Leukotriene C4 and D4 contract rat glomerular mesangial cells

Kidney International, Vol. 30 (1986), pp. 524—531

Leukotriene C4 and D4 contract rat glomerular mesangial cells MICHAEL S. SIMONSON and MICHAEL J. DUNN Deparment of Medicine, Case Western Reserve University, Division of Nephrology, University Hospitals of Cleveland, Cleveland, Ohio, USA

rophages, and mononticlear cells Ill. LTB4 causes adhesion and chemotaxis of leukocytes and stimulates aggregation, enzyme release, and the generation of reactive oxygen intermediates in neutrophils. The sulfidopeptide leukotrienes, originally known as slow reacting substance of anaphylaxis, increase permeabil-

Leukotriene C4 and D4 contract rat glomerular mesangial cells. The sulfidopeptide leukotrienes, LTC4 and LTD4, have vasoconstrictor

effects in the kidney, reducing both renal blood flow and the glornerular

filtration rate. As one mechanism regulating the glonierular filtration rate, mesangial cell contraction may reduce the capillary surface area, thereby lowering the ultrafiltration coefficient. Using image analysis microscopy to quantify changes in cell morphology, we found that LTC4 and LTD4 (lx 10 2M to lx 10 6M) reduced the cross—sectional area of cultured inesangial cells from rat glomeruli. The response to LTC4 and LTD4 (l06M), as measured by the percentage of responding

ity in posteapillary venules and contract a variety of smooth muscle preparations in vitro and in vivo [3, 4]. Leukotrienes also stimulate prostaglandin and thromboxane synthesis in many cells [5—7], although the exact contribution of these

cells (30 to 35%), the maximum decrease in cross—sectional area (25 to

32%), and the time course was identical to that for angiotensin 11 (l0M). The contraction induced by LTD4 was attenuated by an LTD4 receptor antagonist (4R ,SS ,6Z-nor-LTD ). Also, preineubation with colehicine prevented LTC4-induced contraction. Leukotriene B4, a non-sulfidopeptide leukotriene that stimulates ehemotaxis and chemokinesis, had negligible agonist activity. Mesangial cells cultured on less adhesive teflon membranes were more responsive to LTD4 (62% of cells responded) than cells euttured on glass or polystyrene (35% of cells responded). Mesangial cell contraction was not merely a shape—change as a result of cell damage, since cellular injury was not documented by lactate dehydrogenase release and proliferation of mesangial cells was not retarded by LTC4. Furthermore, the contraction was independent of cell size. Because leukotrienes stimulate cyclooxygenase products in other cells, we examined the ability of the sulfidopeptide leukotrienes to stimulate prostaglandin and thromboxane synthesis. LTC4 and LTD4 did not stimulate PGE formation, the major eyelooxygenase product of rat mesangial cells. Although LTC4, but not LTD4, stimulated a small amount of thromboxane synthesis, a thromboxane synthetase inhibitor (UK-38485) and a receptor antagonist (EP-092) did not alter leukotriene—mediated contraction. The contraction of mesangial cells

exposed to LTC4 and LTD4 may contribute to the reduction in the ultrafiltration coefficient seen in some types of glomerular injury.

cyclooxygenase products to the actions of LTC4 and LTD4 is uncertain. Evidence from several sources suggests a role for lettkotrienes in modulating renal function. LTC4 administration, to

the rat in vivo, reduced renal blood flow (RBF) and the glomerular filtration rate (GFR) by almost 50% [81. In the isolated perfused rat kidney, LTC4 and LTD4 caused vasoconstriction that was blocked by a leukotriene receptor antagonist [91. Similarly, LTC4 and L'l'D4 were vasoconstrictor in the pig kidney perfused in situ [101. There are also specific receptors

for LTC4 in rat glomeruli [II]. After glomerular immune injttry, infiltration by both neutrophils and monocytes could produce high concentrations of leukotrienes in the glomerulus. As one mechanism regulating the GFR, mesangial cell contraction is thought to reduce the capillary surface area, thus decreasing the ultrafiltration coefficient [12—14]. In this study, therefore, we used image analysis

microscopy to examine the effects of leukotrieries on the contraction of cultured rat mesangial cells. We also examined whether leukotricnes stimulated cyclooxygenase products and

whether prostaglandins or thromboxane were important in Leukotrienes are eicosanoids formed from cellular phosphoglycerides by the oxidative metabolism of arachidonic acid via the lipoxygenase pathway [11. Both leukotriene B4 (LTB4)t and

leukotriene—mediated contraction. Methods

the sulfidopeptide leukotrienes, LlC4 and [,TD4, act as

Materials

autacoids mediating immediate hypersensitivity reactions and inflammation [1, 21. They are synthesized by many organs and cells including the lung, polymorphonuclear leukocytes, mac-

Fetal bovine serum, RPMI 1640 media, and trypsin were obtained from K,C. Biological. Sodium pyruvate, disodium

Abbreviations used in this paper are: LT, leukotriene(s): LTB4, ITC4, LTD4, leukotriene B4, C4, D4; ANG ti, angiotensin II; CSA,

Biochemicals (Indianapolis, Indiana, USA). Fatty acid free bovine serum albumin, synthetic angiotensin II, and calcium

cross—sectional area; KHH, Krebs—Henseleit H EPES; GFR, glomerular filtration rate; K1, ultrafiltration coefficient; POE2, prostaglandin E,; TxA7, thromboxane A2; 1.DH, lactate dehydrogenase.

Received for publication July 23, 1985 and in revised form January 17, 1986

© 1986 by the International Society of Nephrology

NADH, and lactate dehydrogenase (#127221) in glycerol from porcine muscle were purchased from Boehringer Mannheim

ionophore, A23 187, were from Calhiochem—Behring (La Jolla, California, USA). Colchicine, L-serine, and sodium tetrahorate were obtained from the Sigma Chemical Company (St. Louis, Missouri, USA). Collagenase 'fype I was from Cooper Biomed-

ical (Palo Alto, California, USA) and ITS premix was from Collaborative Research. 524

Leukotrienes contract mesangial cells

When we verified the purity of the leukotrienes by HPLC [15], single, distinct peaks were evident. Stock solutions of LTC4 and LTD4 were prepared in water, using glass vessels only, and the exact concentrations were determined spectro-

525

entific, Elkhart, Indiana, USA) at a density of 3 to 6 x IO cells/cm2; however, for biochemical experiments, cells were subcultured into 6—well multiwell plates only at a density of 1 to

3 x iO cells/cm2. Fourteen different cell lines were used in this photometrically using a molar extinction coefficient of 40,000 study. In several experiments mesangial cells were cultured on M'cm' at 282 nm. All stock solutions were stored at —70°C teflon membranes in a Bionique tissue culture chamber [21] under argon. Fresh solutions were prepared for each experi- (Corning #26 100). ment by diluting an aliquot of the stock with Krebs—Henseleit Morphometric analysis HEPES (KHH). Stock solutions of 4R,5S,6Z-2-nor-LTD1 were prepared in 20 m sodium carbonate. Morphometric measurements were made using a Unicomp EP-092 [16], a more potent analog of the thromboxane Image Analyzer (Southern Micro Instruments, Atlanta, Georreceptor antagonist EP-045 [17], was used in this study. For gia, USA) with a Nikon Diaphot phase—contrast inverted mieach experiment a solution of the sodium salt was prepared by croscope. The Unicomp Image Analyzer is a semi-automatic dissolving the free acid in ethanol, adding an equivalent amount image analyzer as described by Bradbury [22]. We could not of NaOH, mixing, and evaporating the ethanol with N2. This use a bicarbonate—buffered solution because the cells were left a residue of the sodium salt that was soluble in KHH. All exposed to air throughout the experiments. Therefore, we other chemicals were reagent grade from standard commercial modified Krebs—Henseleit bicarbonate—buffered saline by replacing the sodium bicarbonate with 20 mrvi Hepes (Na suppliers. Hepes/Hepes, pH 7.4), reducing the Ca2 concentration to 1.25 Cell culture m, and by adding 5.5 m glucose and 0.2% (wt/vol) fatty acid Primary rat mesangial cell lines were started from isolated free albumin (buffer designated as Krebs—Henseleit Hepes: glomeruli that were partially digested with collagenase to form KHH). For each experiment the incubation well was clamped "glomerular cores" [18, 19]. All primary cultures were pre- to the microscope stage, and using a 20x objective, a field pared using standard aseptic techniques. Kidneys were re- chosen at random that contained four to seven isolated cells. moved from four ether—anesthetized rats (150 to 200 g male The only criterion for choosing a field was easily discernible cell Sprague—Dawley; Zivic—Miller, Allison Park, Pennsylvania, boundaries; cells were not selected on the basis of size, shape, USA), decapsulated, the medulla dissected out, and the cortex or density. minced. The cortical paste was passed through a 106 m metal Next, the monolayer was gently washed (two times) with 2 ml sieve, and the material was collected from the underside into 20 of KHH, and the experiment started. Cells were incubated for ml of Hank's balanced salt solution containing penicillin (100 30 minute basal and experimental periods during which sequenunits/mI) and streptomycin (100 g/ml). This suspension was tial photographs (Polaroid CT 35 mm transparency film, ASA then passed through a 21 g needle and passed through a 74 tm 125) were taken to measure cross—sectional area (CSA). At —20 nylon sieve. Glomeruli were collected from the top surface of minutes, the supernate was gently aspirated and replaced with the 74 tm sieve and suspended in approximately 6 ml of the 2 ml of KHH, or in experiments where inhibitors were used Hank's solution. After centrifugation at 50 g for two minutes, to with 2 ml KHH containing the inhibitors. At zero time, the pellet the glomeruli, the method of Lovett et al [19] was used to supernate was again aspirated and replaced with 2 ml of KHH prepare the "glomerular cores" by collagenase digestion of the containing the appropriate agonist. We determined that gentle aspiration was necessary to avoid disturbing the cells. Pictures glomerular epithelial cells. The mesangial cells were grown in RPMI 1640 containing of an objective micrometer (Nikon, 0.01 mm) were included to

15% fetal bovine serum, 100 units/ml penicillin, 100 g/ml calibrate the image analyzer. The cells had to be critically streptomycin, 5 g/ml insulin, 5 g/ml transferrin, and 5 ng/mI focused using the magnifier on the observation turret, and the selenium at 37°C in 5% CO2 and 95% air. Primary cultures were

incubated for eight to ten days and then subcultured. Media in all cultures were changed every two to three days. Mesangial cells were characterized by their typical morphology under both phase—contrast and electron microscopy [20], and by their contractile response to the vasoconstrictive peptide, ANG II. Unlike fibroblasts, mesangial cells proliferated in medium containing D-valine, The mesangial cytotoxin, mitomycin C (10 /Lg/ml), caused our cells to round up and detach, whereas the glomerular epithelial cytotoxin, puromycin aminonucleoside (100 g/ml), induced no morphologic changes. Mesangial cells did not stain for the endothelial surface marker, factor VIII, and contamination by monocytes and macrophages was excluded using a non-specific esterase stain. Furthermore, these cells did

field was refocused in the same manner before each picture was taken.

Two filters were placed in the microscope's light path: a green filter (Nikon #79094) to enhance contrast on the black and white film and a heat—absorbing filter (Nikon #76940) to prevent light from increasing the monolayer temperature. When the experiment was completed, the slides were developed, projected (500 to 700x), and the cell outlines traced on paper. Cross—sectional area was then measured for each time point using a digitizing—tablet with the image analyzer. All images were traced in the stream mode (that is, continuous vs. point— to-point). Because the initial size of each cell varied, the results were analyzed in a paired manner. The CSA for each time point of not express Ia antigen. Both morphologic and biochemical the experimental period was expressed as a percentage of the experiments were performed on mesangial cells from passages mean for the four basal measurements. Taking into account 1 to 4. For morphologic experiments, cells were subcultured small fluctuations (that is, <5%) in basal measurements and the into either polystyrene six—well multiwell plates (Costar) or reproducibility of the image analysis, we decided that any two—chamber glass tissue culture slides (Lab—Tek, Miles Sci-

reduction of CSA greater than 7.0% would be considered

526

Simonson and Dunn

significant, that is, cell contraction. We also made other morphometric measurements including perimeter, form factor (47r area/perimeter2), maximum diameter, angle of the maximum diameter, and Feret lengths in both the x and y axes [23, 24]. Unlike CSA, no other measurement gave consistent results, although, in most cases (>90%) the perimeter was also reduced

after addition of a contractile stimulus. When ten replicate measurements were made of a circle with a mean CSA of 7255 p.m2, the coefficient of variation was 1.1%. The coefficient of variation increased slightly to 2.0% when irregular ccli outlines were traced.

Prostaglandin responses to leukotrienes Mesangial cell monolayers were washed twice with 2 ml of

100 C',

0 90 'C' C')

'U CU

0

80

U a)

j 70 011

20 30 0 10 KHH to remove the media. Incubations consisted of a 15 minute basal period followed by a 15 minute experimental Time, mm period, both at 37°C. At the end of the basal period, half the Fig. 1. Time course of contraction by mesangial cells that responded KHH (I ml) was removed to measure basal prostaglandin to ANG II. Data are expressed as mean SEM for six separate

levels, and another 1 ml containing the agonist was added. At the end of the experimental period, 1 ml was again removed to measure proslaglandins. In this way prostaglandin synthesis in each well could be corrected for basal synthesis. For measuring cellular protein, each dish was washed twice with albumin—free KHH, incubated with 0.1% (wt/vol) sodium dodecyl sulfate for one hour at room temperature, and proteins were determined by a modification [25] of the Lowry method using bovine serum albumin as the standard. Prostaglandins and thromboxane B2 were measured by radioimmunoassay as previously described

experiments. Control cells were incubated with KHH alone. All mea-

surements with ANG II were statistically different (P < 0.01) from control cells.

decided to quantify the morphologic changes in two ways: (a) by calculating the percentage of cells that showed a significant response; and (b) by measuring the maximum change in CSA for each of those responsive cells. In control experiments in which KHI-I was added during the in this laboratory [26]. All compounds used were checked experimental period there was no change in mesangial cell independently to determine whether they had any effect on the CSA. ANG 11(1 x 106M) induced a time—dependent decrease in CSA in 41% of the cells. The maximum decrease in CSA was radioimmunoassays. 28.0 SEM, N = 6 separate experiments). 4.4% (mean Similarly, the perimeter was reduced by 9.4 4.3%, There was Lactate dehydrogenase assay Mesangial cells cultured in six—well dishes were washed no significant change in either morphologic descriptor in any of twice with Kl-IH without albumin, and then either KHH or the non-responding cells, In addition, we did three experiments KHH containing various concentrations of LTC4 were added to with 109M ANG II, and both the percentage of cells responding the cells. After 15 minutes, the supernatant media were col- (27%) and the maximum change in CSA (15.4 2.7) were less lected and immediately frozen. To measure the total cellular than with 106M ANG II. Figure 1 illustrates the time course of content of lactate dehydrogenase (LDH), a separate well in ANG Il-mediated contraction. A decrease was evident at 5 each experiment was incubated for 15 minutes, after which the minutes, with the maximum contraction occurring between 10 cells were scraped off the dish, homogenized with a Potter to 20 minutes. In experimental periods of up to 50 minutes, no Elvehjcm tissue grinder, frozen and thawed three times in an additional contraction was seen after 30 minutes. acetone/dry—ice bath, and stored frozen. LDH activity in the Effects of leukotrienes on mesangial cell morphology supcrnates and homogenates was measured spectrophotoFigure 2 illustrates typical changes in mesangial cell morpholmetrically under saturating conditions at 25°C by the method of Bergmeyer and Berut [27], using 6.22 mrsf 'cmi as the mil- ogy induced by LTC4 or LTD4. Also shown in Figure 2 is a limolar absorption coefficient for NADH at 340 nm. Cellular non-responsive cell that remained unchanged throughout the protein was measured as described above and LDH activity experiment. Table I summarizes the morphologic effects of expressed as mole pyruvate consumed/min/g cell protein. incubating mesangial cells with different types and concentraLTC4 had no effect on the LDH assay in control incubations tions of leukotrienes. The sulfldopcptide leukotrienes, LTC4 with both total mcsangial cell homogenates and authentic LDH and LTD4, caused a reduction in CSA. Concentrations of LTC4 and LTD4 of I x l02M induced significant contraction in only from porcine muscle. a few cells. When the concentration of LTC4 or LTD4 was increased from l02M to l0 9M, the percentage of responsive Results cells increased; however, the maximum decrease in CSA inMorphologic changes in mesangial cells incuhaled with creased between 10 9M and l06M LTC4 or LTD4. The perimANG II eter of the mesangial cells was also reduced, but without a clear We validated and characterized our measurements of mesan- trend over the range of LTC4 and LTD4 concentrations. In gial contraction using the image analysis system with ANG II as addition, while lO6M LTI)4 did not cause more cells to respond the contractile stimulus. After preliminary experiments we than 106M LTC4, the maximum decrease in CSA was greater

527

Leukotrienes contract mesangial cells

Fig. 2. Photomicrograph of cultured mesangial cells exposed to LTc4 (106M). A. Four cells at —20 mm of the basal period; B, the same four cells

after a 20 mm incubation with LTC4. Note that the cell at the upper right did not contract (that is, only 4.6%). (x 770).

Table 1. Changes in mesangial cell morphology in response to leukotrienesa Maximum

Concentration

% Cells responding

decrease in CSAb

Maximum decrease in perimeter

0

—

—

traction stimulated by I x l06M ANG II (49% of cells

Control —

(3)

responding, N =

LTC4 (4) (5)

16 31

18.2 16.2

4.5 2.2

(6)

33

24.8

3.6c

I02M

(4)

11

3.1

109M

(2) (2)

35

23.0 20.3

12

18.0

(2)

35

31.8

(2) (2)

0

10'2M

l09M 106M LTD4

109M + 105M nor-LTD1 106M LTB4 109M 106M

4R,5S ,6Z-2-nor-LTD1 attenuates LTD4-induced bronchoconstriction in guinea pigs. Table 1 shows that nor-LTD1 antagonized the response of mesangial cells to LTD4 (109M) as the percentage of cells responding fell from 35% for LTD4 alone to 12% for LTD4 with nor-LTD1. Nor-LTD1 did not inhibit con-

6

14.8 11.9 13.7

2.5 2.3 1.7

4.1 7.2

12.0 18.7 9.3

6.9

2.1

14.1

2.6

1.8

—

— 19.0

3.0

13.9

Data are mean SEM. Number of experiments in parenthesis; 4 to

7 cells per experiment. All changes of CSA and perimeter were statistically different from control (P < 0.01). X basal — maximum I, experimental < 100 X basal

Paired t-test on raw data, P < 0.05, 105M compared to i0M.

3

separate experiments), suggesting that

nor-LTD1 did not independently inhibit cell contraction.

To determine whether mesangial cells respond to LTC4, LTD4, or both, in three experiments we preincubated cells for 10 minutes with 1 mivi serine borate complex, a transition—state inhibitor of y-glutamyltranspeptidase that blocks conversion of

LTC4 to LTD4. Serine borate complex (1 mM) completely blocked LTC4 to LTD4 conversion in isolated rat glomeruli [11]. The percentage of cells responding to 1 X l06M LTC4 fell from 33% to 23% in the presence of 1 mrvi serine borate. Serine borate did not diminish contraction induced by 1 )< 106M LTD4 (42% cells responding, N = 3). The attenuated contraction to LTC4 in

the presence of serine borate complex suggests that although LTC4 and LTD4 can stimulate contraction independently, some of the response seen with LTC4 may depend on partial conversion to LTD4. The time course of contraction in response to LTC4 is shown in Figure 3. The time course is similar to that for ANG II and,

in almost all cells that responded, a decrease of CSA was with LTD4. When mesangial cells were incubated with l09M LTB4, there was no change in cell morphology. Only at 106M LTB4 was there a barely detectable response with only 6% of the cells responding. A structural analog of LTD4, 4R,5S,6Z-2-nor-LTD1 [28], acts as an LTD4 receptor antagonist in vitro and in vivo [29]. This analog antagonizes LTD4-induced contraction in guinea pig

evident at five minutes. The time course for LTD4-mediated contraction was identical to LTC4, except that the maximum decrease was greater for LTD4 than LTC4 at 106M. Unlike ANG II, some of the contracted cells exposed to LTC4 began to return to their basal CSA between 20 to 30 minutes; neverthe-

less, no cell relaxed completely to the basal CSA. These data are consistent with the sulfidopeptide leukotrienes causing morphologic changes in mesangial cells, thought to be the result tracheal spirals and lung parenchymal strips. Furthermore, of contraction, that are analogous to those caused by ANG II,

528

Si,nonson (1/UI Dunn

Table 2. The effect of leukotriene C4 on mesangial cell proliferationa

Cell density (cells xlO1/cm2) after subculture

Addition to media

100

Control 90

24 hr

72 hr

0.54 2.96 0.16 4.61 + 1.68 0.67 4.50 + 0.76 7.44 1.35 0.1! 3.73 0.33 7.46 1.58

2.03

I.TC4, 10 9M 2.25

LTC4, 10 M 1.92

8-

48 hr

96 hr 9.00 7.73

0.38

7.58

0.82

1.36

0

Data are mean SEM of triplicate determinations l'rom each of three separate experiments. Cell density was measured in situ using image analysis microscopy; in this way, cell growth in the same well could be followed without disturbing the cells. LTC4 was added daily atler the first measurement of cell density at 24 hrs.

80 Ii)

1)

70 01

0

10

20

30

Time, rn/n Fig. 3. Time course for inesungial ce/I contraction in repon S e to LTC4. Data are expressed as mean SEM for live experiments at tO 9M and six experiments at 10 6M. Four to seven cells were assessed in each experiment. LTD4 produced a time course indistinguishable from UI C4

at corresponding concentrations.

Contraction of mesangial celLs cultured on teflon membranes The

artificial surface upon which cells are cultured can

influence cell adhesiveness. Cells cultured on teflon substrates are less adhesive than cells cultured on glass. Mcsangial cells grew well on teflon membranes and had the same morphology as cells cultured on either plastic or glass. The variation in basal measurements of cells cultured on teflon was greater than with cells cultured on plastic or glass. We attribute this variation to the reduced anchorage of mesangial cells grown on teflon and not to the slightly reduced optical quality of teflon membranes versus glass. When inesangial cells, cultured on teflon, were incubated with LTD4 (10 6M)the percentage of cells responding increased to 62% (as compared to 35% on glass); however, the maximum change in CSA and the time course of shape changes were unchanged.

Table 3. Effects of LTC4 on prostaglandin and thromboxane synthesis in rat mesangial cells'

PGE

TxB

6-keto-PGF1,,

ng/mg protein/IS mm

Control I,TC4, 10 2M LTC4, 10 9M LTC4, 10 6M A23 187, tO 5M

0.26 * 0.11 0.25 0.20 0.31

26.0 *

(1.19

0.10 0.16 4.5

ND ND 0.7 0.23 1.3

12.0

0.40 4.7

ND ND ND ND ND

Data arc mean SUM, N — 5 (duplicates in each experiment: ND, none detectable).

both control and LTC4-exposed cells was less than 1.6% of the

total cellular activity. Moreover LDH activity was not increased in cells incubated with 1 x 10 'M ANG 11(6.2 2.1). These results suggest that the sulfidopeptide leukoirienes do not compromise the viability or integrity of cultured rat mesangial cells. Prolifrra tion of cells exposed to leukotrienes To confirm the lack of leukotriene—mediated toxicity we measured proliferation of cells incubated with LTC4. Mesangial cells grown in six—well multiwell plates were fed daily for four

days with media alone or media containing 10 9M or 10 6M LTC4. Cell density in each well was measured using the image Inhibition of contraction by c'olchu'ine analyzer (Table 2). Cells incubated with LTC4 proliferated at a Colchicine is a tuhulin binding drug that inhibits the addition faster rate on days two to three than cells receiving media of tubulin to microtubules, leading to microtubule depolymeri- alone, and by day four when the cells approached confluence, zation and inhibition of muscle contraction. In control incuba- the density of L'I'C4-exposed and control cells was equivalent. tions with 1 x 10 7M colchicine, only one cell out of 22 (4.5%) had a small (14%) but significant decrease in CSA. Preincuhating mesangial cells for ten minutes with colchicine prevented contraction in response to 1 x 10 6M L'I'C4 (3.0% cells respond-

ing, N 3 experiments, vs. 33.0% for LTC4 alone). Measurements of extracellular LDJ—I in cells exposed to L'I'C4

Lack of correlation between cell size and Contraction

We were concerned that the initial size of a cell might influence the contractile response. Analysis of the changes of CSA as a function of cotitrol size dismissed this concern. The mean basal CSA for contracting cells was 2296 153.8 ,i.Lm2 (mean SEM, N — 65) as compared to 2577 98 ni2 (N 226)

To determine it the reduction in CSA in the presence of sulfidopeptide leukotrienes was a result of cell injury, we measured LDH activity in the supernatant media of cells incubated with L'I'C4. Mesangial cells incubated with LTC4

for all cells. Furthermore, the amount of contraction in re sponding cells was independent of cell size. There was no

above control cells (data are mean SEM: control — 4.2 1 .6 jmoles pyruvate consurned/min/g protein vs. LTC4 10 6M, 2.1 2.0; LTC4 10 9M, 6.2 2.1; LTC4 10 2M 4.2 1.7; total cellular LDH, 312.0 62.0). LDH activity in the supernates of

Effect of LTC4 on (yc'looxygenase products and the role o,f TxA, in contraction Because leukotrienes stimulate the synthesis of cyclooxygenase products in other tissues, we examined the effect of LTC4

(10 2 to 10 6M) had no increase in extracellular LDH activity

correlation (r —--0.124) between the basal CSA of responding cells and the maximum decrease in CSA.

Leukotrienes contract mesangial cells

100

Cs

% cells responding

90

LTC4, 106M

-n

33%

0

529

the glomerular explants. Finally, the morphologic changes we saw in mesangial cells were similar to those seen in cultured vascular smooth muscle cells exposed to ANG II [34, 351, and norepinephrine [35]. The most significant finding of our study was that LTC4 and

LTD4 contracted mesangial cells. The threshold for the leukotriene response (1012M) is similar to the contractile threshold in guinea pig lung parenchymal tissues [36] and ileum

80

[41. LTC4 and LTD4 had identical effects in terms of the

LTC4, 10- 6M

(I)

0

percentage of cells responding, the maximum decrease in CSA,

+

and the time course of contraction. Contraction induced by

EP092, 105M

70

LTC4 and LTD4 appeared to be specific and agonist dependent as a LTD4 receptor antagonist attenuated the contraction stim-

01' 0

10

20 Time,

30

mm

Fig. 4. Time course for contraction of mesangial cells exposed to LTC4 and cells preincubated with EP-092 and then incubated with LTC4 and EP-092. The data are presented as mean SEM for six experiments with LTC4 alone and three experiments with LTC4 and EP-092. There was no significant difference between the two conditions at any time point. The percentage of cells responding was also unchanged by EP-092. In control incubations, EP-092 had no independent effect on cell CSA.

and LTD4 on mesangial cell prostaglandin and thromboxane synthesis. LTC4 and LTD4 in concentrations of 1 X 10— 12 to 1 x 106M did not stimulate PGE2 synthesis (Table 3). LTC4, but not LTD4, slightly stimulated TxB2 synthesis. The calcium ionophore, A23187, had the expected stimulatory action on the mesangial cell production of PGE2 and TxB2 (Table 3). To determine what role the low levels of vasoconstrictive

ulated by LTD4, and LTB4 did not alter CSA. Low receptor affinity of the LTD4 receptor antagonist, nor-LTD1, probably explains the incomplete blockade of LTD4-induced contraction. Even with high concentrations of nor-LTD1, enough LTD4 may

occupy its receptor to stimulate some contraction, and the

contraction once stimulated proceeds to completion. Likewise, in other systems in vivo and in vitro, nor-LTD1 inhibited, but not completely blocked, LTD4-induced contraction [28, 29]. Because LTC4 and LTD4 directly increase vascular permeability in post capillary venules [1], we were concerned that these

leukotrienes might cause a non-specific, cytotoxic response with alterations of CSA secondary to cell injury and partial detachment. The experiments measuring LDH release and proliferation in cells exposed to LTC4 suggest strongly that the

sulfidopeptide leukotrienes do not damage mesangial cells. Indeed, if the morphologic changes were caused by some non-specific, cytotoxic effect, then one would expect to see a larger percentage, if not all the cells, respond. Cultured mesangial cells and smooth muscle cells respond

similarly to leukotrienes. LTC4 (1 X 108M) contracted hamster smooth muscle cells (DDT1 MF2) in culture, and a leukotriene receptor antagonist, FPL 55712 (105M), inhibited contraction [37]. The morphologic changes resemble mesangial cell contraction. Also, this cell line had single high affinity, saturable, and decrease in CSA was affected. In addition, tJK-38485 (1 x specific binding sites for LTC4 with a Kd of 21 4 flM [371, or 105M) did not alter the percentage of cells responding to LTC4 5 nM [38], similar to the specific LTC4 binding sites in isolated (1 x 106M) and caused only a slight reduction in the maximum rat glomeruli [11] and well within concentrations that stimulate decrease in CSA (17.0 2.9% for LTC4 and UK-38485 vs. 24.8 mesangial cell contraction. Furthermore, Krilis and coworkers 3.6% for LTC4 alone). Both EP-092 and UK-38485 had no found specific LTC4 receptors on segments of longitudinal independent effect on CSA at the concentrations used. These smooth muscle from guinea pig ileum, dissociated cells from data do not support a role for TxA2 in mediating the morpho- these segments, and in subcellular fractions [39]. Our experiments with serine borate complex suggest LTC4, logic changes caused by LTC4. without conversion to LTD4, can cause contraction. The rapid

TxA2 might play in LTC4-mediated contraction, we used

EP-092, a TxA2 receptor antagonist, and UK-38485, a thromboxane synthetase inhibitor. Figure 4 shows that when mesangial cells were incubated with LTC4 in the presence of EP-092, neither the percentage of cells responding nor the maximum

Discussion

conversion of LTC4 to LTD4 is catalyzed by the enzyme

y-glutamyltranspeptidase [401 (EC 2.3.2.2). This membrane bound enzyme is found in high concentrations in the kidney of various species, including the rat [41]. In some cells, however, there appear to be specific receptors for both LTC4 and LTD4 [21, and therefore mesangial cells may respond to LTC4 as well as LTD4 via separate receptors, but partial conversion of LTC4 to LTD4 may enhance the response to LTC4. There are many possible explanations of the partial responthat 41% of the cells responded. This value agrees with the siveness of mesangial cells to sulfidopeptide leukotrienes, that work of Mahieu et al [31] and Tanaka et al [331. Using similar is, only one—third of the cultured cells contracted. Cultured techniques, Tanaka and co-workers [331 found a faster time mesangial cells may lose leukotriene receptors. A decline in course with ANG II, a difference possibly attributable to the receptor number, affinity, or receptor linkage to intracellular use of mesangial cells in primary culture as they emerge from mediators could lead to a loss of responsiveness. Furthermore,

Image analysis microscopy permits the rapid, quantitative

evaluation of microscopical images. Image analyzers have been used to study the morphology of a variety of cells and tissues [22—24, 30]. We successfully applied this technique to measure morphologic changes in cultured rat mesangial cells exposed to vasoconstrictors. Using ANG 11(1 X 106M), a vasoconstrictor peptide that decreases the ultrafiltration coefficient in vivo [141 and contracts mesangial cells in vitro [12, 13, 31, 32], we found

530

Simonson and Dunn

when the cells are maintained in culture, they may lose part of their contractile apparatus or the cell—substrate interaction could retard or completely inhibit contraction in some cells. Most mammalian cells synthesize extracellular matrices that bind directly to artificial surfaces such as plastic and glass. These matrices are essential for cell growth and differentiation in vitro. Both our experiments with cells cultured on teflon, and Tanaka's (33] experiments with less adherent primary cultures, suggest that cell—substrate interactions influence the response

of mesangial cells to vasoconstrictors. Specifically, when mesangial cells are cultured on less adhesive teflon surfaces, the percentage of cells responding to LTD4 was greater than with cells cultured on glass.

The contribution of cyclooxygenase products to leukotriene—mediated contraction in mesangial cells appears negligible. Similarly, kinetic analyses of TxA2 synthesis in guinea pig lung strips, exposed to the sulfidopeptide leukotrienes, suggest

that TxA2 plays little or no role in the contractile response to leukotrienes [42, 43]. LTD4-mediated contraction in our experiments was not accompanied by enhanced TxA2 production. Although LTC4 did stimulate a small amount of TxA2, the contraction was unaffected by an inhibitor of TxA2 synthesis (UK-38485) or a TxA7 receptor blocker (EP-t392). Our results with UK-38485 and EP-092 are consistent with experiments in

isolated perfused rat kidney in which OKY-1581, another thromboxane synthetase inhibitor, and indomethacin failed to inhibit the vasoconstriction caused by LTC4 and LTD4 19]. The failure of LTC4 and LTD4 to increase PGE2 production contrasts with the potent stimulation of PGE2 by angiotensin II and arginine vasopressin [26]. Vasodilatory prostaglandins antagonize the reduction of GFR and K1 by ANG II and arginine vasopressin [44]. Presumably, without the counterbalancing effects of PGE2, the mesangial contraction and the reduction in K1 caused by leukotrienes could he profound [45]. It is unknown

whether glomerular mesangial or epithelial cells synthesize leukotrienes. Infiltrating polymorphonuclear leukocytes and monocytes during immune injury to the glomerulus could

2, LEwis RA, AUSTEN KF: The biologically active leukotrienes. J Clin Invest 73:889—897, 1984

3. Huoovis P, DAHLEN SE, GusTAussoN L, HAMMAR5TROM 5, SAMuEL550N B: Biological profile of leukotriencs C4 and D4. Acta PhysiolScand 110:331—333, 1980 4. Sip.ois P, Roy 5, TETRAIJLF JP, BORuEAT P, PICARn 5, CORn' EJ:

Pharmacological activity of leukotrienes A4, B4, C4 and D4 on

selected guinea—pig, rat, rabbit, and human smooth muscles. Prostaglandins Med 7:327—340, 1981

5. FEURERsTEIN N, FoEon, RAMwEL1, P: Leukotrienes C4 and D4 induce prostaglandin and thromboxane release from rat peritoneal macrophages. Br J Phorrnacol 72:389—391, 1981 6. FoLco G, HAN550N G, GRAN5TROM E: Leukotriene C4 stimulates TxA2 formation in isolated sensititized guinea pig lungs. Biochem Pharmacol 30:2491—2493, 1981 7. CRAMER EB, PoLooE L, PAwLowsKI NA, COHN ZA, SCOTT WA:

Leukotriene C promotes prostacyclin synthesis by human endothehal cells. Proc NatI Acad Sci 80:4109—4113, 1983 8. BADR K, BAvLI5 C, PFEFFERJ, PFEPFERM, SOBERMAN RJ, LEwis

RA, AUSTEN KF, COREY EJ, BRENNER B: Renal and systemic hemodynamic responses to intravenoos infusion of leukotriene C4 in the rat. Circ Res 54:492—499, 1984 9. ROSENTHAL A, PACE—A5CIAK CR: Potent vasoconstrietion of the isolated perfused rat kidney by Icukotrienes C4 and D4. Can J PIp's Pharmacol 61 :325—328, 1983

10. PIPER PJ, STANTON A, McLEon L: The actions of leukotrienes C4 and D4 in the porcine renal vascular bed. Prostaglandins 29:61—73, 1985

Ii. BALLERMANN BJ, LEwis RA, COREY EJ, AUSTEN KF, BRENNER

BM: Identification and characterization of leukotriene C4 receptors in isolated rat renal glomeruli. Circ Res 56:324—330, 1985 12. AU5IELL0 DA, KREISBERG JJ, ROY C, KARNOVSKY MJ: Contrac-

tion of cultured rat glomerular cells afier stimulation with angintensin II and arginine vasopressin. J Clin Invest 65:754—760, 1980 13. BRENNER BM: Control of glomerular function by intrinsic contractile elcments. Fed Proc 42:3045—3079, 1983 14. DwORKIN LD, IcHIKAwA I, BRENNER B: Hormonal modulation of glomerular function. Am J Physiol 244:F95—Fl04, 1983 15. LEVINE L, MORGAN R, LEwIs R, AU5TEN K, CLARK D, MARFAT

A, COREY F: Radioimmunoassay of the leukotrienes of slow

reacting substance of anaphylaxis. Proc Nail Acad Sd 78: 7692—7696, 1981

16. ARMSTRONG RA, JONES RL, PEESAPATI V, WILL SG, WILSON,

NH: Competitive antagonism at thromboxane receptors in human platelets. Br J Pharmacol 84:595—607, 1985

release LTC4 and LTD4 within the glomerulus. In fact, Tucker

17. JONES RL, PEESAPATI V, WILSON NH: Antagonism of the

et al found that leukocytes are necessary for lowering K1 in glomerular immune injury [46]. The potential role of these leukotrienes to reduce GFR in glomerulonephritis can be evaluated when effective lipoxygenase inhibitors become available for in vivo use.

saphenous vein and guinea—pig trachea. Br J Pharmacol 76:

Acknowledgments Part of this work was presented in December, 1984 to the American

Society of Nephrology and was published as an abstract in Kidney International 27:256, 1985. This work was supportcd by thc NIH, HL

22563. We are grateful to Linda Goldberg and Norma Minear for secretarial help. UK-38485, an inhibitor of thromboxane synthesis, was a gift from Pfizer Pharmaceuticals, New York, USA. Synthetic LTC4, LTD4, and a structural analog of LTD4 (4R,55,6Z-nor-LTD1) were gifts from Dr. John Gleason of SmithKline and French Laboratories, Philadelphia, Pennsylvania, USA. EP-092 used in this study was a gift from Dr. R,L. Jones, Department of Pharmacology, Edinburgh University.

Reprint requests to Dr. Michael J. Dunn, Division of Nephrology, University Hospitals of Cleveland, Cleveland, Ohio 44106, USA.

References 1. SAMUELSsON B: Lcukotricncs: Mediators of immediate hypersensitivity reactions and inflammation. Science 220:568—575, 1983

thromboxane—sensitive contractile systems of the rabbit aorta, dog 423-438, 1982 18. STRIKER GE, KILLEN P0, FARRIN FM: Human glomerular cells in

vitro: Isolation and characterization. Transplant Proc 12(suppl l):88—99, 1980

19. LOVETT D, STERZEL RB, KA5HGARIAN M, RYAN JL: Neutral

protciuase activity produced in vitro by cells of the glomerular mesangium. Kidney Int 23:342—349, 1983 20. KREI5BERu J, KARNOv5KY MJ: Glomerular cells in culture. Kidney Int 23:439—447, 1983

21. GABRIDGE MG: The chamber/dish: An improved vessel for cell and explant culture. In vitro 17:91—97, 1981

22. BRADHURY 5: Microscopical image analysis: Problems and approaches. JMicrosc 115:137—150, 1978 23. BRADBURY 5: Quantitative image analysis, in Analytical and Quantitative Methods in Microscopy, edited by MEEK GA, EIDER HY, Cambridge, Cambridge University Press, 1977, pp. 91—116

24. HUANO HK: Biomedical image processing. CRC Crit Rev Bioengineering 5:185—271, 1981 25. MARKwELL MK, HAA5 SM, BIEIIER LL, TOLBERT NE: A modifi-

cation of the Lowry procedure to simplify protein determination in membrane and hipoprotein samples. Anal Biochem 87:206—210, 1978

26. SCHARSCHMIDT LA, DUNN MJ: Prostaglandin synthesis by rat glomerular mesangial cells in culture. J Clin Invest 71:1756—1764, 1983

Leukotrienes contract mesangial cells 27. BERGMEYER HU, BERNT E: Lactate dehydrogenase, in Methods of

Enzymatic Analysis, edited by Bergmeyer HU, Deerfield Beach, Verlag Chemie Tnt., 1974, pp. 574—579 28. GLEASON JG, Ku TW, MCCARTHY ME, WEICI-IMAN BM, HOLDEN D, OSBORN RR, ZABKO—POTAPOVICH B, BERKOWITZ B, WASSER-

MAN MA: 2-nor-leukotriene analogs: Antagonists of the airway and

vascular smooth muscle effects of leukotnene C4, D, and E4. Biochem Biophys Res Comm 117:732—739, 1983 29. WEICHMAN BM, WASSERMAN MA, HOLDEN DA, OSBORN RR, WOODWARD DF, Ku TW, GLEASON JG: Antagonism of the pul-

monary effects of the peptide leukotrienes by a leukotriene D4 analog. J Pharmacol Exp Ther 227:700—705, 1983 30. SIM0Ns0N MS: Image analysis microscopy: Principles and applications. Biotechniques 3:484—493, 1985

31. MAHIEU PR, FOIDART JB, DuBoIs CH, DECHENNE CA, DEI-IENEFFE J: Tissue culture of normal rat glomeruli: Contractile

activity of the cultured mesangial cells. Invest Cell Pathol 3:121—128, 1980

32. FOIDART J, SRAER J, DELARuE F, MAHIEU P, ARDAILLOU R:

Evidence for mesangial glomerular receptors for angiotensin II linked to mesangial cell contractility. FEBS Lett 121:333—339, 1980 33. TANAKA T, FUJIWARA Y, ORITA A, SASAKI E, KITAMURA H, ABE

H: The functional characteristics of cultured rat mesangial cell. Jpn Circ J 48:1017—1029, 1984

34. PENIT J, FAURE M, JARD S: Vasopressin and angiotensin II receptors in rat aortic smooth muscle cells in culture. Am J Physiol 244:E72—E82, 1983

35. Ivs HE, SCHULTZ GS, GALARDY RE, JAMIESON JD: Preparation of functional smooth muscle cells from the rabbit aorta. J Exp Med 148:1400—1413, 1978

36. DRAZEN JM, AUSTEN KF, LewIs RA, CLARK DA, GoTo G, MARFAT A, COREY EJ: Comparative airway and vascular activities

of leukotrienes C-l and D in vivo and in vitro. Proc Nail Acad Sci USA 77:4354—4358, 1980

531

37. CLARK MA, COOK M, MONG S, HOGABOOM G, SI-IORR R, STADEL

J, CROOKE S: Leukotriene C4 ([3H1-LTC4) binding to membrane isolated from a hamster smooth muscle cell line (DDT1MF2). Life Sci 35:441—448, 1984

38. KRILIS S, LEWIS RA, COREY EJ, AUSTEN KF: Specific receptors

for leukotriene C4 on a smooth muscle cell line. J Clin Invest 72:1516—1519, 1983

39. Kuis S, LEWIS RA, COREY EJ, AuSTEN KF: Specific binding of leukotriene C4 to ileal segments and subcellular fractions of ileal smooth muscle cells. Proc Nail Acad Sci 81:4529—4533, 1984 40. PACE—ASCIAK CR, SMITH WL: Enzymes in the biosynthesis and

catabolism of the eicosanoids: Prostaglandinds, thromboxanes, leukotrienes, and hydroxy fatty acids, in The Enzymes, Vol. 16, edited by BOYER PD, New York, Academic Press, 1983, p. 544—603.

41. MEISTERAS, TATE S, GRIFFITH OW: y-glutamyltranspeptidase, in

Methods in Enzymology, vol. 77, edited by JAKOBY WB, New York, Academic Press, 1981, p. 237—253 42. AUSTEN KF, COREY EJ, DRAZEN JM, LEITCH AG: The effect of indomethacin on the contractile response of the guinea—pig paren-

chymal strip to leukotrienes B4, C4, D4, and E4. Br J Pharmacol 80:47—53, 1983

43. CREESE BR, BACH MK, FITZPATRICK FA, BOTHWELL WM: Leukotriene—induced contraction and thromboxane production in guinea—pig lung parenchymal strips. EurJ Pharm 102:197—204, 1984 44. BAYLIS C, BRENNER BM: Modulation by prostaglandin synthesis inhibitors of the action of exogenous angiotensin II on glomerular ultrafiltration in the rat. Circ Res 43:889—898, 1978 45. SCHARSCHMIDT LA, LIANOS E, DUNN MJ: Arachidonate metabolites and the control of glomerular function. Fed Proc 42:3058—3063, 1983

46. TUCKER BJ, GUSHWA LC, WILSON CB, BLANTZ RC: Effect of leukocyte depletion on glomerular dynamics during acute glomerular immune injury. Kidney mi 28:28—35, 1985