The effects of plasma from IGA nephropathy patients on vascular prostacyclin and platelet cyclic AMP production

Prostaglandins Leukotrienes and Essential Fatty Acids (1989) 35, 9-13 0 Lon@tan Group UK Ltd 1989

The Effects of Plasma from IGA Nephropathy Patients on Vascular Prostacyclin and Platelet Cyclic AMP Production S. TORI*,

J. NAGY+, C. TORDAY*,

2. HAVASS*

and C. BERECZKY*

*Depa/tments of Paediatrics and *Experimental Surgery, Albert Szentgyiirgi University School, Szeged 6725, 14 Korzjnyi Str., 2nd Department of Medicine, ‘University Medical Hungary (reprint request to ST)

Medical School,

P&s,

Abstract - The effects of plasma from 10 IgA nephropathy patients and from ten controls were studied on vascular prostacyclin (PG12) production, the cyclic AMP (CAMP) level and the aggregation of normal platelets. The ability of the plasma to support PGl*-like activity (PSA) was significantly lower in the group of patients (18.0 k 13.3%) than in the controls (52.6 + 12.9%). The concentration of 6-keto-PGF, alpha in the supernatant of the vascular tissue was also lower following incubation with patient plasma than with control plasma (p < 0.001). The reduced PG12 released by the patient plasma led to a significantly lower platelet CAMP than that following the control plasma (p c 0.01). There was a significantly positive correlation between the 6-keto-PGF, a~phaand the plasma PSA, and also between both the plasma PSA and 6-keto-PGF I alphs concentrations and the platelet CAMP level. These findings suggest that a vascular PG12 defect may cause a reduced CAMP production and an uninhibited aggregation of platelets, which might play a role in the pathogenesis of IgA nephropathy.

Introduction

Prostacyclin (PGL), the main cycle-oxygenase product of the large vessel endothelium, is released into the blood-stream at the luminal surface and by raising platelet cyclic AMP (CAMP). inhibits platelet activation (1, 15). CAMP is perhaps the most important inhibitory regulator of platelet function. An increase in platelet CAMP leads to sequestration of intraplatelet calcium, reducing the availability of cytosolic calcium for platelet activation processes (5).

A fall in vessel wall PGI? generation could affect the vascular tone and the vessel-platelet interaction, which could lead to thromboembolic problems in large vessels or down-stream in microvessels. An increasing number of data are available on the possible role of a PGIZ defect in microangiopathies of various origins. Remuzzi et al. (17) first reported deficient PGIz production in haemolytic uraemic syndrome and thrombotic thrombocytopenic purpura, and this has been confirmed by several groups (2. 8. 20).

10

PROSTAGLANDINS

A decrease in PGI1! stimulating plasma factor was observed in Cyclosporin-A treatment (14), and in diabetic microangiopathy (17, 22). Recently we reported on a reduced ability of the plasma to support PGII-like activity (PSA) in patients with IgA-nephropathy (23) and the related disease Henoch-SchGnlein purpura (HSP) (21, 23). The plasma concentration of 6-ketoPGF , alpha(a stable PGI2 metabolite) was significantly lower in HSP than in controls (21). In our present study we have investigated the effects of PG12 (released from vascular tissue on the action of plasma from IgA nephropathy patients and controls) on the aggregation and CAMP production of normal platelets. Materials and Methods Ten patients with IgA glomerulonephritis (aged 36.3 f 9.8 years, mean + SD), and 10 controls matched for age and sex, were studied. The control individuals were free of renal, cardiovascular, pulmonary and inflammatory diseases. The serum creatinine was 151.9 + 23.7 prno , and the urinary protein was 1.28 * 0.63 $ 24 hours in patients with IgA nephropathy. The average follow-up time for these patients was 12.3 + 4.6 years. In all cases, the blood pressure, plasma albumin concentrations, serum electrolyte levels, platelet counts and blood film showed no abnormalities. After informed consent had been obtained, 10 ml of venous blood was taken from the patients and from the controls. Blood for estimation of PSA was anticoagulated with 3.8% trisodium citrate in a ratio of 9:l. Platelet-poor plasma obtained for PSA estimation was prepared within 20 minutes of collection, by centrifugation at 2000 g for 10 minutes at 4°C. All plasma samples were stored at -40°C until examination. Platelet-rich plasma for aggregation studies was prepared from normal adults by centrifugation of titrated plasma at 800 g for 10 minutes at room temperature. The final platelet count was adjusted to 250-300 x lo’/1 by dilution with autologous platelet-poor plasma. No patient or control had been taking any drug known to alter the prostaglandin metabolism for at least 2 weeks before the study. Estimation of ability of plasma to support vascular PGIJike activity (PSA): Plasma PSA

was assessed by measurement of the platelet antiaggregatory activity by the method of

LEUKOTRIENES

AND ESSENTIAL

FATTY

ACIDS

Moncada et al. (13). Human umbilical arterial rings were obtained from freshly delivered umbilical cords. The umbilical arteries were freed from all surrounding tissue, cut into rings 1 mm in lenght (30-50 mg wet weight) and kept in Ringer’s buffer (pH 7.4) at 0°C for not more than 60 minutes. The rings were then incubated in 1 ml of TRIS buffer (pH 8.6) for 5 minutes at 37”C, and the PSA (platelet antiaggregatory activity) of 100 ~1 of the supernatant was tested. This was added to 200 ~1 of platelet-rich plasma in a Chronolog aggregometer. The mixture was incubated for one minute at 37°C after which collagen (Hormonchemie), at a final concentration of 2 pg/ml, was added. The t-ate of aggregation was recorded using a Radelkis recorder. The rings were then washed several times with Ringer’s buffer at 37”C, until no antiaggregatory activity could be detected. These “exhausted” rings were next incubated with 1 ml of test platelet-poor plasma at 37°C for 20 minutes. PSA was then assessed as before. The ability of each test platelet-poor plasma to produce PSA was expressed by calculation of the percentage difference in inhibition of platelet aggregation obtained with the same “exhausted” ring before and after the addition of test platelet-poor plasma (21). Radioimmunoassay studies of ti-keto-PGF, alphain the supernatant of umbilical arterial rings: The

urements of 6-keto-PGF, +,ha were carried out on the supernatants of fresh and “exhausted” umbilical arterial rings incubated in 1 ml of TRIS-buffer (pH 8.6) for 5 minutes and following a 20-minute incubation of the “exhausted” rings with a mixture of 1 ml of test platelet-poor plasma (obtained from the patients and from the controls) and 1 ml of TRIS-buffer, by radioimmunoassay using a modification of the method of Mitchell (12). The measurement was performed without extraction procedures, as many extraction techniques may add further impurities (7). Our method gives good reproducibility and recovery of added 6-keto-PGF, alpha. and has a sensitivity of 5 pg/ml. This assay has the following cross-reactivities at 50% normalized percent bound: 6-keto-PGF] aloha lOO%, PGF? +,,,a 0.8%, PGE? 1.4%, PGF, alpha2.0%. PGE, 1.2%. Measurement of platelet CAMP by binding assay method: Following platelet aggregation, platelets

were exposed and CAMP was released by cyto-

VASCULAR PROSTACYCLIN AND PLATELET CYCLIC AMP PRODUCTION

lysis. During platelet preparation, platelet CAMP was protected from CAMP-dependent phosphodiesterase (PDE) with a diluting buffer containing (in final concentrations) 0.5 mM IBMX (3isobutyl-I-methylxanthine, Sigma 15879, a specific PDE inhibitor), 4 mM EDTA (Sigma) and 50 mM TRIS-HCl buffer (pH 7.5), and the mixture was centrifugated at 800 g for 10 minutes at 4°C. The sediment was lysed in a h;Ipotonic solution containing 4 mM EDTA (Sigma), 8 mM Theophylline (Sigma) and 20 mM TRIS-HCI buffer (pH 7.5). For more perfect exposure of the platelets, cytolysis was completed by boiling of this mixture for 2 minutes in a hot water bath (6). Protein-free supernatant was obtained by centrifugation for 30 minutes at 6000 g in a refrigerated Janetzky K 24 centrifuge. CAMP was determined in 50 ~1 of the supernatant of the lysed platelets with a CAMP binding assay kit (Amersham TRK 432). The composition of the buffer used for binding assay corresponded to that applied for cytolysis. Radioactive samples were measured in a tritontoluene cocktail in a tritium programme with a Beckman LS 100 C liquid scintillation counter. All platelet aggregation, CAMP and 6-ketoPGF, alphastudies were carried out in duplicate. Statistical analysis was performed with the Student t test and the Sperman rank correlation test.

cbnhl la-ml

11

cmhdl6A-6n

hhm

Ia-6N

Fig 1A. Plasma PGI,-supporting activity (PSA) IGA-GN: IgA nephropathy Control (n = 10): X + SD = 57.7 + 12.9% IGA-GN (n = 10): f f SD = 18.0 + 13.3% (p < 0.01) Fig 1B. Concentration of 6-keto-PGFt alpha in the supernatant of umbilical arterial rings following incubation with test (control/patient) plasma Control: ir + SD = 20.2 f 4.06 pg/ml, IGA-GN: 9.2 + 5.6 pg/ml (p < 0.01) Fig IC. The level of the platelet CAMP following incubation of normal platelets with the supernatant of arterial rings regenerated with test (control/patient) plasma Control: ic + SD = 62.06 + 24.3 pmol cAMP/109 platelets IGA-GN: X t SD = 27.7 f 20.4 pm01 cAMP/lO’ platelets (p < 0.05)

between

the groups was significant (p <

Measurement

of platelet CAMP by binding assay:

ence

0.01). Results The ability of plasma to support PGtJike activity (PSA)

was reduced in 9 of the 10 patients with IgA nephropathy studied (18.0 + 13.3%) (mean + SD). In 4 cases, the PSA was very low (< 10%) or undetectable. The difference between the results for the IgA nephropathy group and the adult control group (52.7 + 12.9%) was significant (p < 0.01). Results are given in Figure 1A. Radioimmunoassay studies of 6-keto-PGF, alphain the supernatant of umbilical arterial rings: The

level of 6-keto-PGF, illphain the supernatant of fresh umbilical arterial rings was immeasurably high (> 4OOUpg/ml). Following “exhaustion” it fell to 7.1 +- 1.2 p&/ml (n = 10 samples). The changes following incubation of the “exhausted” with the plasma of IgA nephropathy ;r?&ts (9.2 + 5.6 pg/ml) or control cases (20.2 + 4.06 pg/ml) are given in Figure 1B. The differ-

The value of CAMP was 210.9 + 25.9 pmol cAMP,/lO’ platelets following incubation of platelets with the supernatant of fresh umbilical arterial rings. This decreased to 43.2 + 6.2 pmol cAMP/‘lOY platelets after the “exhaustion” of PGIz-producing ability (n = 10 samples). The difference was significant (p < 0.01). There was a significant difference (p < 0.05) between the platelet CAMP results following incubation with the supernatant of umbilical arterial rings, which had been regenerated with control (62.06 rt 24.3 pmol CAMP/IO’ platelets) or patient plasma (27.7 + 20.4 pmol cAMP/lOY platelets) previously (Fig. 1C). There was a significant correlation between plasma PSA (patients and controls) and the values of 6-keto-PGF, alpha measured in the supernatant of the “exhausted” umbilical arterial rings following incubation with test platelet-poor plasma (patients and controls) (r = 0.84. p < 0.05). A significant correlation was also found

PROSTAGLANDINS

J/

_

0

I

20

40

*

I

60

.

I

60

.

.

too PSAWI

Fig 2. Correlation between the plasma PSA and the platelet CAMP values following incubation of normal platelets with the supematant of arterial rings regenerated with test (control/patient) plasma x IgA nephropathy l Control r = 0.81, p < 0.05

between the plasma PSA and the platelet CAMP values (r = 0.81, p < 0.05). Discussion The production of PGI2 from vascular tissue is a complex multistep event. Phospholipase, cyclooxygenase and PGI2 synthetase are the principal enzymes responsible for this production. However, little is known about the regulation of the enzymatic steps, and normal plasma might contain some regulatory factors. Some authors (11, 16, 17) have found that normal plasma has a stimulatiang activity. Subsequent attempts to characterize this activity have revealed the presence of a factor with a molecular weight of < 10 000, which apparently acts both by enhancing the release of endogenous arachidonic acid from all membranes and by preventing self-deactivation of the vascular cycle-oxygenase system (4). Various abnormalities could be at the origin of a PG12 deficiency, including an absence or low level of PGIz-stimulating factor (17), an accelerated degradation of PGIZ (3), a depressed preservation of the PGI2 effect (21), an inhibition of PGIz production (9) or PGI*-like activity (21), or defective serum binding of PG12, which may reduce the PG12 availability at damaged vascular sites (24). It has been

LECIKOTRIENES

AND ESSENTIAL

FATTY

ACIDS

suggested that at least part of this inhibitory activity may be secondary to the presence of circulating lipid peroxides (9), which are known to inhibit PGI2 synthesis. Further characterization of this ‘inhibitory activity is obviously required. This defect in PSA is not specific to a single type of vascular disease and has been demonstrated in a variety of different conditions, as diverse as sickle cell anaemia (19), acute glomerulonephritis (2, 10) and Henoch-Schonlein purpura (21). Recently, we reported a decreased plasma PSA in patients with IgA nephropathy (23), which was associated with a high level of low-density lipoprotein and a diminished concentration of high-density lipoprotein in the plasma of these patients as compared with controls. However, the levels of arachidonic acid and its precursors were not lower in the plasma of patients than in the controls. Therefore, the decreased PGIz synthesis may play an important role in the pathogenesis of IgA nephropathy, but it can not be explained by a low level of PG precursors. In our present study we confirmed our previous findings concerning the low plasma PSA level in the patients with IgA nephropathy relative to that in the controls. This was in accordance with the significantly lower concentration of 6-keto- PGFi alpha released by the effect of patient plasma from the “exhausted” umbilical arterial rings. 6-keto- PGFi alp,,a has a very diminished inhibitory activity on platelet aggregation; nevertheless, a lower level of an inactive metabolite of PGIz may reflect a reduced PGIZ production. A further observation was the significantly lower concentration of CAMP released from the platelets in response to the supernatant of vascular rings incubated with patient plasma. On the action of the control plasma, the rings produced more PSA, which increased the platelet CAMP concentration and inhibited the aggregation of these platelets. These findings suggest that a defect of vascular PC12 production may cause a reduced platelet CAMP release with an increased platelet aggregability, which might play a role in the pathogenesis of IgA nephropathy. Further study is required to observe the changes in platelet CAMP level in these patients. It is very likely that a decreased vascular PGIz formation is a secondary phenomenon in the development of vasculitis. The primary event is the vascular damage caused by immune complex deposition.

VASCULAR

PROSTACYCLIN

AND PLATELET

CYCLIC

13

AMP PRODUCTION

Apart from the oxidation products of the cycle-oxygenase pathway, hydroxy fatty acids (hydroxyeicosatetraaenic acids) and leukotrienes are synthesized via the lipoxygenase pathway, which can also influence the pathophysiological events in glomerulonephritis (10). Therefore, deficient PGIz production is not the only factor responsible for the glomerular diseases. Nevertheless. the absence of PGIz could play an important role in the progression of the angiopathy.

References Brace L D, Venton D L, Le Breton G C. Thromboxane Az/prostaglandin H, mobilizes calcium in human platelets. Circulation (Suppl) 64: 1198, 1982. Chamone D A. Proesmans W C. Monnens L A, Binda ki Muaka P. Vermylen J R. Reversible deficient prostacyclin release in childhood hemolytic uremic syndrome. Int J Ped Nephrol 3: 13-16, 1982. Chen Y C. Hall E R, McLeod B, Wu K K. Accelerated prostacyclin degradation in thrombotic thrombocytopenic purpura. Lancet ii: 267-269. 1981. Deckmyn H. Zoja C, Arnout J, Todisco A, Van den Bulcke F, D’Hondt L, Hendrickx N. Gresele P. Vermylen J. Partial isolation and function of the prostacyclin regulating plasma factor. Clinical Science 69: 383-393. 1985. Fenstein M B. Egan J J. Shaah R J. White J. The cytoplasmic concentration of free calcium in platelets is controlled by stimulators of cyclic AMP production (PGDZ. PGE,. forskolin). Biochem Biophys Res Commun 113: 598-604, 1983. Furakawa H. Bilezikian J P. Loeb J N. Effects of oabain and isoproterenol on potassium influx in the turkey erythrocyte. Quantitative relation to ligand binding and cyclic AMP generation. Biochimica et Biophysics Acta 598: 345-356. 1980. Greaves M. Preston F E. Plasma 6-keto-PGF, .,iph.,: fact or fiction. Thromb Res 26: 145-157. 1982. Joorgensen K A. Pedersen R S. Familial deficiency of prostacyclin production stimulating factor in the hemolytic uremic syndrome of childhood. Thromb Res 21: 311-316, 1081. Levin M, Elkon K B, Nokes T J C, Buckle A M. Dillon M J. Hardisty R M, Barratt T M. Inhibitor of prostacyclin production in sporadic hemolytic uremic syndrome. Arch Dis Child 58: 703-708, 1983.

IO. Lianos

E A.

metabolites

Biosynthesis

and role

in glomerulonephritis.

of arachidonic Nephron

acid

37: 73-77.

1984. Il.

McIntyre D E. Pearson J D. Gordon J L. Localisation and stimulation of prostacyclin production in vascular cells. Nature 271: 549-551. 1978. for 6-keto12. Mitchell M D. A sensitive radio-immunoassay PGF,, Preliminary observations on circulating concentrations. Prostaglandins Med 1: 13-21. 1978. 13. Moncada S, Higgs E A. Vane J R. Human arterial and venous tissue generate prostacyclin (prostaglandin X) a potent inhibitor of platelet aggregation. Lancet i. 18-20. lY70. 13. Neild G H. Rocche G, Imberti L, Fumagalli F, Brown Z, Remuzzi G. Williams D G. Effect of Cyclosporin .4 on prostacyclin synthesis by vascular tissue. Thromb Rcs 32: 373-379. 1983. 1.5 Owen N E. Le Breton G C. Ca” mobilization in blood platelets as visualized by chlortetracycline fluorescence. Am J Physiol 241: H613-H619, 1981. 16 Pommerantz K B, Tall A R. Feinmark S J, Cannon P J. Stimulation of vascular smooth muscle cell prostacyclin and prostaglandin E: synthesis by plasma high and low density lipo-protein. Circ Res 54: 554-565, 1984. 17. Remuzzi G, Marchesi D, Mecca G. Misiani R, Livio M. De Gaetano G. Haemolytic uraemic syndrome: deficiency of plasma factor(s) regulating prostacyclin activity? Lancet ii: 871-872. 1978. M. Uchino H. Decreased plasma 18. Sensaki S. Okuma prostacyclin-regenerating activities in diabetics. Prostaglandin Leukotriene and Med 20: 197-207. 1985. 19. Stuart M J. Sills R H. Deficiency of plasma prostacyclin or PGI, regenerating ability in sickle cell anacmia. Brit J Haematol 48: 545-550. 1981. platelet 20. Stuart M J. Spitzer R E. Coppe D. Abnormal and vascular prostaglandin synthesis in an infant with hemolytic uremic syndrome. Pediatrics 71: 120-114. lox.?. Turi S, Belch J J F, Beattie T J. Forbes C’ 0. Abnormalities of vascular prostaglandins in Henoch-Schiinlein purpura. Arch Dis Child 61: 173-177. 1986. 22. T’uri S. Magyari M. Nemeth M. Bereczky Cs. Plasma factors influencing prostacyclin-like activity in patients with diabetic microangiopathy. Prostaglandin Leukotriene and Med in press. 23. Turi S. Nagy J. Haszon I. Havass Z. Nemeth M. Bereczky Cs. Disturbances of lipoprotein and PGI?-metabolism in IgA nephropathy and Henoch-Schonlein purpura. In Contributions to Nephrology: Pardiatric Nephrology Ed Boda D. Turi S. Karger Base1 in press. 24 Wu K K. Hall E R. Rossi E C. Papp A C. Serum prostacyclin binding defects in thrombotic thrombocytopenic purpura. J Clin Invest 75: 16X-174. 1085

21.