Agents and Actions, vol. 34, 1/2 (1991)
0065-4299/91/020077-04 $1.50 +0.20/0 9 ~991 Birkh~tuserVedag, Basel
Antiflammin-2 ( H D M N K V L D L ) does not inhibit phospholipase A 2 activities W C . Hope 1., B.J. Patel 1 and D.R. Bolin 2 Department of Allergyand InflammationResearch and 2 Department of Peptide Research, Hoffmann-LaRoche Inc., Nutley, NJ 07110, USA
Abstract
A basic nonapeptide P2 (antiflammin-2, H D M N K V L D L ) which is identical to a portion of the amino acid sequence (residues 246-254) of lipocortin I, has been described to have antiinflammatory activity in a rat paw edema model (Nature 335:726-730 [1988]). P2 (0.05 ~tM) was also reported to inhibit porcine pancreatic phospholipase Az (PLA2). The effect of synthetic P2 (98% pure) on PLA2 was evaluated in two assay systems, Using porcine pancreatic PLA 2 and phosphatidylcholine/deoxycholate mixed micellar substrate, P2 (0.005-50 ~tM) had no effect on PLA2 activity, even in the presence of 2-mercaptoethanol to prevent peptide oxidation. In another assay, using human synovial fluid PLA 2 as the enzyme and [14C]-oleate-labelled E. coli substrate, P2 (0.005-50 ~tM) had no significant effect on PLA z activity. A reported PLA2 inhibitor, manoalide, was a potent inhibitor of PLA z in both assay systems. On the basis of these results, we conclude that P2 is devoid of PLA2 inhibitory activity.
Introduction
Phospholipase A2 (PLA2) may play a key role in inflammatory diseases by releasing arachidonic acid from membrane phospholipids and thus triggering the production of inflammatory lipid mediators such as leukotrienes, prostagtandins and platelet activating factor. Inhibitors of PLA 2 activity may have potential therapeutic use as antiinflammatory agents. A basic nonapeptide, P2 (antiflammin-2) which has an amino acid sequence ( H D M N K V L D L ) that occurs in a region of a reported PLA2-inhibitory protein, lipocortin I, was reported to have antiinflammatory activity in carrageenan-induced rat paw edema models [l, 2]. P2 was also reported to be a potent inhibitor of porcine pancreatic PLA 2 activity (90% inhibition * Author for correspondence.
at 0.05 gM) [1], human polymorphonuclear leukocyte sonicate PLA 2 activity (74% inhibition at 0.2 laM [3] and PLA2 activity in extracts of human psoriatic epidermis (47% inhibition at 0.05 gM) [4]. On the other hand, it has been suggested that P2 does not inhibit porcine pancreatic PLA 2 activity [5, 6] or carrageenan-induced rat paw edema [6]. Therefore, in order to critically evaluate P2 as a putative PLA 2 inhibitor, we studied the effects of synthetic P2 (98% pure) on PLA 2 activity in two assay systems: 1) porcine pancreatic PLA 2 using micellar substrate and 2) human synovial fluid PLA 2 using cell membrane substrate. Materials and methods
The synthetic peptide P2 (antiflammin-2, purchased from Peptide Technologies Corp., Washington D.C.) was analyzed and its amino acid con-
78
Agents and Actions, vol. 34, 1/2 (1991)
EFFECT OF P2 AND MANOALIDEON PORCINE PANCREATICPHOSPHOLIPASEAa ACTIVITY BO Q P2 ( ANTIFLAMMIN-2) 0 MANOALIDE
60 r
T
0
O
z
"'
40
J
0 l-.-
/
Q'l.--I
t
tt
_I o 9- r
'-r"
z
fao
o Ill..
~
-20
)
-
2 .
.00t Fibre1
|
t.,.,,I
9
.01
- - ,
,,.,,I
.
.1
9
I
.....
I
t
. . . . . . . .
I
. . . . . . . .
t0
I
100
,
.
,
t000
CONCENTRATION ( ~M )
The effects of peptide P2 and manoalide on porcine pancreatic phospholipase A z activity. Enzyme was preincubated with P2 or mauoalide, substrate was added and the mixture was incubated as described in Methods. D M S O (2%), which was used to solubilize manoalide, had no effect on PLA 2 activity. The results are presented as the percent inhibition o f P L A 2 activity (10.8__+0.3% hydrolysis, m e a n _ SE of two experiments, n = 9 or 14). Each point represents the mean percent inhibition of duplicate or triplicate determinations. The specific activity of the enzyme was 1.4 gmol [1 - 14C]-arachidonic acid released m i n - 1 mg protein- t using this assay method.
tent and sequence, FAB-mass spectrum, and purity by HPLC (98%) were confirmed. Porcine pancreatic PLA2 (Boehringer Mannheim) was assayed using 10 IxM L-a-l-palmitoyl-2-[114C]-arachidonyl phosphatidylcholine (from NEN, Boston, MA) as the substrate in mixed micelles with 1 m M sodium deoxycholate as previously described [1]. Enzyme (5 nM) was preincubated with P2 or manoalide (a reported PLA 2 inhibitor) for 5 min at 37 ~ micellar substrate/detergent was added and the mixture (total volume 100 gl, containing 100raM Tris-HC1, pH 8.0, 1 m M Ca + + and 100 m M Na +) was incubated for 4 min at 37 ~ The lipid extraction of the reaction mixture and TLC analysis oflipids to resolve product from substrate was performed as previously reported [7, 8]. Human synovial fluid PLA2 (HSF-PLA2) activity was assayed using [1-14C]-oleate-labelled, autoclaved E. coli substrate (20,000 dpm/ml = 2 x 109 cells/ml = 18 gM phospholipids) as described earlier [7, 8].
Results
Manoalide, a reported PLA a inhibitor which was used for comparative purposes, inhibited porcine pancreatic PLA2 activity in a concentration-dependent manner, using phosphatidylcholine/deoxycholate micellar substrate (Fig. 1). A maximum of 4 4 - 65 % inhibition of enzyme activity was observed at the higher manoalide concentrations (1-100 gM). In contrast, peptide P2 concentrations of 0.005- 500 gM had no significant effect on this pancreatic enzyme activity. Furthermore, P2 did not inhibit the latter enzyme even in the presence of a reducing agent (i.e., 50 I~M 2-mercaptoethanol) to prevent potential oxidation of this peptide (data not shown). In the other assay system, the reference compound, manoalide was a potent inhibitor of HSF-PLA2 activity (ICso 5 gM), using the cell membrane substrate, 14C-oleate-labelled E. eoli (Fig. 2). At 30 gM manoalide, HSF-PLAz activity was completely abolished. In comparison, P2, at concentra-
79
Agents and Actions, vol, 34, 1/2 (1991)
EFFECT OF P2 AND MANOALIDEON HUMAN SYNOVIAL FLUID PHOSPHOLIPASEAa ACTIVITY
p.--:~-~
t00 /
80 El_
~
Z o
llJ co
,--,.,-,
" .J
Hz "-
co o
"- ,' --
9 P2 ( ANTIFLAMMIN-2 )
/"
C) MANOALIDE
,
60
/ /
40
}/
20
~
,,,,,' T./~
-20 .00! Figure2
.01
,t t CONCENTRATION #M )
(
10
t00
The effects of peptide P2 and manoalide on human synovial fluid phosphotipase A 2 activity. Enzyme (0,2% human synovial fluid, v/v) was preincubated with P2 or manoalide for 5 min at 37 ~ radiolabelled E. coli substrate was added and the mixture was incubated as indicated in Methods. The results are presented as the percent inhibition of HSF-PLA 2 activity, corrected for nonenzymatic substrate hydrolysis, Each point represents the mean percent inhibition of 5-6 determinations from two experiments.
tions of 0.005-50 gM, had no significant effect on HSF-PLA 2 activity. Discussion Several reports on the PLA2-inhibitory activity of the nonapeptide P2 (antiflammin-2), using a variety of enzyme sources, phospholipid substrates, and assay methodologies, are in direct conflict (1,3-6). Since the purity of the P2 utilized may have contributed to these discrepancies, we confirmed the identity and purity (98 %) of a synthetic sample of this peptide. After following an assay method which closely mimicked that of the earliest report on P2 [1], we were unable to demonstrate any significant PLAE-inhibitory activity of P2 using porcine pancreatic PLA 2 and a phospholipid/ detergent mixed micellar substrate. Thus, our results on the effect of P2 on the pancreatic enzyme activity do not agree with those of the earliest report [1], but are in agreement with those of two other reports [5, 6]. It has been suggested that P2
might be inactivated by oxidation of its methionine residue in position 3 and may be effective only in the presence of a reducing agent such as 2-mercaptoethanol [3]. In contrast, we found that P2 did not inhibit porcine pancreatic PLA 2 activity, even in the presence of 2-mercaptoethanol to prevent potential oxidation of this peptide. The lack of PLA 2inhibitory activity of P2 was confirmed in the other PLA 2 assay system, which used a different enzyme source (HSF-PLA2) , type of substrate (cell membrane-associated phospholipids) and assay conditions. Although the original report on P2 [1] described PLAa-inhibitory activity at 0.05 ~tM, the present study shows that highly pure P2 at concentrations of up to 1,000-10,000 fold higher than this do not inhibit PLA 2 activities in two different assay systems. It has previously been reported that a structurally related basic nonapeptide P1 (antiflammin-1, M Q M K K V L D S ) either does [1, 3] or does not [5, 6, 8, 9] inhibit PLA 2 activities. The reasons for these conflicting reports on the PLA 2inhibitory activity of these nonapeptides are not
80
known, but it is possible that the purity of P1 and P2 and/or the design of the enzyme assay systems may influence this activity. On the basis of the present results, using two assay systems, we conclude that peptide P2 has no PLA/-inhibitory activities. References [1] L. Miele, E. Cordella-Miele, A. Facchiano and A. B. Mukherjee, Novel anti-inflammatory peptides from the region of highest similarity between uteroglobin and lipocortin L Nature 335, 726-730 (1988). [2] A. Ialenti, P. M. Doyle, G. N. Hardy, D. S. E. Simpkin and M. Di Rosa, Anti-inflammatory effects of vasocortin and nonapeptide fragments of uteroglobin and lipocortin I (antiflamrains). Agents and Actions 29, 48-49 (1990). [3] G. Camussi, C. Tetta, F. Bussolino and C. Baglioni, Antiinflammatory peptides (antiflammins) inhibit synthesis of platelet-activating factor, neutrophil aggregation and chemotaxis, and intradermal inflammatory reactions. J. Exp. Med. 171, 913-927 (1990).
Agents and Actions, vol. 34, 1/2 (1991) [4] P. H. Cartwright, E. llderton, J. M. Sowden and H. J. Yardley, Inhibition of phospholipase A 2 in extracts of lesion-free psoriatic epidermis by the nonapeptide HDMNKVLDL which corresponds to lipocortin I residues 246-254. Br. J. Dermatol. 122, 277-278 (1990). [5] J. van Binsbergen, A. J. Slotboom, A. J. Aarsman and G. H. de Haas, Synthetic peptide from lipocortin I has no phospholipase A 2 inhibitory activity. FEBS Lett. 247, 293-297 (1989). [6] F. M~irki, J. Pfeilschifter, H. Rink and I. Wiesenberg, "Antiflammins": Two nonapeptide fragments of uteroglobin and lipocortin 1 have no phospholipase A2-inhibitory and anti-inflammatory activity. FEBS Lett. 264, 171-175 (1990). [7] W. C. Hope, B. J. Patel, C. Fiedler-Nagy and B. H. Wittreich. Retinoids inhibit phospholipase A 2 in human synovial fluid and arachidonic acid release from rat peritoneal maerophages. Inflammation 14. 543-559 (1990). [8] W. C. Hope, B. J. Patel and D. R. Bolin, P1 peptide (antiflamrain-l) does not inhibit phospholipase A z activities. FASEB J. 4, A2230 (1990). [9] D. J. Masters, A. S. Dutta and R. M. McMillan, Is antiflamrain P1 a phospholipase A 2 inhibitor? Br. J. Pharmacol. 98, 849P (1989).
0065-4299/91/020077-04 $1.50 +0.20/0 9 ~991 Birkh~tuserVedag, Basel
Antiflammin-2 ( H D M N K V L D L ) does not inhibit phospholipase A 2 activities W C . Hope 1., B.J. Patel 1 and D.R. Bolin 2 Department of Allergyand InflammationResearch and 2 Department of Peptide Research, Hoffmann-LaRoche Inc., Nutley, NJ 07110, USA
Abstract
A basic nonapeptide P2 (antiflammin-2, H D M N K V L D L ) which is identical to a portion of the amino acid sequence (residues 246-254) of lipocortin I, has been described to have antiinflammatory activity in a rat paw edema model (Nature 335:726-730 [1988]). P2 (0.05 ~tM) was also reported to inhibit porcine pancreatic phospholipase Az (PLA2). The effect of synthetic P2 (98% pure) on PLA2 was evaluated in two assay systems, Using porcine pancreatic PLA 2 and phosphatidylcholine/deoxycholate mixed micellar substrate, P2 (0.005-50 ~tM) had no effect on PLA2 activity, even in the presence of 2-mercaptoethanol to prevent peptide oxidation. In another assay, using human synovial fluid PLA 2 as the enzyme and [14C]-oleate-labelled E. coli substrate, P2 (0.005-50 ~tM) had no significant effect on PLA z activity. A reported PLA2 inhibitor, manoalide, was a potent inhibitor of PLA z in both assay systems. On the basis of these results, we conclude that P2 is devoid of PLA2 inhibitory activity.
Introduction
Phospholipase A2 (PLA2) may play a key role in inflammatory diseases by releasing arachidonic acid from membrane phospholipids and thus triggering the production of inflammatory lipid mediators such as leukotrienes, prostagtandins and platelet activating factor. Inhibitors of PLA 2 activity may have potential therapeutic use as antiinflammatory agents. A basic nonapeptide, P2 (antiflammin-2) which has an amino acid sequence ( H D M N K V L D L ) that occurs in a region of a reported PLA2-inhibitory protein, lipocortin I, was reported to have antiinflammatory activity in carrageenan-induced rat paw edema models [l, 2]. P2 was also reported to be a potent inhibitor of porcine pancreatic PLA 2 activity (90% inhibition * Author for correspondence.
at 0.05 gM) [1], human polymorphonuclear leukocyte sonicate PLA 2 activity (74% inhibition at 0.2 laM [3] and PLA2 activity in extracts of human psoriatic epidermis (47% inhibition at 0.05 gM) [4]. On the other hand, it has been suggested that P2 does not inhibit porcine pancreatic PLA 2 activity [5, 6] or carrageenan-induced rat paw edema [6]. Therefore, in order to critically evaluate P2 as a putative PLA 2 inhibitor, we studied the effects of synthetic P2 (98% pure) on PLA 2 activity in two assay systems: 1) porcine pancreatic PLA 2 using micellar substrate and 2) human synovial fluid PLA 2 using cell membrane substrate. Materials and methods
The synthetic peptide P2 (antiflammin-2, purchased from Peptide Technologies Corp., Washington D.C.) was analyzed and its amino acid con-
78
Agents and Actions, vol. 34, 1/2 (1991)
EFFECT OF P2 AND MANOALIDEON PORCINE PANCREATICPHOSPHOLIPASEAa ACTIVITY BO Q P2 ( ANTIFLAMMIN-2) 0 MANOALIDE
60 r
T
0
O
z
"'
40
J
0 l-.-
/
Q'l.--I
t
tt
_I o 9- r
'-r"
z
fao
o Ill..
~
-20
)
-
2 .
.00t Fibre1
|
t.,.,,I
9
.01
- - ,
,,.,,I
.
.1
9
I
.....
I
t
. . . . . . . .
I
. . . . . . . .
t0
I
100
,
.
,
t000
CONCENTRATION ( ~M )
The effects of peptide P2 and manoalide on porcine pancreatic phospholipase A z activity. Enzyme was preincubated with P2 or mauoalide, substrate was added and the mixture was incubated as described in Methods. D M S O (2%), which was used to solubilize manoalide, had no effect on PLA 2 activity. The results are presented as the percent inhibition o f P L A 2 activity (10.8__+0.3% hydrolysis, m e a n _ SE of two experiments, n = 9 or 14). Each point represents the mean percent inhibition of duplicate or triplicate determinations. The specific activity of the enzyme was 1.4 gmol [1 - 14C]-arachidonic acid released m i n - 1 mg protein- t using this assay method.
tent and sequence, FAB-mass spectrum, and purity by HPLC (98%) were confirmed. Porcine pancreatic PLA2 (Boehringer Mannheim) was assayed using 10 IxM L-a-l-palmitoyl-2-[114C]-arachidonyl phosphatidylcholine (from NEN, Boston, MA) as the substrate in mixed micelles with 1 m M sodium deoxycholate as previously described [1]. Enzyme (5 nM) was preincubated with P2 or manoalide (a reported PLA 2 inhibitor) for 5 min at 37 ~ micellar substrate/detergent was added and the mixture (total volume 100 gl, containing 100raM Tris-HC1, pH 8.0, 1 m M Ca + + and 100 m M Na +) was incubated for 4 min at 37 ~ The lipid extraction of the reaction mixture and TLC analysis oflipids to resolve product from substrate was performed as previously reported [7, 8]. Human synovial fluid PLA2 (HSF-PLA2) activity was assayed using [1-14C]-oleate-labelled, autoclaved E. coli substrate (20,000 dpm/ml = 2 x 109 cells/ml = 18 gM phospholipids) as described earlier [7, 8].
Results
Manoalide, a reported PLA a inhibitor which was used for comparative purposes, inhibited porcine pancreatic PLA2 activity in a concentration-dependent manner, using phosphatidylcholine/deoxycholate micellar substrate (Fig. 1). A maximum of 4 4 - 65 % inhibition of enzyme activity was observed at the higher manoalide concentrations (1-100 gM). In contrast, peptide P2 concentrations of 0.005- 500 gM had no significant effect on this pancreatic enzyme activity. Furthermore, P2 did not inhibit the latter enzyme even in the presence of a reducing agent (i.e., 50 I~M 2-mercaptoethanol) to prevent potential oxidation of this peptide (data not shown). In the other assay system, the reference compound, manoalide was a potent inhibitor of HSF-PLA2 activity (ICso 5 gM), using the cell membrane substrate, 14C-oleate-labelled E. eoli (Fig. 2). At 30 gM manoalide, HSF-PLAz activity was completely abolished. In comparison, P2, at concentra-
79
Agents and Actions, vol, 34, 1/2 (1991)
EFFECT OF P2 AND MANOALIDEON HUMAN SYNOVIAL FLUID PHOSPHOLIPASEAa ACTIVITY
p.--:~-~
t00 /
80 El_
~
Z o
llJ co
,--,.,-,
" .J
Hz "-
co o
"- ,' --
9 P2 ( ANTIFLAMMIN-2 )
/"
C) MANOALIDE
,
60
/ /
40
}/
20
~
,,,,,' T./~
-20 .00! Figure2
.01
,t t CONCENTRATION #M )
(
10
t00
The effects of peptide P2 and manoalide on human synovial fluid phosphotipase A 2 activity. Enzyme (0,2% human synovial fluid, v/v) was preincubated with P2 or manoalide for 5 min at 37 ~ radiolabelled E. coli substrate was added and the mixture was incubated as indicated in Methods. The results are presented as the percent inhibition of HSF-PLA 2 activity, corrected for nonenzymatic substrate hydrolysis, Each point represents the mean percent inhibition of 5-6 determinations from two experiments.
tions of 0.005-50 gM, had no significant effect on HSF-PLA 2 activity. Discussion Several reports on the PLA2-inhibitory activity of the nonapeptide P2 (antiflammin-2), using a variety of enzyme sources, phospholipid substrates, and assay methodologies, are in direct conflict (1,3-6). Since the purity of the P2 utilized may have contributed to these discrepancies, we confirmed the identity and purity (98 %) of a synthetic sample of this peptide. After following an assay method which closely mimicked that of the earliest report on P2 [1], we were unable to demonstrate any significant PLAE-inhibitory activity of P2 using porcine pancreatic PLA 2 and a phospholipid/ detergent mixed micellar substrate. Thus, our results on the effect of P2 on the pancreatic enzyme activity do not agree with those of the earliest report [1], but are in agreement with those of two other reports [5, 6]. It has been suggested that P2
might be inactivated by oxidation of its methionine residue in position 3 and may be effective only in the presence of a reducing agent such as 2-mercaptoethanol [3]. In contrast, we found that P2 did not inhibit porcine pancreatic PLA 2 activity, even in the presence of 2-mercaptoethanol to prevent potential oxidation of this peptide. The lack of PLA 2inhibitory activity of P2 was confirmed in the other PLA 2 assay system, which used a different enzyme source (HSF-PLA2) , type of substrate (cell membrane-associated phospholipids) and assay conditions. Although the original report on P2 [1] described PLAa-inhibitory activity at 0.05 ~tM, the present study shows that highly pure P2 at concentrations of up to 1,000-10,000 fold higher than this do not inhibit PLA 2 activities in two different assay systems. It has previously been reported that a structurally related basic nonapeptide P1 (antiflammin-1, M Q M K K V L D S ) either does [1, 3] or does not [5, 6, 8, 9] inhibit PLA 2 activities. The reasons for these conflicting reports on the PLA 2inhibitory activity of these nonapeptides are not
80
known, but it is possible that the purity of P1 and P2 and/or the design of the enzyme assay systems may influence this activity. On the basis of the present results, using two assay systems, we conclude that peptide P2 has no PLA/-inhibitory activities. References [1] L. Miele, E. Cordella-Miele, A. Facchiano and A. B. Mukherjee, Novel anti-inflammatory peptides from the region of highest similarity between uteroglobin and lipocortin L Nature 335, 726-730 (1988). [2] A. Ialenti, P. M. Doyle, G. N. Hardy, D. S. E. Simpkin and M. Di Rosa, Anti-inflammatory effects of vasocortin and nonapeptide fragments of uteroglobin and lipocortin I (antiflamrains). Agents and Actions 29, 48-49 (1990). [3] G. Camussi, C. Tetta, F. Bussolino and C. Baglioni, Antiinflammatory peptides (antiflammins) inhibit synthesis of platelet-activating factor, neutrophil aggregation and chemotaxis, and intradermal inflammatory reactions. J. Exp. Med. 171, 913-927 (1990).
Agents and Actions, vol. 34, 1/2 (1991) [4] P. H. Cartwright, E. llderton, J. M. Sowden and H. J. Yardley, Inhibition of phospholipase A 2 in extracts of lesion-free psoriatic epidermis by the nonapeptide HDMNKVLDL which corresponds to lipocortin I residues 246-254. Br. J. Dermatol. 122, 277-278 (1990). [5] J. van Binsbergen, A. J. Slotboom, A. J. Aarsman and G. H. de Haas, Synthetic peptide from lipocortin I has no phospholipase A 2 inhibitory activity. FEBS Lett. 247, 293-297 (1989). [6] F. M~irki, J. Pfeilschifter, H. Rink and I. Wiesenberg, "Antiflammins": Two nonapeptide fragments of uteroglobin and lipocortin 1 have no phospholipase A2-inhibitory and anti-inflammatory activity. FEBS Lett. 264, 171-175 (1990). [7] W. C. Hope, B. J. Patel, C. Fiedler-Nagy and B. H. Wittreich. Retinoids inhibit phospholipase A 2 in human synovial fluid and arachidonic acid release from rat peritoneal maerophages. Inflammation 14. 543-559 (1990). [8] W. C. Hope, B. J. Patel and D. R. Bolin, P1 peptide (antiflamrain-l) does not inhibit phospholipase A z activities. FASEB J. 4, A2230 (1990). [9] D. J. Masters, A. S. Dutta and R. M. McMillan, Is antiflamrain P1 a phospholipase A 2 inhibitor? Br. J. Pharmacol. 98, 849P (1989).
