534
Agents and Actions vol. 5/5 (1975) Birkhfiuser Verlag, Basel
Platelets, Arthus-Type Reactions and Inflammatory Mediators by F.B. UBATUBA a n d S.H. FERREIRA1) The Wellcome Research Laboratories, Langley Court, Beckenham, Kent, BR3 3BS, England
Abstract The release of inflammatory mediators and the appearance of necrohaemorrhagle lesions induced by subcutaneous implantation of sponges containing antiplatelet serum globulins were studied in control and thromboeytopenie rats. In thromboeytopenie animals, antlplatelet globulins caused a greater release of prostaglanding-like material and 5-hydroxytryptamine as well as larger inflammatory lesions. Thus, platelet integrity is not necessary for the induction of lesions by antiplatelet globulins and the mediators in the sponge exudates must have originated from leucocytes or damaged tissues. Mast ceils seem no to be involved in the production of 5-hydroxytryptamine in this inflammatory reaction. It is suggested that in idiopathic thromboeytopenie purpura (ITP) an Arthus-type hypersensitivity reaction, rather than the thrombocytopenia itself, is the cause of the vaseniar lesions and their manifesta-
tions.
T h e early phase of the i n f l a m m a t o r y response to a s u b - p l a n t a r injection of anti-rat platelet serum does n o t depend o n the presence of circulating platelets [1]. The oedema, which was n o t due to the release of i n f l a m m a t o r y mediators from platelets could have been caused by direct vascular d a m a g e resulting from a crossreaction of the antiplatelet antibodies with b l o o d vessels endothelial antigens. I n the present study we have investigated the possible role of the mediators released from platelets i n the late phase of the i n f l a m m a t o r y reaction i n d u c e d by partially purified antiplatelet antibodies, b o t h in n o r m o t h r o m b o c y t e m i c a n d t h r o m b o c y t o p e n i c rats. I n order to correlate the degree of the i n f l a m m a t i o n with the release of mediators (histamine, 5-hydroxytryptamine a n d prostaglandins) we used the s u b c u t a n e o u s sponge technique, recently developed for the
detection of prostaglandins i n d u c e d exudates [2].
in
carrageenin-
Material and methods
(a) Experimental schedule Male albino rats (Charles River), 180-220 g were rendered thrombocytopenic by the injection of antiplate]et serum in the penis vein under light ether anaesthesia. The animals were left to recover from the shock for 30 minutes. Sponges impregnated with the phlogogen agent were then implanted subcutaneously. Twenty-four hours later the animals were lightly anaesthetized before decapitation and blood samples were taken by cardiac puncture for platelet counts. The sponges were then gently removed for collection of the exudates. Control animals, injected intravenouslywith normal rabbit serum, were implanted with sponges impregnated either with saline, normal rabbit serum globulins or normal rabbit serum. (b) Platelet counts
The blood samples (0.2 ml) preserved in small vials containing 5 mg of solid EDTA were shaken for 3-5 minutes before counting. Recently filtered ammonium oxalate (IX w/v) was used as diluent in haematological pipettes. Platelets were counted by phase contrast microscopy [3] in improved Neubauer ruled chambers. At least 300 cells were counted in normal blood samples. In thrombocytopenic animals the cells were counted in 400 small squares. (c) Treatment with compound 48/80
The drug was always dissolved in saline and injected intraperitoneally low down on the abdominal wall. The rats received 200 ~xgof the drug in the first day, 300 ~Lg in the second and 2 x 400 [xg in the third. This dosage completely degranulated tissue mast cells. The rats were implanted with the sponges 24 hours after the last injec~) Present address: Department of Pharmacology, Faculdade de Medicina de Ribeirfio Preto, $5.o Paulo, Brazil.
Platelets, Arthus-Type Reactions and Inflammatory Mediators
tion. Mast cells were studied in a mesenteric sheet fixed in 8 0 ~ alcohol and stained with toluidine blue (0.1 ~ w/v).
(d) Phlogogens Antiplatelet serum (APS) and antiplatelet serum globulins (APSG) were used in sponges to produce the inflammatory reaction. APS. Antibodies to rat platelets were raised in healthy New Zealand rabbits of about 2 kg body weight. The rat platelets used for the immunization were separated from the blood of large animals collected under ether anaesthesia. Heparin (500 IU/rat) was injected into the penis vein and the animals were exsanguinated through the abdominal aorta. Polycarbonate syringes and polypropylene needles were used for arterial puncture, the blood being immediately transferred after collection to a polyethylene bottle immersed in chipped ice. The blood was diluted with 0.2 volumes ofTRIS-EDTA-Saline (TES) containing NaCI (0.154 M), TRIS-HC1 buffer pH 7.4 (0.05 M ) and EDTA-2Na (0.001 M ) and centrifuged in 15 ml polyethylene tubes at 260g for 20 minutes at room temperature. The platelet-rich plasma was removed with a polyethylene pipette and the original volume of the blood was restored with chilled TES. After gentle mixing by inversion, the tubes were centrifuged again as above and the top layer removed. After pooling the platelet-rich supernatants the platelets were sedimented and washed at least 6 times with chilled TES solution. Low speed centrifugation (80-90 g) for 1 minute was intercalated in the washing cycles in order to remove residual red cells and leucocytes. Stained films (May-Grtinwald-Giemsa) of the final suspension of ceils were prepared for a check of the homogeneity of the platelet cell population. The concentration of cells in the final suspension was roughly adjusted to 1012 platelets/ml. Suspensions of platelets in TES were mixed with complete Freund's adjuvant for priming and with incomplete adjuvant for the boosters. Stable water-in-oil emulsions (1 : 1 v/v) were obtained by injecting the aqueous phase, in small portions, into the oily adjuvant and the mixing completed by repeated transfer between two glass syringes. Initially 1 ml of the emulsion was distributed into the hind foot pads and in several small sites in the back (intra- and subcutaneously), in each rabbit. After 20 to 30 days, a series of 4 injections, using incomplete adjuvant and spaced by two week intervals were made only in the back skin. The animals were thereafter injected intravenously at monthly intervals with a suspension of platelets in saline. Ten to twelve days after each booster dose, the animals were bled from an ear vein, the blood being collected under reduced pressure. After standing for 1 hour at 37~ the blood was kept at 4~ for at least 24 hours, to allow the clot to retract completely. The red cells present were removed by centrifugation and the clear serum was decomplemented by heating at 56~ for 30 minutes. After filtration through a 0.22/z Millipore membrane, the serum was stored frozen at - 2 0 ~ Serum samples from healthy untreated rabbits were similarly processed. The antiplatelet activity of the immuneserum was tested by its ability to produce throm-
535
bocytopenia, accompanied by a fall in blood pressure of anaesthetized rats, 1 to 1.5 minutes after intravenous administration. The batch of serum used in the present study produced regularly (0.3 ml/rat) a thrombocytopenia of less than l0 s platelets per ~zl of blood. APSG. Ammonium sulphate precipitation was used in the isolation of the globulin fraction from APS. One volume of active antiserum was slowly precipitated by the drop by drop addition of 1/8 volume of a saturated solution of ammonium sulphate recently adjusted to pH 7.8 with diluted NaOH solution. The suspension was gently stirred (magnetic bar) for 3 additional hours at room temperature. The precipitated proteins were collected by centrifugation and the supernatant discarded. After the precipitate was dissolved in borate-buffered saline (BBS; pH 7.8) to the original volume, the precipitation was repeated twice as above but the stirring shortened to 1 hour. The final precipitate was dissolved in a small volume of BBS and the resulting solution dialyzed (Visking cellulose tubing) against several changes of BBS at 4~ The dialysis was interrupted when the solution showed no more sulphate ions in the barium test. The final globulin fraction was centrifuged for 20 minutes at 5000 g in an MSE High-Speed 18 centrifuge set at 4~ and the insoluble material discarded. The volume was adjusted with BBS to a convenient value and the solution was filtered through a 0.22 Ix. Millipore membrane. The protein content of the final solution was determined by Lowry's method using bovine serum albumin as standard and adjusted to 4 mg protein/ml. The fraction was stored frozen at - 20~ NRSG. Globulin fractions from normal rabbit serum were prepared as above.
(e) Collection of the inflammatory exudates in sponges Polyester sponges measuring 4 x 1.5 x 0.5 cm were thoroughly washed in tap water and in distilled water before sterilization by autoclaving at l l 0 ~ for 15 minutes. Immediately before use the sponges were soaked in 1.5 ml of saline, NRS, APS, NRSG or APSG and implanted in one side of the abdominal mid-line under ether anaesthesia. The skin flaps were closed with suture clips and the animals were left to recover before being returned to a cage. Exudates were removed from the sponges by gentle washing and squeezing in 7 ml of heparinized saline in 10 ml graduated tubes, the final volume being measured to 0.1 ml. A sample (2 ml) of the diluted exudate was removed and vortexed with 0.4 lnl 6 M trichloracetic acid. The precipitated proteins were removed by centrifugation and the clear supernatant transferred to small specimen vials and stored frozen at - 2 0 ~ until assayed for histamine and 5-hydroxytryptamine. Distilled water was added to the remaining fluid to make a volume of 10 ml and the pH adjusted to 3 with concentrated formic acid. Ethanol (5 ml) and chloroform (25 ml) were added and the tubes shaken for 15 minutes to extract the acidic lipids. The chloroform was separated by centrifugation at 1000g for 15 minutes, filtered through Whatman 1 paper into a 100 ml pear shaped flask and
536
Platelets, Arthus-Type Reactions and Inflammatory Mediators
then evaporated at 35 ~ in a Buchi Rotavapor. The last traces of formic acid were removed from the residue by the evaporation of 2 ml ethanol and the flask was flushed with N2 for 5 minutes. The residue was taken up in Krebs' bicarbonate solution for bioassay. (f) Prostaglandin (PG) bioassay PG-like activity was measured on rat stomach strips superfused with Krebs' bicarbonate solution containing a mixture of antagonists in order to increase the specificity of the assay [4]. PGEe was used as standard in a bracketing assay mode, the activity in the exudate being expressed in PGE2-equivalents per ml.
(g) Fluorometry of histamine and 5-hydroxytryptamine Commercial o-phtaldialdehyde (OPT) was recrystallized from n-heptane after boiling for 1 hour under reflux with decolourizing charcoal. Traces of solvent were removed from the pale yellowish needles kept for a few hours in a vacuum dessicator over paraffin turnings. The dried OPT was stored in an amber bottle at 4~ All glassware used was routinely soaked for a few hours in a mixture (t:1, v/v) of concentrated HCI and HNO3. After thoroughly washing in tap and distilled water the glassware was dried at room temperature. Fluorescence measurements were made in 10 m m quartz cells using a Hitachi-Perkin Elmer (model MPF2A) spectrophotofluorometer. Histamine. Histamine was determined after condensation with OPT at pH 12.5 and low temperature[5-7]. The method was briefly as follows: the acidic extracts were shaken with diethyl ether (saturated with water) in order to remove the trichloracetic acid. Adequate volumes of the aqueous phase (5 to 100 ~tl) were then mixed in glass stoppered tubes with 2 ml of 0.05 N N a O H and 0.4 ml of a freshly prepared solution of OPT in methanol (0.05~ w/v). All tubes were cooled together in an ice-salt bath at - 15 ~ to - 18 ~ for 4-5 hours. The fluorescence was developed by the addition of 0.025 ml of 0.5 N HzSO4 (final pH 2) and measurements were made at 355/445 n m (uncorrected wavelength settings). Blanks for the reagent and the extracts were included in each series analysed. Working standards were prepared immediately before use by dilution of a stock solution of histamine di-hydmchloride (1 mg/ml as the base), stored frozen at - 20~ This solution was stable at least for 4 weeks. 5-Hydroxytryptamine. 5-HT was measured by the formation of a fluorophore with OPT in strong acid [8]. The reagent was always recently prepared by adding a solution of 50 mg of recrystallized OPT in 10 ml of absolute ethanol to 100 ml of 8 N HC1 [9]. Convenient aliquots of the trichloracetic extracts (25-250 ~zl) were directly added to 3 ml of the reagent in glass stoppered tubes. All the tubes were heated together in a boiling water bath for 10 minutes and then cooled in chipped ice. The mixtures were vigorously shaken for 1 minute with 2 ml of chloroform and the phases separated by centrifugation. The fluorescence was measured in the aqueous phase at 355/465 n m (uncorrected wavelength settings).
A blank for the reagent was included in each series processed. Working standards were prepared immediately before use by dilution of a stock solution of 5-HTcreatinesulphate (1 mg/ml as the base), stored frozen at -20~ This solution was stable for at least 2 weeks.
(h) Chromatography of the extracts In order to identify the prostaglandins present in the sponge exudates, the extracts were chromatographed in small silicic acid columns (3-4 g). The residues left after evaporation of the chloroform were taken into Mixture I ll0] and applied to the columns. Increased amounts of ethyl acetate in toluene were used to develop the chromatograms. The resultant fractions were evaporated to dryness in a Buchi Rotavapor at 35 ~ and the residues taken up in Krebs' bicarbonate solution for the assay on rat stomach strips. The prostaglandins in sponge exudates behaved essentially as PGE~ (80~oo) and PGF2~ (20%).
Drugs and chemicals The following drugs and compounds were used: Compound 48/80 (N-methyl homoanisylamine formaldehyde condensation product) from Wellcome Reagents Ltd., Histamine di-hydrochloride from Merck, Sharp and Dohme, 5-Hydroxytryptamine creatinesulphate from Sigma Chemical Co., Heparin from Evans Medical Ltd., U.K., Silicic acid (Unisil, 100-200 mesh) from Clarkson Chemical Co. Inc., Williamsport, Pc, USA, Freund's adjuvant from DIFCO. The dyestuffs were from Gurr Ltd. and other chemicals were commercially available analytical grade reagents. Statistics All comparisons between means were based on Student's t-test. The leveI of significance taken was 5%. Values given in the figures are Mean d: S.E.M. Results
Content of inflammatory mediators in 24-hour sponge exudates Prostaglandins, 5-hydroxytryptamine and histamine were detected in polyester sponges implanted subcutaneously in normothromboc y t e m i c r a t s ( F i g . 1). T h e r e w a s n o s i g n i f i c a n t difference in the content of the mediators studied in exudates induced either by NRSG- or APSGsponges, although the content of prostaglandins and histamine was significantly higher than that in saline-sponges. In thrombocytopenic rats ( F i g . 2) t h e c o n t e n t o f p r o s t a g l a n d i n s a n d 5hydroxytryptamine in APSG-sponge exudates was significantly higher than that in NRSGsponge exudates. Thus, the anti-rat platelets immunoglobulins caused a greater release of prostaglandins and 5-hydroxytryptamine but not of histamine. The APSG was a much stronger releaser of prostaglandins and 5-hydroxytryp-
Platelets, Arthus-Type Reactions and Inflammatory Mediators tamine in thrombocytopenic rats (compare Fig. 1 and 2). In order to evaluate the contribution of mast cells to the release of mediators by APSG, a group of thrombocytopenic rats was treated with compound 48/80 before the implantation of APSG-sponges. Figure 3 shows that there was no significant difference between the content of prostaglandins and 5-hydroxytryptamine in exudates from both groups of animals. Notwithstanding, the treatment significantly reduced the histamine content of the exudates. In a l l 48/80 treated rats, no granulated mast cells were seen in the mesentery preparations made at the moment the sponges were removed.
Inflammatory reaction produced by NRSG- and APSG-sponges Figure 4 shows a striking difference in the necro-haemorrhagic lesions induced by the
537
flammatory reaction. The A P S G produced a milder reaction in normothrombocytemic rats. However, the addition of rat platelets (1.4 • 109 cells) to APS-sponges produced in n o r m o t h r o m bocytemic animals a lesion equivalent to that observed when APSG-sponges alone were implanted in thrombocytopenic animals. The addition of platelets to NRS-sponges caused a mild lesion similar to that induced by NRSG-sponges. Thus, A g - A b complexes formed in sponges induced in normothrombocytemic rats a lesion similar to that observed when APSG-sponges were implanted in thrombocytopenic rats. Discussion
The results presented here show that A P S G was able to induce in normothrombocytemic rats an increased release of prostaglandin-like material as well as a less pronounced inflammatory lesion when compared with the effects observed in thrombocytemic animals. These results strong(BLOOD
250 (n=9)
BLOOD PLATELET5
pG E2- Equiv. ng .ml" 1
H ~g .ml"i
1
(n=iO) 2OO
800.
4
,oo
[
"[
I000-
(n=iO)
AOO-
(n:lO)
5i (n=10)
[
%
800
600-
/~00 100
= < i03.~I -I) H Bg. ml" I
[ T
I S A L I N E = 8,82 .+0.35 (xl0S.~tl-I) N R S G = 8.78 -+0.~5 tAPSG = 6.05 .+0.93
8-HT ng .ml"I
PLATELET5 S-HT ng. ml" i
P G E 2" Equiv. ng .ml -I
(n~lO)
l
3-
2-
-F
200 -
i-
I I
S NRS AP$ -G -G
S NRS A P S -G -G
S N R S A"PS -G -G
01
9
NRS APS -G -G
'
NRS APS -G -G
O
NRS APS -G -G
Figure 1
Figure 2
Content of prostaglandins (PGE2-equiv.), 5-hydroxytryptamine (5-HT), and histamine (H) in 24-hour sponge exudates induced by saline (S), normal rabbit serum globulins (NRSG), and anti-rat platelets serum globulins (APSG) in normothrombocytemic rats. The bars represent the means of n exudates • S.E.M.
Content of prostaglandins (PGE2-equiv.), 5-hydroxytryptamine (5-HT), and histamine (H) in 24-hour sponge exudates induced by normal rabbit serum globulins (NRSG), and anti-rat platelets serum globulins (APSG) in thrombocytopenic rats. The bars represent the means of n exudates ~z S.E.M.
538
Platelets, Arthus-Type Reactions and Inflammatory Mediators
APSG when compared with NRSG, in thrombocytopenic animals. This result indicates the participation of an immunoglobulin in the inly suggest that platelet integrity is not necessary for either the release of mediators or the production of the inflammatory lesions. Intravenous administration of APS and APSG in rats caused a very fast (ca. 1 minute) and prolonged (4-5 days) disappearance of circulating platelets [11]. Furthermore, when rat platelets were mixed at 37~ in vitro with specific APS and reconstituted lyophilized guinea-pig serum as a source of complement, they were rapidly agglutinated before lysis occurred, lysis which was nearly completed within a couple of hours [11], On the other hand, it is known that the half-life of rabbit antiplatelet antibodies is very short in the circulation of animals of other species [12]. One would then expect that in the animals of our experiments a
In conclusion, the lesions observed in the late phase of the inflammation induced in rats by antiplatelet serum is not consequent to the in the APSG preparation are certainly heterogeneous and possibly directed to a multitude of antigenic platelet components. Thus, Ag-Ab complexes other than those involved in platelets' lysis may possibly be circulating and eventually reach the inflammed sites. This might explain the mild but definite inflammation observed in thrombocytopenic animals implanted with NRSG-sponges. The presence of an excess of antibodies in APSG-sponges, reacting with platelet antigens, or Ag-Ab soluble complexes reaching the interstitial spaces, may cause their insolubilization and triggering an Arthus-type hypersensitivity reaction (reverse). This suggestion is supported by the fact that necro-haemorrhagic lesions were induced by the addition of platelets to APSG-sponges implanted in normothrombocytemic rats and that such lesions were similar to that observed when APSG-sponges (BLOOD PLATELETS -
Agents and Actions vol. 5/5 (1975) Birkhfiuser Verlag, Basel
Platelets, Arthus-Type Reactions and Inflammatory Mediators by F.B. UBATUBA a n d S.H. FERREIRA1) The Wellcome Research Laboratories, Langley Court, Beckenham, Kent, BR3 3BS, England
Abstract The release of inflammatory mediators and the appearance of necrohaemorrhagle lesions induced by subcutaneous implantation of sponges containing antiplatelet serum globulins were studied in control and thromboeytopenie rats. In thromboeytopenie animals, antlplatelet globulins caused a greater release of prostaglanding-like material and 5-hydroxytryptamine as well as larger inflammatory lesions. Thus, platelet integrity is not necessary for the induction of lesions by antiplatelet globulins and the mediators in the sponge exudates must have originated from leucocytes or damaged tissues. Mast ceils seem no to be involved in the production of 5-hydroxytryptamine in this inflammatory reaction. It is suggested that in idiopathic thromboeytopenie purpura (ITP) an Arthus-type hypersensitivity reaction, rather than the thrombocytopenia itself, is the cause of the vaseniar lesions and their manifesta-
tions.
T h e early phase of the i n f l a m m a t o r y response to a s u b - p l a n t a r injection of anti-rat platelet serum does n o t depend o n the presence of circulating platelets [1]. The oedema, which was n o t due to the release of i n f l a m m a t o r y mediators from platelets could have been caused by direct vascular d a m a g e resulting from a crossreaction of the antiplatelet antibodies with b l o o d vessels endothelial antigens. I n the present study we have investigated the possible role of the mediators released from platelets i n the late phase of the i n f l a m m a t o r y reaction i n d u c e d by partially purified antiplatelet antibodies, b o t h in n o r m o t h r o m b o c y t e m i c a n d t h r o m b o c y t o p e n i c rats. I n order to correlate the degree of the i n f l a m m a t i o n with the release of mediators (histamine, 5-hydroxytryptamine a n d prostaglandins) we used the s u b c u t a n e o u s sponge technique, recently developed for the
detection of prostaglandins i n d u c e d exudates [2].
in
carrageenin-
Material and methods
(a) Experimental schedule Male albino rats (Charles River), 180-220 g were rendered thrombocytopenic by the injection of antiplate]et serum in the penis vein under light ether anaesthesia. The animals were left to recover from the shock for 30 minutes. Sponges impregnated with the phlogogen agent were then implanted subcutaneously. Twenty-four hours later the animals were lightly anaesthetized before decapitation and blood samples were taken by cardiac puncture for platelet counts. The sponges were then gently removed for collection of the exudates. Control animals, injected intravenouslywith normal rabbit serum, were implanted with sponges impregnated either with saline, normal rabbit serum globulins or normal rabbit serum. (b) Platelet counts
The blood samples (0.2 ml) preserved in small vials containing 5 mg of solid EDTA were shaken for 3-5 minutes before counting. Recently filtered ammonium oxalate (IX w/v) was used as diluent in haematological pipettes. Platelets were counted by phase contrast microscopy [3] in improved Neubauer ruled chambers. At least 300 cells were counted in normal blood samples. In thrombocytopenic animals the cells were counted in 400 small squares. (c) Treatment with compound 48/80
The drug was always dissolved in saline and injected intraperitoneally low down on the abdominal wall. The rats received 200 ~xgof the drug in the first day, 300 ~Lg in the second and 2 x 400 [xg in the third. This dosage completely degranulated tissue mast cells. The rats were implanted with the sponges 24 hours after the last injec~) Present address: Department of Pharmacology, Faculdade de Medicina de Ribeirfio Preto, $5.o Paulo, Brazil.
Platelets, Arthus-Type Reactions and Inflammatory Mediators
tion. Mast cells were studied in a mesenteric sheet fixed in 8 0 ~ alcohol and stained with toluidine blue (0.1 ~ w/v).
(d) Phlogogens Antiplatelet serum (APS) and antiplatelet serum globulins (APSG) were used in sponges to produce the inflammatory reaction. APS. Antibodies to rat platelets were raised in healthy New Zealand rabbits of about 2 kg body weight. The rat platelets used for the immunization were separated from the blood of large animals collected under ether anaesthesia. Heparin (500 IU/rat) was injected into the penis vein and the animals were exsanguinated through the abdominal aorta. Polycarbonate syringes and polypropylene needles were used for arterial puncture, the blood being immediately transferred after collection to a polyethylene bottle immersed in chipped ice. The blood was diluted with 0.2 volumes ofTRIS-EDTA-Saline (TES) containing NaCI (0.154 M), TRIS-HC1 buffer pH 7.4 (0.05 M ) and EDTA-2Na (0.001 M ) and centrifuged in 15 ml polyethylene tubes at 260g for 20 minutes at room temperature. The platelet-rich plasma was removed with a polyethylene pipette and the original volume of the blood was restored with chilled TES. After gentle mixing by inversion, the tubes were centrifuged again as above and the top layer removed. After pooling the platelet-rich supernatants the platelets were sedimented and washed at least 6 times with chilled TES solution. Low speed centrifugation (80-90 g) for 1 minute was intercalated in the washing cycles in order to remove residual red cells and leucocytes. Stained films (May-Grtinwald-Giemsa) of the final suspension of ceils were prepared for a check of the homogeneity of the platelet cell population. The concentration of cells in the final suspension was roughly adjusted to 1012 platelets/ml. Suspensions of platelets in TES were mixed with complete Freund's adjuvant for priming and with incomplete adjuvant for the boosters. Stable water-in-oil emulsions (1 : 1 v/v) were obtained by injecting the aqueous phase, in small portions, into the oily adjuvant and the mixing completed by repeated transfer between two glass syringes. Initially 1 ml of the emulsion was distributed into the hind foot pads and in several small sites in the back (intra- and subcutaneously), in each rabbit. After 20 to 30 days, a series of 4 injections, using incomplete adjuvant and spaced by two week intervals were made only in the back skin. The animals were thereafter injected intravenously at monthly intervals with a suspension of platelets in saline. Ten to twelve days after each booster dose, the animals were bled from an ear vein, the blood being collected under reduced pressure. After standing for 1 hour at 37~ the blood was kept at 4~ for at least 24 hours, to allow the clot to retract completely. The red cells present were removed by centrifugation and the clear serum was decomplemented by heating at 56~ for 30 minutes. After filtration through a 0.22/z Millipore membrane, the serum was stored frozen at - 2 0 ~ Serum samples from healthy untreated rabbits were similarly processed. The antiplatelet activity of the immuneserum was tested by its ability to produce throm-
535
bocytopenia, accompanied by a fall in blood pressure of anaesthetized rats, 1 to 1.5 minutes after intravenous administration. The batch of serum used in the present study produced regularly (0.3 ml/rat) a thrombocytopenia of less than l0 s platelets per ~zl of blood. APSG. Ammonium sulphate precipitation was used in the isolation of the globulin fraction from APS. One volume of active antiserum was slowly precipitated by the drop by drop addition of 1/8 volume of a saturated solution of ammonium sulphate recently adjusted to pH 7.8 with diluted NaOH solution. The suspension was gently stirred (magnetic bar) for 3 additional hours at room temperature. The precipitated proteins were collected by centrifugation and the supernatant discarded. After the precipitate was dissolved in borate-buffered saline (BBS; pH 7.8) to the original volume, the precipitation was repeated twice as above but the stirring shortened to 1 hour. The final precipitate was dissolved in a small volume of BBS and the resulting solution dialyzed (Visking cellulose tubing) against several changes of BBS at 4~ The dialysis was interrupted when the solution showed no more sulphate ions in the barium test. The final globulin fraction was centrifuged for 20 minutes at 5000 g in an MSE High-Speed 18 centrifuge set at 4~ and the insoluble material discarded. The volume was adjusted with BBS to a convenient value and the solution was filtered through a 0.22 Ix. Millipore membrane. The protein content of the final solution was determined by Lowry's method using bovine serum albumin as standard and adjusted to 4 mg protein/ml. The fraction was stored frozen at - 20~ NRSG. Globulin fractions from normal rabbit serum were prepared as above.
(e) Collection of the inflammatory exudates in sponges Polyester sponges measuring 4 x 1.5 x 0.5 cm were thoroughly washed in tap water and in distilled water before sterilization by autoclaving at l l 0 ~ for 15 minutes. Immediately before use the sponges were soaked in 1.5 ml of saline, NRS, APS, NRSG or APSG and implanted in one side of the abdominal mid-line under ether anaesthesia. The skin flaps were closed with suture clips and the animals were left to recover before being returned to a cage. Exudates were removed from the sponges by gentle washing and squeezing in 7 ml of heparinized saline in 10 ml graduated tubes, the final volume being measured to 0.1 ml. A sample (2 ml) of the diluted exudate was removed and vortexed with 0.4 lnl 6 M trichloracetic acid. The precipitated proteins were removed by centrifugation and the clear supernatant transferred to small specimen vials and stored frozen at - 2 0 ~ until assayed for histamine and 5-hydroxytryptamine. Distilled water was added to the remaining fluid to make a volume of 10 ml and the pH adjusted to 3 with concentrated formic acid. Ethanol (5 ml) and chloroform (25 ml) were added and the tubes shaken for 15 minutes to extract the acidic lipids. The chloroform was separated by centrifugation at 1000g for 15 minutes, filtered through Whatman 1 paper into a 100 ml pear shaped flask and
536
Platelets, Arthus-Type Reactions and Inflammatory Mediators
then evaporated at 35 ~ in a Buchi Rotavapor. The last traces of formic acid were removed from the residue by the evaporation of 2 ml ethanol and the flask was flushed with N2 for 5 minutes. The residue was taken up in Krebs' bicarbonate solution for bioassay. (f) Prostaglandin (PG) bioassay PG-like activity was measured on rat stomach strips superfused with Krebs' bicarbonate solution containing a mixture of antagonists in order to increase the specificity of the assay [4]. PGEe was used as standard in a bracketing assay mode, the activity in the exudate being expressed in PGE2-equivalents per ml.
(g) Fluorometry of histamine and 5-hydroxytryptamine Commercial o-phtaldialdehyde (OPT) was recrystallized from n-heptane after boiling for 1 hour under reflux with decolourizing charcoal. Traces of solvent were removed from the pale yellowish needles kept for a few hours in a vacuum dessicator over paraffin turnings. The dried OPT was stored in an amber bottle at 4~ All glassware used was routinely soaked for a few hours in a mixture (t:1, v/v) of concentrated HCI and HNO3. After thoroughly washing in tap and distilled water the glassware was dried at room temperature. Fluorescence measurements were made in 10 m m quartz cells using a Hitachi-Perkin Elmer (model MPF2A) spectrophotofluorometer. Histamine. Histamine was determined after condensation with OPT at pH 12.5 and low temperature[5-7]. The method was briefly as follows: the acidic extracts were shaken with diethyl ether (saturated with water) in order to remove the trichloracetic acid. Adequate volumes of the aqueous phase (5 to 100 ~tl) were then mixed in glass stoppered tubes with 2 ml of 0.05 N N a O H and 0.4 ml of a freshly prepared solution of OPT in methanol (0.05~ w/v). All tubes were cooled together in an ice-salt bath at - 15 ~ to - 18 ~ for 4-5 hours. The fluorescence was developed by the addition of 0.025 ml of 0.5 N HzSO4 (final pH 2) and measurements were made at 355/445 n m (uncorrected wavelength settings). Blanks for the reagent and the extracts were included in each series analysed. Working standards were prepared immediately before use by dilution of a stock solution of histamine di-hydmchloride (1 mg/ml as the base), stored frozen at - 20~ This solution was stable at least for 4 weeks. 5-Hydroxytryptamine. 5-HT was measured by the formation of a fluorophore with OPT in strong acid [8]. The reagent was always recently prepared by adding a solution of 50 mg of recrystallized OPT in 10 ml of absolute ethanol to 100 ml of 8 N HC1 [9]. Convenient aliquots of the trichloracetic extracts (25-250 ~zl) were directly added to 3 ml of the reagent in glass stoppered tubes. All the tubes were heated together in a boiling water bath for 10 minutes and then cooled in chipped ice. The mixtures were vigorously shaken for 1 minute with 2 ml of chloroform and the phases separated by centrifugation. The fluorescence was measured in the aqueous phase at 355/465 n m (uncorrected wavelength settings).
A blank for the reagent was included in each series processed. Working standards were prepared immediately before use by dilution of a stock solution of 5-HTcreatinesulphate (1 mg/ml as the base), stored frozen at -20~ This solution was stable for at least 2 weeks.
(h) Chromatography of the extracts In order to identify the prostaglandins present in the sponge exudates, the extracts were chromatographed in small silicic acid columns (3-4 g). The residues left after evaporation of the chloroform were taken into Mixture I ll0] and applied to the columns. Increased amounts of ethyl acetate in toluene were used to develop the chromatograms. The resultant fractions were evaporated to dryness in a Buchi Rotavapor at 35 ~ and the residues taken up in Krebs' bicarbonate solution for the assay on rat stomach strips. The prostaglandins in sponge exudates behaved essentially as PGE~ (80~oo) and PGF2~ (20%).
Drugs and chemicals The following drugs and compounds were used: Compound 48/80 (N-methyl homoanisylamine formaldehyde condensation product) from Wellcome Reagents Ltd., Histamine di-hydrochloride from Merck, Sharp and Dohme, 5-Hydroxytryptamine creatinesulphate from Sigma Chemical Co., Heparin from Evans Medical Ltd., U.K., Silicic acid (Unisil, 100-200 mesh) from Clarkson Chemical Co. Inc., Williamsport, Pc, USA, Freund's adjuvant from DIFCO. The dyestuffs were from Gurr Ltd. and other chemicals were commercially available analytical grade reagents. Statistics All comparisons between means were based on Student's t-test. The leveI of significance taken was 5%. Values given in the figures are Mean d: S.E.M. Results
Content of inflammatory mediators in 24-hour sponge exudates Prostaglandins, 5-hydroxytryptamine and histamine were detected in polyester sponges implanted subcutaneously in normothromboc y t e m i c r a t s ( F i g . 1). T h e r e w a s n o s i g n i f i c a n t difference in the content of the mediators studied in exudates induced either by NRSG- or APSGsponges, although the content of prostaglandins and histamine was significantly higher than that in saline-sponges. In thrombocytopenic rats ( F i g . 2) t h e c o n t e n t o f p r o s t a g l a n d i n s a n d 5hydroxytryptamine in APSG-sponge exudates was significantly higher than that in NRSGsponge exudates. Thus, the anti-rat platelets immunoglobulins caused a greater release of prostaglandins and 5-hydroxytryptamine but not of histamine. The APSG was a much stronger releaser of prostaglandins and 5-hydroxytryp-
Platelets, Arthus-Type Reactions and Inflammatory Mediators tamine in thrombocytopenic rats (compare Fig. 1 and 2). In order to evaluate the contribution of mast cells to the release of mediators by APSG, a group of thrombocytopenic rats was treated with compound 48/80 before the implantation of APSG-sponges. Figure 3 shows that there was no significant difference between the content of prostaglandins and 5-hydroxytryptamine in exudates from both groups of animals. Notwithstanding, the treatment significantly reduced the histamine content of the exudates. In a l l 48/80 treated rats, no granulated mast cells were seen in the mesentery preparations made at the moment the sponges were removed.
Inflammatory reaction produced by NRSG- and APSG-sponges Figure 4 shows a striking difference in the necro-haemorrhagic lesions induced by the
537
flammatory reaction. The A P S G produced a milder reaction in normothrombocytemic rats. However, the addition of rat platelets (1.4 • 109 cells) to APS-sponges produced in n o r m o t h r o m bocytemic animals a lesion equivalent to that observed when APSG-sponges alone were implanted in thrombocytopenic animals. The addition of platelets to NRS-sponges caused a mild lesion similar to that induced by NRSG-sponges. Thus, A g - A b complexes formed in sponges induced in normothrombocytemic rats a lesion similar to that observed when APSG-sponges were implanted in thrombocytopenic rats. Discussion
The results presented here show that A P S G was able to induce in normothrombocytemic rats an increased release of prostaglandin-like material as well as a less pronounced inflammatory lesion when compared with the effects observed in thrombocytemic animals. These results strong(BLOOD
250 (n=9)
BLOOD PLATELET5
pG E2- Equiv. ng .ml" 1
H ~g .ml"i
1
(n=iO) 2OO
800.
4
,oo
[
"[
I000-
(n=iO)
AOO-
(n:lO)
5i (n=10)
[
%
800
600-
/~00 100
= < i03.~I -I) H Bg. ml" I
[ T
I S A L I N E = 8,82 .+0.35 (xl0S.~tl-I) N R S G = 8.78 -+0.~5 tAPSG = 6.05 .+0.93
8-HT ng .ml"I
PLATELET5 S-HT ng. ml" i
P G E 2" Equiv. ng .ml -I
(n~lO)
l
3-
2-
-F
200 -
i-
I I
S NRS AP$ -G -G
S NRS A P S -G -G
S N R S A"PS -G -G
01
9
NRS APS -G -G
'
NRS APS -G -G
O
NRS APS -G -G
Figure 1
Figure 2
Content of prostaglandins (PGE2-equiv.), 5-hydroxytryptamine (5-HT), and histamine (H) in 24-hour sponge exudates induced by saline (S), normal rabbit serum globulins (NRSG), and anti-rat platelets serum globulins (APSG) in normothrombocytemic rats. The bars represent the means of n exudates • S.E.M.
Content of prostaglandins (PGE2-equiv.), 5-hydroxytryptamine (5-HT), and histamine (H) in 24-hour sponge exudates induced by normal rabbit serum globulins (NRSG), and anti-rat platelets serum globulins (APSG) in thrombocytopenic rats. The bars represent the means of n exudates ~z S.E.M.
538
Platelets, Arthus-Type Reactions and Inflammatory Mediators
APSG when compared with NRSG, in thrombocytopenic animals. This result indicates the participation of an immunoglobulin in the inly suggest that platelet integrity is not necessary for either the release of mediators or the production of the inflammatory lesions. Intravenous administration of APS and APSG in rats caused a very fast (ca. 1 minute) and prolonged (4-5 days) disappearance of circulating platelets [11]. Furthermore, when rat platelets were mixed at 37~ in vitro with specific APS and reconstituted lyophilized guinea-pig serum as a source of complement, they were rapidly agglutinated before lysis occurred, lysis which was nearly completed within a couple of hours [11], On the other hand, it is known that the half-life of rabbit antiplatelet antibodies is very short in the circulation of animals of other species [12]. One would then expect that in the animals of our experiments a
In conclusion, the lesions observed in the late phase of the inflammation induced in rats by antiplatelet serum is not consequent to the in the APSG preparation are certainly heterogeneous and possibly directed to a multitude of antigenic platelet components. Thus, Ag-Ab complexes other than those involved in platelets' lysis may possibly be circulating and eventually reach the inflammed sites. This might explain the mild but definite inflammation observed in thrombocytopenic animals implanted with NRSG-sponges. The presence of an excess of antibodies in APSG-sponges, reacting with platelet antigens, or Ag-Ab soluble complexes reaching the interstitial spaces, may cause their insolubilization and triggering an Arthus-type hypersensitivity reaction (reverse). This suggestion is supported by the fact that necro-haemorrhagic lesions were induced by the addition of platelets to APSG-sponges implanted in normothrombocytemic rats and that such lesions were similar to that observed when APSG-sponges (BLOOD PLATELETS -
