Design, Synthesis, and Development of pyrazolo[1,5-a]pyrimidine Derivatives as a Novel Series of Selective PI3Kδ Inhibitors: Part I—Indole Derivatives

Phosphoinositide 3-kinase δ (PI3Kδ), a member of the class I PI3K family, is an essential signaling biomolecule that regulates the differentiation, proliferation, migration, and survival of immune cells. The overactivity of this protein causes cellular dysfunctions in many human disorders, for example, inflammatory and autoimmune diseases, including asthma or chronic obstructive pulmonary disease (COPD). In this work, we designed and synthesized a new library of small-molecule inhibitors based on indol-4-yl-pyrazolo[1,5-a]pyrimidine with IC50 values in the low nanomolar range and high selectivity against the PI3Kδ isoform. CPL302253 (54), the most potent compound of all the structures obtained, with IC50 = 2.8 nM, is a potential future candidate for clinical development as an inhaled drug to prevent asthma.


Introduction
PI3Ks (phosphoinositide 3-kinases) are a family of lipid kinases that can perform the phosphorylation reaction of the hydroxyl group at the 3-position of the phosphatidylinositol ring. More specifically, they are capable of catalyzing the phosphorylation reaction of 4,5-phosphatidylinositol diphosphate (PIP2) to 3,4,5-phosphatidylinositol triphosphate (PIP3) [1][2][3]. This family of kinases consists of three classes (I, II, and III) in terms of the structure and affinity for the substrate. Most class Is of PI3Ks have been described in the literature. PI3K I consist of heterodimeric proteins: PI3Kα, PI3Kβ, PI3Kγ, and PI3Kδ [1][2][3][4]. Each of them is involved in different functions and cellular processes, such as proliferation, migration, cytokine production, or apoptosis [1][2][3][4]. Cells involved in the body's immune response, such as macrophages, neutrophils, T, and B cells, highly expressed PI3Kγ and PI3Kδ [1][2][3][4][5]. The role of PI3Kδ as the co-stimulator between T to B cell interactions was also reported [6,7]. In addition, two other subunits, PI3Kα and PI3Kβ, are involved in normal embryogenesis or metabolism regulation. Therefore, PI3Kδ has been identified as was also reported [6,7]. In addition, two other subunits, PI3Kα and PI3Kβ, are involved in normal embryogenesis or metabolism regulation. Therefore, PI3Kδ has been identified as an attractive and promising therapeutic target for the treatment of cancer, autoimmune and inflammatory diseases [8][9][10][11][12][13][14].
One of the manifestations of inflammatory diseases is asthma, a chronic illness with a spectrum of respiratory symptoms burdensome for patients [15][16][17]. It was reported that PI3Kδ is involved in the regulation of allergic asthma development processes, such as activation of cytokines expression by Th2 cells, activation of antibodies production (e.g., IgE) by B cells, activation of basophils, and accumulation following the migration of eosinophil in the lungs [2,15,18]. Thus far, several selective PI3Kδ inhibitors have been developed, to name only: Idelalisib (PI3Kδ selective) or Duvelisib (PI3Kδ and γ selective; Figure 1) [15,[19][20][21]. Unfortunately, the toxicity and side effects caused by these candidates' low selectivity in systemic action exclude them from the group of potential future therapeutics for asthma management [15,22,23]. Therefore, new approaches focused on developing safe, selective PI3Kδ inhibitors designed to be conveniently delivered by inhalation remain an unfulfilled challenge [15,23]. Rich expression of PI3Kδ by lung epithelial cells provides the rationale for the new drug design against asthma as the alternative for patients poorly responding to current treatments.  The therapeutic application of PI3Kδ inhibition at the molecular level utilizes particular interactions of the respective inhibitors within the p110δ subunit of the ATP binding site [24,25]. Several binding protein key sites are involved in this mechanism: the affinity pocket, the hinge pocket, and a hydrophobic region located below the non-conserved part of the enzyme's active site [25][26][27]. Numerous active PI3Kδ inhibitors are characterized by the interactions with a conserved tyrosine residue (Tyr-876) and hydrogen bonds with Lys-833 located at the binding pocket [27,28]. Most selective PI3Kδ inhibitors, however, form a specific hydrogen bond between two critical amino acids: Trp-760 and Met-752 [24,28,29]. In addition, opening the pocket between the Trp-812 and Met-804 has been identified as a selectivity improvement operation [25]. Moreover, PI3Kδ selectivity strongly depends on the interaction with Trp-760, for which a 'tryptophan shelf' term was coined [6,24,25]. Binding to Asp-787 was also observed.
Most of the pan-PI3K inhibitors hold in their molecular structure bicyclic cores such as thienopyrimidines (GDC-0941), purines, pyridopyrimidines, or furopyrimidines ( Figure 1) [6,27]. The enormous activity and selectivity potential have been associated with the presence of the morpholine ring in the "morpholine-pyrimidine" system (marked in red in Figure 1) [6]. In the hinge-binding mechanism motif, the morpholine ring plays a role as an H-bond acceptor. The heteroaromatic or aromatic ring (marked in green in Figure 1), placed in a "meta"-like position to the morpholine ring, takes up space within the affinity pocket of the enzyme (binding to Val-828) [6,25,27]. This mutual interaction enhances the activity and selectivity of designed inhibitors. Moreover, the heterocyclic system (marked in blue in Figure 1) occupying the pocket responsible for the kinase's specificity drives the selectivity of the designed compounds [6,25,27].
In our work, utilizing known "morpholine-pyrimidine" structure-PI3Kδ-activity relationship and bicyclic pyrazolo [1,5-a]pyrimidine core, we developed a novel library of compounds focused on future COPD treatment. More specifically, we were fixed on the substitution of morpholine at the C(7) position leading to the 7-(morpholin-4-yl) pyrazolo [1,5a]pyrimidine structural motif. According to mentioned in the above paragraphs' correlations, we focused on the pyrazolo[1,5-a]pyrimidine core as probably the most promising structure (including the nitrogen atom in the five-membered ring), especially with the morpholine moiety in the appropriate position (to create the "morpholine-pyrimidine" system). We noticed that based on the structure of inhibitors as the candidates for the treatment of COPD or Asthma, cores based on bicyclic rings five-six-membered are more potent than six-six-membered, such as in CDZ 173 or UCB-5857. Moreover, we hoped that a five-six-membered ring, similar to pan-inhibitor GDC-0941 with appropriate modifications, could improve and increase the selectivity for isoform δ and thus becomes a selective PI3Kδ inhibitor. As a result, we obtained a selection of indole derivatives with improved potency and selectivity towards PI3Kδ inhibition. Moreover, we observed that 5-indole-pyrazolo [1,5-a]pyrimidine turned out to be the most promising core for future SAR studies.

Chemistry
The final compounds of our design were obtained in three different multistage approaches. The appropriate aminopyrazole derivatives (available commercially or synthesized) were used as the respective starting materials to provide the final inhibitors utilizing mainly the Buchwald-Hartwig reaction, the Suzuki coupling, or the Dess-Martin periodinane oxidation as the crucial synthetic steps.

Synthesis of Compounds 5-3
2-Methyl pyrazolo [1,5-a]pyrimidine derivatives were obtained in a multi-step reaction according to Scheme 1. 5-Amino-3-methylpyrazole was reacted with diethyl malonate in the presence of a base (sodium ethanolate) to obtain dihydroxy-heterocycle 1 (89% yield). Then, 2-methylpyrazolo[1,5-a]pyrimidine-5,7-diol (1) was subjected to the chlorination reaction with phosphorus oxychloride to give 5,7-dichloro-2-methylpyrazolo[1,5-a]pyrimidine (2) (61% yield). Structure 3 was prepared from 2 in a nucleophilic substitution reaction using morpholine in the presence of potassium carbonate at room temperature (94% yield). The selectivity of the reaction results from the strong reactivity of the chlorine atom at position 7 of the pyrazolo[1,5-a]pyrimidine core [50]. 4-{5-Chloro-2-methylpyrazolo[1,5a]pyrimidin-7-yl}morpholine (3) is the key intermediate in the preparation of a series of Supplementary compounds 5-13. Depending on the R 1 substituent, the final compounds were prepared from 3 using two types of coupling reactions: either the Buchwald-Hartwig or the Suzuki coupling reaction. Benzimidazole derivatives 5-7 were synthesized by carrying out the three-step reaction: again, the Buchwald-Hartwig reaction (average yield of 61%), amidation, following the final cyclization step. The corresponding amides 5-7 were prepared in the presence of EDCI and HOBt from the appropriate carboxylic acids and amine 4, resulting from the Buchwald-Hartwig synthesis by the heterocycle ring closure in the presence of glacial acetic acid. Since this synthetic route requires no intermediate purification, the observed yields are satisfactory in the 74-77% range. A separate synthetic route was chosen for compound 9, obtained in two steps by the Buchwald-Hartwig reaction with a masked aminopyrazole (54% yield), followed by the final deprotection of intermediate 8 (89% yield). Derivatives 10-13 were prepared by the Suzuki reaction of compound 3 with the respective esters or boronic acids in the presence of a palladium catalyst with yields in the range of 55-61%.

Synthesis of Compounds 23-45
The synthesis of Supplementary compounds 23-45 was more complicated and required several additional steps. The first three steps leading to compound 16 were performed based on the available literature data [51][52][53][54]. Initially, the reaction of benzyl alcohol with ethyl bromoacetate in the presence of sodium hydride gave the corresponding ether 14 (Scheme 2) with a 76% yield. Then the beta-ketoester derivative 15 was prepared by reaction with acetonitrile under basic conditions using 2,5 M n-butyllithium solution at a lower temperature of −78 • C. Compound 15 was subsequently condensed with hydrazine to give the corresponding aminopyrazole derivative 16 in satisfying 87% yield after two steps, as depicted in Scheme 2. The experiences gained in the previous synthetic route could be successfully extrapolated to accomplish the next four steps of the synthesis. Reaction of diethyl malonate with the aminopyrazole derivative 16 gave 2-[(benzyloxy)methyl]pyrazolo[1,5-a]pyrimidine-5,7-diol (17, 84% yield). Chlorination of 17 with phosphorus oxychloride provided the corresponding dichloro-derivative: 2-[(benzyloxy)methyl]-5,7-dichloropyrazolo [1,5-a]pyrimidine (18) in 38% yield. A selective and efficient (92% yield) substitution of the C(7)-chlorine atom in the heteroaromatic core with morpholine gave the analog of 3 (Scheme 1) as intermediate 19.
Applying the Suzuki coupling conditions to 19 with indole-4-boronic acid pinacol ester led to benzyl masked alcohol 20 in 83% yield. Classical deprotection conditions (gaseous hydrogen over palladium catalyst on activated charcoal) of the benzyloxy group provided compound 21 in 66% The synthesis of compounds 23-45 was more complicated and required several additional steps. The first three steps leading to compound 16 were performed based on the available literature data [51][52][53][54]. Initially, the reaction of benzyl alcohol with ethyl bromoacetate in the presence of sodium hydride gave the corresponding ether 14 (Scheme 2) with a 76% yield. Then the beta-ketoester derivative 15 was prepared by reaction with acetonitrile under basic conditions using 2,5 M n-butyllithium solution at a lower temperature of −78 °C. Compound 15 was subsequently condensed with hydrazine to give the corresponding aminopyrazole derivative 16 in satisfying 87% yield after two steps, as depicted in Scheme 2. The experiences gained in the previous synthetic route could be successfully extrapolated to accomplish the next four steps of the synthesis. Reaction of diethyl malonate with the aminopyrazole derivative 16 gave 2-[(benzyloxy)methyl]pyrazolo[1,5-a]pyrimidine-5,7-diol (17, 84% yield). Chlorination of 17 with phosphorus oxychloride provided the corresponding dichloro-derivative: 2-[(benzyloxy)methyl]-5,7-dichloropyrazolo[1,5-a]pyrimidine (18) in 38% yield. A selective and efficient (92% yield) substitution of the C(7)-chlorine atom in the heteroaromatic core with morpholine gave the analog of 3 (Scheme 1) as intermediate 19.

Synthesis of Compounds 49-51 and 53-55
An essential intermediate 19 (Scheme 2) was also successfully used to prepare another set of compounds functionalized at the C(5) position to explore more deeply the structureactivity relationship of this particular core. The synthesis of another subset of substituted pyrazolo [1,5-a]pyrimidines is shown in Scheme 3. Due to the same reaction types, the synthesis pathways of examples Supplementary 49-51 and 53-55 were similar to the synthesis of the previous compounds (23-45, Scheme 2), the difference being the order of the Suzuki reaction and the reductive amination reaction sequence in the multistage synthesis pathway. After deprotection of the hydroxyl group of 19, compound 46 was oxidized to aldehyde 47 (Scheme 3). The following steps included a reductive amination reaction with the carefully selected, based on in silico calculations, amines: (2-(4-piperidyl)-2-propanol or N-t-butylpiperazine followed by a Suzuki coupling to provide Supplementary 49-51 and 53-55, respectively (Scheme 3). An essential intermediate 19 (Scheme 2) was also successfully used to prepare another set of compounds functionalized at the C(5) position to explore more deeply the structure-activity relationship of this particular core. The synthesis of another subset of substituted pyrazolo [1,5-a]pyrimidines is shown in Scheme 3. Due to the same reaction types, the synthesis pathways of examples 49-51 and 53-55 were similar to the synthesis of the previous compounds (23-45, Scheme 2), the difference being the order of the Suzuki reaction and the reductive amination reaction sequence in the multistage synthesis pathway. After deprotection of the hydroxyl group of 19, compound 46 was oxidized to aldehyde 47 (Scheme 3). The following steps included a reductive amination reaction with the carefully selected, based on in silico calculations, amines: (2-(4-piperidyl)-2-propanol or N-t-butylpiperazine followed by a Suzuki coupling to provide 49-51 and 53-55, respectively (Scheme 3).

Docking Study
Several approaches have been described leading to various structural docking theories explaining the selectivity of PI3Kδ inhibitors [25,27]. Opening the specificity pocket between the two amino acids, Trp-812 and Met-804, and adopting the appropriate shape within the protein combined with additional correlations, allows the identification of much more selective PI3Kδ inhibitors from all PI3K Class I isoforms [25,27,34]. It was reported that there are many meaningful interactions between ligand and protein in the enzyme's active site [6,24,27]. First is the hydrogen bond of the morpholine from pyrazolo[1,5-a]pyrimidine derivative in the hinge-binding motif [6,[24][25][26]. More precisely, the hydrogen bonding between the oxygen atom from the morpholine mentioned above the ring and amino acid Val-828 was crucial in the hinge region. It has been suggested that indole derivatives in the C(5) position of the core of pyrazolo[1,5-a]pyrimidine may form an additional hydrogen bond with Asp-787 (another important interaction in many selective inhibitors, most with the affinity pocket) [25]. For this reason, indole heterocycle-based inhibitors are more selective for PI3Kδ than other PI3K isoforms. In addition, a suitable substituent of this structure, which can extend into the solvent, can improve the solubility, ADME properties, and potency of the final compounds [25]. 7

Docking Study
Several approaches have been described leading to various structural docking theories explaining the selectivity of PI3Kδ inhibitors [25,27]. Opening the specificity pocket between the two amino acids, Trp-812 and Met-804, and adopting the appropriate shape within the protein combined with additional correlations, allows the identification of much more selective PI3Kδ inhibitors from all PI3K Class I isoforms [25,27,34]. It was reported that there are many meaningful interactions between ligand and protein in the enzyme's active site [6,24,27]. First is the hydrogen bond of the morpholine from pyrazolo [1,5-a]pyrimidine derivative in the hinge-binding motif [6,[24][25][26]. More precisely, the hydrogen bonding between the oxygen atom from the morpholine mentioned above the ring and amino acid Val-828 was crucial in the hinge region. It has been suggested that indole derivatives in the C(5) position of the core of pyrazolo[1,5-a]pyrimidine may form an additional hydrogen bond with Asp-787 (another important interaction in many selective inhibitors, most with the affinity pocket) [25]. For this reason, indole heterocyclebased inhibitors are more selective for PI3Kδ than other PI3K isoforms. In addition, a suitable substituent of this structure, which can extend into the solvent, can improve the solubility, ADME properties, and potency of the final compounds [25].
Our work is focused on the pyrazolo[1,5-a]pyrimidine scaffold and appropriate further optimization with different C(5) substituents.
An example of our approach showing the possible binding site of compound 13 with the kinase is presented in Figure 2. The docking procedure utilizes the PI3Kδ protein (PDB: 2WXP) and the Auto-Dock Vina program [55]. Compound 13 (magenta) binds similarity to protein as referent compound GDC-0941 (orange, Figure 2). More specifically, the oxygen atom in the morpholine ring forms a hydrogen bond with the amino acid (Val-828) in the hinge region of the enzyme (the importance of this interaction has been explained before). Moreover, the indole system's hydrogen atom (NH) is involved in forming the hydrogen bond with the carbonyl oxygen in Asp-787 in the affinity pocket of the kinase ( Figure 2). (v) N-t-butylpiperazine, sodium triacetoxyborohydride, DCM, RT, 16 h, 53%; (vi) boronic acid pinacol ester, tetrakis(triphenylphosphino)palladium (0), 2M aq Na 2 CO 3 , DME, reflux, 16 h, 68-77%.
Our work is focused on the pyrazolo[1,5-a]pyrimidine scaffold and appropriate further optimization with different C(5) substituents.
An example of our approach showing the possible binding site of compound 13 with the kinase is presented in Figure 2. The docking procedure utilizes the PI3Kδ protein (PDB: 2WXP) and the Auto-Dock Vina program [55]. Compound 13 (magenta) binds similarity to protein as referent compound GDC-0941 (orange, Figure 2). More specifically, the oxygen atom in the morpholine ring forms a hydrogen bond with the amino acid (Val-828) in the hinge region of the enzyme (the importance of this interaction has been explained before). Moreover, the indole system's hydrogen atom (NH) is involved in forming the hydrogen bond with the carbonyl oxygen in Asp-787 in the affinity pocket of the kinase ( Figure 2).
Among the structures 24, 36, and 37 additional features were found in our in silico model compared to 13 and similars. Compared to compound 23, higher activity and selectivity can be explained by interactions with the tryptophan shelf (2WXP: Trp-760) in PI3Kδ, as described by Sutherlin et al. [25]. For those compounds, the distance between the R 2 substituent and the tryptophan's indole ring is significantly shorter ( Figure 3A). Moreover, the additional hydrogen bond of the hydroxyl group in (2-(piperidin-4-yl) propan-2-ol) (36) with Lys-708 was observed ( Figure 3B). On the other hand, for a derivative containing tert-butylpiperazine (37), strong hydrophobic interactions with tryptophan (Trp-760) were found, which may cause the withdrawal of the indole ring of 37 from the enzyme affinity pocket. Most likely, this situation is observed due to the lack of interaction with tyrosine (Tyr-813) and aspartic acid (Asp-787) in the mentioned pocket ( Figure 3B). Among the structures 24, 36, and 37 additional features were found in our in silico model compared to 13 and similars. Compared to compound 23, higher activity and selectivity can be explained by interactions with the tryptophan shelf (2WXP: Trp-760) in PI3Kδ, as described by Sutherlin et al. [25]. For those compounds, the distance between the R 2 substituent and the tryptophan's indole ring is significantly shorter ( Figure 3A). Moreover, the additional hydrogen bond of the hydroxyl group in (2-(piperidin-4-yl) propan-2-ol) (36) with Lys-708 was observed ( Figure 3B). On the other hand, for a derivative containing tert-butylpiperazine (37), strong hydrophobic interactions with tryptophan (Trp-760) were found, which may cause the withdrawal of the indole ring of 37 from the enzyme affinity pocket. Most likely, this situation is observed due to the lack of interaction with tyrosine (Tyr-813) and aspartic acid (Asp-787) in the mentioned pocket ( Figure 3B).   Among the structures 24, 36, and 37 additional features were found in our in silico model compared to 13 and similars. Compared to compound 23, higher activity and selectivity can be explained by interactions with the tryptophan shelf (2WXP: Trp-760) in PI3Kδ, as described by Sutherlin et al. [25]. For those compounds, the distance between the R 2 substituent and the tryptophan's indole ring is significantly shorter ( Figure 3A). Moreover, the additional hydrogen bond of the hydroxyl group in (2-(piperidin-4-yl) propan-2-ol) (36) with Lys-708 was observed ( Figure 3B). On the other hand, for a derivative containing tert-butylpiperazine (37), strong hydrophobic interactions with tryptophan (Trp-760) were found, which may cause the withdrawal of the indole ring of 37 from the enzyme affinity pocket. Most likely, this situation is observed due to the lack of interaction with tyrosine (Tyr-813) and aspartic acid (Asp-787) in the mentioned pocket ( Figure 3B).

Biological Evaluation In Vitro PI3 Kinase Inhibition Assays
To verify whether the 7-(morpholin-4-yl) pyrazolo[1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1.

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13).

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13).

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13).

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13).

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13). 43

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13).

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13).

Biological Evaluation
In Vitro PI3 Kinase Inhibition Assays To verify whether the 7-(morpholin-4-yl) pyrazolo [1,5-a] pyrimidine system can inhibit PI3δ kinase, the synthesized compounds 6-13 were tested for inhibition of selected PI3Kδ and PI3Kα kinases activity. Enzymatic tests have been used, and the results are presented in Table 1. The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC50 value was measured for compound 7 (IC50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13). The activity of these compounds ranged from 45 µM to 0.5 µM for the PI3K δ isoform and from over 60 µM to 1.06 µM for the PI3Kα isoform, and thus the α/δ selectivity ranged from 1 to 30 (Table 1). Among all benzimidazole derivatives synthesized, the most promising activity with the low PI3Kδ IC 50 value was measured for compound 7 (IC 50 = 0.47 µM) ( Table 1). On the other hand, compounds 5 and 6, keeping benzimidazole derivatives within their structures, show significantly lower activity against the PI3Kδ isoform than compound 7 (IC 50 value of 3.56 µM and 2.30 µM, respectively), regardless of better selectivity against the PI3Kα isoform (α/δ) (9.9 for 5 and 11 for 6). We observed that compounds with a monocyclic 5-or 6-membered heteroaromatic ring (9-11) turned out to be less active and thus showed a lower enzyme inhibition potential than the other bicyclic structures. Structures 12 and 13 bearing conjugated bicyclic system as the R 1 substituent presented a similar activity to the benzimidazole derivatives. The most active were compounds having R 1 substituents in the form of 2-difluoromethylbenzimidazole (7) and indole (13). Specifically, their IC 50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization.
Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2).  Specifically, their IC50 value against PI3Kδ was 0.475 µM and 0.772 µM, respectively. Due to the much better α/δ selectivity of compound 13 over compound 7 (α/δ = 30 and α/δ = 2.2, respectively), we have chosen the indole derivatives for further optimization. Compared to compound 13, significantly more sterically demanding derivatives were designed and synthesized as the next optimization step. While the indole fragments were preserved, many different cyclic amines were linked to the scaffold core through a methylene linkage as an R 2 substituent (Table 2). IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( Table 2). Moreover, the presence of the second ring within the R 2 substituent (compounds 39-40 and 42-45) did not improve PI3Kδ activity compared to previously obtained compounds 24 or 29. Finally, only large aliphatic substituents within piperazine or piperidine rings gain the PI3Kδ potency and respective selectivity.
We observed that the best results were achieved for two compounds being the rep- IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( Table 2). Moreover, the presence of the second ring within the R 2 substituent (compounds 39-40 and 42-45) did not improve PI3Kδ activity compared to previously obtained compounds 24 or 29. Finally, only large aliphatic substituents within piperazine or piperidine rings gain the PI3Kδ potency and respective selectivity.
We observed that the best results were achieved for two compounds being the rep- IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( Table 2). Moreover, the presence of the second ring within the R 2 substituent (compounds 39-40 and 42-45) did not improve PI3Kδ activity compared to previously obtained compounds 24 or 29. Finally, only large aliphatic substituents within piperazine or piperidine rings gain the PI3Kδ potency and respective selectivity. 42 13 IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( Table 2). Moreover, the presence of the second ring within the R 2 substituent (compounds 39-40 and 42-45) did not improve PI3Kδ activity compared to previously obtained compounds 24 or 29. Finally, only large aliphatic substituents within piperazine or piperidine rings gain the PI3Kδ potency and respective selectivity. 51 12 IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( Table 2). Moreover, the presence of the second ring within the R 2 substituent (compounds 39-40 and 42-45) did not improve PI3Kδ activity compared to previously obtained compounds 24 or 29. Finally, only large aliphatic substituents within piperazine or piperidine rings gain the PI3Kδ potency and respective selectivity. 56 18 IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( Table 2). Moreover, the presence of the second ring within the R 2 substituent (compounds 39-40 and 42-45) did not improve PI3Kδ activity compared to previously obtained compounds 24 or 29. Finally, only large aliphatic substituents within piperazine or piperidine rings gain the PI3Kδ potency and respective selectivity. IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( IC50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC50 above 50 nM were additionally checked for the potency of the PI3Kα isoform. The synthesis of the new group of pyrazolo[1,5-a] pyrimidine derivatives (depicted in Scheme 2) required additional steps related to the functionalization of the C(2)-position of the heteroaromatic core. Firstly, a group of derivatives with differing sizes of heterocycle rings and different chemical properties of substituents (23-31) was synthesized ( Table  2). We noted that structures containing monocyclic five-membered rings (25)(26) and morpholine (28) turned out to be less potent PI3Kδ inhibitors than compound 13 ( Table 1). The mesylpiperazine group present in the GDC0941 Reference [34] did not significantly improve the activity of structurally similar compound 23 from our library (the IC50 value of that example for PI3Kδ and PI3Kα was 0.4 µM and 2.35 µM, respectively). Urea-derivatives, 30 and 31, also showed moderate activity. The most potent compounds in this group ( Table 2) turn out to be the analogs of N,N-dimethyl-4-aminopiperidine (24), and 4-(Nmethylpiperazin-1ylo)piperidine (29). Both, 24 and 29, showed promising inhibitory activity against PI3Kδ (37 nM and 52 nM respectively) and selectivity against other isoforms (α/δ = 172; β/δ = 389; γ/δ = 1332 for 24 and α/δ = 301 for 29). Careful structural analysis around the R 2 substituent of the examples provided in Table 2 led us to several conclusions. Relatively modest activities of the compounds containing the methyl group, aromatic ring, or ester group at the C(4)-position of the heterocyclic ring misled us towards the synthesis of piperazine and piperidine analogs(32-45) ( Table 2). Moreover, the presence of the second ring within the R 2 substituent (compounds 39-40 and 42-45) did not improve PI3Kδ activity compared to previously obtained compounds 24 or 29. Finally, only large aliphatic substituents within piperazine or piperidine rings gain the PI3Kδ po- 29 6320 218 IC 50 values were determined as the mean based on two independent experiments. For compounds with PI3Kδ IC 50 above 0.5 µM, the activity for the remaining isoforms was not determined. Compounds with PI3Kδ IC 50 above 50 nM were additionally checked for the potency of the PI3Kα isoform.
We observed that the best results were achieved for two compounds being the representatives of two different modifications. More specifically 2-(piperidin-4-yl) propan-2-ol (compound 36 of piperidine modification series) and N-tert-butylpiperazine (compound 37 of piperazine modification series) exhibit high activities towards the PI3Kδ (IC 50 = 6.6 and 13.0 nM, respectively) and appreciable selectivities towards other isoforms (α/δ = 1217; β/δ = 332; γ/δ = 1223 for 36 and α/δ = 1889; β/δ = 829; γ/δ > 9091 for 37; Table 2). As the hit to lead optimization route continued, several indole and azaindole derivatives at the C(5) position were introduced to the existing scaffold. While preserving the most active amino groups, we prepared the piperidine derivatives series (summarized in Table 3) and piperazine derivatives series (covered in Table 4). From all the synthesized structures, the N-tert-butylpiperazine derivatives (37, 53, 54, 55, Table 4) show the highest PI3Kδ activity, greater than the piperidyl-propanol analogs shown in Table 3 (36,49,50,51). The presence of the fluorine atom in the C(5)-position of the indol fragment causes a slight decrease in activity against the PI3Kδ isoform in both groups without affecting the selectivity toward other isoforms. The introduction of the nitrogen atom to the indole ring at position 7 caused a slight decrease in the activity of compound 51 (Table 3), which was almost doubled in the case of 55 (Table 4). Moreover, slight decreases in activity related to the PI3Kα isoform were observed for these structures. An introduction of a nitrogen atom in the 6-position of the indole caused a decrease in activity derivate 50 but a 10-fold improvement for 54. Decreased selectivity against the PI3Kα isoform was also observed for the azaindole structures (50, 51, 53, 54) despite the good activity in the nanomolar range (IC 50 value: 2.8-45 nM). the hit to lead optimization route continued, several indole and azaindole derivatives at the C(5) position were introduced to the existing scaffold. While preserving the most active amino groups, we prepared the piperidine derivatives series (summarized in Table  3) and piperazine derivatives series (covered in Table 4). From all the synthesized structures, the N-tert-butylpiperazine derivatives (37, 53, 54, 55, Table 4) show the highest PI3Kδ activity, greater than the piperidyl-propanol analogs shown in Table 3 (36,49,50,51). The presence of the fluorine atom in the C(5)-position of the indol fragment causes a slight decrease in activity against the PI3Kδ isoform in both groups without affecting the selectivity toward other isoforms. The introduction of the nitrogen atom to the indole ring at position 7 caused a slight decrease in the activity of compound 51 (Table 3), which was almost doubled in the case of 55 (Table 4). Moreover, slight decreases in activity related to the PI3Kα isoform were observed for these structures. An introduction of a nitrogen atom in the 6-position of the indole caused a decrease in activity derivate 50 but a 10-fold improvement for 54. Decreased selectivity against the PI3Kα isoform was also observed for the azaindole structures (50,51,53,54) despite the good activity in the nanomolar range (IC50 value: 2.8-45 nM). the hit to lead optimization route continued, several indole and azaindole derivatives at the C(5) position were introduced to the existing scaffold. While preserving the most active amino groups, we prepared the piperidine derivatives series (summarized in Table  3) and piperazine derivatives series (covered in Table 4). From all the synthesized structures, the N-tert-butylpiperazine derivatives (37, 53, 54, 55, Table 4) show the highest PI3Kδ activity, greater than the piperidyl-propanol analogs shown in Table 3 (36,49,50,51). The presence of the fluorine atom in the C(5)-position of the indol fragment causes a slight decrease in activity against the PI3Kδ isoform in both groups without affecting the selectivity toward other isoforms. The introduction of the nitrogen atom to the indole ring at position 7 caused a slight decrease in the activity of compound 51 (Table 3), which was almost doubled in the case of 55 (Table 4). Moreover, slight decreases in activity related to the PI3Kα isoform were observed for these structures. An introduction of a nitrogen atom in the 6-position of the indole caused a decrease in activity derivate 50 but a 10-fold improvement for 54. Decreased selectivity against the PI3Kα isoform was also observed for the azaindole structures (50,51,53,54) despite the good activity in the nanomolar range (IC50 value: 2.8-45 nM). the hit to lead optimization route continued, several indole and azaindole derivatives at the C(5) position were introduced to the existing scaffold. While preserving the most active amino groups, we prepared the piperidine derivatives series (summarized in Table  3) and piperazine derivatives series (covered in Table 4). From all the synthesized structures, the N-tert-butylpiperazine derivatives (37, 53, 54, 55, Table 4) show the highest PI3Kδ activity, greater than the piperidyl-propanol analogs shown in Table 3 (36,49,50,51). The presence of the fluorine atom in the C(5)-position of the indol fragment causes a slight decrease in activity against the PI3Kδ isoform in both groups without affecting the selectivity toward other isoforms. The introduction of the nitrogen atom to the indole ring at position 7 caused a slight decrease in the activity of compound 51 (Table 3), which was almost doubled in the case of 55 (Table 4). Moreover, slight decreases in activity related to the PI3Kα isoform were observed for these structures. An introduction of a nitrogen atom in the 6-position of the indole caused a decrease in activity derivate 50 but a 10-fold improvement for 54. Decreased selectivity against the PI3Kα isoform was also observed for the azaindole structures (50,51,53,54) despite the good activity in the nanomolar range (IC50 value: 2.8-45 nM). the hit to lead optimization route continued, several indole and azaindole derivatives at the C(5) position were introduced to the existing scaffold. While preserving the most active amino groups, we prepared the piperidine derivatives series (summarized in Table  3) and piperazine derivatives series (covered in Table 4). From all the synthesized structures, the N-tert-butylpiperazine derivatives (37, 53, 54, 55, Table 4) show the highest PI3Kδ activity, greater than the piperidyl-propanol analogs shown in Table 3 (36,49,50,51). The presence of the fluorine atom in the C(5)-position of the indol fragment causes a slight decrease in activity against the PI3Kδ isoform in both groups without affecting the selectivity toward other isoforms. The introduction of the nitrogen atom to the indole ring at position 7 caused a slight decrease in the activity of compound 51 (Table 3), which was almost doubled in the case of 55 (Table 4). Moreover, slight decreases in activity related to the PI3Kα isoform were observed for these structures. An introduction of a nitrogen atom in the 6-position of the indole caused a decrease in activity derivate 50 but a 10-fold improvement for 54. Decreased selectivity against the PI3Kα isoform was also observed for the azaindole structures (50,51,53,54) despite the good activity in the nanomolar range (IC50 value: 2.8-45 nM). the hit to lead optimization route continued, several indole and azaindole derivatives at the C(5) position were introduced to the existing scaffold. While preserving the most active amino groups, we prepared the piperidine derivatives series (summarized in Table  3) and piperazine derivatives series (covered in Table 4). From all the synthesized structures, the N-tert-butylpiperazine derivatives (37, 53, 54, 55, Table 4) show the highest PI3Kδ activity, greater than the piperidyl-propanol analogs shown in Table 3 (36,49,50,51). The presence of the fluorine atom in the C(5)-position of the indol fragment causes a slight decrease in activity against the PI3Kδ isoform in both groups without affecting the selectivity toward other isoforms. The introduction of the nitrogen atom to the indole ring at position 7 caused a slight decrease in the activity of compound 51 (Table 3), which was almost doubled in the case of 55 (Table 4). Moreover, slight decreases in activity related to the PI3Kα isoform were observed for these structures. An introduction of a nitrogen atom in the 6-position of the indole caused a decrease in activity derivate 50 but a 10-fold improvement for 54. Decreased selectivity against the PI3Kα isoform was also observed for the azaindole structures (50,51,53,54) despite the good activity in the nanomolar range (IC50 value: 2.8-45 nM). We have found that two compounds: 37 and 54, from the entire synthesized library showed the best activity and selectivity for PI3Kδ. Based on all parameters, these structures showed the highest selectivity, the lowest IC 50 values, and the most promising other parameters [15]. Consequently, those two selected examples were tested by flow cytometry towards the proliferation of B lymphocytes capabilities. Both showed very high potency in inhibiting B cell proliferation with IC 50 values of 20 nM and 19 nM, respectively (Table 5). Moreover, compound 54 had better kinetic solubility at pH 7.4 than compound 37 (>500 and 444 µM respectively) ( Table 5). We also observed that the presence of nitrogen atom in the 6-azaindole ring of 54 molecule results in higher metabolic stability in murine and human microsomes (for details, see Table 5).  IC50 values were determined as the mean based on two independent experiments.
We have found that two compounds: 37 and 54, from the entire synthesized library showed the best activity and selectivity for PI3Kδ. Based on all parameters, these structures showed the highest selectivity, the lowest IC50 values, and the most promising other parameters [15]. Consequently, those two selected examples were tested by flow cytometry towards the proliferation of B lymphocytes capabilities. Both showed very high potency in inhibiting B cell proliferation with IC50 values of 20 nM and 19 nM, respectively (Table 5). Moreover, compound 54 had better kinetic solubility at pH 7.4 than compound 37 (>500 and 444 µM respectively) ( Table 5). We also observed that the presence of nitrogen atom in the 6-azaindole ring of 54 molecule results in higher metabolic stability in murine and human microsomes (for details, see Table 5).  We have found that two compounds: 37 and 54, from the entire synthesized library showed the best activity and selectivity for PI3Kδ. Based on all parameters, these structures showed the highest selectivity, the lowest IC50 values, and the most promising other parameters [15]. Consequently, those two selected examples were tested by flow cytometry towards the proliferation of B lymphocytes capabilities. Both showed very high potency in inhibiting B cell proliferation with IC50 values of 20 nM and 19 nM, respectively (Table 5). Moreover, compound 54 had better kinetic solubility at pH 7.4 than compound 37 (>500 and 444 µM respectively) ( Table 5). We also observed that the presence of nitrogen atom in the 6-azaindole ring of 54 molecule results in higher metabolic stability in murine and human microsomes (for details, see Table 5).  We have found that two compounds: 37 and 54, from the entire synthesized library showed the best activity and selectivity for PI3Kδ. Based on all parameters, these structures showed the highest selectivity, the lowest IC50 values, and the most promising other parameters [15]. Consequently, those two selected examples were tested by flow cytometry towards the proliferation of B lymphocytes capabilities. Both showed very high potency in inhibiting B cell proliferation with IC50 values of 20 nM and 19 nM, respectively (Table 5). Moreover, compound 54 had better kinetic solubility at pH 7.4 than compound 37 (>500 and 444 µM respectively) ( Table 5). We also observed that the presence of nitrogen atom in the 6-azaindole ring of 54 molecule results in higher metabolic stability in murine and human microsomes (for details, see Table 5).  We have found that two compounds: 37 and 54, from the entire synthesized library showed the best activity and selectivity for PI3Kδ. Based on all parameters, these structures showed the highest selectivity, the lowest IC50 values, and the most promising other parameters [15]. Consequently, those two selected examples were tested by flow cytometry towards the proliferation of B lymphocytes capabilities. Both showed very high potency in inhibiting B cell proliferation with IC50 values of 20 nM and 19 nM, respectively (Table 5). Moreover, compound 54 had better kinetic solubility at pH 7.4 than compound 37 (>500 and 444 µM respectively) ( Table 5). We also observed that the presence of nitrogen atom in the 6-azaindole ring of 54 molecule results in higher metabolic stability in murine and human microsomes (for details, see Table 5).  We have found that two compounds: 37 and 54, from the entire synthesized library showed the best activity and selectivity for PI3Kδ. Based on all parameters, these structures showed the highest selectivity, the lowest IC50 values, and the most promising other parameters [15]. Consequently, those two selected examples were tested by flow cytometry towards the proliferation of B lymphocytes capabilities. Both showed very high potency in inhibiting B cell proliferation with IC50 values of 20 nM and 19 nM, respectively (Table 5). Moreover, compound 54 had better kinetic solubility at pH 7.4 than compound 37 (>500 and 444 µM respectively) ( Table 5). We also observed that the presence of nitrogen atom in the 6-azaindole ring of 54 molecule results in higher metabolic stability in murine and human microsomes (for details, see Table 5).   and were used without additional purification. Solvents were purified according to standard procedures if required. Air or moisture-sensitive reactions were carried out under an argon atmosphere. All reac-tion progresses were routinely checked by thin-layer chromatography (TLC). TLC was performed using silica gel coated plates (Kieselgel F254) and visualized using UV light. Flash chromatography was performed using Merck silica gel 60 (230-400 mesh ASTM). 1 H NMR spectra were acquired on a Varian Inova 300 MHz NMR spectrometer, JOEL JNMR-ECZS 400 MHz spectrometer, JOEL JNMR-ECZR 600 MHz spectrometer, and Bruker DRX 500 NMR spectrometer with 1 H being observed at 300 MHz, 400 MHz, 600 MHz, and 500 MHz, respectively. 13 C NMR spectra were recorded similarly at 75 MHz, 101 MHz, 151 MHz, and 126 MHz, frequencies for 13 C, respectively. Due to the poor solubility of some final compounds, usual characterization by 13 C NMR was omitted. Chemical shifts for 1 H and 13 C NMR spectra were reported in δ (ppm) using tetramethylsilane as an internal standard or according to the residual undeuterated solvent signal (2.50 ppm for DMSO-d 6  To the cooled to 0 • C POCl 3 (90 mL, 0.963 mol), compound 1 (15.2 g, 0.092 mol) was added. The reaction was carried out at reflux for 24 h. The reaction mixture was cooled to room temperature and poured into the water with ice. The mixture was quenched with a 6 M sodium hydroxide solution to pH 6. The aqueous phase was extracted with ethyl acetate, and after separation, the organic phase was dried with anhydrous sodium sulfate. After filtration of the drying agent and evaporation of the solvent, the residue was purified by column chromatography (0-40% ethyl acetate gradient in heptane) to give compound 2 (11.4 g, 0.056 mol) obtained as an off-white solid with 61% yield. 1 (3) To the solution of compound 2 (2.0 g, 9.9 mmol) in acetone (50 mL), potassium carbonate (1.64 g, 11.9 mmol), and morpholine (1.35 mL, 15.5 mmol) were added. The reaction was carried out at room temperature for 1.5 h. Then water (100 mL) was added to the reaction mixture, and the precipitated white solid was filtered off. The obtained solid was washed with water (50 mL) and water/acetone mixture (2/1, v/v) (50 mL), then dried. Compound 3 (2.36 g, 0.09 mol) was obtained as a white solid with 94% yield. 1 (4) The mixture of compound 3 (1.0 g, 3.96 mmol), benzene-1,2-diamine (1.31 g, 11.9 mmol), cesium carbonate (3.87 g, 11.9 mmol), tris(dibenzylideneacetone)dipalladium(0) (0.181 g, 0.20 mmol), 9,9-dimethyl-4,5-bis(diphenylphosphine)xanthene (0.229 g, 0.40 mmol) and dry toluene (40 mL) were introduced to the reaction Schlenk flask. The mixture was flushed with argon and stirred at 110 • C for 24 h. After cooling to room temperature, the reaction mixture was filtered through Celite ® , and the solid was washed with ethyl acetate. The filtrate was concentrated under reduced pressure using an evaporator. The residue was resolved and purified by column chromatography (50-100% ethyl acetate gradient in heptane) to give the title compound 4 (0.78 g, 2.4 mmol) with 61% yield. 1  In the solution of compound 4 (1.0 eq) dissolved in dry DCM (10 mL/1g of compound 4), the carboxylic acid (2.0 eq), HOBt × H 2 O (1.2 eq), EDCI × HCl (2.4 eq), and TEA (3.0 eq) were added. The whole reaction mixture was stirred at room temperature for 48 h. To the reaction, mixture water was added, and organic and water phases were separated. The aqueous phase was washed three times with DCM. Combined organic phases were dried over anhydrous sodium sulfate. After the drying agent was filtered off and the solvent evaporated, the reaction mixture was dissolved in glacial acetic acid. The reaction mixture was refluxed for 24 h. Then the reaction mixture was cooled and concentrated under reduced pressure. The residue was diluted with water and neutralized with a saturated sodium bicarbonate solution. The aqueous phase was extracted three times with ethyl acetate. Combined organic phases were dried over sodium sulfate. Once the drying agent was filtered off, the solvent was evaporated under reduced pressure using an evaporator. The reaction mixture was purified by column chromatography.

General Procedure for the Suzuki Reaction
To the solution of compound 3 (1.0 eq) dissolved in 1,2-dimethoxyethane (DME) (10 mL/1 g of compound 3), boronic acid pinacol ester or boronic acid (1.5 eq), tetrakis(triph enylphosphino)palladium (0) (0.2 eq) and 2M aqueous sodium carbonate solution (2.0 eq) were added. The reaction mixture was refluxed overnight. Then, the reaction mixture was cooled to room temperature, filtered through the pad of Celite ® , and obtained solid washed with ethyl acetate. The filtrate was concentrated under reduced pressure using an evaporator and the residue was purified by column chromatography.

Docking Study
The docking procedure was performed in the PI3K δ protein from Protein Data Bank (PDB: 2WXP) using the Auto-Dock Vina program [55]. All figures with examples of 3D modeling of a possible binding mode of selected compounds were prepared based on the calculated pK a from the Instant JChem 21.13.0 program [57]. More specifically, all structures depicted in the respective figures have not had protons added, but the appropriate protonation state has been maintained.

In Vitro Kinase Inhibition Assay for PI3K
Tested compounds were dissolved in 100% DMSO, and obtained solutions were serially diluted in 1× reaction buffer. The recombinant kinase solution was diluted in a reaction mixture comprising 5× reaction buffer, respective compound solution (1 mM sodium diacetate 4,5-bisphosphate phosphatidylinositol (PIP2) solution in 40 mM Tris buffer), and water. In a 96-wells plate, 5 µL of compound solutions and 15 µL of the kinase solution in the reaction mixture were added per well. To initiate the interaction of chemical compounds to be tested with the enzyme, the plate was incubated for 10 min at a suitable temperature in a plate thermostat with orbital shaking at 600 rpm. Negative control wells contained all the above reagents except tested compound and kinase, and positive control wells contained all the above reagents except tested compounds. The enzymatic reaction was initiated by adding 5 µL of 150 µM ATP solution. Subsequently, the plate was incubated for 1 h at 25 or 30 • C (depending on the PI3K isoform tested) in a plate thermostat with orbital shaking of the plate contents at 600 rpm. The reaction conditions are combined in the table below (Table 6). Detection of ADP formed in the enzymatic reaction was then performed using ADP-Glo Kinase Assay™ (Promega, Madison, WI, USA). To the wells of a 96-well plate, 25 µL of ADP-Glo Reagent™ was added, and the plate was incubated for 40 min at 25 • C in a plate thermostat with orbital shaking at 600 rpm. Then 50 µL of Kinase Detection Reagent were added to each well, and the plate was incubated for 40 min at 25 • C in a plate thermostat with orbital shaking at 600 rpm. Once the incubation was complete, the luminescence intensity was measured using a Victor Light luminometer (Perkin Elmer, Inc., Waltham, MA, USA). IC 50 values were determined based on luminescence intensity measured in wells containing tested compounds at different concentrations in relation to control wells. These values were calculated with Graph Pad 5.03 software by fitting the curve using non-linear regression. Each compound was tested at least in quadruplicates (4 wells) on two 96-well plates utilizing at least 4 wells for each control. Averaged results of inhibition activity respective to specific isoforms of PI3K kinases for tested compounds are presented as IC 50 values in Tables 1-4.