Synthesis and Biological Evaluation of 3-Amino-4,4-Dimethyl Lithocholic Acid Derivatives as Novel, Selective, and Cellularly Active Allosteric SHP1 Activators

Src homology 2 domain-containing protein tyrosine phosphatase 1 (SHP1), a non-receptor member of the protein tyrosine phosphatase (PTP) family, negatively regulates several signaling pathways that are responsible for pathological cell processes in cancers. In this study, we report a series of 3-amino-4,4-dimethyl lithocholic acid derivatives as SHP1 activators. The most potent compounds, 5az-ba, showed low micromolar activating effects (EC50: 1.54–2.10 μM) for SHP1, with 7.63–8.79-fold maximum activation and significant selectivity over the closest homologue Src homology 2 domain-containing protein tyrosine phosphatase 2 (SHP2) (>32-fold). 5az-ba showed potent anti-tumor effects with IC50 values of 1.65–5.51 μM against leukemia and lung cancer cells. A new allosteric mechanism of SHP1 activation, whereby small molecules bind to a central allosteric pocket and stabilize the active conformation of SHP1, was proposed. The activation mechanism was consistent with the structure–activity relationship (SAR) data. This study demonstrates that 3-amino-4,4-dimethyl lithocholic acid derivatives can be selective SHP1 activators with potent cellular efficacy.


Introduction
No commercial PTP-targeted drug has been approved for decades. The biological and clinical discovery of PTP inhibitors is hampered by off-target profiles, poor pharmacokinetic properties, or the emergence of drug resistance. In contrast, SHP1 was found to negatively regulate various pathological processes, including hematopoietic lineage differentiation, tumorigenesis, and immune response, which has led to the development of SHP1 activators. SHP1 is localized mainly in the cytoplasm and is expressed on multiple cell types, including hematopoietic, epithelial, and mesenchymal cells. Low expression of SHP1 protein was observed in patients with hepatocellular carcinoma [1], prostate cancer [2], multiple sclerosis [3], and a variety of hematopoietic diseases, such as leukemia [4], lymphoma [5], and multiple myeloma [6], as well as myelodysplastic syndrome [7], whereas SHP1 is up-regulated in patients with nasopharyngeal carcinoma [8] and ovarian cancer [9]. SHP1 dephosphorylates the signal transducer and activator of transcription 3 (STAT3) [10], extracellular signal-regulated kinase (ERK) [11], protein kinase B (Akt) [12], and other signaling components [13][14][15][16], thereby contributing to a decrease in cancer cell proliferation and survival. SHP1 suppresses angiogenesis [17,18] and enhances the antitumor efficacy of chemotherapeutic agents [16,19]. SHP1 also inhibits inflammation [20] SHP1 dephosphorylates the signal transducer and activator of transcription 3 (STAT3) [10], extracellular signal-regulated kinase (ERK) [11], protein kinase B (Akt) [12], and other signaling components [13][14][15][16], thereby contributing to a decrease in cancer cell proliferation and survival. SHP1 suppresses angiogenesis [17,18] and enhances the antitumor efficacy of chemotherapeutic agents [16,19]. SHP1 also inhibits inflammation [20] and osteoclastogenesis [21,22], suggesting that it has therapeutic potential for treating inflammation-mediated diseases and bone diseases. Thus, targeting SHP1 activation could represent a promising therapeutic approach for the above diseases and open a new avenue for PTP drug discovery.

Results and Discussion
We initiated our efforts by investigating C-3 substituents. Usually, compounds containing the aliphatic -NH-group have better water solubility and stronger hydrogenbonding ability. We replaced the C-3 hydroxyl group with an amino group to afford the compounds 5aa-af (Scheme 1). Lithocholic acid (6) was subjected to esterification and oxidation via a combination of 2-iodoxybenzoic acid (IBX) and trifluoroacetic acid to afford α,β-unsaturated ketone compound 8. Compound 9 was obtained by C-4 methylation accompanied by a rearrangement of the double bond. Reductive amination of the C-3 carbonyl group afforded compound 10. Compound 5aa was prepared by reducing the C-24 carboxylate group of compound 10. The N-alkylation of compound 10 and the subsequent LiAlH4 reduction formed compounds 5ab-af. Lithocholic acid derivatives, which contain the above steroid scaffold, have been found to inhibit the growth of cancer cells [42][43][44][45]. Two lithocholic acid derivatives, ursodeoxycholic acid and obeticholic acid, are approved drugs for the treatment of primary biliary cholangitis. Furthermore, obeticholic acid is currently in clinical trials for nonalcoholic steatohepatitis (NASH) as a farnesoid X receptor (FXR) agonist. Here, we replaced the carboxyl group of lithocholic acid with a hydroxyl group, modified the C-3, C-7 positions to obtain a series of novel structures, and evaluated their biological activities for SHP1 activation.

Results and Discussion
We initiated our efforts by investigating C-3 substituents. Usually, compounds containing the aliphatic -NH-group have better water solubility and stronger hydrogen-bonding ability. We replaced the C-3 hydroxyl group with an amino group to afford the compounds 5aa-af (Scheme 1). Lithocholic acid (6) was subjected to esterification and oxidation via a combination of 2-iodoxybenzoic acid (IBX) and trifluoroacetic acid to afford α,β-unsaturated ketone compound 8. Compound 9 was obtained by C-4 methylation accompanied by a rearrangement of the double bond. Reductive amination of the C-3 carbonyl group afforded compound 10. Compound 5aa was prepared by reducing the C-24 carboxylate group of compound 10. The N-alkylation of compound 10 and the subsequent LiAlH 4 reduction formed compounds 5ab-af. The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate. The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate. The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate. The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate.  The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate.  The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate.  The SHP1-activating effects of the synthetic compounds were examined by the 6,8difluoro-4-methylumbelliferyl phosphate (DiFMUP) assay. As shown in Table 1, the compounds 5aa-ac bearing the C-3 hydrophilic amino or extended hydroxyl groups showed better activity. The C-3 alkylamino substitution of compounds 5ad-af showed a decline in the activity. These results indicated that the C-3 free amino group benefits activity. Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate.  Given that C-3 free amino substitution appeared to have a better effect on potency, we next synthesized 3-amino compounds, 5ag-ak, with different C-7 substituents (Schemes 2 and 3). Compound 5aa underwent C-3 Boc protection and C-24 TBS protection to afford compound 13, which then oxidized into compound 14 using sodium dichromate. Compound 5ag was finally produced through deprotection. Compounds 5ah-ai were prepared by C-7 oximation or the reduction of compound 5ag. Compound 10 underwent C-3 Boc protection and C-7 oxidation to afford compound 16. The reductive amination of the C-7 carbonyl group resulted in compound 17. Compound 5aj was prepared by the C-3 deprotection and C-24 reduction of compound 17. The treatment of compound 16 with NaBH 4 gave compound 19. The reaction with NaH and MeI failed to In addition, we replaced the hydroxyl group at C-24 with amide or amine to form compounds 5al-am (Scheme 4). Compound 15 was hydrolyzed into carboxylic acid 22, which then reacted with methylamine hydrochloride in the presence of N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDCI) to afford amide 23. Compound 5al was finally produced through deprotection. Compound 12 was transformed into aldehyde 24 by IBX oxidation. The subsequent reductive amination of 24 with n-propylamine and C-3 deprotection resulted in compound 5am.  In addition, we replaced the hydroxyl group at C-24 with amide or amine to form compounds 5al-am (Scheme 4). Compound 15 was hydrolyzed into carboxylic acid 22, which then reacted with methylamine hydrochloride in the presence of N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDCI) to afford amide 23. Compound 5al was finally produced through deprotection. Compound 12 was transformed into aldehyde 24 by IBX oxidation. The subsequent reductive amination of 24 with n-propylamine and C-3 deprotection resulted in compound 5am. We envisioned that the introduction of secondary/tertiary alcohol groups into the side chain would probably increase the activity. The compounds 5an-av were designed and synthesized, as described in Scheme 5. The treatment of compounds 24 and 15 with Grignard reagents, followed by C-3 deprotection, furnished compounds 5an-av.   In addition, we replaced the hydroxyl group at C-24 with amide or amine to form compounds 5al-am (Scheme 4). Compound 15 was hydrolyzed into carboxylic acid 22, which then reacted with methylamine hydrochloride in the presence of N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDCI) to afford amide 23. Compound 5al was finally produced through deprotection. Compound 12 was transformed into aldehyde 24 by IBX oxidation. The subsequent reductive amination of 24 with n-propylamine and C-3 deprotection resulted in compound 5am. We envisioned that the introduction of secondary/tertiary alcohol groups into the side chain would probably increase the activity. The compounds 5an-av were designed and synthesized, as described in Scheme 5. The treatment of compounds 24 and 15 with Grignard reagents, followed by C-3 deprotection, furnished compounds 5an-av. We envisioned that the introduction of secondary/tertiary alcohol groups into the side chain would probably increase the activity. The compounds 5an-av were designed and synthesized, as described in Scheme 5. The treatment of compounds 24 and 15 with Grignard reagents, followed by C-3 deprotection, furnished compounds 5an-av. To verify the influence of changing the side chain length on a compound's biological activity, we removed two ethylene spacers between the hydroxyl group and the remainder of the scaffold to design compound 5aw (Scheme 6). Commercially available 21-hydroxy-20-methylpregn-4-en-3-one (28) was initially protected with an Ac-group, methylated with KO t Bu and MeI to furnish compound 30. The following reductive amination and C-22 deprotection resulted in compound 5aw. To verify the influence of changing the side chain length on a compound's biological activity, we removed two ethylene spacers between the hydroxyl group and the remainder of the scaffold to design compound 5aw (Scheme 6). Commercially available 21-hydroxy- To verify the influence of changing the side chain length on a compound's biological activity, we removed two ethylene spacers between the hydroxyl group and the remainder of the scaffold to design compound 5aw (Scheme 6). Commercially available 21-hydroxy-20-methylpregn-4-en-3-one (28) was initially protected with an Ac-group, methylated with KO t Bu and MeI to furnish compound 30. The following reductive amination and C-22 deprotection resulted in compound 5aw.  Table 3 shows that the replacement of the C-24 hydroxyl group with an amide or alkylamino moiety (compounds 5al-am) attenuated the activity. Most of the secondary and tertiary alcohols exerted good efficacies, with EC50 values from 12.88 μM to 38.11 μM and 5.73-to 6.94-fold maximum activation. Ethyl substitution (compounds 5ao and 5at) exhibited more potent biological activity than methyl substitution (compounds 5an and 5as). The activation property of compound 5aw was decreased compared to compound 5aa. It appeared that compounds with different side chain lengths affected the activity. These data implied that the C-24 terminal position deserves further modification, and the introduction of proper groups may increase biological activity.  Table 3 shows that the replacement of the C-24 hydroxyl group with an amide or alkylamino moiety (compounds 5al-am) attenuated the activity. Most of the secondary and tertiary alcohols exerted good efficacies, with EC 50 values from 12.88 µM to 38.11 µM and 5.73-to 6.94-fold maximum activation. Ethyl substitution (compounds 5ao and 5at) exhibited more potent biological activity than methyl substitution (compounds 5an and 5as). The activation property of compound 5aw was decreased compared to compound 5aa. It appeared that compounds with different side chain lengths affected the activity. These data implied that the C-24 terminal position deserves further modification, and the introduction of proper groups may increase biological activity.       Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged 38  Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged structures 5ax-ba. The syntheses of compounds 5ax-ba are outlined in Schemes 7 and 8.  Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged structures 5ax-ba. The syntheses of compounds 5ax-ba are outlined in Schemes 7 and 8.  Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged structures 5ax-ba. The syntheses of compounds 5ax-ba are outlined in Schemes 7 and 8. Treating compound 32 with isocyanate compound 33 produced diphenyl urea intermedi-  Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged structures 5ax-ba. The syntheses of compounds 5ax-ba are outlined in Schemes 7 and 8. Treating compound 32 with isocyanate compound 33 produced diphenyl urea intermediate 34. Compound 5ax was made in one step from compounds 5aa and 34 in the presence 12.88 ± 0. 30 6.50 ± 0.12

5au
Molecules 2023, 28, x FOR PEER REVIEW 7 of 32 Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged structures 5ax-ba. The syntheses of compounds 5ax-ba are outlined in Schemes 7 and 8. Treating compound 32 with isocyanate compound 33 produced diphenyl urea intermediate 34. Compound 5ax was made in one step from compounds 5aa and 34 in the presence of Et3N. Compound 12 underwent C-24 bromination, coupling with compound 35, and  Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged structures 5ax-ba. The syntheses of compounds 5ax-ba are outlined in Schemes 7 and 8. Treating compound 32 with isocyanate compound 33 produced diphenyl urea intermediate 34. Compound 5ax was made in one step from compounds 5aa and 34 in the presence of Et3N. Compound 12 underwent C-24 bromination, coupling with compound 35, and 35 Urea exhibits unique hydrogen-binding capabilities and becomes an important functional group for drug-target interactions. We employed the rational design strategy of incorporating a diphenyl urea active substructure into the steroids to obtain the merged structures 5ax-ba. The syntheses of compounds 5ax-ba are outlined in Schemes 7 and 8. Treating compound 32 with isocyanate compound 33 produced diphenyl urea intermediate 34.
Compound 5ax was made in one step from compounds 5aa and 34 in the presence of Et 3 N. Compound 12 underwent C-24 bromination, coupling with compound 35, and then deprotection to generate compound 5ay. Compound 14 was selectively deprotected and brominated with CBr4 and PPh3 to furnish compound 38. Next, coupling with compound 34, followed by C-7 NaBH4 reduction, yielded compound 40. Finally, removal of the Boc protecting group resulted in compound 5az. The C-3 protection and C-22 bromination of compound 5aw gave compound 42. The subsequent coupling and deprotection afforded compound 5ba.
It is evident from Table 4 that the position in the steroid scaffold for the urea substituents specifies the SHP1 activity. With the placing of 4-chloro-3-(trifluoromethyl)phenyl urea in the C-3 position, the compound 5ax showed no activity. As the diphenyl urea group was placed in the C-24 position and the other positions were kept constant, the activating effects of compounds 5ay-ba were increased several-fold compared to those of compound 5aa. These data support the idea that the urea group probably contacts with both the hydrophilic and the hydrophobic halves of the SHP1 active site and that a flexible Compound 14 was selectively deprotected and brominated with CBr4 and PPh3 to furnish compound 38. Next, coupling with compound 34, followed by C-7 NaBH4 reduction, yielded compound 40. Finally, removal of the Boc protecting group resulted in compound 5az. The C-3 protection and C-22 bromination of compound 5aw gave compound 42. The subsequent coupling and deprotection afforded compound 5ba.
It is evident from Table 4 that the position in the steroid scaffold for the urea substituents specifies the SHP1 activity. With the placing of 4-chloro-3-(trifluoromethyl)phenyl urea in the C-3 position, the compound 5ax showed no activity. As the diphenyl urea group was placed in the C-24 position and the other positions were kept constant, the activating effects of compounds 5ay-ba were increased several-fold compared to those of compound 5aa. These data support the idea that the urea group probably contacts with both the hydrophilic and the hydrophobic halves of the SHP1 active site and that a flexible Compound 14 was selectively deprotected and brominated with CBr 4 and PPh 3 to furnish compound 38. Next, coupling with compound 34, followed by C-7 NaBH 4 reduction, yielded compound 40. Finally, removal of the Boc protecting group resulted in compound 5az. The C-3 protection and C-22 bromination of compound 5aw gave compound 42. The subsequent coupling and deprotection afforded compound 5ba.
It is evident from Table 4 that the position in the steroid scaffold for the urea substituents specifies the SHP1 activity. With the placing of 4-chloro-3-(trifluoromethyl)phenyl urea in the C-3 position, the compound 5ax showed no activity. As the diphenyl urea group was placed in the C-24 position and the other positions were kept constant, the activating effects of compounds 5ay-ba were increased several-fold compared to those of compound 5aa. These data support the idea that the urea group probably contacts with both the hydrophilic and the hydrophobic halves of the SHP1 active site and that a flexible side chain may enhance proper orientation. Interestingly, the 7-hydroxyl compound 5az displayed 3-fold higher maximum activation than compound 5ay though their EC 50 values were similar. Compound 5ba showed better activity than compound 5ay, which indicates that the short length of the side chain benefits the activity. In particular, 5ba is more than 32-fold more selective for SHP1 over the closest homologue SHP2 (EC 50 > 50 µM). In order to ameliorate the solubility of the most potent compound, 5ba, this class of derivatives was prepared in a hydrochloride salt form for further modifications. side chain may enhance proper orientation. Interestingly, the 7-hydroxyl compound 5az displayed 3-fold higher maximum activation than compound 5ay though their EC50 values were similar. Compound 5ba showed better activity than compound 5ay, which indicates that the short length of the side chain benefits the activity. In particular, 5ba is more than 32-fold more selective for SHP1 over the closest homologue SHP2 (EC50 > 50 μM). In order to ameliorate the solubility of the most potent compound, 5ba, this class of derivatives was prepared in a hydrochloride salt form for further modifications. The structure-activity relationships (SARs) for the 3-amino-4,4-dimethyl lithocholic acid derivatives are summarized below. The presence of a primary amine at the C-3 position (5aa) seems to be important for the interaction of the compound with the binding site of SHP1. The substitutions of the aliphatic amine groups at the C-3 position (5ab-af) exhibited lower activities, and the compound with bulky phenyl urea substitution at the C-3 position (5ax) was inactive, indicating that the binding site cavity of SHP1 near the C-3 position is relatively small. The compounds with substitutions at the C-7 position (5ag-aj) had weaker potency. The primary alcohol moiety in a side chain can be replaced by other chemical groups while maintaining or increasing potency. The compounds with diphenyl urea moieties in a side chain (5ay-ba) markedly increased their potency, indicating that the hydrogen bond and hydrophobic interactions play an important role in the binding pocket.
The anti-tumor effects of compounds 5az-ba which showed higher maximum activation ability were examined by cell viability assay using acute lymphoblastic leukemia (ALL) cell line RS4;11, acute promyelocytic leukemia (APL) cell line NB4, and non-small cell lung cancer (NSCLC) cell line NCI-H1299. As shown in Table 5, 5az-ba exhibited significant cytotoxicity to effectively inhibit cancer cell viability; this was particularly the case with compound 5az against RS4;11 cells, with an IC50 value of less than 2 μM. side chain may enhance proper orientation. Interestingly, the 7-hydroxyl compound 5az displayed 3-fold higher maximum activation than compound 5ay though their EC50 values were similar. Compound 5ba showed better activity than compound 5ay, which indicates that the short length of the side chain benefits the activity. In particular, 5ba is more than 32-fold more selective for SHP1 over the closest homologue SHP2 (EC50 > 50 μM). In order to ameliorate the solubility of the most potent compound, 5ba, this class of derivatives was prepared in a hydrochloride salt form for further modifications. The structure-activity relationships (SARs) for the 3-amino-4,4-dimethyl lithocholic acid derivatives are summarized below. The presence of a primary amine at the C-3 position (5aa) seems to be important for the interaction of the compound with the binding site of SHP1. The substitutions of the aliphatic amine groups at the C-3 position (5ab-af) exhibited lower activities, and the compound with bulky phenyl urea substitution at the C-3 position (5ax) was inactive, indicating that the binding site cavity of SHP1 near the C-3 position is relatively small. The compounds with substitutions at the C-7 position (5ag-aj) had weaker potency. The primary alcohol moiety in a side chain can be replaced by other chemical groups while maintaining or increasing potency. The compounds with diphenyl urea moieties in a side chain (5ay-ba) markedly increased their potency, indicating that the hydrogen bond and hydrophobic interactions play an important role in the binding pocket.
The anti-tumor effects of compounds 5az-ba which showed higher maximum activation ability were examined by cell viability assay using acute lymphoblastic leukemia (ALL) cell line RS4;11, acute promyelocytic leukemia (APL) cell line NB4, and non-small cell lung cancer (NSCLC) cell line NCI-H1299. As shown in Table 5, 5az-ba exhibited significant cytotoxicity to effectively inhibit cancer cell viability; this was particularly the case with compound 5az against RS4;11 cells, with an IC50 value of less than 2 μM. side chain may enhance proper orientation. Interestingly, the 7-hydroxyl compound 5az displayed 3-fold higher maximum activation than compound 5ay though their EC50 values were similar. Compound 5ba showed better activity than compound 5ay, which indicates that the short length of the side chain benefits the activity. In particular, 5ba is more than 32-fold more selective for SHP1 over the closest homologue SHP2 (EC50 > 50 μM). In order to ameliorate the solubility of the most potent compound, 5ba, this class of derivatives was prepared in a hydrochloride salt form for further modifications. The structure-activity relationships (SARs) for the 3-amino-4,4-dimethyl lithocholic acid derivatives are summarized below. The presence of a primary amine at the C-3 position (5aa) seems to be important for the interaction of the compound with the binding site of SHP1. The substitutions of the aliphatic amine groups at the C-3 position (5ab-af) exhibited lower activities, and the compound with bulky phenyl urea substitution at the C-3 position (5ax) was inactive, indicating that the binding site cavity of SHP1 near the C-3 position is relatively small. The compounds with substitutions at the C-7 position (5ag-aj) had weaker potency. The primary alcohol moiety in a side chain can be replaced by other chemical groups while maintaining or increasing potency. The compounds with diphenyl urea moieties in a side chain (5ay-ba) markedly increased their potency, indicating that the hydrogen bond and hydrophobic interactions play an important role in the binding pocket.
The anti-tumor effects of compounds 5az-ba which showed higher maximum activation ability were examined by cell viability assay using acute lymphoblastic leukemia (ALL) cell line RS4;11, acute promyelocytic leukemia (APL) cell line NB4, and non-small cell lung cancer (NSCLC) cell line NCI-H1299. As shown in Table 5, 5az-ba exhibited significant cytotoxicity to effectively inhibit cancer cell viability; this was particularly the case with compound 5az against RS4;11 cells, with an IC50 value of less than 2 μM. side chain may enhance proper orientation. Interestingly, the 7-hydroxyl compound 5az displayed 3-fold higher maximum activation than compound 5ay though their EC50 values were similar. Compound 5ba showed better activity than compound 5ay, which indicates that the short length of the side chain benefits the activity. In particular, 5ba is more than 32-fold more selective for SHP1 over the closest homologue SHP2 (EC50 > 50 μM). In order to ameliorate the solubility of the most potent compound, 5ba, this class of derivatives was prepared in a hydrochloride salt form for further modifications. The structure-activity relationships (SARs) for the 3-amino-4,4-dimethyl lithocholic acid derivatives are summarized below. The presence of a primary amine at the C-3 position (5aa) seems to be important for the interaction of the compound with the binding site of SHP1. The substitutions of the aliphatic amine groups at the C-3 position (5ab-af) exhibited lower activities, and the compound with bulky phenyl urea substitution at the C-3 position (5ax) was inactive, indicating that the binding site cavity of SHP1 near the C-3 position is relatively small. The compounds with substitutions at the C-7 position (5ag-aj) had weaker potency. The primary alcohol moiety in a side chain can be replaced by other chemical groups while maintaining or increasing potency. The compounds with diphenyl urea moieties in a side chain (5ay-ba) markedly increased their potency, indicating that the hydrogen bond and hydrophobic interactions play an important role in the binding pocket.
The anti-tumor effects of compounds 5az-ba which showed higher maximum activation ability were examined by cell viability assay using acute lymphoblastic leukemia (ALL) cell line RS4;11, acute promyelocytic leukemia (APL) cell line NB4, and non-small cell lung cancer (NSCLC) cell line NCI-H1299. As shown in Table 5, 5az-ba exhibited significant cytotoxicity to effectively inhibit cancer cell viability; this was particularly the case with compound 5az against RS4;11 cells, with an IC50 value of less than 2 μM. side chain may enhance proper orientation. Interestingly, the 7-hydroxyl compound 5az displayed 3-fold higher maximum activation than compound 5ay though their EC50 values were similar. Compound 5ba showed better activity than compound 5ay, which indicates that the short length of the side chain benefits the activity. In particular, 5ba is more than 32-fold more selective for SHP1 over the closest homologue SHP2 (EC50 > 50 μM). In order to ameliorate the solubility of the most potent compound, 5ba, this class of derivatives was prepared in a hydrochloride salt form for further modifications. The structure-activity relationships (SARs) for the 3-amino-4,4-dimethyl lithocholic acid derivatives are summarized below. The presence of a primary amine at the C-3 position (5aa) seems to be important for the interaction of the compound with the binding site of SHP1. The substitutions of the aliphatic amine groups at the C-3 position (5ab-af) exhibited lower activities, and the compound with bulky phenyl urea substitution at the C-3 position (5ax) was inactive, indicating that the binding site cavity of SHP1 near the C-3 position is relatively small. The compounds with substitutions at the C-7 position (5ag-aj) had weaker potency. The primary alcohol moiety in a side chain can be replaced by other chemical groups while maintaining or increasing potency. The compounds with diphenyl urea moieties in a side chain (5ay-ba) markedly increased their potency, indicating that the hydrogen bond and hydrophobic interactions play an important role in the binding pocket.
The anti-tumor effects of compounds 5az-ba which showed higher maximum activation ability were examined by cell viability assay using acute lymphoblastic leukemia (ALL) cell line RS4;11, acute promyelocytic leukemia (APL) cell line NB4, and non-small cell lung cancer (NSCLC) cell line NCI-H1299. As shown in Table 5, 5az-ba exhibited significant cytotoxicity to effectively inhibit cancer cell viability; this was particularly the case with compound 5az against RS4;11 cells, with an IC50 value of less than 2 μM. side chain may enhance proper orientation. Interestingly, the 7-hydroxyl compound 5az displayed 3-fold higher maximum activation than compound 5ay though their EC50 values were similar. Compound 5ba showed better activity than compound 5ay, which indicates that the short length of the side chain benefits the activity. In particular, 5ba is more than 32-fold more selective for SHP1 over the closest homologue SHP2 (EC50 > 50 μM). In order to ameliorate the solubility of the most potent compound, 5ba, this class of derivatives was prepared in a hydrochloride salt form for further modifications. The structure-activity relationships (SARs) for the 3-amino-4,4-dimethyl lithocholic acid derivatives are summarized below. The presence of a primary amine at the C-3 position (5aa) seems to be important for the interaction of the compound with the binding site of SHP1. The substitutions of the aliphatic amine groups at the C-3 position (5ab-af) exhibited lower activities, and the compound with bulky phenyl urea substitution at the C-3 position (5ax) was inactive, indicating that the binding site cavity of SHP1 near the C-3 position is relatively small. The compounds with substitutions at the C-7 position (5ag-aj) had weaker potency. The primary alcohol moiety in a side chain can be replaced by other chemical groups while maintaining or increasing potency. The compounds with diphenyl urea moieties in a side chain (5ay-ba) markedly increased their potency, indicating that the hydrogen bond and hydrophobic interactions play an important role in the binding pocket.
The anti-tumor effects of compounds 5az-ba which showed higher maximum activation ability were examined by cell viability assay using acute lymphoblastic leukemia (ALL) cell line RS4;11, acute promyelocytic leukemia (APL) cell line NB4, and non-small cell lung cancer (NSCLC) cell line NCI-H1299. As shown in Table 5, 5az-ba exhibited significant cytotoxicity to effectively inhibit cancer cell viability; this was particularly the case with compound 5az against RS4;11 cells, with an IC50 value of less than 2 μM. The structure-activity relationships (SARs) for the 3-amino-4,4-dimethyl lithocholic acid derivatives are summarized below. The presence of a primary amine at the C-3 position (5aa) seems to be important for the interaction of the compound with the binding site of SHP1. The substitutions of the aliphatic amine groups at the C-3 position (5ab-af) exhibited lower activities, and the compound with bulky phenyl urea substitution at the C-3 position (5ax) was inactive, indicating that the binding site cavity of SHP1 near the C-3 position is relatively small. The compounds with substitutions at the C-7 position (5ag-aj) had weaker potency. The primary alcohol moiety in a side chain can be replaced by other chemical groups while maintaining or increasing potency. The compounds with diphenyl urea moieties in a side chain (5ay-ba) markedly increased their potency, indicating that the hydrogen bond and hydrophobic interactions play an important role in the binding pocket.
The anti-tumor effects of compounds 5az-ba which showed higher maximum activation ability were examined by cell viability assay using acute lymphoblastic leukemia (ALL) cell line RS4;11, acute promyelocytic leukemia (APL) cell line NB4, and non-small cell lung cancer (NSCLC) cell line NCI-H1299. As shown in Table 5, 5az-ba exhibited significant cytotoxicity to effectively inhibit cancer cell viability; this was particularly the case with compound 5az against RS4;11 cells, with an IC 50 value of less than 2 µM. Based on these activity data, we selected the crystal structure of SHP1 in the open conformation (PDB ID: 3PS5) and the 3-β-diastereomer of 5ba for the docking study (see Supplementary Materials). Compound 5ba binds at the central allosteric site through a network of hydrogen bonds and hydrophobic interactions (Figure 2). More importantly, this site is closest to the reported interface of three domains [46], and the binding of compounds tightens the inter-domain connection, which thereby stabilizes the active conformation of SHP1.
Molecules 2023, 28, x FOR PEER REVIEW 10 of 32 Based on these activity data, we selected the crystal structure of SHP1 in the open conformation (PDB ID: 3PS5) and the 3-β-diastereomer of 5ba for the docking study (see Supplementary Materials). Compound 5ba binds at the central allosteric site through a network of hydrogen bonds and hydrophobic interactions (Figure 2). More importantly, this site is closest to the reported interface of three domains [46], and the binding of compounds tightens the inter-domain connection, which thereby stabilizes the active conformation of SHP1. The binding pocket is mainly composed of a hydrophobic pocket and a hydrophilic pocket ( Figure 3). The narrow hydrophobic pocket is made up of Val2, Trp4, Phe27, and Val76, with Met1 as gatekeeper. The depth of the hydrophobic pocket is approximately 10 Å, and the width is approximately 6 Å. Four hydrophilic residues, namely Arg217, Asp481, Asp483, and Lys486, embrace the larger hydrophilic pocket from one side. The hydrophilic pocket provides the interaction between the protein and the amino head of the compound. Compound 5ba possesses an L-shaped structure, and the phenyl urea tail fits well in the hydrophobic pocket ( Figure 4). The binding pocket is mainly composed of a hydrophobic pocket and a hydrophilic pocket (Figure 3). The narrow hydrophobic pocket is made up of Val2, Trp4, Phe27, and Val76, with Met1 as gatekeeper. The depth of the hydrophobic pocket is approximately 10 Å, and the width is approximately 6 Å. Four hydrophilic residues, namely Arg217, Asp481, Asp483, and Lys486, embrace the larger hydrophilic pocket from one side. The hydrophilic pocket provides the interaction between the protein and the amino head of the compound. Compound 5ba possesses an L-shaped structure, and the phenyl urea tail fits well in the hydrophobic pocket ( Figure 4).
As shown in Figure 5, the 4-chloro-3-(trifluoromethyl)phenyl ring occupies the hydrophobic pocket through hydrophobic interactions with Met1, Phe27, Val76, and Glu77. The urea group of 5ba forms two hydrogen bonds with Thr80 (3.7 Å) and Lys97 (2.9 Å), which could explain why the potency of 5ba increased ∼23 times than 5aw. The phenoxy group forms a hydrogen bond with Gln81 (3.1 Å) and forms a hydrophobic interaction with Thr80. The 3-amino group of the steroid parent nucleus creates two hydrogen bonds with Asn472 (1.9, 2.9 Å), which accounts for the potency of the series of newly synthesized compounds. The C-4 methyl group forms a hydrophobic interaction with Asp481. The C-21 methyl group forms a hydrophobic interaction with Pro314.   As shown in Figure 5, the 4-chloro-3-(trifluoromethyl)phenyl ring occupies the hydrophobic pocket through hydrophobic interactions with Met1, Phe27, Val76, and Glu77. The urea group of 5ba forms two hydrogen bonds with Thr80 (3.7 Å) and Lys97 (2.9 Å), which could explain why the potency of 5ba increased ∼23 times than 5aw. The phenoxy group forms a hydrogen bond with Gln81 (3.1 Å) and forms a hydrophobic interaction with Thr80. The 3-amino group of the steroid parent nucleus creates two hydrogen bonds with Asn472 (1.9, 2.9 Å), which accounts for the potency of the series of newly synthesized compounds. The C-4 methyl group forms a hydrophobic interaction with Asp481. The C-21 methyl group forms a hydrophobic interaction with Pro314.   As shown in Figure 5, the 4-chloro-3-(trifluoromethyl)phenyl ring occupies the hydrophobic pocket through hydrophobic interactions with Met1, Phe27, Val76, and Glu77. The urea group of 5ba forms two hydrogen bonds with Thr80 (3.7 Å) and Lys97 (2.9 Å), which could explain why the potency of 5ba increased ∼23 times than 5aw. The phenoxy group forms a hydrogen bond with Gln81 (3.1 Å) and forms a hydrophobic interaction with Thr80. The 3-amino group of the steroid parent nucleus creates two hydrogen bonds with Asn472 (1.9, 2.9 Å), which accounts for the potency of the series of newly synthesized compounds. The C-4 methyl group forms a hydrophobic interaction with Asp481. The C-21 methyl group forms a hydrophobic interaction with Pro314. Structure modification at the C-3 position (5ab-af) dramatically attenuated the potency, suggesting that -NH 2 is the key group for activity retention. This could also be explained by the reduced hydrogen bond interactions of the amino group with Asn472. Those compounds lacking a urea group only occupy and bind the entrance of the pocket with less interactions, leading to reduced potency compared to 5ay-ba. The three methyl groups at C-4 and C-19 incur steric hindrance to the urea 3-NH of compound 5ax, making it energetically unfavorable to engage Asn472, which is consistent with the loss of potency. As shown in Figure 6A, Thr80, Gln81, and Lys97 are on the surface of the auto-inhibited SHP1, which allows the free access and binding of compounds. The long and flexible BG loop plays a pivotal role in the conformational change and activation of SH2 domaincontaining proteins [46,47]. The positively charged head group of the compound may interact with the negatively charged LQDRDG motif on the BG loop of the N-SH2 domain, which causes movement of the BG loop. Then, the "switch" N-SH2 domain shifts away, moving from one side of the PTP domain to the other, leading to open conformation and thus the activation of SHP1. In contrast, no similar pockets were observed in the center of the homologue SHP2 in open conformation ( Figure 6B). These observations probably explain the selectivity of 5ba for SHP1 over SHP2. Structure modification at the C-3 position (5ab-af) dramatically attenuated the potency, suggesting that -NH2 is the key group for activity retention. This could also be explained by the reduced hydrogen bond interactions of the amino group with Asn472. Those compounds lacking a urea group only occupy and bind the entrance of the pocket with less interactions, leading to reduced potency compared to 5ay-ba. The three methyl groups at C-4 and C-19 incur steric hindrance to the urea 3-NH of compound 5ax, making it energetically unfavorable to engage Asn472, which is consistent with the loss of potency. As shown in Figure 6A, Thr80, Gln81, and Lys97 are on the surface of the auto-inhibited SHP1, which allows the free access and binding of compounds. The long and flexible BG loop plays a pivotal role in the conformational change and activation of SH2 domain-containing proteins [46,47]. The positively charged head group of the compound may interact with the negatively charged LQDRDG motif on the BG loop of the N-SH2 domain, which causes movement of the BG loop. Then, the "switch" N-SH2 domain shifts away, moving from one side of the PTP domain to the other, leading to open conformation and thus the activation of SHP1. In contrast, no similar pockets were observed in the center of the homologue SHP2 in open conformation ( Figure 6B). These observations probably explain the selectivity of 5ba for SHP1 over SHP2.

Chemical Synthesis
All reagents and solvents were obtained from commercial suppliers and used without further purification unless otherwise indicated. Melting points (mps) were taken in open capillaries on a WRS-2 melting point system. The 1 H NMR and 13 C NMR spectra were recorded using TMS as the internal standard on a Bruker Ascend 400 spectrometer at 400 and 100 MHz, respectively. The 1 H NMR chemical shifts were reported in parts per million (ppm) relative to the centerline of the singlet signal of the solvent molecule (7.26 ppm for CDCl3 or 3.31 ppm for CD3OD). The 13 C NMR chemical shifts were reported in ppm relative to the centerline at 77.16 ppm for CDCl3 or at 49.00 ppm for CD3OD. The data were reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, dd = doublet of doublets, m = multiplet), coupling constant (Hz), and integration. Evaporation was carried out in vacuo using a rotating evaporator. Silica gel chromatography was performed using silica gel (200-300 mesh). Reactions were monitored by TLC with detection by phosphomolybdic acid visualization or UV light (λ = 254 nm). All reac-

Chemical Synthesis
All reagents and solvents were obtained from commercial suppliers and used without further purification unless otherwise indicated. Melting points (mps) were taken in open capillaries on a WRS-2 melting point system. The 1 H NMR and 13 C NMR spectra were recorded using TMS as the internal standard on a Bruker Ascend 400 spectrometer at 400 and 100 MHz, respectively. The 1 H NMR chemical shifts were reported in parts per million (ppm) relative to the centerline of the singlet signal of the solvent molecule (7.26 ppm for CDCl 3 or 3.31 ppm for CD 3 OD). The 13 C NMR chemical shifts were reported in ppm relative to the centerline at 77.16 ppm for CDCl 3 or at 49.00 ppm for CD 3 OD. The data were reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, dd = doublet of doublets, m = multiplet), coupling constant (Hz), and integration. Evaporation was carried out in vacuo using a rotating evaporator. Silica gel chromatography was performed using silica gel (200-300 mesh). Reactions were monitored by TLC with detection by phosphomolybdic acid visualization or UV light (λ = 254 nm). All reactions involving air-or moisture-sensitive reagents were performed under a nitrogen atmosphere using anhydrous solvents. The purity of the final compounds was >95%, as deduced by 1 H NMR spectra. High-resolution mass spectroscopy (HRMS) was performed on a time-of-flight instrument with electrospray ionization (ESI) in the positive ionization mode. 10S,13R,17R)-3-hydroxy-10,13-dimethylhexadecahydro-1Hcyclopenta[a]phenanthren-17-yl)pentanoate (7) TsOH . H 2 O (47.5 mg, 0.25 mmol) was added to a stirred solution of lithocholic acid 6 (1.88 g, 5 mmol) in MeOH (20 mL). The mixture was refluxed for 2 h, and TLC indicated the consumption of the starting material. MeOH was removed in vacuo, and aqueous NaHCO 3 solution was added. The mixture was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, and concentrated to afford compound 7 (1.95 g, 100% yield) as a white solid. The mp was 105.3-106.5 • C. in DMSO (20 mL) were combined. The mixture was warmed to 50 • C, and all the solids were dissolved. After stirring for 2 h under a N 2 atmosphere, the reaction was completed, as indicated by TLC. The reaction mixture was cooled and added to an aqueous NaHSO 3 solution (20 mL). The mixture was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with water, brine, dried over Na 2 SO 4 , filtered, and concentrated. The crude material was purified by silica gel column chromatography using PE/EtOAc (10:1, v/v) to afford compound 8 (1.02 g, 53% yield) as a white solid. The mp was 124.2-126.1 • C. Potassium tert-butoxide (1.18 g, 10.56 mmol) was added in batches to the suspension of compound 8 (1.02 g, 2.64 mmol) in absolute tert-butyl alcohol (50 mL). The mixture was stirred at r.t. for 30 min to form a clear yellow solution and then added to MeI (1.5 g, 10.56 mmol) dropwise. The reaction was stirred at r.t. for 12 h in the dark under a N 2 atmosphere, and the starting material disappeared, as monitored by TLC. Upon addition of an aqueous Na 2 S 2 O 3 solution (5 mL) for quenching the excess MeI, the mixture was evaporated and water (30 mL) was added. The suspension was acidified to pH < 6 with an aqueous HCl solution. The mixture was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, and concentrated. The crude material was dissolved in MeOH (20 mL), and TsOH . H 2 O (47.5 mg, 0.25 mmol) was added. The mixture was refluxed for 2 h, and the MeOH was removed in vacuo. An aqueous NaHCO 3 solution was added, and it was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, concentrated, and purified by silica gel column chromatography using PE/EtOAc (10:1, v/v) to obtain compound 9 (547 mg, 50% yield) as a white solid. The mp was 133.6-135.8 • C. 1 4-((10R,13R,17R)-3-amino-4,4,10,13-tetramethyl-tetradecahydro-1Hcyclopenta[a]phenanthren-17-yl)pentanoate (10) Compound 9 (547 mg, 1.32 mmol) was dissolved in a saturated NH 4 OAc/EtOH solution (20 mL). It was added to NaBH 3 CN (166 mg, 2.64 mmol) and NH 3 . H 2 O (0.8 mL). The mixture was refluxed for 12 h under a N 2 atmosphere, and TLC indicated the consumption of the starting material. The solvents were removed in vacuo and water (20 mL) was added. The mixture was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, concentrated, and purified by silica gel column chromatography using DCM/MeOH/ LiAlH 4 (11 mg, 0.3 mmol) was added in batches at 0 • C to a dry THF solution (20 mL) of compound 10 (116 mg, 0.28 mmol). The mixture was stirred at r.t. for 12 h under a N 2 atmosphere. Water (11 µL) was added to quench the reaction, then a 15% NaOH solution (11 µL) and then water (33 µL). The suspension was filtrated, and the solid residue was washed with THF (10 mL). The solvents were removed in vacuo, and water (30 mL) was added to the residue. The mixture was extracted with DCM (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, concentrated, and purified by silica gel column chromatography using DCM/MeOH/NH 3 . H 2 O (100:5:0.5, v/v/v) to obtain compound 5aa (81 mg, 75% yield) as a white solid. The mp was 163.5-166.3 • C.        4-((10R,13R,17R)-3-((3-chloropropyl)amino)-4,4,10,13-tetramethyltetradecahydro-1H-cyclopenta[a]phenanthren-17-yl)pentan-1-ol (5ad) Compound 5ad was synthesized in a 79% yield as a white solid using a similar procedure to that in 3.1.5. The mp was 162.7-165.5 • C. 1     3.1.17. Tert-butyl((10R,13R,17R)-17-((R)-5-((tert-butyldimethylsilyl)oxy)pentan-2-yl)-4,4,10,13-tetramethyl-tetradecahydro-1H-cyclopenta[a]phenanthren-3-yl)carbamate (13) TBSCl (226 mg, 1.5 mmol) was added to a solution of compound 12 (366 mg, 0.75 mmol) and imidazole (204 mg, 3 mmol) in dry DMF (8 mL). The reaction mixture was stirred at 80 • C for 6 h under a N 2 atmosphere, and the starting material disappeared, as monitored by TLC. The mixture was added with an aqueous NaOH solution (20 mL) and extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with water, brine, dried over Na 2 SO 4 , filtered, concentrated, and recrystallized in 5 mL of EtOAc to obtain compound 13 (451 mg, 100% yield) as a white solid. The mp was 183.4-185.1 • C. 1  An aqueous 4N HCl solution (5 mL) was added to the solution of compound 14 (167 mg, 0.32 mmol) in THF (5 mL). The reaction was refluxed for 2 h, and TLC indicated the consumption of the starting material. The suspension was basified to pH > 8 with aqueous NaOH solution. The solvent THF was removed in vacuo, and water (30 mL) was added. The mixture was extracted with DCM (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, and concentrated to obtain compound 5ag (121 mg, 94% yield) as a white solid. The mp was 151.3-153.7 • C. 1 (10R,13R,17R)-3-amino-17-((R)-5-hydroxypentan-2-yl)-4,4,10,13-tetramethyltetradecahydro-7H-cyclopenta[a]phenanthren-7-oxime (5ah) NH 2 OH . HCl (35 mg, 0.5 mmol) and NaOAc (136 mg, 1 mmol) was added to a solution of compound 5ag (40 mg, 0.1 mmol) in EtOH (5 mL). The mixture was refluxed for 2 h. When the reaction was completed, as indicated by TLC, the solvent EtOH was removed in vacuo, and an aqueous NaOH solution (10 mL) was added. The mixture was extracted with DCM (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, concentrated, and recrystallized in MeCN to obtain compound 5ah ( (10R,13R,17R)-3-amino-17-((R)-5-hydroxypentan-2-yl)-4,4,10,13-tetramethyltetradecahydro-1H-cyclopenta[a]phenanthren-7-ol (5ai) NaBH 4 (19 mg, 0.5 mmol) was added to the solution of compound 5ag (40 mg, 0.1 mmol) in MeOH (10 mL). The mixture was stirred at r.t. for 2 h under a N 2 atmosphere. When the reaction was completed, as indicated by TLC, HOAc (0.1 mL) was added, and the solvent was removed in vacuo. An aqueous NaOH solution (10 mL) was added, and the mixture was extracted with DCM (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, concentrated, and recrystallized in EtOAc to obtain compound 5ai (  Compound 17 was synthesized in a 74% yield as a white solid using a similar procedure to that in 3.1.4. and was used for the next reaction without further purification. 3.1.25. Methyl(4R)-4-((10R,13R,17R)-3,7-diamino-4,4,10,13-tetramethyl-tetradecahydro-1H-cyclopenta[a]phenanthren-17-yl)pentanoate (18) Compound 17 (160 mg, 0.3 mmol) was dissolved in an anhydrous EtOAc/HCl solution (10 mL). The mixture was stirred at r.t. for 4 h. Water (20 mL) was added, and the suspension was basified to pH > 8 with an aqueous NaOH solution. The mixture was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, and concentrated to obtain compound 18 (118 mg, 92% yield) as a white solid. 1 (19) Compound 19 was synthesized in a 92% yield as a white solid using a similar procedure to that in 3. Sixty percent NaH (30 mg, 0.75 mmol) and MeI (107 mg, 0.75 mmol) were added to a solution of compound 19 (133 mg, 0.25 mmol) in dry DMF (10 mL). The mixture was stirred at r.t. for 2 h under a N 2 atmosphere, and the starting material disappeared. An aqueous Na 2 S 2 O 3 solution (5 mL) was added to quench the excess MeI. Water (30 mL) was added, and the mixture was extracted with EtOAc (3 × 30 mL). The combined organic layer was washed with water, brine, dried over Na 2 SO 4 , filtered, concentrated, and recrystallized in 5 mL of EtOH to obtain compound 20 (76 mg, 61% yield) as a white solid and used for the next reaction without further purification.  (21) Compound 21 was synthesized in a 92% yield as white solid using a similar procedure to that in 3.1.25 and used for the next reaction without further purification.  13 (22) An aqueous 2N NaOH solution (1 mL) was added to the solution of compound 15 (516 mg, 1.0 mmol) in MeOH (5 mL). The reaction was stirred at r.t. for 2 h, and TLC indicated the consumption of the starting material. Water (20 mL) was added, and the suspension was acidified to pH < 6 with aqueous HCl solution. The solvent MeOH was removed in vacuo. The mixture was extracted with DCM (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, and concentrated to obtain compound 22 (470 mg, 94% yield) as a white solid. The mp was 243.1-245.5 • C. Compound 5al was synthesized in a 93% yield as a white solid using a similar procedure to that in 3.1.25. The mp was >300 • C. 1 (25) NaBH(OAc) 3 (85 mg, 0.4 mmol) at 0 • C was added to a solution of compound 24 (92 mg, 0.2 mmol) and n-propylamine (18 mg, 0.3 mmol) in DCM (10 mL). The mixture was stirred at r.t. for 2 h under a N 2 atmosphere, and the starting material disappeared. Water (30 mL) was added, and the mixture was extracted with DCM (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, concentrated, and purified by silica gel column chromatography using DCM/MeOH/ The compound 5am was synthesized in a 90% yield as a white solid using a similar procedure to that in 3.1.25. mp >300 • C. 1   A 1M MeMgCl solution in THF (1.1 mL, 1.1 mmol) was added dropwise at 0 • C to a dry THF solution (20 mL) of compound 24 (243 mg, 0.5 mmol) under a N 2 atmosphere. The mixture was stirred at r.t. for 12 h under a N 2 atmosphere. When the reaction was completed, as indicated by TLC, an aqueous NH 4 Cl solution (2 mL) was added, and the solvents were removed in vacuo. Water (30 mL) was added, and the mixture was extracted with DCM (3 × 30 mL). The combined organic layer was washed with brine, dried over Na 2 SO 4 , filtered, concentrated, and purified by silica gel column chromatography using DCM/MeOH (30:1, v/v) to obtain compound 26a (130 mg, 52% yield) as a white solid. The mp was 200.8-202.5 • C. 1  Compound 26d was synthesized in a 45% yield as a white solid using a similar procedure to that in 3.1.37. and was used for the next reaction without further purification.

CCK8 Cell Viability Assay
Acute lymphoblastic leukemia (ALL) cell line RS4;11 (2.5 × 10 4 cells per well), acute promyelocytic leukemia (APL) cell line NB4 (1 × 10 4 cells per well), and non-small cell lung cancer (NSCLC) cell line NCI-H1299 (1.5 × 10 3 cells per well) were seeded into 96-well plates. After treatment with different concentrations of drugs, the cells were incubated in quadruple for 72 h. Thereafter, a 10 µL CCK8 solution was added to each well. After a 3 h incubation at 37 • C, the plates were read for absorbance at 450 nm and 650 nm using a SpectraMax Molecular Devices microplate reader. The final data were calibrated by OD450 nm-OD650 nm. The inhibition rates of proliferation were calculated with the following equation: Inhibition ratio = (OD DMSO -OD Compd )/(OD DMSO -OD blank ) × 100%. (1) The concentrations of the compounds that inhibited cell growth by 50% (IC 50 ) were calculated using GraphPad Prism version 5.0. NVP-2 was used as a positive control.

Molecular Docking
The Glide (Maestro 10.2) package of Schrödinger suite 2015 (https://www.schrodinger. com, accessed on 1 September 2015) was employed for the molecular docking investigation in SP mode. The SHP1 crystal structure (open conformation, PDB ID: 3PS5, 3.10 Å of resolution) was downloaded from the Protein Data Bank (https://www.rcsb.org, accessed on 1 March 2021). The receptors and ligands were prepared using the Protein Preparation Wizard and LigPrep modules embedded in the Schrödinger suite, respectively. All water molecules and sulfate ions were removed, and hydrogen atoms were added. The optimized geometry of the ligands with minimum energy was attained using an OPLS 2005 force field. The grid box was set to search all over the protein for binding sites. All the other parameters were adjusted as a default. The graphics were generated by PyMOL 2.6.0 (https://pymol.org/2, accessed on 1 November 2022) and UCSF Chimera 1.16.

Statistical Analysis
The data are presented as the mean ± SEM unless otherwise noted. Statistical significance was calculated using the two-tailed Student's t-test; a p-value < 0.05 was considered statistically significant. Statistical analysis was conducted using GraphPad Prism.

Conclusions
In summary, a new series of 3-amino-4,4-dimethyl lithocholic acid derivatives were designed, synthesized, and evaluated for SHP1 activation ability. The SAR data presented herein demonstrated that the introduction of a free amino group at the C-3 position was essential for its potency. The primary alcohol moiety in the side chain can be replaced by other chemical groups while maintaining or significantly increasing potency. A large, >9fold increase in potency was observed with the hybrid of compound 5aa and the diphenyl urea group. Combining these observations led to the synthesis of the diphenyl urea hybrid molecules 5az-ba, which proved to be highly potent SHP1 activators, with EC 50 values of 2.10 and 1.54 µM, respectively. The compounds 5az-ba also showed good selectivity in the SHP2 activation assay. In the in vitro cellular assays, 5az-ba showed potent activities with IC 50 values up to 1.65 µM against acute lymphoblastic leukemia cell line RS4;11, acute promyelocytic leukemia cell line NB4, and lung cancer cell line NCI-H1299. The molecular modeling study revealed that compound 5ba gains hydrogen bond interactions with Thr80, Gln81, Lys97, and Asn472. The deeper and closer binding of the compound to the central allosteric pocket of SHP1 may explain the stronger activation. All of the studies presented here support the proposal that 5az-ba may provide good references for the development of new anticancer drugs targeting SHP1 activation, and they deserve further research.