Efficacy of Novel Aminooxyacetic Acid Prodrugs in Colon Cancer Models

biomolecules

Article Efﬁcacy of Novel Aminooxyacetic Acid Prodrugs in Colon Cancer Models: Towards Clinical Translation of the Cystathionine β-Synthase Inhibition Concept

Mark R. Hellmich 1,*, Celia Chao 1, Katalin Módis 1,2, Ye Ding 3, John R. Zatarain 1, Ketan Thanki 1, Manjit Maskey 1 , Nadiya Druzhyna 2, Ashley A. Untereiner 2 , Akbar Ahmad 2, Yu Xue 3, Haiying Chen 3, William K. Russell 4, Jianmei Wang 5, Jia Zhou 3,* and Csaba Szabo 2,6,*

1 Department of Surgery, University of Texas, Medical Branch, Galveston, TX 77555, USA; [email protected] (C.C.); [email protected] (K.M.); [email protected] (J.R.Z.); [email protected] (K.T.); [email protected] (M.M.) 2 Department of Anesthesiology, University of Texas, Medical Branch, Galveston, TX 77555, USA; [email protected] (N.D.); [email protected] (A.A.U.); [email protected] (A.A.) 3 Department of Pharmacology and Toxicology, University of Texas, Medical Branch, Galveston, TX 77555, USA; [email protected] (Y.D.); [email protected] (Y.X.); [email protected] (H.C.) 4 Department of Biochemistry and Molecular Biology, University of Texas, Medical Branch, Galveston, TX 77555, USA; [email protected] 5 College of Pharmacy, University of North Texas Health Science Center, Fort Worth, TX 76107, USA; [email protected] 6 Citation: Hellmich, M.R.; Chao, C.; Chair of Pharmacology, Section of Medicine, University of Fribourg, CH-1700 Fribourg, Switzerland * Correspondence: [email protected] (M.R.H.); [email protected] (J.Z.); [email protected] (C.S.) Módis, K.; Ding, Y.; Zatarain, J.R.; Thanki, K.; Maskey, M.; Druzhyna, N.; Untereiner, A.A.; Ahmad, A.; et al. Abstract: Upregulation of hydrogen sulfide (H2S) biosynthesis, at least in part related to the upregu- Efficacy of Novel Aminooxyacetic lation of cystathionine β-synthetase (CBS) in cancer cells, serves as a tumor-promoting factor and has Acid Prodrugs in Colon Cancer emerged as a possible molecular target for antitumor drug development. To facilitate future clinical Models: Towards Clinical Translation translation, we have synthesized a variety of novel CBS-targeting, esterase-cleavable prodrugs based of the Cystathionine β-Synthase on the structure of the prototypical CBS inhibitor aminooxyacetic acid (AOAA). The pharmacological Inhibition Concept. Biomolecules 2021, properties of these compounds were evaluated in cell-free assays with recombinant human CBS 11, 1073. https://doi.org/10.3390/ protein, the human colon cancer cell line HCT116, and in vivo using various tumor-bearing mice biom11081073 models. The prodrug YD0251 (the isopropyl ester derivative of AOAA) was selected for detailed characterization. YD0251 exhibits improved antiproliferative efficacy in cell culture models when Academic Editor: Vladimir compared to AOAA. It is up to 18 times more potent than AOAA at suppressing HCT116 tumor N. Uversky growth in vivo and is effective when administered to tumor-bearing mice either via subcutaneous

Received: 30 June 2021 injection or oral gavage. Patient-derived xenografts (PDTXs) with higher levels of CBS protein grew Accepted: 19 July 2021 significantly larger than tumors with lower levels, and YD0251 treatment inhibited the growth of Published: 21 July 2021 PDTXs with elevated CBS, whereas it had no significant effect on PDTXs with low CBS protein levels. The toxicity of YD0251 was assessed in mice subjected to subchronic administration of Publisher’s Note: MDPI stays neutral supratherapeutic doses the inhibitor; no significant alteration in circulating markers of organ injury with regard to jurisdictional claims in or histopathological alterations were noted, up to 60 mg/kg/day × 5 days. In preparation to a published maps and institutional affil- future theranostic concept (to match CBS inhibitor therapy to high-CBS expressors), we identified a iations. potential plasma marker of CBS-expressing tumors. Colon cancer cells produced significant levels

of lanthionine, a rare metabolic intermediate of CBS-mediated H2S biosynthesis; forced expression of CBS into non-transformed epithelial cells increased lanthionine biogenesis in vitro and in vivo (measured in the urine of tumor-bearing mice). These current results may be useful to facilitate the Copyright: © 2021 by the authors. translation of a CBS inhibition-based antitumor concept into the clinical space. Licensee MDPI, Basel, Switzerland. This article is an open access article Keywords: hydrogen sulﬁde; anticancer; biomarker; prodrug distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Biomolecules 2021, 11, 1073. https://doi.org/10.3390/biom11081073 https://www.mdpi.com/journal/biomolecules Biomolecules 2021, 11, 1073 2 of 17

1. Introduction Twenty-five years ago, Abe and Kimura proposed that the endogenous gaseous me- diator hydrogen sulfide (H2S) could function as a physiological neuromodulator in the brain [1]. Subsequent studies by them and others have established important physiological roles for H2S in a variety of biological processes involving the nervous, cardiovascular, gas- trointestinal, endocrine, and immune systems [2–6]. The principal sources of endogenously synthesized H2S in mammalian cells are the two transsulfuration pathway enzymes CBS and cystathionine γ-lyase (CSE), and a third enzyme, 3-mercaptopyruvate sulfurtransferase (3-MST) [2–8]. Given its importance in normal physiology, it is not surprising that dysregulation of H2S homeostasis is implicated in the pathophysiology of many diseases, with some illnesses including atherosclerosis, colitis, type 2 diabetes, psoriasis, and hepatocellular carcinoma exhibiting local and/or systemic suppression of physiological levels of H2S, while other maladies (e.g., various forms of shock and inflammation, and Down syndrome) are associated with an aberrant upregulation of H2S production [8–15]. Cancers of the colon [16], ovaries [17], prostate [18], and breast [19] belong to the group of human diseases associated with a pathological upregulation of CBS and CBS-dependent H2S production. In 2013, we first reported the aberrant upregulation of CBS in human colon tumor tissue and colon cancer-derived cell lines and demonstrated that inhibition of CBS expression and/or activity had antitumor effects [16]. We went on to show that CBS expression was increased in human adenomatous polyps and that experimental upregulation of CBS in a premalignant colonic epithelial cell line, NCM356, caused extensive metabolic reprogramming and induction of an invasive tumorigenic phenotype, support- ing the conclusion that enhanced CBS activity not only contributes to the progression and the spread of established cancer, but also promotes colon carcinogenesis [20]. More recently, we have demonstrated that resistance to the chemotherapeutic agent 5-fluorouracil was associated with increased CBS protein levels and that inhibition of CBS activity restored cancer cell chemosensitivity [21]. Given this growing body of experimental and clinical evidence for the role of H2S as a pro-cancer bioenergetic, proliferative, and survival factor [16–26], we have pursued the development of novel inhibitors of cancer cell CBS activity as potential future anti-cancer therapies. To date, we have conducted medium-throughput screening of various clinically approved and pharmacologically well-characterized small molecules and identified several compounds that may be potentially suitable for repurposing as CBS inhibitors [27]. We have also synthesized and tested derivatives of the prototypical CBS inhibitor aminooxy- acetate (AOAA) in different cellular and animal models of colon cancer [16,21,28]. The studies presented herein represent a continuation of these efforts, with a focus towards the potential future translation of CBS-inhibitors for the treatment of colorectal cancers.

2. Materials and Methods 2.1. Materials All chemicals were obtained from Sigma-Aldrich, St. Louis, MO, USA, unless other- wise stated. For the synthesis of the various AOAA prodrugs, all commercially available starting materials and solvents were reagent grade and used without further purification. Reactions were performed under a nitrogen atmosphere in dry glassware with magnetic stirring. Preparative column chromatography was performed using silica gel 60, particle size 0.063–0.200 mm (70–230 mesh, flash). Analytical TLC was carried out on silica gel 60 F254 plates (Merck, Darmstadt, Germany). Nuclear magnetic resonance (NMR) spectra, recorded on a Bruker 300 (1H, 300 MHz; 13C, 75 MHz) spectrometer (Billerica, MA, USA), were used to confirm the identity of the synthesized compounds. 1H and 13C NMR spectra (see supplementary) were recorded with tetramethylsilane (TMS) as an internal reference. Chemical shifts downfield from TMS were expressed in ppm, and J values were given in Hz. High-resolution mass spectra (HRMS) were obtained using a Thermo Fisher LTQ Orbitrap Elite mass spectrometer. Parameters included the following: nano electrospray Biomolecules 2020, 10, x 3 of 17

Biomolecules 2020, 10, x 3 of 17

mass spectrometer. Parameters included the following: nano electrospray ionization (ESI) voltage of 1.8 kV, capillary temperature of 275 °C, and resolution of 60,000; ionization was Biomolecules 2021, 11, 1073 3 of 17 achievedmass spectrometer.by positive mode. Parameters NMR spectra included of thethe synthesizedfollowing: nano compounds electrospray are shown ionization in the (ESI) Supportingvoltage ofInformation 1.8 kV, capillary section. temperature of 275 °C, and resolution of 60,000; ionization was achievedOne of theby positivedesign goals mode. for NMR the spectranew AOAA-based of the synthesized prodrugs compounds was to improve are shown lipo- in the philicitySupportingionization (i.e., increase (ESI) Information voltage ClogP of section. values) 1.8 kV, relative capillary to temperature AOAA to enhance of 275 ◦ cellularC, and resolution uptake accord- of 60,000; ing toionization Lipinski'sOne of was the“Rule achieved design of Five”. goals by Key positive for syntheti the mode.newc methods AOAA-based NMR spectraused to prodrugs ofgenerate the synthesized wasthe representative to improve compounds lipo- prodrugsphilicityare shown are (i.e., shown in increase the Supportingin Figure ClogP 1. values) InformationThe amino relative group section. to AOAA of AOAA to enhance was protected cellular withuptake Boc accord- to giveing the to keyOne Lipinski's intermediate of the design“Rule S1.of goals Five”. The for substitution Key the newsyntheti AOAA-based reactionc methods between prodrugsused tothe generate acid was toof improve S1the and representative various lipophilic- brominated/chlorinatedprodrugsity (i.e., increase are shown ClogP reagentsin Figure values) generated1. relative The amino todi AOAAfferent group S2 toof enhanceesters,AOAA including was cellular protected uptakethe prodrugs with according Boc to withgiveto a Lipinski’scarbamate the key intermediate “Rule group. of Five”. S1. The Key substitution synthetic methods reaction used between to generate the acid the of S1 representative and various brominated/chlorinatedprodrugsFinally, the are prodrugs shown in withFigurereagents an1. aminooxy Thegenerated amino group groupdifferent in of the AOAAS2 formesters, was of including trifluoroacetic protected the with prodrugs acid Boc to (TFA)withgive salt thea carbamatewere key obtained intermediate group. by Boc-deprotection S1. The substitution with reaction TFA in betweenCH2Cl2 ( thev/v acid1:4). of S1 and various brominated/chlorinatedFinally, the prodrugs reagents with an generated aminooxy different group in S2 the esters, form including of trifluoroacetic the prodrugs acid (TFA)with a salt carbamate were obtained group. by Boc-deprotection with TFA in CH2Cl2 (v/v 1:4).

Figure 1. Synthetic routes applied to synthesize novel AOAA derivative prodrugs, with structures shown in Table 1.

FigureFigure 1. 1. SyntheticSynthetic routes routes applied appliedSince much to to synthesi synthesize of theze follow-upnovel AOAA efficacy derivative and prodrugs, safety work with focused structures on shown YD0251 in Table (the1 1.. isopentyl ester of AOAA), a second, more facile chemical synthetic route was developed suitable for scale-upSinceFinally, much thefrom prodrugs of gram the follow-upto with kilogram an aminooxyefficacy quantities and group of safety prodrug in the work form (Figure focused of trifluoroacetic 2, left on panel).YD0251 acidIn (the this (TFA) iso- method,pentylsalt were SOCl ester obtained2 of(2.18 AOAA), g, by 18.30 Boc-deprotection a second, mmol) morewas addedfacile with chemical TFAto a inmixture CH synthetic2Cl 2of( vcarboxymethoxylamine/ routev 1:4). was developed suit- hemihydrochlorideable forSince scale-up much (1.0 offrom the g, gram follow-up9.15 mmol)to kilogram efficacy in 3-pentanol quantities and safety (5 of workmL) prodrug at focused 0 °C. (Figure The on YD0251resulting 2, left panel). (the mixture isopentyl In this wasmethod,ester stirred, of AOAA),firstSOCl at2 room(2.18 a second, g,temperature 18.30 more mmol) facile for was2 h chemical and added then syntheticto for a anmixture additional route of wascarboxymethoxylamine 16 developedh at 100 °C suitable(the reactionhemihydrochloridefor scale-up progress from was gram monitored(1.0 g, to 9.15 kilogram mmol)by NMR). quantities in 3-pentanol The solvent of prodrug (5 mL)was at (Figureremoved 0 °C. The2, leftunder resulting panel). reduced mixture In this pressurewasmethod, stirred, to achieve SOCl first2 the(2.18at room desired g, 18.30 temperature product mmol) YD0251 wasfor 2 added h andas a thenwhite to a mixturefor solid an additional(1.41 of carboxymethoxylamine g, 86%), 16 h which at 100 was °C (the ◦ recrystallizedreactionhemihydrochloride progress from 3-pentanolwas (1.0 monitored g, 9.15 (5 mmol)mL) by NMR).(F inigure 3-pentanol The 2, solventright (5 mL)panel), was at removed 0washedC. The under resultingwith reduced10% mix- EtOAc/hexanepressureture was to stirred, achievesolution first the (25 atdesired mL), room and product temperature dried YD0251 under for as 2vacuum. ha white and then solidTLC: for (1.41Rf an = g, additional0.4 86%), ~ 0.5 which (20% 16 hwas at ◦ EtOAc/hexanes).recrystallized100 C (the reaction 1H-NMRfrom progress3-pentanol (300 MHz, was (5 D monitored 2O):mL) δ 4.80(Figure by (m, NMR). 1H),2, right 4.66 The (s, solventpanel), 2H), 1.54 waswashed (m, removed 4H), with 0.77 under 10% (t, J EtOAc/hexanereduced= 7.5 Hz, pressure 6H). 13C-NMRsolution to achieve (25 (75 mL), theMHz, desiredand D2 O):dried productδ 169.6, under YD025180.79, vacuum. 70.58, as a TLC:25.67, white Rf25.67, solid = 0.4 (1.418.73, ~ 0.58.73. g, 86%),(20% which was recrystallized from 3-pentanol+ (5 mL) (Figure2, right panel), washed with HRMSEtOAc/hexanes). Calcd for C7H 116H-NMRNO3: [M (300 + H] MHz, 162.1130; D2O): found δ 4.80 162.1124. (m, 1H), 4.66 (s, 2H), 1.54 (m, 4H), 0.77 10% EtOAc/hexane solution (25 mL), and dried under vacuum. TLC: Rf = 0.4~0.5 (20% (t,The J = structures7.5 Hz, 6H). of 13C-NMR the successfully (75 MHz, synt Dhesized2O): δ 169.6, prodrug 80.79, test 70.58, compounds 25.67, 25.67, (together 8.73, 8.73. EtOAc/hexanes). 1H-NMR (300 MHz, D O): δ 4.80 (m, 1H), 4.66 (s, 2H), 1.54 (m, 4H), 0.77 withHRMS the reference Calcd for compound C7H16NO3 :AOAA) [M + H] are+ 162.1130; show2 n foundin Table 162.1124. 1. YD0382 and YD0452 were designed(t, J = as 7.5 potentially Hz, 6H). 13C-NMR hypoxia-activatable (75 MHz, Dpr2odrugsO): δ 169.6, based 80.79, on their 70.58, particular 25.67, 25.67, ester 8.73,struc- 8.73. The structures of the successfully+ synthesized prodrug test compounds (together ture.withHRMS the Calcd reference for C 7compoundH16NO3: [M AOAA) + H] 162.1130;are show foundn in Table 162.1124. 1. YD0382 and YD0452 were designed as potentially hypoxia-activatable prodrugs based on their particular ester structure.

Figure 2. Streamlined synthesis of YD0251 suitable for scale-up at the gram to kilogram level. Reagents and conditions: (a) ◦ 3-pentanol, SOCl2, rt, 2 h, 100 C, 16 h. BiomoleculesBiomolecules 2020 2020, 10,, 10, x,, xx 4 4of of 17 17 Biomolecules BiomoleculesBiomolecules 2020 2020 2020, ,10 ,10 10, ,x ,x x 44 4of ofof 17 1717 Biomolecules Biomolecules Biomolecules Biomolecules 2020 2020 2020 2020, 10,, ,10 ,,10 x10,, ,xx ,,x xx 4 4of4 4 of of17of 17 17 17 Biomolecules 2020, 10, x 4 of 17 Biomolecules 2020, 10, x 4 of 17

Biomolecules 2021, 11, 1073 4 of 17 FigureFigureFigure 2. 2. 2.Streamlined Streamlined Streamlined synthesis synthesis synthesis of of ofYD0251 YD0251 YD0251 suitable suitable suitable for for for scale-up scale-up scale-up at at atthe the the gram gram gram to to tokilogram kilogram kilogram level. level. level. Reagents Reagents Reagents and and and conditions: conditions: conditions: FigureFigureFigure 2. 2. 2. Streamlined Streamlined Streamlined synthesis synthesis synthesis of of of YD0251 YD0251 YD0251 suitable suitable suitable for for for scale-up scale-up scale-up at at at the the the gram gram gram to to to kilogram kilogram kilogram level. level. level. Reagents Reagents Reagents and and and conditions: conditions: conditions: FigureFigure 2. 2. Streamlined Streamlined2 synthesis synthesis of of YD0251 YD0251 suitable suitable for for scale-up scale-up at at the the gram gram to to kilogram kilogram level. level. ReagentsReagents Reagents andand and conditions:conditions: conditions: (a)(a)FigureFigure Figure3-pentanol, 3-pentanol, 2.2. 2. StreamlinedStreamlined Streamlined SOCl SOCl2, 2rt,,2 rt, synthesis synthesis2 synthesis 2h, h, 100 100 °C, of°C,of of 16YD0251 YD0251 16YD0251 h. h. suitablesuitable suitable forfor for scale-upscale-up scale-up atat at thethe the gramgram gram toto to kilogramkilogram kilogram level.level. level. ReagentsReagents Reagents andand and conditions:conditions: conditions: (a)(a)(a) 3-pentanol, 3-pentanol, 3-pentanol, SOCl SOCl SOCl2,2 ,rt, ,rt, rt, 2 2 2h, h, h, 100 100 100 °C, °C, °C, 16 16 16 h. h. h. (a)(a)Figure 3-pentanol, 3-pentanol, 2. Streamlined SOCl SOCl2, 2rt,,, 2rt,rt, synthesis2 22h, h,h, 100 100100 °C, °C,°C,of 16 YD0251 1616 h. h.h. suitable for scale-up at the gram to kilogram level. Reagents and conditions: (a)(a)(a)(a) 3-pentanol,3-pentanol, 3-pentanol,3-pentanol, SOClSOCl SOClSOCl22,, 2 rt,,,rt, rt,rt, 22 22 h,h, h,h, 100100 100100 °C,°C, °C,°C, 1616 1616 h.h. h.h. Table(a) 3-pentanol, 1. Comparison SOCl2 ,of rt, the 2 h, relative 100 °C, inhibitor 16 h. effects of the different AOAA-based prodrugs to the reference compound, TableTableTable 1. Table1. 1.Comparison Comparison Comparison 1. Comparison of of ofthe the the relative ofrelative relative the relativeinhibitor inhibitor inhibitor inhibitor effects effects effects of effects of ofthe the the di ofdi fferentdifferentfferent thefferent different AOAA-based AOAA-based AOAA-basedAOAA-based AOAA-based prodrugs prodrugs prodrugsprodrugs prodrugs to toto tothe the thethe reference toreference referencereference the reference compound, compound, compound,compound, compound, TableTableTableTable 1. 1. 1. Comparison 1.Comparison Comparison Comparison of of of the ofthe the the relative relative relative relative inhibitor inhibitor inhibitor inhibitor effects effects effects effects of of of the ofthe the the di di different fferentdifferentfferent AOAA-based AOAA-based AOAA-based AOAA-based prodrugs prodrugs prodrugs prodrugs to to to the tothe the the reference reference reference reference compound, compound, compound, compound, AOAA.AOAA.TableTableAOAA.Table 1.1. 1. ComparisonComparison Comparison ofof of thethe the relativerelative relative inhibitorinhibitor inhibitor effectseffects effects ofof of thethe the didi differentfferentfferent AOAA-basedAOAA-based AOAA-based prodrugsprodrugs prodrugs toto to thethe the referencereference reference compound,compound, compound, AOAA.AOAA.AOAA. AOAA. AOAA.TableAOAA.AOAA. 1. Comparison of the relative inhibitor effects of the different AOAA-based prodrugs to the reference compound, AOAA. CBSCBSCBSCBS Enzyme EnzymeEnzyme Enzyme HCT116 HCT116HCT116HCT116 Proliferation ProliferationProliferation Proliferation Tumor TumorTumorTumor Homogen. Homogen.Homogen. Homogen. Liver LiverLiverLiver Homogen. Homogen.Homogen. Homogen. CBSCBSCBS Enzyme Enzyme Enzyme HCT116HCT116HCT116 Proliferation Proliferation ProliferationHCT116 TumorTumorTumor Homogen. Homogen. Homogen. Liver LiverLiver Homogen. Homogen. Homogen. CompoundCompound CategoryCategory Structure Structure (%(%CBSCBS InhibitionCBS Inhibition Enzyme Enzyme Enzyme (% (% HCT116 HCT116InhibitionHCT116 Inhibition Proliferation Proliferation Proliferation at at 10, 10, 30, 30, CBS TumorCBSTumorTumor Activity Activity Homogen. TumorHomogen. Homogen. (% (% CBS CBSLiver LiverLiver Activity Activity Homogen. Homogen. Homogen.Liver (% (% CompoundCompoundCompound CategoryCategoryCategory Structure Structure Structure (%(%(%CBS Inhibition Inhibition Inhibition Enzyme (% (%(% Inhibition Inhibition InhibitionProliferation at at at 10, 10, 10, 30, 30, 30, CBSCBSCBS Activity Activity Activity (% (% (% CBSCBSCBS Activity Activity Activity (% (% (% CompoundCompoundCompound CategoryCategoryCategory Structure Structure Structure (%(% CBSInhibition(% Inhibition Inhibition Enzyme (% (% HCT116Inhibition(% Inhibition Inhibition Proliferation at at 10, at 10, 30,10, 30, 30, CBS TumorCBSCBS Activity ActivityHomogen. Activity Homogen. (% (% (% CBS CBSLiverCBS Activity Activity Homogen.ActivityHomogen. (% (% (% CompoundCompoundCompoundCompound CategoryCategoryCategoryCategory Structure Structure Structure at(%(%atat (%100at Inhibition100100Inhibition Inhibition100Inhibition µM) µM)µM) µM) 100, (% 100,(%100,100,(% Inhibition300Inhibition Inhibition300(%300 300 µM InhibitionµMµM µM for forfor for at at48 at 48 10,48 10, 48 h)10, h) h) 30,h)30, at 30, CBSCBSInhibition)CBSInhibition)Inhibition)Inhibition) ActivityActivity Activity (%(% (% CBSCBSInhibition)CBSInhibition)Inhibition)Inhibition) ActivityActivity Activity (%(% (% atatat 100 100 100 µM) µM) µM) 100,100,100, 300 300 300 µM µM µM for for for 48 48 48 h) h) h) Inhibition)Inhibition)Inhibition) Inhibition)Inhibition)Inhibition) Compound Category Structure (%atatat Inhibition100 100 100 µM) µM) µM)µ (%100,100,100, Inhibition 300 300 300 µM µM µM for for atfor 48 10,48 48 h) h)30, h) CBSInhibition)Inhibition)CBSInhibition)Inhibition) Activity Activity (% CBSInhibition)Inhibition)Inhibition)Inhibition) ActivityCBS Activity (% at 100 M) 10, 30, 100, (% Inhibition) at 100 µM) 100,300 300µ µMM for 48 h) h) (%Inhibition) Inhibition) Inhibition) AOAAAOAAAOAAAOAA ReferenceReferenceReferenceReference 92%92%92%92% 0, 0, 0,0, 0, 0,0, 0, 0,36%0, 36%36% 36% 0% 0% 0% 0% 36% 36% 36% 36% AOAAAOAAAOAAAOAA ReferenceReferenceReferenceReference 92%92%92%92% 0, 0, 0,0, 0, 0, 0,0, 0, 0, 0,36% 36% 0,36% 36% 0% 0% 0% 0% 36% 36% 36% 36% AOAAAOAAAOAA ReferenceReferenceReference 92%92%92% 0, 0, 0, 0,0, 0, 0,0, 0, 36%36% 36% 0% 0% 0% 36% 36% 36%

AOAAAOAA ReferenceReference 92%92% 0,0, 0, 0,0, 0,36% 36% 0% 0% 36% 36%

YD0171 Prodrug 44% 4, 12, 34, 55% 68% 15% YD0171YD0171YD0171 ProdrugProdrugProdrug 44%44%44%44% 4, 4, 4,12, 4, 12,12, 12, 34, 34,34, 34, 55% 55%55% 55% 68% 68% 68% 68% 15% 15% 15% 15% YD0171YD0171YD0171YD0171 ProdrugProdrugProdrugProdrug 44%44%44%44% 4, 4, 4,12, 12, 4,12, 34,12, 34, 34, 55%34, 55% 55% 55% 68% 68% 68% 68% 15% 15% 15% 15% YD0171YD0171YD0171YD0171 ProdrugProdrugProdrugProdrug 44%44%44%44% 4, 4, 4, 12,12, 12, 12, 34,34, 34, 34, 55%55% 55% 55% 68% 68% 68% 68% 15% 15% 15% 15% YD0171 Prodrug 44% 4, 12, 34, 55% 68% 15%

YD0222YD0222YD0222YD0222 ProdrugProdrugProdrugProdrug 17%17%17%17% 0, 0, 0,0, 0, 0,0, 0, 0,30%0, 30%30% 30% 68% 68% 68% 68% 67% 67% 67% 67% YD0222YD0222YD0222 ProdrugProdrugProdrug 17%17%17% 0, 0, 0,0, 0, 0,0, 0, 0,30% 30% 30% 68% 68% 68% 67% 67% 67% YD0222YD0222YD0222YD0222 ProdrugProdrugProdrugProdrug 17%17%17%17% 0, 0, 0, 0, 0,0, 0, 0, 0, 0, 30% 0,30% 30% 30% 68% 68% 68% 68% 67% 67% 67% 67% YD0222 Prodrug 17% 0, 0, 0, 30% 68% 67%

YD0223 Prodrug 4% 0, 0, 0, 15% 50% 62% YD0223YD0223YD0223 ProdrugProdrugProdrug 4%4%4%4% 0, 0, 0,0, 0, 0,0, 0, 0,15%0, 15%15% 15% 50% 50% 50% 50% 62% 62% 62% 62% YD0223YD0223YD0223 ProdrugProdrugProdrug 4%4%4% 0, 0, 0,0, 0, 0,0, 0, 0,15% 15% 15% 50% 50% 50% 62% 62% 62% YD0223YD0223YD0223YD0223 ProdrugProdrugProdrugProdrug 4%4%4%4% 0, 0, 0, 0, 0,0, 0, 0, 0, 0, 15% 0,15% 15% 15% 50% 50% 50% 50% 62% 62% 62% 62% YD0223 Prodrug 4% 0, 0, 0, 15% 50% 62%

YD0239YD0239YD0239YD0239 ProdrugProdrugProdrugProdrug 1%1%1%1% 0, 0, 0,8, 0, 8, 8,21,8, 21,21, 21, 41% 41%41% 41% 63% 63% 63% 63% 0% 0% 0% 0% YD0239YD0239YD0239 ProdrugProdrugProdrug 1%1%1% 0, 0, 0,8, 8, 8,21, 21, 21, 41% 41% 41% 63% 63% 63% 0% 0% 0% YD0239YD0239YD0239YD0239 ProdrugProdrugProdrugProdrug 1%1%1%1% 0, 0, 0, 0,8, 8, 8, 21, 8,21, 21, 21, 41% 41% 41% 41% 63% 63% 63% 63% 0% 0% 0% 0% YD0239 Prodrug 1% 0, 8, 21, 41% 63% 0%

YD0242 Prodrug 0% 0, 0, 2, 65% 0% 60% YD0242YD0242YD0242 ProdrugProdrugProdrug 0%0%0%0% 0, 0, 0,0, 0, 0,2,0, 2, 2,65%2, 65%65% 65% 0% 0% 0% 0% 60% 60% 60% 60% YD0242YD0242YD0242YD0242YD0242 ProdrugProdrugProdrugProdrugProdrug 0%0%0% 0% 0, 0, 0,0, 0, 0, 0,2, 2, 0, 2,0,65% 65% 2,65% 2, 65% 65% 0% 0% 0% 0% 0% 60% 60% 60% 60% 60% YD0242YD0242YD0242 ProdrugProdrugProdrug 0%0%0% 0, 0, 0, 0,0, 0, 2,2, 2, 65%65% 65% 0% 0% 0% 60% 60% 60% YD0242 Prodrug 0% 0, 0, 2, 65% 0% 60%

YD0246 Prodrug 0% 0, 0, 0, 43% 54% 73% YD0246YD0246YD0246YD0246 ProdrugProdrugProdrugProdrug 0%0%0%0% 0% 0, 0, 0,0, 0, 0, 0, 0, 0,43%0, 43%43% 0,43% 43% 54% 54% 54% 54% 54% 73% 73% 73% 73% 73% YD0246YD0246YD0246 ProdrugProdrugProdrug 0%0%0% 0, 0, 0,0, 0, 0,0, 0, 0,43% 43% 43% 54% 54% 54% 73% 73% 73% YD0246YD0246YD0246 ProdrugProdrugProdrug 0%0%0% 0, 0, 0, 0, 0, 0, 0, 0, 0, 43% 43% 43% 54% 54% 54% 73% 73% 73% YD0246 Prodrug 0% 0, 0, 0, 43% 54% 73%

YD0251YD0251YD0251 ProdrugProdrug 10%10% 5, 5, 5, 5, 11, 11, 52% 52% 42% 42% 54% 54% YD0251YD0251 ProdrugProdrugProdrug 10%10%10%10% 5, 5, 5, 5, 5, 11, 5,11,11, 11, 52%52% 52% 52% 42% 42% 42% 42% 54% 54% 54% 54% YD0251YD0251YD0251YD0251 ProdrugProdrugProdrugProdrug 10%10%10%10% 5, 5, 5,5, 5, 5, 5,11, 11,5, 11, 52%11, 52% 52% 52% 42% 42% 42% 42% 54% 54% 54% 54% YD0251YD0251YD0251 ProdrugProdrugProdrug 10%10%10% 5, 5, 5, 5,5, 5, 11,11, 11, 52%52% 52% 42% 42% 42% 54% 54% 54% YD0251 Prodrug 10% 5, 5, 11, 52% 42% 54%

YD0335YD0335YD0335YD0335YD0335 ProdrugProdrugProdrugProdrugProdrug 0%0%0%0% 0% 0, 0, 0,0, 0, 0,16,0, 16,16, 0,16, 54% 16, 54%54% 54% 54% 46% 46% 46% 46% 46% 59% 59% 59% 59% 59% YD0335YD0335YD0335 ProdrugProdrugProdrug 0%0%0% 0, 0, 0,0, 0, 0,16, 16, 16, 54% 54% 54% 46% 46% 46% 59% 59% 59% YD0335YD0335YD0335 ProdrugProdrugProdrug 0%0%0% 0, 0, 0, 0, 0, 0, 16, 16, 16, 54% 54% 54% 46% 46% 46% 59% 59% 59% YD0335 Prodrug 0% 0, 0, 16, 54% 46% 59%

YD0343YD0343YD0343YD0343 ProdrugProdrugProdrugProdrugProdrug 0%0%0%0% 0, 0, 0,0, 0, 0,0, 0, 0,24%0, 24%24% 24% 0% 0% 0% 0% 0% 0% 0% 0% YD0343YD0343YD0343YD0343 ProdrugProdrugProdrugProdrug 0%0%0% 0% 0, 0, 0,0, 0, 0,0, 0, 0,24% 24% 24% 0, 24% 0% 0% 0% 0% 0% 0% 0% 0% YD0343YD0343YD0343 ProdrugProdrugProdrugProdrugProdrug 0%0%0% 0, 0, 0, 0, 0, 0, 0, 0, 0, 24% 24% 24% 0% 0% 0% 0% 0% 0% YD0343 Prodrug 0% 0, 0, 0, 24% 0% 0%

YD0381YD0381YD0381YD0381 ProdrugProdrugProdrugProdrugProdrug 1%1%1%1% 0, 0, 0,3, 0, 3, 3,13,3, 13,13, 13, 41% 41%41% 41% 81% 81% 81% 81% 61% 61% 61% 61% YD0381YD0381YD0381YD0381 ProdrugProdrugProdrugProdrug 1%1%1% 1% 0, 0, 0,3, 3, 0,3,13, 13, 13, 3, 41% 41% 13,41% 41% 81% 81% 81% 81% 61% 61% 61% 61% YD0381YD0381YD0381 ProdrugProdrugProdrugProdrugProdrug 1%1%1% 0, 0, 0, 3, 3, 3, 13, 13, 13, 41% 41% 41% 81% 81% 81% 61% 61% 61% YD0381 Prodrug 1% 0, 3, 13, 41% 81% 61%

ProdrugProdrug (Hypoxia- (Hypoxia- YD0382YD0382 ProdrugProdrugProdrug (Hypoxia- Prodrug(Hypoxia- (Hypoxia- 20%20% 0, 0, 0, 0, 0, 0, 50% 50% 0% 0% 0% 0% YD0382YD0382 ProdrugProdrugProdrugactivated) (Hypoxia- (Hypoxia- (Hypoxia- 20%20%20% 0, 0, 0, 0,0, 0, 0,0, 0, 50%50% 50% 0% 0% 0% 0% 0% 0% YD0382YD0382YD0382YD0382YD0382 ProdrugProdrugProdrugactivated)activated)activated)activated)(Hypoxia- (Hypoxia-(Hypoxia- (Hypoxia- 20%20%20%20% 0, 0, 0,0, 0, 0, 0,0, 0, 0, 0,50% 50% 0,50% 0, 50% 50% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% YD0382YD0382YD0382 Prodrugactivated)activated)activated)activated) (Hypoxia- 20%20%20% 0, 0, 0, 0,0, 0, 0,0, 0, 50%50% 50% 0% 0% 0% 0% 0% 0% YD0382 activated)activated)activated)activated) 20% 0, 0, 0, 50% 0% 0% activated)

YD0450YD0450YD0450YD0450 ProdrugProdrugProdrugProdrugProdrug 3%3%3%3% 0, 0, 0,0, 0, 0,0, 0, 0,59%0, 59%59% 59% 16% 16% 16% 16% 73% 73% 73% 73% YD0450YD0450YD0450 ProdrugProdrugProdrug 3%3%3% 0, 0, 0,0, 0, 0,0, 0, 0,59% 59% 59% 16% 16% 16% 73% 73% 73% YD0450YD0450YD0450YD0450 ProdrugProdrugProdrugProdrugProdrugProdrug 3%3%3%3% 0, 0, 0, 0, 0,0, 0, 0, 0, 0, 59% 0,59% 59% 59% 16% 16% 16% 16% 73% 73% 73% 73% YD0450 Prodrug 3% 0, 0, 0, 59% 16% 73% ProdrugProdrug (Hypoxia- (Hypoxia- YD0452YD0452 ProdrugProdrugProdrug (Hypoxia- Prodrug(Hypoxia- (Hypoxia- 50%50% 0, 0, 0, 0, 0, 0, 50% 50% 0% 0% 0% 0% YD0452YD0452 ProdrugProdrugProdrugactivated) (Hypoxia- (Hypoxia- (Hypoxia- 50%50%50% 0, 0, 0, 0,0, 0, 0,0, 0, 50%50% 50% 0% 0% 0% 0% 0% 0% YD0452YD0452YD0452YD0452YD0452 ProdrugProdrugProdrugactivated)activated)activated)activated)(Hypoxia- (Hypoxia-(Hypoxia- (Hypoxia- 50%50%50%50% 0, 0, 0,0, 0, 0, 0,0, 0, 0, 0,50% 50% 0,50% 0, 50% 50% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% YD0452YD0452YD0452 Prodrugactivated)activated)activated)activated) (Hypoxia- 50%50%50% 0, 0, 0, 0,0, 0, 0,0, 0, 50%50% 50% 0% 0% 0% 0% 0% 0% YD0452 activated)activated)activated)activated) 50% 0, 0, 0, 50% 0% 0% activated)

Biomolecules 2020, 10, Prodrugx Prodrug 5 of 17 YD0541YD0541 ProdrugProdrugProdrugProdrug 1%1% 0, 0, 0, 0, 0, 0, 19% 19% 0% 0% 38% 38% YD0541YD0541 (PEGylated)ProdrugProdrugProdrug 1%1%1% 0, 0, 0, 0,0, 0, 0,0, 0, 19%19% 19% 0% 0% 0% 38% 38% 38% YD0541YD0541YD0541YD0541YD0541 (PEGylated)(PEGylated)(PEGylated)(PEGylated)ProdrugProdrugProdrug(PEGy- 1%1%1% 1% 0, 0, 0,0, 0, 0, 0,0, 0, 0, 0,19% 19% 0,19% 0, 19% 19% 0% 0% 0% 0% 0% 38% 38% 38% 38% 38% YD0541YD0541YD0541 (PEGylated)(PEGylated)(PEGylated)(PEGylated)Prodrug 1%1%1% 0, 0, 0, 0,0, 0, 0,0, 0, 19%19% 19% 0% 0% 0% 38% 38% 38% YD0541 (PEGylated)(PEGylated)(PEGylated)lated) 1% 0, 0, 0, 19% 0% 38% (PEGylated)

YD0706YD0706 ProdrugProdrug 0%0% 0, 0,0, 0,0, 0,0% 0% 0% 0% 42% 42%

2.2. Measurement of the Catalytic Activity of Recombinant Human CBS

Full-length recombinant human CBS was produced as described [28]. H2S production by CBS was measured using the 7-azido-4-methylcoumarin (AzMC) method [28] in a 96- well black plate format. The total assay volume was 200 μL per well and consisted of 170 μL of reaction solution containing final concentrations of 200 mM Tris HCl (pH 8.0), 5 μM pyridoxal 5′-phosphate (PLP), 10 mM glutathione, and 0.5 mg/mL BSA; 10 μL test compound; 10 μL CBS (5 μg); and 10 μL of a mixture of AzMc (10 μM) and the CBS substrates L-cysteine and homocysteine (each at 2.5 mM final concentration). The reaction mixture was incubated at 37 °C for 1 h, and AzMc fluorescence read at 450 nm (λex = 365 nm). Percent inhibition of CBS activity of each test compound (at 100 μM) is shown in Table 1 as the mean value of n = 3 determinations.

2.3. Effect of Test Compounds on Cell Proliferation and Viability The human colorectal carcinoma cell line, HCT116, was cultured in McCoy’s 5A medium supplemented with 10% FBS, 100 IU/mL penicillin, and 100 mg/mL streptomycin as described [16]. Cells were grown in a 37 °C, 5% CO2 atmosphere. For assessment of cell proliferation, the xCELLigence Real-time Cell Analyzer (Agilent Technologies, Santa Clara, CA, USA) was used, as described [16]. Briefly, HCT116 cells were cultured until approximately 70% confluence. Cells were detached by Trypsin–EDTA and re-suspended in fresh culture media at a concentration of 60,000 cells/mL. Then, 100 μL of cell suspension was added to each well (6,000 cells/well) of an E-plate after recording the background with 100 μL of cell culture media per well. The E-plates are specially designed 96-well microtiter plates containing interdigitated microelectrodes to non-invasively monitor the cell proliferation by measuring the relative change in the electrical impedance of the cell monolayer, a unitless parameter named the “cell index”. Once plated, the cells were treated either with vehicle (i.e., test compound diluent) or different concentrations of the test compounds (10–1000 μM). The change in electrical impedance (i.e., proliferation rate) was monitored for 48 h. The percent inhibitions of vehicle-treated cell cultures for each concentration of the test compound (at 10, 30, 100, and 300 μM) at 48 h are shown in Table 1 as the mean values of n = 3 determinations to compare the effects of YD0251 and 5- fluorouracil (5-FU) on the viability of control and multidrug-resistant HCT116 cells, generated and characterized as previously described [21]. Cells were incubated either with various concentrations of YD0251 or 5-FU for 24 h, and mitochondrial-dependent conversion of MTT to formazan was measured as described [16,21].

2.4. Ex Vivo Assessment of Tumor and Liver CBS Activity Following In Vivo Treatment with Test Compounds The animal studies conducted in the current paper were approved by UTMB’s IACUC and were conducted using protocols 890102I and 8901012J (“In vivo studies in GI cancer growth, approved 1 January 2016 and valid until 31 December 2021). Nu/nu Balb/C mice (4–6 per test compound) were implanted with HCT116 cells subcutaneously as described [16]. When tumor size reached ~150 mm3, animals were treated either with vehicle or a test compound (2.5 mg/kg) via s.q. injection every 6 h for 24 h (total dose 10 mg/kg). Following treatment, tumor and liver tissues were harvested and protein homogenates prepared for ex vivo CBS activity measurements using the AzMC-based H2S production assay as described [28]. Inhibition of CBS activity was calculated as follows: ((vehicle-treated CBS activity—AOAA or prodrug-treated CBS activity)/vehicle-treated CBS activity) × 100. Results are shown in Table 1. Each value is the mean of n = 3 determinations. Biomolecules 2021, 11, 1073 5 of 17

The structures of the successfully synthesized prodrug test compounds (together with the reference compound AOAA) are shown in Table1. YD0382 and YD0452 were designed as potentially hypoxia-activatable prodrugs based on their particular ester structure.

2.2. Measurement of the Catalytic Activity of Recombinant Human CBS

Full-length recombinant human CBS was produced as described [28]. H2S production by CBS was measured using the 7-azido-4-methylcoumarin (AzMC) method [28] in a 96-well black plate format. The total assay volume was 200 µL per well and consisted of 170 µL of reaction solution containing final concentrations of 200 mM Tris HCl (pH 8.0), 5 µM pyridoxal 50-phosphate (PLP), 10 mM glutathione, and 0.5 mg/mL BSA; 10 µL test compound; 10 µL CBS (5 µg); and 10 µL of a mixture of AzMc (10 µM) and the CBS substrates L-cysteine and homocysteine (each at 2.5 mM final concentration). The reaction mixture was incubated at 37 ◦C for 1 h, and AzMc fluorescence read at 450 nm (λex = 365 nm). Percent inhibition of CBS activity of each test compound (at 100 µM) is shown in Table1 as the mean value of n = 3 determinations.

2.3. Effect of Test Compounds on Cell Proliferation and Viability The human colorectal carcinoma cell line, HCT116, was cultured in McCoy’s 5A medium supplemented with 10% FBS, 100 IU/mL penicillin, and 100 mg/mL streptomycin ◦ as described [16]. Cells were grown in a 37 C, 5% CO2 atmosphere. For assessment of cell proliferation, the xCELLigence Real-time Cell Analyzer (Agilent Technologies, Santa Clara, CA, USA) was used, as described [16]. Briefly, HCT116 cells were cultured until approximately 70% confluence. Cells were detached by Trypsin–EDTA and re-suspended in fresh culture media at a concentration of 60,000 cells/mL. Then, 100 µL of cell suspension was added to each well (6000 cells/well) of an E-plate after recording the background with 100 µL of cell culture media per well. The E-plates are specially designed 96-well microtiter plates containing interdigitated microelectrodes to non-invasively monitor the cell proliferation by measuring the relative change in the electrical impedance of the cell monolayer, a unitless parameter named the “cell index”. Once plated, the cells were treated either with vehicle (i.e., test compound diluent) or different concentrations of the test compounds (10–1000 µM). The change in electrical impedance (i.e., proliferation rate) was monitored for 48 h. The percent inhibitions of vehicle-treated cell cultures for each concentration of the test compound (at 10, 30, 100, and 300 µM) at 48 h are shown in Table1 as the mean values of n = 3 determinations to compare the effects of YD0251 and 5-fluorouracil (5-FU) on the viability of control and multidrug-resistant HCT116 cells, generated and characterized as previously described [21]. Cells were incubated either with various concentrations of YD0251 or 5-FU for 24 h, and mitochondrial-dependent conversion of MTT to formazan was measured as described [16,21].

2.4. Ex Vivo Assessment of Tumor and Liver CBS Activity Following In Vivo Treatment with Test Compounds The animal studies conducted in the current paper were approved by UTMB’s IACUC and were conducted using protocols 890102I and 8901012J (In vivo studies in GI cancer growth, approved 1 January 2016 and valid until 31 December 2021). Nu/nu Balb/C mice (4–6 per test compound) were implanted with HCT116 cells subcutaneously as described [16]. When tumor size reached ~150 mm3, animals were treated either with vehicle or a test compound (2.5 mg/kg) via s.q. injection every 6 h for 24 h (total dose 10 mg/kg). Following treatment, tumor and liver tissues were harvested and protein homogenates prepared for ex vivo CBS activity measurements using the AzMC- based H2S production assay as described [28]. Inhibition of CBS activity was calculated as follows: ((vehicle-treated CBS activity—AOAA or prodrug-treated CBS activity)/vehicle- treated CBS activity) × 100. Results are shown in Table1. Each value is the mean of n = 3 determinations. Biomolecules 2021, 11, 1073 6 of 17

2.5. Assessment of the Efﬁcacy of YD0251 and AOAA in Mice Bearing HCT116 Subcutaneous Xenografts Tumors were initiated by injecting Nu/nu Balb/C mice with HCT116 cells (1 × 106 cells/injection) subcutaneously in the right or left rear ﬂank. Then, 7–10 days after tumor initiation, animals were treated either with vehicle or test compounds via s.q. injection or oral gavage at different doses once a day for 20 days. Tumor size (volume) was measured transcutaneously with a caliper every 2–3 days as described [16]. Results are presented as the mean ± SEM values of n = 7–9 determinations.

2.6. Assessment of the Efficacy of YD0251 in Mice Bearing Subcutaneous Patient-Derived Tumor Xenografts (PDTXs) Freshly resected colorectal cancer tissue was collected under an IRB-approved protocol as part of a discarded tissue biobanking protocol. At the time of tissue collection, the establishment of this protocol did not require specific patient informed consent. Cancer tissues were dissected by a surgical pathologist and immediately placed into a sterile tube containing cold DMEM supplemented with 200 U/mL penicillin, 200 µg/mL streptomycin, 0.25 µg/mL amphotericin B, and 50 µg/mL gentamicin (Invitrogen; Carlsbad, CA, USA). After several washing steps in this medium, a small piece was minced into ~2–3 mm3 pieces under sterile conditions for xenografting into the flanks of 2 or 3 Nu/nu Balb/C mice. This initial engraftment was designated “passage zero” (P0). The studies described here used PDTXs at passages 3 to 5 (i.e., P3 to P5). To test the effects of the prodrugs on PDTX growth in vivo, once the tumor measured ~100 mm along 2 axes, animals were treated daily for up to 3 weeks either with vehicle or test compounds at a dose of 6 mg/kg (100 µL, subcutaneous injection). Tumor growth was monitored by transcutaneous measurements with a caliper every 2–3 days as described [16]. PDTXs from 6 different patients were used. The level of CBS protein expression in each PDTX was determined by Western blotting as described [16].

2.7. Assessment of the Safety and Tolerability of YD0251 in Mice Balb/C mice were subjected to intraperitoneal injections of YD0251 (total daily doses: 2 mg/kg/day, 6 mg/kg/day, 20 mg/kg/day, and 60 mg/kg/day) for 7 consecutive days. On the morning of the eighth day, the experiment was terminated, blood was collected and analyzed using the Vetscan VS2 Chemistry Analyzer (Abaxis/Zoetis, Parsippany, NJ, USA), and histopathological assessment of the liver and kidney was conducted in a blinded fashion as described [28]. The data on plasma markers are presented as mean ± SEM of n = 5 mice per group; histological images are presented as representative examples.

2.8. Sample Preparation and Detection of Transsulfuration Pathway Metabolites by Mass Spectrometry Sample preparation: 20 µL of working internal standard solution was added to 150 µL of each urine sample or cell extract. An additional 90 µL of cold acetonitrile with 0.1% FA was added, and the resulting solution vortexed for 5 min. After vortexing, tubes were centrifuged for 5 min at 13,000 rpm, and 200 µL of each supernatant was transferred to an HPLC glass vial with a glass insert for analysis. Liquid chromatography–mass spectrometry (LCMS) analysis was performed on a Sciex 6500 QTRAP (Framingham, MA, USA) coupled with an Agilent 1260 ultra-high pressure liquid chromatography (UHPLC) system. Separation was accomplished by using a Waters Atlantis Hillic Silica column (150 mm × 2.1 mm, 5 µM). The pump ﬂow rate was set at 0.5 mL/min with the gradient: 0–1.5 min, 95% B; 3–10.5 min, 80% B; 11–13 min, 60% B; 13.1–20 min, 95% B. Mobile phase A composition consisted of 1000 mL water + 10 mL of 1 M ammonium formate + 2 mL formic acid. Mobile phase B consisted of 950 mL acetonitrile + 40 mL water + 10 mL of 1 M ammonium formate + 2 mL formic acid. The transsulfuration pathway metabolites were separated by UHPLC and monitored using scheduled multiple reaction monitoring (MRM) mode [29]. Heavy stable isotope- Biomolecules 2021, 11, 1073 7 of 17

labeled internal standards for cystathionine, serine, and homocysteine were used for internal calibration.

2.9. Statistics All the results were expressed as mean ± standard error (SEM) of at least 3 independent experiments. Differences among means were considered signiﬁcant if p ≤ 0.05. Normal distribution was determined using either the Shapiro–Wilk test or D’Agostino and Pearson test. To analyze normally distributed data, we used either an unpaired t-test (two- tailed), one-way ANOVA, or 2-way ANOVA, with Tukey’s multiple comparison post-hoc tests. For analyses of non-parametric data, we used the Mann–Whitney test (two-tailed). Statistical calculations were performed using GraphPad Prism 9 (GraphPad Software Inc., San Diego, CA, USA).

3. Results and Discussion 3.1. AOAA-Based Prodrugs Are Weak CBS Inhibitors in the Absence of Bioactivation We initially evaluated the relative potency of each AOAA-based prodrug against the reference compound (i.e., authentic AOAA) in a cell-free assay using puriﬁed recombinant human CBS enzyme [27]. H2S production was measured by the AzMC method, with the results summarized in Table1. We expected that, unlike AOAA, the prodrug structures would be too large to tightly interact with the PLP binding pocket of CBS without esterase- mediated bioactivation. Consequently, their inhibitory potency would be weaker than AOAA. Indeed, simply replacing the hydroxy group of AOAA with a methyl ester (YD0171) reduced the percent inhibition of recombinant CBS-mediated H2S production from ~92% for the reference compound to only ~42% for the prodrug (Table1). In all, 10 of the 15 AOAA-based prodrugs exhibited less than 10% inhibition of recombinant CBS activity when used at a concentration of 100 µM (Table1), demonstrating that in the absence of enzymatic cleavage, the prodrug structures are generally poor CBS inhibitors.

3.2. Bioactivation Enhances the Inhibitory Activity of the AOAA-Based Prodrugs YD0171 and YD0251 To assess the role of bioactivation in prodrug-mediated CBS inhibition, we incubated AOAA, YD0171, or YD0251 (final concentrations, 30 µM) in mouse plasma or a cytosolic extract from HCT116 cancer cells for up to 2 h at 37 ◦C. We selected these biofluids because it is well known that rodent plasma contains relatively high levels of carboxylesterase activity when compared to other species [30–32], and therefore it would be a good positive control. In addition, the cancer cell extract is relevant to our overall goal of developing antitumor therapeutics. At the start of the incubation period (i.e., time = 0) and at 45 min and 2 h (120 min), the recombinant CBS protein was added to the biofluid/AOAA or prodrug mixtures, and H2S production measured using AzMC as described above. Vehicle-treated control plasma and cytosolic extracts “spiked” with recombinant CBS at each time point were used to determine maximum CBS activity and to calculate the percent inhibition of CBS activity by each test compound. At the initial time point (t = 0), the HCT116 cytosolic extract and mouse plasma samples containing AOAA exhibited between 70% and 80% inhibition of vehicle-treated control CBS activity, respectively (Figure3). The inhibitory efficacy of AOAA incubated in the cytosolic extract declined over time, presumably due to either metabolic inactivation or perhaps binding to non-CBS proteins, whereas its inhibitory potency remained stable in plasma (Figure3). Biomolecules 2020, 10, x 8 of 17

Biomolecules 2021, 11, 1073 8 of 17

Figure 3. Comparison of the CBS inhibitory activity of AOAA, YD0251, or YD0171 before and after incubation in cytosolic Figure 3. Comparison of the CBS inhibitory activity of AOAA, YD0251, or YD0171 before and after extracts of HCT116incubation cells in (left cytosolic panel) orextracts mouse of plasma HCT116 (right cells panel). (left panel) Each compound or mouse wasplasma added (right to the panel). biofluid Each to give a final concentrationcompound of 30 µM. was At timeadded 0 min,to the 45 biofluid min, or to 2h, give the a recombinant final concentration CBS was of added 30 μM. and At CBS-mediatedtime 0 min, 45 Hmin,2S production was measuredor 2 using h, the AzMC. recombinant Inhibition CBS of was CBS added activity and was CBS-mediated calculated as follows: H2S production ((vehicle-treated was measured CBS activity—AOAA using or prodrug-treatedAzMC. CBS Inhibition activity)/vehicle-treated of CBS activity was CBS calculated activity) × as100. follows: Data points ((vehicle-treated are mean values CBS activity—AOAA± SEM from 3 independent experiments.or prodrug-treated CBS activity)/vehicle-treated CBS activity) × 100. Data points are mean values ± SEM from 3 independent experiments. The high level of CBS inhibitory activity at time zero was expected for AOAA given The high levelthe fact of CBS that inhibitory it does not activity require at esterase-mediated time zero was expected bioactivation. for AOAA In contrast, given the cytosolic the fact that it doesextracts notand require plasma esterase-med samples containingiated bioactivation. the prodrugs In contrast, exhibited the little cytosolic to no CBS inhibitory extracts and plasmaactivity samples at time containing zero; <15% the for prodrugs YD0171 and exhibited ~0% for little YD0215, to no butCBS showed inhibitory gradually increas- activity at timeing zero; inhibition <15% for at YD0171 45 min and and ~0% 2 h, consistentfor YD0215, with but esterase-mediatedshowed gradually bioactivation increas- (Figure3). ing inhibition atAs 45 expected, min andboth 2 h, consistent compounds with were esterase-mediated activated in mouse bioactivation plasma, with (Figure YD0171 inhibiting 3). As expected,60% both of compounds control CBS were activity acti byvated 2 h in and mouse YD0251 plasma, inhibiting with YD0171 ~40% (Figure inhibit-3). However, in ing 60% of controlthe cancerCBS activity cell cytosolic by 2 h and extracts, YD0251 the inhibiting inhibitor efficacy~40% (Figure of YD0251 3). However, was almost two times in the cancer cellgreater cytosolic than extracts, that of YD0171 the inhibitor at the efficacy final timepoint of YD0251 (Figure was 3almost), suggesting two times that YD0251 may greater than thatbe of preferable YD0171 at to the YD0171 final fortimepoint targeting (Figure tumor 3), CSB suggesting activity. that YD0251 may be preferable to YD0171 for targeting tumor CSB activity. 3.3. AOAA Prodrugs Show Enhanced Antiproliferative Efficacy Compared to AOAA In Vitro 3.3. AOAA ProdrugsHCT116 Show Enhanced cells were Antiprolif treatederative over Efficacy a 48-h time Compared course to either AOAA with In Vitro vehicle (100% growth HCT116 cellscontrol) were ortreated different over concentrations a 48-h time course of AOAA either orwith an vehicle AOAA-based (100% growth prodrug (i.e., 10, 30, µ control) or different100, 300, concentrations and 1000 M).of AOAA In designing or an AOAA-based the different prodrug prodrug (i.e., structures, 10, 30, 100, we reasoned that 300, and 1000 μincreasingM). In designing the lipophilicity the different (i.e., ClogPprodrug value) structures, of AOAA we with reasoned the addition that inof ester- and/or amide-linked functional groups (i.e., R-groups) would enhance the cellular uptake of creasing the lipophilicity (i.e., ClogP value) of AOAA with the addition of ester- and/or the novel structures and thus increase their relative inhibitor potencies when compared amide-linked functional groups (i.e., R-groups) would enhance the cellular uptake of the to the unmodified reference compound. For example, two prodrugs with amide-linked novel structures and thus increase their relative inhibitor potencies when compared to the carbamate groups (YD0381 and YD0239) were found to inhibit HCT116 cell proliferation unmodified reference compound. For example, two prodrugs with amide-linked carba- by ~3% and 8%, respectively, at a concentration of 30 µM, and by ~41% at 300 µM(Table1 ). mate groups (YD0381 and YD0239) were found to inhibit HCT116 cell proliferation by By comparison, the reference compound, AOAA, had no effect on cell replication at ~3% and 8%, respectively, at a concentration of 30 μM, and by ~41% at 300 μM (Table 1). the lower concentrations and only inhibited proliferation by ~36% at 300 µM(Table1 ). By comparison, the reference compound, AOAA, had no effect on cell replication at the The free aminooxy forms of YD0381 (i.e., YD0382) and YD0239 (i.e., YD0242) were also lower concentrations and only inhibited proliferation by ~36% at 300 μM (Table 1). The not effective inhibitors at low concentrations, but induced more inhibition than their free aminooxy forms of YD0381 (i.e., YD0382) and YD0239 (i.e., YD0242) were also not carbamate forms at 300 µM, inhibiting ~50% and ~65% of vehicle-treated cell proliferation, effective inhibitors at low concentrations, but induced more inhibition than their carba- respectively (Table1 ). This observation is not surprising given that, similarly to AOAA, mate forms at 300 μM, inhibiting ~50% and ~65% of vehicle-treated cell proliferation,+ re- the free aminooxy groups of these compounds are protonated (NH3 ) at physiological pH, spectively (Tablewhich, 1). This relative observation to their carbamateis not surprising forms, reducesgiven that, their similarly lipophilicity to AOAA, and likely the slows cellular + free aminooxy uptake.groups However,of these compounds once inside theare cell,protonated YD0382 and(NH YD02423 ) at physiological only require pH, enzyme-mediated which, relativecleavage to their carbamate of one ester-linked forms, reduces functional their group lipophilicity to generate and AOAA,likely slows which cel- may account for lular uptake. However,their increased once inside inhibitory the activitycell, YD0382 at the and highest YD0242 concentration. only require enzyme- mediated cleavage Ofof one all the ester-linked novel compounds functional analyzed group to in generate the HCT116 AOAA, cell proliferation which may assay, YD0171 account for theirand increased YD0251 inhibitory showed the activity greatest at potency,the highest inhibiting concentration. cell growth at a concentration as low Of all the novelas 10 µcompoundsM (Table1). analyzed As shown in the in FigureHCT1164, YD0251cell proliferation inhibited assay, HCT116 YD0171 cell proliferation and YD0251 showedat all concentrationsthe greatest potency, tested inhibiting by 32 h, with cell growth greater at than a concentration 50% inhibition as low of vehicle control proliferation observed at 100 µM. At a concentration of 300 µM, YD0251 exerted a cytostatic Biomolecules 2020, 10, x 9 of 17

Biomolecules 2021, 11, 1073 9 of 17 all concentrations tested by 32 h, with greater than 50% inhibition of vehicle control proliferation observed at 100 μM. At a concentration of 300 μM, YD0251 exerted a cytostatic effect on cell proliferation across the 48-h time course (Figure 4). By comparison, it took a effect on cell proliferation across the 48-h time course (Figure4). By comparison, it took a 10-fold higher10-fold concentrat higherion concentration of AOAA of(i.e., AOAA 1000 (i.e., μM) 1000 to µachieveM) to achieve ~50% ~50%inhibition inhibition of of HCT116 HCT116 cell proliferationcell proliferation at 32 ath 32(Figure h (Figure 4). 4).

Figure 4. EffectsFigure of different4. Effects concentrations of different concentrations of AOAA (left of panel) AOAA and (left YD0251 panel) (right and panel) YD0251 on HCT116(right panel) cell proliferationon In Vitro. HCT116HCT116 cells cell were proliferation treated either In Vitro. with HCT116 vehicle cells (i.e., were test treated compound either diluent) with vehicle or different (i.e., test concentrations compound of the test compoundsdiluent) (10–1000 or differentµM), and concentratio proliferationns of wasthe test monitored compounds for 48(10–1000 h. Proliferation μM), and proliferation rates were monitored was mon- using the xCELLigenceitored Real-time for 48 Cell h. Proliferation Analyzer as rates described were monitored [17]. Data using are shown the xCELLigence as mean of Real-timen = 6 technical Cell Analyzer replicates at each time point. as described [17]. Data are shown as mean of n = 6 technical replicates at each time point.

Thus, these dataThus, indicate these datathat YD0251 indicate is that more YD0251 potent is than more AOAA potent in than inhibiting AOAA CBS in inhibiting CBS in cell-based systemsin cell-based (although systems it is (although less potent it than is less AOAA potent in than the AOAAisolated in CBS the enzyme). isolated CBS enzyme). These propertiesThese of YD0251 properties are of similar YD0251 to areYD0 similar171 [28], to an YD0171 earlier-generation [28], an earlier-generation CBS inhibitor CBS inhibitor prodrug, previouslyprodrug, synthesized previously and synthesized characterized and characterized by us in various by uscolon in variouscancer cell colon lines cancer cell lines in vitro and colonin vitro cancerand colonmodels cancer in vivo. models Thein data vivo presented. The data above presented are consistent above are consistentwith with the the hypothesishypothesis that YD0251 that is YD0251 converted is converted to AOAA to by AOAA intra- byand intra- extracellular and extracellular esterases, esterases, and and ultimately,ultimately, it is AOAA it isthat AOAA is responsi that isble responsible for CBS inhibition. for CBS inhibition. While the Whileprodrug the strat- prodrug strategy egy is useful tois improve useful to cell improve uptake cell and uptake thus cell-based and thus cell-basedpotency, the potency, selectivity the selectivityof YD0251 of YD0251 to to CBS is identicalCBS is to identical the selectivity to the selectivity of AOAA, of AOAA,i.e., not i.e., particularly not particularly selective. selective. Indeed, Indeed, AOAA AOAA is knownis known to also to inhibit also inhibit CSE (another CSE (another H2S-producing H2S-producing enzyme), enzyme), as well as as well a variety as a variety of PLP- of PLP-dependentdependent enzymes enzymes that are that unrelated are unrelated to H2S biosynthesis to H2S biosynthesis (e.g., GABA-transferase (e.g., GABA-transferase and and various transaminases)various transaminases) [26]. These [26 ].additional These additional actions, actions,on a variety on a of variety additional of additional en- enzymes, zymes, may inmay fact in contribute fact contribute to the tocellular the cellular or in vivo or in effects vivo effects of YD0251. of YD0251.

3.4. Ex Vivo Assessment3.4. Ex Vivo of Tumor Assessment and Liver of Tumor CBS and Activity Liver after CBS In Activity Vivo Treatment after In Vivo with Treatment AOAA with AOAA Prodrugs Prodrugs In a separate subset of in vivo experiments, tumor xenografts were initiated by in- In a separate subset of in vivo experiments, tumor xenografts were initiated by injecting 1 × 106 HCT116 cells subcutaneously in the right or left ﬂank of Nu/nu Balb/C jecting 1 × 106 HCT116 cells subcutaneously in the right or left flank of Nu/nu Balb/C mice. mice. When the tumor reached a size of ~150 mm3, the animals were acutely treated either When the tumor reached a size of ~150 mm3, the animals were acutely treated either with with vehicle or the different test compounds (i.e., prodrugs) at a dose of 2 mg/kg, injected vehicle or the different test compounds (i.e., prodrugs) at a dose of 2 mg/kg, injected sub- subcutaneously every 6 h for 24 h (total dose; 10 mg/kg). In total, 4–6 mice were used cutaneously everyto evaluate 6 h for each24 h (total test compound. dose; 10 mg At/kg). the In end total, of 4–6 the inmice vivo weretreatment used to period,eval- tumor and uate each testliver compound. tissue homogenates At the end wereof the prepared in vivo treatment for ex vivo period, CBS activity tumor assays and liver using the AzMC tissue homogenatesmethod. were The data,prepared expressed for ex as vi percentvo CBS inhibition activity assays of vehicle-treated using the controlAzMC CBS activity, method. The data,are shown expressed in the as last percent two columns inhibition of of Table vehicle-treated1. control CBS activity, are shown in the lastTreating two columns mice with of aTable total 1. of 10 mg/kg of AOAA over 24 h had no effect on the tumor Treating CBSmice activity with a total ex vivo of 10 (Table mg/kg1). Inof otherAOAA words, over the24 h amount had no ofeffect CBS on activity the tumor measured ex vivo CBS activity exin tumorvivo (Table homogenates 1). In other from word AOAA-treateds, the amount mice of wasCBS equivalentactivity measured to the activity ex measured vivo in tumorin homogenates the tumor homogenates from AOAA-treated from vehicle-treated mice was equivalent mice, suggesting to the activity that meas- the HCT116 tumor ured in the tumortissue homogenates took up and/or from retained vehicle-treated relatively mice, little AOAAsuggesting during that the the acute HCT116in vivo treatment tumor tissue tookperiod. up and/or By comparison, retained relatively the liver homogenateslittle AOAA during from AOAA-treated the acute in vivo mice treat- exhibited ~36% ment period. lessBy comparison, CBS activity the than liver the homo liversgenates from vehicle-treated from AOAA-tre mice.ated YD0242, mice exhibited YD0541, and YD0706 ~36% less CBSbehaved activity similarly than the tolivers AOAA from in vehicle-treated that they inhibited mice. vehicle-treated YD0242, YD0541, liver and CBS activity, but did not inhibit tumor CBS activity, as measured by tissue homogenate H2S production (Table1 ). Three test compounds (YD0343, YD0382, and YD0452) had no effects on either Biomolecules 2020, 10, x 10 of 17

YD0706 behaved similarly to AOAA in that they inhibited vehicle-treated liver CBS activ- Biomolecules 2021, 11, 1073 10 of 17 ity, but did not inhibit tumor CBS activity, as measured by tissue homogenate H2S production (Table 1). Three test compounds (YD0343, YD0382, and YD0452) had no effects on either tumor or liver ex vivo H2S production; several other test compounds (YD0246, YD0251,tumor or and liver YD0335), ex vivo H inhibited2S production; both tumor several and other liver test CBS compounds activity. (YD0246,On average, YD0251, the inhi- and bitionYD0335), of liver inhibited CBS by both these tumor compounds and liver was CBS ~15% activity. greater On than average, that themeasured inhibition in the of livercor- respondingCBS by these tumor compounds tissue homogenates. was ~15% greater Greater than inhibition that measured of liver in theCBS corresponding than tumor activity tumor wastissue observed homogenates. with 9 of Greater the 15 test inhibition compounds of liver (Table CBS 1), than suggesting tumor activity that the wasliver observed cells are morewith 9efficient of the 15 than test cancer compounds cells (Tableat taking1), suggestingup and/or thatbioactivating the liver cells most are of more the prodrug efficient structures.than cancer Only cells two at taking test compounds up and/or (YD0171 bioactivating and YD0239) most of thewere prodrug more effective structures. at inhib- Only itingtwo testtumor compounds CBS activity (YD0171 than liver and using YD0239) the wereacute more dosing effective regimen. at inhibitingTumor homogenates tumor CBS fromactivity YD0171-treated than liver using mice the exhibited acute dosing~68% less regimen. CBS activity Tumor than homogenates tumor homogenates from YD0171- from vehicle-treatedtreated mice exhibited control ~68% animals, less CBS whereas activity liver than tumorhomogenates homogenates from fromthe same vehicle-treated animals showedcontrol animals,only an ~15% whereas reduction liver homogenates of control liver from CBS the activity same animals (Table showed1). Similarly, only anYD0239 ~15% reducedreduction tumor of control ex vivo liver CBS CBS activity activity by (Table ~63%1). relative Similarly, to vehicle-treated YD0239 reduced control tumor tumors, ex vivo whileCBS activity exerting by no ~63% inhibitory relative activity to vehicle-treated on liver CBS. control tumors, while exerting no inhibitory activity on liver CBS. 3.5. YD0251 Exhibits Improved Antitumor Potency In Vivo When Compared to AOAA 3.5. YD0251 Exhibits Improved Antitumor Potency In Vivo When Compared to AOAA To assess the antitumor efficacy of selected AOAA-based prodrugs with long-term To assess the antitumor efficacy of selected AOAA-based prodrugs with long-term dosing (i.e., 14–28 days), we initially focused on YD0251 and YD0239 because these com- dosing (i.e., 14–28 days), we initially focused on YD0251 and YD0239 because these compounds were significantly more potent than AOAA in inhibiting HCT116 cell prolifera- pounds were significantly more potent than AOAA in inhibiting HCT116 cell proliferation tion in vitro (Table 1). However, unexpectedly, a preliminary in vivo evaluation of YD0239 in vitro (Table1). However, unexpectedly, a preliminary in vivo evaluation of YD0239 showed a significant (p <0.005) increase in the size of HCT116 tumor xenograft growth in showed a significant (p < 0.005) increase in the size of HCT116 tumor xenograft growth mice treated for 24 days (6 mg/kg/daily), when compared to vehicle-treated animals (Fig- in mice treated for 24 days (6 mg/kg/daily), when compared to vehicle-treated animals ure 5). The mechanism behind this enhancement is unclear but may be due to a hitherto (Figure5). The mechanism behind this enhancement is unclear but may be due to a hitherto unknown secondary enzymatic target of this particular molecule. unknown secondary enzymatic target of this particular molecule.

Figure 5. Effects of YD0239 on HCT116 tumor xenograft growth In Vivo. HCT116 tumor-bearing Figure 5. Effects of YD0239 on HCT116 tumor xenograft growth In Vivo. HCT116 tumor-bearing Nu/nuNu/nu Balb/C Balb/C mice mice were were treated treated with with either either vehicl vehiclee or YD0239 or YD0239 (6 mg/kg/daily) (6 mg/kg/daily) via s.q. via injection s.q. injection once aonce day afor day 24 for days. 24 days.Tumor Tumor size (volume) size (volume) was meas was measuredured transcutaneously transcutaneously with witha caliper a caliper every every 2–3 days.2–3 days. (n = 6 (n mice/group,= 6 mice/group, ** p < **0.01p < showing 0.01 showing significant signiﬁcant enhancement enhancement of tumor of tumor size). size).

InIn the the remainder remainder of of our our studies, studies, we we focuse focusedd on on the further characterization of of the the antitumorantitumor properties properties of of YD0251. YD0251. Our Our results results de demonstratedmonstrated that that YD0251 YD0251 is is 3 3 to to 18 18 times moremore potent potent than than AOAA AOAA inin vivo. vivo. Treating Treating HCT116 HCT116 tumor-bearing tumor-bearing mice mice via via subcutaneous subcutaneous injectioninjection with with either either a adaily daily dose dose of of0.5, 0.5, 1, or 1, or3 mg/kg 3 mg/kg YD0251 YD0251 markedly markedly inhibited inhibited the rate the ofrate xenograft of xenograft growth growth (Figure (Figure 6A). Importantly,6A). Importantly, the lower the lower doses dosesof YD0251 of YD0251 were as were effica- as ciousefﬁcacious as treating as treating with 9 with mg/kg 9 mg/kg of AOAA of AOAA (Figure (Figure 6A). 6Furthermore,A). Furthermore, YD0251 YD0251 at doses at doses of 1 andof 1 3 and mg/kg 3 mg/kg effectively effectively inhibited inhibited in vivo intumor vivo growthtumor both growth in the both subcutaneous in the subcutaneous injection injection and the oral gavage therapeutic regimens (Figure6A,B). and the oral gavage therapeutic regimens (Figure 6A,B). Biomolecules 20202021, 1011, x 1073 1111 of of 17 17

Figure 6. 6. YD0251YD0251 is is more more potent potent than than AOAA AOAA at at inhibiting inhibiting HCT HCT116116 tumor tumor xenograft xenograft growth growth In In Vivo. Vivo. The The compounds compounds were were administered daily at the indicated doses by subcutaneous subcutaneous injection ( A,,B)) and and oral oral gavage gavage ( (CC).). Data Data are are expressed expressed as as ratio ratio of tumor weight to mouse body weight at harvest following 21 days of treatment. YD0251 YD0251 did did not not affect affect the the weight weight of the animals overover thethe course course of of the the experiments, experiments, while while AOAA AOAA induced induced a signiﬁcant a significant weight lossweight by Day loss 31 by (D Day). (n =31 6 mice/group,(D). (n = 6 mice/group,* p < 0.05, † p *

3.6. YD0251 Inhibits Patient-Derived Tumor Xenograft Growth 3.6. YD0251 Inhibits Patient-Derived Tumor Xenograft Growth Next, we tested the growth inhibitory efficacy of YD0251 using patient-derived tu- Next, we tested the growth inhibitory efficacy of YD0251 using patient-derived tu- mormor xenografts xenografts (PDTXs). (PDTXs). Low-passage Low-passage PDTXs PDTXs have have emerged emerged as as an important preclinical modelmodel for oncologicaloncological drug drug discovery discovery because because they they maintain maintain the histologicalthe histological features, features, gene geneexpression expression profile, profile, and mutationand mutation status status of the of patient’s the patient’s original original tumor tumor tissue tissue [33]. [33]. ColorectalColorectal tumorstumors resectedresected from from six six different different patients patients were were used used to develop to develop the PDTXs the PDTXsfor these for studies. these studies. Based Based on their on levelstheir levels of CBS, of CBS, the six the PDTXs six PDTXs were were divided divided into into two twogroups groups of three of three (Figure (Figure7A,B). 7A,B). Moreover, Moreover the relative, the relative growth-inhibitory growth-inhibitory effect of effect YD0251 of YD0251was significantly was significantly more pronounced more pronounced in the in high the CBS-expressorhigh CBS-expressor cohort cohort than than in the in lowthe lowCBS-expressor CBS-expressor cohort cohort (Figure (Figure7F). 7F).

3.7. YD0251 Exerts Antiproliferative Effects in Multidrug-Resistant Colon Cancer Cells The efficacy of many anticancer drugs diminishes when cancer cells assume a multidrug resistant phenotype. We have recently established a 5-FU-resistant HCT116 human colon cancer cell line by serial passage in the presence of increasing 5-FU concentrations [21]. These cells also exhibit a partial resistance to an unrelated chemotherapeutic agent, oxaliplatin. The multidrug-resistant cells exhibit a significant increase in the expression of the drug-metabolizing cytochrome P450 enzymes CYP1A2 and CYP2A6 and they also upregulate CBS (as well as another major H2S-generating enzyme, 3-MST) [21]. The 5- FU-resistant cells exhibited decreased sensitivity to AOAA but remained sensitive to the antiproliferative effect of benserazide (a recently identified, potentially repurposable CBS inhibitor) [21]. Herein we compared the effects of YD0251 and 5-FU treatments on the viability of multidrug-resistant and parental (normal) HCT116 cell lines. As shown in Figure8A, 5-FU caused a concentration-dependent decrease in the percentage of viable parental HCT116 but had no effect on the viability of resistant cells. In contrast, YD0251 reduced the viability of both resistant and parental cells (Figure8B). BiomoleculesBiomolecules2021 2020,,11 10,, 1073 x 1212 of 1717

FigureFigure 7.7. InIn VivoVivo efficacyefficacy ofof YD0251YD0251 onon thethe growthgrowth ofof PDTXPDTX inin nudenude mice.mice. ((AA)) ExampleExample WesternWestern blotblot showingshowing varyingvarying levelslevels ofof CBSCBS proteinprotein inin PDTXsPDTXs fromfrom 66 differentdifferent patients.patients. ((BB)) ComparisonComparison ofof thethe relativerelative levelslevels ofof CBSCBS proteinprotein expressionexpression inin thethe variousvarious tumors.tumors. ((CC––EE)) TheThe effect effect of of YD0251 YD0251 (subcutaneous (subcutaneous dosing, dosing, 6 6 mg/kg/day) mg/kg/day) is shown inin nudenude micemice bearingbearing humanhuman patient-derivedpatient-derived xenograftsxenografts exhibiting high high ( (CC,,DD)) or or low low (E (E) )CBS CBS expression. expression. Note Note that that YD0251 YD0251 was was efficacious efficacious in in6/8 6/8 patients; patients; it was it was efficacious efficacious in the in thePDTX's PDTX’s of all of high all high CBS CBS expressors expressors and and in 50% in 50% of PDTX's of PDTX’s of low of lowCBS CBS expressors expressors (n = mice/group, * p < 0.05, ** p < 0.01, *** p < 0.001, ns: not significant). It was found that 1 PDTX with high CBS levels and 2 (n = mice/group, * p < 0.05, ** p < 0.01, *** p < 0.001, ns: not significant). It was found that 1 PDTX with high CBS levels and PDTXs with low levels of CBS possessed activating mutations of the KRAS gene. The KRAS status of the tumors did not Biomolecules2 PDTXs 2020, with10, x low levels of CBS possessed activating mutations of the KRAS gene. The KRAS status of the tumors did13 not of 17 correlate with the growth inhibitory efficacy of YD0251. Overall, high CBS-expressing PDTXs grew faster and the effect of correlateYD0251 was with more the growth pronounced inhibitory in them efficacy than of in YD0251. low CBS-expressing Overall, high PDTXs CBS-expressing (F). PDTXs grew faster and the effect of YD0251 was more pronounced in them than in low CBS-expressing PDTXs (F). 3.7. YD0251 Exerts Antiproliferative Effects in Multidrug-Resistant Colon Cancer Cells The efficacy of many anticancer drugs diminishes when cancer cells assume a multidrug resistant phenotype. We have recently established a 5-FU-resistant HCT116 human colon cancer cell line by serial passage in the presence of increasing 5-FU concentrations [21]. These cells also exhibit a partial resistance to an unrelated chemotherapeutic agent, oxaliplatin. The multidrug-resistant cells exhibit a significant increase in the expression of the drug-metabolizing cytochrome P450 enzymes CYP1A2 and CYP2A6 and they also upregulate CBS (as well as another major H2S-generating enzyme, 3-MST) [21]. The 5-FU- resistant cells exhibited decreased sensitivity to AOAA but remained sensitive to the antiproliferative effect of benserazide (a recently identified, potentially repurposable CBS FigureFigureinhibitor) 8. 8. TheThe [21]. inhibitory inhibitory Herein effect effect we ofcompared YD0251YD0251 onon the cell cell effects viability viability of is YD0251is largely largely maintained andmaintained 5-FU in treatments multidrug-resistantin multidrug-re- on the sistantHCT116viability HCT116 cells. of multidrug-resistant cells. HCT116 HCT116 cells cells were wereand maintained parentalmaintain in ed(normal) a in 5-FU-containing a 5-FU-containing HCT116 cell medium lines.medium for As 2 forshown months. 2 months. in CellFig- Cellassumedure assumed 8A, 5-FU a drug-resistant a drug-resistant caused a concentration-dependent phenotype phenotype and an lostd lost their their responsiveness responsivenessdecrease in tothe to 5-FU percentage 5-FU (A (),A as), as assessed ofassessed viable by by thepa- theMTTrental MTT assay HCT116assay 24 24 h h afterbut after had incubation incubation no effectwith with on variousthevarious viabilit concentrationsconcentrationsy of resistant of YD0251. cells. In ( B (contrast,B) )YD0251 YD0251 YD0251 treatment treatment re- resultedresultedduced in the in a aconcentration-dependentviability concentration-dependent of both resistant reduction reduction and parental in in cell cell viability cells viability (Figure (n ( n= =5, 5,**8B).** p p< <0.01). 0.01).

3.8.3.8. YD0251 YD0251 at atDoses Doses at atWhich Which It ItExerts Exerts Antica Anticancerncer Effects Effects Does Does Not Not In Induceduce Significant Signiﬁcant Organ Organ InjuryInjury ToTo begin begin evaluating evaluating the the potential potential inin vivo vivo toxicitytoxicity of of YD0251, YD0251, we we administered administered inin- creasingcreasing doses doses to tomice mice via via intraperitoneal intraperitoneal injection injection and and measured measured plasma plasma markers markers of or- of ganorgan damage damage as well as well as liver as liver and and kidney kidney histol histology.ogy. As As shown shown in inFigure Figure 9,9 ,no no significant signiﬁcant changeschanges were were noted noted in in either either plasma plasma marker marker or tissue histologyhistology inin animalsanimals receivingreceiving up up to to60 60 mg/kg/day mg/kg/day of of YD0215 for 7 days.

Figure 9. Effect of YD0251 treatment on plasma markers of organ damage and tissue histology. Mice were subjected to intraperitoneal injection of YD0251 in 2 divided doses (total daily doses: 2 mg/kg/day, 6 mg/kg/day, 20 mg/kg/day, and 60 mg/kg/day) for 7 consequent days (1 week). On the morning of the eighth day, the experiment was terminated, and blood was collected and analyzed by the Vetscan system (ALT: alanine aminotransferase; ALP: alkaline phosphatase; TBIL: total bilirubin; BUN: blood urea nitrogen). C: vehicle control group. The mean ± SEM of n = 5 mice per group Biomolecules 2020, 10, x 13 of 17

Figure 8. The inhibitory effect of YD0251 on cell viability is largely maintained in multidrug-resistant HCT116 cells. HCT116 cells were maintained in a 5-FU-containing medium for 2 months. Cell assumed a drug-resistant phenotype and lost their responsiveness to 5-FU (A), as assessed by the MTT assay 24 h after incubation with various concentrations of YD0251. (B) YD0251 treatment resulted in a concentration-dependent reduction in cell viability (n = 5, ** p < 0.01).

3.8. YD0251 at Doses at Which It Exerts Anticancer Effects Does Not Induce Significant Organ Injury To begin evaluating the potential in vivo toxicity of YD0251, we administered increasing doses to mice via intraperitoneal injection and measured plasma markers of organ damage as well as liver and kidney histology. As shown in Figure 9, no significant Biomolecules 2021, 11, 1073 changes were noted in either plasma marker or tissue histology in animals receiving13 of up 17 to 60 mg/kg/day of YD0215 for 7 days.

FigureFigure 9. EffectEffect of of YD0251 YD0251 treatment treatment on onplasma plasma markers markers of organ of organ damage damage and tissue and tissue histology. histology. Mice Micewere weresubjected subjected to intraperitoneal to intraperitoneal injection injection of YD0251 of YD0251 in 2 in divided 2 divided doses doses (total (total daily daily doses: doses: 2 2mg/kg/day, mg/kg/day, 6 mg/kg/day, 6 mg/kg/day, 20 mg/kg/day 20 mg/kg/day,, and 60 and mg/kg/day) 60 mg/kg/day) for 7 conseq for 7 consequentuent days (1 daysweek). (1 week).On the Onmorning the morning of the eighth of the day, eighth the day,experiment the experiment was terminated, was terminated, and blood and was blood collected was collectedand analyzed and analyzedby the Vetscan by the system Vetscan (ALT: system alanine (ALT: aminotransfe alanine aminotransferase;rase; ALP: alkaline ALP: phosphatase; alkaline phosphatase; TBIL: total TBIL: bili- totalrubin; bilirubin; BUN: blood BUN: urea blood nitrogen). urea nitrogen). C: vehicle C: cont vehiclerol group. control The group. mean The ± SEM mean of± n SEM= 5 mice of n per= 5 group mice per group is shown. In addition, there were no detectable histological alterations in any of the groups (representative H&E pictures are shown).

3.9. Identification of Lanthionine as a Potential Urinary Biomarker of High CBS-Expressor Tumors Finally, we tested the theranostic concept that lanthionine [16] might serve as a biomarker for tumors with high levels of CBS expression. Lanthionine is a natural nonpro- teogenic amino acid, an analogue of cysteine, consisting of two alanine residues crosslinked on their β-carbon atoms by a thioether linkage. The condensation of two molecules of cysteine (β-replacement reaction) produces H2S, as well as lanthionine as a side product. Additionally, the other transsulfuration enzyme (CSE) does not contribute to lanthionine production and there are currently no known enzymes that produce lanthionine (other than CBS) [34,35]. Lanthionine was readily detectable by mass spectrometry in the culture supernatants from HCT116 cells, and inhibition of CBS activity with AOAA reduced lanthionine levels by ~80% (Figure 10A). Furthermore, forced expression of recombinant CBS in NCM356 cells increased lanthionine production over 12-fold and enhanced their tumorigenicity in vivo (Figure 10B,C, respectively). Importantly, we detected significant higher level of lanthionine in the urine of mice bearing the CBS-overexpressing NCM356 tumor, as compared to mice bearing the control NCM356 tumor (Figure 10D). These data provide a rationale to explore the use of plasma or urinary lanthionine as a marker of high CBS- expressor tumors in humans with the goal of identifying patients that could potentially benefit from CBS inhibitory therapy. Biomolecules 2020, 10, x 14 of 17

is shown. In addition, there were no detectable histological alterations in any of the groups (representative H&E pictures are shown).

3.9. Identification of Lanthionine as a Potential Urinary Biomarker of High CBS-Expressor Tumors Finally, we tested the theranostic concept that lanthionine [16] might serve as a biomarker for tumors with high levels of CBS expression. Lanthionine is a natural nonpro- teogenic amino acid, an analogue of cysteine, consisting of two alanine residues crosslinked on their β-carbon atoms by a thioether linkage. The condensation of two molecules of cysteine (β-replacement reaction) produces H2S, as well as lanthionine as a side product. Additionally, the other transsulfuration enzyme (CSE) does not contribute to lanthionine production and there are currently no known enzymes that produce lanthionine (other than CBS) [34,35]. Lanthionine was readily detectable by mass spectrometry in the culture supernatants from HCT116 cells, and inhibition of CBS activity with AOAA reduced lanthionine levels by ~80% (Figure 10A). Furthermore, forced expression of recombinant CBS in NCM356 cells increased lanthionine production over 12-fold and enhanced their tumorigenicity in vivo (Figure 10B,C, respectively). Importantly, we detected significant higher level of lanthionine in the urine of mice bearing the CBS-overexpressing NCM356 tumor, as compared to mice bearing the control NCM356 tumor (Figure 10D). These data provide a rationale to explore the use of plasma or urinary lanthionine as a marker of high CBS-ex- Biomolecules 2021, 11, 1073 14 of 17 pressor tumors in humans with the goal of identifying patients that could potentially benefit from CBS inhibitory therapy.

Figure 10.10. Lanthionine is a potentialpotential biomarkerbiomarker of CBS-expressingCBS-expressing coloncolon cancercancer cells.cells. (A) InhibitionInhibition ofof CBSCBS suppressessuppresses lanthioninelanthionine productionproduction in in HCT116 HCT116 cells. cells. (B ()B Forced) Forced overexpression overexpression of CBSof CBS increases increa lanthionineses lanthionine production production in NCM356 in NCM356 cells. (cells.C) Effect (C) ofEffect CBS overexpressionof CBS overexpression on NCM356 on tumorNCM356 volume tumor 20 daysvolume after 20 implantation. days after implanta CBS-overexpressingtion. CBS-overexpressing NCM356 cells NCM356 cells grow larger tumors than parental cells in nude mice. (D) Comparison of urine lanthionine levels NCM356 grow larger tumors than parental cells in nude mice. (D) Comparison of urine lanthionine levels NCM356 tumor-bearing tumor-bearing mice 1 day and 20 days after tumor cell implantation (n = 5, ** p < 0.01). mice 1 day and 20 days after tumor cell implantation (n = 5, ** p < 0.01). 4. Conclusions 4. Conclusions In conclusion, the current study identified and characterized the isopropyl ester de- rivativeIn conclusion,of AOAA, YD0251, the current as a novel study CBS identified inhibitor and molecule characterized (AOAA the prodrug) isopropyl with ester sig- derivative of AOAA, YD0251, as a novel CBS inhibitor molecule (AOAA prodrug) with nificant antiproliferative efficacy in cell culture models of colon cancer (including efficacy Biomolecules 2020, 10, x significant antiproliferative efficacy in cell culture models of colon cancer15 (includingof 17 efficacy in multi-drug resistant variants of human colon cancer cell lines), and with significant in multi-drug resistant variants of human colon cancer cell lines), and with significant antitumor effects in tumor-bearing mice models (Figure 11). antitumor effects in tumor-bearing mice models (Figure 11). The current report also shows the diversity in human colon cancer tissues in terms of degree of CBS expression; colon cancers which exhibit high CBS expression proliferate faster when placed in PDTX models than the low CBS-expressor PDTX counterparts, and respond better to the CBS inhibitor’s antiproliferative effect. YD0251 was effective when administered both by subcutaneous injection and oral administration, predicting the oral bioavailability of the molecule. At the doses of YD0251 which were effective in the tumor- bearing mice, as well as at several fold higher doses, the molecule was well tolerated, as evidenced by the fact that repeat administration of relatively high doses did not produce signs of organ injury.

Figure 11. Mechanism of YD0251’s action in colon cancer cells. YD0251, due to its high lipophilicity, readily enters cancer Figure 11. Mechanism of YD0251’s action in colon cancer cells. YD0251, due to its high lipophilicity, cells. After conversionreadily enters by esterases,cancer cells. AOAA After is conversion released, which by esterases, in turn inhibitsAOAA is intratumor released, which CBS activity, in turn decreases inhibits tumoral H2S levels, andintratumor exerts antitumor CBS activity actions, decreasesin vitro tumoraland in vivo H2S. Theselevels, effects and exerts are most antitumo pronouncedr actions in in tumors vitro and with in higher CBS expression. vivo. These effects are most pronounced in tumors with higher CBS expression.

Finally, with anThe eye current towards report future also theran showsostic the applications diversity in human(i.e., matching colon cancer CBS in- tissues in terms hibitor therapyof to degree patients of CBSwith expression;high CBS-expr colonessing cancers tumors), which the exhibit current high study CBS identifies expression proliferate faster when placed in PDTX models than the low CBS-expressor PDTX counterparts, and lanthionine (a metabolic intermediate of CBS-mediated H2S biosynthesis) as a potential biochemical markerrespond of betterCBS upregulation. to the CBS inhibitor’s In summary, antiproliferative the current findings effect. YD0251provide wasaddi- effective when tional proof-of-conceptadministered that bothCBS byis a subcutaneous significant antitumor injection target and oral in colon administration, cancer and predictingsup- the oral bioavailability of the molecule. At the doses of YD0251 which were effective in the tumor- port continued efforts to investigate and eventually clinically develop novel antitumor bearing mice, as well as at several fold higher doses, the molecule was well tolerated, as therapeutics based on CBS inhibition; perhaps in conjunction with the validation of a suit- evidenced by the fact that repeat administration of relatively high doses did not produce able biomarker (e.g., lanthionine) to identify a high CBS-expressor (and likely a best-re- signs of organ injury. sponder) colon cancer target population. Finally, with an eye towards future theranostic applications (i.e., matching CBS inhibitor therapy to patients with high CBS-expressing tumors), the current study identiﬁes Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1. lanthionine (a metabolic intermediate of CBS-mediated H2S biosynthesis) as a potential bio- Author Contributions: Conceptualization: M.R.H., C.C., J.Z. and C.S.; methodology: M.R.H. C.C., W.K.R., J.Z. and C.S; formal analysis: M.R.H. and C.S.; investigation: C.C., K.M., Y.D., J.R.Z., K.T., M.M., N.D., A.A.U., A.A., Y.X., H.C., W.K.R., J.W. and J.Z.; resources: M.R.H., J.Z. and C.S.; data curation: M.R.H. and C.S.; writing—original draft preparation: C.S. and M.R.H.; writing—review and editing, M.R.H., J.Z. K.M. and C.C.; supervision: M.R.H., C.C., W.K.R., J.Z. and C.S.; project administration: M.R.H., J.Z. and C.S.; funding acquisition: M.R.H., C.C., K.M., W.K.R., J.Z. and C.S. All authors have read and agreed to the published version of the manuscript. Funding: The laboratories of Szabo, Hellmich, and Zhou in the field of CBS and cancer were or currently are funded by the Cancer Prevention Research Institute of Texas (CPRIT), Texas, USA (DP150074), the National Institutes of Health, Bethesda, USA (R41CA213463), and the Swiss Kreb- sliga (KLS–4504–08–2018). Zhou is also partly supported by the John D. Stobo, M.D. Distinguished Chair Endowment Fund at UTMB. Zatarain and Thanki were supported by a T32 training grant (DK007639) from the National Institutes of Health, Bethesda, USA. Modis was funded by the De- partment of Defense Career Development Award (W81XWH2010641) and the American Cancer So- ciety Research Scholar Grant (RSG–21–027–01–CSM). The Russell and the UTMB Mass Spectrome- try Facility are funded in part by CPRIT, Texas USA, (RP190682). Institutional Review Board Statement: Discarded human surgical samples were collected was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of the UTMP as a part of a discarded tissue, biobanking protocol. Informed Consent Statement: The subjects whose discarded surgical specimens were collected from were not part of a clinical trial; individual patient informed consent was not required at the time of the IRB approval. Biomolecules 2021, 11, 1073 15 of 17

chemical marker of CBS upregulation. In summary, the current ﬁndings provide additional proof-of-concept that CBS is a signiﬁcant antitumor target in colon cancer and support continued efforts to investigate and eventually clinically develop novel antitumor therapeutics based on CBS inhibition; perhaps in conjunction with the validation of a suitable biomarker (e.g., lanthionine) to identify a high CBS-expressor (and likely a best-responder) colon cancer target population.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/biom11081073/s1. Author Contributions: Conceptualization: M.R.H., C.C., J.Z. and C.S.; methodology: M.R.H., C.C., W.K.R., J.Z. and C.S.; formal analysis: M.R.H. and C.S.; investigation: C.C., K.M., Y.D., J.R.Z., K.T., M.M., N.D., A.A.U., A.A., Y.X., H.C., W.K.R., J.W. and J.Z.; resources: M.R.H., J.Z. and C.S.; data curation: M.R.H. and C.S.; writing—original draft preparation: C.S. and M.R.H.; writing—review and editing, M.R.H., J.Z., K.M. and C.C.; supervision: M.R.H., C.C., W.K.R., J.Z. and C.S.; project administration: M.R.H., J.Z. and C.S.; funding acquisition: M.R.H., C.C., K.M., W.K.R., J.Z. and C.S. All authors have read and agreed to the published version of the manuscript. Funding: The laboratories of Szabo, Hellmich, and Zhou in the field of CBS and cancer were or currently are funded by the Cancer Prevention Research Institute of Texas (CPRIT), Texas, USA (DP150074), the National Institutes of Health, Bethesda, USA (R41CA213463), and the Swiss Krebsliga (KLS–4504–08–2018). Zhou is also partly supported by the John D. Stobo, M.D. Distinguished Chair Endowment Fund at UTMB. Zatarain and Thanki were supported by a T32 training grant (DK007639) from the National Institutes of Health, Bethesda, USA. Modis was funded by the Department of Defense Career Development Award (W81XWH2010641) and the American Cancer Society Research Scholar Grant (RSG–21–027–01–CSM). The Russell and the UTMB Mass Spectrometry Facility are funded in part by CPRIT, Texas USA, (RP190682). Institutional Review Board Statement: Discarded human surgical samples were collected was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of the UTMP as a part of a discarded tissue, biobanking protocol. Informed Consent Statement: The subjects whose discarded surgical specimens were collected from were not part of a clinical trial; individual patient informed consent was not required at the time of the IRB approval. Data Availability Statement: The data presented in this study are available on request from the corresponding authors. Acknowledgments: The technical assistance of Gabor Olah and Gabor Törö is appreciated. Ciro Coletta has also participated in early-stage experiments that conceptually belong to the current project—he tragically passed away in 2015; he is gone but not forgotten. Conflicts of Interest: The novel structure of the compound YD0171 has been patented; the patent is owned by the University of Texas Medical Branch (UTMB), and some of the authors of the current paper (JZ, MH and CS) are inventors on this patent. These authors would benefit from the potential commercialization of this molecule according to UTMB’s standard technology transfer and commercialization policies. The other authors declare no conflict of interest.

References 1. Abe, K.; Kimura, H. The possible role of hydrogen sulfide as an endogenous neuromodulator. J. Neurosci. 1996, 16, 1066–1071. [CrossRef] 2. Kimura, H. Hydrogen Sulfide as a Neuromodulator. Mol. Neurobiol. 2002, 26, 013–020. [CrossRef] 3. Szabó, C. Hydrogen sulphide and its therapeutic potential. Nat. Rev. Drug Discov. 2007, 6, 917–935. [CrossRef][PubMed] 4. Li, L.; Rose, P.; Moore, P.K. Hydrogen Sulfide and Cell Signaling. Annu. Rev. Pharmacol. Toxicol. 2011, 51, 169–187. [CrossRef] 5. Wang, R. Physiological Implications of Hydrogen Sulfide: A Whiff Exploration That Blossomed. Physiol. Rev. 2012, 92, 791–896. [CrossRef][PubMed] 6. Paul, B.D.; Snyder, S.H. H 2 S: A Novel Gasotransmitter that Signals by Sulfhydration. Trends Biochem. Sci. 2015, 40, 687–700. [CrossRef] 7. Panthi, S.; Chung, H.–J.; Jung, J.; Jeong, N.Y. Physiological Importance of Hydrogen Sulfide: Emerging Potent Neuroprotector and Neuromodulator. Oxidative Med. Cell. Longev. 2016, 2016, 1–11. [CrossRef] Biomolecules 2021, 11, 1073 16 of 17

8. Szabo, C.; Papapetropoulos, A. International Union of Basic and Clinical Pharmacology. CII: Pharmacological Modulation of H2S Levels: H2S Donors and H2S Biosynthesis Inhibitors. Pharmacol. Rev. 2017, 69, 497–564. [CrossRef] 9. Zezos, P.; Papaioannou, G.; Nikolaidis, N.; Vasiliadis, T.; Giouleme, O.; Evgenidis, N. Hyperhomocysteinemia in ulcerative colitis is related to folate levels. World J. Gastroenterol. 2005, 11, 6038–6042. [CrossRef] 10. Kim, J.; Hong, S.J.; Park, J.H.; Park, S.Y.; Kim, S.; Cho, E.Y.; Do, I.–G.; Joh, J.–W.; Kim, D.S. Expression of cystathionine β–synthase is downregulated in hepatocellular carcinoma and associated with poor prognosis. Oncol. Rep. 2009, 21, 1449–1454. [CrossRef] [PubMed] 11. Akbulut, S. Increased levels of homocysteine in patients with ulcerative colitis. World J. Gastroenterol. 2010, 16, 2411–2416. [CrossRef][PubMed] 12. Jain, S.K.; Bull, R.; Rains, J.L.; Bass, P.F.; Levine, S.N.; Reddy, S.; McVie, R.; Bocchini, J.A. Low Levels of Hydrogen Sulfide in the Blood of Diabetes Patients and Streptozotocin–Treated Rats Causes Vascular Inflammation? Antioxid. Redox Signal. 2010, 12, 1333–1337. [CrossRef][PubMed] 13. Alshorafa, A.K.; Guo, Q.; Zeng, F.; Chen, M.; Tan, G.; Tang, Z.; Yin, R. Psoriasis Is Associated with Low Serum Levels of Hydrogen Sulfide, a Potential Anti–inflammatory Molecule. Tohoku J. Exp. Med. 2012, 228, 325–332. [CrossRef][PubMed] 14. Zheng, S.; Yang, W.; Wu, C.; Lin, D.; Ding, R.; Sun, L.; Jiang, L. Association of ulcerative colitis with transcobalamin II gene polymorphisms and serum homocysteine, vitamin B12, and folate levels in Chinese patients. Immunogenetics 2017, 69, 421–428. [CrossRef][PubMed] 15. Szabo, C. The re-emerging pathophysiological role of the cystathionine-β-synthase—Hydrogen sulfide system in Down syndrome. FEBS J. 2020, 287, 3150–3160. [CrossRef][PubMed] 16. Szabo, C.; Coletta, C.; Chao, C.; Módis, K.; Szczesny, B.; Papapetropoulos, A.; Hellmich, M.R. Tumor–derived hydrogen sulfide, produced by cystathionine–β–synthase, stimulates bioenergetics, cell proliferation, and angiogenesis in colon cancer. Proc. Natl. Acad. Sci. USA 2013, 110, 12474–12479. [CrossRef] 17. Bhattacharyya, S.; Saha, S.; Giri, K.; Lanza, I.; Nair, K.S.; Jennings, N.B.; Rodriguez–Aguayo, C.; Lopez–Berestein, G.; Basal, E.; Weaver, A.L.; et al. Cystathionine Beta–Synthase (CBS) Contributes to Advanced Ovarian Cancer Progression and Drug Resistance. PLoS ONE 2013, 8, e79167. [CrossRef] 18. Guo, H.; Gai, J.–W.; Wang, Y.; Jin, H.–F.; Du, J.–B.; Jin, J. Characterization of Hydrogen Sulfide and Its Synthases, Cystathionine β–Synthase and Cystathionine γ–Lyase, in Human Prostatic Tissue and Cells. Urology 2012, 79, 483.e1–483.e5. [CrossRef] 19. Sen, S.; Kawahara, B.; Gupta, D.; Tsai, R.; Khachatryan, M.; Roy–Chowdhuri, S.; Bose, S.; Yoon, A.; Faull, K.; Farias–Eisner, R.; et al. Role of cystathionine β–synthase in human breast Cancer. Free. Radic. Biol. Med. 2015, 86, 228–238. [CrossRef] 20. Phillips, C.M.; Zatarain, J.R.; Nicholls, M.E.; Porter, C.; Widen, S.G.; Thanki, K.; Johnson, P.; Jawad, M.U.; Moyer, M.P.; Randall, J.W.; et al. Upregulation of Cystathionine–β–Synthase in Colonic Epithelia Reprograms Metabolism and Promotes Carcinogenesis. Cancer Res. 2017, 77, 5741–5754. [CrossRef] 21. Untereiner, A.; Pavlidou, A.; Druzhyna, N.; Papapetropoulos, A.; Hellmich, M.R.; Szabo, C. Drug resistance induces the upregulation of H2S–producing enzymes in HCT116 colon cancer cells. Biochem. Pharmacol. 2018, 149, 174–185. [CrossRef] [PubMed] 22. Hellmich, M.R.; Coletta, C.; Chao, C.; Szabo, C. The Therapeutic Potential of Cystathionine β–Synthetase/Hydrogen Sulfide Inhibition in Cancer. Antioxid. Redox Signal. 2015, 22, 424–448. [CrossRef] 23. Szabo, C. Gasotransmitters in cancer: From pathophysiology to experimental therapy. Nat. Rev. Drug Discov. 2016, 15, 185–203. [CrossRef][PubMed] 24. Szabo, C. Hydrogen Sulfide, an Endogenous Stimulator of Mitochondrial Function in Cancer Cells. Cells 2021, 10, 220. [CrossRef] [PubMed] 25. Augsburger, F.; Szabo, C. Potential role of the 3–mercaptopyruvate sulfurtransferase (3–MST)—Hydrogen sulfide (H2S) pathway in cancer cells. Pharmacol. Res. 2020, 154, 104083. [CrossRef] 26. Zuhra, K.; Augsburger, F.; Majtan, T.; Szabo, C. Cystathionine–β–synthase: Molecular Regulation and Pharmacological Inhibition. Biomolecules 2020, 10, 697. [CrossRef] 27. Druzhyna, N.; Szczesny, B.; Olah, G.; Módis, K.; Asimakopoulou, A.; Pavlidou, A.; Szoleczky, P.; Gerö, D.; Yanagi, K.; Törö, G.; et al. Screening of a composite library of clinically used drugs and well–characterized pharmacological compounds for cystathionine β–synthase inhibition identifies benserazide as a drug potentially suitable for repurposing for the experimental therapy of colon cancer. Pharmacol. Res. 2016, 113, 18–37. [CrossRef][PubMed] 28. Chao, C.; Zatarain, J.R.; Ding, Y.; Coletta, C.; Mrazek, A.A.; Druzhyna, N.; Johnson, P.; Chen, H.; Hellmich, J.L.; Asimakopoulou, A.; et al. Cystathionine–β–Synthase Inhibition for Colon Cancer: Enhancement of the Efficacy of Aminooxyacetic Acid via the Prodrug Approach. Mol. Med. 2016, 22, 361–379. [CrossRef] 29. Perna, A.F.; Pane, F.; Sepe, N.; Fontanarosa, C.; Pinto, G.; Zacchia, M.; Trepiccione, F.; Anishchenko, E.; Ingrosso, D.; Pucci, P.; et al. Lanthionine and Other Relevant Sulfur Amino Acid Metabolites: Detection of Prospective Uremic Toxins in Serum by Multiple Reaction Monitoring Tandem Mass Spectrometry. Breast Cancer 2019, 2007, 9–17. [CrossRef] 30. Bahar, F.G.; Ohura, K.; Ogihara, T.; Imai, T. Species Difference of Esterase Expression and Hydrolase Activity in Plasma. J. Pharm. Sci. 2012, 101, 3979–3988. [CrossRef][PubMed] 31. Yoshigae, Y.; Imai, T.; Horita, A.; Otagiri, M. Species differences for stereoselective hydrolysis of propranolol prodrugs in plasma and liver. Chirality 1997, 9, 661–666. [CrossRef] Biomolecules 2021, 11, 1073 17 of 17

32. Berry, L.M.; Wollenberg, L.; Zhao, Z. Esterase activities in the blood, liver and intestine of several preclinical species and humans. Drug Metab. Lett. 2009, 3, 70–77. [CrossRef][PubMed] 33. Chao, C.; Widen, S.G.; Wood, T.G.; Zatarain, J.R.; Johnson, P.; Gajjar, A.; Gomez, G.; Qiu, S.; Thompson, J.; Spratt, H.; et al. Patient–derived Xenografts from Colorectal Carcinoma: A Temporal and Hierarchical Study of Murine Stromal Cell Replacement. Anticancer Res. 2017, 37, 3405–3412. [CrossRef][PubMed] 34. Singh, S.; Padovani, D.; Leslie, R.A.; Chiku, T.; Banerjee, R. Relative Contributions of Cystathionine β–Synthase and γ– Cystathionase to H2S Biogenesis via Alternative Trans–sulfuration Reactions. J. Biol. Chem. 2009, 284, 22457–22466. [CrossRef] [PubMed] 35. Majtan, T.; Krijt, J.; Sokolová, J.; Kˇrížková, M.; Ralat, M.A.; Kent, J.; Gregory, J.F.; Kožich, V.; Kraus, J.P. Biogenesis of Hydrogen Sulﬁde and Thioethers by Cystathionine Beta–Synthase. Antioxid. Redox Signal. 2018, 28, 311–323. [CrossRef]