Regulatory Peptides, 28 (1990) 39-46 Elsevier
39
REGPEP 00887
N-terminally extended substance P is released together with substance P from rat spinal cord Gudrun Toresson 1, Carola de las Carreras ~ Leif Bertilsson 1 and Ernst Brodin 2 1Department of Clinical Pharmacology at the Karolinska Institute, Huddinge Hospital, Huddinge and 2Department of Pharmacology, Karolinska Institute, Stockholm (Sweden) (Received 25 August 1989; revised version received and accepted 20 November 1989)
Key words: Substance P; N-terminally extended substance P; Rat spinal cord; Release
Summary The release of different forms of substance P-like immunoreactivity (SP-LI) from superfused slices of rat spinal cord was studied. The released SP-LI was characterized by reverse-phase high-performance liquid chromatography and radioimmunoassay with two antisera directed to the C- and N-terminal parts of SP, respectively. The SP-LI detected in the superfusates with the C-terminally directed antiserum was found to consist of (undeca) SP, SP-sulfoxide and a late eluting component which was not detectable with the N-terminally directed antiserum. This component was also found in neutral extracts of the spinal cord. Upon trypsin digestion, it produced SP-LI detectable with both C- and N-terminally directed antiserum which also coeluted with SP. From these results we conclude that this form of SP-LI most likely corresponds to an N-terminally extended form of SP. An increase of the potassium concentration in the superfusion fluid from 5 to 50 mM evoked an increased overflow of both SP and the N-terminally extended SP. The present results indicate that N-terminally extended SP is released by a calcium-dependent mechanism together with SP from terminals in the spinal cord in response to potassium stimulation.
Correspondence: E. Brodin, Department of Pharmacology, Karolinska Institute, S-10401 Stockholm, Sweden. 0167-0115/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
40 Introduction
The tachykinins are neuropeptides characterized by a common C-terminal sequence of amino acids, and a number of biological actions [ 1]. Substance P (SP), the most studied of the group, is currently accepted as a neurotransmitter or neuromodulator [2,3]. Two new members of the tachykinin family, neurokininA (NKA) and neurokinin B (NKB) (for references see Ref. 3) as well as N-terminally extended forms of NKA, neuropeptide K [4] and neuropeptide V [5] have recently been identified in mammals. Neuropeptide K and neuropeptide 7 represent posttranslational products of fl- and 7-preprotachykinin (PPT) A [6,7], respectively. Recently we reported that an N-terminally extended form of SP occurs in human cerebrospinal fluid (CSF) [8]. This or a compound very similar to it, is also found in rat spinal cord [8,9]. The N-terminally elongated SP is likely to represent a specific posttranslational product of preprotachykinins. All preprotachykinins containing the SP-sequence (~-, fl-, and 7-PPT A) are identical in amino acid sequence in the region N-terminal to SP [ 10]. Thus all of them could generate the same N-terminally extended form of SP. The question if the elongated forms of tachykinins only represent partially processed precursor forms without other functional significance, or if they may possess transmitter functions of their own, remains open. The aim of the present study was to determine whether the N-terminally elongated SP is released together with SP from the rat spinal cord in vitro upon potassium stimulation.
Materials and Methods
Release experiments The experimental procedures used in the release experiments have been described previously [11,12]. Male Sprague-Dawley rats (body weight 200 g) were decapitated and the spinal cords were quickly dissected out. The tissue was immediately sliced (0.4 ram) using a McIlwain tissue chopper (Micle laboratory engineering, Gomshall, U.K.) and transferred to superfusion chambers (one spinal cord in each chamber) perfused with Krebs-Ringer solution (in raM; NaC1 138, KC1 5, CaCI2 1, MgC12 1, NaH2PO 4 1, NaHCO3 11, glucosel 1) which contained 0.2% bovine serum albumin and 0.03 % bacitracin and which was adjusted to pH 7.4 by bubbling with 95% 02 and 5 % CO 2. The sodium concentration was lowered during potassium (50 raM) stimulation periods to keep constant osmolarity. The flow rate was 0.2 ml/min and the temperature was 37 °C. The chambers were perfused during a wash out period of 1 h before samples were collected. In one initial experiment, the concentration of potassium was increased from 5 to 50 mM during two separate 3 min periods. From each of two chambers a total of eight fractions (1.8 ml in each fraction) were collected. The matched superfusate fractions from the two chambers were pooled, frozen immediately on dry ice and kept frozen until subjected to high-performance liquid chromatography (HPLC). In other experiments, the calcium dependence of the potassium-evoked release of substance P-like immuno-
41 reactivity (SP-LI) was examined. Ten chambers containing one spinal cord each were used. The release of SP-LI was stimulated twice as described above. The chambers were perfused with Krebs-Ringer solution containing no calcium and 7 mM magnesium during the first stimulation period and normal Krebs-Ringer solution containing 1 mM calcium during the second stimulation. Four 2.4 ml superfusate fractions from each chamber were collected and frozen on dry ice. Before HPLC analysis, matched fractions from two chambers were pooled. Chromatography Reverse-phase HPLC was performed using a Supelcosil LC-18-DB column (15 cm x 4.6 mm, 5/~m, Supelco). The samples were injected without pretreatment into the HPLC system via a precolumn (5 c m x 4.6 ram, packed with Lichroprep RP-18, 25-40 #m, Merck) [8]. The samples were then eluted at a flow rate of 1 ml/min with a gradient of acetonitrile containing 0.08 ~ trifluoroacetic acid in water, see Fig 2. 1-ml fractions were collected, evaporated to dryness in a vacuum centrifuge, and analysed by radioimmunoassay (RIA). The retention times for synthetic peptides were determined by HPLC and subsequent RIA in separate analyses. R adioimmunoassay SP-LI was measured using two different antisera, i.e. SP 2 and SP N1. Antiserum SP 2 is directed against the C-terminus of substance P and antiserum SP N1 against the N-terminus (Refs. 8, 9 and Brodin et al. unpublished data). The cross-reactivity of antiserum SP 2 with various SP-related peptides, NKA and NKB using SP as a 100~/o reference was as follows: SP sulphoxide (SP-ox) 56~, SP (3-11) 147~o, SP (4-11) 22~o, SP (1-9) less than 0.1~, SP-free acid and SP-Gly-Lys less than 0.01~o, SP (7-11), SP (8-11), SP (1-7), SP (1-4), NKA and NKB up to 0.001~o. The cross reactivity of antiserum SP N1 with SP-ox was 103~o, SP-free acid 106~o, SP-GIy-Lys 100~o, SP (1-9) 77~o, SP (1-7) 16~o, SP (1-4), SP (3-11) and SP (8-11) less than 0.001~. The RIA method has been described previously [8]. The IC5o value (the amount of unlabelled peptide needed to displace the radiollgand by 50 ~ ) was 2.1 fmol SP/100#I sample, and the limit of detection was 0.15 fmol SP/100/~1 sample. The inter-assay coefficient of variation was 14~o. All values are expressed in equivalents of synthetic (undeca) SP. The synthetic peptides were obtained from Peninsula Laboratories (CA, U.S.A.). Spinal cord extracts Following the initial release experiment, the spinal cord slices from each chamber were desintegrated by mixing in 1 ml Krebs-Ringer solution, and frozen on dry ice. After thawing, the homogenates were centrifuged and the supernatants were subjected to HPLC/RIA. For trypsin treatment, HPLC fractions 30-33 from two runs on the chromatograph were collected and evaporated to dryness. The results of one experiment are shown in Fig. 3. The incubation medium consisted of 400 #I 0.1 M ammonium bicarbonate buffer (pH 7,8) and 10 #g of trypsin (TPCK-treated, Sigma). After an incubation period of 4 h at 37 °C the reaction was stopped by adding 50/zl of 2 M acetic acid and heating the sample to 100 °C for 5 min. A control experiment where the sample
42 20D SP • C
15-
o~m ~
10"
N-terminally extended SP
5"
~
15
m
i
30
n
45
60
~
75
90 5 0 mM K +
Fig. 1. Release of SP-LI from slices of rat spinal cord (data from one experiment). The concentration of potassium (K + ) in the superfusion fluid was increased during two 3 rain periods as indicated. The levels of SP and N-terminally elongated SP were determined in the superfusates by HPLC and RIA (See Fig. 2) with a C-terminally directed antiserum (SP 2).
SP 20
Basal
20
82
I~-stimulation
SP ~15
15
S
O 4.s O
~ 10
10,
-I a.I
% CH3CN 5O
OJ
-~-~
2O
30
40
0
2'0
3'o
40
Fraction number Fig. 2. HPLC analysis of superfusates of rat spinal cord slices collected at basal conditions and after a K ÷ (50 raM) stimulation. The figure shows chromatograms of samples collected between 51-60 and 60-69 rain. in Fig. 1. The samples were eluted with an acetonitrile (CH3CN) gradient, as shown with the broken line. SP-LI was determined with the C.terminally directed antiserum (SP 2). The detection limit was 0.15 fmol SP-LI/HPLC fraction, The arrows indicate the elution positions of synthetic SP and SP sulfoxide (SP-ox).
43 was incubated and heated in acid under identical conditions but without trypsin was also performed. Results
An increase of the potassium concentration in the superfusion medium from 5 to 50 mM induced an increased overflow of SP-LI measured with the C-terminally directed antiserum SP 2 (Fig. 1). H P L C characterization of the SP-LI measured with this antiserum in the spinal cord superfusates collected at basal and stimulated conditions revealed three components (Fig. 2). The major immunoreactive component eluted in the position of SP, and a minor peak was found in the position of SP-sulfoxide (SP-ox). In addition to these two components there was a late component eluting in fractions 32-33, which corresponded to the elution position of the N-terminally extended SP previously identified in human CSF [8] and in neutral extracts of rat spinal cord [8,9]. This component was not detected using the N-terminally directed antiserum SP N1 (results not shown). A component detected with antiserum SP 2 (Fig. 3A) but not with SP N1 (not shown) and which eluted in fraction 32-33, could also be demonstrated in the spinal cord extracts. Treatment of this component (pooled fractions 30-33) with trypsin with subsequent HPLC/RIA analysis yielded a component which coeluted with SP (Fig. 3B). Incubation of this late eluting component under identical conditions but in the absence of trypsin, had no effect on this component and gave not rise to any immunoreactivity in the elution position of SP (not shown). When analysed separately, by combined H P L C and RIA, the overflow of the late eluting component of SP-LI, was found to be increased together with that of (undeca) SP in response to superfusion with Krebs-Ringer buffer containing 50 mM potassium (Figs. 1, 2 and 4). The second stimulation with potassium induced a smaller release of both components compared to the In'st stimulation (Fig. 1). The potassium-induced release of the late eluting component was found to be calcium dependent (Fig. 4). In the absence of calcium and presence of 7 mM magnesium, 50 mM potassium had no significant effect on the release of this component (Fig. 4). A
B
(Trypsin)
~-
SP-ox SP
SP
O
#
t
t
!1 2
~ "2_
o
;o
20
,
3'o
,
,o
!] o
" 10
2o
30
4u"
Fraction number
Fig. 3. HPLC of a rat spinal cord extract (A). The HPLC fractions were analysedwith the C-terminally directed antiserum(SP 2). Fractions correspondingto the late elutingpeak (indicatedin A) were collected followingchromatographyof another sample of the extract, treated with trypsin and rechromatographed (B). The detectionlimitwas 0.15 fmolSP-LI/HPLCfraction.The elutionpositions of syntheticSP and SP sult'oxide(SP-ox) are indicated. The acetonitrilegradient was the same as in Fig. 2.
44
c
E
0.8
o a.
m
0.6p < 0.001
"o "o f-
p
39
REGPEP 00887
N-terminally extended substance P is released together with substance P from rat spinal cord Gudrun Toresson 1, Carola de las Carreras ~ Leif Bertilsson 1 and Ernst Brodin 2 1Department of Clinical Pharmacology at the Karolinska Institute, Huddinge Hospital, Huddinge and 2Department of Pharmacology, Karolinska Institute, Stockholm (Sweden) (Received 25 August 1989; revised version received and accepted 20 November 1989)
Key words: Substance P; N-terminally extended substance P; Rat spinal cord; Release
Summary The release of different forms of substance P-like immunoreactivity (SP-LI) from superfused slices of rat spinal cord was studied. The released SP-LI was characterized by reverse-phase high-performance liquid chromatography and radioimmunoassay with two antisera directed to the C- and N-terminal parts of SP, respectively. The SP-LI detected in the superfusates with the C-terminally directed antiserum was found to consist of (undeca) SP, SP-sulfoxide and a late eluting component which was not detectable with the N-terminally directed antiserum. This component was also found in neutral extracts of the spinal cord. Upon trypsin digestion, it produced SP-LI detectable with both C- and N-terminally directed antiserum which also coeluted with SP. From these results we conclude that this form of SP-LI most likely corresponds to an N-terminally extended form of SP. An increase of the potassium concentration in the superfusion fluid from 5 to 50 mM evoked an increased overflow of both SP and the N-terminally extended SP. The present results indicate that N-terminally extended SP is released by a calcium-dependent mechanism together with SP from terminals in the spinal cord in response to potassium stimulation.
Correspondence: E. Brodin, Department of Pharmacology, Karolinska Institute, S-10401 Stockholm, Sweden. 0167-0115/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
40 Introduction
The tachykinins are neuropeptides characterized by a common C-terminal sequence of amino acids, and a number of biological actions [ 1]. Substance P (SP), the most studied of the group, is currently accepted as a neurotransmitter or neuromodulator [2,3]. Two new members of the tachykinin family, neurokininA (NKA) and neurokinin B (NKB) (for references see Ref. 3) as well as N-terminally extended forms of NKA, neuropeptide K [4] and neuropeptide V [5] have recently been identified in mammals. Neuropeptide K and neuropeptide 7 represent posttranslational products of fl- and 7-preprotachykinin (PPT) A [6,7], respectively. Recently we reported that an N-terminally extended form of SP occurs in human cerebrospinal fluid (CSF) [8]. This or a compound very similar to it, is also found in rat spinal cord [8,9]. The N-terminally elongated SP is likely to represent a specific posttranslational product of preprotachykinins. All preprotachykinins containing the SP-sequence (~-, fl-, and 7-PPT A) are identical in amino acid sequence in the region N-terminal to SP [ 10]. Thus all of them could generate the same N-terminally extended form of SP. The question if the elongated forms of tachykinins only represent partially processed precursor forms without other functional significance, or if they may possess transmitter functions of their own, remains open. The aim of the present study was to determine whether the N-terminally elongated SP is released together with SP from the rat spinal cord in vitro upon potassium stimulation.
Materials and Methods
Release experiments The experimental procedures used in the release experiments have been described previously [11,12]. Male Sprague-Dawley rats (body weight 200 g) were decapitated and the spinal cords were quickly dissected out. The tissue was immediately sliced (0.4 ram) using a McIlwain tissue chopper (Micle laboratory engineering, Gomshall, U.K.) and transferred to superfusion chambers (one spinal cord in each chamber) perfused with Krebs-Ringer solution (in raM; NaC1 138, KC1 5, CaCI2 1, MgC12 1, NaH2PO 4 1, NaHCO3 11, glucosel 1) which contained 0.2% bovine serum albumin and 0.03 % bacitracin and which was adjusted to pH 7.4 by bubbling with 95% 02 and 5 % CO 2. The sodium concentration was lowered during potassium (50 raM) stimulation periods to keep constant osmolarity. The flow rate was 0.2 ml/min and the temperature was 37 °C. The chambers were perfused during a wash out period of 1 h before samples were collected. In one initial experiment, the concentration of potassium was increased from 5 to 50 mM during two separate 3 min periods. From each of two chambers a total of eight fractions (1.8 ml in each fraction) were collected. The matched superfusate fractions from the two chambers were pooled, frozen immediately on dry ice and kept frozen until subjected to high-performance liquid chromatography (HPLC). In other experiments, the calcium dependence of the potassium-evoked release of substance P-like immuno-
41 reactivity (SP-LI) was examined. Ten chambers containing one spinal cord each were used. The release of SP-LI was stimulated twice as described above. The chambers were perfused with Krebs-Ringer solution containing no calcium and 7 mM magnesium during the first stimulation period and normal Krebs-Ringer solution containing 1 mM calcium during the second stimulation. Four 2.4 ml superfusate fractions from each chamber were collected and frozen on dry ice. Before HPLC analysis, matched fractions from two chambers were pooled. Chromatography Reverse-phase HPLC was performed using a Supelcosil LC-18-DB column (15 cm x 4.6 mm, 5/~m, Supelco). The samples were injected without pretreatment into the HPLC system via a precolumn (5 c m x 4.6 ram, packed with Lichroprep RP-18, 25-40 #m, Merck) [8]. The samples were then eluted at a flow rate of 1 ml/min with a gradient of acetonitrile containing 0.08 ~ trifluoroacetic acid in water, see Fig 2. 1-ml fractions were collected, evaporated to dryness in a vacuum centrifuge, and analysed by radioimmunoassay (RIA). The retention times for synthetic peptides were determined by HPLC and subsequent RIA in separate analyses. R adioimmunoassay SP-LI was measured using two different antisera, i.e. SP 2 and SP N1. Antiserum SP 2 is directed against the C-terminus of substance P and antiserum SP N1 against the N-terminus (Refs. 8, 9 and Brodin et al. unpublished data). The cross-reactivity of antiserum SP 2 with various SP-related peptides, NKA and NKB using SP as a 100~/o reference was as follows: SP sulphoxide (SP-ox) 56~, SP (3-11) 147~o, SP (4-11) 22~o, SP (1-9) less than 0.1~, SP-free acid and SP-Gly-Lys less than 0.01~o, SP (7-11), SP (8-11), SP (1-7), SP (1-4), NKA and NKB up to 0.001~o. The cross reactivity of antiserum SP N1 with SP-ox was 103~o, SP-free acid 106~o, SP-GIy-Lys 100~o, SP (1-9) 77~o, SP (1-7) 16~o, SP (1-4), SP (3-11) and SP (8-11) less than 0.001~. The RIA method has been described previously [8]. The IC5o value (the amount of unlabelled peptide needed to displace the radiollgand by 50 ~ ) was 2.1 fmol SP/100#I sample, and the limit of detection was 0.15 fmol SP/100/~1 sample. The inter-assay coefficient of variation was 14~o. All values are expressed in equivalents of synthetic (undeca) SP. The synthetic peptides were obtained from Peninsula Laboratories (CA, U.S.A.). Spinal cord extracts Following the initial release experiment, the spinal cord slices from each chamber were desintegrated by mixing in 1 ml Krebs-Ringer solution, and frozen on dry ice. After thawing, the homogenates were centrifuged and the supernatants were subjected to HPLC/RIA. For trypsin treatment, HPLC fractions 30-33 from two runs on the chromatograph were collected and evaporated to dryness. The results of one experiment are shown in Fig. 3. The incubation medium consisted of 400 #I 0.1 M ammonium bicarbonate buffer (pH 7,8) and 10 #g of trypsin (TPCK-treated, Sigma). After an incubation period of 4 h at 37 °C the reaction was stopped by adding 50/zl of 2 M acetic acid and heating the sample to 100 °C for 5 min. A control experiment where the sample
42 20D SP • C
15-
o~m ~
10"
N-terminally extended SP
5"
~
15
m
i
30
n
45
60
~
75
90 5 0 mM K +
Fig. 1. Release of SP-LI from slices of rat spinal cord (data from one experiment). The concentration of potassium (K + ) in the superfusion fluid was increased during two 3 rain periods as indicated. The levels of SP and N-terminally elongated SP were determined in the superfusates by HPLC and RIA (See Fig. 2) with a C-terminally directed antiserum (SP 2).
SP 20
Basal
20
82
I~-stimulation
SP ~15
15
S
O 4.s O
~ 10
10,
-I a.I
% CH3CN 5O
OJ
-~-~
2O
30
40
0
2'0
3'o
40
Fraction number Fig. 2. HPLC analysis of superfusates of rat spinal cord slices collected at basal conditions and after a K ÷ (50 raM) stimulation. The figure shows chromatograms of samples collected between 51-60 and 60-69 rain. in Fig. 1. The samples were eluted with an acetonitrile (CH3CN) gradient, as shown with the broken line. SP-LI was determined with the C.terminally directed antiserum (SP 2). The detection limit was 0.15 fmol SP-LI/HPLC fraction, The arrows indicate the elution positions of synthetic SP and SP sulfoxide (SP-ox).
43 was incubated and heated in acid under identical conditions but without trypsin was also performed. Results
An increase of the potassium concentration in the superfusion medium from 5 to 50 mM induced an increased overflow of SP-LI measured with the C-terminally directed antiserum SP 2 (Fig. 1). H P L C characterization of the SP-LI measured with this antiserum in the spinal cord superfusates collected at basal and stimulated conditions revealed three components (Fig. 2). The major immunoreactive component eluted in the position of SP, and a minor peak was found in the position of SP-sulfoxide (SP-ox). In addition to these two components there was a late component eluting in fractions 32-33, which corresponded to the elution position of the N-terminally extended SP previously identified in human CSF [8] and in neutral extracts of rat spinal cord [8,9]. This component was not detected using the N-terminally directed antiserum SP N1 (results not shown). A component detected with antiserum SP 2 (Fig. 3A) but not with SP N1 (not shown) and which eluted in fraction 32-33, could also be demonstrated in the spinal cord extracts. Treatment of this component (pooled fractions 30-33) with trypsin with subsequent HPLC/RIA analysis yielded a component which coeluted with SP (Fig. 3B). Incubation of this late eluting component under identical conditions but in the absence of trypsin, had no effect on this component and gave not rise to any immunoreactivity in the elution position of SP (not shown). When analysed separately, by combined H P L C and RIA, the overflow of the late eluting component of SP-LI, was found to be increased together with that of (undeca) SP in response to superfusion with Krebs-Ringer buffer containing 50 mM potassium (Figs. 1, 2 and 4). The second stimulation with potassium induced a smaller release of both components compared to the In'st stimulation (Fig. 1). The potassium-induced release of the late eluting component was found to be calcium dependent (Fig. 4). In the absence of calcium and presence of 7 mM magnesium, 50 mM potassium had no significant effect on the release of this component (Fig. 4). A
B
(Trypsin)
~-
SP-ox SP
SP
O
#
t
t
!1 2
~ "2_
o
;o
20
,
3'o
,
,o
!] o
" 10
2o
30
4u"
Fraction number
Fig. 3. HPLC of a rat spinal cord extract (A). The HPLC fractions were analysedwith the C-terminally directed antiserum(SP 2). Fractions correspondingto the late elutingpeak (indicatedin A) were collected followingchromatographyof another sample of the extract, treated with trypsin and rechromatographed (B). The detectionlimitwas 0.15 fmolSP-LI/HPLCfraction.The elutionpositions of syntheticSP and SP sult'oxide(SP-ox) are indicated. The acetonitrilegradient was the same as in Fig. 2.
44
c
E
0.8
o a.
m
0.6p < 0.001
"o "o f-
p
