Transcellular biosynthesis of lipoxin A4 during adhesion of platelets and neutrophils in experimental immune complex glomerulonephritis

Kidney International, Vol. 47 (1995), pp.

1295—1302

Transcellular biosynthesis of lipoxin A4 during adhesion of platelets and neutrophils in experimental immune complex glomerulonephritis AIKATERINA PAPAYIANNI, CHARLES N. SERHAN, M. LAURIE PHILLIPS, HELMUT G. RENNKE,

and HUGH R. BRADY Renal and Hematology/Oncology Divisions, Department of Medicine and Department of Pathology, Brigham & Women's Hospita4 and Renal Section, Medical Service, Brockton/ West Roxbuiy Veterans Affairs Medical Center, Harvard Medical School, Boston, Massachusetts, and the Cytel Corporation, San Diego, California, USA

Transcellular biosynthesis of lipoxin A4 during adhesion of platelets and neutrophils in experimental immune complex glomerulonephritis. Polymorphonuclear neutrophils are important effectors of injury in host defense and inflammation. Many inflammatory diseases are self-limiting, raising the possibility that compounds are generated in vivo during the course of inflammation that inhibit neutrophil recruitment and tissue destruction. Lipoxins, a more recent addition to the families of bioactive eicosanoids, are potential candidates in this regard. Lipoxins are generated via pathways that initially involve the dual lipoxygenation of arachidonic acid and are potent inhibitors of several neutrophil trafficking events

in Vitro. Here, we present evidence that lipoxin A4 is generated in rat kidneys during experimental immune complex-mediated glomerulonephritis in vivo. Renal lipoxin A4 levels were markedly reduced by prior depletion of animals of either neutrophils or platelets, suggesting that most lipoxin A4 generated in vivo was derived from transcellular biosynthetic pathways during platelet-neutrophil interactions. Electron microscopic examination of glomerulonephritic kidneys revealed areas of intimate contact between neutrophils and platelets within the lumen of glomerular capillaries. P-selectin on platelets is an important mediator of platelet-neutrophil adhesion in vitro and in vivo. Prior treatment of animals with a blocking monoclonal antibody (mAb) against P-selectin (mAb CY1747), but not an isotype-matched non-blocking control mAb (mAb PNB1.6), caused striking inhibition of lipoxin A4 generation without

attenuating neutrophil recruitment. Anti-P-selectin niAb also blunted transcellular lipoxin A4 generation during coincubations of activated neutrophils and platelets in vitro. Together, these data suggested that adhesion of activated platelets and neutrophils through P-selectin within glomerular capillaries enhances transcellular lipoxin A4 formation by

Leukocyte adhesion to endothelial cells and other tissue components is essential for leukocyte recruitment in host defense, inflammation, thrombosis, and other vascular events [reviewed in

1]. In addition, leukocyte adhesion may promote tissue injury during inflammation by facilitating the generation of reactive oxygen species, degranulation, antigen presentation, and the generation of pro-inflammatory mediators such as leukotrienes [1, 2]. Since inflammatory responses often resolve without therapeutic intervention, it is likely that inhibitors of leukocyte adhesion are generated in vivo that limit leukocyte recruitment and prevent

unchecked destruction of host tissue. Several lines of in vitro evidence suggest that lipoxins may be important lipid-derived modulators of leukocyte trafficking in inflammation [3—9]. Lipoxins are eicosanoids formed via biosynthetic pathways that initially involve the dual lipoxygenation of arachidonic acid by either 5- and 15-lipoxygenases or 5- and 12-lipoxygenases (hence,

the trivial name litixygenase interaction products) [3, 4]. They possess biochemical structures and spectra of bioactions distinct from other classes of lipoxygenase-derived eicosanoids [3, 4]. We

recently demonstrated that lipoxin A4 (5S,6R,15S-trihydroxy7,9,13-trans-11-cis-eicosatetraenoic acid)(LXA4), at nanomolar con-

transfer of arachidonate intermediates between adherent platelets and

centrations inhibits leukotriene-induced neutrophil-endothelial cell adhesion supported by 2 integrins and P-selectin via distinct actions with neutrophils and endothelial cells, respectively [5]. In addition, LXA4 attenuates neutrophil chemotaxis in vitro and neutrophil migration across endothelial and epithelial cell mono-

neutrophils. Lipoxin A4 inhibits neutrophil chemotaxis, adhesion to endo-

layers induced by a variety of lipid and/or peptide mediators [6—9].

priming lipoxygenase pathways of adherent cells and/or by facilitating the

thelial cells and diapedesis in some in vitro systems, and exposure of neutrophils to lipoxin A4 er vivo in the present study resulted in striking attenuation of their recruitment to inflamed glomeruli. These observations

provide evidence that neutrophil adhesion, in addition to facilitating neutrophil recruitment during the initial stages of inflammation, also promotes transcellular generation of lipid-derived signals that regulate neutrophil trafficking and may contribute to the suppression of neutrophilmediated tissue injury.

In the present study, we assessed LXA4 generation in acute glomerulonephritis, a paradigm of renal inflammation, and defined the relative contributions of leukocyte and transcellular biosynthetic pathways to LXA4 generation in this setting. We explored the influence of cell-cell adhesion on these biosynthetic events and examined the impact of LXA4 on neutrophil trafficking in acute glomerular inflammation.

Methods Received for publication November 2, 1994 and in revised form December 19, 1994 Accepted for publication December 19, 1994

© 1995 by the International Society of Nephrology

Induction of experimental immune complex-mediated glomerulonephritis

Acute immune complex-mediated glomerulonephritis was induced in the left kidneys of adult male Sprague-Dawley rats (150 1295

1296

Papayianni et al: Lipoxin synthesis in glomerulonephritis

to 175 g) by perfusion of left renal artery with 0.8 mg of a

Fig. 1. Morphologic characteristics and renal LXA4 levels in the Con-A!

kidney is relatively spared of disease in this model and serves as an

ulus depicted in Panel A (hematoxylin and eosin) shows hypercellularity of

conjugate of Concanavalin-A and native ferritin (Con-A/ferritin) ferritin model of acute immune complex-mediated glomendonephritis. A-C. Structural details of the inflammatory process in the glomerulus of rat with for two minutes and subsequent intravenous injection of rabbit active glomerulonephritis 15 minutes after administration of anti-ferritin anti-ferritin antibody (2 ml), as detailed previously [101. The right antibody. D and E. Equivalent structures in control animals. The glomer-

internal control. Tissue fixation, staining light, immunofluores- the glomerular tuft predominantly due to infiltrating neutrophils. The inflammatory cells are positive for dichloroacetate esterase, as evidenced cence and electron microscopy, and assessment of glomerular by the red reaction product, depicted in Panel B (see Fig. 3 for quantitaneutrophil infiltration by the dichloroacetate esterase reaction tion). The inflammation is initiated by immune complex formation along were performed as reported previously [101. In some experiments, glomerular and to a lesser extent peritubular capillary walls, as illustrated animals were selectively depleted of either neutrophils or platelets in Panel C, which shows characteristic granular deposits containing rabbit

by treatment with anti-neutrophil or anti-thrombocyte sera, respectively (each 1.5 ml i.p.; Accurate Chemical and Scientific

immunoglobulin G (IgG) (direct immunofluorescence microscopy with FITC-anti-rabbit IgG). F. LXA4 levels, as determined by ELISA [11], in

renal homogenates prepared fifteen minutes after induction of acute Corp., Westbury, NY, USA) 18 hours prior to induction of GN. In glomerulonephritis, a time at which platelet and neutrophil infiltration has

other experiments designed to define the role of P-selectin in neutrophil recruitment and transcellular LXA4 generation, animals were infused intravenously with a blocking anti-P-selectin

occurred [10]. The induction of glomerulonephritis was associated with a marked increase in renal LXA4 levels, a change not observed in shamoperated controls. Rats were selectively depleted of platelets or neutrophils to define the relative contributions of neutrophil or transcellular mAb (CY 1747, also designated PB1.3, 4 mg/kg of body wt, IgGi) or pathways to LXA4 generation. Both neutrophil depletion and platelet an isotype-matched non-blocking anti-P-selectin mAb (PNB1.6, depletion reduced renal LXA4 levels, suggesting that most LXA4 was

IgGi) five minutes prior to administration of anti-ferritin antibody.

formed by transcellular routes by neutrophil-platelet interactions. A

Lipoxin A4 levels

not platelet-depleted animals, if most LXA4 was generated either by neutrophils alone or through their interactions with other cells within

LXA4 levels were monitored in renal homogenates (Polytron PT-MR3000, 3 strokes, 4°C), as an index of LXA4 generation, using a recently developed ELISA for LXA4 (Oxford Biomedical Research, Inc., Oxford, MI, USA) [11]. This ELISA has a lower limit of detection for LXA4 of 90.8 fmol/ml and cross-reactivities

reduction in LXA4 levels would be expected in neutrophil-depleted, but glomeruli. Magnifications are 510X. LXA4 levels are means SE and are expressed as ng/kidney and an asterisk represents a statistically significant reduction (°P < 0.05 for neutrophil depletion and ** < 0.025 for platelet depletion) when compared with acute glomerulonephritis.

with other lipoxygenase products as follows: LXB4 1%; 15-HETE

0.1%; 5-HETE, 12-HETE, leukotrienes, each <0.1% [11]. For assessment of transcellular LXA4 generation in vitro, rat neutrophils and platelets were isolated from heparinized and citrated venous blood, respectively, as previously described for human

Kidneys were harvested 15 minutes after induction of disease and

their radioactivity was measured in a y-counter (LKB Clinig-

cells [12]. LXA4 levels were assessed by ELISA in supernatants of amma) as an index of neutrophil recruitment.

coincubations of neutrophils and platelets (neutrophil : platelet ratio 1:100; platelet density 3 x 109/ml) 20 minutes after stimuMonoclonal antibodies lation with the chemotactic peptide fMLP (10 M) and thrombin (1 U/mI) in the presence of anti-P-selectin (CY 1747) or control CY1747 (PB1.3) and PNB1.6 are murine monoclonal antibodies against P-selectin whose specificity and functional characterrnAb (each 10 ig/ml). istics have been described previously [14]. Briefly, monoclonal Assessment of neutrophil recruitment during acute antibodies (both IgGi isotype) were raised by immunization of glomerulonephritis in vivo RBF/DnJ mice with thrombin-activated human platelets and partially purified human P-selectin, and isolation of hybridoma Lipoxins may potentially modulate neutrophil recruitment in clones. Both CY1747 and PNB1.6 react with recombinant human vivo directly by affecting neutrophil chemotaxis and/or adhesion to P-selectin, purified P-selectin and thrombin-activated, but not endothelial cells, or indirectly by influencing vascular tone and intraglomerular shear stress (Discussion). To explore the possi- resting platelets, as determined by ELISA, immunoblotting and flow cytometry [14]. In contrast, they do not react with either bility that LXA4 directly inhibits neutrophil trafficking in glomer- E-selectin or L-selectin. Of particular relevance to the present ulonephritis, rat neutrophils were isolated from heparinized blood study, both monoclonal antibodies also react with activated rat by standard procedures of Ficoll-Hypaque density gradient cen- platelets, but not resting rat platelets or rat neutrophils, as trifugation followed by dextran sedimentation, as described predetermined by flow cytometry. CY1747 inhibits adhesion of viously for human neutrophils [12, 13]. Contaminating red blood activated rat platelets to rat neutrophils in mixed cell suspensions cells were removed by hypotonic lysis and neutrophils were suspended in Dulbecco's phosphate-buffered saline (PBS) at a in vitro, whereas PNB1.6 does not. density of 5 x 106/ml. Neutrophils were radiolabeled with 111in-

dium by incubation with 11indium oxine (1 Ci/106 cells; Mediphysics, IL, USA) for thirty minutes at room temperature, washed three times in PBS and the final preparation was suspended in PBS at a concentration of 5 X 106 cells/ml. Radiolabeled neutrophils (10 X 106 in 2 ml of PBS) were exposed to LXA4 (10—v M, 15 mm, 37°C) or its diluent, washed and injected

into rats five minutes prior to injection of anti-ferritin antibody.

Statistics

Data are expressed as mean SEM. Differences between group means were determined by analysis of variance and comparisons between groups were made using a two-tailed Student's t-test. A P value of <0.05 was taken to represent a statistically significant difference between group means.

Papayjanni et al. Lipoxin synthesis in glomerulonephritis

1297

rars

.fl;:t A

a

•

4H—

'I

I

V

fl_a S • .j

'

4'

it!

. 4"?er ,g a N

Z. q I

'Mi— —

' C, 0j''

fr.

'p — 2-

14

',

— 4'

r

.

S

t5

0.

a

'1> C'

a4 ,,r 4'

.1' I-

E F Normal

(N_ 6) Sham .operated

(N=3) Co ntra late ra I

I

I

(N=6) Glomerulonephrilis (GN)

(N= 6) GN•PMN depletion IN = 5) GN-platelel depletion

(N= 5)

02

4

6 8 10

Renal LXA4 levels ny/kidney

1298

ea1a wain Papayianni et al: Lipoxin synthesis in glomendonephritis

Fig. 2. Electron microscopic examination of the Con Alferritin model of acute immune complex-mediated glomerulonephritis. The representative electron micrograph in A shows a glomerular capillary loop from a glomerulonephritic animal with infiltrating neutrophils (PMN) and platelets (P) in close contact and partially occluding the lumen. B. An equivalent examination of a normal animal (CL = capillary lumen; US = urinary space; Ep = visceral epithelial cell; RBC = red blood cell; End = endothelial cell). Kidneys were prepared from glomerulonephritic animals 15 minutes after administration of anti-ferritin antibody. Magnifications are 6600X.

Results

synthetase activity" of their 12-lipoxygenase with leukotriene A4

Detection of LXA4 levels in glomerulonephritic kidneys in vivo

(LTA4), an epoxide intermediate produced by neutrophil 5-li-

In the Con-A/ferritin model of immune complex-mediated glomerulonephritis, the Con-Alferritin conjugate localizes predominantly to the glomerular endothelium and lamina rara interna of glomerular basement membrane [10]. Subsequent administration of anti-ferritin antibody causes rapid immune complex formation, complement activation, infiltration of glomeruli by platelets and neutrophils, and functional renal impairment and morphologic injury that bear many of the hallmarks of human acute glomerulonephritis (Fig. 1, A-C) [10, 15]. LXA4 levels were

poxygenase [4, 12, 16, 17]. In the present study, rats were depleted of either neutrophils or platelets, using specific antisera, to define the relative contributions of neutrophil and transcellular pathways

to LXA4 formation in acute glomerulonephritis. The number of circulating leukocytes and platelets after each of these maneuvers was as follows: circulating leukocytes in 10311.d: basal 6.5

0.7 (18

3% neutrophils), anti-neutrophil sera 3.1 0.9 (0.7 0.3% 3.2% neutrophils), anti-thrombocyte sera 4.9 0.8 (23.7 94, neutrophils); circulating platelets in 1031p1: basal 752 51. anti-thrombocyte sera 4 1, anti-neutrophil sera 462 monitored by ELISA in homogenates of glomerulonephritic Neutrophil depletion and platelet depletion were independently kidneys fifteen minutes after induction of disease (Fig. 1, A-C). associated with marked reduction in renal LXA4 levels (Fig. iF), LXA4 was detected in nanogram quantities in glomerulonephritic suggesting that most LXA4 was generated within glomerular kidneys (Fig. iF). Significantly lower levels of LXA4 were de- capillaries by neutrophil-platelet interactions. tected in contralateral kidneys which are relatively spared of disease in this model, and LXA4 was virtually undetectable in kidneys of either normal or sham-operated animals. Biosynthesis Platelet-neutrophil adhesion through P-selectin promotes of LXA4 occurred rapidly, being evident within five minutes of transcellular LXA4 generation in vivo injection of anti-ferritin antibody (data not shown). Electron microscopic examination of glomerulonephritic kidContribution of transcellular biosynthetic pathways to LXA4 neys, performed in parallel with analyses of renal LXA4 generageneration in acute glomerulonephritis tion, revealed platelets and neutrophils in close proximity within Isolated human leukocytes in vitro generate LXA4, upon acti- the lumen of glomerular capillaries (Fig. 2A). Of interest, areas of vation with receptor-mediated stimuli, through the actions of their focal contact between platelets and neutrophils were frequently 5- and 15-lipoxygenases [3, 4]. LXA4 generation is significantly observed (Fig. 2A). enhanced and new biosynthetic pathways are evoked in vitro when P-selectin is a major platelet ligand that supports plateletneutrophils are stimulated in the presence of platelets [4, 12, 16, neutrophil adhesion in vitro, modulates platelet-neutrophil inter17]. In the latter setting, platelets, which lack 5- and 15-lipoxyge- actions during inflammation and thrombosis, and appeared a nase activity, form LXA4 through the "15-oxygenase and LX likely substrate for platelet-neutrophil adhesion in our study [14,

Papayianni Ct at: Lipoxin synthesis in glomerulonephrztis

A

1299

Fig. 3. Impact of anti-P-selectin monoclonal antibody (mAb) on renal

In vivo

Glomerulonephritis (ON)

LXA4 levels and neutrophil infiltration in glomerulonephritis in vivo and LXA4 levels in neutrophil-platelet coincubations in vitro. A. Influence of adhesion-blocking anti-P-selectin mAb (N = 4) and control non-blocking mAb (N = 3) on renal LXA4 levels in Con-A/ferritin glomerulonephritis. Anti-P-selectin mAb inhibited transcellular LXA4 generation in vivo (*P

ON + anti-P-selectin mAb

< 0.05). B. The impact of anti-P-selectin mAb (N = 8) and platelet depletion (N = 6) on glomerular neutrophil infiltration (expressed as

Normal

neutrophils per cross section of glomerulus) in Con-Alferritin glomerulonephritis in vivo. For purposes of comparison, the effect of neutrophil depletion on glomerular neutrophil counts is also provided (N = 4).

ON + control mAb

0.0 B

Platelet depletion, but not anti-P-selectin mAb afforded inhibition of neutrophil infiltration (*P < 0.005), suggesting that anti-P-selectin mAb

12.0 4.0 8.0 Renal LXA4 levels, nglkidney

blocked transcellular LXA4 generation by influencing platelet-neutrophil interactions within the lumen of inflamed glomerular capillaries and not by inhibiting neutrophil recruitment. C. The influence of anti-P-selectin (N = 11) and control mAb (N 9) on transcellular LXA4 generation during

In vivo

coincubations of rat platelets and neutrophils in vitro. As with acute

Normal

glomerulonephritis in vivo, LXA4 levels were significantly lower in Vitro (*P < 0.005) in the presence of anti-P-selectin mAb, providing additional

evidence that P-selectin promotes transcellular LXA4 generation by

Glomerulonephritis (ON)

platelet-neutrophil interaction within a local milieu. Anti-P-selectin mAb (CY 1747, also designated PB1.3, 4 mg/kg of body wt, [14]) waS adminis-

ON + anti-P-selectin mAb

tered intravenously five minutes prior to administntion of anti-ferritin ON + PMN depletion ON + platelet depletion

0.0

0.5

1 .0

1 .5

PMN/glomerulus In vitro a,

a, C (V

U

2.0 The average number of neutrophils per cross section of glomerulus was calculated by analysis of 20 glomeruli per kidney by two independent

15 5

antibody. Importantly, this mAb did not alter circulating leukocyte (6.7 71 103/1.d). 0.5 103/pi; 18.3 1.4% neutrophils) or platelet counts (805 LXA4 was measured by ELISA of renal homogenates prepared 15 minutes after induction of disease, as described in Figure 1. Platelet and neutrophil depletion were achieved using specific antisera as described in Figure 1.

:r

—5

observers in blinded fashion on paraffin sections stained by the dichloroacetate esterase reaction [10]. LXA4 levels were assessed by ELISA in supernatants of coincubations of neutrophils and platelets (neutrophil: platelet ratio 1:100; platelet density 3 x 109/ml) 20 minutes after stimulation with the chemotactic peptide fMLP (l0- M) and thrombin (1 U/mI) in the presence of anti-P-selectin (CY 1747) or control mAb (each 10 sgIml).

Co

a, —15

> a,

—25

x-J

—35

*

—45

Control mAb Anti-P-selectin mAb

vations suggested that platelet-neutrophil adhesion through Pselectin promotes transcellular LXA4 generation within the local intravascular milieu of inflamed glomeruli and not by facilitating recruitment of neutrophils, the cell source of the arachidonate intermediate (LTA4) for platelet LXA4 biosynthesis.

To explore this hypothesis further, the impact of anti-P-selectin 18—231. A specific monoclonal antibody (mAb) against an adhesion epitope on P-selectin [141 was used to probe the role of this mAb on transcellular LXA4 biosynthesis was assessed in vitro ligand in LXA4 biosynthesis in acute glomerulonephritis. Impor- during coincubations of activated neutrophils and platelets. LXA4 tantly, this mAb did not alter the number of circulating leukocytes biosynthesis was initiated by the simultaneous addition of the (6.7 0.5 i0/.tl; 18.3 1.4% neutrophils) or platelets (805 71 chemotactic peptide fMLP (10—v M), a well-characterized recep103/pd) when compared to untreated animals (vide supra). Prior tor-mediated stimulus for neutrophil activation, and thrombin (1

exposure of rats to this mAb, but not an isotype-matched nonblocking control mAb, was associated with a striking decrement in

renal LXA4 levels during glomerular inflammation (Fig. 3A). Importantly, anti-P-selectin mAb blocked LXA4 production with-

out affecting intrarenal deposition of Con-Alferritin or subsequent formation of immune complexes, as determined by immunofluorescence microscopy (data not shown). Peak platelet infiltration precedes peak neutrophil infiltration in the Con-Alferritin model of glomerulonephritis [10] and neu-

U/mI), a potent physiological stimulus for platelet activation. LXA4 was not detectable in suspensions of platelets alone and there was a striking increase in LXA4 generation in neutrophilplatelet coincubations (0.44 0.09 ng/ml; N = 11) when compared with suspensions of neutrophils alone (0.06 0.01 nglml; N = 3). These findings suggest tlat rat platelets, like human platelets [4, 121, possess "LX synthetase activity" and transform

the neutrophil-derived arachidonate intermediate LTA4 to LXA4 during cell-cell interactions. Again, transcellular biosynthesis of trophil recruitment appeared dependent, at least in part, on the LXA4 was significantly blunted by anti-P-selectin mAb, but not an presence of platelets in the present study as glomerular neutrophil isotype-matched control antibody (Fig. 3C). Thus, parallel in vivo infiltration was reduced in glomerulonephritis following platelet and in vitro studies support a role for P-selectin in transcellular depletion (Fig. 3B). However, glomerular neutrophil infiltration LXA4 generation by platelet-neutrophil interacfions within a local was not reduced by anti-P-selectin mAb (Fig. 3B). These obser- inflammatory milieu.

1300

Papayianni et al: Lipoxin synthesis in glbmenLlonephritis

16000'

12000

detection of LXA4 released into the systemic circulation by

i m* N8

8000

glomerulonephritic kidneys. The striking reduction in renal LXA4 levels following neutrophil depletion and platelet depletion provide compelling evidence for transcellular eicosanoid biosynthesis in vivo and support previous in vitro predictions that transcellular pathways may amplify the profile and levels of mediators generated within the vascular lumen during multicellular inflammatory responses [2, 15, 24—27]. It is likely that the residual LXA4 levels

detected in platelet-depleted and neutrophil-depleted animals reflect LXA4 biosynthesis by other leukocytes and/or by transcel-

4000

N=4

lular routes during interactions of these leukocytes with other glomerular cells that express appropriate lipoxygenase activity [27].

Electron microscopic analysis of inflamed glomeruli demonstrated neutrophil adhesion to endothelial cells, in keeping with the well-defined role of adhesion in leukocyte recruitment [1, 28].

0 Control

ON i-Vehicle ON + LXA4

A more striking and novel pathologic finding was the direct visualization of areas of platelet-leukocyte contact within capillary

Fig. 4. Influence of LXA4 on neutrophil trafficking to glomerulonephritic kidneys in vivo. '11lndium-labeled neutrophils (10 X 10 in 2 ml of PBS) were exposed to LXA4 (10 M, 15 mm, 37°C) or its diluent, washed and injected into rats five minutes prior to injection of anti-ferritin antibody. Exposure of tm11indium-labeled neutrophils to LXA4 did not affect their levels of radioactivity when compared with exposure to diluent. Kidneys were harvested 15 minutes after induction of disease and their radioactivity was measured in a y-counter (LKB Clinigamma) as an index of neutrophil recruitment. Data are means SE. *Denotes a statistically significant difference (P < 0.001) when compared with renal radioactivity in glomerulonephritic rats injected with diluent-treated neutrophils. Exrat neutrophils to LXA4 resulted in marked posure of inhibition of their trafficking to glomerulonephritic kidneys, supporting

lumens. The latter finding was even more intriguing given that renal LXA4 generation was blunted by mAb against P-selectin, a major ligand on platelets that supports platelet-neutrophil adhesion. Together, these observations suggested that P-selectinmediated platelet-neutrophil adhesion facilitates transcellular lipoxin biosynthesis in acute glomerular inflammation. Whereas platelets appeared to contribute to early neutrophil infiltration in acute glomerulonephritis, consistent with other investigations [29] it is unlikely that P-selectin promoted transcellular LXA4 biosynthesis simply by facilitating neutrophil recruitthe contention that this eicosanoid may function as an endogenous ment. In keeping with this interpretation, the number of neutroinhibitor of leukocyte recruitment in vivo. phils within inflamed glomeruli was reduced by prior depletion of LXA4 inhibits neutrophil recruitment to inflamed glomeruli in vivo

The influence of LXA4 on neutrophil accumulation in glomerulonephritic kidneys was assessed to determine if LXA4 inhibits neutrophil recruitment during inflammation in vivo. To this end,

neutrophils were isolated from healthy rats, radiolabeled with

experimental animals of platelets, but not by treatment with anti-P-selectin mAb. These findings suggest that platelets promote neutrophil infiltration by presenting adhesion molecules other than P-selectin to neutrophils and/or by elaborating chemical stimuli of neutrophil recruitment, and that platelet-neutrophil adhesion through P-selectin then augments transcellular LXA4 biosynthesis within a local inflammatory milieu. They also indicate

that endothelial cell P-selectin is not a pivotal molecule in

neutrophil recruitment in this model. Indeed, immunohistochemical staining revealed little up-regulation of endothelial cell Pselectin expression within the temporal framework of our study markedly increased radioactivity when compared with those of (data not shown). There are several potential mechanisms by which P-selectinnormal animals (Fig. 4), consistent with histologic assessments of glomerular neutrophil infiltration (Figs. 1 and 2). Prior exposure mediated adhesion could enhance LXA4 biosynthesis during local of neutrophils to LXA4 was associated with inhibition of neutro- platelet-neutrophil interactions in vivo. By approximating neutro11mindium, treated with LXA4 (10 M) or its diluent, and injected intravenously five minutes prior to administration of anti-ferritin antibody. The kidneys of glomerulonephritic animals displayed

phil recruitment, supporting a modulatory role for LXA4 on

phil and platelet cell membrane lipid bilayers, adhesion may

potentially facilitate the transfer of the arachidonate intermediate LTA4 from neutrophils to platelets. In this regard, it is noteworthy Discussion that LTA4 undergoes rapid non-enzymatic hydrolysis to biologiAcute immune complex-mediated glomerulonephritis is a par- cally inactive products in aqueous media and that its stability is adigm of acute renal inflammation. Our results indicate that significantly prolonged in a lipid environment [30]. Neutrophil LXA4 is generated in the kidney in the Con-A/ferritin model of adhesion through P-selectin may also augment leukocyte 5-lipoxyimmune complex-mediated glomerulonephritis in nanogram genase pathways such that adherent neutrophils release greater quantities, amounts that are likely to have potent modulatory quantities of LTA4 upon activation. Indeed, P-selectin-mediated effects on neutrophil recruitment (vide infra). Significantly more adhesion was recently reported to enhance neutrophil oxidative LXA4 was detected in glomerulonephritic kidneys than in con- bursts in vitro, suggesting that engagement of this ligand may tralateral kidneys, which are relatively spared of disease, impli- prime neutrophil functional responses through "outside-in" sigcating inflamed glomeruli as the major source of LXA4 in our nalling events [31]. It is likely that both mechanisms could model. It is likely that the LXA4 detected in contralateral kidneys contribute to the stimulatory action of P-selectin on transcellular reflects some, albeit lower grade, inflammation on that side or eicosanoid generation in vivo. neutrophil trafficking in vivo.

1301

Papayianni et al: Lipoxin synthesis in glomen4lonephritis

Acknowledgments

The identification of LXA4 as a product of neutrophil-platelet

interactions was particularly provocative in view of the selflimiting nature of renal injury in glomerulonephritis induced by Con-A/ferritin [10] and recent evidence implicating LXA4 as a potential inhibitor of leukocyte trafficking [5—9]. Neutrophils express specific high affinity G-protein-coupled receptors for LXA4 [31] and engagement of these receptors stimulates rapid remodeling of cell membrane lipids without provoking homotypic or heterotypic neutrophil adhesion. Exposure of neutrophils to nanomolar concentrations of LXA4 in vitro attenuates neutrophil chemotaxis, adhesion to endothelial cells, and/or migration across endothelium induced by leukotrienes and other chemoattractants [5—8]. Our results provide evidence that LXA4 may also inhibit neutrophil recruitment during

This work was supported by NIH grants to CNS, HGR and HRB and the Veterans Administration to HRB. CNS is also an Established Inves-

tigator of the American Heart Association. AP is the recipient of a Fellowship from the Juvenile Diabetes Foundation of America. Portions of this work were presented at the 26th Annual Meeting of the American Society of Nephrology and published in abstract form JAm Soc Nephrol 4:627, 1993). We thank Drs. R. S. Cotran and B. M. Brenner for helpful discussions.

Reprint requests to Hugh R. Brady, M.D., Ph.D., FRCPI, Renal Section, Brocktonl West Roxbuiy Veterans Affairs Medical Center, 1400 VFW Parkway, Boston, Massachusetts 02132, USA.

References

inflammation in vivo. Indeed, it has been postulated that lipoxins and leukotrienes exert opposing actions on neutrophil trafficking within the vascular lumen in a manner akin to the counter-regulatory actions of prostacyclin and thromboxane on platelet activation [2, 331. The potential importance of LXA4 as an immunomodulator is further supported by the recent detec-

1. HARLAN JM, Liu DY: Adhesion: Its Role in Inflammato,y Disease. New

tion of this eicosanoid in human inflammatory and allergic diseases and vascular events [reviewed in 4]. Whereas our results implicate neutrophil-platelet transcellular biosynthetic routes as an important initial source of LXA4 during inflam-

CN: Leukotrienes and lipoxins: Structures, biosynthesis and biological

mation, it is likely that cytokine-stimulated induction of 12- and 15-lipoxygenase activity in other cell-types further amplifies the magnitude and efficiency of lipoxin generation in these settings

[34, 35]. The formation of LXA4 within the vascular lumen during platelet-neutrophil interactions would ideally situate this eicosanoid for modulation of leukocyte recruitment. In the specific context of glomerular inflammation, intravascular generation of LXA4 may also benefit renal blood flow and

York, W.H. Freeman and Co, 1992, p 202 A, SERHAN CN: Leukocyte adhesion promotes biosynthesis of lipoxygenase products by transcellular routes.

2. BRADY HR, PAPAYIANNI

Kidney mt 45(Suppl 45):S90—S97, 1994

3. SAMUELSSON B, DAHLEN S-E, LINDGREN JA, ROUZER CA, SERHAN

effects. Science 237:1171—1176, 1987 4. SERHAN CN: Lipoxin biosynthesis and its impact in inflammatory and vascular events. Biochim Biophys Acta 1212:1—25, 1994 5. PAPAYIANNI A, SERHAN CN, BRADY HR: Lipoxins inhibit adhesion of

neutrophils and endothelial cells mediated by P-selectin and /2inregrins through distinct actions on PMN and EC. (abstract) JAm Soc Nephrol 5:792, 1994 6. LEE TH, HORTON CE, KYAN-AUNG U, HASKARD D, CREA AEG, SPUR

BW: Lipoxin A4 and lipoxin B4 inhibit chemotactic responses of human neutrophils stimulated by leukotriene B4 and N-formyl-Lmethionyl-L-leucyl-L-phenylalanine. Clin Sci 77:195—203, 1989 7. BRADY HR, PERSSON U, BALLERMANN BJ, BRENNER BM, SERHAN

CN: Leukotrienes stimulate neutrophil adhesion to mesangial cells: Modulation with lipoxins. Am J Physiol 259:F809—F815, 1990

glomerular filtration, since LXA4 induces endothelium-

8. HEDQVIST P, RAUD J, PALMERTZ U, HAEGGSTROM J, NicoL..oU KC,

dependent vasodilatation and opposes the intrarenal vasoconstrictive actions of peptidoleukotrienes [361, which are probable important pro-inflammatory vasoactive mediators in acute glomerulonephritis.

1989 9. C0LGAN SP, SERHAN CN, PARKOS CA, DELP-ARcHER C, MADARA JL:

In summary, we present evidence that LXA4 is generated within the vascular lumen during glomerular inflammation by transcel-

lular routes during neutrophil-platelet interactions and that adhesion of platelets and neutrophils through platelet P-selectin promotes transcellular LXA4 biosynthesis in this setting. Expo-

sure of neutrophils to LXA4 results in inhibited neutrophil trafficking to glomerulonephritic kidneys, suggesting that cell-cell

interactions may promote the generation of endogenous lipidderived inhibitors that act on leukocyte trafficking during the evolution of multicellular inflammatory responses to limit tissue destruction. In this regard, our results add to an expanding body of evidence implicating products of the 12- and 15-lipoxygenase pathways (lipoxins, 15-hydroxyeicosatetraenoic acid) as "braking

DAHLEN S-E: Lipoxin A4 inhibits leukotriene B4-induced inflammation in the hamster cheek pouch. Acta Physiol Scand 137:571—572,

Lipoxin A4 modulates transmigration of human neutrophils across intestinal epithelial monolayers. J Clin Invest 92:75—82, 1993 10. FRIES JWU, MENDRICK DL, RENNKE HG: Determinants of immune complex-mediated glomerulonephritis. Kidney mt 34:333—345, 1988 11. LEVY BD, BERTRAM S, TA! HH, ISRAEL E, FISHER A, DRAZEN JM, SERHAN CN: Agonist-induced lipoxin A4 generation: Detection by a novel lipoxin A4-ELISA. Lipids 28:1047-1053, 1993 12. FIORE S, SERIIAN CN: Formation of lipoxins and leukotrienes during

receptor-mediated interactions of human platelets and recombinant human granulocyte/macrophage colony-stimulating factor-primed neutrophils. J Exp Med 172:451-1457, 1990 13. BRADY HR, SERHAN CN: Adhesion promotes transcellular leukotriene biosynthesis during neutrophil-glomerular endothelial cell in-

teractions: Inhibition by antibodies against CD18 and L-selectin. Biochem Biophys Res Commun 186:307—1314, 1992 14. MULLIGAN MS, POLLEY MJ, BAYER RJ, NUNN MF, PAULSON JC,

WARD PA: Neutrophil-dependcnt acute lung injury. Requirement for

(GMP-140). J Clin Invest 90:1600—1607, 1992 signals" in the evolution of host defense and inflammatory 15. P-selectin COUSER WG: Mediation of immune glomerular injury. J Am Soc

responses [37—42]. Definitive demonstration of a role for LXA4 as

a modulator of neutrophil recruitment will require the development of pharmacologic or molecular strategies for specific inhibition of LXA4 biosynthesis or bioactivity. The elucidation of the signals generated by transcellular biosynthesis that down-regulate multicellular inflammatory responses in vivo may suggest novel strategies for therapeutic intervention in leukocyte-mediated human diseases.

Nephrol 1:13—29, 1990 16. EDENIUS C, HAEGGSTROM J, LINDGREN JA: Transcellular convertion

of endogenous arachidonic acid to lipoxins in mixted human plateletgranulocyte suspensions. Biochem Biophys Res Commun 157:801—807, 1988 17. SERHAN CN, SHEPPARD K-A: Lipoxin formation during human neu-

trophil-platelet interactions. Evidence for the transformation of leu-

kotriene A4 by platelet 12-lipoxygenase in vitro. J Clin Invest 85:772— 780, 1990

1302

Papayianni et al: Lipoxin synthesis in glomerulonephritis

18. Hsu-LIN S-C, BERMAN CL, FURIE BC, AUGUST D, FUmE B: A platelet

membrane protein expressed during platelet activation and secretion. Studies using a monoclonal antibody specific for thrombin-activated platelets. J Biol Chem 259:9121—9126, 1984 19. LARSEN E, CELl A, GILBERT GE, FURIE BC, ERBAN JK, BONFANTI R,

WAGNER DD, Fumu B: PADGEM protein: A receptor that mediates the interaction of activated platelets with neutrophils and monocytes. Cell 59:305—312, 1989 20. PALABRICA T, LOBB R, FURIE BC, ARONOVITZ M, BENJAMIN C, Hsu

Y-M, SAJEN SA, FURIE B: Leukocyte accumulation promoting fibrin

deposition is mediated in vivo by P-selectin on adherent platelets. Nature 359:848—851, 1992 21. WEYRICH AS, MA X-I, LEFER DJ, ALBERTINE KH, LEFER AM: In vivo

neutralization of P-selectin protects feline heart and endothelium in myocardial ischemia and reperfusion injury. J Gun Invest 91:2620— 2629, 1993 22. MULLIGAN MS, PAULSON JC, DEFREES S, ZHENG Z-L, LOWE JB, WARD PA: Protective effects of oligosaccharides in P-selectin-dependent lung injury. Nature 364:149—151, 1993 23. MAYADAS TN, JOHNSON RC, RAYBURN H, HYNES RU, WAGNER DD:

Leukocyte rolling and extravasation are severely compromised in P-selectin-deficient mice. Cell 74:541—554, 1993 24. MARCUS AJ: Transcellular metabolism of eicosanoids. Prog Hemost Thromb 8:127—142, 1986 25. BREZINSKI DA, NESTO RW, SERHAN CN: Angioplasty triggers intra-

coronary leukotrienes and lipoxin A4. Impact of aspirin therapy. Circulation 86:56—63, 1992

26. SERHAN CN: Cell-cell interactions in the generation of eicosanoids: Charting the routes and products of transcellular biosynthesis. J Lab Clin Med 121:372—374, 1993 27. IMAI E, HOOVER RL, MAKITA N, FUNK CD, BADR KF: Localization

and relative abundance of 5-lipoxygenase (LO), 15-LO, 12-LO, and leukotriene A4 hydrolase (LTA4-OHASE) gene expression in cultured glomerular cells. (abstract) JAm Soc Nephrol 1:443, 1990 28. BRADY HR: Leukocyte adhesion molecules and kidney diseases. Kidney mt 45:1285—1300, 1994

31. NAGATA K, TSUJI T, T000KORI N, K.ATAGIRI Y, TANOUE K, YAMAZAKI H, HANAI N, IRIMUZA T: Activated platelets induce

superoxide anion release by m000cytes and neutrophils through

P-selectin (CD62). J Immunol 151:3067—3073, 1993 32. FI0RE S, MADDOX JF, PEREZ HD, SERHAN CN: Identification of a human cDNA encoding a functional high affinity lipoxin A4 receptor. J Exp Med 180:253—260, 1994 33. BADR KF: 15-lipoxygenase products as leukotriene antagonists: Ther-

apeutic potential in glomerulonephritis. Kidney Int 42 (Suppl 38): S101—S108, 1992 34. KATOB T, LAJous FG, MAKITA N, BADR KF: Co-regulated expression

of glomerular 12/15-lipoxygenase and interleukin-4 mRNAs in rat nephrotoxic serum nephritis. Kidney mt 46:341—349, 1994 35. CRUET EN, NASSAR GM, BADR KF, LAKKIS FG: A potential role for

interleukin-13 in experimental glomerulonephritis. (abstract) J Am Soc Nephrol 5:744, 1994 36. BADR KF, DEBOER DK, SCHWARTZBERG M, SERHAN CN: Lipoxin A4

antagonizes cellular and in vivo actions of leukotriene D4 in rat glomerular mesangial cells: Evidence for competition at a common receptor. Proc Nati Acad Sci USA 86:3438—3442, 1989 37. SPECTOR AA, GORDON JA, MOORE SA: Hydroxyeicosatetraenoic acids (HETE). Prog Lipid Res 27:271—323, 1988 38. BREZINSKI ME, SERHAN CN: Selective incorporation of (15S)-eicosatetraenoic acid in phosphatidylinositol of human neutrophils: Agonistinduced deacylation and transformation of stored hydroxyeicosanoids. Proc NatlAcad Sci USA 87:6248—6252, 1990 39. FISHER DB, CHRISTMAS JW, BADR KF: Fifteen-S-hydroxyeicosatetra-

enoic acid (15-S-HETE) specifically antagonizes the chemotactic action and glomerular synthesis of leukotriene B4 in the rat. Kidney mt 41:1155—1160, 1992 40. TAKATA 5, MATSUBARA M, ALLEN PG, JANMEY PA, SERHAN CN,

BRADY HR: Remodeling of neutrophil phospholipids with 15(S)hydroxyeicosatetraenoic acid inhibits leukotriene B4-induced neutrophil migration across endothelium. J Clin Invest 93:499—508, 1994 41. TAKATA S, PAPAYANN1 A, MATSUBARA M, JIMENEZ W, PRONOVOST

C, COUSER WG: Platelets mediate neutrophil-dependent immune

PH, BIwY HR: 15-Hydroxyeicosatetraenoic acid inhibits neutrophil migration across cytokine-activated cndothelium. Am J Patliol 145:

complex nephritis in the rat. J Clin Invest 82:1225—1235, 1988 30. FIORE S, SERHAN CN: Phospholipid bilayers enhance the stability of

541—549, 1994 42. BRADY HR, PAPAYIANNI A, SERMAN CN: Potential vascular roles for

leukotriene A4 and epoxytetraenes: Stabilization of eicosanoids by

lipoxins in the "Stop programs" of host defense and inflammation.

liposomes. Biochem Biophys Res Commun 159:477—481, 1989

Trends Cardiovasc Med (in press)

29. JOHNSON Ri, ALPERS CE, PRrrzI P, SCHULZE M, BAKER P, PRUCI-INO