Brain Research, 581 (1992) 273-282 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/92/$05.00
273
BRES 17771
Decreased neuropeptide-converting enzyme activities in cerebrospinal fluid during acute but not chronic phases of collagen induced arthritis in rats Stefan Persson a, Claes Post a'c, R i k a r d H o l m d a h l b and Fred N y b e r g a Departments of apharmacology, and bMedical and Physiological Chemistry, University of Uppsala, Uppsala (Sweden) and CDepartment of Pharmacology, Draco AB, Lund (Sweden) (Accepted 7 January 1992)
Key words: Rat; Pain; Collagen II-induced arthritis; Dynorphin converting enzyme; Substance P endopeptidase; Dynorphin; Substance P
We investigated the effects of collagen II-induced arthritis on two cerebrospinal fluid (CSF) enzymes converting dynorphin A and substance P (SP), namely dynorphin-converting enzyme (DCE) and substance P endopeptidase (SPE). The products generated by these enzymes are the bioactive fragments Leu-enkephalin-Arg 6 and substance P1-7, respectively. The strain used (DA rats) is very sensitive towards induction of arthritis. The collagen arthritis is a chronic autoimmune arthritis induced by native rat collagen type II (CII). Following intradermal injection of CII into the tailbase, CSF was sampled on day 21 (acute arthritis) and day 38 (chronic arthritis). Control rats were untreated because the strain used developed an acute and self-limited arthritis (adjuvant arthritis) when administered vehicle (i.e. incomplete Freund's adjuvant). The DCE activity was significantly lowered in the acute phase of arthritis (P < 0.05) when analysed with two-factor analysis of variance (ANOVA). The enzyme converting SP (SPE) also showed a significant decrease in the acute phase of arthritis (P < 0.05). These results demonstrate that both DCE and SPE are affected in the acute phase of arthritis. A functional role of these enzymes in processing pain-related neuropeptides is therefore implicated. INTRODUCTION Peptides participating in neurotransmission or neuromodulation are probably inactivated through enzymatic hydrolysis. This assumption is based on the observation that neuropeptides are rapidly degraded by blood, extracellular fluid and tissue enzymes. To date, there is no convincing evidence that peptidergic signals should be inactivated through energy-dependent uptake processes. This is an important distinction between peptidergic and classical neurotransmission in addition to the differences in biosynthesis. Whereas it is possible to pharmacologically manipulate the classical neurotransmission by preventing reuptake processes (e.g. inactivation mechanism for catecholamines) or by degradation through enzymatic hydrolysis, as occurs in the inactivation of acetylcholine by acetylcholinesterase, it seems that the only way to modulate peptidergic neurotransmission is to prevent extraneuronal metabolism. Owing to their putative function in pain transmission or modulation dynorphins and substance P (SP) (for structure, see Table I) have received much attention, and more recently interest has been focused on the enzymatic degradation and conversion mechanisms of these
peptides. Both of these peptide systems are present in fibers with high innervation in the dorsal horn of the spinal cord. In conformity with many other peptides exhibiting neural or hormonal activity, both the dynorphins and SP are derived from large precursors (preprohormones) by sequential proteolytic degradation. The preprohormone is generally biologically inactive, whereas the primary product released by so-called processing enzymes may possess biological activity. More commonly, however, further enzymatic conversion is necessary to generate the mature peptide, which is the case for dynorphin A and SP as well. Furthermore, converting enzymes may modify or change the actions of a particular bioactive peptide, thus resulting in a modulatory effect 7' 39,41.
Prodynorphin-derived peptides are found within the dorsal horn of the spinal cord, with especially dense concentrations in laminae I - I I and V 35. Several different pain models have indicated that spinal dynorphin may be involved in the modulation, or perhaps transmission, of pain-related signals. In a unilateral experimental inflammation model, a large increase in the spinal concentration of dynorphin AI_ 8 and dynorphin A has been observed 24'32'37'47. A n increase in the level of m R N A
Correspondence: S. Persson, Department of Pharmacology, University of Uppsala, P.O. Box 591, S-751 24 Uppsala, Sweden.
274 encoding the p r o d y n o r p h i n precursor precedes the corresponding increase in dynorphin 8'24'25'47. This increase occurs unilaterally and is confined to spinal segments receiving afferents innervating the inflamed hindpaw. A similar elevation of dynorphin A levels in dorsal spinal cord in polyarthritic rats has been described. H o w e v e r , the increases in m R N A and peptide levels are not restricted to a particular spinal segment due to the unspecific action of the inflammation 32'33. F u r t h e r m o r e , the pharmacology of dynorphin has revealed some discrepancies. A l t h o u g h agonists for the opioid ~c-receptor (a putative r e c e p t o r for the dynorphins) produce analgesia, it has been difficult to demonstrate spinal effects of the dynorphins because intrathecal administration leads to paralysis and cell death 3'1°'46. H o w e v e r , the fact that des-Tyrl-dynorphin (which lacks opioid activity) also induces hindlimb paralysis indicates that non-opioid sites may be involved in this action of the parent peptide 46. The u n d e c a p e p t i d e SP is considered to be a putative transmitter or m e d i a t o r of nociception in the spinal cord 16'53. It is localized in small, lightly myelinated or unmyelinated A 6 and C fibers which terminate in the superficial laminae (I-II) of the dorsal horn 22. A variety of noxious stimuli of these small afferent fibers results in the release of SP into the dorsal horn 2'9. In both behavioural and electrophysiological studies, intrathecal injections of SP have given effects consistent with nociception 16'3°. H o w e v e r , little is known about the involvement of SP neurons in chronic nociception. A n increase in the spinal SP concentrations 27, and, m o r e recently, an enhanced release of SP 43 and the gene expression of preprotachykinin A (a precursor for SP) have been r e p o r t e d in adjuvant-induced arthritis 36. The elevated levels of these pain-modulating or transmitter peptides as a reaction to the pain that follows noxious stimulation and tissue injury m a y reflect regulation on several different levels between transcription and the final peptide inactivation process. A s m e n t i o n e d above, it is evident that the preprotachykinin A and prodynorphin precursor transcription is increased in different models of inflammation, which in part could be responsible for the elevated SP and dynorphin concentrations in the dorsal spinal cord. A n o t h e r possible contributing factor is regulation in enzymatic activities from the initial processing enzymes to the final converting or inactivating enzymes. In this study we have focused our attention towards two converting enzymes n a m e d dynorphin-converting enzyme ( D C E ) and substance P e n d o p e p t i d a s e (SPE). D C E transform the m e m b e r s of the endogenous opioid peptide family p r o d y n o r p h i n , also called p r o e n k e p h a l i n B, (dynorphin A , dynorphin B and a - n e o e n d o r p h i n ) into L e u - e n k e p h a l i n - A r g 6 (see Table I) 41. This conver-
sion changes the r e c e p t o r profile from ~c-selectivity (dynorphins) to 6-selectivity (enkephalins). SPE has been named for its capacity to cleave SP at the Phe7-Phe s and, to a lesser extent, at the Phe8-Gly 9 b o n d (see Table I) 39. Recent studies concerning these two CSF enzymes has shown an increased D C E - and SPE-activity in opiatetolerant rats 44, whereas a reduced DCE-activity has been r e p o r t e d in CSF from women at term pregnancy 29. These findings and the nature of the substrates suggest that these proteases may play a role in nociception. Therefore, we have investigated whether the SPE and D C E activities are affected in the cerebrospinal fluid (CSF) during acute and chronic phases of arthritis in D A rats. The strain used is very sensitive towards induction of arthritis and immunization with autologous type II collagen, the major c o m p o n e n t of cartilage, emulsified in mineral oil induces a chronic arthritic disease 18'26. This model gives us the possibility to investigate the influence of dynorphin and SP regulatory enzymes during both acute and chronic phases of arthritis. MATERIALS AND METHODS Chemicals
Thiorphan and all peptides used in this study were purchased from Bachem Feinchemikalien AG (Bubendorf, Switzerland) except for substance PI-7 (SP1-7) and Tyr-SPI_7 which was prepared by Dr. G. Lindeberg, Dept. of Immunology, University of Uppsala, Sweden and Dr. J.M. Stewart, Dept. of Biochemistry, University of Colorado, Denver, CO, USA, respectively. The protease inhibitors phosphoramidon and amastatin were purchased from Sigma (St. Louis, MO, USA), whereas captopril and guanidinoethylmercaptosuccinic acid (GEMSA) were supplied by Squibb (Princeton, NJ, USA) and Calbiochem Corp. (La Jolla, CA, USA), respectively. All other chemicals and solvents were of analyticalreagent grade from usual sources. SPI_7 and Leu-enkephalin-Arg6 (Leu-enk-Arg6) were both iodinated by the chloramine-T method. 1.8 nmol peptide (Tyr-SPI_7 or Leu-enk-Arg6) in 20 pl water, 20 pl sodium phosphate buffer (0.2 M), 10 pl chloramine-T solution (0.2 mg/ml) and 0.5 mCi NalZSI were allowed to react for 60 s before 100 #1 15% acetonitrile/water in 0.04% trifluoroacetic acid (TFA) was added to terminate the reaction. The iodinated peptide was purified by HPLC using a reversed-phase Silica Gel C-18 column (Ultropac TSK ODS-120T, 4.6 x 250 mm, particle size 5 pm). Elution was carried out with a linear gradient of acetonitrile (15-50%) containing 0.04% TFA and the monoiodinated peptide was isolated. Animal experiments
Dark Agouti (DA) rats, originally obtained from Harlan Olac Ltd (Oxon, UK), were bred and kept at the Biomedical Center in Uppsala. All experiments were performed on female rats at an age of 10-12 weeks. The guidelines proposed by the ethical committee of the IASP5 for investigating experimental pain in animals have been followed. The duration of the experiment was as short as possible and the number of animals involved was kept to a minimum. Large cages were used for animal housing and the floors were covered with sawdust. There were never more than 3 rats per cage, thus minimizing the possibility of painful contact between rats45. For induction of collagen-induced arthritis, an emulsion of 150 #g native rat collagen type II (CII) 19 with 150/~1 incomplete Freund's adjuvant (IFA) (Difco, Detroit, USA) was injected intradermally
275 TABLE I
Structure of dynorphin A and substance P and their bioactive fragments generated by dynorphin-converting enzyme (DCE) and substance P endopeptidase (SPE) Dyn A, dynorphin A; enk, enkephalin; SP, substance P.
Peptide
Structure
Receptor selectivity
Dyn A
Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile-Arg-Pro- x Lys-Leu-Lys-Trp-Asp-Asn-Gln
Determination of dynorphin-converting enzyme activity
Leu-enkArg 6 Tyr-Gly-Gly-Phe-Leu-Arg SP
Arg-Pro-Lys-Pro-Gln-Gln-PhePhe-Gly-Leu-Met-NH 2
SPI_7
Arg-Pro-Lys-Pro-Gln-Gln-Phe
the addition of the hydrogen peroxide solution the absorbance was measured in a spectrophotometer at 600 nm. Rat hemoglobin dissolved and diluted in artificial CSF (Na + 149.9 mM; K + 2.6 mM; Ca 2+ 1.3 mM; Mg 2+ 0.9 mM; C1- 134.6 mM; HCO 3- 20.2 mM; HPO42- 1.0 mM, adjusted to pH 7.4) was used as standard. The detection limit of the hemoglobin assay was determined as the blank signal plus three standard deviations of the blank, e.g., the calculated intercept plus the standard deviation of the residuals. These calculations were based on the unweighted least-squares method. The limit of detection was estimated at 0.13/~g//~l. All CSF samples which contained measurable amounts of hemoglobin were excluded from this study.
6 NK1
at the base of the tail. Control rats were untreated since an acute self-limited arthritis is induced in D A rats by IFA (Holmdahl et al., in preparation). Arthritis development was followed by a macroscopic scoring system for the four paws, ranging from 0 to 3 (1 = swelling and/or redness of one toe or finger joint, 2 = two or more joints involved and 3 = severe arthritis in the entire paw).
CSF sampling On day 21 after the immunization, i.e, in the acute phase of arthritis, and two and a half weeks later (day 38) in the chronic phase, the animals were anesthetized with 3.1 ml/kg of an aqueous solution containing chloral hydrate, 42.5 g/l; sodium pentobarbital, 9.7 g/l; magnesium chloride, 21 g/l; propylene glycol, 428 g/l and ethanol 90 g/1. The atlanto-occipitat membrane was exposed and a 30 gauge needle was inserted through the midline of the membrane. By gently aspirating with a syringe connected to a polyethylene tubing, approximately 100-125/~1 CSF was collected in the tubing and transferred to an Eppendorf tube. Prior to analysis the collected samples were kept frozen at -70°C. In order to exclude CSF samples which contained plasma proteases, due to blood contamination, the samples were examined visually and by a biochemical method 2s'49. Briefly, 10/ll CSF was incubated in glass tubes at room temperature together with 2 ml 3, 3', 5, 5'-tetramethylbenzidine (5 mg/ml) in 90% (v/v) glacial acetic acid and 2 ml 0.3% (v/v) hydrogen peroxide. Exactly 10 min after
TABLE II
Cross-reactivity of the substance PI-z antiserum (46 C) with various substance P (SP) fragments and other related peptides SP, substance P.
Peptide
Cross-reactivity (%)
SP SP5_11 sP2_ 7 SP3_7 SPl_ s SPl_ 9 Neurokinin A Neurokinin B
2 8 1 0.2 7 0.2 13 12
The enzyme activity was monitored by measuring the formation of Leu-enk-Arg 6 from synthetic dynorphin A, using a radioimmunoassay (RIA) specific for the product. To prevent further degradation of the released fragment by aminopeptidases, angiotensinconverting enzyme (ACE) and carboxypeptidases, the protease inhibitors amastatin, captopril and GEMSA, respectively, were added to the incubation mixture. These enzymes are known to be present in the cerebrospinal fluid (CSF) 15"17. Incubations were performed in duplicates in Eppendorf tubes at 37°C with 0.1 ~g (= 36 pmol) of dynorphin A and 20/A CSF together with the inhibitors amastatin (20 ~M), GEMSA (1 mM) and captopril (20/tM) in a final volume of 40/~1 buffered at pH 7.4 with 50 mM Tris-HC1. Prior to the addition of substrate, the CSF was preincubated for 15 min at 37°C together with the inhibitors. The incubation proceeded for 2 h and the reaction was terminated by the addition of 0.5 ml ice-cold methanol, followed by centrifugation and evaporation of the supernatant in a Speed-Vac centrifuge (Savant, Hicksville, NY, USA).
Determination of substance P endopeptidase activity The conversion of synthetic substance P to its N-terminal fragment SPI_ 7 was followed by a RIA specific for the heptapeptide product. Ten/~1 CSF was preincubated for 15 min at 37°C together with 20/tM amastatin. Following the addition of 0.1/~g (74 pmol) of substance P the incubation was performed in duplicates in Eppendorf tubes at 37°C for 1 h in a final volume of 40/A buffered at pH 7.4 with 50 mM Tris-HCl. After completed incubation, samples were treated in accordance with those in the DCE assay described above.
Generation of antisera All antisera were generated in rabbits. Leu-enk-Arg 6 (1 mg) and thyroglobulin (5 mg) were dissolved in 300/~l 100 mM phosphate buffer, pH 7.4, and cooled to 0°C. Glutaraldehyde in 25% aqueous solution was diluted with ice-cold distilled water (1:100, v/v) and 180 pl was added dropwise. The reaction mixture was stirred for 30 min at 0°C followed by 24 h at room temperature. After extensive dialysis against 0.9% NaCl, the peptide-thyroglobulin conjugates were ready for injection into rabbits. A 500/d aliquot of the conjugate, corresponding to 100 pg of peptide, was emulsified with an equal volume of Freund's complete adjuvant. The animals were injected intracutaneously on their backs at multiple sites. Booster doses in volumes of 250 pl corresponding to 50 pg of peptide were emulsified with an equal volume of Freund's incomplete adjuvant and injected subcutaneously at intervals of 3-4 weeks. The crossreactivity of the enkephalyl hexapeptide antiserum (Benj G) with Met-enkephalin-Arg6 was 10% and with Met-enkephalin-Arg6 sulfoxide and Met-enkephalin-Arg6-Phe7, around 1%. With all other proenkephalin and prodynorphin-derived opioid peptides the crossreactivity was less than i%. The final dilution was 1:2800. Antiserum to SPI_ 7 (46 C) was raised in an analogous fashion. However, in this case dicyclohexylcarbodiimide was used as a coupling agent 5°. In the latter study, tritiated 2,4-dehydroproline-SPl_ 7 was used as radiolabel, but we here used the iodinated Tyr-SPl_7, which shows a different cross-reactivity pattern (Table II). The final dilution used under assay conditions was 1:450,000.
276
Radioimmunoassay (RIA) methods
Statistical analysis
The dried eluates were dissolved and diluted in a 1:1 MeOH/0.1 M HC1 mixture and 25 kd aliquots of samples and standards were added in triplicate to incubation vials together with i00/~1 of antisera and iodinated peptide (4,500 cpm), respectively. The antisera and labelled peptide were diluted in assay buffer containing 0.1% gelatin, 0.1% bovine serum albumin, 0.8% NaCI, 0.93% EDTA in 50 mM sodium phosphate (pH 7.4). The vials were then incubated at 4°C for 24 h. The bound and free peptides were separated by adding 200/al active charcoal. For the Leu-enk-Arg 6 assay, a charcoal suspension containing 500 mg active charcoal and 50 mg dextran T 70 dissolved in 200 ml 50 mM sodium phosphate (pH 7.4) was used, and in the SPI_ 7 assay the ratio charcoal/dextran T 70 was 750 mg/75 mg in the above buffer. After 10 min of incubation, the mixture was centrifuged for 1 min in a Beckman Microfuge8. A 300/~1 aliquot of the supernatant was removed and counted in a y-counter. The intraassay coefficient of variation for one RIA was calculated from eight quality controls (triplicates) evenly distributed within the RIA and in the same concentration interval as the sampies. The coefficient was found to be 5.6% and 10.8% for the SPI_ 7 and Leu-enk-Arg6 RIA, respectively. The interassay coefficient of variation was calculated from the ICs0 values of the standard curves, and was 12.3% (n = 8) and 13.1% (n = 10), respectively. The minimum detectable amount of the RIA was 0.3 and 14 fmol per tube, according to Signorella et a l : s, and 50% tracer displacement was attained at around 7 and 350 fmol per tube, respectively.
Statistical analyses of the recorded enzymatic activities were performed on means of duplicate determinations using two-factor analysis of variance (ANOVA) with factor interaction. All calculations were carried out on log-transformed data, since, when tested for normality, data were found to be log-normal distributed.
RESULTS
Accuracy o f enzyme measurements In o r d e r to investigate the c o n d i t i o n s necessary for p r e v e n t i n g f u r t h e r d e g r a d a t i o n of the e n z y m e p r o d u c t s ( L e u - e n k - A r g 6 and SPI_7) the effect o f d i f f e r e n t inhibitors on the a m o u n t of p r o d u c t g e n e r a t e d by D C E and S P E versus t i m e was investigated. Possible changes in o t h e r C S F p r o t e a s e s and the lack of suitable inhibitors against these activities m a y give a l t e r e d rates of p r o d u c t f o r m a t i o n , leading to m i s i n t e r p r e t a t i o n s c o n c e r n i n g the e n z y m e s studied. Fig. 1 illustrates the t i m e c o u r s e of L e u - e n k - A r g 6 form a t i o n in the p r e s e n c e of v a r i o u s p r o t e a s e inhibitors. If no i n h i b i t o r is a d d e d , only a small increase in p r o d u c t
Reversed-phase chromatography on SMART system
f o r m a t i o n is o b s e r v e d during the first 120 min. T h e r e a f -
CSF samples (100/d) from control and arthritic rats were analyzed by the SMART system (Pharmacia LKB Biotechnology, Uppsala, Sweden). The computer controlled micropreparative chromatography system was equipped with built-in detector cells for UV (UV-MII with 214 nm optics) and conductivity measurements. The conductivity scale was set by calibration with eluent A (100%) and eluent B (0%); A, 0.04% trifluoroacetic acid (TFA); B, 0.034% TFA in 60% acetonitrile. The column used was a/aRPC C2lC18, PC 3.2/3 (particle size 3/~m, 120/~; 30 x 3.2 mm I.D.). Elution was performed with a gradient of acetonitrile (0-60%) containing TFA at a flow-rate of 240/A/min.
ter, the c u r v e declines. H o w e v e r , w h e n the a m i n o p e p t i dase i n h i b i t o r a m a s t a t i n and the e n k e p h a l i n c o n v e r t a s e inhibitor GEMSA
are p r e s e n t , a m o r e p l a n a r shape is
s e e n after 120 m i n of incubation. A m u c h m o r e pron o u n c e d effect is o b s e r v e d w h e n the A C E i n h i b i t o r captopril
is a d d e d
GEMSA.
in
combination
with
amastatin
and
Inhibitors of the n e u t r a l e n d o p e p t i d a s e ( N E P
or E C 3.4.24.11), such as p h o s p h o r a m i d o n
and thior-
p h a n (each a d d e d in 2 0 / ~ M ) , had n o significant effect (data not shown).
8 ----o--- Control -----or-- Amastatin+GEMSA 70 A m ~
V [ /
1.6
/
1.4
Consequently,
--..o-- Control
amastatin,
GEMSA
T
Captolmil+Am~tatin J
1.2 ~ 1.0 2
0.8
~ 0.6
0.4
1 00
0.2 30
60
90
120
150
180
Time (min)
Fig. 1. Time-course of dynorphin A conversion to Leu-enk-Arg 6 in rat CSF in the absence and presence of both amastatin (20/zM), GEMSA (1 mM) and combined with captopril (20 ktM). 15/~1 CSF was preincubated for 15 min at 37°C with the different inhibitors used. After the addition of 0.1 /~g dynorphin A incubation was performed in a final volume of 40 kd, buffered at pH 7.4 by 50 mM Tris-HCl. The reaction was terminated by the addition of 0.5 ml ice-cold methanol and samples were evaporated before radioimmunoassay. Vertical lines show standard error (S.E.), n = 3.
30
60
90 Time (rain)
120
150
180
Fig. 2. Time-course of SP conversion to SPI_7 in rat CSF in the absence and presence of captopril (20/~M) alone or combined with amastatin (20 ktM). 10/~1 CSF (diluted × 3) was preincubated for 15 min at 37°C with the different inhibitors used. After the addition of 0.1 ktg SP incubation was performed in a final volume of 40 :d, buffered at pH 7.4 by 50 mM Tris-HC1. For further details see legend to Fig. 1 and the text. Vertical lines show standard error (S.E.), n = 3.
277 12-
150.
10-
125.
8.
E * 100
day 21 p.i.
~
4
day 38 p.i.
I~
,
~ 75 ff 50
14.
25
20 10
. . --- , .... 15
, .... 20
, .... 25
, .... 30
, .... 35
40
Control
CIA
Control
CIA
n=8
n--7
n=9
n=7
Days after immunization
Fig. 3. T h e development of arthritis after immunization with type II collagen (collagen induced arthritis), C I A , n . = 10). T h e develo p m e n t of arthritis was followed by a macroscopic scoring system for the four paws, ranging from 0 to 3 (1 = swelling and/or redness of one toe or finger joint, 2 = two or more joints involved and 3 = severe arthritis in the entire paw). T h e arrows indicate time points when CSF were collected. Serum was collected a t day 20 after immunization. M e a n + S.E.M. ,
and captopril were used in the DCE assay. Fig. 2 shows the corresponding time course for SPI_ 7 formation and how this is influenced by captopril alone or by captopril combined with amastatin. It is evident that the SPE activity is effectively inhibited by the ACE inhibitor captopril. The proline residue adjacent to the N-terminal arginine in SPI_ 7 (for structure, see Table I) should make the heptapeptide stable against aminopeptidases. To test whether aminopeptidases have an effect upon SPI_ 7 formation, amastatin was used in combination with captopril. Indeed, the addition of amastatin seems to have a slight stabilizing effect on the product formation with time. If no inhibitor is used an almost linear relationship is observed during the first two hours of incubation. Therefore, in the SPE assay, amastatin
6 0 t 4--T'-day21p'i"
~
4
day38p, i.
Fig. 5. Substance P endopeptidase (SPE) activity in CSF from untreated rats (control; open bar) and rats i m m u n i z e d with type II collagen (collagen induced arthritis, CIA, shaded bar). CSF was sampled day 21 and day 38 after the immunization. Enzymatic activity was m e a s u r e d as the rate of formation of substance P1-7 from substance P. M e a n + S.E.M., p.i. = post immunization.
was added to prevent possible influences by aminopeptidases. Our results also indicated that none of the inhibitors discussed above influenced the RIA by itself. Levels of DCE and SPE in CSF from arthritic rats D A rats with collagen-induced arthritis developed a severe and chronic arthritis. Macroscopic signs of the arthritis development in the four paws started at day 15. The severity increased fast to day 20 and thereafter a slow increase was observed during the whole experiment, i.e. unti 1 day 38 after immunization (Fig. 3). All rats immunized with CII developed an anti-CII antibody response (185 + 147/zg/ml at day 20 after immunization, n = 10) whereas no antibodies were detected in control rats. CSF samples were collected both during the acute phase (day 21) and during the chronic phase (day 38). DCE and SPE activities in CSF collected from control and rats with collagen-induced arthritis, in both acute and chronic phase, were analysed with a two fac-
I~ ]
50-
500¸
400,450'
. .-"'"
350 ~ 300
30 .
273
BRES 17771
Decreased neuropeptide-converting enzyme activities in cerebrospinal fluid during acute but not chronic phases of collagen induced arthritis in rats Stefan Persson a, Claes Post a'c, R i k a r d H o l m d a h l b and Fred N y b e r g a Departments of apharmacology, and bMedical and Physiological Chemistry, University of Uppsala, Uppsala (Sweden) and CDepartment of Pharmacology, Draco AB, Lund (Sweden) (Accepted 7 January 1992)
Key words: Rat; Pain; Collagen II-induced arthritis; Dynorphin converting enzyme; Substance P endopeptidase; Dynorphin; Substance P
We investigated the effects of collagen II-induced arthritis on two cerebrospinal fluid (CSF) enzymes converting dynorphin A and substance P (SP), namely dynorphin-converting enzyme (DCE) and substance P endopeptidase (SPE). The products generated by these enzymes are the bioactive fragments Leu-enkephalin-Arg 6 and substance P1-7, respectively. The strain used (DA rats) is very sensitive towards induction of arthritis. The collagen arthritis is a chronic autoimmune arthritis induced by native rat collagen type II (CII). Following intradermal injection of CII into the tailbase, CSF was sampled on day 21 (acute arthritis) and day 38 (chronic arthritis). Control rats were untreated because the strain used developed an acute and self-limited arthritis (adjuvant arthritis) when administered vehicle (i.e. incomplete Freund's adjuvant). The DCE activity was significantly lowered in the acute phase of arthritis (P < 0.05) when analysed with two-factor analysis of variance (ANOVA). The enzyme converting SP (SPE) also showed a significant decrease in the acute phase of arthritis (P < 0.05). These results demonstrate that both DCE and SPE are affected in the acute phase of arthritis. A functional role of these enzymes in processing pain-related neuropeptides is therefore implicated. INTRODUCTION Peptides participating in neurotransmission or neuromodulation are probably inactivated through enzymatic hydrolysis. This assumption is based on the observation that neuropeptides are rapidly degraded by blood, extracellular fluid and tissue enzymes. To date, there is no convincing evidence that peptidergic signals should be inactivated through energy-dependent uptake processes. This is an important distinction between peptidergic and classical neurotransmission in addition to the differences in biosynthesis. Whereas it is possible to pharmacologically manipulate the classical neurotransmission by preventing reuptake processes (e.g. inactivation mechanism for catecholamines) or by degradation through enzymatic hydrolysis, as occurs in the inactivation of acetylcholine by acetylcholinesterase, it seems that the only way to modulate peptidergic neurotransmission is to prevent extraneuronal metabolism. Owing to their putative function in pain transmission or modulation dynorphins and substance P (SP) (for structure, see Table I) have received much attention, and more recently interest has been focused on the enzymatic degradation and conversion mechanisms of these
peptides. Both of these peptide systems are present in fibers with high innervation in the dorsal horn of the spinal cord. In conformity with many other peptides exhibiting neural or hormonal activity, both the dynorphins and SP are derived from large precursors (preprohormones) by sequential proteolytic degradation. The preprohormone is generally biologically inactive, whereas the primary product released by so-called processing enzymes may possess biological activity. More commonly, however, further enzymatic conversion is necessary to generate the mature peptide, which is the case for dynorphin A and SP as well. Furthermore, converting enzymes may modify or change the actions of a particular bioactive peptide, thus resulting in a modulatory effect 7' 39,41.
Prodynorphin-derived peptides are found within the dorsal horn of the spinal cord, with especially dense concentrations in laminae I - I I and V 35. Several different pain models have indicated that spinal dynorphin may be involved in the modulation, or perhaps transmission, of pain-related signals. In a unilateral experimental inflammation model, a large increase in the spinal concentration of dynorphin AI_ 8 and dynorphin A has been observed 24'32'37'47. A n increase in the level of m R N A
Correspondence: S. Persson, Department of Pharmacology, University of Uppsala, P.O. Box 591, S-751 24 Uppsala, Sweden.
274 encoding the p r o d y n o r p h i n precursor precedes the corresponding increase in dynorphin 8'24'25'47. This increase occurs unilaterally and is confined to spinal segments receiving afferents innervating the inflamed hindpaw. A similar elevation of dynorphin A levels in dorsal spinal cord in polyarthritic rats has been described. H o w e v e r , the increases in m R N A and peptide levels are not restricted to a particular spinal segment due to the unspecific action of the inflammation 32'33. F u r t h e r m o r e , the pharmacology of dynorphin has revealed some discrepancies. A l t h o u g h agonists for the opioid ~c-receptor (a putative r e c e p t o r for the dynorphins) produce analgesia, it has been difficult to demonstrate spinal effects of the dynorphins because intrathecal administration leads to paralysis and cell death 3'1°'46. H o w e v e r , the fact that des-Tyrl-dynorphin (which lacks opioid activity) also induces hindlimb paralysis indicates that non-opioid sites may be involved in this action of the parent peptide 46. The u n d e c a p e p t i d e SP is considered to be a putative transmitter or m e d i a t o r of nociception in the spinal cord 16'53. It is localized in small, lightly myelinated or unmyelinated A 6 and C fibers which terminate in the superficial laminae (I-II) of the dorsal horn 22. A variety of noxious stimuli of these small afferent fibers results in the release of SP into the dorsal horn 2'9. In both behavioural and electrophysiological studies, intrathecal injections of SP have given effects consistent with nociception 16'3°. H o w e v e r , little is known about the involvement of SP neurons in chronic nociception. A n increase in the spinal SP concentrations 27, and, m o r e recently, an enhanced release of SP 43 and the gene expression of preprotachykinin A (a precursor for SP) have been r e p o r t e d in adjuvant-induced arthritis 36. The elevated levels of these pain-modulating or transmitter peptides as a reaction to the pain that follows noxious stimulation and tissue injury m a y reflect regulation on several different levels between transcription and the final peptide inactivation process. A s m e n t i o n e d above, it is evident that the preprotachykinin A and prodynorphin precursor transcription is increased in different models of inflammation, which in part could be responsible for the elevated SP and dynorphin concentrations in the dorsal spinal cord. A n o t h e r possible contributing factor is regulation in enzymatic activities from the initial processing enzymes to the final converting or inactivating enzymes. In this study we have focused our attention towards two converting enzymes n a m e d dynorphin-converting enzyme ( D C E ) and substance P e n d o p e p t i d a s e (SPE). D C E transform the m e m b e r s of the endogenous opioid peptide family p r o d y n o r p h i n , also called p r o e n k e p h a l i n B, (dynorphin A , dynorphin B and a - n e o e n d o r p h i n ) into L e u - e n k e p h a l i n - A r g 6 (see Table I) 41. This conver-
sion changes the r e c e p t o r profile from ~c-selectivity (dynorphins) to 6-selectivity (enkephalins). SPE has been named for its capacity to cleave SP at the Phe7-Phe s and, to a lesser extent, at the Phe8-Gly 9 b o n d (see Table I) 39. Recent studies concerning these two CSF enzymes has shown an increased D C E - and SPE-activity in opiatetolerant rats 44, whereas a reduced DCE-activity has been r e p o r t e d in CSF from women at term pregnancy 29. These findings and the nature of the substrates suggest that these proteases may play a role in nociception. Therefore, we have investigated whether the SPE and D C E activities are affected in the cerebrospinal fluid (CSF) during acute and chronic phases of arthritis in D A rats. The strain used is very sensitive towards induction of arthritis and immunization with autologous type II collagen, the major c o m p o n e n t of cartilage, emulsified in mineral oil induces a chronic arthritic disease 18'26. This model gives us the possibility to investigate the influence of dynorphin and SP regulatory enzymes during both acute and chronic phases of arthritis. MATERIALS AND METHODS Chemicals
Thiorphan and all peptides used in this study were purchased from Bachem Feinchemikalien AG (Bubendorf, Switzerland) except for substance PI-7 (SP1-7) and Tyr-SPI_7 which was prepared by Dr. G. Lindeberg, Dept. of Immunology, University of Uppsala, Sweden and Dr. J.M. Stewart, Dept. of Biochemistry, University of Colorado, Denver, CO, USA, respectively. The protease inhibitors phosphoramidon and amastatin were purchased from Sigma (St. Louis, MO, USA), whereas captopril and guanidinoethylmercaptosuccinic acid (GEMSA) were supplied by Squibb (Princeton, NJ, USA) and Calbiochem Corp. (La Jolla, CA, USA), respectively. All other chemicals and solvents were of analyticalreagent grade from usual sources. SPI_7 and Leu-enkephalin-Arg6 (Leu-enk-Arg6) were both iodinated by the chloramine-T method. 1.8 nmol peptide (Tyr-SPI_7 or Leu-enk-Arg6) in 20 pl water, 20 pl sodium phosphate buffer (0.2 M), 10 pl chloramine-T solution (0.2 mg/ml) and 0.5 mCi NalZSI were allowed to react for 60 s before 100 #1 15% acetonitrile/water in 0.04% trifluoroacetic acid (TFA) was added to terminate the reaction. The iodinated peptide was purified by HPLC using a reversed-phase Silica Gel C-18 column (Ultropac TSK ODS-120T, 4.6 x 250 mm, particle size 5 pm). Elution was carried out with a linear gradient of acetonitrile (15-50%) containing 0.04% TFA and the monoiodinated peptide was isolated. Animal experiments
Dark Agouti (DA) rats, originally obtained from Harlan Olac Ltd (Oxon, UK), were bred and kept at the Biomedical Center in Uppsala. All experiments were performed on female rats at an age of 10-12 weeks. The guidelines proposed by the ethical committee of the IASP5 for investigating experimental pain in animals have been followed. The duration of the experiment was as short as possible and the number of animals involved was kept to a minimum. Large cages were used for animal housing and the floors were covered with sawdust. There were never more than 3 rats per cage, thus minimizing the possibility of painful contact between rats45. For induction of collagen-induced arthritis, an emulsion of 150 #g native rat collagen type II (CII) 19 with 150/~1 incomplete Freund's adjuvant (IFA) (Difco, Detroit, USA) was injected intradermally
275 TABLE I
Structure of dynorphin A and substance P and their bioactive fragments generated by dynorphin-converting enzyme (DCE) and substance P endopeptidase (SPE) Dyn A, dynorphin A; enk, enkephalin; SP, substance P.
Peptide
Structure
Receptor selectivity
Dyn A
Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile-Arg-Pro- x Lys-Leu-Lys-Trp-Asp-Asn-Gln
Determination of dynorphin-converting enzyme activity
Leu-enkArg 6 Tyr-Gly-Gly-Phe-Leu-Arg SP
Arg-Pro-Lys-Pro-Gln-Gln-PhePhe-Gly-Leu-Met-NH 2
SPI_7
Arg-Pro-Lys-Pro-Gln-Gln-Phe
the addition of the hydrogen peroxide solution the absorbance was measured in a spectrophotometer at 600 nm. Rat hemoglobin dissolved and diluted in artificial CSF (Na + 149.9 mM; K + 2.6 mM; Ca 2+ 1.3 mM; Mg 2+ 0.9 mM; C1- 134.6 mM; HCO 3- 20.2 mM; HPO42- 1.0 mM, adjusted to pH 7.4) was used as standard. The detection limit of the hemoglobin assay was determined as the blank signal plus three standard deviations of the blank, e.g., the calculated intercept plus the standard deviation of the residuals. These calculations were based on the unweighted least-squares method. The limit of detection was estimated at 0.13/~g//~l. All CSF samples which contained measurable amounts of hemoglobin were excluded from this study.
6 NK1
at the base of the tail. Control rats were untreated since an acute self-limited arthritis is induced in D A rats by IFA (Holmdahl et al., in preparation). Arthritis development was followed by a macroscopic scoring system for the four paws, ranging from 0 to 3 (1 = swelling and/or redness of one toe or finger joint, 2 = two or more joints involved and 3 = severe arthritis in the entire paw).
CSF sampling On day 21 after the immunization, i.e, in the acute phase of arthritis, and two and a half weeks later (day 38) in the chronic phase, the animals were anesthetized with 3.1 ml/kg of an aqueous solution containing chloral hydrate, 42.5 g/l; sodium pentobarbital, 9.7 g/l; magnesium chloride, 21 g/l; propylene glycol, 428 g/l and ethanol 90 g/1. The atlanto-occipitat membrane was exposed and a 30 gauge needle was inserted through the midline of the membrane. By gently aspirating with a syringe connected to a polyethylene tubing, approximately 100-125/~1 CSF was collected in the tubing and transferred to an Eppendorf tube. Prior to analysis the collected samples were kept frozen at -70°C. In order to exclude CSF samples which contained plasma proteases, due to blood contamination, the samples were examined visually and by a biochemical method 2s'49. Briefly, 10/ll CSF was incubated in glass tubes at room temperature together with 2 ml 3, 3', 5, 5'-tetramethylbenzidine (5 mg/ml) in 90% (v/v) glacial acetic acid and 2 ml 0.3% (v/v) hydrogen peroxide. Exactly 10 min after
TABLE II
Cross-reactivity of the substance PI-z antiserum (46 C) with various substance P (SP) fragments and other related peptides SP, substance P.
Peptide
Cross-reactivity (%)
SP SP5_11 sP2_ 7 SP3_7 SPl_ s SPl_ 9 Neurokinin A Neurokinin B
2 8 1 0.2 7 0.2 13 12
The enzyme activity was monitored by measuring the formation of Leu-enk-Arg 6 from synthetic dynorphin A, using a radioimmunoassay (RIA) specific for the product. To prevent further degradation of the released fragment by aminopeptidases, angiotensinconverting enzyme (ACE) and carboxypeptidases, the protease inhibitors amastatin, captopril and GEMSA, respectively, were added to the incubation mixture. These enzymes are known to be present in the cerebrospinal fluid (CSF) 15"17. Incubations were performed in duplicates in Eppendorf tubes at 37°C with 0.1 ~g (= 36 pmol) of dynorphin A and 20/A CSF together with the inhibitors amastatin (20 ~M), GEMSA (1 mM) and captopril (20/tM) in a final volume of 40/~1 buffered at pH 7.4 with 50 mM Tris-HC1. Prior to the addition of substrate, the CSF was preincubated for 15 min at 37°C together with the inhibitors. The incubation proceeded for 2 h and the reaction was terminated by the addition of 0.5 ml ice-cold methanol, followed by centrifugation and evaporation of the supernatant in a Speed-Vac centrifuge (Savant, Hicksville, NY, USA).
Determination of substance P endopeptidase activity The conversion of synthetic substance P to its N-terminal fragment SPI_ 7 was followed by a RIA specific for the heptapeptide product. Ten/~1 CSF was preincubated for 15 min at 37°C together with 20/tM amastatin. Following the addition of 0.1/~g (74 pmol) of substance P the incubation was performed in duplicates in Eppendorf tubes at 37°C for 1 h in a final volume of 40/A buffered at pH 7.4 with 50 mM Tris-HCl. After completed incubation, samples were treated in accordance with those in the DCE assay described above.
Generation of antisera All antisera were generated in rabbits. Leu-enk-Arg 6 (1 mg) and thyroglobulin (5 mg) were dissolved in 300/~l 100 mM phosphate buffer, pH 7.4, and cooled to 0°C. Glutaraldehyde in 25% aqueous solution was diluted with ice-cold distilled water (1:100, v/v) and 180 pl was added dropwise. The reaction mixture was stirred for 30 min at 0°C followed by 24 h at room temperature. After extensive dialysis against 0.9% NaCl, the peptide-thyroglobulin conjugates were ready for injection into rabbits. A 500/d aliquot of the conjugate, corresponding to 100 pg of peptide, was emulsified with an equal volume of Freund's complete adjuvant. The animals were injected intracutaneously on their backs at multiple sites. Booster doses in volumes of 250 pl corresponding to 50 pg of peptide were emulsified with an equal volume of Freund's incomplete adjuvant and injected subcutaneously at intervals of 3-4 weeks. The crossreactivity of the enkephalyl hexapeptide antiserum (Benj G) with Met-enkephalin-Arg6 was 10% and with Met-enkephalin-Arg6 sulfoxide and Met-enkephalin-Arg6-Phe7, around 1%. With all other proenkephalin and prodynorphin-derived opioid peptides the crossreactivity was less than i%. The final dilution was 1:2800. Antiserum to SPI_ 7 (46 C) was raised in an analogous fashion. However, in this case dicyclohexylcarbodiimide was used as a coupling agent 5°. In the latter study, tritiated 2,4-dehydroproline-SPl_ 7 was used as radiolabel, but we here used the iodinated Tyr-SPl_7, which shows a different cross-reactivity pattern (Table II). The final dilution used under assay conditions was 1:450,000.
276
Radioimmunoassay (RIA) methods
Statistical analysis
The dried eluates were dissolved and diluted in a 1:1 MeOH/0.1 M HC1 mixture and 25 kd aliquots of samples and standards were added in triplicate to incubation vials together with i00/~1 of antisera and iodinated peptide (4,500 cpm), respectively. The antisera and labelled peptide were diluted in assay buffer containing 0.1% gelatin, 0.1% bovine serum albumin, 0.8% NaCI, 0.93% EDTA in 50 mM sodium phosphate (pH 7.4). The vials were then incubated at 4°C for 24 h. The bound and free peptides were separated by adding 200/al active charcoal. For the Leu-enk-Arg 6 assay, a charcoal suspension containing 500 mg active charcoal and 50 mg dextran T 70 dissolved in 200 ml 50 mM sodium phosphate (pH 7.4) was used, and in the SPI_ 7 assay the ratio charcoal/dextran T 70 was 750 mg/75 mg in the above buffer. After 10 min of incubation, the mixture was centrifuged for 1 min in a Beckman Microfuge8. A 300/~1 aliquot of the supernatant was removed and counted in a y-counter. The intraassay coefficient of variation for one RIA was calculated from eight quality controls (triplicates) evenly distributed within the RIA and in the same concentration interval as the sampies. The coefficient was found to be 5.6% and 10.8% for the SPI_ 7 and Leu-enk-Arg6 RIA, respectively. The interassay coefficient of variation was calculated from the ICs0 values of the standard curves, and was 12.3% (n = 8) and 13.1% (n = 10), respectively. The minimum detectable amount of the RIA was 0.3 and 14 fmol per tube, according to Signorella et a l : s, and 50% tracer displacement was attained at around 7 and 350 fmol per tube, respectively.
Statistical analyses of the recorded enzymatic activities were performed on means of duplicate determinations using two-factor analysis of variance (ANOVA) with factor interaction. All calculations were carried out on log-transformed data, since, when tested for normality, data were found to be log-normal distributed.
RESULTS
Accuracy o f enzyme measurements In o r d e r to investigate the c o n d i t i o n s necessary for p r e v e n t i n g f u r t h e r d e g r a d a t i o n of the e n z y m e p r o d u c t s ( L e u - e n k - A r g 6 and SPI_7) the effect o f d i f f e r e n t inhibitors on the a m o u n t of p r o d u c t g e n e r a t e d by D C E and S P E versus t i m e was investigated. Possible changes in o t h e r C S F p r o t e a s e s and the lack of suitable inhibitors against these activities m a y give a l t e r e d rates of p r o d u c t f o r m a t i o n , leading to m i s i n t e r p r e t a t i o n s c o n c e r n i n g the e n z y m e s studied. Fig. 1 illustrates the t i m e c o u r s e of L e u - e n k - A r g 6 form a t i o n in the p r e s e n c e of v a r i o u s p r o t e a s e inhibitors. If no i n h i b i t o r is a d d e d , only a small increase in p r o d u c t
Reversed-phase chromatography on SMART system
f o r m a t i o n is o b s e r v e d during the first 120 min. T h e r e a f -
CSF samples (100/d) from control and arthritic rats were analyzed by the SMART system (Pharmacia LKB Biotechnology, Uppsala, Sweden). The computer controlled micropreparative chromatography system was equipped with built-in detector cells for UV (UV-MII with 214 nm optics) and conductivity measurements. The conductivity scale was set by calibration with eluent A (100%) and eluent B (0%); A, 0.04% trifluoroacetic acid (TFA); B, 0.034% TFA in 60% acetonitrile. The column used was a/aRPC C2lC18, PC 3.2/3 (particle size 3/~m, 120/~; 30 x 3.2 mm I.D.). Elution was performed with a gradient of acetonitrile (0-60%) containing TFA at a flow-rate of 240/A/min.
ter, the c u r v e declines. H o w e v e r , w h e n the a m i n o p e p t i dase i n h i b i t o r a m a s t a t i n and the e n k e p h a l i n c o n v e r t a s e inhibitor GEMSA
are p r e s e n t , a m o r e p l a n a r shape is
s e e n after 120 m i n of incubation. A m u c h m o r e pron o u n c e d effect is o b s e r v e d w h e n the A C E i n h i b i t o r captopril
is a d d e d
GEMSA.
in
combination
with
amastatin
and
Inhibitors of the n e u t r a l e n d o p e p t i d a s e ( N E P
or E C 3.4.24.11), such as p h o s p h o r a m i d o n
and thior-
p h a n (each a d d e d in 2 0 / ~ M ) , had n o significant effect (data not shown).
8 ----o--- Control -----or-- Amastatin+GEMSA 70 A m ~
V [ /
1.6
/
1.4
Consequently,
--..o-- Control
amastatin,
GEMSA
T
Captolmil+Am~tatin J
1.2 ~ 1.0 2
0.8
~ 0.6
0.4
1 00
0.2 30
60
90
120
150
180
Time (min)
Fig. 1. Time-course of dynorphin A conversion to Leu-enk-Arg 6 in rat CSF in the absence and presence of both amastatin (20/zM), GEMSA (1 mM) and combined with captopril (20 ktM). 15/~1 CSF was preincubated for 15 min at 37°C with the different inhibitors used. After the addition of 0.1 /~g dynorphin A incubation was performed in a final volume of 40 kd, buffered at pH 7.4 by 50 mM Tris-HCl. The reaction was terminated by the addition of 0.5 ml ice-cold methanol and samples were evaporated before radioimmunoassay. Vertical lines show standard error (S.E.), n = 3.
30
60
90 Time (rain)
120
150
180
Fig. 2. Time-course of SP conversion to SPI_7 in rat CSF in the absence and presence of captopril (20/~M) alone or combined with amastatin (20 ktM). 10/~1 CSF (diluted × 3) was preincubated for 15 min at 37°C with the different inhibitors used. After the addition of 0.1 ktg SP incubation was performed in a final volume of 40 :d, buffered at pH 7.4 by 50 mM Tris-HC1. For further details see legend to Fig. 1 and the text. Vertical lines show standard error (S.E.), n = 3.
277 12-
150.
10-
125.
8.
E * 100
day 21 p.i.
~
4
day 38 p.i.
I~
,
~ 75 ff 50
14.
25
20 10
. . --- , .... 15
, .... 20
, .... 25
, .... 30
, .... 35
40
Control
CIA
Control
CIA
n=8
n--7
n=9
n=7
Days after immunization
Fig. 3. T h e development of arthritis after immunization with type II collagen (collagen induced arthritis), C I A , n . = 10). T h e develo p m e n t of arthritis was followed by a macroscopic scoring system for the four paws, ranging from 0 to 3 (1 = swelling and/or redness of one toe or finger joint, 2 = two or more joints involved and 3 = severe arthritis in the entire paw). T h e arrows indicate time points when CSF were collected. Serum was collected a t day 20 after immunization. M e a n + S.E.M. ,
and captopril were used in the DCE assay. Fig. 2 shows the corresponding time course for SPI_ 7 formation and how this is influenced by captopril alone or by captopril combined with amastatin. It is evident that the SPE activity is effectively inhibited by the ACE inhibitor captopril. The proline residue adjacent to the N-terminal arginine in SPI_ 7 (for structure, see Table I) should make the heptapeptide stable against aminopeptidases. To test whether aminopeptidases have an effect upon SPI_ 7 formation, amastatin was used in combination with captopril. Indeed, the addition of amastatin seems to have a slight stabilizing effect on the product formation with time. If no inhibitor is used an almost linear relationship is observed during the first two hours of incubation. Therefore, in the SPE assay, amastatin
6 0 t 4--T'-day21p'i"
~
4
day38p, i.
Fig. 5. Substance P endopeptidase (SPE) activity in CSF from untreated rats (control; open bar) and rats i m m u n i z e d with type II collagen (collagen induced arthritis, CIA, shaded bar). CSF was sampled day 21 and day 38 after the immunization. Enzymatic activity was m e a s u r e d as the rate of formation of substance P1-7 from substance P. M e a n + S.E.M., p.i. = post immunization.
was added to prevent possible influences by aminopeptidases. Our results also indicated that none of the inhibitors discussed above influenced the RIA by itself. Levels of DCE and SPE in CSF from arthritic rats D A rats with collagen-induced arthritis developed a severe and chronic arthritis. Macroscopic signs of the arthritis development in the four paws started at day 15. The severity increased fast to day 20 and thereafter a slow increase was observed during the whole experiment, i.e. unti 1 day 38 after immunization (Fig. 3). All rats immunized with CII developed an anti-CII antibody response (185 + 147/zg/ml at day 20 after immunization, n = 10) whereas no antibodies were detected in control rats. CSF samples were collected both during the acute phase (day 21) and during the chronic phase (day 38). DCE and SPE activities in CSF collected from control and rats with collagen-induced arthritis, in both acute and chronic phase, were analysed with a two fac-
I~ ]
50-
500¸
400,450'
. .-"'"
350 ~ 300
30 .
