc Natl Acad SciUSA93: 8374–8378, 1996 31. Babior BM, Kipnes RS, Curnutte JT: Biological defense mech- 12. Lord GM, Matarese G, Howard JK, Baker RJ, Bloom SR, anisms. The production by leukocytes of superoxide, a potential Lechler RI: Leptin modulates the T-cell immune response and bactericidal agents. J Clin Invest 52: 741–744, 1973 2372 Journal of the American Society of Nephrology J Am Soc Nephrol 15: 2366–2372, 2004

32. Bass DA, Parce JW, Dechatelet LR, Szejda P, Seeds MC, 46. Stossel TP: The mechanical responses of white blood cells. In: Thomas M: Flow cytometric studies of oxidative product forma- Inflammation: Basic Principles and Clinical Correlates, 2nd Ed., tion by neutrophils: A graded response to membrane stimulation. edited by Gallin JI, Goldstein IM, Snyderman R, New York, J Immunol 130: 1910–1917, 1983 Raven Press, 1992, pp 459–475 33. Ghio M, Ottonello L, Contini P, Amelotti M, Mazzei C, Indiveri 47. Meshulam T, Proto P, Diamond RD, Melnick DA: Calcium F, Puppo F, Dallegri F: Transforming growth factor-␤1insu- modulation and chemotactic response: Divergent stimulation pernatants from stored red blood cells inhibits neutrophil loco- of neutrophil chemotaxis and cytosolic calcium response by motion. Blood 102: 1100–1107, 2003 the chemotactic peptide receptor. J Immunol 137: 1954–1960, 34. Salant DJ, Glover AM, Anderson R, Meyers AM, Rabkin R, 1986 Myburgh JA, Rabson AR: Depressed neutrophil chemotaxis in 48. Zigmond SH, Slonczewski JL, Wilde MW, Carson M: Polymor- patients with chronic renal failure and after renal transplantation. phonuclear leukocyte locomotion is insensitive to lowered cytoplas- J Lab Clin Med 88: 536–545, 1976 mic calcium levels. Cell Motil Cytoskeleton 9: 184–189, 1988 35. Brown CC, Gallin JI: Chemotactic disorders. Hematol Oncol 49. Fabbri E, Spisani S, Biondi C, Barbin L, Colamussi ML, Cariani Clin North Am 2: 61–79, 1988 A, Traniello S, Torrini I, Ferretti ME: Two for-Met-Leu-Phe- 36. Devreores PN, Zigmond SH: Chemotaxis in eukaryotic cells: A focus on leukocytes and dictyostelium. Ann Rev Cell Biol 4: OMe analogues trigger selective neutrophil responses. A differ- 2ϩ 649–686, 1988 ential effect on cytosolic free Ca . Biochim Biophys Acta 1359: 37. United States Renal Data System: USRDS 2003 Annual Data 233–240, 1997 Report, Bethesda, National Institutes of Health, Diabetes and 50. Senior RM, Skogen WF, Griffin GL, Wilner GD: Effects of Digestive and Kidney Diseases, 1998 derivatives upon the inflammatory response. Stud- 38. Terpenning MS, Bradley SF: Why aging leads to increased ies with human fibrinopeptide B. J Clin Invest 77: 1014–1019, susceptibility to infection. Geriatrics 46: 77–80, 1991 1986 39. Hoen B, Paul-Dauphin ADH, Kessler M: EPIBACDIAL: A multi- 51. Haines KA, Kolasinski SL, Cronstein BN, Reibman J, Gold LI, center prospective study of risk factors for bacteremia in chronic Weissmann G: Chemoattraction of neutrophils by Substance P hemodialysis patients. J Am Soc Nephrol 9: 869–876, 1998 and transforming growth factor-␤1 is inadequately explained by 40. Haag-Weber, M Horl WH: Dysfunction of polymorphonuclear current models of lipid remodeling. J Immunol 151: 1491–1499, leukocytes in uremia. Semin Nephrol 16: 192–201, 1996 1993 41. Vanholder R, De Smet R, Glorieux G, Argiles A, Baurmeister U, 52. Zabeau L, Lavens D, Peelman F, Eyckerman S, Vanderkerck- Brinet P, Clark W, Cohen G, De Deyn PP, Deppish R, Des- hove J, Tavernier J: The ins and outs of leptin receptor activation. camps-Latscha B, Henle T, Jorres A, Lemke HD, Massy ZA, FEBS Lett 546: 45–50, 2003 Passlick-Deetjen J, Rodriguez M, Stegmayr B, Stenvinkel P, 53. Soriano SF, Serrano A, Hernanz-Falcon P, Martin de Ana A, Tetta C, Wanner C, Zidek W: Review on uremic toxins: Classi- Monterrubio M, Martinez C, Rodriguez-Frade JM, Mellado M: fication, concentration, and interindividual variability. Kidney Int Chemokines integrate JAK/STAT and G-protein pathways dur- 63: 1934–1943, 2003 ing chemotaxis and calcium flux responses. Eur J Immunol 33: 42. Leher RI, Ganz T, Selsted ME, Babior BM, Curnutte JT: Neu- 1328–1333, 2003 trophils and host defense. Ann Intern Med 109: 127–142, 1988 54. Bjorbaek C, Uotani S, da Silva B, Flier JS: Divergent signalling 43. Richardson RM, Haribabu B, Ali H, Snyderman R: Cross-desen- capacities of the long and short isoforms of the leptin receptor. sitization of chemoattractant receptors occurs at multiple levels: Evidence for a role for inhibition of phospholipase C activity. J Biol Chem 272: 32686–32695, 1997 J Biol Chem 27: 27829–27833, 1995 55. Hannigan M, Zhan L, Ai Y, Huang CK: The role of p38 MAP ␤ 44. Uhing RJ, Snyderman R: Chemoattract stimulus-response cou- kinase in TGF- 1-induced signal transduction in human neutro- pling. In: Inflammation: Basic Principles and Clinical Corre- phils. Biochem Biophys Res Commun 246: 55–58, 1998 lates, 3rd Ed., edited by Gallin JI, Snyderman R, Philadelphia, 56. Molad Y, Haines KA, Anderson DC, Buyon JP, Cronstein Lippincott Williams & Wilkins, 1999, pp 607–626 BN: Immunocomplexes stimulate different signalling events 45. Wilkinson PC, Allan RB: Assay systems for measuring leuko- to chemoattractants in the neutrophil and regulate L- cyte locomotion: An overview. In: Leukocyte Chemotaxis, edited and beta 2-integrin expression differently. Biochem J 299: by Gallin JI, Quie PG, New York, Raven Press, 1975, pp 1–24 881– 887, 1994