- Email: [email protected]
Novel pyrazolo[1,5-a]pyridines with improved aqueous solubility as p110α-selective PI3 kinase inhibitors
Accepted Manuscript Novel Pyrazolo[1,5-a]pyridines with Improved Aqueous Solubility as p110αSelective PI3 Kinase Inhibitors Jackie D. Kendall, Anna C. Giddens, Kit Yee Tsang, Elaine S. Marshall, Claire L. Lill, Woo-Jeong Lee, Sharada Kolekar, Mindy Chao, Alisha Malik, Shuqiao Yu, Claire Chaussade, Christina Buchanan, Stephen M.F. Jamieson, Gordon W. Rewcastle, Bruce C. Baguley, William A. Denny, Peter R. Shepherd PII: DOI: Reference:
S0960-894X(16)31243-4 http://dx.doi.org/10.1016/j.bmcl.2016.11.078 BMCL 24473
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date: Revised Date: Accepted Date:
3 November 2016 23 November 2016 24 November 2016
Please cite this article as: Kendall, J.D., Giddens, A.C., Yee Tsang, K., Marshall, E.S., Lill, C.L., Lee, W-J., Kolekar, S., Chao, M., Malik, A., Yu, S., Chaussade, C., Buchanan, C., Jamieson, S.M.F., Rewcastle, G.W., Baguley, B.C., Denny, W.A., Shepherd, P.R., Novel Pyrazolo[1,5-a]pyridines with Improved Aqueous Solubility as p110αSelective PI3 Kinase Inhibitors, Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/ j.bmcl.2016.11.078
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Novel Pyrazolo[1,5-a]pyridines with Improved Aqueous Solubility as p110α-Selective PI3 Kinase Inhibitors Jackie D. Kendall,*,a,b Anna C. Giddens,a Kit Yee Tsang,a Elaine S. Marshall,a Claire L. Lill,c WooJeong Lee,c Sharada Kolekar,c Mindy Chao,c Alisha Malik,c Shuqiao Yu, c Claire Chaussade,b,c,d Christina Buchanan,b,c Stephen M.F. Jamieson,a,b Gordon W. Rewcastle,a,b Bruce C. Baguley,a,b William A. Denny,a,b Peter R. Shepherd.b,c a
Auckland Cancer Society Research Centre, School of Medical and Health Sciences, The University
of Auckland, Private Bag 92019, Auckland 1142, New Zealand. b
Maurice Wilkins Centre for Molecular Biodiscovery, The University of Auckland, Private Bag
92019, Auckland 1142, New Zealand. c
Department of Molecular Medicine and Pathology, School of Medical and Health Sciences, The
University of Auckland, Private Bag 92019, Auckland 1142, New Zealand. d
Present address: Galderma R&D, Nestle Skin Health, Sophia Antipolis, France.
*To whom correspondence should be addressed. E-mail: [email protected].
Abstract As part of our investigation into pyrazolo[1,5-a]pyridines as novel p110α selective PI3 kinase inhibitors, we report a range of analogues with improved aqueous solubility by the addition of a basic amine. The compounds demonstrated comparable p110α potency and selectivity to earlier compounds but with up to 1000x greater aqueous solubility, as the hydrochloride salts. The compounds also displayed good activity in a cellular assay of PI3 kinase activity.
Keywords PI3 kinase; PI3K; p110α; pyrazolo[1,5-a]pyridine; sulfonohydrazide; aqueous solubility.
Phosphoinositide-3-kinases (PI3 kinases) are lipid kinases which phosphorylate the 3’-hydroxyl group of phosphatidylinositol 4,5-diphosphate (PIP2) to phosphatidylinositol 3,4,5-triphosphate (PIP3). PIP3 subsequently recruits pleckstrin homology (PH) domain containing proteins, primarily protein kinase B (PKB, also known as Akt) to the cell membrane. Once recruited, PKB is phosphorylated and activated, leading to a cascade of cell signalling which control a range of cellular processes like cell proliferation, growth and survival.
The PI3 kinases are split into three sub-families (class I, II and III), and class I is further split into class Ia and Ib based upon their mechanism of activation and sequence homology. The class Ia PI3 kinases are heterodimeric, consisting of a catalytic subunit (p110α, p110β or p110δ) in complex with a regulatory subunit.1 PIK3CA, the gene encoding for p110α, is often over-expressed and mutated in many cancer types. Two of the most common of these mutations (E545K and H1047R) have been confirmed as activating mutations and hence increase levels of PIP3. 2 Mutations in p110β and p110δ are much less common.1 Inhibitors of PI3 kinase are widely regarded as an important new strategy in cancer treatment,3, 4 where they cover a wide range of structural types and isoform selectivity profiles. 5 The different isoforms each have a different purpose: p110α in solid tumours, p110β in thrombosis and p110δ in haematological cancers.5 Indeed the p110δ-selective inhibitor idelalisib (CAL-101), has now been approved for use in the clinic for relapsed chronic lymphocytic leukaemia (CLL), relapsed follicular B-cell non-Hodgkin lymphoma (NHL) and relapsed small lymphocytic leukaemia (SLL).6 However, few highly p110α-selective PI3 kinase inhibitors have been reported, notably A66 7 and closely related alpelisib, also known as BYL719 (Figure 1) 8 which is currently in phase II clinical trials. Alpelisib is showing promise in combination with aromatase inhibitor letrozole, where 9 out of 26 patients achieved a lack of disease progression at 6 months in a phase Ib study in breast cancer. 9 The majority of these patients had ER+/PIK3CA mutant breast cancer. These results suggest that the development of isoform selective PI3 kinase inhibitors is still of benefit in the clinic encouraged us to continue in our study of p110α-selective PI3 kinase inhibitors.
Figure 1. Structures of some p110α-selective PI3 kinase inhibitors.
In our previous papers we had made a series of pyrazolo[1,5-a]pyridine compounds and shown that they have good PI3 kinase potency and selectivity for the p110α isoform. 10, 11 In particular, compound 1 had a p110α IC50 of only 1.3 nM and selectivity of around 40-fold over p110β and p110δ. This compound also showed tumour growth inhibition in a HCT-116 human tumour xenograft model in mice.11 However, this compound was deemed not sufficiently water-soluble for further development.
In the current work we aimed to improve the aqueous solubility of these compounds and investigate whether this improvement would result in better efficacy in vivo.
The previous work showed us that the benzene ring was tolerant to substitution, and molecular modelling indicated that such substitutions would be pointing toward solvent, 11 hence we chose to incorporate a range of basic amine groups off that ring to improve aqueous solubility. The aldehyde derived hydrazones were all prepared starting from either aldehyde 2a or 2b10 according to Scheme 1. They were converted to fluorophenyl compounds 3a-c, 7a-b by our previously published method of condensation with methylhydrazine sulfate in the presence of 2,6-lutidine followed by addition of either a sulfonyl chloride or an acyl chloride. Substitution of the fluoride with alcohols was achieved by initial deprotonation with sodium hydride to make 4a-c, 5a, 6a, and with amines in THF or DMSO, to afford 4d-v, 5b-d, 6b-d, 8a-b.
Scheme 1. Reagents: i. MeHNNH 2.H2SO4, 2,6-lutidine, MeOH then arylsulfonyl chloride; ii. alcohol, NaH, THF; iii. amine, THF or DMSO; iv. MeHNNH 2.H2SO4, 2,6-lutidine, MeOH then acyl chloride.
The ketone derived hydrazones were made by a slightly different method, where methyl ketones 9a-b were reacted firstly with hydrazine hydrate followed by sulfonylation to afford 10a-b (Scheme 2). Next, fluoro substitution with amines gave 11a-f, and finally N-methylation was achieved by reaction with diazomethane solution to give 12a-f.
Scheme 2. Reagents: i. N2H4.H2O, MeOH, reflux; ii. 2-fluoro-5-nitrobenzenesulfonyl chloride, pyridine, CH2Cl2; iii. amine, THF; iv. CH2N2 solution in Et2O, THF. The compounds were all assayed against the three class I PI3 kinase isoforms: p110α, p110β and p110δ, as well as their effect on the cell proliferation in two early passage cell lines: NZB5 and NZOV9. The NZB5 cell line was derived from a brain tumour (medulloblastoma) and has the wildtype gene for p110α, and the NZOV9 cell line was derived from an ovarian tumour (poorly differentiated endometrioid adenocarcinoma) and contains a mutant p110α with a single amino acid substitution in the kinase domain (Y1021C).
The compounds with the amine side chain linked through either oxygen (4a-c) or a secondary amine (4d-f) all had moderate to good p110α potency and selectivity over the other isoforms, however the cell potency was markedly less. In contrast, the compounds with an NMe linker between the aryl ring and the amine side chain (4g-v) gave much better results against both p110α and in the cell proliferation assay. In general, the more basic alkyl amines (4g-4l) had slightly higher IC50s for p110α than the less basic morpholines (4m, n), pyridines (4o-t) and imidazoles (4u, v), although dimethylaminopropyl compound 4j is an exception to this with IC50 16 nM. All of these compounds demonstrated selectivity over the p110β and p110δ isoforms. The potency in the cell proliferation
assay gave more variable results however. Six compounds had IC 50 <10 nM in the cell line NZB5 (4j, 4r-v) although there were large differences observed between the two cell lines.
We had previously found that replacement of the aryl nitro group with cyano gave only slightly reduced potency against p110α,11 however in this instance, the comparison of 5a-d with their nitro analogues 4b, 4i, 4n and 4o, respectively, gave higher IC50s against p110α by between 2-fold and 20fold. Similarly, replacement of the cyano pyrazolo[1,5-a]pyridine substituent with bromo (6a-d) which we had previously shown was well tolerated10 gave higher p110α IC50s, although 6b was unexpectedly potent in the cell proliferation assay against both cell lines suggesting the possibility of off-target activity. Acyl hydrazides 8a and 8b were both much less potent than their corresponding sulfonyl hydrazide analogues 4i and 5b against p110α and both cell lines although 8a did demonstrate modest p110δ selectivity.
Finally, ketone hydrazones 11a-f and 12a-f all had poor activity in the cell proliferation assay even though the isolated enzyme potencies varied widely. Pyridine compounds 11c, 11f, 12c and 12f were the most potent of these compounds, and in particular 11f and 12c showed excellent selectivity for p110α over p110β (800-fold and 40-fold, respectively) and p110δ (2000-fold and 100-fold, respectively). The poor cell potency however suggests that these compounds may suffer from a lack of cell penetration.
Table 1. Inhibition of PI3 kinase isoforms and cell proliferation.
IC50a (nM) Cpd
Str
R1
R2
R3
1b
p110α
p110β
p110δ
NZB5
NZOV9
1.3
51
54
60
140
4a
A
CN
NO2
O(CH2)2NMe2
75
>1000
>1000
100
170
4b
A
CN
NO2
O(CH2)2(4-morph)
21
220
94
440
410
4c
A
CN
NO2
OCH2(2-pyridyl)
6.0
5600
950
1300
660
4d
A
CN
NO2
NH(CH2)2NMe2
430
>1000
>1000
750
640
4e
A
CN
NO2
NH(CH2)2(4-morph)
9
>1000
27
9400
>20000
4f
A
CN
NO2
NHCH2(2-pyridyl)
25
>1000
350
1100
680
4g
A
CN
NO2
NMe(CH2)2NH2
50
>1000
470
29
57
4h
A
CN
NO2
NMe(CH2)2NHMe
65
1000
280
24
11
4i
A
CN
NO2
NMe(CH2)2NMe2
46
670
230
49
55
4j
A
CN
NO2
NMe(CH2)3NMe2
16
76
100
4
25
4k
A
CN
NO2
NMe(CH2)2NEt2
75
330
66
500
340
4l
A
CN
NO2
NMe(CH2)2(1-piperid)
79
4200
450
92
49
4m
A
CN
NO2
NMe(CH2)2(4-morph)
19
1000
74
260
210
4n
A
CN
NO2
NMe(CH2)3(4-morph)
26
510
170
71
560
4o
A
CN
NO2
NMeCH2(2-pyridyl)
10
450
130
90
330
4p
A
CN
NO2
NMeCH2(3-pyridyl)
5.8
530
160
540
630
4q
A
CN
NO2
NMeCH2(4-pyridyl)
15
390
150
44
480
4r
A
CN
NO2
NMe(CH2)2(2-pyridyl)
9
410
60
5
410
4s
A
CN
NO2
NMe(CH2)2(3-pyridyl)
9
180
55
<2
740
4t
A
CN
NO2
NMe(CH2)2(4-pyridyl)
17
160
61
<2
250
4u
A
CN
NO2
NMe(CH2)2(1-imid)
31
150
100
5
45
4v
A
CN
NO2
NMe(CH2)2(4-imid)
18
740
88
3
57
5a
A
CN
CN
O(CH2)2(4-morph)
360
>1000
680
400
390
5b
A
CN
CN
NMe(CH2)2NMe2
81
2000
380
55
47
5c
A
CN
CN
NMe(CH2)3(4-morph)
240
>1000
370
21
360
5d
A
CN
CN
NMeCH2(2-pyridyl)
40
130
960
6a
A
Br
NO2
O(CH2)2(4-morph)
51
270
550
6b
A
Br
NO2
NMe(CH2)2NMe2
510
<2
<2
6c
A
Br
NO2
NMe(CH2)3(4-morph)
370
94
670
6d
A
Br
NO2
NMeCH2(2-pyridyl)
180
>1000
93
580
8a
B
CN
NO2
NMe(CH2)2NMe2
580
230
5600
3100
8b
B
CN
CN
NMe(CH2)2NMe2
630
2900
6200
11a
C
CN
H
NMe(CH2)2NMe2
51
460
2000
11b
C
CN
H
NMe(CH2)3(4-morph)
333
2400
9500
11c
C
CN
H
NMeCH2(2-pyridyl)
4.8
207
11
97
3200
11d
C
Br
H
NMe(CH2)2NMe2
208
>1000
>1000
420
420
11e
C
Br
H
NMe(CH2)3(4-morph)
448
900
3200
11f
C
Br
H
NMeCH2(2-pyridyl)
7.3
5600
292
96
2800
12a
C
CN
Me
NMe(CH2)2NMe2
328
>1000
658
1800
860
12b
C
CN
Me
NMe(CH2)3(4-morph)
>1000
3300
2400
12c
C
CN
Me
NMeCH2(2-pyridyl)
5.2
575
2200
2300
12d
C
Br
Me
NMe(CH2)2NMe2
>1000
>1000
3500
930
12e
C
Br
Me
NMe(CH2)3(4-morph)
>1000
5100
3800
12f
C
Br
Me
NMeCH2(2-pyridyl)
1040
2600
3000
a
2900
900
9400
All IC50 values are the mean of duplicate or triplicate measurements.
350
68
676
b
Data from ref.11
The most potent compounds against both p110α and at least one of the two cell lines were assessed for their aqueous solubility (Table 2). As hydrochloride salts, the three pyridine compounds 4r, 4s and 4t had the lowest solubility of those tested at <50 µg/mL. It should be noted however that these still show an improvement over compound 1 at only 1.2 µg/mL.11 Much better solubility was seen with dimethylamino compound 4j (1200 µg/mL) and imidazoles 4u and 4v (1000 µg/mL and 900 µg/mL, respectively).
Table 2. Aqueous solubility of selected compounds as their hydrochloride salts. Compound 1
a
Aqueous solubility (µg/mL)
a
1.2
4j
1200
4r
5.7
4s
24
4t
45
4u
1000
4v
900
Compound 1 is neutral and hence not a salt. Data from ref.11
The three most soluble compounds were then tested in an assay of their cellular potency by measuring the inhibition of phosphorylation of PKB at Ser473, a down-stream marker of PI3 kinase activity, in HCT-116 cells containing the H1047R mutation of p110α (Figure 1). The IC50s for 4j, 4u and 4v were 72 nM, 111 nM and 67 nM, respectively. This compares favourably with compound 1 with IC50 82 nM in the same assay.11
Figure 2. Inhibition of phosphorylation of PKB at Ser473 in HCT-116 cells.
Finally, compounds 4j and 4v were assessed for their tolerability in mice, and disappointingly both compounds were found to be more toxic than 1, and so were not evaluated further. This toxicity may indeed be related to the disparity between the potencies in the p110α enzyme assay and the cell proliferation assay. A comparison of the IC50s against p110α and the NZB5 and NZOV9 cell lines for 4j (0.25 and 1.6 for NZB5/p110α and NZOV9/p110α, respectively), 4v (0.17 and 3.2, respectively) and 1 (46 and 108, respectively) would suggest that 4j and 4v are behaving differently in cells to 1, and the lower IC50s for 4j and 4v in cells may be an indicator of off-target activity resulting in the observed toxicity. Clearly, there are a number of other factors operating in cells which are not fully understood.
In conclusion, we made a set of novel analogues of compound 1 and found that we could greatly improve the aqueous solubility by the addition of a basic amine group while retaining good PI3 kinase activity and p110α selectivity. Our earlier work had directed us to investigate substitution off the 2position of the benzene ring. We found this position to be tolerant of quite large basic substituents, which is consistent with the molecular modelling indicating that substituents at this position would point out towards solvent. Unfortunately, these more soluble compounds were not suitable for further evaluation in an in vivo efficacy model due to their toxicity.
Acknowledgements
The authors would like to thank Sisira Kumara for the aqueous solubility measurements, and Maruta Boyd for the NMR spectra. This work was funded by the Health Research Council of New Zealand (grant number 06/062), Maurice Wilkins Centre for Molecular Biodiscovery, and Pathway Therapeutics Inc.
Supplementary Data
Experimental details for the synthesis and characterisation of target compounds and their intermediates. Experimental details for all assays.
References
1.
Cui, W.; Cai, Y.; Zhou, X. Pathology 2014, 46, 169.
2.
Ciraolo, E.; Morello, F.; Hirsch, E. Curr. Med. Chem. 2011, 18, 2674.
3.
Thorpe, L. M.; Yuzugullu, H.; Zhao, J. J. Nat. Rev. Cancer 2015, 15, 7.
4.
Wang, X.; Ding, J.; Meng, L. Acta Pharmacol. Sin. 2015, 36, 1170.
5.
Sabbah, D. H.; Hu, J.; Zhong, H. A. Curr. Top. Med. Chem. 2016, 16, 1413.
6.
Markham, A. Drugs 2014, 74, 1701.
7.
Sun, M.; Hillmann, P.; Hofmann, B. T.; Hart, J. R.; Vogt, P. K. Proc. Natl. Acad. Sci. U. S. A.
2010, 107, 15547. 8.
Furet, P.; Guagnano, V.; Fairhurst, R. A.; Imbach-Weese, P.; Bruce, I.; Knapp, M.; Fritsch,
C.; Blasco, F.; Blanz, J.; Aichholz, R.; Hamon, J.; Fabbro, D.; Caravatti, G. Bioorg. Med. Chem. Lett. 2013, 23, 3741. 9.
Mayer, I. A.; Abramson, V.; Formisano, L.; Balko, J. M.; Estrada, M. V.; Sanders, M.; Juric,
D.; Solit, D.; Berger, M. F.; Won, H.; Li, Y.; Cantley, L. C.; Winer, E. P.; Arteaga, C. L. Clin. Cancer Res. 2016, http://dx.doi.org/10.1158/1078-0432.CCR-16-0134. 10.
Kendall, J. D.; O'Connor, P. D.; Marshall, A. J.; Frédérick, R.; Marshall, E. S.; Lill, C. L.;
Lee, W.-J.; Kolekar, S.; Chao, M.; Malik, A.; Yu, S.; Chaussade, C.; Buchanan, C.; Rewcastle, G. W.; Baguley, B. C.; Flanagan, J. U.; Jamieson, S. M. F.; Denny, W. A.; Shepherd, P. R. Bioorg. Med. Chem. 2012, 20, 69. 11.
Kendall, J. D.; Giddens, A. C.; Tsang, K. Y.; Frédérick, R.; Marshall, E. S.; Singh, R.; Lill, C.
L.; Lee, W.-J.; Kolekar, S.; Chao, M.; Malik, A.; Yu, S.; Chaussade, C.; Buchanan, C.; Rewcastle, G. W.; Baguley, B. C.; Flanagan, J. U.; Jamieson, S. M. F.; Denny, W. A.; Shepherd, P. R. Bioorg. Med. Chem. 2012, 20, 58.
Graphical abstract
S0960-894X(16)31243-4 http://dx.doi.org/10.1016/j.bmcl.2016.11.078 BMCL 24473
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date: Revised Date: Accepted Date:
3 November 2016 23 November 2016 24 November 2016
Please cite this article as: Kendall, J.D., Giddens, A.C., Yee Tsang, K., Marshall, E.S., Lill, C.L., Lee, W-J., Kolekar, S., Chao, M., Malik, A., Yu, S., Chaussade, C., Buchanan, C., Jamieson, S.M.F., Rewcastle, G.W., Baguley, B.C., Denny, W.A., Shepherd, P.R., Novel Pyrazolo[1,5-a]pyridines with Improved Aqueous Solubility as p110αSelective PI3 Kinase Inhibitors, Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/ j.bmcl.2016.11.078
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Novel Pyrazolo[1,5-a]pyridines with Improved Aqueous Solubility as p110α-Selective PI3 Kinase Inhibitors Jackie D. Kendall,*,a,b Anna C. Giddens,a Kit Yee Tsang,a Elaine S. Marshall,a Claire L. Lill,c WooJeong Lee,c Sharada Kolekar,c Mindy Chao,c Alisha Malik,c Shuqiao Yu, c Claire Chaussade,b,c,d Christina Buchanan,b,c Stephen M.F. Jamieson,a,b Gordon W. Rewcastle,a,b Bruce C. Baguley,a,b William A. Denny,a,b Peter R. Shepherd.b,c a
Auckland Cancer Society Research Centre, School of Medical and Health Sciences, The University
of Auckland, Private Bag 92019, Auckland 1142, New Zealand. b
Maurice Wilkins Centre for Molecular Biodiscovery, The University of Auckland, Private Bag
92019, Auckland 1142, New Zealand. c
Department of Molecular Medicine and Pathology, School of Medical and Health Sciences, The
University of Auckland, Private Bag 92019, Auckland 1142, New Zealand. d
Present address: Galderma R&D, Nestle Skin Health, Sophia Antipolis, France.
*To whom correspondence should be addressed. E-mail: [email protected].
Abstract As part of our investigation into pyrazolo[1,5-a]pyridines as novel p110α selective PI3 kinase inhibitors, we report a range of analogues with improved aqueous solubility by the addition of a basic amine. The compounds demonstrated comparable p110α potency and selectivity to earlier compounds but with up to 1000x greater aqueous solubility, as the hydrochloride salts. The compounds also displayed good activity in a cellular assay of PI3 kinase activity.
Keywords PI3 kinase; PI3K; p110α; pyrazolo[1,5-a]pyridine; sulfonohydrazide; aqueous solubility.
Phosphoinositide-3-kinases (PI3 kinases) are lipid kinases which phosphorylate the 3’-hydroxyl group of phosphatidylinositol 4,5-diphosphate (PIP2) to phosphatidylinositol 3,4,5-triphosphate (PIP3). PIP3 subsequently recruits pleckstrin homology (PH) domain containing proteins, primarily protein kinase B (PKB, also known as Akt) to the cell membrane. Once recruited, PKB is phosphorylated and activated, leading to a cascade of cell signalling which control a range of cellular processes like cell proliferation, growth and survival.
The PI3 kinases are split into three sub-families (class I, II and III), and class I is further split into class Ia and Ib based upon their mechanism of activation and sequence homology. The class Ia PI3 kinases are heterodimeric, consisting of a catalytic subunit (p110α, p110β or p110δ) in complex with a regulatory subunit.1 PIK3CA, the gene encoding for p110α, is often over-expressed and mutated in many cancer types. Two of the most common of these mutations (E545K and H1047R) have been confirmed as activating mutations and hence increase levels of PIP3. 2 Mutations in p110β and p110δ are much less common.1 Inhibitors of PI3 kinase are widely regarded as an important new strategy in cancer treatment,3, 4 where they cover a wide range of structural types and isoform selectivity profiles. 5 The different isoforms each have a different purpose: p110α in solid tumours, p110β in thrombosis and p110δ in haematological cancers.5 Indeed the p110δ-selective inhibitor idelalisib (CAL-101), has now been approved for use in the clinic for relapsed chronic lymphocytic leukaemia (CLL), relapsed follicular B-cell non-Hodgkin lymphoma (NHL) and relapsed small lymphocytic leukaemia (SLL).6 However, few highly p110α-selective PI3 kinase inhibitors have been reported, notably A66 7 and closely related alpelisib, also known as BYL719 (Figure 1) 8 which is currently in phase II clinical trials. Alpelisib is showing promise in combination with aromatase inhibitor letrozole, where 9 out of 26 patients achieved a lack of disease progression at 6 months in a phase Ib study in breast cancer. 9 The majority of these patients had ER+/PIK3CA mutant breast cancer. These results suggest that the development of isoform selective PI3 kinase inhibitors is still of benefit in the clinic encouraged us to continue in our study of p110α-selective PI3 kinase inhibitors.
Figure 1. Structures of some p110α-selective PI3 kinase inhibitors.
In our previous papers we had made a series of pyrazolo[1,5-a]pyridine compounds and shown that they have good PI3 kinase potency and selectivity for the p110α isoform. 10, 11 In particular, compound 1 had a p110α IC50 of only 1.3 nM and selectivity of around 40-fold over p110β and p110δ. This compound also showed tumour growth inhibition in a HCT-116 human tumour xenograft model in mice.11 However, this compound was deemed not sufficiently water-soluble for further development.
In the current work we aimed to improve the aqueous solubility of these compounds and investigate whether this improvement would result in better efficacy in vivo.
The previous work showed us that the benzene ring was tolerant to substitution, and molecular modelling indicated that such substitutions would be pointing toward solvent, 11 hence we chose to incorporate a range of basic amine groups off that ring to improve aqueous solubility. The aldehyde derived hydrazones were all prepared starting from either aldehyde 2a or 2b10 according to Scheme 1. They were converted to fluorophenyl compounds 3a-c, 7a-b by our previously published method of condensation with methylhydrazine sulfate in the presence of 2,6-lutidine followed by addition of either a sulfonyl chloride or an acyl chloride. Substitution of the fluoride with alcohols was achieved by initial deprotonation with sodium hydride to make 4a-c, 5a, 6a, and with amines in THF or DMSO, to afford 4d-v, 5b-d, 6b-d, 8a-b.
Scheme 1. Reagents: i. MeHNNH 2.H2SO4, 2,6-lutidine, MeOH then arylsulfonyl chloride; ii. alcohol, NaH, THF; iii. amine, THF or DMSO; iv. MeHNNH 2.H2SO4, 2,6-lutidine, MeOH then acyl chloride.
The ketone derived hydrazones were made by a slightly different method, where methyl ketones 9a-b were reacted firstly with hydrazine hydrate followed by sulfonylation to afford 10a-b (Scheme 2). Next, fluoro substitution with amines gave 11a-f, and finally N-methylation was achieved by reaction with diazomethane solution to give 12a-f.
Scheme 2. Reagents: i. N2H4.H2O, MeOH, reflux; ii. 2-fluoro-5-nitrobenzenesulfonyl chloride, pyridine, CH2Cl2; iii. amine, THF; iv. CH2N2 solution in Et2O, THF. The compounds were all assayed against the three class I PI3 kinase isoforms: p110α, p110β and p110δ, as well as their effect on the cell proliferation in two early passage cell lines: NZB5 and NZOV9. The NZB5 cell line was derived from a brain tumour (medulloblastoma) and has the wildtype gene for p110α, and the NZOV9 cell line was derived from an ovarian tumour (poorly differentiated endometrioid adenocarcinoma) and contains a mutant p110α with a single amino acid substitution in the kinase domain (Y1021C).
The compounds with the amine side chain linked through either oxygen (4a-c) or a secondary amine (4d-f) all had moderate to good p110α potency and selectivity over the other isoforms, however the cell potency was markedly less. In contrast, the compounds with an NMe linker between the aryl ring and the amine side chain (4g-v) gave much better results against both p110α and in the cell proliferation assay. In general, the more basic alkyl amines (4g-4l) had slightly higher IC50s for p110α than the less basic morpholines (4m, n), pyridines (4o-t) and imidazoles (4u, v), although dimethylaminopropyl compound 4j is an exception to this with IC50 16 nM. All of these compounds demonstrated selectivity over the p110β and p110δ isoforms. The potency in the cell proliferation
assay gave more variable results however. Six compounds had IC 50 <10 nM in the cell line NZB5 (4j, 4r-v) although there were large differences observed between the two cell lines.
We had previously found that replacement of the aryl nitro group with cyano gave only slightly reduced potency against p110α,11 however in this instance, the comparison of 5a-d with their nitro analogues 4b, 4i, 4n and 4o, respectively, gave higher IC50s against p110α by between 2-fold and 20fold. Similarly, replacement of the cyano pyrazolo[1,5-a]pyridine substituent with bromo (6a-d) which we had previously shown was well tolerated10 gave higher p110α IC50s, although 6b was unexpectedly potent in the cell proliferation assay against both cell lines suggesting the possibility of off-target activity. Acyl hydrazides 8a and 8b were both much less potent than their corresponding sulfonyl hydrazide analogues 4i and 5b against p110α and both cell lines although 8a did demonstrate modest p110δ selectivity.
Finally, ketone hydrazones 11a-f and 12a-f all had poor activity in the cell proliferation assay even though the isolated enzyme potencies varied widely. Pyridine compounds 11c, 11f, 12c and 12f were the most potent of these compounds, and in particular 11f and 12c showed excellent selectivity for p110α over p110β (800-fold and 40-fold, respectively) and p110δ (2000-fold and 100-fold, respectively). The poor cell potency however suggests that these compounds may suffer from a lack of cell penetration.
Table 1. Inhibition of PI3 kinase isoforms and cell proliferation.
IC50a (nM) Cpd
Str
R1
R2
R3
1b
p110α
p110β
p110δ
NZB5
NZOV9
1.3
51
54
60
140
4a
A
CN
NO2
O(CH2)2NMe2
75
>1000
>1000
100
170
4b
A
CN
NO2
O(CH2)2(4-morph)
21
220
94
440
410
4c
A
CN
NO2
OCH2(2-pyridyl)
6.0
5600
950
1300
660
4d
A
CN
NO2
NH(CH2)2NMe2
430
>1000
>1000
750
640
4e
A
CN
NO2
NH(CH2)2(4-morph)
9
>1000
27
9400
>20000
4f
A
CN
NO2
NHCH2(2-pyridyl)
25
>1000
350
1100
680
4g
A
CN
NO2
NMe(CH2)2NH2
50
>1000
470
29
57
4h
A
CN
NO2
NMe(CH2)2NHMe
65
1000
280
24
11
4i
A
CN
NO2
NMe(CH2)2NMe2
46
670
230
49
55
4j
A
CN
NO2
NMe(CH2)3NMe2
16
76
100
4
25
4k
A
CN
NO2
NMe(CH2)2NEt2
75
330
66
500
340
4l
A
CN
NO2
NMe(CH2)2(1-piperid)
79
4200
450
92
49
4m
A
CN
NO2
NMe(CH2)2(4-morph)
19
1000
74
260
210
4n
A
CN
NO2
NMe(CH2)3(4-morph)
26
510
170
71
560
4o
A
CN
NO2
NMeCH2(2-pyridyl)
10
450
130
90
330
4p
A
CN
NO2
NMeCH2(3-pyridyl)
5.8
530
160
540
630
4q
A
CN
NO2
NMeCH2(4-pyridyl)
15
390
150
44
480
4r
A
CN
NO2
NMe(CH2)2(2-pyridyl)
9
410
60
5
410
4s
A
CN
NO2
NMe(CH2)2(3-pyridyl)
9
180
55
<2
740
4t
A
CN
NO2
NMe(CH2)2(4-pyridyl)
17
160
61
<2
250
4u
A
CN
NO2
NMe(CH2)2(1-imid)
31
150
100
5
45
4v
A
CN
NO2
NMe(CH2)2(4-imid)
18
740
88
3
57
5a
A
CN
CN
O(CH2)2(4-morph)
360
>1000
680
400
390
5b
A
CN
CN
NMe(CH2)2NMe2
81
2000
380
55
47
5c
A
CN
CN
NMe(CH2)3(4-morph)
240
>1000
370
21
360
5d
A
CN
CN
NMeCH2(2-pyridyl)
40
130
960
6a
A
Br
NO2
O(CH2)2(4-morph)
51
270
550
6b
A
Br
NO2
NMe(CH2)2NMe2
510
<2
<2
6c
A
Br
NO2
NMe(CH2)3(4-morph)
370
94
670
6d
A
Br
NO2
NMeCH2(2-pyridyl)
180
>1000
93
580
8a
B
CN
NO2
NMe(CH2)2NMe2
580
230
5600
3100
8b
B
CN
CN
NMe(CH2)2NMe2
630
2900
6200
11a
C
CN
H
NMe(CH2)2NMe2
51
460
2000
11b
C
CN
H
NMe(CH2)3(4-morph)
333
2400
9500
11c
C
CN
H
NMeCH2(2-pyridyl)
4.8
207
11
97
3200
11d
C
Br
H
NMe(CH2)2NMe2
208
>1000
>1000
420
420
11e
C
Br
H
NMe(CH2)3(4-morph)
448
900
3200
11f
C
Br
H
NMeCH2(2-pyridyl)
7.3
5600
292
96
2800
12a
C
CN
Me
NMe(CH2)2NMe2
328
>1000
658
1800
860
12b
C
CN
Me
NMe(CH2)3(4-morph)
>1000
3300
2400
12c
C
CN
Me
NMeCH2(2-pyridyl)
5.2
575
2200
2300
12d
C
Br
Me
NMe(CH2)2NMe2
>1000
>1000
3500
930
12e
C
Br
Me
NMe(CH2)3(4-morph)
>1000
5100
3800
12f
C
Br
Me
NMeCH2(2-pyridyl)
1040
2600
3000
a
2900
900
9400
All IC50 values are the mean of duplicate or triplicate measurements.
350
68
676
b
Data from ref.11
The most potent compounds against both p110α and at least one of the two cell lines were assessed for their aqueous solubility (Table 2). As hydrochloride salts, the three pyridine compounds 4r, 4s and 4t had the lowest solubility of those tested at <50 µg/mL. It should be noted however that these still show an improvement over compound 1 at only 1.2 µg/mL.11 Much better solubility was seen with dimethylamino compound 4j (1200 µg/mL) and imidazoles 4u and 4v (1000 µg/mL and 900 µg/mL, respectively).
Table 2. Aqueous solubility of selected compounds as their hydrochloride salts. Compound 1
a
Aqueous solubility (µg/mL)
a
1.2
4j
1200
4r
5.7
4s
24
4t
45
4u
1000
4v
900
Compound 1 is neutral and hence not a salt. Data from ref.11
The three most soluble compounds were then tested in an assay of their cellular potency by measuring the inhibition of phosphorylation of PKB at Ser473, a down-stream marker of PI3 kinase activity, in HCT-116 cells containing the H1047R mutation of p110α (Figure 1). The IC50s for 4j, 4u and 4v were 72 nM, 111 nM and 67 nM, respectively. This compares favourably with compound 1 with IC50 82 nM in the same assay.11
Figure 2. Inhibition of phosphorylation of PKB at Ser473 in HCT-116 cells.
Finally, compounds 4j and 4v were assessed for their tolerability in mice, and disappointingly both compounds were found to be more toxic than 1, and so were not evaluated further. This toxicity may indeed be related to the disparity between the potencies in the p110α enzyme assay and the cell proliferation assay. A comparison of the IC50s against p110α and the NZB5 and NZOV9 cell lines for 4j (0.25 and 1.6 for NZB5/p110α and NZOV9/p110α, respectively), 4v (0.17 and 3.2, respectively) and 1 (46 and 108, respectively) would suggest that 4j and 4v are behaving differently in cells to 1, and the lower IC50s for 4j and 4v in cells may be an indicator of off-target activity resulting in the observed toxicity. Clearly, there are a number of other factors operating in cells which are not fully understood.
In conclusion, we made a set of novel analogues of compound 1 and found that we could greatly improve the aqueous solubility by the addition of a basic amine group while retaining good PI3 kinase activity and p110α selectivity. Our earlier work had directed us to investigate substitution off the 2position of the benzene ring. We found this position to be tolerant of quite large basic substituents, which is consistent with the molecular modelling indicating that substituents at this position would point out towards solvent. Unfortunately, these more soluble compounds were not suitable for further evaluation in an in vivo efficacy model due to their toxicity.
Acknowledgements
The authors would like to thank Sisira Kumara for the aqueous solubility measurements, and Maruta Boyd for the NMR spectra. This work was funded by the Health Research Council of New Zealand (grant number 06/062), Maurice Wilkins Centre for Molecular Biodiscovery, and Pathway Therapeutics Inc.
Supplementary Data
Experimental details for the synthesis and characterisation of target compounds and their intermediates. Experimental details for all assays.
References
1.
Cui, W.; Cai, Y.; Zhou, X. Pathology 2014, 46, 169.
2.
Ciraolo, E.; Morello, F.; Hirsch, E. Curr. Med. Chem. 2011, 18, 2674.
3.
Thorpe, L. M.; Yuzugullu, H.; Zhao, J. J. Nat. Rev. Cancer 2015, 15, 7.
4.
Wang, X.; Ding, J.; Meng, L. Acta Pharmacol. Sin. 2015, 36, 1170.
5.
Sabbah, D. H.; Hu, J.; Zhong, H. A. Curr. Top. Med. Chem. 2016, 16, 1413.
6.
Markham, A. Drugs 2014, 74, 1701.
7.
Sun, M.; Hillmann, P.; Hofmann, B. T.; Hart, J. R.; Vogt, P. K. Proc. Natl. Acad. Sci. U. S. A.
2010, 107, 15547. 8.
Furet, P.; Guagnano, V.; Fairhurst, R. A.; Imbach-Weese, P.; Bruce, I.; Knapp, M.; Fritsch,
C.; Blasco, F.; Blanz, J.; Aichholz, R.; Hamon, J.; Fabbro, D.; Caravatti, G. Bioorg. Med. Chem. Lett. 2013, 23, 3741. 9.
Mayer, I. A.; Abramson, V.; Formisano, L.; Balko, J. M.; Estrada, M. V.; Sanders, M.; Juric,
D.; Solit, D.; Berger, M. F.; Won, H.; Li, Y.; Cantley, L. C.; Winer, E. P.; Arteaga, C. L. Clin. Cancer Res. 2016, http://dx.doi.org/10.1158/1078-0432.CCR-16-0134. 10.
Kendall, J. D.; O'Connor, P. D.; Marshall, A. J.; Frédérick, R.; Marshall, E. S.; Lill, C. L.;
Lee, W.-J.; Kolekar, S.; Chao, M.; Malik, A.; Yu, S.; Chaussade, C.; Buchanan, C.; Rewcastle, G. W.; Baguley, B. C.; Flanagan, J. U.; Jamieson, S. M. F.; Denny, W. A.; Shepherd, P. R. Bioorg. Med. Chem. 2012, 20, 69. 11.
Kendall, J. D.; Giddens, A. C.; Tsang, K. Y.; Frédérick, R.; Marshall, E. S.; Singh, R.; Lill, C.
L.; Lee, W.-J.; Kolekar, S.; Chao, M.; Malik, A.; Yu, S.; Chaussade, C.; Buchanan, C.; Rewcastle, G. W.; Baguley, B. C.; Flanagan, J. U.; Jamieson, S. M. F.; Denny, W. A.; Shepherd, P. R. Bioorg. Med. Chem. 2012, 20, 58.
Graphical abstract