Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14 DOI 10.1007/s00210-013-0940-6
ORIGINAL ARTICLE
Identification of small molecule NPR-B antagonists by high throughput screening — potential use in heart failure Trond Bach & Stine Bergholtz & Jon Riise & Eirik Qvigstad & Tor Skomedal & Jan-Bjørn Osnes & Finn Olav Levy
Received: 5 July 2013 / Accepted: 12 November 2013 / Published online: 3 December 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract We found previously that stimulation of natriuretic peptide receptor (NPR)-B by C-type natriuretic peptide (CNP) in failing rat ventricle potentiates β1-adrenoceptor (β1-AR)mediated inotropic response to noradrenaline through cGMPmediated inhibition of phosphodiesterase (PDE) 3, thereby enhancing cAMP-mediated signalling. Increased cAMPmediated signalling is deleterious in chronic heart failure (HF; basis for the use of β-blockers in HF) and we propose to consider NPR-B antagonists as new HF treatment in addition to conventional therapy. Since there is no NPR-Bselective antagonist available for clinical studies, we aimed at identifying a novel small molecule (non-peptide) NPR-B antagonist. An assay was developed and high throughput screening performed on a chemical library of about 20,000 small molecule compounds (30 % NPR-B inhibition with various degrees of selectivity towards NPR-A (>5 % NPR-A inhibition). For the purpose of pharmacological comparison we chose to analyse hit compounds of two different groups; selective NPR-B inhibition (i.e. ≥25 % inhibition of NPR-B and ≤5 % inhibition of NPR-A) and partially selective NPR-B inhibition (i.e. compounds inhibiting the responses of both NPR-B and NPR-A, but preferentially NPR-B; Fig. 1).
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
Group 1 Selective NPR-B inhibition: C12
C17
C37
Group 2 Partially selective NPR-B inhibition: C10
C15
C18
Fig. 1 Molecular structures of the compounds tested. C12 (C22H14ClN3O5S), C17 (C22H29NO), C37 (C29H 33ClN2O 2), C10 (C17H19ClN2), C15 (C25H35NO3) and C18 (C22H27NO)
Noncompetitive and reversible inhibition of NPR-B response in HEK293 cells
mediated a similar 30 % reduction (C17, 29 %±4 % (SEM); C37, 33 %±4 %; C12, 31 %±4 %) of the maximal cGMP production without significantly changing the EC50 values for CNP (Fig. 2a, left panel). C17 and C37 showed little or no effect on the NPR-A response (Fig. 2a, right panel) but surprisingly, C12 amplified the effect of BNP stimulation of NPR-A (Fig. 2a, right panel). Since the effect was only seen at higher BNP concentrations, C12 is not an NPR-A agonist per se, but somehow amplifies the effect of BNP. The mechanism is still not clear. C37 turned out to be a prescribed drug, loperamide (an opioid agent reducing gastrointestinal motility and secretion, used against diarrhoea), and was included here to report an unexpected, novel effect of this prescribed drug (Fig. 2a).
The selective NPR-B inhibitors C17, C37 and C12 mediate a moderate reduction of the efficacy of CNP-induced cGMP production and no changes of the EC50 values
The partially selective NPR-B antagonists C15, C18 and C10 mediate a strong reduction of the maximum cGMP production and no changes of the EC50 values for CNP
We tested the three compounds that showed NPR-B selectivity (Fig. 1). The concentration-response curves of NPR-B and NPR-A stimulation with CNP and BNP, respectively, in the presence of 10 μM compounds are shown in Fig. 2. The selective NPR-B peptide antagonist, P19 (Deschenes et al. 2005), was included as a control antagonist in the NPR-B experiments. P19 shifted the concentration-response curves of NPR-B stimulation with CNP to higher concentrations, consistent with competitive antagonism (Deschenes et al. 2005). The three selective NPR-B antagonists C17, C37 and C12
The three partially selective NPR-B antagonists, C15, C18 and C10 (Fig. 1), were tested as explained above. C15 and C18 mediated a 55 %±3 % and 44 %±4 % reduction of the maximal NPR-B response, respectively, whereas only 23 %± 4 % and 15 %±4 % of the NPR-A response, respectively (Fig. 2b). C10 inhibited the NPR-B response strongly by 72 % ±3 % and the NPR-A response by 39 %±4 %. These results suggest a noncompetitive or an irreversible competitive inhibition by the compounds, since none of the compounds evoked a shift in the EC50 values of CNP.
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
A
120
Control C12 C17 C37 P19
100
cGMP (% of control)
Fig. 2 Concentration–response curves for cGMP production in HEK293 cell lines expressing NPR-B or NPR-A. cGMP production was evoked by CNP stimulation of NPR-B (a, b, left panels) and by BNP stimulation of NPR-A (a, b, right panels), in the presence or absence of 10 μM of compounds a C17, C37 and C12, and b C15, C18 and C10. P19 was used as a control antagonist when testing the NPRB response. Data are presented normalized to control (100 %). Data points are means±SEM (n =6)
9
Control C12 C17 C37
200
150
80
60
100 40 50
20
0
0
-10
-9
-8
-7
-6
-5
-10
-9
Log [CNP] (M)
B
120
cGMP (% of control)
-7
-6
-5
-6
-5
Log [BNP] (M)
120
Control C10 C15 C18 P19
100
-8
Control C10 C15 C18
100
80
80
60
60
40
40
20
20
0
0 -10
-9
-8
-7
Log [CNP] (M)
Reversibility of the effects To further characterize the nature of the inhibition by the compounds we tested whether the inhibition was reversible. Washout experiments were performed for compound C10, C15 and C18, showing the largest inhibition, and C37, showing NPR-B selectivity. The cells were first incubated for 20 min with the compounds and then subjected to a 2× 5 min and 1×10 min wash. As seen in Fig. 3a, b, the inhibitory effect of the compounds upon NPR-B response were nearly completely removed after washing. The results suggest that the compounds bind reversibly to some part of the receptors. Thus, the inhibition exerted by these compounds can be charaterized as noncompetitive and reversible. Testing the specificity of the compounds We tested whether the six compounds (Fig. 1) had unspecific effects on receptor systems (transfected 5-HT4 receptors,
-6
-5
-10
-9
-8
-7
Log [BNP] (M)
endogenous β-ARs and EP receptors in HEK293 cells) coupled to adenylyl cyclase as measured in an adenylyl cyclase enzyme activity assay. No effects of the compounds (tested at 10 μM) were observed on these receptor systems or adenylyl cyclase directly (not shown), hence no lack of selectivity was revealed in this respect. Neither did the compounds alone (in the absence of CNP for NPR-B cells or BNP for NPR-A cells) show any effects in the AlphaScreen assay (not shown). The potency of the compounds is independent of agonist concentration The potencies of the compounds (C10, C12, C15 and C17) were analysed at high (0.5 μM) and low (50 nM) concentrations of CNP (Fig. 4) and BNP (data not shown). For NPR-B, the IC50 value of each compound was about the same at high and low CNP concentrations as opposed to the expected and clear reduction of the IC50
10
A
B 120
120
Control C10 C10 wash C15 C15 wash
100
cGMP (% of control)
Fig. 3 Concentration–response curves for cGMP production in HEK293 cell lines expressing NPR-B, evoked by CNP stimulation of NPR-B in the presence or absence of 10 μM of compounds a C10 and C15, and b C18 and C37. Samples indicated “wash” were subjected to a 2×5 min and 1×10 min wash in stimulation buffer following incubation with antagonists. Cells used for the other samples also underwent the same washing procedure but before antagonists were added. Data are presented normalized to control (100 %). Data points are means±SEM (n =4–6)
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
Control C18 C18 wash C37 C37 wash
100
80
80
60
60
40
40
20
20
0
0
-10
-9
-8
-7
Log [CNP] (M)
of the competitive, reversible antagonist P19 at low agonist concentration (Fig. 4a, b). Thus, it is likely that the compounds do not compete with CNP for binding to the receptor, but bind outside the agonist binding site of NPRB. The effects of the compounds on NPR-A exhibited similar IC50 (or, for C12, actually EC50) at the two BNP concentrations (data not shown). These results are consistent with a noncompetitive antagonism.
The characteristics of the competitive NPR-B antagonist, P19, are not changed in the presence of the other compounds, suggesting different binding sites The competitive NPR-B antagonist P19 was added to NPRexpressing HEK293 cells at increasing concentrations. When NPR-B was stimulated at a fixed low concentration of CNP (50 nM), adding 15 μM of the non-peptide compounds did not significantly alter the IC50 of P19, but the responses to CNP were reduced in a proportional fashion (Fig. 5a). Thus, the compounds did not compete with P19 for receptor binding and must bind at a different site than P19. Since P19 is a competitive antagonist of NPR-B, these results indicate that the compounds bind outside the agonist binding site of the receptor. Even though P19 was published as a selective NPR-B antagonist (Deschenes et al. 2005), our experiments showed that P19 stimulated NPR-A in HEK293 cells in the same concentration range as it inhibited NPR-B (Fig. 5b). Thus, P19 is an NPR-A agonist. Whether the non-peptide compounds bind to NPR-A at a different site than P19 is less clear from these data since EC50 cannot be accurately determined, although the compounds seem to exert variable shifts of the concentration-response curve of P19 (Fig. 5b) .
-6
-5
-10
-9
-8
-7
-6
-5
Log [CNP] (M)
Compound C10 inhibits the potentiating effect of NPR-B stimulation on β1-AR-mediated inotropic response in heart muscle Stimulation of NPR-B by CNP has been shown to potentiate β1-AR-mediated inotropic response to noradrenaline in rat heart left ventricle strips and a low degree β1-AR stimulation demasks a marked inotropic effect of CNP (Qvigstad et al. 2010). To test if our hit compounds would reverse this cardioexcitatory effect of NPR-B stimulation, we studied the contractile response in rat myocardial strips in the presence of C10. C10 was chosen for its strong inhibitory effect of the NPR-B response in HEK293 cells. The heart muscle strips were treated with a submaximal concentration of noradrenaline (40 nM) in the absence or presence of 10 μM of C10 and an increasing concentration of CNP was then added (Fig. 6a). Contractility was measured following stabilisation of the response to each concentration of CNP. The concentration–response curve of CNP was clearly shifted in a parallel manner to higher concentrations in the presence of C10 (Fig. 6b). The results indicate that C10 inhibits activation of the NPR-B receptor in a functional heart preparation. The compound alone had no effect on the contractile response (not shown). The compound did not significantly reduce the maximal response to CNP in muscle strips, which may seem to be different from the result in HEK293 cells. Thus, in myocardial tissue the attenuated stimulation of NPR-B still permitted full inotropic effects by increasing concentrations of CNP in the presence of low grade β1-AR stimulation. This may be explained by a relatively high susceptibility of PDE3 to be inhibited by cGMP (see the “Discussion and conclusions”). The result indicates that C10 has an antagonistic effect upon NPR-B also in a functional heart muscle preparation. Furthermore, the result suggests that the compound mediates the effect through an NPR-B receptor specific antagonism.
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
cGMP (% of 0.5 µM CNP stimulation)
A
11
The latter finding could be of clinical importance, since NPRB stimulation was previously found to potentiate β1-ARmediated inotropic response to noradrenaline in left ventricle (Qvigstad et al. 2010).
120
100
80
Selective noncompetitive, reversible inhibition of NPR-B vs NPR-A
60
40 + + + + +
20
P19 C10 C12 C15 C17
0 -8
-7
-6
-5
-4
In the screen we identified three compounds that showed selective antagonism of NPR-B vs NPR-A (C17, C37 and C12). These compounds mediated a similar reduction of the maximal NPR-B response (∼30 %), without significantly changing the EC50 values for CNP (Fig. 2a). The three partially selective NPR-B antagonists C15, C18 and C10
Log [antagonist] (M)
A
120
120
cGMP (% of 50 nM CNP stimulation)
cGMP (% of 50 nM CNP stimulation)
B
100
80
60
40 + + + + +
20
P19 C10 C12 C15 C17
100 No compound + C10 + C12 + C15 + C17 + C18 + C37
80
60
40
20
0
0
-8
-7
-6
-5
-7
Log [antagonist] (M)
Fig. 4 Inhibition of NPR-B response (cGMP production) in HEK293 cell lines expressing NPR-B, in the presence of increasing concentrations of antagonists (P19, C10, C12, C15 and C17), at a fixed high (0.5 μM; panel a) or low (50 nM; panel b) agonist (CNP) concentration. All data points were normalized to a 100 % maximal [cGMP] value, estimated from samples stimulated with 0.5 μM or 50 nM CNP only (a and b, respectively), indicated by the dotted horizontal lines. Data points are means±SEM (n =6)
Discussion and conclusions High throughput screening of a library of about 20,000 small molecule compounds revealed several potential NPR-B antagonists, based on their ability to inhibit CNP-stimulated cGMP production in an NPR-B-expressing HEK293 cell line. Six compounds were selected for further characterization of which three showed selective NPR-B inhibition and three showed partially selective inhibition of NPR-B vs NPR-A. The compounds mediated noncompetitive, reversible inhibition of the NPR-B response and most likely bind to allosteric sites outside the agonist binding site of NPR-B. We also show that compound C10 inhibited the NPR-B response both in HEK293 cells and in a functional heart muscle preparation.
-6
-5
-4
-5
-4
Log [P19] (M)
-4
B cGMP (% of 50 nM BNP stimulation)
-8
250
No compound + C10 + C12 + C15 + C17 + C18 + C37
200
150
100
50
0 -8
-7
-6
Log [P19] (M)
Fig. 5 Effects on cGMP production in HEK293 cell lines expressing NPR-B and NPR-A receptors (panel a and b, respectively) by the presence of 15 μM of the indicated compounds and increasing concentrations of the competitive antagonist P19, at a fixed, low agonist concentration (50 nM CNP in a, 50 nM BNP in b). All data points were normalized to a 100 % maximal [cGMP] value, estimated from samples stimulated with 50 nM CNP or 50 nM BNP only (a and b, respectively), indicated by the dotted horizontal lines. Data points are means±SEM (n =4–6)
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
A
Contractility ((dF/dt)max, arbitrary scale)
12
i)
_ _ ii)
B
Log [CNP] (M)
Fig. 6 a Original tracings from contracting rat myocardial muscle strips (representative experiment); the potentiating effect of NPR-B stimulation on β1-AR-mediated inotropic response was inhibited by compound C10. Strips were exposed to compound C10 (10 μM; ii) or DMSO control (i) prior to a submaximal dose of noradrenaline (NA, 40 nM). Note that the Y axes of (i) and (ii) are independent. After stabilization of the β1-ARmediated inotropic response to NA, cumulative doses of CNP were added (log (M), total concentrations). All additions are indicated by arrows. The maximal response of the control strip was reached after addition of the
fourth dose of CNP (−7.25), whereas the C10-treated strip reached maximum after the sixth dose of CNP (−6.25). Timolol (β-AR antagonist) was finally added to both strips, effectively reversing the contractility responses. b Normalized concentration-response curves for CNP on β1AR-mediated inotropic response to noradrenaline (NA, 40 nM). The concentration–response curve to CNP (filled circles) was shifted in a parallel manner to higher concentrations in the presence of C10 (open circles). Asterisk represents statistical significance with Student’s t test (p
ORIGINAL ARTICLE
Identification of small molecule NPR-B antagonists by high throughput screening — potential use in heart failure Trond Bach & Stine Bergholtz & Jon Riise & Eirik Qvigstad & Tor Skomedal & Jan-Bjørn Osnes & Finn Olav Levy
Received: 5 July 2013 / Accepted: 12 November 2013 / Published online: 3 December 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract We found previously that stimulation of natriuretic peptide receptor (NPR)-B by C-type natriuretic peptide (CNP) in failing rat ventricle potentiates β1-adrenoceptor (β1-AR)mediated inotropic response to noradrenaline through cGMPmediated inhibition of phosphodiesterase (PDE) 3, thereby enhancing cAMP-mediated signalling. Increased cAMPmediated signalling is deleterious in chronic heart failure (HF; basis for the use of β-blockers in HF) and we propose to consider NPR-B antagonists as new HF treatment in addition to conventional therapy. Since there is no NPR-Bselective antagonist available for clinical studies, we aimed at identifying a novel small molecule (non-peptide) NPR-B antagonist. An assay was developed and high throughput screening performed on a chemical library of about 20,000 small molecule compounds (30 % NPR-B inhibition with various degrees of selectivity towards NPR-A (>5 % NPR-A inhibition). For the purpose of pharmacological comparison we chose to analyse hit compounds of two different groups; selective NPR-B inhibition (i.e. ≥25 % inhibition of NPR-B and ≤5 % inhibition of NPR-A) and partially selective NPR-B inhibition (i.e. compounds inhibiting the responses of both NPR-B and NPR-A, but preferentially NPR-B; Fig. 1).
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
Group 1 Selective NPR-B inhibition: C12
C17
C37
Group 2 Partially selective NPR-B inhibition: C10
C15
C18
Fig. 1 Molecular structures of the compounds tested. C12 (C22H14ClN3O5S), C17 (C22H29NO), C37 (C29H 33ClN2O 2), C10 (C17H19ClN2), C15 (C25H35NO3) and C18 (C22H27NO)
Noncompetitive and reversible inhibition of NPR-B response in HEK293 cells
mediated a similar 30 % reduction (C17, 29 %±4 % (SEM); C37, 33 %±4 %; C12, 31 %±4 %) of the maximal cGMP production without significantly changing the EC50 values for CNP (Fig. 2a, left panel). C17 and C37 showed little or no effect on the NPR-A response (Fig. 2a, right panel) but surprisingly, C12 amplified the effect of BNP stimulation of NPR-A (Fig. 2a, right panel). Since the effect was only seen at higher BNP concentrations, C12 is not an NPR-A agonist per se, but somehow amplifies the effect of BNP. The mechanism is still not clear. C37 turned out to be a prescribed drug, loperamide (an opioid agent reducing gastrointestinal motility and secretion, used against diarrhoea), and was included here to report an unexpected, novel effect of this prescribed drug (Fig. 2a).
The selective NPR-B inhibitors C17, C37 and C12 mediate a moderate reduction of the efficacy of CNP-induced cGMP production and no changes of the EC50 values
The partially selective NPR-B antagonists C15, C18 and C10 mediate a strong reduction of the maximum cGMP production and no changes of the EC50 values for CNP
We tested the three compounds that showed NPR-B selectivity (Fig. 1). The concentration-response curves of NPR-B and NPR-A stimulation with CNP and BNP, respectively, in the presence of 10 μM compounds are shown in Fig. 2. The selective NPR-B peptide antagonist, P19 (Deschenes et al. 2005), was included as a control antagonist in the NPR-B experiments. P19 shifted the concentration-response curves of NPR-B stimulation with CNP to higher concentrations, consistent with competitive antagonism (Deschenes et al. 2005). The three selective NPR-B antagonists C17, C37 and C12
The three partially selective NPR-B antagonists, C15, C18 and C10 (Fig. 1), were tested as explained above. C15 and C18 mediated a 55 %±3 % and 44 %±4 % reduction of the maximal NPR-B response, respectively, whereas only 23 %± 4 % and 15 %±4 % of the NPR-A response, respectively (Fig. 2b). C10 inhibited the NPR-B response strongly by 72 % ±3 % and the NPR-A response by 39 %±4 %. These results suggest a noncompetitive or an irreversible competitive inhibition by the compounds, since none of the compounds evoked a shift in the EC50 values of CNP.
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
A
120
Control C12 C17 C37 P19
100
cGMP (% of control)
Fig. 2 Concentration–response curves for cGMP production in HEK293 cell lines expressing NPR-B or NPR-A. cGMP production was evoked by CNP stimulation of NPR-B (a, b, left panels) and by BNP stimulation of NPR-A (a, b, right panels), in the presence or absence of 10 μM of compounds a C17, C37 and C12, and b C15, C18 and C10. P19 was used as a control antagonist when testing the NPRB response. Data are presented normalized to control (100 %). Data points are means±SEM (n =6)
9
Control C12 C17 C37
200
150
80
60
100 40 50
20
0
0
-10
-9
-8
-7
-6
-5
-10
-9
Log [CNP] (M)
B
120
cGMP (% of control)
-7
-6
-5
-6
-5
Log [BNP] (M)
120
Control C10 C15 C18 P19
100
-8
Control C10 C15 C18
100
80
80
60
60
40
40
20
20
0
0 -10
-9
-8
-7
Log [CNP] (M)
Reversibility of the effects To further characterize the nature of the inhibition by the compounds we tested whether the inhibition was reversible. Washout experiments were performed for compound C10, C15 and C18, showing the largest inhibition, and C37, showing NPR-B selectivity. The cells were first incubated for 20 min with the compounds and then subjected to a 2× 5 min and 1×10 min wash. As seen in Fig. 3a, b, the inhibitory effect of the compounds upon NPR-B response were nearly completely removed after washing. The results suggest that the compounds bind reversibly to some part of the receptors. Thus, the inhibition exerted by these compounds can be charaterized as noncompetitive and reversible. Testing the specificity of the compounds We tested whether the six compounds (Fig. 1) had unspecific effects on receptor systems (transfected 5-HT4 receptors,
-6
-5
-10
-9
-8
-7
Log [BNP] (M)
endogenous β-ARs and EP receptors in HEK293 cells) coupled to adenylyl cyclase as measured in an adenylyl cyclase enzyme activity assay. No effects of the compounds (tested at 10 μM) were observed on these receptor systems or adenylyl cyclase directly (not shown), hence no lack of selectivity was revealed in this respect. Neither did the compounds alone (in the absence of CNP for NPR-B cells or BNP for NPR-A cells) show any effects in the AlphaScreen assay (not shown). The potency of the compounds is independent of agonist concentration The potencies of the compounds (C10, C12, C15 and C17) were analysed at high (0.5 μM) and low (50 nM) concentrations of CNP (Fig. 4) and BNP (data not shown). For NPR-B, the IC50 value of each compound was about the same at high and low CNP concentrations as opposed to the expected and clear reduction of the IC50
10
A
B 120
120
Control C10 C10 wash C15 C15 wash
100
cGMP (% of control)
Fig. 3 Concentration–response curves for cGMP production in HEK293 cell lines expressing NPR-B, evoked by CNP stimulation of NPR-B in the presence or absence of 10 μM of compounds a C10 and C15, and b C18 and C37. Samples indicated “wash” were subjected to a 2×5 min and 1×10 min wash in stimulation buffer following incubation with antagonists. Cells used for the other samples also underwent the same washing procedure but before antagonists were added. Data are presented normalized to control (100 %). Data points are means±SEM (n =4–6)
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
Control C18 C18 wash C37 C37 wash
100
80
80
60
60
40
40
20
20
0
0
-10
-9
-8
-7
Log [CNP] (M)
of the competitive, reversible antagonist P19 at low agonist concentration (Fig. 4a, b). Thus, it is likely that the compounds do not compete with CNP for binding to the receptor, but bind outside the agonist binding site of NPRB. The effects of the compounds on NPR-A exhibited similar IC50 (or, for C12, actually EC50) at the two BNP concentrations (data not shown). These results are consistent with a noncompetitive antagonism.
The characteristics of the competitive NPR-B antagonist, P19, are not changed in the presence of the other compounds, suggesting different binding sites The competitive NPR-B antagonist P19 was added to NPRexpressing HEK293 cells at increasing concentrations. When NPR-B was stimulated at a fixed low concentration of CNP (50 nM), adding 15 μM of the non-peptide compounds did not significantly alter the IC50 of P19, but the responses to CNP were reduced in a proportional fashion (Fig. 5a). Thus, the compounds did not compete with P19 for receptor binding and must bind at a different site than P19. Since P19 is a competitive antagonist of NPR-B, these results indicate that the compounds bind outside the agonist binding site of the receptor. Even though P19 was published as a selective NPR-B antagonist (Deschenes et al. 2005), our experiments showed that P19 stimulated NPR-A in HEK293 cells in the same concentration range as it inhibited NPR-B (Fig. 5b). Thus, P19 is an NPR-A agonist. Whether the non-peptide compounds bind to NPR-A at a different site than P19 is less clear from these data since EC50 cannot be accurately determined, although the compounds seem to exert variable shifts of the concentration-response curve of P19 (Fig. 5b) .
-6
-5
-10
-9
-8
-7
-6
-5
Log [CNP] (M)
Compound C10 inhibits the potentiating effect of NPR-B stimulation on β1-AR-mediated inotropic response in heart muscle Stimulation of NPR-B by CNP has been shown to potentiate β1-AR-mediated inotropic response to noradrenaline in rat heart left ventricle strips and a low degree β1-AR stimulation demasks a marked inotropic effect of CNP (Qvigstad et al. 2010). To test if our hit compounds would reverse this cardioexcitatory effect of NPR-B stimulation, we studied the contractile response in rat myocardial strips in the presence of C10. C10 was chosen for its strong inhibitory effect of the NPR-B response in HEK293 cells. The heart muscle strips were treated with a submaximal concentration of noradrenaline (40 nM) in the absence or presence of 10 μM of C10 and an increasing concentration of CNP was then added (Fig. 6a). Contractility was measured following stabilisation of the response to each concentration of CNP. The concentration–response curve of CNP was clearly shifted in a parallel manner to higher concentrations in the presence of C10 (Fig. 6b). The results indicate that C10 inhibits activation of the NPR-B receptor in a functional heart preparation. The compound alone had no effect on the contractile response (not shown). The compound did not significantly reduce the maximal response to CNP in muscle strips, which may seem to be different from the result in HEK293 cells. Thus, in myocardial tissue the attenuated stimulation of NPR-B still permitted full inotropic effects by increasing concentrations of CNP in the presence of low grade β1-AR stimulation. This may be explained by a relatively high susceptibility of PDE3 to be inhibited by cGMP (see the “Discussion and conclusions”). The result indicates that C10 has an antagonistic effect upon NPR-B also in a functional heart muscle preparation. Furthermore, the result suggests that the compound mediates the effect through an NPR-B receptor specific antagonism.
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
cGMP (% of 0.5 µM CNP stimulation)
A
11
The latter finding could be of clinical importance, since NPRB stimulation was previously found to potentiate β1-ARmediated inotropic response to noradrenaline in left ventricle (Qvigstad et al. 2010).
120
100
80
Selective noncompetitive, reversible inhibition of NPR-B vs NPR-A
60
40 + + + + +
20
P19 C10 C12 C15 C17
0 -8
-7
-6
-5
-4
In the screen we identified three compounds that showed selective antagonism of NPR-B vs NPR-A (C17, C37 and C12). These compounds mediated a similar reduction of the maximal NPR-B response (∼30 %), without significantly changing the EC50 values for CNP (Fig. 2a). The three partially selective NPR-B antagonists C15, C18 and C10
Log [antagonist] (M)
A
120
120
cGMP (% of 50 nM CNP stimulation)
cGMP (% of 50 nM CNP stimulation)
B
100
80
60
40 + + + + +
20
P19 C10 C12 C15 C17
100 No compound + C10 + C12 + C15 + C17 + C18 + C37
80
60
40
20
0
0
-8
-7
-6
-5
-7
Log [antagonist] (M)
Fig. 4 Inhibition of NPR-B response (cGMP production) in HEK293 cell lines expressing NPR-B, in the presence of increasing concentrations of antagonists (P19, C10, C12, C15 and C17), at a fixed high (0.5 μM; panel a) or low (50 nM; panel b) agonist (CNP) concentration. All data points were normalized to a 100 % maximal [cGMP] value, estimated from samples stimulated with 0.5 μM or 50 nM CNP only (a and b, respectively), indicated by the dotted horizontal lines. Data points are means±SEM (n =6)
Discussion and conclusions High throughput screening of a library of about 20,000 small molecule compounds revealed several potential NPR-B antagonists, based on their ability to inhibit CNP-stimulated cGMP production in an NPR-B-expressing HEK293 cell line. Six compounds were selected for further characterization of which three showed selective NPR-B inhibition and three showed partially selective inhibition of NPR-B vs NPR-A. The compounds mediated noncompetitive, reversible inhibition of the NPR-B response and most likely bind to allosteric sites outside the agonist binding site of NPR-B. We also show that compound C10 inhibited the NPR-B response both in HEK293 cells and in a functional heart muscle preparation.
-6
-5
-4
-5
-4
Log [P19] (M)
-4
B cGMP (% of 50 nM BNP stimulation)
-8
250
No compound + C10 + C12 + C15 + C17 + C18 + C37
200
150
100
50
0 -8
-7
-6
Log [P19] (M)
Fig. 5 Effects on cGMP production in HEK293 cell lines expressing NPR-B and NPR-A receptors (panel a and b, respectively) by the presence of 15 μM of the indicated compounds and increasing concentrations of the competitive antagonist P19, at a fixed, low agonist concentration (50 nM CNP in a, 50 nM BNP in b). All data points were normalized to a 100 % maximal [cGMP] value, estimated from samples stimulated with 50 nM CNP or 50 nM BNP only (a and b, respectively), indicated by the dotted horizontal lines. Data points are means±SEM (n =4–6)
Naunyn-Schmiedeberg's Arch Pharmacol (2014) 387:5–14
A
Contractility ((dF/dt)max, arbitrary scale)
12
i)
_ _ ii)
B
Log [CNP] (M)
Fig. 6 a Original tracings from contracting rat myocardial muscle strips (representative experiment); the potentiating effect of NPR-B stimulation on β1-AR-mediated inotropic response was inhibited by compound C10. Strips were exposed to compound C10 (10 μM; ii) or DMSO control (i) prior to a submaximal dose of noradrenaline (NA, 40 nM). Note that the Y axes of (i) and (ii) are independent. After stabilization of the β1-ARmediated inotropic response to NA, cumulative doses of CNP were added (log (M), total concentrations). All additions are indicated by arrows. The maximal response of the control strip was reached after addition of the
fourth dose of CNP (−7.25), whereas the C10-treated strip reached maximum after the sixth dose of CNP (−6.25). Timolol (β-AR antagonist) was finally added to both strips, effectively reversing the contractility responses. b Normalized concentration-response curves for CNP on β1AR-mediated inotropic response to noradrenaline (NA, 40 nM). The concentration–response curve to CNP (filled circles) was shifted in a parallel manner to higher concentrations in the presence of C10 (open circles). Asterisk represents statistical significance with Student’s t test (p
